LONGMANS AND CO., 39, PATEENOSTEB SOW. s. QDASITCH, 16, PICCADILLT. D�LA� AND CO., 37, SOHO SQUAEE, W.nbsp;KEGAN PAUL AND CO., CHASING CROSS EOAD, W.C.

HEETFORD :

PREFACE

The first part of tliis Catalogue of tlie Wealden Plants contained figures and descriptions of the Algae, Oharaceae,nbsp;Equisetinae, and Filicinae; the present Tolume is devotednbsp;to the Cycadeae and the Coniferae.

In the Author�s conclusions he gives a summary (pp. 233-241) of the Wealden flora comprised in these pages,nbsp;from which it appears that the Thallophyta are represented hy 2 sp.; the Charophyta hy 1 sp.; the Bryophytanbsp;hy 1 sp.; the Equisetinae by 3 sp.; the Filicinae by 23 sp.;nbsp;the Cycadeae hy 24 sp.; the Coniferae hy 17 sp.; uncertainnbsp;forms, 5 species ; total, 76 species.

Mr. Seward considers that � the general characters of the vegetation certainly seem to point to a tropical climate,nbsp;and there can he little doubt that the temperature wasnbsp;considerably higher than the Wealden districts enjoy at thenbsp;present day� (p. 239). He further adds that, �Lookingnbsp;at the Wealden plants collectively, we notice a very strikingnbsp;agreement with the flora of the underlying Jurassic strata,

and it would be difficult to point to any well-marked or essential difference between the plant-life of the twonbsp;periods. The evidence of palseobotany certainly favoursnbsp;tbe inclusion of tbe Wealden rocks in the Jurassic series.�nbsp;Mr. Arthur Smith Woodward informs me that the fishes ofnbsp;the Wealden beds bear testimony to the same Jurassic alliance.nbsp;We are thus led to conclude that whereas the palmonto-logical evidence, derived from the more purely marinenbsp;deposits, would induce us to place the Wealden beds withnbsp;the overlying and newer Cretaceous series�the peculiarnbsp;estuary, or lake conditions, of these mostly fresh-waternbsp;deposits, full of remains of terrestrial organisms, bothnbsp;of plants and animals, would, by their close relationshipnbsp;with the underlying and older Purbecks and Oolites, fixnbsp;a Jurassic date to this ancient land surface upon whichnbsp;the Wealden flora once flourished.

AUTHOR�S PREFACE.

In the present volume the same method of treatment has been followed as in Part I.

My thanks are again due to Mr. George Murray, Mr. Carruthers, and to the Assistants of the Geological and Botanical departments generally; also tonbsp;Mr. Rufford and Mr. C. Davies Sherborn.

To the Director of the Royal Gardens, Kew, I am indebted for the facilities afforded me of repeatedlynbsp;examining the exceptionally large collection of cycadeannbsp;plants in the Kew Herbarium.

I wish also to gratefully acknowledge communications from the late Marquis of Saporta. Sir William Dawson, Prof. Kathorst, Prof. Lester Ward, and others;nbsp;and to express my thanks to Miss Gertrude Woodward

for tlie great care and artistic skill witk whicL. ske kas executed tke litkograpkic plates.

I am indekted to Mr. Grepp, of tke Botanical Department, for tke negative from wkick Plate VIII. kas keen printed.

NOTES.

The names of authors in the footnotes, �when follo'wed by a number in brackets, or quot;without a number, will he found innbsp;the list of works at the end of the present �volume (Part II.);nbsp;those followed by A. in brackets will be found in the bibliography of Part I.

The great majority of specimens described in Part II. are from Eoclesbourne and Fairlight, near Hastings, and form partnbsp;of the Eufford Collection.

In addition to these, there are a few fossils from the Beckles Collection, and from the collections of Mantell, Dawson, andnbsp;others.

EEEATA.

P. 65. For Otozamites Klipsteinii var. superha, read Otozainites Klipsteinii var. superbus.

P. 68. For Otozamites Klipsteinii var. longifolia, read Otozamites Klipsteinii var. longifolius.

LIST OF FIGUEES IN THE TEXT.

Group SPERMAPHYTA (PHAYEROGAMIA).

In Engler and Prantl�s invaluaWe work, Die natiirlichen PJlamen-familien^ we find certain innovations as regards the classification of plants; conspicuous among the changes suggested are the termsnbsp;i/��ibryophyta zoidiogama and EmbTyophyta sipJionogama, the formernbsp;being applied to the Bryophyta and Pteridophyta, and in the latternbsp;are included the Phanerogamia.

The researches of Hofmeister, and the more recent investigations of Strashurger and others, have brought to light a multitude ofnbsp;facts, by which we have been led to a more exact knowledge as tonbsp;the natural affinities between the several plant groups. Developmental study, and our more accurate perception of the homologiesnbsp;existing between the difierent families, have tended to emphasizenbsp;the points of contact between the various divisions of the vegetablenbsp;kingdom. Any system of classification is to be welcomed whichnbsp;best enables us to give expression to recognized leading characteristics, and at the same time to bring out in a concise phraseologynbsp;the differences and resemblances between class and class. Engler�snbsp;new terms, if not used to supersede the older and widely knownnbsp;designations, may at least be recognized as marking a definitenbsp;advance towards a better understanding of phylogenetic problems.

In dealing with fossil plants we have constantly to face the difficulties of classification. With some writers there is a tendencynbsp;to strain the known points of resemblance between living andnbsp;extinct forms, and to include both in one family or sub-class; innbsp;other cases, needless isolation may he given to fossil genera hynbsp;separating them from existing types. Undoubtedly the mostnbsp;natural plan is to endeavour as far as possible to fit together thenbsp;representatives of Palaeozoic, Mesozoic, and Cainozoic genera, withnbsp;present day plants, in a common scheme of classification. It isnbsp;obviously impossible in the vast majority of fossil specimens,nbsp;to discover anything of those characters on which a modern

classification of plants is based; but we have to discriminate as best we can between valueless and important taxonomic features,nbsp;and to accept within legitimate limits the assistance of evidencenbsp;founded on analogy. To exclude fossil plants from a classificationnbsp;based on living types would be at once thoroughly- unscientific andnbsp;unnatural. Eecent botany and the botany of past ages have toonbsp;often been treated from different standpoints, and the great aim ofnbsp;palseobotanical study has thus been entirely lost sight of. Thenbsp;more we recognize the fact that plant-life, with its innumerablenbsp;problems awaiting solution, is not confined within the limits of onenbsp;age in the history of the earth, the sooner ought we to attain tonbsp;a natural system of classification.

The more important characters of the Spermaphyta {Embryophyta siphonogama) may be thus briefly summarized : �

In the great majority of cases the body of the plant is differentiated into root, stem, and leaves. The embryo is formed as the result of fertilization, by means of a pollen-grain tube, of annbsp;egg-cell enclosed in a macrospore; the fertilized egg-cell developsnbsp;into an embryo, which more or less completely fills up thenbsp;macrospore and macrosporangium. The seed may or may notnbsp;be enclosed in an ovary. The gametophyte (sexual or oophorenbsp;generation) is considerably reduced, and tbe sporophyte (asexualnbsp;or sporophore generation) has become much more conspicuous thannbsp;in the Pteridophyta.

Class GYMNOSPERM^.

Seeds naked, not enclosed in an ovary. Fertilization of the egg-cell by means of a pollen-tube. Vegetative structures capablenbsp;of secondary growth in thickness.

Stem rarely branched, leaves large and generally pinnate. In the recent genera flowers always dioecious, and without a perianth.

The Order Cycadacem, like the Marattiaeem among ferns, affords an instance of a series of plants of which few survive at thenbsp;present day, but which was abundantly represented in the vege-

tation of former periods. The recent cycads are usually divided into nine genera and two families: the Cyoadea, including onenbsp;genus, Ci/cos; and the Zamieai, with the genera Zamia, Cerato-zamia, Macrozamia, Dioon, Encephalartos, Stangeria, Bowenia, andnbsp;Microfiycm}

None of the living cycads occur outside tropical or subtropical regions. In Tertiary times the family does not appear to havenbsp;had a wide distribution, nor to have been represented by manynbsp;genera; possibly, however, a closer acquaintance with extra-European Tertiary strata may bring to light a greater numbernbsp;of cycads from these beds than are at present known. In thenbsp;Mesozoic period cycads occupied a prominent position, and hadnbsp;an extended geographical range. The Jurassic strata affordnbsp;abundant evidence that cycadean plants reached their maximumnbsp;development in that era; less numerous in the Triassic vegetation,nbsp;the Cycadace(B dwindle down to a few representatives in thenbsp;Permian and Carboniferous floras.

Before giving a summary of the earlier geological history of this exceedingly interesting section of the Gymnosperrna, we may takenbsp;note of some of the difficulties which beset any attempt to trace thenbsp;geological history of cycadean plants. As in the case of ferns, andnbsp;indeed of all fossil plants, so here again we have to deal in nearlynbsp;every instance with detached and isolated specimens of stems,nbsp;fronds, flowers, and seeds. The fronds are often abundant enough,nbsp;and their preservation frequently good; but the characters whichnbsp;are made use of in generic and specific determinations are such asnbsp;preclude any certain conclusions as to precise botanical affinity.nbsp;The nature of cycadean flowers, and their manner of occurrencenbsp;on the plant, separated as they are from the sterile fronds, presentnbsp;an obstacle to exact determination. On the other hand, the frondsnbsp;alone afford, in many instances, convenient data on which to foundnbsp;a provisional classification; their form and general habit of growthnbsp;are fairly uniform, and they do not present the same strikingnbsp;variation in leaf form which constitutes one of the many difficultiesnbsp;associated with the fronds of fossil ferns. Among recent cycadsnbsp;we have a phyllopodium exhibiting, in the majority of species,nbsp;certain distinct and easily recognized characters; usually a pinnatenbsp;structure, with stout and more or less closely set segments

traversed either by a single midrib or by a number of equal and parallel Veins. There are, however, certain variations from thenbsp;familiar oycadean type, even in some of the living genera. Innbsp;the South African genus Stangeria^ originally described in 1835nbsp;as a fern, the pinnae possess a fern-like venation, forming anbsp;strongly marked contrast to the usual Zamia or Cycm type.nbsp;Writing of this plant in 1854, Smith� notes that the peculiarnbsp;character of the leaf segments renders untenable the criterion ofnbsp;venation, usually relied upon in discriminating between fossil fernsnbsp;and oycads. Among fossil leaves there are various genera whichnbsp;have been assigned to cycads or ferns according to the preferencenbsp;of different authors. The well-known genus Nihsonia has beennbsp;placed by Schenk and others among the FilicitKB, hut it is usuallynbsp;referred to as an extinct member of the Cycadacece; the widelynbsp;distributed Tmniopteris has been assigned to both ferns and cycads,nbsp;but it is generally regarded as a genus of fossil ferns. The genusnbsp;Dictyozamites^ and numerous others might he cited as examplesnbsp;of doubtful forms which cannot with any certainty be assignednbsp;either to the Pteridophyta or Gymnosperma.

In a recent work on the Coal-Measures of Gard, Grand�Eury^ includes certain leaf forms in the class of gymnosperms, hut bynbsp;other writers these have usually been described as ferns. To settlenbsp;such doubtful cases as these, Bornemann � undertook a minutenbsp;comparative examination of the epidermal cells of recent fernsnbsp;and oycads, and found what he considered fairly safe guidesnbsp;in the rectangular or wavy outlines of the. epidermal cells ofnbsp;the leaves of these two sets of plants. Schenk� has followednbsp;Bornemann�s example in making use of this anatomical characternbsp;in the case of carbonized epidermal tissues of doubtful fossilnbsp;leaves, but the fern-like wavy walls in the epidermal cells ofnbsp;Stangeria leaves preclude any trustworthy reliance on such anbsp;method of separating ferns and cycads.

� Hooker, Bot. Mag. PI. 5121, vol. xv. [3] 1859. Reference given to Kunze, etc.

As a general rule, the fronds of recent cycads are simply pinnate; but in the Australian genus Bowenia,^ with itsnbsp;leaves, we have an exception to this rule; and it is by no meansnbsp;improbable that this character may have been shared by manynbsp;extinct genera. The late Dr. Star, of Vienna, expressed hisnbsp;belief that the well-known Coal-Measure fossils Neurvpteris andnbsp;�Alethopteris should he included in the list of Palaeozoic cycads,nbsp;and this opinion was partly founded on the resemblance of thenbsp;Carboniferous fronds to the branched leaf of Bowema. The absencenbsp;of any clearly proved fructification in these so-called ferns hasnbsp;been referred to by Stur and others in favour of a cycadeannbsp;relationship. Kidston^ has recently recorded the occurrence ofnbsp;a fertile Neuropteris frond, but the facts he publishes cannot benbsp;regarded as finally settling the position of these genera. Henbsp;figures a terminal portion of a specimen �ending in a number ofnbsp;dichotomous branchlets, the ultimate divisions being about 8 mm.nbsp;long, and bearing the fruit at their summits.�� Unfortunately thenbsp;very small pinnules associated with this fragment do not furnishnbsp;all the evidence one could desire as to the real nature of thenbsp;specimens.

Another aberrant form of a recent frond is afforded hy the Australian cycad Maorozamia lieteromera, Moore,^ in which thenbsp;pinnm are in some varieties of the species repeatedly forked,nbsp;reminding one to some extent of the Mesozoic species of Baiera.nbsp;Specimens of Macrozamia lieteromera, var. Narrabri, and var. glanca,nbsp;in the Royal Gardens, Hew, show very clearly this striking andnbsp;unusual character in cycadean fronds. (PI. XIII. Pigs. 1 and 2.)

A further variation in the form of cycadean leaves is seen in such species as Zamia Skinneri, Warscew, Z. piota (=Z. muricata,nbsp;Willd.y Z. Wallisii, A. Gr., etc. -. the pinnse of these forms reachnbsp;an unusually large size, and differ in shape from those of mostnbsp;members of the family. A single pinna of Z. W^allisii in the Rewnbsp;Herbarium measures 37 X 13 cm.; the lamina is traversed bynbsp;a few prominent and forked veins, and exhibits another peculiaritynbsp;in the possession of a short petiole. If we have to rely on leaves

alone we must necessarily expect to fall into error, but it is important not to bind ourselves too closely to the more commonnbsp;forms of cycadean fronds in endeavouring to determine the leavesnbsp;of extinct species. Seeing that the existing genera of cycads arenbsp;obviously but a few remnants of a once vigorous and numerousnbsp;family, we should not neglect the less known and more aberrantnbsp;forms of fronds in our comparisons of fossil and recent specimens.

quot;VVe are accustomed to include in the Cycadacem a large number of Jurassic and Lower Cretaceous fronds which possess some morenbsp;or less close external resemblance to those of living species. Thatnbsp;such determinations are correct we have no absolute proof, butnbsp;can only trust to the distinctly cycadean form which the leavesnbsp;present. It is possible that among such Mesozoic genera therenbsp;are included some which should rather come under the head ofnbsp;Bennettitem, a group of plants nearly allied to the true cycads,nbsp;but which possess certain peculiarities of structure of sufficientnbsp;importance to exclude them from the Cycadacem as at presentnbsp;defined. Silicified stems from the Upper Jurassic and Lowernbsp;Cretaceous rocks of England, France, Italy, America, and othernbsp;places, agree in anatomical structure with the stems of recentnbsp;cycads, but in organic connection with some of these fossil formsnbsp;there has been found a special type of inflorescence, showing a morenbsp;highly organized and specialized structure than is afforded by thenbsp;flowers of existing Cycadem or Zamiem. Our knowledge of thenbsp;vegetative and reproductive structures of Bennettites is mainlynbsp;due to the researches of Carruthers,^ Solms-Laubuch,² and morenbsp;recently Lignier.� The Bennettitem inflorescence presents certainnbsp;points of contact with the Coniferm, and the characters it possessesnbsp;in common with and distinct from those of cycadean flowersnbsp;suggest that � the Bennettitem are posterior to the Cycadacem, atnbsp;least as regards the reproductive structures.� As Lignier has saidnbsp;in his recent paper, we may perhaps regard the Bennettitem asnbsp;a family which has been derived with the cycads from commonnbsp;ancestors. We have still to learn what forms of frond werenbsp;possessed by these stems. Carruthers� speaks of a � remarkable

� Lig-nier. (For abstract of this paper see Nature, October 18, 1894, p. 594.)

oycadean lea� � from tie Lower Greensand, wiich he suggests may possibly represent a frond of Bennettites; the specimen referred tonbsp;is not in organic connection, nor in any close association, withnbsp;a stem, and therefore no satisfactory conclusion can be drawn asnbsp;to its real nature. As yet we can only reply to the question as tonbsp;what was the precise form of Bennettites leaves by mere guesses,nbsp;founded on no surer basis than a vague suspicion of probability.nbsp;The leaf-scars on the surface of the stems suggest a frond ofnbsp;cycadean habit; and in all probability many of the Mesozoic leavesnbsp;which we are accustomed to connect with true cycadean stemsnbsp;should be referred to Bennettites. To include all cycad-like frondsnbsp;in the Cycaiaeem as defined for existing species, would almostnbsp;certainly result in assigning many fossil leaves to a wrong position.nbsp;Possibly the better plan would be to assign such fossil fronds asnbsp;niay reasonably be referred to oycadean plants, to some morenbsp;comprehensive Natural Order than that of the Cyoadacea.

This brings us to the question of intermediate forms, and the association of oycadean structure with several of these syntheticnbsp;types lends an increased interest to the past history of cycads, andnbsp;at the same time enhances the difficulty of systematic treatment.nbsp;The Upper Carboniferous genus Myeloxylon {^Sten%elia, G�ppert,nbsp;Xyelepteris, Renault), found in England, Erance, and Germany,nbsp;has been assigned by several writers to the FilieineB, and placednbsp;in the Marattiaeem or Ophioglossacea; others prefer to includenbsp;it with the cycads. The structure of the vascular bundles ofnbsp;Myeloxylon petioles' is in some respects typical of recent cycads;nbsp;the spiral protoxylem elements being on that side of the xylemnbsp;facing the phloem. The bundles are collateral in form, and oftennbsp;accompanied by mechanical or stereome elements. The fundamentalnbsp;tissue contains numerous secretory canals, and in some cases strandsnbsp;of stereome. One of the most readily recognized features is thenbsp;hypodermal tissue, made up of alternating bands of thick wallednbsp;fibres and thin parenchymatous cells. Occasionally the petiolarnbsp;axis is found to be branched, and small Peeopteris-Vi^B pinnulesnbsp;have been observed attached to a slender Myeloxylon midrib. Thisnbsp;discovery by Renault of pinnules in connection with Myeloxylon

' Seward (1). Keferences givea to otker papers ; see also Zeiller (1), p. 290, pi. xxvii. fig. 1.

appears to be confirmed by some specimens in the Binney Collection ¹ ² of Coal-Measure plants. In one instance this form of petiole hasnbsp;been found inserted on a stem of Medullosa Leuckarti, G�pp. andnbsp;Stenz., a plant with distinctly cycadean characteristics. Probablynbsp;we may regard Myeloxylon as a synthetic or intermediate formnbsp;exhibiting cycadean and fern characters, but more nearly allied tonbsp;existing Cycadece than to the Filicince. In the Coal-Measure genusnbsp;Lyginodendron,'^ originally described in detail by Williamson innbsp;1873, we have another important link in the chain of cycadeannbsp;phylogeny. A revision of the English specimens of this plant, andnbsp;an examination of fresh material by Williamson and Scott, hasnbsp;brought into greater prominence the clearly defined cycadeannbsp;features exhibited by the Lyginodendron stems. It has recentlynbsp;been shown by these observers that Williamson�s genus Kaloxylonnbsp;represents the root of Lyginodendron, and we have previouslynbsp;learned that Rachiopteri� aspera, Will., with its sphenopteroidnbsp;pinnules, is a branch of the same plant.^ This is, again, an instancenbsp;of cycadean and pteridophytic characters combined in a syntheticnbsp;genus. The presence of secondary vascular tissue in Lyginodendronnbsp;lends additional interest to this instance of fern-cycad alliance.nbsp;In speaking of the occurrence of diploxyloid structure in thisnbsp;genus, Bertrand and Eenault ² regard the existence of such a typenbsp;of vascular bundle in the petioles of recent cycads as a remnantnbsp;of an ancestral structure.

The same diploxyloid arrangement occurs on an extended scale in the Permo-Carboniferous genus Poroxylon,^ and must be lookednbsp;upon as an important aid in any attempt to trace the lines ofnbsp;development of the Cycadacece. Eenault has founded the genusnbsp;Cycadoxylon � on a fragment of a silicified branch from Autun,nbsp;in which the structure of the wood and fundamental tissue bearsnbsp;a distinct resemblance to a young cycadean stem. He suggestsnbsp;that this type may find its true position between cycads and

^ Williamson (1, part iv.). The name was proposed by Gourlie in 1843. (Williamson, p. 393.) See also Solms-Laubach (A.), Fossil Botany, p. 358.

CordaitecR. TJnger�s genus Cordaites,^ with its large parallel veined leaves and tall woody stem, affords another example ofnbsp;the occurrence of cyoadean structures in association withnbsp;anatomical features suggestive of another set of plants; in thisnbsp;case it is with the Conifers that cycadean characters appear tonbsp;he combined. In the Mesozoic floras we have Carruthers� genusnbsp;Bennettites, to which reference has already been made, with itsnbsp;combination of cycadean and coniferous characters. Anothernbsp;and less accurately known plant, Williamsonta,'^ offers a difficultnbsp;problem to the palaeobotanist; but here, again, we have probablynbsp;to deal with a synthetic type closely allied to BennetUtes.

Enough has been said to show the promising character of the study of the geological history of cycads, and we may notnbsp;unreasonably entertain the hope, that we are within a measurablenbsp;distance of deciphering some of the earlier chapters in the recordsnbsp;of cyoadean development.

Before considering the questions of terminology and the details of generic and specific determination of fossil cycadean fronds, wenbsp;may briefly pass in review the recorded facts as to the past historynbsp;of the CyeadacecB, and especially such as have reference to thenbsp;representatives of this order in Palaeozoic times. In 1868nbsp;Carruthers^ expressed the opinion that �no satisfactory evidencenbsp;exists of the occurrence of Cycadece in any Palaeozoic formation.�nbsp;It is true that the facts we at present possess do not allow usnbsp;to affirm that the Palseozoio strata contain examples of plantsnbsp;which exhibit typical cyoadean structure, and of sucb a kind asnbsp;to warrant their inclusion in the Cycadacece as at present defined.

It has already been shown that certain typical features of cyoad structure are met with in various Permo-Carboniferous genera,nbsp;hut these are associated with other morphological charactersnbsp;which are unknown among recent representatives of this classnbsp;of gymnosperms. It would, indeed, he a matter of surprise ifnbsp;we found in Palaeozoic strata a perfectly typical cyoadean genus.nbsp;In the case of Jurassic plants we speak unhesitatingly of cycadnbsp;leaves, although we cannot as a rule support such assertions withnbsp;facts of anatomical details or floral structure If external resem-

blanoe of leaf form is to be trusted at all, we must admit the existence in Tipper Palseozoio rooks of a few fossil fronds, whichnbsp;have as much claim as those from Jurassic strata to be classednbsp;among the Cycadacea. In reviewing the evidence in favour ofnbsp;Palaeozoic oycads, we may for convenience sake consider Permiannbsp;and Carboniferous specimens together.

In 1848 Grutbier� figured and described a Eothliegende plant from Eheinsdorf, near Zwickau, which he designated Pterophyllumnbsp;Cottmanum. The figure reminds one to some extent of Gtenisnbsp;falcate, L. and H., but the pinnae show no trace of any anastomosing venation; the specimen cannot well be excluded from thenbsp;provisional oycadean genus Pterophyllum. Eichwald ^ has figurednbsp;a portion of a frond from the Carboniferous rocks of Konznetzknbsp;in the Altai Hills, under the name of Pterophyllum inflexum; thisnbsp;also seems to conform to the recognized characters of Pterophyllum.nbsp;Carruthers � has referred to some stems described by Eichwald fromnbsp;Eussian Permian rooks, but is of opinion that they cannot benbsp;accepted as satisfactory examples of Palaeozoic cycads; the samenbsp;author also calls attention to the specimens described by Presl andnbsp;Guillard as cycadean stems, and shows that they have no claim tonbsp;be placed among fossil cycads. Schmalhausen � has more recentlynbsp;figured a stem fragment from the Permian of Kargala in Orenburg,nbsp;which he refers to Schimper�s species Clathraria strigata, butnbsp;regards the specimen as a stem of Cordaites lancifolius, Schmalh.nbsp;In 1864 Sandberger� recorded a species of Pterophyllum, P.nbsp;hlechniodes, from the Upper Coal-Measures of Holzplatze, nearnbsp;Oppenau; the specimen seems to have been reasonably placednbsp;among oycadean fronds. Goppert,� in 1843, described what henbsp;considered to be the oldest known cycadean frond; this imperfectnbsp;fragment from Konigshiitte, in Silesia, he named Pterophyllumnbsp;gonorrachis. Two other specimens were recorded by the samenbsp;author from Palaeozoic strata as Cycadites gyrosus and Cycaditesnbsp;taxodinus; � the former is a small and imperfect specimen which

' (A.) Verstein. Eoth. Sachsen, p. 21, pi. viii. fig. 7. 2 Vol. i. p. 215, pi. XV. figs. 5, 6.

it is hardly possible to definitely refer to either cycads or ferns; the latter specimen, from the Culm beds, is more distinct, but stillnbsp;by no means a satisfactory proof of the existence of a cyeadeannbsp;species in the Culm flora. Solms-Laubachconsiders that G�ppertnbsp;was probably justified in referring the last-named species to thenbsp;Cyoaiece. An examination of the type specimens in the Breslaunbsp;Museum of these two species of Cycadites led me to regardnbsp;C. gyrosus as too imperfect for identification, and suggested thenbsp;possibility that C. taxodinw might perhaps be regarded as anbsp;fragment of a coniferous branch. If the evidence for Carboniferousnbsp;cycads rested simply ou G�ppert�s specimens it would he of littlenbsp;value; hut there have been many more perfect examples recordednbsp;from this formation. From the Permo-Carboniferous rooks ofnbsp;France we have several records of cyeadean fronds. Ihe genusnbsp;Pterophyllum has been discovered in the Upper Carboniferous bedsnbsp;of Montchanin (Saone-et-Loire), and the fragment is figured bynbsp;Saporta and Marion as Pterophyllum, GrdnA'PuTynnutti, Sap. etnbsp;Mar.^; the form of the pinnse and their manner of attachment to thenbsp;rachis support this determination. Another species is recorded bynbsp;Renault, under the name of Bphenoamp;atnites Rochei,^ from the Permiannbsp;of Autun; the figure of this plant, given by Saporta and Marion,'¹ ²nbsp;suggests a strong likeness to Noeggerathia, and it may be that if,nbsp;as some believe, the latter genus must be assigned to the FihciniB,nbsp;the same position should be given to Renault�s species. Noeggera-thia may be left for the present as one of those doubtful formsnbsp;which cannot he definitely assigned to any clearly defined position.nbsp;From the Commentry coal-field, from which so many interestingnbsp;additions have been made by Renault and Zeiller to the Coal-Measures flora, we have several new species of cyeadean leaves.nbsp;Zamite� carbonarius, Ren. and ZeilL,� is the name given to thenbsp;largest of a set of frond fragments from a particular locality innbsp;this coal-field; the type specimen consists of a portion of a stiffnbsp;rachis hearing a few alternately placed oval pinn�, and the formnbsp;of the segments is not unlike that of Noeggerathia. In addition to

this species, the same authors institute five other specific names' for isolated pinnm which do not appear to afiord any distinctnbsp;indication of specific difierence. Potoni�,* in his recent work onnbsp;the Permian flora of Thiiringen, includes all these five speciesnbsp;under Zamites carhonarius, and an examination of the flgurednbsp;pinnae certainly lends support to this view. Zeiller � has defendednbsp;Eenault�s determination, on the ground that there are certainnbsp;differences in the venation and form of the pinnae which arenbsp;hardly consistent with the suggested inclusion under a singlenbsp;species; he is, however, willing to admit that possibly Zamitesnbsp;regularis may he identical with Z. Planehardi. Whatever may benbsp;the specific value of these Commentry specimens, Zeiller regardsnbsp;them as undoubtedly fragments of the same generic form, and thenbsp;discovery of more perfect specimens leads him to found a newnbsp;genus, Plagio%amites, as more suitable for their reception thannbsp;Zamites. In speaking of the resemblance between Plagioamp;amitesnbsp;and Noeggerathia, Zeiller expresses an opinion in favour ofnbsp;including the latter genus among cycads, using the term cyoads *nbsp;in a wide sense. This opinion is partly based on the closenbsp;similarity between Noeggerathia and Plagiozamites on the onenbsp;hand, and on the marked resemblance between the latter genusnbsp;and Zamites on the other. The form of the fronds certainlynbsp;favours this view, hut such reasoning from external resemblancenbsp;cannot he accepted as conclusive when we are dealing with cycadsnbsp;and ferns. In all these cases we must he prepared to find anbsp;combination of pteridophytio and cycadean characters, and if wenbsp;were in possession of the facts of anatomical structure, we shouldnbsp;possibly be quite unable to decide definitely for one or other ofnbsp;these two groups of plants.

The Commentry flora has furnished an exceedingly fine specimen of the genus Pterophyllum^�P. FayoU, Een. and Zeill. Thisnbsp;example is unusually large and well preserved, and there can benbsp;little or no hesitation in accepting it as a Palaeozoic cycadeannbsp;frond, having an equally strong claim to be described as such as

the Mesozoic representatives of the same genus. The genus itself is merely a provisional one, and rests on external characters ofnbsp;vegetative structures, hut the cyoadeau habit is sufficiently obviousnbsp;to lend confidence to the generally accepted botanical positionnbsp;assigned to this and other cy cad-like leaves. Portions of giganticnbsp;leaves are figured by Renault and Zeiller from the Commentrynbsp;coal-field under the generic name Titanophyllum,'- and it isnbsp;suggested that possibly these may belong to Calpoxylon stems,nbsp;�which have been referred on anatomical grounds to the Cycadacece,nbsp;but these and many other leaf forms must remain in the list ofnbsp;plantie incertce sedis until additional facts are available. Renaultnbsp;has recently described another species of Permian cycad, Ptero-phyllum Comlrayi^ quot;which shows a fairly close resemblance tonbsp;P. Jaegeri, Brong. Enough has been said to sho-w that in Permo-Carboniferous times there existed certain forms of leaf structures,nbsp;which must he assigned with the numerous Mesozoic fronds to thenbsp;provisional genera of extinct cycads. The large number of seedsnbsp;from this geological horizon, with their well-preserved structurenbsp;and variety of external form, are naturally a source of difficultynbsp;as regards systematic position. There are distinct indications ofnbsp;cycadean affinity in many of the silicified gymnospermous seeds;nbsp;some belong, no doubt, to Cordaites, whilst others may be morenbsp;correctly placed in the Coniferm. The seeds of the recent genusnbsp;Ginkgo show some points of contact with those of cycads, andnbsp;among the seeds of Palreozoio plants it would not surprise usnbsp;to find cycadean and coniferous characteristics represented in thenbsp;same species. We cannot well do more than speak of these doubtfulnbsp;fossils as examples of Palaeozoic gymnospermous seeds, many ofnbsp;which distinctly resemble the seeds of recent cycads.. Grand�Eury�nbsp;includes many such fossils in the family Noeggerattiacem, a subsection of gymnosperms; the choice of this name is not a verynbsp;happy one, seeing that we know so little as to the actual positionnbsp;of Sternberg�s genus Noeggerathia.

Ascending the geologic series from the Permian to the Epper Jurassic strata, we find a gradual increase in the number andnbsp;variety of cycadean fronds, and in the Wealden vegetation the

Cycaiacem were represented by many large and striking species. Further reference �will be made to the Lower Cretaceous cycads innbsp;the general review of the Wealden flora at the end of this volume.nbsp;Throughout the Cretaceous and Tertiary series we have evidencenbsp;of a decline in the relative importance and numerical proportionnbsp;of the CyeadacecB. It has been suggested that possibly the paucitynbsp;of species may in some measure be explained by our very imperfectnbsp;acquaintance with tropical and subtropical Cretaceous and Tertiarynbsp;plant-hearing strata ;' it may he that the rocks of these eras werenbsp;deposited under cliraatal conditions which were not favourable tonbsp;a rich development of cycads. Heer^ has described various frondnbsp;fragments from Tertiary beds which are not particularly satisfactory as records of cyoadean species. The two species Nilssonianbsp;Serotina, Heer, and N. pygmaa, Heer, from the Miocene flora ofnbsp;Sachalin Island, are both founded on fragments which may possiblynbsp;belong to that doubtful genus iu which they have heen placed.nbsp;From the Hpper Fresh-water Molasse of Schatfhausen, the samenbsp;author describes a structureless stem as Cyeadites Escheri, Heer; ^nbsp;the appearance of the scale-covered surface lends some support tonbsp;this determination, but the specimen is too imperfect to be of anynbsp;particular importance. Heer figures a fragment of a frond fromnbsp;Lausanne under the name Zamitea {I)ioon ?) tertiariua,^ foundednbsp;on a poor and fragmentary specimen. Three species of Tertiarynbsp;cycads are figured by Saporta and Marion in their V�volutionnbsp;du r�gne v�g�tal: ¹ ² one of these is assigned to Zamioatrohua�Z.nbsp;Saportanus, Schimp., and may possibly be rightly described asnbsp;a cyoadean cone, but its precise nature cannot be definitelynbsp;ascertained. The other two species, Zamitea epibiua, Sap., andnbsp;Encephalmrtoa Gorceixianua, Sap., are most probably true cycads.nbsp;Ettingshausen�s Hew Zealand specimen, described as Zamitea sp. ?nbsp;cannot be accepted as trustworthy evidence of a Tertiary cycad.�nbsp;From Australia the same author records Anomo%amitea Muelleri,^nbsp;Ett., a species based on small fragments of what may be a

� (A.) FI. Tert. Helvet, p. 46, pi. xv. and pi. xvi. fig. 1. ^ Lee Phanerogames, vol. i. p. 116.

cycadean leaf. Anotlier possible Tertiary cycad is described by Ettingsbausen from the Miocene beds of Leoben; to this thenbsp;name GeraU%amia Sojfmanni, Ett., has been assigned.'^ The singlenbsp;imperfect pinna �which is figured by the author of the species,nbsp;does not afford sufficient evidence that it belongs to this particularnbsp;recent genus. Granting its cycadean nature, and even this entailsnbsp;a considerable amount of faith, there is surely no reason why thenbsp;fragment should not be referred to some other genus than the onenbsp;cho.sen; one might suggest � cycadean pinna ? � as a more fittingnbsp;term than C. Hojfmanni. Gdppert�s Tertiary species of anbsp;Greenland cycad, Zamites arcticu�? is founded on a fairly goodnbsp;specimen, and certainly appears to be correctly included amongnbsp;the Cycadacem. These few examples of fragments described bynbsp;various writers as cycadean fronds, sufficiently demonstrate thenbsp;meagre relics of this order of gymnosperms in Tertiary rocks.

In his Monograph on the Jurassic cycads, Saporta ^ has given a useful and critical summary of the history of the literature onnbsp;fossil CycadacecB, to which is added a series of definitions of thenbsp;chief characters by which the several genera of fronds may benbsp;recognized. Certain suggested emendations of some of thesenbsp;diagnoses will be found under the head of the respective generanbsp;in the descriptive part of this Catalogue.

Without following the gradual additions to our knowledge of fossil cycadean fronds during the last sixty or seventy years,nbsp;or attempting to discuss the numerous classifications proposed bynbsp;various writers, it may serve a useful purpose to draw attentionnbsp;to some of the difficulties and possible sources of error associatednbsp;with the investigation of the past history of cycads.

The characters generally made use of in the separation of distinct genera of fossil cycadean leaves may be enumerated asnbsp;follows: (i.) The method of attachment of the pinn� to thenbsp;rachis, and whether persistent or deciduous, (ii.) The nature ofnbsp;the base of the pinnrn, auriculate or gradually tapered, etc., thenbsp;presence or absence of a distinct basal callosity, (iii.) The pinnanbsp;apex, whether truncate, acuminate, etc. (iv.) Venation, (v.) Thenbsp;angle of insertion of the pinnae on the rachis; the alternate

or opposite arrangement of the pinnae, (vi.) The form of the epidermal cell-walls, (vii.) Presence of spines on the segmentnbsp;margin. In addition to these more detailed characters, the formnbsp;of the frond as a whole, whether simple, pinnate, or hipinnate,nbsp;and the shape of the individual pinnae, long, narrow, broadlynbsp;oval, etc., are important characters to he kept in view.

In G�ppert�s valuable paper on fossil cycads,^ the wholesome warning is given that to define generic characters within suchnbsp;narrow limits as are often adopted, results in an unnecessarynbsp;multiplication of genera, and tends to confusion and to increasenbsp;the difficulties of determination. Allusion has already been madenbsp;to the numerous leaves, the affinities of which cannot he definitelynbsp;settled until further data are forthcoming. As regards the genusnbsp;Nihsonia, some writers have argued for its inclusion among ferns,nbsp;hut others prefer to consider it an unusual form of cycadeannbsp;frond. Tmniopter�, Neuropteris, Noeggerathia, and a host of othernbsp;leaves must for the present he left in a somewhat doubtfulnbsp;position. The genus Slangerites, instituted by Bornemann,� hasnbsp;been used by a few authors as a convenient term for certainnbsp;Tceniopteris-Vike leaves, hut the name seems unnecessary, andnbsp;may he ranked among those misleading titles which suggest anbsp;relationship to a living genus which is not supported by factsnbsp;of any taxonomic value. Saporta, in speaking of this genus,nbsp;remarks that the author of the term Stangerites � a ajout� a cenbsp;qui s��tait fait avant lui une confusion reellement inextricable etnbsp;p�riss� de diffioult�s la synonymie des principals esp�ces, decritesnbsp;d�apr�s leurs feuilles seulement.� ¹

In the recent species Stangeria paradoxa (Moore), it is worthy of note that we have pinnae with entire margins, and others withnbsp;deeply out lobes extending to the midrib; some of the deeplynbsp;divided laminae suggest in a slight measure a Nilssonia form ofnbsp;leaf. In a small plant of Cyeas eireinalis, L., in the Eoyalnbsp;Gardens, Kew, I noticed an abnormal form of leaf structure atnbsp;the base of a young frond, suggesting another example of annbsp;approach to the Nilssonia type of leaf. Instead of the ordinarynbsp;uninerved and separate pinnae characteristic of Cycas, this

particular specimen sliowed a lamina on either side of the hasal part of the raohis, having the appearance of several pinn� fusednbsp;together laterally, the position of each segment being indicatednbsp;by a strong vein.

In every classification which is based on artificial characters, and which gives us provisional genera, there must necessarilynbsp;be inconsistencies, and in all probability plants possessing no closenbsp;relationship will often he included in the same genus. Amongnbsp;fossil ferns this is especially the case; as regards cycads, althoughnbsp;not perhaps to au equal extent, there are the same difficultiesnbsp;to he encountered owing to the isolated and fragmentary naturenbsp;of the specimens on which determinations are based. It may,nbsp;perhaps, be possible to add to the convenience of classification,nbsp;or to minimise the danger of conveying wrong impressions bynbsp;ill-chosen names, by adopting some more admittedly provisionalnbsp;classification than is at present employed. An attempt to modifynbsp;onr present system, which is too often inadequate and unsatisfactory, will be more appropriately undertaken after the Wealdennbsp;and Jurassic genera have been subjected to a detailed treatment.nbsp;For the present, attention may be drawn to some of the obstaclesnbsp;in the way of accurate determination of fossil fronds.

As regards the manner of attachment of pinnae to the rachis : among recent genera there are some in which the pinn� arenbsp;readily detached from the rachis by a well-marked line of articulation ; e.g. in such forms as Zamia furfuracea, Ait., with broadnbsp;oval pinn�, and other species of the same genus. In species ofnbsp;Encephalartos, Ceratozamia, Eioon, etc., there are distinct andnbsp;sharply defined soars left on the axis of the frond on the fallnbsp;of the pinn�; in others, again, the pinn� are persistent. Amongnbsp;fossil forms, the rachis soars and detached pinn� with clearly outnbsp;bases evidently point to a deciduous habit; hut it is often a matternbsp;of great difficulty to decide definitely as to the existence of suchnbsp;a character, and it is quite unsafe to trust to a feature of thisnbsp;bind as an essential character in generic classification. It is bynbsp;no means easy in some cases to distinguish the true aurioulatenbsp;base of a pinna, from a cordate form produced by the crushingnbsp;and flattening of a thick and leathery segment, hornemann hasnbsp;called attention to this possible source of error, and points tonbsp;the absence of any true aurioulate base in the pinn� of recentnbsp;fronds. In examining herbaria specimens of some Encephalartos

fronds, such as E. Coffer, Miq., and other species �with broad stout pinnse, one frequently notices that the basal portions ofnbsp;the segments have been depressed in such a way as to presentnbsp;in surface view the appearance of a distinct auriculate base.nbsp;In some of the examples of Otozamites Klipdeinii (Dunk.)nbsp;var. superha, described in the present volume, this has probablynbsp;been the case; but, thanks to the large number of excellentnbsp;specimens in the Eufford Collection, it is perfectly clear thatnbsp;the pinnse of this striking plant possessed auriculate bases.nbsp;The absence or presence of a callosity is often a question ofnbsp;considerable uncertainty among fossil leaves, and the existencenbsp;of a basal thickening, often none too distinct in the segments ofnbsp;recent species, can only be satisfactorily made out in exceedinglynbsp;well-preserved specimens. In some cases there is a distinctnbsp;wrinkling of the coaly surface layer in the position where anbsp;callosity would naturally occur, and this may no doubt havenbsp;sometimes resulted from a callosity in the living pinna, but innbsp;others the same appearance may be due to mere bending of thenbsp;frond segments in the process of fossilization.

It has been shown by more than one writer how easily the manner of attachment of the pinnae to the rachis may be obscurednbsp;by the frond being seen from its under side. In the case of Eioonnbsp;a view of the upper face of the leaf would lead one to refer itnbsp;to such a genus as Dioonites; but if the lower surface werenbsp;exposed to view Pterophyllum would be the most appropriatenbsp;genus. In a species like Maorozamia Benisoni, Moor and Meull.,nbsp;in which the pinnse are attached along a median line on thenbsp;upper face of the rachis, the same pinnse seen from below arenbsp;apparently inserted laterally on the axis, and show no signs ofnbsp;decurrent bases. Braun�s figures of Zamites {�Otozamites) hrevi-folius, Braun,' as seen from above and below, bring out verynbsp;clearly the striking contrast between the two views ; the samenbsp;kind of difference is well shown in Beistmantel�s figures ofnbsp;Ptilophyllum aoutifolium, var. maximum, from the Eajmahal Hillsnbsp;of India.'*

The comparative breadth of the pinna base is a character which varies considerably according to the position of the segment on

the rachis, whether towards the tip or the lower part of the leaf, or according to the age of the frond. The terminal pinnm arenbsp;often strongly decurrent at the base, whilst the lower segmentsnbsp;have a uniform width; a young frond of Cycas media, Br.,nbsp;shows pinnse with no indication of tapering towards the rachis,nbsp;hut the older and broader segments are distinctly narrowed.

Stress is often laid on the form of the pinna apex, whether truncate, acute, etc. In the typical form of Pterophjllum thenbsp;pinnae have truncated apices, but specimens are occasionallynbsp;referred to this common provisional genus in which the apices ofnbsp;the segments are clearly not truncate. Bornemann defines thenbsp;genus as possessing pinnm which may he either straight at the tipnbsp;or obliquely truncate, and this wider definition is probably thenbsp;most satisfactory. In such a specimen as that of Otozamiiegnbsp;GUppertianua (Dunk.), figured in PI. I. Pig- 2, some of thenbsp;Pinnffi are more or less truncate at the tip, and others regularlynbsp;acuminate. In the examples of Zamites Puchianus (Btt.) in thenbsp;British Museum Collection, the variation in the apical terminationsnbsp;of the pinnse has proved a difficulty, some specimens havingnbsp;gradually tapering segments, and others showing obtusely terminated apices, but the occurrence of some intermediate formsnbsp;throws doubt on the value of such a feature as a leading specificnbsp;characteristic.! In dried fronds of Cycas revoluta, Thunb., it is notnbsp;Uncommonly found that in many of the pinnse the pointed spinynbsp;upex has been replaced by a rounded termination, with a slightnbsp;median depression at the end of the single vein. As a rule,nbsp;however, the pinnae of recent fronds maintain a fairly uniformnbsp;mode of termination in the same species. The venation is notnbsp;always readily made out even in fairly good specimens; the thicknbsp;Coriaceous pinnse of some recent species, with their indistinct veins,nbsp;prepare us for a similar difficulty in dealing with fossil leaves.nbsp;It is well known that the lower surface of a pinna often showsnbsp;'rery distinct venation, while the veins on the upper surface arenbsp;quite obscure. In Cycas we have a convenient venation character,nbsp;'which is taken as the essential feature of the fossil genus Cycadites'nbsp;but in this case, as we shall see later in describing the genus,nbsp;frequent mistakes have been made in the determination of specimens, which apparently rest on such a readily recognized character

as the presence or absence of a midrib. Schenk has pointed out Dunker�s error with regard to the supposed Wealden species ofnbsp;Cycadites, C. Morruianm, Dunk., and a careful examination of thenbsp;English material confirms Schenk�s correction. In some recentnbsp;species of Cycas the midrib is by no means obvious on the uppernbsp;surface of the pinn�; e.g. in a dried specimen of Cycas Cairnsiananbsp;(MuelL), the upper convex surface of a pinna presents an appearance suggestive of a few parallel veins, no doubt due to wrinkling,nbsp;rather than of a single midrib. In C. Beddomei, Dyer,� thenbsp;margins of the pinnae are strongly revolute, and a cast of thenbsp;lower surface of a pinna would show too longitudinal ridgesnbsp;separated by a distinct groove, the latter being formed by thenbsp;projecting central vein. On the other hand, the tendency tonbsp;a revolute margin in the long, narrow, linear pinnae of othernbsp;genera than Cycas, often leads to an appearance which mightnbsp;easily be mistaken for a stout midrib in fossil specimens of suchnbsp;a leaf. The under surface of the pinnae of Encephalartosnbsp;Ghellinchii, Lem. (PI. XIII. Fig. 3), Zamia angustifoUa, Jacq., etc.,nbsp;shows a narrow median groove separating the revolute edges of thenbsp;narrow segments, and this same folding might readily give rise tonbsp;a midrib-like character in the segments of fossil fronds. In a fewnbsp;exceptional cases there is an anastomosis of the veins in cycadeannbsp;leaves; among fossil fronds Lindley and Hutton established thenbsp;genus Ctenis, for � all leaves having the general character ofnbsp;Cycadem, but with veins connected by forks or transverse bars.�nbsp;As regards living genera, some authors refer to Bowenia andnbsp;Stangeria as having anastomosing veins, but the occurrence ofnbsp;anastomosis in the segments of the former genus is denied bynbsp;Engler ^ and others. The proximity and number of the veins innbsp;a pinna are characters of no little value in the separation of specificnbsp;forms, but the difficulty of eliminating the effects of fossilizationnbsp;and the different appearances presented by the upper and lowernbsp;faces, render it difficult to arrive at any very trustworthy conclusion as to venation characters. In speaking of cycadean venation,nbsp;Bornemann ^ suggests that the characteristic veins of Zamia have

usually been overlooked as a means of identifioation. The inclination of pinnae to the raohis, and their alternate or opposite disposition are characters �which have been used as the basis ofnbsp;specific determination, but such features as these are likely to provenbsp;misleading unless used with great caution. In one part of a frondnbsp;the pinnse may be distinctly opposite, and in another alternate.nbsp;The same kind of �variation in the angle of insertion of a segmentnbsp;to the raohis, is readily seen in the large fronds of such recentnbsp;species as Cerato%amia mexicana, Brong., Macrozamia Macleayi, Miq.,nbsp;and many others ; also among fossils in the larger specimens ofnbsp;Zamites Buchianus (Ett,), etc. A comparison of the young andnbsp;old fronds of many oyoads reveals the same striking difference asnbsp;regards the inclination of the pinnee. The open or closely setnbsp;arrangement of pinn� is another misleading character; e.g. in annbsp;old frond of JSncephalartos longifoltus, Lehm., the pinn� are foinbsp;the most part in contact �with one another, hut the young frondnbsp;presents a distinctly open habit, 'with the pinn� much morenbsp;openly arranged. In Otczamites Gopperftanus (Dunk.) there isnbsp;the same difference in this respect between the upper and lowernbsp;portions of the same specimen, e.g. PI. I- Pigs. 1 and 2.

The form of the epidermal cells is a character of doubtful quot;'''alue, and at the same time one �which can only be made use ofnbsp;Under favourable conditions of fossilization. The custom of associating spiny margins with the pinn� of JSnoephalartos has led annbsp;American �writer to adopt this feature as the leading characteristicnbsp;cf his genus ISncepJialartopsis.^ Pontaine has founded this newnbsp;genus on some very fragmentary and imperfect pinn� with spinousnbsp;margins and anastomosing veins. None of the figured fiaginentsnbsp;ufford any clue as to the nature of the pinna base, or as to thenbsp;manner of insertion on the rachis. The material is hopelesslynbsp;inadequate for the institution of a new genus. The fact of thenbsp;fragments possessing anastomosing veins deters Pontaine fromnbsp;including them in the recent genus EueeplialdTios; as it is, lienbsp;prefers to institute a ne^w term, and to consider the species asnbsp;probably a �prototype� of the recent genus. It is true one isnbsp;accustomed to associate spiny pinn� �with species of BJncepJuilartos,nbsp;hut there are several forms of that genus in which no indicationnbsp;ef such a character is found; and on the other hand, spiny pinn�

are met with in Dioon eduh, Lind., and to a certain extent in Zamia Lindeni, etc. Newberry' has doubtfully referred a smallnbsp;portion of a frond from the Ehsetio beds of Honduras to the genusnbsp;Micephalartos, hut expresses his hesitation as to the trne positionnbsp;of the specimen by adding a query to the generic name. It isnbsp;suggested by Newberry that the Miocene cycad named by Saportanbsp;Encephalartos Gorceiximus, does not correspond so closely with anynbsp;living member of the genus as does the Honduras specimen; henbsp;adds: �This correspondence in the form of the pinnules is sonbsp;close that I felt warranted in placing our fossil provisionally in thenbsp;genus Zncephalartos. The fructification will of course be necessarynbsp;for a demonstration of generic identity, and has not yet beennbsp;obtained.� In Lesquereux� posthumous monograph on the Dakotanbsp;flora, there is a fragment figured and described as a new species,nbsp;under the name Enceplialartos cretaceus, Lesq.^; but this is anothernbsp;example of what we may regard as the utterly unwarrantable usenbsp;of a recent generic name, and the institution of a new species onnbsp;absolutely insufiicient data. It does not seem to have been generallynbsp;recognized that the living species of Encephalartos present a greatnbsp;variety of leaf form, from the long and narrow pinnae of suchnbsp;species as E. OhellincMi, Lem. (PI. XIII. Pig. 3), and E.nbsp;cycadifoUus, Lehm. (PI. XIII. Fig. 6), through E. Lehmanni,nbsp;Lehm., etc., to E. Gaffer, Micp, and E. horridus, Lehm. Therenbsp;is a very striking difference between the young and old frondsnbsp;of E. cycadifoUus : in the former the pinnae are much more obliquenbsp;to the raehis, and have not assumed the stiff and straight characternbsp;which is so pronounced in the latter. Many of the Mesozoicnbsp;cycadean fronds present a striking similarity to Encephalartosnbsp;leaves, but it would be exceedingly rash to apply the name ofnbsp;the recent genus to even the best of these fronds, and still morenbsp;unwise to make use of it for the merest fragments of isolatednbsp;pinnae.

It will be most convenient to consider the Wealden specimens referred to the Cycadacem under the headings Frondes and Trunci;nbsp;and also to describe such seeds and reproductive structures as maynbsp;possibly be included among cycadean fossils. Unfortunately thenbsp;isolated mode of occurrence of leaves, stems, and seeds does not

allow, in tlie great majority of cases, of any certain conclusions as to the relation of the detached members one to another.

Otozamite� Klipsteinii (Dunk.), var. superha mihi. Otozamites Klipsteinii (Dunk.), var. longifolia mihi.nbsp;Otozamites cf. 0. Klipsteinii (Dunk.).nbsp;Otozamites sp., cf. 0. Reibeiroanus, Heer.

Sternberg proposed this name in 1825 for three fossil plants the Lower Cretaceous of Hor in Scania, and one from Eadnitznbsp;m Bohemia. He defined the genus as follows: � Folia pinnati-fida seu pinnata, nervis validis simplicibus e rhachi horizontaliternbsp;exeuntibus.�

Sternberg�s species Cycadites Nilssoni had been previously figured by Nilsson in 1820,� but he left the plant unnamed;nbsp;fihis species is now included in the genus Nilssonia. Anothernbsp;uf Sternberg�s species, C. linearis, is no doubt, as Presl firstnbsp;�uggested,^ a fragment of some fossil stem. Cycadites palmatus,nbsp;Sternb., from Eadnitz, is probably a fragment of Cordaites, andnbsp;C'. iamiafolius suggests a coniferous twig. In 1824 Nilsson^

* Sternberg (A.), Flor. Vorwelt, fase. vii. p. 194. � (2), p. 143, pi. ii. bis. figs. 4 and 6.

uninerved and apparently palmately-arranged segments; this he described as probably a Filioite. Brongniart ^ afterwards referrednbsp;Nilsson�s plant to Cycadites, on account of the resemblance ofnbsp;the leaf segments to the pinnae of the recent genus Cycas. Asnbsp;Schenk ^ has pointed out, Nilsson�s figure in all probabilitynbsp;represents an Aralia leaf, and the fossil is certainly not a speciesnbsp;of Cycadites. In Brongniart�s Prodrome'^ we have the followingnbsp;definition of the genus Cycadites :�

� Feuilles pinn�es, d pinnules lin�aires, enti�res adh�rentes par toute leur base, travers�es par une seule nervure moyenne,nbsp;�paisse ; point de nervures secondaires.�

He regards the single-veined linear pinnae as the important feature, and in spite of the fact that the first specimen to benbsp;included under this generic name was incorrectly determined,nbsp;this definition of Cycadites has been generally adhered to.

Sohimper, Saporta, and other authors have, in the main, adopted Brongniart�s diagnosis. We may perhaps most conveniently define Cycadites as follows :��

Frond pinnate, pinnae alternate or opposite, linear, lanceolate, entire, with a single median vein; attached to the rachis bynbsp;the entire base, the lower margin of which may be slightlynbsp;decurrent on the frond axis, or slightly narrowed towards thenbsp;point of attachment.

It is better to confine our definition to the frond characters, and thus frame it in such a manner that it practically includesnbsp;those fossil fronds which have a cycadean habit, and resemblenbsp;the recent Cycas in the possession of uninerved segments. Innbsp;several cases Cycadites fronds have been found in close association with characteristic Cycas-Yiks carpellary leaves; but in thenbsp;majority of specimens we have only sterile fronds, and it isnbsp;better, therefore, to have some definition which enables us tonbsp;give such leaves a place in a convenient genus, which does notnbsp;depend upon special characters of fertile leaves.

The genus Cycadites, as defined by most writers since the days of Brongniart, possesses easily recognized characters, andnbsp;ought not to present any very serious difficulty in the way of

g6iieric dstermination. When we come to examine the various plant fragments which have been figured as representatives ofnbsp;the genus at different geological horizons, it becomes apparentnbsp;that the mere acceptance of a list of Cycadites species as an indexnbsp;of the past history of the genus would undoubtedly lead us intonbsp;error. In any case it would he rash to maintain that a recordnbsp;of even the most perfectly preserved specimens of the Cycaditesnbsp;type of frond, affords an epitome of the geological history of thenbsp;genus Cycas, The occurrence of fossil carpellary leaves verynbsp;similar to, or practically identical with, those of Cycas, lendsnbsp;confirmation to the position assigned to many of the Cycaditesnbsp;fronds; but as regards other species we can only express thenbsp;opinion that they are parts of a plant which closely resemblesnbsp;in habit, and probably in structure, the living genus. It hasnbsp;already been pointed out that the pinnoe of Cycas circinalis, L.,nbsp;may occasionally he united laterally and assume a form suggestive,nbsp;in some degree, of Nilssonia or Pterophyllum. No great weightnbsp;can be attached to this single instance of such lateral fusion, hutnbsp;it is worth noting as having a possible connection with somenbsp;of the fossil leaf forms which present little resemblance tonbsp;recent fronds. Saporta has called attention to the similaritynbsp;between some Cycadites species and Nilssonia, and one of Heer snbsp;species, C. Biclcsoni^ from the Cretaceous of Greenland, seemsnbsp;to possess pinnm which are either in contact with one another,nbsp;or actually united by the margins.

Berger figured a fragment as Cycadites alatus, Berg.,* and compared it with Nilssonia irevis, Brong., the same plant beingnbsp;afterwards renamed by Goppert Nilssonia Bergeri.^ As regardsnbsp;the first record of Cycadites in Palaeozoic rocks, it cannot be saidnbsp;that there is any very decided evidence of the occurrence of thisnbsp;genus, hut Goppert�s C. taxodinus is by no means such a doubtfulnbsp;representative of the genus as several of the species describednbsp;from newer beds. Goppert�s G. gyrosus may perhaps be a portionnbsp;of a young frond with its pinurn circinately rolled, hut it is notnbsp;enough to establish the existence of Cycadites in Carboniferousnbsp;times. Sterzel has recently figured an imperfectly preserved

Heer (A. 3.), FI. foss. Arct. vol. iii. pt. ii. p. 99, pis. xxvii. and xxviii. Berger, p. 22, pi. iii. figs. 5 and 6.

impression from the Middle Eothliegende of Possendorf, Saxony/ which he speaks of as � Cycadites ? or Walchia sp.,� hut doesnbsp;not consider it suiSciently distinct to allow of accurate identification. The figure entirely justifies Sterzel�s doubtful attitude.nbsp;In the Mesozoic beds Cycadites fronds become more abundant;nbsp;a list of most of the species has been given by Solms-Laubach.^nbsp;A few of the so-called Cycadites species call for special mention.nbsp;The Jurassic specimens described by Leckenby, from Cloughton,nbsp;as Cycadites zamioides^ are probably, as Nathorst suggests,nbsp;fragments of a conifer.nbsp;nbsp;nbsp;nbsp;The type specimen of Leckenby�s

species in the Woodwardian Museum, Cambridge, bears a label on which Nathorst has written, � A conifer of the genusnbsp;Palissya �; and Eichards,¹ ² who examined the specimens a fewnbsp;years ago, adopts this view. In the case of some small indistinctnbsp;impressions, it is often very difficult to decide between a twig ofnbsp;a conifer with its spirally arranged leaves extended in onenbsp;plane, and a small cycadean frond with its uninerved pinnsenbsp;inserted on the two sides of a raohis. A branch of Cephalotaxusnbsp;Fortunei, Hook, might very easily be mistaken for Cycadites ifnbsp;found in a fossil state with the details of structure imperfectlynbsp;preserved. Heer has described several species of Cycadites fromnbsp;Arctic localities, hut the figures do not inspire confidence in hisnbsp;determinations. Cycadites Dichsom�� may very probably be a truenbsp;Cycadites; C. sibiricus, Heer,� and C. gramineus, Heer,� fromnbsp;the Jurassic rocks of Siberia, are both founded on the merestnbsp;fragments of single pinnae, and cannot be taken as trustworthynbsp;records. The institution of species on such minute fragmentsnbsp;as the figures represent, is to be greatly deplored; the resultnbsp;can only be either to mislead those who are willing to acceptnbsp;all fossil species described by well-known authors, or to deternbsp;the more sceptical from attaching any importance to fossil plant

� Leckenby (A.), Quart. Jouru. Geol. Soc. vol. xx. 1864, p. 77, pi. viii. fig. 1.

6 Heer, FI. foss. Arct. vol. iii. pt. ii. p. 97, pis. xxyii. and xxviii.; and vol. vL pi. xiv. fig. 10.

determinations whicli do not rest on other characters than those of external form. Another species from the same beds, C. 1nbsp;planieoata, Heer,^ is founded on imperfect pinnse, hut in thisnbsp;case Heer definitely admits the doubtful value of the name.nbsp;I�he specimen described by the same authority from the Tertiarynbsp;beds of Schaffhausen as Cycadiies Eacheri'^ is very likely anbsp;eycadean stem, but the genus Gycadites has been restricted tonbsp;fossil fronds, and Heer�s stem fragment should he referred tonbsp;some other genus, in order to avoid the confusion likely to arisenbsp;from using the name in a more comprehensive sense. Dawsonnbsp;has described some fronds from the Middle Cretaceous of thenbsp;B-ocky Mountains, which he names Cycadites JJnjuga,^ andnbsp;compares with Heer�s Cycadites Eicksoni. The two figures ofnbsp;the Canadian specimens do not appear to agree as regards thenbsp;Cycadites form of pinna: in Dig. 2 each pinna appears to havenbsp;several parallel veins, and the general habit seems differentnbsp;from that in Dig. 2l; if Dig. 2 be an accurate representationnbsp;of the specimen, and the vein-like lines are not the draughtsman�s shading, it could hardly he accepted as a true Cycadites.nbsp;Beistmantel figures a fragment under the name of Cycaditesnbsp;lt;^onstrictus, Deist.,� and speaks of a midrib in the hasally constricted pinnm ; the figure does not show any distinct midrib,nbsp;and leaves one in doubt as to the wisdom of choosing the genusnbsp;Cycadites.

1,�Cycadites Romeri, Schenk.

1871. Cycadites Mdmeri, Schenk, Palseontographica, vol. xix. p. 229, pi. xxxii. figs. 1, la.

* nbsp;nbsp;nbsp;Feistniantel, Gond. Flor. vol. i. pt. iv. p. 25, pl. vii. fig. 10.nbsp;� Schenk (A. 2), Palseontographica, vol. xix. p. 229.

�Folia petiolata pinnata, petiolus validus, segmenta linearia pateatia Integra alterna basi dilatata breviter deourrente sessilia,nbsp;cent. longa, 2 mm. lata, uninervia, nervus medianus validus.�

He refers to the recent species Oyoas Siamensis, Miq., as most nearly allied to the fossil frond. This is the only example ofnbsp;Cycadites recognized by Schenk among the North Germannbsp;Wealden plants; the specimens referred by Dunker to thatnbsp;genus being without the characteristic single vein in the pinnae.nbsp;The English specimens, for which the name Cycadites Saportm,nbsp;sp. nov., is proposed, differ from the present species in theirnbsp;narrower and more approximately disposed pinnae, inclinednbsp;almost at right angles to the raohis. The scanty material innbsp;the Kuiford Collection referred to Cycadites R�meri enables us,nbsp;however, to add one or two points to the original diagnosis bynbsp;Schenk :�

Frond pinnate, rachis strong, pinnae linear and narrow, obliquely and laterally attracted to the rachis, entire, alternate,nbsp;with slightly broadened and somewhat decurrent base, singlenbsp;median vein, apices acuminate and terminating in a sharp point.

V. 2738. Fig. 1.

An imperfect specimen, showing several partially preserved pinnte, the largest of which has a length of cm., as innbsp;Schenk�s specimen, and a breadth of 3 mm.

In the type specimen of C. R�meri the pinnse are broken at the apices, but in the English example the sharply acuminatenbsp;tips are clearly preserved, and correspond closely with thosenbsp;in C. Saportm, Cycas revoluta, Thunb., etc. Each pinna isnbsp;traversed by a median groove, which must prohahly be regardednbsp;as the midrib seen from the under side, but it should be noticednbsp;that there are in some of the pinnse slight variations in thenbsp;breadth of the groove, and it occasionally departs somewhat

from a strictly median course. The appearance, indeed, is such as to suggest a folding over of the pinnfe margins. It hasnbsp;already been pointed out in the introductory remarks on fossilnbsp;cycadean fronds, how the pinnss of such recent species asnbsp;J^ncephalartos GhelUncUi, Lem., may become folded over untilnbsp;a narrow median groove is left in the middle of the lowernbsp;surface of the segment, representing the line of separation ofnbsp;the recurved edges (PI. XIII. Fig. 3). On the other hand, wenbsp;may have a similar curling over in the pinnae of a true Cycas;nbsp;fgt;ut in the present specimen the narrow line is for the most partnbsp;perfectly median and of uniform breadth, and cannot well benbsp;attributed to any other cause than the presence of a central vein.nbsp;At one corner of the specimen there are three pinnse, whichnbsp;clearly demonstrate a folding over of the margins, but this isnbsp;in itself no proof of the absence of a single vein. These pinnanbsp;fragments are in oblique contact with what appears to be anbsp;portion of the rachis, and if we may regard the two asnbsp;organically connected, the segments exhibit the same charactersnbsp;as regards the form and attachment of the base as Schenk hasnbsp;described in the German examples. Ecclesbourne. Rufford Coll.

2.�Cycadites Saportse, sp. nov.

The difficulty of recognizing the essential character of Cycadites the pinnae of fossil fronds has made itself felt in no smallnbsp;degree in dealing with the present series of specimens. Thenbsp;figures and descriptions given by Eomer and Dunker of thenbsp;(lycadites-Ytke leaves have to be viewed in the light of Schenk�snbsp;Diore recent statements,^ based on an examination of Bunker�snbsp;fype specimens. If we leave out of consideration those portionsnbsp;�f cycadean fronds which are figured by Dunker as Cycadites

Morridanus, Dunk.,' and wkich present a close agreement in habit with the English specimens, we have only C. R�meri, Schenk,nbsp;among Wealden fronds with which to compare C. Saportm. Thenbsp;differences between C. R�meri and the present species are, Inbsp;believe, too well marked to admit of a single specific designation.

In view of the exceptionally large size of the Ecclesbourne fronds, and the satisfactory manner of preservation, it is better tonbsp;adopt a new specific term, and I have ventured to identify thenbsp;name of the Marquis of Saporta with this new form of Lowernbsp;Cretaceous cycad.

Since the above was written the Marquis of Saporta�s promised Monograph on the Flore fossile du Portugal has been published.nbsp;A review of this valuable contribution to Mesozoic palseobotanynbsp;is given in the latter part of the present volume. Among thenbsp;very few remains of cycadean fronds described by Saporta, onenbsp;form of Cyeadites appears under the name of C. tenuiseetus^ Sap.,nbsp;and the figures of the frond fragments show a very distinct resemblance to the English specimens which I have referred to the newnbsp;species, C. Saportm. Possibly the Portuguese and British plantsnbsp;should be placed in one species, but for the present at least,nbsp;there are certain differences to be noticed which hardly justifynbsp;this adoption of Saporta�s specific name. In the English frondsnbsp;the pinnse are somewhat stouter, the tips more sharply acuminate,nbsp;and the general habit of the leaf appears to be rather stiller thannbsp;in C. tenuisectus.

Erond pinnate, linear, of uniform breadth; rachis broad and flattened, marked with obliquely placed lines, terminating proxi-mally in a broadened and swollen base. Pinnae of uniform breadth,nbsp;alternate or subopposite, attached to the upper surface of thenbsp;rachis, and inserted at right angles or slightly oblique tonbsp;the frond axis, the bases of the two rows of the pinnae almostnbsp;in contact; average length of the pinnae 6-7 cm., and 1-15 mm.nbsp;in breadth; bases slightly broadened and contiguous, apices ofnbsp;the long linear pinnae terminating in a sharp point; single mediannbsp;vein in each segment.

* Dunker (A. 2), 'Wealdenbildung, p. 16, pi. vii. fig. 1. ^ (1), p. 171, pi. xxxii. figs. 1-4 and 6.

for a specimen from the !Nortli German quot;Wealden beds, and thus dedned it¹ ²; �C. foliis pinnatis sublinearibus, pinnis numerosisnbsp;linearibus approximatis apice obtusiuscnlis medio costatis basi subnbsp;dilatatis.�

He speaks of tbe pinnffi as possessing a strong midrib, and his figure shows this character very clearly. Dunker has reflgurednbsp;Homer�s original specimen, and here again the pinnae appearnbsp;to have a distinct median vein; he points out that Mantell snbsp;(^yoadites Brongniarti ^ should be placed in the genus Nilssonia, asnbsp;it does not conform to the accepted definition of Cyclt;idites, Thisnbsp;Tilgate fossil� is now referred to as JDioonites Bfong'niciTti (Mant.).

In 1852 Ettingshausen ² obtained a portion of a cycadean frond from near Teschen, in Silesia, and referred it to Romer snbsp;species, but at the same time expressing the opinion that itnbsp;represented a form intermediate between C. Brongniarti, Rom.,nbsp;and C, Morrisiauus, Dunk. This is certainly not the same speciesnbsp;as Romer�s type, and should, as Schenk suggests, be placed innbsp;another species; he speaks of it as G. Seerii, Schenk, andnbsp;expresses the opinion that possibly C. Brongniarti, Rom., maynbsp;fie simply a partially developed frond of C. Morrisianus? Subsequently the same author includes both G. Brongniarti andnbsp;G. Morrisianus as synonyms of Bioomtes Dunkerianus (Gopp.).nbsp;He states that the type specimen of G. Morrisianus, Dunk., showsnbsp;1²0 indication of a midrib, and must therefore be referred to Ptero-or Bioonites instead of to Gycadites. There is the samenbsp;absence of a median vein, according to Schenk, in the segmentsnbsp;of G. Brongniarti, Rom., and this must, therefore, be also excludednbsp;from the genus Gycadites.^ It is not quite clear if Schenk is herenbsp;speaking of Romer�s original specimen; if he refers to the figurednbsp;specimen as it appears in the illustrations of Romer and Dunker,nbsp;the figures are certainly at variance with Schenk s description,nbsp;ere is the same apparent contradiction between figure and

� (A. 4), Ath. k.-k. geol. Eeichs. vol. i. Abth. iii. No. 2, 1852, p. 20, Pfi i- fig. 9.

diagnosis in Bioonites ahietinus' (Gopp.), as represented in pi. xxxvii. fig. 1 of Schenk�s monograph. Sohimper retainsnbsp;Eomer�s species, and unites with it PteropTiyllum Biinkerianum,nbsp;Giipp., as figured by Dunker,^ but in this case the figure showsnbsp;very clearly that the venation is not of the Cyoadiies type. Innbsp;the face of Schenk�s statements, we cannot, then, accept any ofnbsp;Bunker�s figures of what he describes as species of Cycaditesnbsp;as really examples of that genus; and, as Saporta � points out, thenbsp;only representatives of Cycadites so far known for beds of approximately Wealden age are C. Romeri, Schenk, and G. Meerii,nbsp;Schenk. In the English specimens referred to the new speciesnbsp;C. Saportm the preservation is fortunately good, and leaves nonbsp;doubt as to the existence of a true midrib in the pinnae.nbsp;C. Romeri agrees to some extent with this species, but differsnbsp;in its broader pinnae and their disposition on the frond axis. Thenbsp;specimens of C. Saportce are unusually large, and hence enablenbsp;us to obtain a good idea as to the general habit of the frond;nbsp;if it were not for this fact one might be inclined to include themnbsp;under Schenk�s species. The plant described by Braun from thenbsp;Jurassic sandstones of Steinstedt as C. rectangular isdiffers innbsp;its shorter and broader pinnae, and in the fact that they are morenbsp;distinctly at right angles to the rachis. Some of Saporta�s figuresnbsp;of what he regards as C. rectangularis, are much more likenbsp;C. Saporta than the type specimen figured by Braun; of. especially pi. xiii. figs. 1 and 3 of the Flore Jarassiqiie. Saportanbsp;includes C. pectinatus, Berg., as a synonym of Braun�s species,nbsp;and adopts the term rectangularis in preference to the older namenbsp;pectinatus, because of the use of the latter term by Bindley andnbsp;Hutton in connection with the genus Zamites; ^ perhaps hardlynbsp;a sufficiently sound argument to overrule the priority of Berger�snbsp;term.

' (A. 2), p. 234. Goppert, and not Miquel, appears to be author of the two specific names ahietinus and Bunkerianus; Schenk refers both these speciesnbsp;to Miquel.

evidently belongs to a plant of very similar habit to that of O'- Saportm; it differs mainly in the greater breadth of thenbsp;pinnae, so far at least as it is possible to judge from Berger�snbsp;figure.

Another species 'which bears a still closer resemblance to Saportm, is C. Rajmalialensis, Old., described by Oldham,nbsp;from Indiad The figures and description of this plant are innbsp;close agreement with the Eccleshourne Wealden species ; the seg-nients of the Indian frond seem to be rather more closely arrangednbsp;nud somewhat shorter than in the English form. It would,nbsp;however, be somewhat unwise to refer the lYealden specimens

Oldham�s species, considering the geological age of the two plants, and the less perfect preservation of CycadiUs Rajmahalensis.nbsp;-fn this, as in many other cases of fossil plants, we have to speaknbsp;cautiously as to the relationship of individual members of differentnbsp;fioras, and must trust rather to the comparative study of the florasnbsp;S'S a whole, than to the apparent identity of isolated elements.

Trautsohold�s specimen of Cycadites acinaciformu, Traut.,^ is similar to the present species, but probably not identical with it.nbsp;Schenk suggests that the Russian species is probably identical withnbsp;P'lcopteris decipiens, Traut., and must be placed with the ferns,nbsp;ft is difficult to speak with much confidence as to the naturenbsp;cf the specimen figured as C. acinaoiformts, but it certainly bearsnbsp;^ strong likeness to the oycadean genus.�

This exceptionally fine specimen shows one frond 60 cm. in icD-gth, and a second 38 cm. long, the latter being inclined tonbsp;the former in such a way as to suggest but little displacementnbsp;from their original position of growth on the parent stem. Thenbsp;larger frond, as represented on a small scale in the photograph,nbsp;shows a striking uniformity in the length and breadth of thenbsp;numerous closely set pinnse. lu several of the segments thenbsp;sharp apices are clearly preserved. Judging by other specimens,nbsp;lu which the pinn� are longer and more nearly at right angles

�� Trautschold (A. 3), Nouv. M�m. Soc. Rat. Moscou, vol. xiii. 1876, p. 34, Ph xxi. fig. 1.

to the rachis, it is probable that this large example may represent a frond not quite fully developed. The stout rachis,nbsp;about T4 cm. in breadth, and especially that of the smallernbsp;frond, shows numerous obliquely running longitudinal lines.nbsp;The contiguous and slightly broadened bases of the pinnae arenbsp;very distinct on portions of the larger frond. In some placesnbsp;the segments, adhering together by their contiguous bases, havenbsp;been torn en masse from the axis of the leaf.

The stout and distinct median vein is well marked throughout. Near Hastings.nbsp;nbsp;nbsp;nbsp;Mujford Coll.

Frond 13-5 cm. in length; rachis 1 cm. broad. The long and contiguous pinnae are attached at right angles to one edge ofnbsp;the flattened axes. Midrib distinctly preserved, as in Fig. 5a;nbsp;also the sharply acuminate tips of the segments. Longest pinnanbsp;11 cm. The generalnbsp;nbsp;nbsp;nbsp;appearance ofnbsp;nbsp;nbsp;nbsp;this specimen is indicative

of an older frond than nbsp;nbsp;nbsp;V. 2777.nbsp;nbsp;nbsp;nbsp;Ecclesbourne.nbsp;nbsp;nbsp;nbsp;Rujford Coll.

24 cm. long. At the two extreme ends only one row of pinnae has been preserved, and the impression of the broad rachisnbsp;is shown on the surface of the rook. In other places the twonbsp;rows of pinn� are almost in contact, as in the portion represented in PL IV. Fig. 5, andnbsp;nbsp;nbsp;nbsp;therenbsp;nbsp;nbsp;nbsp;is very littlenbsp;nbsp;nbsp;nbsp;of the rachis

visible between the nbsp;nbsp;nbsp;bases ofnbsp;nbsp;nbsp;nbsp;thenbsp;nbsp;nbsp;nbsp;two sets ofnbsp;nbsp;nbsp;nbsp;pinn�. The

arrangement and general appearance of the segments bear a marked resemblance to Dioonites Bankerianus (G�pp.). Eocles-bourne.nbsp;nbsp;nbsp;nbsp;JRufford Coll.

V. 1069. A more terminal portion of a frond. Finn� smaller and more obliquely inclined to the rachis than in most of thenbsp;other specimens. Cf. the terminal portion with the specimennbsp;of Bioonites Bunkerianus (V. 2823) figured in PL II. Fig. 3.nbsp;Ecclesbourne.nbsp;nbsp;nbsp;nbsp;Presented ly P. Rufford, Rsq., 1885.

V. 2124A Several portions of fronds. In one there appears to be the broad and thick basal termination of the petiole fairly

clearly preserved. The midrib and pointed apex well seen in several of the pinnse. Ecolesbourne.nbsp;nbsp;nbsp;nbsp;Itujford Coll.

V. 2924. 25 cm. long. A single row of pinnae attached to one margin of the broad flat raohis; midrib distinct. Ecoles-lgt;ourne.nbsp;nbsp;nbsp;nbsp;Ruffori Coll.

In dealing with such fronds as those figured by Dunker as Gycadites Morrisianus, Dunk., and C. Brongniarti, E�m., andnbsp;afterwards described by Schenk as species of Pterophyllwn ornbsp;^ioonites, we have to face the difficulty of deciding upon the mostnbsp;suitable generic term. The fronds in the Eufford Collection shownbsp;�''quot;cll-marked characters, and leave no doubt as to the form andnbsp;Planner of attachment of the pinnse; we have long, narrow, linearnbsp;O'Ud parallel-veined segments, with acutely pointed tips attachednbsp;ly broad and non-auriculate bases to the upper surface of thenbsp;rachis. Must these be included in Pterophyllum or Bioonites, ornbsp;do the generally received definitions of these genera not admitnbsp;cf the application of either name to the Wealden fronds ? Letnbsp;briefly summarize some of the various definitions of thesenbsp;genera, and note how far they coincide with the characteristicnbsp;features of the present series of specimens. Pterophyllum wasnbsp;defined by Brongniart in 1828 * as a genus characterized by�nbsp;� Eeuilles pinn�es, � pinnules d�une largeur ^ peu pres egale,nbsp;s�ins�rant sur Ie p�tiole par toute la largeur de leur base, tronqu�esnbsp;an sommet; nervures fines, �gales, simple, peu marquees, toutesnbsp;parall�les.� He speaks of the truncate apices of the pinnae asnbsp;nn essential character, but does not insist on a lateral or surfacenbsp;insertion on the frond axis. Pterophyllum Jaegeri, Brong., is spokennbsp;of as one of the species of this genus, and in this instance thenbsp;attachment of the segments is apparently lateral. In the Tableau ^

Brongniart points out that subsequent writers have applied his generic name to plants which do not conform to the originalnbsp;definition. He considers the essential characters to be (i.) a slightnbsp;union of the bases of the pinnae ; (ii.) the quadrilateral, oblong,nbsp;or linear form of the segments ; (iii.) truncately terminated,nbsp;segments; and (iv.) the presence of fine parallel veins not convergentnbsp;at the apex. Morris * speaks of Pterophyllum as including plantsnbsp;with pinnate fronds and suhlinear pinnae, inserted by the wholenbsp;base, with the apices truncate or sometimes acute, etc. MiquePnbsp;keeps closely to Brongniart�s original definition. G�ppert* adoptsnbsp;a wider definition, and includes in this genus plants with obtuselynbsp;and acutely terminated pinnae, etc. Bornemann,* in 1856, definednbsp;the genus as follows:nbsp;nbsp;nbsp;nbsp;� Brond pinnate or deeply pinnatisect,

pinnae approximate, and with the whole base attached to the racliis, short, broad, quadrate, or elongate, straight at the tipnbsp;or obliquely truncate, horizontal or oblique to the rachis; veinsnbsp;parallel.�

Leckenby � assigns the name Pterophyllum to the species P. medianum, Leek., with its Nilssonia-li^Q lamina, which isnbsp;apparently not attached to the side of the frond axis. Schenk,nbsp;in his Fossile Flora der GrenzscMohten � . . . , adopts a very comprehensive definition, and defines the pinnae of Pterophyllum asnbsp;distichous, elongate, or adherent, narrow or broad, apex acute ornbsp;truncate; but he says nothing as to the manner of attachmentnbsp;to the rachis. He includes P. incrustans, G�pp., and P. Braunii,nbsp;G�pp., in the same genus. Heer� prefers the genus Zamites, usednbsp;in an unusually wide sense, for such fronds as his Z. horealis,nbsp;Heer, Z. aoutipennis, Heer, etc., which resemble in general formnbsp;the leaves of the Wealden species originally described by Hunkernbsp;as Cycadites Morrisianus. Schimper separates the fronds withnbsp;irregularly pinnatifid leaves from the true Pterophyllum type, andnbsp;institutes for their reception the genus Anomozamites.^ In the

� (A. 1), FI. foss. Grenz. Keup. Lias, p. 163. � FI. foss. Arct. vol. iii. p. 66, pi. xv.

genus Pteropliyllum he includes fronds with pinnae vertically attached to the side of the rachis, and having truncate apices.nbsp;His genus Ctenophyllum,^ which includes certain forms oftennbsp;referred to Pteropliyllum, is defined as follows: �Folia linealia,nbsp;gracilia; foliolis lateri rachis superiori oblique adfixis, ssepiusnbsp;eppositis, linealibus, obtusis, basi retro folium infrapositum de-fleuntibus, coriaceis, tenuiter et parallele nervosis.�

Saporta departs somewhat from the definitions given by other authors,¹ ² and restricts Pteropliyllum to fronds with pinnse attachednbsp;to the side of the rachis, and which are distinct one from another,nbsp;Dot fused laterally at the base, and having truncate apices.nbsp;Peistmantel,² on the other hand, in speaking of Zamites proximm,nbsp;^'eist., points to the separate pinnae, which are not connectednbsp;the base, as a feature inconsistent with the inclusion of thenbsp;plant in the genus Pteropliyllum. Nathorst ^ has describednbsp;Certain plants from Bjuf as possibly species of Pteropliyllum;nbsp;^ut to express the absence of perfectly satisfactory evidence, henbsp;prefixes a query to the generic name. More recently, thisnbsp;8'Uthor has called attention to the lateral insertion of the pinna3nbsp;an essential character of Pteropliyllum, and a convenientnbsp;distinguishing feature from Nilssonia.^ In Zittel's Sandhueh,^nbsp;the lateral attachment of the pinnae, which may or may not benbsp;distinct at the base, and their rounded or truncate apices are givennbsp;important generic marks. It is suggested that possibly suchnbsp;C� frond as Pteropliyllum Punkerianum, G�pp., ought not to benbsp;included in the genus Pteropliyllum, because of the insertion ofnbsp;the segments on the upper surface of the frond axis. Solms-Laubach² refers to P. Jaegeri, Brong., as an example of onenbsp;form of Pteropliyllum frond, and in another place� calls attentionnbsp;to the Nilssonia-Yike! form of some species of the same genus,nbsp;'Which agree with other examples of Pteropliyllum in the lateralnbsp;insertion of the leaf lamina.

nbsp;nbsp;nbsp;(A. 3), Denkschr. k. Ak. 'Wiss. math.-nat. Cl. vol. Ivii. p. 6.nbsp;' (A.), p. 224.

On the whole, then, the characters generally insisted on seem to be the lateral attachment of the pinnse to the raehis, and bynbsp;many, but by no means all authors, the truncately terminatednbsp;segments. The confusion which has arisen from constant alterations by various writers, and from the not uncommon practicenbsp;of including certain fronds in a particular genus, in spite ofnbsp;obvious discrepancies between the specimens and the genericnbsp;diagnosis, is sufficiently obvious if we glance at some of thenbsp;better known Pterophyllum species as figured by different authors.nbsp;We have such forms as P. incondans, Gopp., P. Punlcerianum,nbsp;Gopp., P. Jaegeri, Brong., P. Braunii, G�pp., etc., included in thenbsp;same genus. It is true that in examining fossil fronds we arenbsp;often unable to decide as to the actual manner of attachment ofnbsp;the pinnae, and are thus driven to leave the specimen as doubtful,nbsp;or to decide as best we may in the face of difficulties inseparablenbsp;from the determination of isolated leaf fragments. We cannotnbsp;always be sure whether we have the frond preserved with itsnbsp;lower or upper side uppermost. It is, however, clear that wenbsp;cannot consistently make use of Brongniart�s genus for suchnbsp;specimens as those before us.

The genus Bioonites of Miquel has been adopted by some authors for these narrowly segmented Wealden fronds. Thisnbsp;again is a generic name which has been made to do duty fornbsp;forms of leaves, which it is difficult to regard as correctly includednbsp;in the same genus, even if the genus be admittedly a provisionalnbsp;and artificial one. Miquel is responsible for the proposal of thisnbsp;name, and for the following definition: ^ � Frondes pinnat�,nbsp;rigid�, crass�. Foliola densa patentissima supreme nunc sub-imbricata, lanoeolata, vel lineari-laneeolata, recta vel subfalcata,nbsp;acuta vel aoutiuscula, basi tota latitudine inserta, infernenbsp;retrorsum subdecurrentia, nervis cum margine parallelis �qua-libus subtus distinctioribus (cum sulculis stomatiferis alter-nantibus).� He included under this name several speciesnbsp;previously described as examples of Pterophyllum and othernbsp;genera. Bornemann adopts Miquel�s genus and extends its usenbsp;to some additional species, but does not make any importantnbsp;alteration in the original diagnosis. Schimpen retains the term

Dioonites, and gives th.e essential characters as follows:' �F olia pinnata, pinnis pro more angustis, lanceolatis, aoutis, ohliquis,nbsp;Iota latitudine insertis, basique leniter pro- et decurrentibus,nbsp;Hervis parallelis.� In Zittel�s Handhuoh'^ the genus is quoted,nbsp;and Pterophyllum Buchianum, Ett., and P. Brongniarti, Schenk,nbsp;are given as two typical species. The former of these has sincenbsp;been, transferred by Nathorst to a new genus, Zamiophyllum,^ onnbsp;the ground that the pinnae are slightly narrowed towards thenbsp;base. Saporta repeats the character of Bioonites ¹ ² as given bynbsp;previous writers, and figures B. Brongniarti as a typical example;nbsp;hut the species referred to by this writer at the end of hisnbsp;definition as the typical form of the genus is I). Kurrii, Schimp.nbsp;Ih Fontaine�s Potomac Flora we find numerous forms includednbsp;Under Miquel�s genus, hut it must be noted that this author,nbsp;U'hile giving what he refers to as Schimper�s definition of thenbsp;genus, speaks of the pinnae as �sometimes expanded at base sonbsp;Hs to extend up and down the rachis.� ^ This is an importantnbsp;alteration, as Schimper describes the pinnae as distinctlynbsp;decurrent, and it is this characteristic which is repeated by thenbsp;Hiajority of writers as one of the essential generic features. Innbsp;his definition of Bioonites Buohianus (Ett.) Fontaine refers tonbsp;Ihe pinnae as slightly narrowed at the base, hut does not regardnbsp;this character as opposed to the adoption of Miquel�s genus. Thenbsp;attachment of the pinnae by the whole of a more or lessnbsp;decurrent base appears to be the chief characteristic generallynbsp;Insisted on. In several definitions of Bioonites no mention isnbsp;Hiade of the place of attachment of the frond segments, whethernbsp;uu the surface or sides of the rachis; in several of the figurednbsp;specimens referred to this genus the pinnae are inserted laterally.nbsp;Some authors have emphasized the fact that the segments mustnbsp;te attached to the upper surface of the rachis, as in B. Brongniarti.nbsp;This position of the pinnae affords one point of difference fromnbsp;Pterophyllum, and in the decnrrent and separate leaves we havenbsp;other features characteristic of Bioonites. Nathorst, in discussing

the generic characters of the plant, to which he applies the name Zamiophylliim Buchianum (Ett.), refers to Miquel�s genus Bioonitesnbsp;as characterized by the attachment of the pinnae to the uppernbsp;surface of the rachis, and by the insertion of the segments almostnbsp;at right angles to the axis; he says nothing as to the decurrentnbsp;bases of the pinnae. If we accept this definition, and departnbsp;from the usually accepted feature of a decurrent pinna base, wenbsp;may well include the quot;Wealden plants under this genus. It isnbsp;certainly not an easy matter to draw a definite line betweennbsp;pinnae attached to the rachis hy the entire base, which is notnbsp;decurrent, and those which are similarly attached, hut with theirnbsp;bases more or less decurrent. In the English examples of thenbsp;species B. Bunherianus (G�pp), the pinnae towards the uppernbsp;end of the frond are distinctly decurrent, hut those occupyingnbsp;a lower position cannot be described as possessing decurrent bases.nbsp;Cf. PL II. Fig. 3, and PI. III. Fig. 6. There are two othernbsp;genera to which reference should be made, which to a certainnbsp;extent agree in their definitions with such fronds as B. Bunherianus, etc., viz. Ctenofhyllum and Ptilophyllum. The formernbsp;genus was instituted by Sohimper ¹ ² to include certain forms ofnbsp;fronds which do not in all essentials comply with the definitionsnbsp;of Otozamites on the one hand, and Bioonites on the other,nbsp;Pterophyllum, pecten, L. and H., being taken as the type species.nbsp;The author of the genus afterwards somewhat modified his originalnbsp;diagnosis, and pointed out that Pterophyllum Braunianum, G�pp.,nbsp;had been erroneously described as a species of Ctenophy�lum. Thenbsp;genus Ptilophyllum, proposed by Morris in ISdO'� for certain Indiannbsp;fronds, can with difficulty be distinguished from Ctenophy�lum.nbsp;It is thus defined : �

�Fronds pinnate; pinnae linear, closely approximated, more or less elongate; base variable in form, oblique, round, imbricate,nbsp;sometimes auricled in the upper and sometimes in the lower part.nbsp;Veins slender, equal, parallel.� �

Goppert long ago expressed the opinion that Morris� term was a needless addition to the list of cycadean genera.² It

tas been found useful by several writers as a convenient name to apply to Indian fronds, but as at present used it does notnbsp;appear to be wholly satisfactory. The genus Ptilophyllum seemsnbsp;to have been almost confined to Asiatic fronds, and the localitynbsp;at a specimen has probably had too great a share in thenbsp;selection of Ptilophyllum in preference to Ctenophyllum as thenbsp;^�iost suitable name. Nathorst* figures and describes a leafnbsp;tragment from Japan as Ptilophyllum cf. cutchense, Morr., butnbsp;it would seem practically impossible to separate such a form asnbsp;this from some English Jurassic fronds usually placed in thenbsp;genus Ctenophyllum.

On the whole perhaps the better course is to retain, at least tor the present, the name Pioonites as the most suitable genericnbsp;designation for the Wealden species B. Bunkerianus (Gopp.).nbsp;^Ve must slightly modify the definition of the genus, and nonbsp;longer insist on the decurrent pinna base as an essential characteristic. The implied relationship to the recent Bioon is thenbsp;least satisfactory feature of Bioonites, but possibly we shall benbsp;8ble, on a future occasion, to suggest some further alteration innbsp;the existing nomenclature of fossil cycadean fronds. We maynbsp;define this genus, using the term Bioonites in a wide andnbsp;provisional sense, as follows :�

Prond pinnate, pinnae at right angles, or more or less obliquely inclined to the raohis, attached to the upper surface of the frondnbsp;axis, bases separate, may or may not be decurrent, not narrowednbsp;towards the point of attachment, apices acuminate, straight ornbsp;slightly truncate, veins parallel.

In dealing with Bioonites, as with many other genera, we �ay easily fall into the error of excluding or including certainnbsp;forms owing to our imperfect knowledge as to the manner ofnbsp;attachment of the pinnae; but it is obviously impossible tonbsp;devise a perfectly satisfactory system, so long as we are limitednbsp;I�y the exigencies of fossilization and the imperfection of thenbsp;frond fragments.

Dioonites Dunkerianus (G�ppert).

1846. Pterophyllum JDunkerianum, Dunker, AYealdenbildung, p. 14, pl. ii. fig. 3, pl. vi. fig. 4.

1848. nbsp;nbsp;nbsp;Pterophyllum Punkerianum., Broun, Index pal. nomencl. p. 1055.nbsp;Cycadites Morrisianus, Bronn, ibid. p. 371.

1850. nbsp;nbsp;nbsp;Pterophyllum Punkerianum^ Unger, Gen. spec, plant. foss. p. 290.nbsp;Cycadites Morrisianus^ Unger, ibid. p. 280.

1851. nbsp;nbsp;nbsp;Pioonites Punkerianus, Miquel, Rangschik, foss. Cycad. p. 212.

1852. nbsp;nbsp;nbsp;Pterophyllum Punkerianum., Ettingshausen, Abh. k.-k. geol. Reichs.

1871. Pioonites Punkerianus, Schenk, Palseontographica, vol. xix. p. 232, pl. xxxvi. figs. 1-5.

G�ppert¹ thus defines the species: �Pt. fronde pinnata, pinnis crassiusculis alternis lineari - acicularibus elongatis pectinato-patentissimis subremotis aeque distantibus 4-5 nervis subacutis.�nbsp;Dunker originally named this plant Nihsonia peeten, but annbsp;inspection of drawings received from Dunker led G�ppert to dissentnbsp;from the original designation. Dunker, in his Wealienhildung,nbsp;makes one or two slight alterations in G�ppert�s diagnosis; henbsp;speaks of the venation as �nervis 3-4 instinctis,� and adds �rhachinbsp;crassa oompressa.� ^ The specimen represented in Bunker�s pl. v.

S- 3 shows the pinn� apparently attached rather to the side than to the middle of the upper surface of the rachis ; the bases arenbsp;slightly swollen, and the apices pointed. In pi. vi. fig. 4 of thenbsp;some author part of the broad rachis is shown, and the approximatenbsp;ond narrowly linear pinnae are inserted at right angles to the frondnbsp;oxis. The specimen figured by Dunker and named by G�ppertnbsp;^^^Tophyllum abietinum, bears such a strong resemblance to Dioonitesnbsp;I^^nherianm as seen from the under side of the frond, that I havenbsp;''ventured to insert this species as a possible synonym. Schenk hasnbsp;previously called attention to this resemblance, and suggests thatnbsp;possibly the similarity may amount to specific identity ; Schenk�snbsp;Hfirire shows a distinct midrib in the pinnae, but this must be annbsp;error in. tJje sketch or some deceptive appearance in the fossil.nbsp;Ilie specimen referred to Dioonites ahietinus, by Hosius andnbsp;^on Marck, is probably a fragment of Zamites Buehianus (Ett.).�nbsp;Schenk adopts Miquel�s generic term Dioonites, which the latternbsp;author proposed in 1851 for this and other species of Pterophyllum.nbsp;lu speaking of the genus Cycadites, reference was made to Schenk�snbsp;substitution of Dioonites or Pterophyllwm for Cycadites, in the casenbsp;'�I certain specimens previously assigned by Bunker to the latternbsp;genus; an examination of the type specimens having convincednbsp;euenk of the absence of a single median vein in the leaf segments,nbsp;und therefore of the erroneous adoption of the same Cycadites.nbsp;Schenk�s figure 1, pi. xxxvi. ^ shows a portion of one side ofnbsp;u frond with closely placed long and narrow pinnie, which in theirnbsp;nianner of attachment suggest a spirally twisted frond axis, suchnbsp;We have in the recent cycad Macrommia spiralis, Miq.; but thisnbsp;�uy Well be an accident of fossilization. Fig. 5 of Schenk showsnbsp;ue same kind of rachis as in Bunker�s fig. 4, pi. vi. The figuresnbsp;the epidermal cells given by this writer show a distinctlynbsp;�undulating outline in the walls, and the presence of numerousnbsp;�tomata.

_ It does not seem quite clear whether Schenk has correctly �ucluded C. Brongniarti, E�m., as a synonym of the presentnbsp;species; � he speaks of E�mer�s species as probably the upper

' (A. 1), PalsBontographioa, vol. xxvi. p. 213, pi. xliv. fig. 199. * (A.. 2), Palseontographica, vol. xix.

portion of a frond of Bioonites Bunkerianm. The evidence of the figures of E�mer and Danker does not, however, sufScientlynbsp;support this view to justify our following Schenk�s examplenbsp;without having examined the type specimen. In the case ofnbsp;Cyoadites Morrisianus, Dunk., there can be little doubt that thenbsp;specimens referred by Dunker to Cycadites must be transferrednbsp;to the genus Bioonites.

It should be pointed out that Ettingshausen had previously suggested the specific identity of Cycadites Morrisianus, Dunk.,nbsp;and Pterophyllum Bunkerianum, Gopp. ; he considered it possiblenbsp;that P. Oopperhanum ought to be included with these two species.'

The specimen figured by Ettingshausen as Cycadites Brongniarti, Eom.,^ has since been placed by Schenk in a new species�nbsp;Cycadites Beerii.^

Among the Ecclesbourne specimens there are several good examples which must be included in G�ppert�s species. At firstnbsp;sight many of them would be referred to Cycadites, and the generalnbsp;habit of the frond shows a striking resemblance to that of Cycaditesnbsp;Saportm, sp. nov., but a closer examination demonstrates that nonbsp;true midrib can be detected, and that the ridge in some of thenbsp;pinnae which closely simulates such a central vein, is merelynbsp;the strongly marked convexity of the upper surface of thenbsp;leaf segments. Among recent cycads the genus Pncephalartosnbsp;affords examples of fronds in which the general habit is strikinglynbsp;similar to that of Bioonites Bunkerianus: E. Ohellinckii, Lem., asnbsp;shown in PI. XIII. Pigs. S-5, possesses pinnae of about the samenbsp;size, and with a very similar mode of attachment, at least asnbsp;regards their almost horizontal position, hut in the lateral insertionnbsp;to the rachis the segments of the recent species differ from thosenbsp;of the fossil frond. In E. Ghellinckii the convex upper surfacenbsp;of the pinn� presents a very similar appearance to that in thenbsp;quot;Wealden frond segments, and the sharply acuminate tips of thenbsp;pinnae is practically identical in the two cases. In PI. XIII.nbsp;Eig. 3 a portion of a frond of this species is represented, naturalnbsp;size; in Eig. 4 the median groove on the under side of a single

pinna is clearly shown; and in Fig. 5 a section of a pinna illustrates the strongly revolute form of the margins. Lemaire�snbsp;figure of this species is very poor, and gives an imperfect ideanbsp;of the habit of the leaf.^

Another species of this recent genus, E. cycaiifoUus, Lehm. (Pi. XIII. Fig. 6), also illustrates a point of contact betweennbsp;oxisting and extinct fronds ; it differs from E. Ohellinchii in itsnbsp;somewhat broader pinnae. It would, however, tend to a misconception of the true nature of the Wealden fronds, if thenbsp;generic term Encephalartos were adopted on the strength of thenbsp;striking similarity as regards the character of the fronds; wenbsp;Unfortunately know nothing as to the flowers and stems ofnbsp;Eioonites Dunkerianus. We may adopt a slightly emended formnbsp;of Schenk�s definition for the present species :�

Frond pinnate, rachis strong, pinnae approximate, thick, linear, entire, alternate or suhopposite; 2-3 mm. broad at the widestnbsp;part, with a length of 11 cm. or more, gradually hut slightlynbsp;narrowed towards the distal ends; the two rows of pinnsenbsp;8-ttached close together to the upper surface of the rachis ; thenbsp;lower margin of the basal end of the pinnse either slightlynbsp;decurrent, especially towards the tip of the frond, or somewhatnbsp;firoadened and bluntly rounded; towards the apex of the frondnbsp;fhe segments are obliquely inclined, and in the lower portionnbsp;almost at right angles, to the axis. Veins usually indistinct,nbsp;parallel equal veins in each pinna.

V. 3218. PI. III. Fig. 6.

23 cm. in length. The upper surface of the pinnse strongly convex; pinnse slightly and gradually tapered towards a pointednbsp;apex. The arrangement of the segments and their somewhatnbsp;broadened bases are very similar to those in Cycadites Saporta.nbsp;Schenk speaks of the pinnse as 4-4j cm. long, hut his figurenbsp;represents some with a length of 8 or 9 cm. In the presentnbsp;specimen the longest pinna has a length of 11 cm., and this doesnbsp;uot include the actual apex. In nearly all the segments it isnbsp;impossible to make out the venation, but in one or two casesnbsp;file parallel veins are visible. Eoclesbourne.nbsp;nbsp;nbsp;nbsp;Rujford Coll.

Cf. V. 1069. Cycadites Sa/portcR. Small specimen, evidently close to the frond apex. The pinnse are much more obliquenbsp;and decurrent than in the previous example. Eccleshourne.

Imperfectly preserved piece of rachis with portions of pinnae on one side; some of the pinnae have well-marked venation andnbsp;acute tips. The figure shows some of the more perfect apices.nbsp;The pinnae have a breadth of about 2 mm., and each is traversednbsp;by numerous veins, in some of the segments as many as ten maynbsp;be counted. A comparison of this specimen with V. 3218 (PI. II.nbsp;Pig. 6) shows some fairly striking differences, and it is notnbsp;improbable that we have to deal with two specific forms; innbsp;V. 2821 the veins are more numerous, and the pinnae are shorternbsp;and proportionately broader than in the other examples referrednbsp;to this species. Eccleshourne.nbsp;nbsp;nbsp;nbsp;Rufford Coll.

V. 2124r. Broad and flat rachis very like that of Cycadites Saportm. On one side the long and narrow pinnee are fairly wellnbsp;preserved, showing occasional signs of venation and a strongnbsp;convex upper surface. Eccleshourne.nbsp;nbsp;nbsp;nbsp;Rujford Coll.

y. 2127. Probably a portion of a frond near the apex. The broad bases of the pinnae and their manner of attachment to thenbsp;rachis clearly seen. Eccleshourne.nbsp;nbsp;nbsp;nbsp;Rvjford Coll.

V. 2361. Probably a fragment of this species; broad pinnse. Eccleshourne.nbsp;nbsp;nbsp;nbsp;Rufford Coll.

V. 2822. Here the pinnse are more oblique to the axis of the frond, and the lower edges of the bases more deourrent, as in thenbsp;terminal fragment shown in PI. II. Fig. 3 (V. 2823). Ecoles-bourne.nbsp;nbsp;nbsp;nbsp;Rujford Coll.

V. 2824. Similar terminal portion to preceding specimen. Focleshourne.nbsp;nbsp;nbsp;nbsp;Rufford Coll.

V. 2916. Two specimens. Raohis 16 cm. long, apparently twisted, showing in the lower portion two alternate or sub-opposite pinnae attached to its upper surface; in the upper partnbsp;the segments are separated by 6 mm. of rachis, suggesting a viewnbsp;of the under side of the frond. Pinnae slightly convex, presentingnbsp;the appearance of a broad midrib. Ecoleshourne. Rujjford Coll.

Dioonites Brongniarti (Mant.).

1833.

1841.

1842. 1844.

1848.

1849.

1850.

1851.

1852.

1854.

1856.

1871.

1874.

1875. 1881.

1889.

Nilssonia Brongniarti, Bronn and Eomer, Leth. geog. Tol. ii. p. 61, pi. xxT�i. fig. 14.

Nilssonia Brongniarti, Ettingshansen, Abh. k.-k. geol. Eeichs. vol. i. Abth. iii. No. 2, p. 23.

Dioonites Brongniarti, Schenk, Palaeontographica, vol. xix. p. 236, pi. xxxii. fig. 2.

Dioonites Brongniarti, Eenault, Cours. bot. foss. vol. i. p. 51, pi. iv. figs. 13 and 14.

Dioonites Kotoei, Yokoyama, Journ. Coll. Sci. Japan, vol. iii. p. 44, pi. vii. fig. 1, and pi. xiv. fig. 14.

In 1833 Mantell described a badly preserved frond fragment from the Tilgate beds of Sussex, which he named Oycadites Brong-niarti, using the term Cycaditesf^aXYiex as a general designationnbsp;indicative of oycadean affinity, and not in accordance with thenbsp;narrow sense in which Brongniart defined the genus. Morris substituted Plerophyllwm for Cycadites, and Dunker called attentionnbsp;to the plant figured by E�mer as C. Brongniarti, which should notnbsp;be confounded with Mantell�s type described under the same name ;nbsp;the latter he suggested should be referred to Nihsonia. Goppertnbsp;adopts Nihsonia as the generic term, and Miquel, Ettingshausen,nbsp;and others follow his example. Schenk, on the other hand,nbsp;points out certain discrepancies between the characters of Mantell�snbsp;species, as further illustrated by subsequently described examplesnbsp;of the same type from the quot;Wealden of North Germany, and thenbsp;genus Nihsonia; he substitutes Miquel�s term Bioonites fornbsp;Bunker�s Nihsonia.'^ Schenk�s specimen is in a better state ofnbsp;preservation than the English example, and shows more clearlynbsp;the manner of attachment of the pinnae. This species differs fromnbsp;Otozamites Ooppertianm (Dunk.), in the absence of an auriculatenbsp;base to the segments, in its coarser veins, and in the segmentsnbsp;being more nearly at right angles to the axis of the frond. quot;Wenbsp;may adopt Schenk�s definition: �Eolia pinnata, segmenta e basinbsp;latiore apicem versus attenuata acuminata lineari-lanceolata Integranbsp;approximata alterna vel opposita, in petioli latere antico sessilia,nbsp;3 mm. usque ad 2'5 cm. longa, 3'5-5 mm. lata, superiora breviora,nbsp;summa brevissima ovata, superiora oblique patentia, media paten-tissima, nervi tenues quinque vel sex tenues sequales parallel!.�

The plant figured by Leokenby ^ as Pterophyllum angustifolium, Leek., from the Oolite of Gristhorpe, shows a marked similaritynbsp;in general appearance to Bioonites Brongniarti.

Yokoyama�s Japanese species, B. Kotoei,^ may probably be included as a synonym of Mantell�s plant; the former authornbsp;mentions the greater number of veins in the pinnae of his plantnbsp;as a distinguishing feature from B. Brongniarti, and speaks of the

latter form as taving 5-6 veins in each leaf segment, whereas in -�. Kotoei there are 7-14. In Schenk�s figure of Mantell�s speciesnbsp;tfiere are eight or nine veins shown, and in the solitary specimennbsp;1*1 the Eufford Collection there appear to be at least eight veins.nbsp;The greater length of the segments is another point referred to bynbsp;Yokoyama as a specific character of his plant; but it is difficultnbsp;on comparing the published figures of the two species to detect anynbsp;distinct difference in this respect. On the whole, I am unable tonbsp;discover any sufficient difference between the two plants to warrantnbsp;the retention of Yokoyama�s specific name. Ptilophyllum oligo-�gt;^�urum, Ten.-Woods,! also agrees closely with the English species.nbsp;In speaking of Pterophyllum Richtlwfeni, Schenk, from China,nbsp;Schenk ^ suggests that probably some of the fragments so namednbsp;may be identical with Bioonites Brongniarti\ there is certainlynbsp;n close correspondence between the two forms, but perhapsnbsp;hardly a sufficiently strong resemblance to justify the inclusionnbsp;the Chinese specimens in the synonoray of Mantell�s species.nbsp;Ihe plant figured by Schenk� as Pterophyllum mquale, Brong., fromnbsp;I^ersia, resembles Bioonites Brongniarti. In Schenk�s specimennbsp;Ihe pinnse appear to be inserted on the upper surface of thenbsp;rachis, and not laterally as the generic term Pterophyllum implies.

V. 2748. 26 cm. long. The alternately disposed pinnae are attached to the middle of the upper face of the frond axis.nbsp;Venation clearly marked. The lower margin of the pinnse isnbsp;**tirved gradually upwards, cutting off the veins obliquely, andnbsp;the upper margin is practically horizontal. The segments arenbsp;somewhat less than those in Mantell�s figure, but there can benbsp;little doubt as to the specific identity of the specimens. Eccles-hourne.nbsp;nbsp;nbsp;nbsp;Buford Coll.

^ hichthofen (A.), China, vol. iv. p. 247, pi. xlvii. fig. 7, and pi. xlviii. 5, 6, and 8.

In 1820 Mlsson¹ ² described and figured certain plant remains from H�r, a small village north of Lund in Scania, and regardednbsp;them as probably fern fronds, but he made no attempt to definenbsp;them specifically. In 1825 Brongniart refigured and describednbsp;some of Nilsson�s specimens under the following specific names ;nbsp;Nilssonia elongata, N. hrevis, N. {?) mqualis, Pterophyllum majus,nbsp;and P. minus. It was on one of Nilsson�s specimens thatnbsp;Brongniart founded his genus Nilssonia, and also Pterophyllum ;nbsp;the two H�r species referred by Brongniart to this latter genusnbsp;have since been transferred by Nathorst^ to Schimper�s genusnbsp;Anomozamites. We have the first complete diagnosis of Nilssonianbsp;in the Prodrome,^ where it is thus defined : � Feuilles pinnees ;nbsp;pinnules rapproch�es, oblongues, plus ou moins along�es,nbsp;arrondies au sommet, adherentes au rachis par toute la largeurnbsp;de leur base, �. nervures parall�les dont quelques-unes sentnbsp;beaucoup plus marqu�es.� In his later work,^ Brongniart retainsnbsp;this name and speaks of Nilssonia as closely allied to Pterophyllum.nbsp;Miquel ^ substituted a new generic term, Hisingera, for some ofnbsp;the species of Nilssonia, and as an example of the new genus henbsp;cites Cycadites Brongniarti, Mant. G�ppert² accepts Brongniart�snbsp;genus in its wide sense, and does not suggest the institution ofnbsp;any sub-genera. In 1856 Bornemann adopted the followingnbsp;definition of Nilssonia:'' �Frondes coriaoeae, pinnatm, vernationenbsp;circinatse, foliola contigua continue tota latitudine inserta, patentia,nbsp;abbreviata, basi passim cohserentia, apice obtusa vel truncata,nbsp;nervis parallelis arcuatis apice confluentibus nonnullis validi-oribus.� This writer points out the difficulty of recognizing the

different kinds of veins in the leaf divisions. Schenk, in Flora der Grenz^chichten' includes Nihsonia among the ferns,nbsp;^�d refers to certain specimens in which the leaves shownbsp;Numerous round projecting structures between the veins, andnbsp;^hich he regards as sporangia or sori; and it is on the strengthnbsp;these appearances, suggestive of fern fructification, that thenbsp;S�nus is excluded from the Cycadaceat. Schenk speaks of thenbsp;^eins as equal and simple, and refers to the epidermal cells asnbsp;.nbsp;nbsp;nbsp;nbsp;the straight walls characteristic of cycads. No great

^Diportance should be attached to any argument based on the of the cell walls, as Schenk himself has admitted; butnbsp;fructification is a much more important feature. Saporta'�nbsp;places NilsHonia in the Cycadaeece, and considers that Schenknbsp;tvas probably deceived by certain leaf parasites, which mightnbsp;^�11 present an appearance closely simulating fern sori. Nathorst �nbsp;nows a similar course, and speaks of our ignorance as tonbsp;actual nature of Schenk�s sori, seeing that no traces ofnbsp;structure have been preserved; he suggests stomata and parasiticnbsp;Hgi as two possible explanations of these sorus-like appearances.nbsp;itns-Laubach � does not accept the proposed explanation asnbsp;satisfactory, and inclines to follow Schenk in classing Niluonianbsp;among the ferns, on the strength of the sorus-like bodies on thenbsp;�at lamina. Nathorst draws special attention to the insertionnbsp;� the leaf segments on the upper surface of the raohis as annbsp;Essential character of the genus; he speaks of the veins as equalnbsp;simple. Various authors have spoken of two kinds of veinsnbsp;the leaves of Niluonia, stouter and finer veins, but Nathorstnbsp;Remarks that Schenk has recognized his mistake with regard tonbsp;� supposed two sets of veins; he mistook folds in the leafnbsp;mina for well-marked simple veins.�� Schimper,� in the firstnbsp;^lume of the Trait, pal. v�g., classes Niluonia with thenbsp;*nbsp;nbsp;nbsp;nbsp;and accepts Schenk�s interpretation of the apparently

fertile specimens ; but in a later work' by this author we find Nihsonia placed close to Pterophyllum in the Cyoadaoece.

Without following in further detail the various descriptions or definitions of this genus, we may thus sum up the chiefnbsp;characters by which the species may best be recognized:�

Frond coriaceous, the lamina more or less deeply pinnatifid, the lines of division generally extending almost to the rachis ;nbsp;segments attached to the upper surface of the axis by thenbsp;entire base, contiguous, usually broad and truncate, but varyingnbsp;considerably in size and shape; apices obtuse or truncate. Veinsnbsp;simple and equal.

In connection with Nilssonia, which may best be considered as a genus of doubtful affinity, but probably cycadean, there arenbsp;three other genera of which some mention must be made,�nbsp;^mmozamites, Ptilozamites, and Pterophyllum. As regards thenbsp;last, Nathorst has on several occasions emphasized the distinctnbsp;difference as regards the manner of attachment of the leafnbsp;segments; in Pterophyllum they are inserted laterally on thenbsp;rachis; in Nihsonia, as in Bioonites, they are attached to thenbsp;upper face of the leaf axis.^ The genus Anomozamites wasnbsp;instituted by Schimper� for certain species of Pterophyllum withnbsp;irregularly pinnatifid leaves, and this term has been generallynbsp;adopted; the veins are described as simple and parallel, and thenbsp;segments as laterally attached. �lathorst, however, has institutednbsp;a genus, Ptilozamites,'^ in which are included pinnate and bipinnatenbsp;fronds, which in habit correspond fairly closely with Anomozamites,nbsp;hut differ in the possession of forked veins which dichotomizenbsp;at the base, and occasionally branch a second time before reachingnbsp;the margin of the leaf segment. The plant originally describednbsp;by Leckenby as Ctenis Leclienhyi,^ the specific name having beennbsp;suggested by Bean, shows very clearly the characters of thenbsp;venation and the branching habit of the frond; this must nownbsp;be included in Hathorst�s Ptilozamites � as suggested by the author

^ Hathorst (2), p. 61, and (A. 3), Denkschr. k. Ak. quot;Wiss. vol. Ivii. p. 45.

of the genus himself. An inspection of l^athorst�s figures of -^nomozamites and Ptilommites species suggests a difficulty innbsp;certain cases in deciding between the two genera. In such a formnbsp;0�S Anomozamites gracilis, Hath.,' we have well-marked branchingnbsp;la some of the veins, and a close approximation in generalnbsp;oppearanoe to other species included in Ptilozamites, cf. e.g.nbsp;-f¹ ². Seeri, Hath.,^ and A. minor ² (Brong.).

Fontaine 1 figures a few small fragments of leaves from the Fotomae flora under the name Anomozamites, but there is hardlynbsp;enough material to justify even a generic determination, and stillnbsp;less to warrant the institution of two new species.

Nilssonia Schaumburgensis (Dunk.).

1844. nbsp;nbsp;nbsp;Pterophyllum Schaumburgense, G�ppert, Foss. Cycad. p. 54.

1846. Pterophyllum Schaumburgense, Bunker, IVealdenbildung, p. 15, pi. i. fig. 7; pi. ii. fig. 1; pi. vi. figs. 5-10.

1848. nbsp;nbsp;nbsp;Pterophyllum Schaumburgense, Broun, Index pal. nomencl. p. 1056.

1850. nbsp;nbsp;nbsp;Pterophyllum Schaumburgense, Unger, Gen. spec, plant, foss. p. 292.

1851. nbsp;nbsp;nbsp;Pterophyllum Schaumburgense, Miquel, Rangschik, foss. Cycad. p. 213.

1852. nbsp;nbsp;nbsp;Pterophyllum Schaumburgense, Ettingshausen, Abh. k.-k. geol. Eeichs.nbsp;vol. i. Abth. iii. No. 2, p. 22.

1869. Anomozamites Schaumburgensis, Schimper, Trait, pal. v�g. vol. ii. p. 141. 1871. Anomozamites Schaumburgensis, Schenk, Palseontograpbica, vol. xix.nbsp;p. 231, pi. xxxiii.

Pterophyllum Schaumburgense, Peyton, Quart. Journ. Geol. Soc. vol. xxxix. Proc. p. 3.

Nilssonia Schaumburgensis, Yokoyama, Journ. Coll. Sci. Japan, vol. vii. pt. iii. p. 227, pi. XX. figs. 12 and 14; pi. xxi. fig. 14 ; pi. xxii.nbsp;figs. 5-7-

� Pterophylluin fronde pinnate vel, rarissirae quidem, profunde pinnatifida, pinnis alternis approximatis sub-obliquis irregularibus,nbsp;oblongo-ovatis, vel quadratis vel rotundatis, infimis snbdecurrenti-bus, nervis crebris tenuibus instructris, rhachi (suptereti ?)nbsp;longitndine striata.�

The specimens fignred by Dunker from the ITortb German Wealden beds show a considerable variation in size and form ; thisnbsp;variable character is also well brought out in the later and morenbsp;perfect figures in Schenk�s monograph. Such a specimen as thatnbsp;represented by Dunker in pi. i. fig. 7' must probably be regardednbsp;as a leaf seen from the under side, thus showing a Pterophyllum-like appearance. Schenk draws attention to the apparent lateralnbsp;attachment of the segments in some of the specimens which arenbsp;seen from the under surface, but notes that there can be no doubtnbsp;as to their actual insertion on the upper surface of the axis.nbsp;Schimper includes this variable Wealden species in his genusnbsp;Anomommites, and Schenk accepts this determination. Peyton hasnbsp;previously recorded the species in the English Wealden beds, butnbsp;no detailed descriptions or figures accompany his note. Jlathorstnbsp;records from Japan specimens of what is most probably the samenbsp;species as the English and North German forms; he refers tonbsp;a previous paper* in which he pointed out the true Nihsonianbsp;character of Dunker�s species, the segments being attached to tbenbsp;upper face of the raohis, and not laterally as in Pterophyllum ornbsp;Annmozamites. A leaf fragment closely resembling the presentnbsp;species is figured by Schenk* from Persia Anomozamites minor,nbsp;Schimp. In Yokoyama�s recent contribution on Mesozoic plantsnbsp;from Kozuke, etc., several specimens are referred to this speciesnbsp;on the authority of Nathorst; the figures suggest a lateral attachment of the unequal segments, but possibly the leaves are shownnbsp;with the under side uppermost. Nathorst�s figures of this speciesnbsp;from Japan represent typical Nihsonia fronds.

* nbsp;nbsp;nbsp;(2), p. 82.

2171.** Pigs. Za and J. Tlie two figured specimens are Examples of the narrower form of the species: in a the almostnbsp;Entire lamina resembles a small example of Tmniopteris; in hnbsp;6 truncate segments are well shown; and in both cases thenbsp;''^nation is distinct. Both specimens are represented twice thenbsp;natural size.nbsp;nbsp;nbsp;nbsp;Eufford Coll.

2171fl'. Fig. Ze. Broader specimen, I'l cm. in breadth. The ttedian groove on the upper surface and the veins are verynbsp;cinct. This example serves as a connecting link between thenbsp;arger forms of the plant as described by Schenk, and the smallernbsp;nglish specimens.

V. 2171. Several specimens. The variation in the size and division of the lamina is well illustrated in these examples.nbsp;Generally speaking, Schenk�s figures represent leaves with morenbsp;I�egular lobes than are found in the English specimens. In somenbsp;cases the segments are numerous and very narrow, in others thenbsp;lamina is almost entire.

V. 2171c. This specimen of one of the narrower forms of the species shows a depression at the distal end of the lamina, andnbsp;presents a very similar appearance to that of Tainiopteru Beyrichiinbsp;(Schenk) as figured by Schenk. Ecclesbourne. Mufford Coll.

V. 2172. This specimen was erroneously included under Tmnio-pteris Beyrichii in Vol. I. (p. 126).

V. 2234lt;;. 5 mm. broad. In a length of 4-2 cm. there are about 23 divisions in the lamina; in the same length of V. 2171S,nbsp;the lamina shows no divisions. At one end the segments graduallynbsp;decrease in size until they almost disappear. Ecclesbourne.

Other specimens : V. 716. Hastings. Dawson Coll. V. 1436. Ecclesbourne. Presented by P. Rkfford, Dsq., 1886. V. 2234.nbsp;Ecclesbourne. Rujford Coll.

Genus OTOZAMITES, Braun.

The name Otozamites, instituted by Braun in 1843, was defined by him as follows: Leaves pinnate, pinnae alternate and approximate, auriculate, and attached by a portion of the base;nbsp;veins radiating from the point of attachment to the margins ofnbsp;the segments.

This author includes Zamites falcatus (Sternb.), Z. RucMandi (Brong.), and Z. hrevifolius (Braun, etc.) as species of his newnbsp;genus ; Z. falcatus was first figured by Sternberg * as Odontopterisnbsp;falcata, and 0. RucMandi was described by Brongniart in 1825 asnbsp;Filieites BucUandi var. Rritannica ^; both of these species werenbsp;assigned by Morris � to his genus Ptilophyllum. Brongniart adoptsnbsp;Braun�s generic name, and points out that Otopteris, Bindley andnbsp;Hutton, corresponds to Otozamites of Braun. This genus affordsnbsp;another example of confusion in nomenclature arising from anbsp;difference of opinion as to the botanical relationship of the fossil

' (A.), Flor. Vorwelt, pi. xxiii. fig. 1, fasc. 5 and 6, p. 78. � Brongniart (4), p. 422.

fronds. Brongniart' says that Otozamites Buchlandi, Brong., was figured by De la Beche¹ ² as a fern from the Lias of Axminster,nbsp;and by Bindley and Hutton as Otopteris obtusa,^ the typicalnbsp;species of the genus. The same writer suggests the advisabilitynbsp;of distinguishing certain leaf forms under another genus, Sphmo-^amites, of which the chief characteristic is the absence of annbsp;nnrioulate base in the pinnae; this genus has come into generalnbsp;and serves a useful purpose as a convenient provisionalnbsp;term. Bornemann^ has suggested that probably some of thenbsp;plants referred to Otozamites are without true auriculate pinnae,nbsp;the apparently eared form being merely a result of pressure onnbsp;the upper surface of the thick pinnae. He removes some ofnbsp;Braun�s species from the genus, and speaks of Otozamites hrevifoliusnbsp;(Sternb.) and 0. graminem (Morr.) {=.Zamites gramineus, Morris)nbsp;as typical species. The following is Bornemann�s emended versionnbsp;of Braun's diagnosis : � � Leaves pinnate ; pinnae approximate,nbsp;alternate, or subopposite, lanceolate, pointed or more or lessnbsp;filunt, auriculate at the base, and attached to the rachis only bynbsp;the lower part, the upper corner of the auriculate base prolongednbsp;�Dd partly covering the rachis. Veins radiate from the point ofnbsp;attachment towards the margin of the pinnae, and are for thenbsp;�ost part dichotomous.� This definition appears to be on thenbsp;'Whole satisfactory, but Bornemann unfortunately errs in describingnbsp;the pinnae as attached to the rachis by the lower portion; thenbsp;Wanner of insertion of some auriculate pinnae cannot correctly benbsp;described according to his definition. Schenk has discussed atnbsp;some length the botanical position of the genus Otopteris in hisnbsp;l^lora Her GrenzschicMen; � he draws attention to a specimen ofnbsp;'Which the segments exhibit a peculiar marginal structure,nbsp;suggestive of a Pteris-\\ke fertile leaf. The structure of thenbsp;opidermal cells is also referred to in support of the inclusion ofnbsp;this genus among the Filicinm-, but in his later writings Schenknbsp;speaks of Otozamites as a member of the CyoadaeecB.

Schimper, in the first volume of his Trait, pal. veg,'- adopts the genus Otopteris, L. and H., but afterwards (vol. ii.) � acceptsnbsp;Braun�s generic name Otozamites; he institutes a sub-genus Rhomho-zamites for Otozamites Beanii and other species, and makes usenbsp;of one of Pomel�s terms, Cyclozamites, for Otozamites Bunhuryanusnbsp;and other forms. This subdivision seems quite unnecessary, andnbsp;tends rather to confusion than to useful classification. Saporta^nbsp;retains Otozamites Bucldandi (Brong.) as the type of the genus,nbsp;hut in his diagnosis mention is made of certain features whichnbsp;set rather narrow limits to the generic characters; the basalnbsp;callosity of the pinnae and the auriculate upper angle of the base,nbsp;are features which do not always appear in fronds which mustnbsp;he referred on general grounds to the genus Otozamites. It isnbsp;true we frequently find that the upper lobe of the pinna basenbsp;is more decidedly auriculate than the lower, but this is notnbsp;universal. Saporta�s figures of some of the species of Otozamitesnbsp;show this quite clearly; e.g., 0. Reglei, Sap.,^ 0. Brongniarti,nbsp;Schimp.,� etc. In describing the characteristics of the variousnbsp;examples of the genus, Saporta points out the numerous variations from the normal type. In discussing the geological historynbsp;of the genus, this author refers to the absence of Otozamites fromnbsp;Wealden and Feocomian strata; since these words were writtennbsp;several examples of Wealden forms have been discovered, andnbsp;the material acquired in recent years shows that the small plantnbsp;figured by Bunker as Cyelopteris Klipsteinii^ is most probably anbsp;species of Otozamites. The groups into which Saporta dividesnbsp;this genus have been adopted by Schimper in Zittel�s Uandbuehi'nbsp;The recognition of certain typical species as representatives ofnbsp;different forms of a genus may in some cases be a convenientnbsp;aid to classification, but there is always the danger of undulynbsp;emphasizing slight and unimportant differences for the sake ofnbsp;such purely arbitrary grouping. In the case of a genus such as

Otozamites we know very little indeed as to its exact botanical position, and for the present, at least, it will probably be betternbsp;not to bind ourselves to any of these subdivisions of the genus.nbsp;We may adopt a definition of Otozamites very similar to thatnbsp;previously quoted from Bornemann, but which gives a more definitenbsp;expression to the variable character of the numerous forms andnbsp;fronds included in Braun�s genus :�

Frond pinnate; pinnae attached to the npper surface of the rachis by a portion of the aurioulate base, base more or lessnbsp;distinctly auriculate, the upper lobe often more prominent thannbsp;the lower; segments may be approximate, imbricate, or distinct,nbsp;^eins numerous and branched, radiating from the point ofnbsp;attachment and cut o�E obliquely by the margin of the pinna; innbsp;the longer and narrower form of pinna the veins are practicallynbsp;parallel to the edges of the segment; pinnse vary from long,nbsp;narrow, and linear-lanceolate, with acute tips, to broadly ovalnbsp;or almost orbicular in form, with bluntly rounded apices.

Solms-Laubach,^ after speaking of the flabelliforin venation of Otozamites, goes on to say that in some forms of this genus thenbsp;xeins of the pinnae conform much less distinctly to that type thannbsp;in others. If we examine the narrower and longer forms ofnbsp;Otozamites pinnse, the veins become more or less parallel to thenbsp;segment margins soon after leaving the aurioulate basal portion.nbsp;In the longer segments of Otozamites gramineus and other speciesnbsp;this is the case, so that the flabelliform character of the venationnbsp;cannot by any means be relied upon as a constant and easilynbsp;recognized characteristic of the genus. It is often a difficultnbsp;leatter to decide whether the pinnse bases are actually auriculate;nbsp;in dealing with some fronds we find it almost impossible tonbsp;draw a well-marked line between Otozamites, Ctenophyllum, andnbsp;I'tilopkyUum, etc.

Otozamites Klipsteinii (Dunk.).

Dunker defines the species as follows: �Cydopteris fronde pinnata, pinnulis alternis sessilibus ? ovato-oblongis sequalibus,nbsp;nervis creberrimis flabellatis tenerrimis.� He points out thatnbsp;the veins are exceedingly delicate, and apparently dichotomousnbsp;in the upper portions, the venation shown in his figures beingnbsp;coarser than it actually is in the specimens. Ettingshausennbsp;figures four detached leaflets which he describes as intermediatenbsp;in character between Cyolopteris Mantelli and C. Klipsteinii; innbsp;his drawings there appears to be a distinct suggestion of anbsp;midrib, but nothing is said in the definition of the species asnbsp;to the existence of a median vein. It has already been pointednbsp;out (quot;Wealden Catalogue, vol. i. p. 131) that these leaflets arenbsp;certainly not typical examples of Sagenopteris Mantelli (Dunk.),nbsp;as Ettingshausen erroneously states. It will be better to leavenbsp;them out of consideration as doubtful fragments.

Schenk figures a single pinna of this species, which shows very clearly an auriculate base and dichotomously spreadingnbsp;veins. The Dufford Collection contains a large number of well-preserved fronds with pinnae of various sizes, and which in somenbsp;cases are clearly identical with the species figured by Dunkernbsp;and Schenk as Cydopteris or Aneimidium Klipsteinii. Thenbsp;number and variety of the specimens present a difficulty asnbsp;regards specific determination. A casual inspection of the frondsnbsp;with large and broadly oval segments would probably lead one tonbsp;institute a new specific name for their reception, but on carefullynbsp;examining and comparing all the examples, it appears to benbsp;impossible to determine definite specific limitations. The frond

figured in PI. I. Pig. 3 must certainly be referred to Bunker�s species; the pinnae agree closely 'with that figured hy Schenk,nbsp;and with the inferior figure hy Bunker. We have a pinnatenbsp;frond, with segments attached by an auriculate base; the venationnbsp;agrees with that of Otozamites, and the general character of thenbsp;leaf points to that genus as the most convenient designation fornbsp;the specimens. Another specimen, PI. II. Pig- 4, possessesnbsp;pinnae of the same form as those in V. 2236, hut differs in itsnbsp;much stouter rachis and in the imbricate disposition of thenbsp;segments. It is difficult to speak confidently as to the relationnbsp;of the several specimens one to another, hut probably we havenbsp;ia PI. II. Pig. 4 the lower portion of a young frond of the samenbsp;form of which PI. I. Pig. 3 represents a terminal fragment.nbsp;Leaving a more detailed notice of these specimens until later,nbsp;must turnnbsp;nbsp;nbsp;nbsp;to the extremely fine examples ofnbsp;nbsp;nbsp;nbsp;thenbsp;nbsp;nbsp;nbsp;larger

fronds, such as nbsp;nbsp;nbsp;those representednbsp;nbsp;nbsp;nbsp;in PI. VII. Pig. 9,nbsp;nbsp;nbsp;nbsp;etc.nbsp;nbsp;nbsp;nbsp;It is

possible there may be two or thi'ee species represented by this splendid seriesnbsp;nbsp;nbsp;nbsp;of fronds, whichnbsp;nbsp;nbsp;nbsp;I have referred tonbsp;nbsp;nbsp;nbsp;thenbsp;nbsp;nbsp;nbsp;genus

Otozamites; in nbsp;nbsp;nbsp;some the pinnaenbsp;nbsp;nbsp;nbsp;are longer and narrower than

iu others, and in some we have a shorter and broader form of ^ogment. On the whole, in the absence of any constant andnbsp;'H'ell-marked differences consistent with separate species, I prefernbsp;fo include nearly all these various forms under one species,nbsp;Otozamites Klipsteinii (Bunk.), and resort to descriptive termsnbsp;for the designation of one or two varieties. We may thusnbsp;define this comprehensive species: Prond pinnate, rachis fairlynbsp;stout, tapering to a slender axis in the terminal portion; pinnaenbsp;attached to the upper surface of the rachis, in the young frondnbsp;probably imbricate, with a more or less well-marked auriculatenbsp;base; veins numerous, radiating from the point of attachmentnbsp;towards the margin of the pinnae; apices obtuse, pinnae alternatenbsp;or subopposite, in the mature fronds almost at right angles tonbsp;the axis, or more or less obliquely inclined.

Before describing the individual specimens referred to this species and its varieties, we may notice some other forms of thenbsp;genus Otozamites, and other plants with which the present examplenbsp;�uy be compared. Otozamites Beanii (L. and H.)' corresponds

�with 0. Klipsteinii in having auriculate pinnae, broader and stouter than the majority of species of the genus. In describing the typenbsp;specimen, Lindley and Hutton ask if the plant may be a pinnatednbsp;leaf of the Cyeadeoidece, but reply with a decided negative; in thenbsp;second volume of the Fossil Flora the generic name Otopteris isnbsp;substituted by the authors of the species for Cyclopteris, undernbsp;which the plant was originally described. In the Leckenby Collection in the Woodwardian Museum, Cambridge, there are severalnbsp;very fine specimens of Otozamites Beanii, which in some instancesnbsp;show a terminal portion with small pinnae very similar to suchnbsp;an apical tip as is represented in PI. I. Fig. 3. In the Yorkshirenbsp;plant the species are, however, very distinctly auriculate, andnbsp;have the upper portion of the base more prominently lobed thannbsp;is the case in 0. Klipsteinii (Dunk.). In one of Leckenby�snbsp;specimens, which possesses a raohis 43�5 cm. long, the largestnbsp;pinna has a length of 3 cm. and a breadth of 1'9 cm., the smallestnbsp;measuring 1 cm. by 7 mm. Some of Zigno�s figures of his speciesnbsp;Olozamites Molinianus' present a striking likeness to 0. Beanii,nbsp;and ought most likely to be referred to that species. Kurr�s figurenbsp;of what he calls Zamites Mandelslohiquot;^ agrees very closely withnbsp;the specimen represented in PI. IV. Fig. 4; the genus Otozamitesnbsp;has been rightly substituted by Schimper for Kurr�s species.nbsp;The larger pinnae of 0. Klipsteinii show a certain resemblancenbsp;to those of some few previously described plants; but in no casenbsp;does the similarity appear sufficiently pronounced to justify anbsp;reference to the same species. In the third volume of thenbsp;Gond'wana Flora of India,^ Feistmantel figures some isolatednbsp;pinnae under the name Glossozamites StoUczkanus, Feist., andnbsp;describes them as probably constituting the largest representativesnbsp;of Schimper�s genus. The form of these large segments is notnbsp;at all unlike that of the Wealden pinnae, but in the latter casenbsp;the auriculate base favours the adoption of the genus Otozamites.nbsp;The plant described many years ago by G�ppert, from a muchnbsp;lower geological horizon, under the name of Cyotopteris frondosa^nbsp;(Gopp.), may be compared, as regards the form of its leaf segments.

the larger specimens of the present species. Sohimper places this Culm species in his genus Cardiopteris, and describes it as annbsp;Unusually large-leaved neuropteroid fern. The venation of thenbsp;large pinnae of the Wealden fronds suggests that of certain fernnbsp;pinnules; and it must be borne in mind in dealing with such leavesnbsp;as those of a neuropteroid or otozamitean type, that we arenbsp;Unable to speak dogmatically as to the botanical position of thenbsp;species.

Among the Jurassic plants figured by Saporta � there occur u few examples of fronds with leaf segments comparable to thosenbsp;uf 0. Klipsteinii\ e.g., 0. deeorus, Sap., and 0. lagotis, Brong.nbsp;A detached segment named by Saporta Sphenozamites Brongniarti,nbsp;ugrees fairly closely with some of the larger pinnse of the Englishnbsp;Wealden specimens, but the French fragments are too small to allownbsp;uf any precise comparison, and hardly worthy of a special specificnbsp;Uame. The frond fragments described by Hosius and von March asnbsp;^terophyllum blechniforme ^ have a certain resemblance to some ofnbsp;the English specimens, but are clearly not specifically identical; thenbsp;generic name Otozamites would seem to be more applicable to thisnbsp;species of Hosius and von Marck than that of Pterophyllum. Thenbsp;Puinas of Fontaine�s species Zamites tenuinervis � may be comparednbsp;U^ith some of the longer segments of Otozamites Klipsteinii, butnbsp;there is not sufficient evidence to justify the inclusion of thenbsp;I�otomac and English forms in the same species.

Finally, we have a somewhat similar form of leaf in some examples of the Palseozoic genus Cordaites ¹ ² recently figured bynbsp;hlrand�Eury from the Coal-field of Gard. Whilst drawing attentionnbsp;to some of the plants previously described, in which a greater ornbsp;less resemblance may be traced to the present species, it mustnbsp;he definitely stated, that we are without any satisfactory evidencenbsp;Us to the exact position in the plant kingdom to which these large-leaved forms should be referred. Having regard to the generalnbsp;habit of the fossil fronds, the apparently stiff nature of the pinnse,nbsp;6fc., the cycads appear to be the more likely plants with which to

compare Olozamites Klipsteinii. Among recent cycadean species we do not meet with fronds possessing pinnse with a well-definednbsp;aurioulate base, but we are not without instances of fronds whichnbsp;in other respects hear a decided resemblance to the Wealden plant.nbsp;Zamicb purpuraoea, Ait., may be cited as one recent species withnbsp;large pinnae comparable to those of the Wealden plant; the youngnbsp;pinnae of this living cycad are very similar in form to the morenbsp;terminal segments of Otozamites Beanii. In this species of Zamia,nbsp;as also in Z. pxjgmma, Sims, and in some other forms, etc., thenbsp;pinnae are distinctly nairowed towards the point of attachmentnbsp;to the rachis, from which they are readily detached, leaving anbsp;well-marked scar. The recent fronds correspond rather morenbsp;closely with such fossil forms as Podozamites Reinii, Geyl., etc.,nbsp;described by Geyler' and others. In Otozamites latifoliiis (Phill.)�nbsp;we have another large-leaved form which may to some extentnbsp;be compared with the present species : the specimens of Phillips�nbsp;species are imperfectly preserved, and do not give any decidednbsp;indication of an aurioulate base; the venation appears to agreenbsp;fairly closely with that of Nathorst�s genus Ptilozamites, and thenbsp;prominence of the veins reminds one of some of the large andnbsp;boldly veined segments of such recent species as Zamia picta,nbsp;Z. Skinneri, quot;Warsz., etc.

V. 2336. PI. I. Pig. 3. This terminal portion of a frond must no doubt be referred to the same species in which thenbsp;detached segments figured by Bunker and Schenk have beennbsp;included. In Bunker�s specimens there is not the same distinctlynbsp;aurioulated base as in this example, hut this may be put downnbsp;to less perfect preservation and possibly inaccurate drawing ; innbsp;Schenk�s leaflet of the same species the aurioulate base is distinct,nbsp;and agrees exactly with that in the English frond. The veinsnbsp;appear to be rather fewer and farther apart in this terminalnbsp;portion than in the pinnae of the larger fronds. Largest pinnanbsp;2 cm. by 9 mm. Eocleshourne.nbsp;nbsp;nbsp;nbsp;Rujford Coll.

V. 2170a. Two detached pinnae ; the venation finer, and exactly like that in Schenk�s example. Ecoleshourne.nbsp;nbsp;nbsp;nbsp;Rufford Coll.

O. Klipsteinii (Dunk.), var. superba mihi. v. 2745�. PI. I. Pig. 4.

S.achis 12'5 cm. in length. Largest pinna 4'2 cm. by 1'6 cm.; �raallest 3'8 by 1�2. The largest pinnae of V. 2336 has a lengthnbsp;2 cm. and a breadth of 9 mm. Some of the pinnae are verynbsp;'clearly preserved, and show excellent venation. The rachisnbsp;^-Ppears to he slender, as in V. 2336. The upper pinnae havenbsp;'��ry slightly aurioulate bases, which are attached by their centralnbsp;portion to the surface of the frond axis. In the lower pinnae thenbsp;^ase is more distinctly lobed. Cf. Pteropliyllum ohlongifolium,nbsp;Olossozamites ollongifolius (K.VLrr)\^ There is not, I amnbsp;inclined to think, sufficient proof of any important differencenbsp;between V. 2336 (PI. I. Pig. 3) and V. 2745� (PI. I. Pig. 4)nbsp;^0 Warrant a specific separation; but as we shall find a gradualnbsp;transition from such specimens as V. 2745fi( to the fronds withnbsp;much larger pinnae, e.g. Pi. YII., it will be convenient to institutenbsp;^ Variety of Dunker�s species under the name superha. Eccles-tiourne.nbsp;nbsp;nbsp;nbsp;Rufford Coll.

2170. PI. VII. Pig. 9. It may be, as suggested in the introductory account of the species, that we have two or threenbsp;�Pecies included under 0. Klipsteinii (Dunk.), but thanks to thenbsp;numerous and well-preserved specimens it is possible to examinenbsp;nn Unusually fine series, and so escape to some extent from thenbsp;dangers of fragmentary and imperfect portions of fronds, whichnbsp;n^ten lead to unnecessary multiplication of specific names. Innbsp;present instance we can trace a gradual transition from thenbsp;�hort and broad segments, such as those represented in PI. VII.nbsp;^i�- 5 (V. 2745), to the large forms such as PI. VII. Figs. 1nbsp;^ud 6 (V. 2122). The leaves with the broader and more ornbsp;less imbricating obliquely set pinnae come very near to 0. Beanii,nbsp;�iid those with more separate, longer, and narrower segmentsnbsp;i^losely resemble the pinnae figured by Pontaine as Zamitesnbsp;^�^n-uinervis, Pont. In fragments of young fronds or in the lowernbsp;part of older fronds the pinnae may have a length of 2 cm. andnbsp;*1 breadth of l'4cm., in the larger segments 8�2 X 2�4 cm. Innbsp;1'bis specimen (V. 2170) the rachis is 23cm. in length; it hasnbsp;lift a hollow mould in the rook, roofed over by the basal ends

of the pinnse; on one side there are nine pinnse, and on the other seven. These show the hroad, obtusely terminated form ofnbsp;the segments very clearly; they are attached to the upper surfacenbsp;of the rachis by the central portion of the distinctly auriculatenbsp;base. Most of the pinnse overlap, the upper edge of each projecting over the lower margin of the pinna next above it; verynbsp;similar in habit to some of the larger specimens of 0. Beanii innbsp;the Leokenby Collection. On the under side of the same slabnbsp;there occur portions of pinnae of another frond, and a section ofnbsp;a rachis mould, about 6 mm. broad and 8 mm. in depth; alsonbsp;a specimen of Cycadites Saportm, sp. nov. Ecclesbourne.

V. 2122a. PL VII. Pig. 2. Rachis imperfectly preserved; portions of six pinnae, the uppermost having the same form asnbsp;the largest of V. 2745 (PI. I. Fig. 4), the lowest and largest,nbsp;6'4 X 3 cm. (PL VII. Pig. 2), showing a slight lobe in the middlenbsp;of the upper margin. In some of the segments the point ofnbsp;attachment is well shown, and the numerous fine spreadingnbsp;veins are well marked in the carbonaceous surface layer.

V. 212Sa. PL VII. Pigs. 4 and 8. Frond 28cm. long.; 11 pinnee on one side. Hollow round mould of rachis, about 4-5 mm.nbsp;in diameter. Short and broad overlapping pinnae, 4�8 x 2'5 cm.,nbsp;and very like those of V. 2740�, V. 2740, etc. The upper andnbsp;smaller pinn�, as shown in the figure, are narrower and longernbsp;and not overlapping; these agree exactly with the segments innbsp;V. 2745fl! (PL I. Pig. 4). Uppermost pinn� about 3 cm. x 1'8 cm.;nbsp;a segment separated from this by 11 cm. and attached to about thenbsp;middle of the specimen, measures 4-5 x 2'5 cm. The two largernbsp;pinnffi (PL VII. Fig. 8) agree very closely with those of V. 2170nbsp;(PL VII. Pig. 9); they have a strongly convex upper surface,nbsp;and show a sharp bending down of the lamina near the pointnbsp;of insertion to the rachis. This bending of the pinna base isnbsp;e.xactly similar to the appearance frequently presented by thenbsp;stiff leathery pinn� of recent species of EncephalartoB, to whichnbsp;allusion has been made in the introductory remarks on fossilnbsp;Cycadacem.

V. 2126. 32 cm. long. Eachis about 4 mm. broad, in the form of a round hollow mould. Compare the largest pinna of thisnbsp;specimen with the smallest of V. 2122.

V. 2740. PI. VII. Fig. 7. Specimen 37cm. long; 13 pinnae on one side. The marked variation in length of pinn� at thenbsp;npper and lower end of the rachis is well seen. Several segmentsnbsp;show the manner of insertion on the frond axis; venation fairlynbsp;lt;iistinct. The pinnae present a strongly convex surface, and arenbsp;slightly inclined to the horizontal plane of the rachis, as innbsp;^eichselia Mantelli (Brong.), (Vol. i. p. 116). At the upper endnbsp;of the axis the pinn� bases are 2-7 cm. apart, towards the lowernbsp;end 2-2-5 cm. apart. The upper pinn� are practically identicalnbsp;�with those of V. 2170 (PL VII. Fig. 9), but the shorter, broader,nbsp;closer pinn� of V. 2170 are somewhat broader and closer thannbsp;aay in V. 2740.

V. 2740a. 30 cm. long. The venation exceedingly well marked. S-aohis 5 mm. broad, flattened. Pinn� vary considerably in sizenbsp;�t the two ends of the frond, 4-8 X 2-5 to 3-5 X 21 cm. Thenbsp;^uriculate base of many of the segments clearly shown. For thenbsp;�iost part the segments overlap one another, oblique to the rachis,nbsp;^ad apparently of a thick and leathery nature. Cf. 0. Beanii.nbsp;^cclesbourne.nbsp;nbsp;nbsp;nbsp;Buffovd Coll,

V. 2912a. PI. VII. Fig. 3. Eachis 33 cm. in length. Lowest pinna 5-7 x 2-7 cm.; uppermost 6-5 X 3. Venation in excellentnbsp;preservation. Apices bluntly rounded ; the lower edge of the basenbsp;appears to be more distinctly lobed than the upper. Smallernbsp;pinn� like the larger ones in V. 2740, and the larger like thosenbsp;�f V. 2912, etc.

V' 2912. This specimen shows six pinn� attached to the rachis, ffiree on each side; pinn� 7 X 2-7 cm. broad, do not overlap,nbsp;inclined to the horizontal plane of the rock surface. Eachisnbsp;striated longitudinally, and shows cross lines at intervals, butnbsp;ffie latter are no doubt simply minute transversely running cracksnbsp;in a mineral substance. Cf. V. 2090, V. 2122, etc. Ecclesbourne.

V- 2745. PI. VII. Fig. 5. Eachis 12 cm. long, with six pinn�. i^inn� short and broad. Compare this specimen with V. 2336nbsp;(I*!- I. Fig. 3) and V. 2745a (PI. I. Fig. 4). Largest pinn�nbsp;^�8X1-6 cm., lowest and smallest I'SXM cm. Venation distinct.nbsp;^f. Upper end of V. 2740a.

V. 27455. 12'5 cm. long. Seven pinnae on one side : largest 4 2 X 16 cm., smallest 3'8 X 1'2 cm. Eaohis appears to be 2�5 mm.nbsp;Venation good, upper pinnae slightly auriculate, in the lower segments the arioulate base is more distinct. Cf. Pterophyllumnbsp;ohlongifolium, Kurr. Eoolesbourne.nbsp;nbsp;nbsp;nbsp;Rufford Coll.

V. 2090. Part of a rachis and four pinnae. Pinnae long, not overlapping, about 3 cm. apart, of similar form to those of V. 2912.nbsp;Eoolesbourne.nbsp;nbsp;nbsp;nbsp;Presented by P. Rujford, Psq., 1889.

V. 2364�. Portion of a single pinna; compare with some of the lower pinnae in V. 2740a; broad and short, 5 x 3'8 cm. Verynbsp;different in appearance from V. 2364, but probably the samenbsp;species ; in V. 2122 we have the same kind of divergence in thenbsp;form of the segments as between the two specimens V. 2364nbsp;and V. 2364lt;?. Ecolesbourne.nbsp;nbsp;nbsp;nbsp;Rujford Coll.

O. Klipsteinii var. longifolia, mihi.

V. 2122. PI. VII. Figs. 1 and 6. Eachis 43 cm. long; 15 pinnae on one side. The pinnae in this specimen are of thenbsp;longer and narrower form. The smallest and lowest segmentnbsp;measures 7-5 X 2-5 cm. ; the uppermost and longest 8'4 X 2 cm.nbsp;The lowest pinna connects such an example as this with V. 2912,nbsp;V. 2090, V. 2740, etc. Pinnae separate, bases in the lower partnbsp;of the frond 3'5 cm. apart, the two uppermost separated bynbsp;2-5 cm. Venation not quite so distinctly seen as in some othernbsp;specimens; the lower pinnae show very clearly the auriculatenbsp;form of base.

V. 21225. This specimen, 20 cm. in length, is very imperfectly preserved. The pinnae are of the longer and narrower type,nbsp;and agree fairly well with those of Zamites tenuinervis, Font.,nbsp;but the distinctly auriculate form of the base and the mannernbsp;of insertion on the rachis seem to sufficiently distinguish thenbsp;specimen from the Potomac species, and to connect it with thenbsp;other forms of Otozamites Klipsteinii var. superba. Possibly itnbsp;would be better to designate such forms as V. 2122, V. 21225,nbsp;V. 2123�, etc., 0. Klipsteinii var. longifolia. Between thisnbsp;specimen and V. 2122 there is the closest similarity. Eccles-hourne.nbsp;nbsp;nbsp;nbsp;Rvfford Coll.

V. 2123a. Here again we have a specimen, imperfectly preserved, which it is diffloiilt to assign to a perfectly satisfactorynbsp;position. The pinnae are long and narrow, and agree with thosenbsp;V. 2122J, etc. ; hut on the other hand this form is nearer tonbsp;that of Zamites tenuinervis, Tont. The bases of the pinnae arenbsp;tess distinctly auriculate, and in some there is a wrinkling of thenbsp;leaf substance near the point of attachment to the frond axis.nbsp;Ilf. Sncephalartos Lehmmni. Ecclesbourne.nbsp;nbsp;nbsp;nbsp;Ruffurd, Coll.

Other specimens referred to this species: V. 1069, V. 2126, 2364, V. 2912^, V. 3160 (a good example of 0. KUpsteiniinbsp;longifolia).

Otozaiuites sp. Cf. O. Klipsteinii (Dunk.).

In PI, II. Pig. 4 is represented one of two specimens, of which is the counterpart of the other. The raohis has a length ofnbsp;12-5 cm., and bears at its upper end six imbricate and obliquelynbsp;attached pinnae. The base of the rachis is swollen, and. showsnbsp;^ clean cut surface by which it was attached to the parent stem ;nbsp;just above the base there are traces of filiform appendages, possiblynbsp;�cale or hair structures, such as occur on the lower portions of manynbsp;Accent cycadean fronds. The bases of the alternately placed pinnaenbsp;Hi in between one another, and are attached to the upper surfacenbsp;the rachis. Saporta^ has figured a very similar specimen asnbsp;^lo%amites sp., in which the base of the petiole shows two fairlynbsp;large leaf-like or stipular (?) structures. The Prench Jurassicnbsp;Specimen closely resembles 0. Beanh (L. and H.), but it is nonbsp;Houbt better to follow Saporta, and retain it as a fragment toonbsp;�niall to be accurately determined. Another somewhat analogousnbsp;Horna is figured by the same author as Otozamites marginatiis^-, butnbsp;dealing with portions of such partially developed fronds, anynbsp;iltetnpt to assign them to specific forms founded on nature frondsnbsp;Unst be attended with no little difficulty and risk. Possibly this

specimen (V. 2734) may be the basal portion of a young partially expanded frond of tbe same species (0. KUpsteinii), of whichnbsp;V. 2336 (PI. I. Pig. 3) is the apical termination of a young frond,nbsp;but it would hardly be wise to do more than offer the suggestion.nbsp;In V. 2336 the rachis appears to be very slender; here, on the othernbsp;hand, it is distinctly strong and broad : how far this constitutes annbsp;important difference, and how far it is due to one specimennbsp;being the terminal portion and the other the hasal portion, or tonbsp;differences in the manner of preservation, is difficult to decide.nbsp;The individual pinnse are very similar to, and possibly identicalnbsp;with, the isolated segment figured by Bunker' and Schenk� andnbsp;now classed with V. 2336. Ecclesbourne.nbsp;nbsp;nbsp;nbsp;Rufford Coll.

Otozamites sp. Cf. O. Reibeiroanus, Heer.

V. 2926. This small specimen shows ? pinnae attached to the upper surface of a rachis. It may belong to the same speciesnbsp;as V. 2734 (PI. II. Pig. 4), but of this there is no obviousnbsp;proof. It agrees very closely with Heer�s Portuguese speciesnbsp;0. Reibeiroanm? The upper lobes of the pinnae are morenbsp;prominent than the lower. Ecclesbourne.nbsp;nbsp;nbsp;nbsp;Rufford Coll.

Otozamites Goppertianus (Bunk.).

1848. nbsp;nbsp;nbsp;Pterophyllum Goppertianum, Broun, Index pal. nomencl. p. 1055.

1850. nbsp;nbsp;nbsp;Pterophyllum Goppertianum, Unger, Gen. spec, plant, foss. p. 290.

1851. nbsp;nbsp;nbsp;Bioonites Goppertianus, Miqnel, Tijdscb. Wis. Nat. Wet. iv. p. 7.

1852. nbsp;nbsp;nbsp;Pterophyllum Goppertianum, Ettingshausen, Abb. k.-k. geol. Beicbs.

1870. nbsp;nbsp;nbsp;Bioonites Goppertianus, Schimper, Trait, pal. v�g. vol. ii. p. 151.

1871. nbsp;nbsp;nbsp;Bioonites Goppertianus, Schenk, Palseontographica, vol. xix. p. 235,

Hunker thus defined the species in 1846: � Pterophyllum Honde impari-pinnata, pinnis oppositis subreotis elongatis lineari-bus aouminatis subremotis, seque distantibus angulo acuto adnatis,nbsp;hervis obsoletis quinis vel senis, rhachi crassa subtereti Irnvi.�

He speaks of each pinna having five or six equal and thin ^�ins, and shows this venation in the slightly magnified portionnbsp;Represented in fig. ha, pi. ii. Bunker�s figures certainly appearnbsp;justify his choice of the genus Pterophyllum as regards thenbsp;Reanuer of attachment and distal terminations of the segments.nbsp;Heference has already been made, in speaking of the variousnbsp;Sources of error connected with fossil oycads, to the very differentnbsp;Appearance presented by a frond when viewed from the uppernbsp;And lower surface. In the present instance it is quite possible,nbsp;And, indeed, considering all the facts before us, I believe verynbsp;probable, that we have in Bunker�s fig. 5 a view of the undernbsp;Anrface of the frond. The rachis is seen to project considerablynbsp;Above the level of the pinnae, and the latter are either attachednbsp;broad bases to its margin, or pass underneath and are unitednbsp;the face of the rachis which is pressed against the rocknbsp;Anrface. The venation, as Bunker describes and figures it, doesnbsp;r^At accurately correspond with that of the English specimensnbsp;^hioh I have ventured to refer to this species. If, however.nbsp;Hunker�s drawing represents an imperfect fragment seen fromnbsp;below, it is very likely that we should find a somewhat differentnbsp;Appearance presented by the veins to that which is seen in thenbsp;RAore perfect pinnae of the English fronds. If specimen V. 2360nbsp;(I*!- I. Pig. 2) be compared with Bunker�s pi. ii. fig. 5, thenbsp;striking resemblance between them in the form and arrangementnbsp;of the pinnm cannot fail to be noticed. The English specimennbsp;Seen from the under side would present an appearance very similarnbsp;fo that of Bunker�s frond. Turning to Schenk�s account of thenbsp;AAme species, we find he follows Miquel in adopting the genericnbsp;Riume Pioonites in preference to Pterophyll�m. Schenk includesnbsp;lA this species the specimen figured by Bunker as Pecopterunbsp;linearis, which the former regards as a badly preserved fragmentnbsp;of Pioonites Goppertianus ; this does not seem probable, so far asnbsp;the figures enable us to form an opinion. This author extendsnbsp;the original diagnosis of the species, and points out that the pinnaenbsp;Are attached to the upper surface of the frond axis; his figure

shows a median groove in the npper surface of the raohis between the two rows of segments. The veins, apparently not very distinctlynbsp;shown, are described as parallel, uniform, and delicate. Schenk�snbsp;fig. 3, pi. xxxiv. probably represents, as in Bunker�s fig. 5,nbsp;the under surface of a frond, but in fig. 4 of Schenk�s plate wenbsp;have a small piece of leaf seen from above, and here the mediannbsp;groove and manner of insertion of the segments are indistinctlynbsp;seen. A comparison of V. 2123 (PI. I. Fig. 1) with Schenk�snbsp;figs. 4 and 4� reveals a fairly close resemblance; the correspondence in the general form of the fronds and pinnae suggestsnbsp;the same species. In the German specimens we have only smallnbsp;and imperfect fragments, but in the Sussex examples the preservation is particularly good, and the details well marked. Thenbsp;pinnae of the English specimens do not in every case shownbsp;a distinctly auriculate base; it is only here and there that thisnbsp;feature can he seen. Considering the difference in the mannernbsp;of preservation in the two sets of specimens, it is not much tonbsp;he wondered at if no trace of the lobed base can be detected innbsp;the more imperfect specimens. Another important point is thenbsp;probability, that Schenk�s fig. 4 is the only specimen from thenbsp;German Wealden in which we have a view of the upper surfacenbsp;of the rachis.

A plant from the Lower Cretaceous rocks of Portugal, originally figured by Morris' as Zamites gramineus var. mundm, and afterwards by Heer * as Otoiamites angustifolius, shows a certain amountnbsp;of resemblance to the present forms; compare especially Heer,nbsp;pi. ix. fig. 10 and fig. 3J. It must be admitted that thenbsp;English specimens have a broader rachis than is apparent innbsp;Bunker�s figures, but this may easily he due to the differentnbsp;manner of fossilization, and cannot he relied upon as an essentialnbsp;difference, considering the nature of the material. Some ofnbsp;Saporta�s figures of Otozamites latior, Sap., resemble in generalnbsp;characters the Wealden species, but differ in some points of detailnbsp;which sufficiently separate the two forms.^ A comparison ofnbsp;Saporta�s pi. xcvii. figs. 1 and 3 with the present specimensnbsp;shows very clearly the different appearance presented by an upper

Md lower view of the same plant. I am inclined to regard some of the examples referred by Saporta to 0. latior as identical withnbsp;Heer�s 0. angustifolius, Heer; cf. Saporta, pi. xcvii. fig. 2, andnbsp;Ifeer, pi. ix. fig. 12, etc. The plant figured by Bartholni,' asnbsp;^tozamites latior, closely resembles the present species. A some-'^hat similar habit is also seen in Leokenby�s Otopteris lanoeolata :nbsp;the type specimen of this species appears to be identical withnbsp;Some other examples which have been assigned to Ctenophyllumnbsp;Poctinata; the base of the pinnae may be very slightly auriculate,nbsp;t'ot the specimens do not afford satisfactory evidence of this.

It will be seen from the two specimens (V. 2123 and V. 2360) figured in PL I. Pigs. 1 and 2, that the frond presents a verynbsp;different appearance in the lower and upper portions. The gracefulnbsp;Md tapering pinn�, with their slightly but distinctly auriculatenbsp;base (Fig. 4), and the grooved broad raohis are perhaps the mostnbsp;striking specific characteristics. The inclusion of these specimensnbsp;iQ Bunker�s species may, perhaps, be an error of judgment, butnbsp;I have endeav'oured to show on what grounds this course has beennbsp;taken. The institution of a new species would have been in somenbsp;Respects more satisfactory than defining afresh an old speciesnbsp;founded on specimens much less perfect than those now before us,nbsp;but, having recognized the strong probability that the apparentnbsp;differences between Bunker�s type and the English examples isnbsp;due to accidents of fossilization, the original name has beennbsp;rotained. We may thus define Oto%amites Goppertianus:�nbsp;Frond pinnate, deciduous; rachis broad, with a surface marked bynbsp;irregular longitudinal lines; pinnae alternate, articulate, attachednbsp;by a slightly auriculate base to the middle of the upper surfacenbsp;of the rachis; towards the base of the frond the pinnae are verynbsp;Harrow and short, and farther apart than those attached to thenbsp;Hiiddle and upper portion of the frond axis; in certain parts ofnbsp;frond the pinnae are approximate and almost imbricate. Lowernbsp;Pinnse very obliquely attached to the rachis, the upper graduallynbsp;becoming inclined at a greater angle to the axis in passing towardsnbsp;fbe apex of th� frond. Pinnae linear-lanceolate, very graduallynbsp;tapering from the base to the apex, occasionally somewhat falcate,nbsp;upices acute and slightly symmetrical, the lower margin of anbsp;�Hgment being sometimes decidedly curved in an upward direction.

the upper edge in some cases almost straight, l^arrow elliptical scars left on the surface of the raohis where the pinnee havenbsp;fallen off. Veins numerous, radiating outwards from the centrenbsp;of the base, but parallel in the greater part of the length ofnbsp;each pinnae.

V. 2360. PI. I. Pig. 2. Rachis S5'5cra. long; exceedingly well preserved at the base, which shows by its clearly cut outlinenbsp;the deciduous habit of the frond. Surface wrinklings or irregularnbsp;striations very distinctly marked on the raohis. The figurenbsp;represents only a portion of the specimen ; the upper part, notnbsp;shown in the drawing, is practically identical with the portion ofnbsp;V. 2123 figured in PL I. Fig. 1. The specimen as a whole showsnbsp;very clearly the great contrast between the loosely arranged, verynbsp;narrow, and short basal pinnee; the somewhat more approximatenbsp;and much longer, gradually tapering, and slightly falcate pinnaenbsp;in the middle of the frond; and the stiffer, more approximate,nbsp;and broader pinnse towards the apex. The lowest pinna is aboutnbsp;1 mm. broad, the broadest about 5 mm. Some of the pinn� havenbsp;a distinctly auriculate base. Cf. Dunker, pi. ii. fig. 5. Eooles-bourne.nbsp;nbsp;nbsp;nbsp;Huford Coll.

V. 2741. Pig. 4. Eachis 18 cm. in length. Part of the lower portion of a frond corresponding to V. 2360 (PI. I. Fig. 2). Thenbsp;gradual increase in the breadth of the pinn� in passing from thenbsp;lower to the upper part of the frond is well marked. Auriculatenbsp;bases very distinct, and venation fairly clear. Ecclesbourne.

V. 2123. PL I. Fig. 1. Eachis 10cm. long; the pinnae more closely arranged, and resemble on a rather larger scale Schenk�snbsp;figures of this species. The aurieulate character of the bases ofnbsp;the segments distinctly indicated, also the median groove in thenbsp;axis of the frond. Numerous fine veins distinct. Cf. the uppernbsp;Unfigured portion of V. 2360.

There is an imperfect specimen of this species in the Museum cf Practical Geology, Jermyn Street, in which the aurieulatenbsp;base and venation characters are fairly well shown.

In describing some Japanese examples of the Wealden cyca-dean frond named by Ettingshausen * Pterophyllum Buchianum, lathorst� draws attention to certain characters in the pinnmnbsp;^hich, he considers, exclude the species from Pterophyllum ornbsp;Igt;ioonites, in which it has generally been placed by other writers.nbsp;He finds that the segments are somewhat narrower towards thenbsp;point of insertion, and apparently laterally attached to the frondnbsp;^xis, thus differing in essential features from the above-namednbsp;genera. The splendid series of specimens in the Eufford Collectionnbsp;Confirms Nathorst�s remarks with regard to the narrowing of thenbsp;pinnae towards the point of attachment to the rachis, hut on thenbsp;other hand, the English examples show a manner of attachmentnbsp;of the pinn� more in accordance with the genus Zamites, thenbsp;pinnae being inserted on the upper surface of the rachis, andnbsp;apparently along two more or less distinct lines. The new genusnbsp;^miiophyllum, proposed by Nathorst for this species as a substitutenbsp;for Pterophyllum or Bioonites, on the ground that the charactersnbsp;do not conform in all essential respects with any existing genera,nbsp;cannot well be retained in the light of the more recent factsnbsp;afforded by the large series of English specimens. The generanbsp;bioonites and Pterophyllum are quite unsuitable for this species.

' (A. 4), Abli. k.-k. geol. Reichs. vol. i. 1852, p. 21, pi. i. fig. 1. ^ (A. 3), Denkschr. k. Ak. Wisa. vol. Ivii. p. 46.

and it would seem that Brongniart�s genus Zamites corresponds most closely in general characters with these large Wealden fronds.

A brief review of some of the numerous definitions proposed for Zamites may suffice to emphasize the generally acceptednbsp;characters, justifying at the same time Saporta�s pertinentnbsp;remarks as to the vague sense in which the name has frequentlynbsp;heen applied. In 1828 Brongniart * proposed Zamites for a fewnbsp;fossil fronds showing some points of difference from his othernbsp;genus Zamia; he mentions Z. Bechii, Brong., as an examplenbsp;of the former, and describes the pinnae as: � Se recouvrantnbsp;mutuellement et passant sur le petiole commun; nervuresnbsp;divergentes arqu�es, souvent bifurqu�es.�

The name Zamia is afterwards given up as likely to prove misleading as regards the relationship of fossil and recentnbsp;leaves, and Zamites is described as characterized by the possessionnbsp;of entire pinnae, not truncate at the apex, and not decurrent,nbsp;hut slightly constricted at the base. Brongniart includes Braun�snbsp;two genera Fodozarnites and Pterozamites as two subsectionsnbsp;of Zamites. Goppert^ uses Zamites in a wide sense, and notesnbsp;the resemblance of some species to the genus Encephalartosnbsp;among recent cycads. The similarity between various fossilnbsp;fronds and species of this living genus has not been sufficientlynbsp;recognized by most palseobotanical authors. Goppert�s definitionnbsp;does not restrict Zamites to fronds with basally constrictednbsp;pinnae, hut includes those with a swollen and auriculatenbsp;base. Pomel,� true to his unfortunate habit of founding newnbsp;genera, proposes Crossozamia for certain species of Zamites, butnbsp;the name has not come into general use. Bornemann* describesnbsp;Zamites as comprising species with a greater or less resemblancenbsp;to the recent Zamias, and defines the genus as follows: � Erondnbsp;pinnate, leathery, pinnae approximate or distant, from ovate tonbsp;slender and linear in form, contracted at the base, entire or toothednbsp;on the upper margin, blunt at the top. Veins of uniform size,nbsp;clearly seen on both sides of the pinnae, dichotomous. Epidermalnbsp;cells having the same structure as in recent Zamias.� Schenk*

2 (1), p. 122.

gives a diagnosis of Zamites in wiiich the veins in the pinnae are described as equal and parallel, and dichotomously convergingnbsp;towards the apex. He thus includes such forms as correspondnbsp;to Braun�s genus Podo%amites. In Zittel�s Mandhuoh' we find anbsp;closer approximation to the definition of Zamites as generally-accepted : the pinnae are described as attached to the upper surfacenbsp;of the raehis, and possessing a basal callosity, rounded or some-ivhat contracted at the base ; the median veins parallel, and thenbsp;outer veins diverging towards the apex of the pinna.

Solms-Laubaoh � notes the occurrence in the Zamitlt;B of pinnae ivhich articulate with and separate from the frond axis, pointednbsp;at the apex, abruptly rounded at the base, and attached obliquelynbsp;to the raehis, which they overlap and cover. Solms pertinentlynbsp;remarks that isolated segments of species of Zamites may easilynbsp;tie mistaken for portions of other fossil genera. Before attemptingnbsp;to modify in any way the definitions of Zamites, it may be wellnbsp;to consider some of those special features which have beennbsp;referred to by several writers, and to see how far such detailsnbsp;of frond structure are likely to serve as trustworthy guides. Asnbsp;regards the basal callosity usually included in definitions of thisnbsp;genus, we must acknowledge the great difficulty to be experienced in deciding definitely as to its existence in many fossilnbsp;leaves. In the process of fossilization the pinnae of a cyoadeannbsp;frond are often fiattened down against the rock, and closelynbsp;adpressed to the surface of the raehis, and thus there may henbsp;produced transverse wrinklings just above the point of attachmentnbsp;of the segments; such wrinklings may easily suggest in a fossilnbsp;specimen the original existence of a basal callosity. The venationnbsp;roay prove useful in determining certain species, hut it is notnbsp;Very often that fossil specimens are sufficiently well preserved tonbsp;admit of a complete diagnosis of the venation character. If, fornbsp;example, we had neither the basal nor the apical portion of anbsp;pinna, it would be practically impossible to discriminate betweennbsp;some of the long narrow Otozamites pinnae, and those of Zamitesnbsp;and other genera.

Nathorst�s recently proposed genus ZamiopJtyllum, to which reference has already been made, was founded partly to include

fronds of which the obliquely inclined pinnse have a somewhat narrowed base. This decrease in breadth towards the point ofnbsp;attachment of a pinna is characteristic of numerous species ofnbsp;Zamia, Ceratotamia, and Maerozaniia, and might well be includednbsp;in a more extended definition of Zamites. The basal contractionnbsp;of the segments is, as Nathorst points out, directly opposed tonbsp;the accepted definitions of Bioonites and Pterophyllum; hut therenbsp;seems no good reason to regard such a feature as at variaucenbsp;with the genus Zamites. This name Zamites was proposed bynbsp;Brongniart at a time when only two genera of living cycadsnbsp;had been recognized, Cycas and Zamia; and the present definitionnbsp;of the genus stamps it rather as a comprehensive and provisionalnbsp;designation for certain frond characters which are now sharednbsp;by various members of the recent Zamiem. If we retain Zamitesnbsp;as usually defined, or in a slightly modified form, it must benbsp;regarded merely as a convenient term to be applied to certainnbsp;fossil fronds in which some of the characteristics of Zamia.,nbsp;Maorozamia, Ceratozamia, or even Bncephalartos may be represented. In the recent Zamia the pinnse are articulated to thenbsp;rachis, and in many forms are readily detached, leaving a distinctnbsp;circular or elliptical scar; this same character is also met withnbsp;in other genera, such as Ceratozamia, Encephalartos, etc. Thenbsp;basal callosity often referred to as one of the important characteristics of Zamites is best seen in some forms of the genusnbsp;Maorozamia, and the manner of insertion and position of thenbsp;pinnm on the rachis in species of Zamites find a parallel livingnbsp;in species of Encephalartos, Ceratozamia, Maorozamia, and Zamia.

The following general definition of Zamites may serve to indicate those characters which are most readily recognized innbsp;fossil fronds ;�

Frond pinnate, pinnae more or less obliquely inclined to the rachis and attached to the upper surface, apices acuminatenbsp;and tapering, or obtusely rounded, the base may be abruptlynbsp;rounded and marked with a callosity near the point of attachment, or the pinnae may be slightly and gradually narrowednbsp;towards the base, margins entire; veins parallel, but slightlynbsp;divergent in the apical portion of each pinna.

Such a definition is perhaps suggestive of a genus with characters expressed in too general terms, and not sufficientlynbsp;limited, but a more complete examination of the different types

of fossil leaves may possibly lead to the institutiou of other genera With more narrowly defined characters.

Zamites Buchianus' (Ett.).

1847. Pterophyllum saxonicum, G�ppert, Nova Acta Ac. C�s. Leop.-Car. vol. xxii. p. 362, pi. xxxviii, fig. 13.

1852. Tterophyll um Buchianum, Ettingshausen, Abh. k.-k. geol. Eeichs. vol. i. Abth. iii. No. 2, p. 21, pi. i. fig. 1.

1856. nbsp;nbsp;nbsp;? Dioonites Buchianus, Bornemann, Organ. Rest. Lettenkobl. p. 57.

1867. Pterophyllum saxonicum, Ettingshausen, Sitz. k. Ak. Wiss. Wien, vol. Iv. Abth. i. p. 11, pi. i. figs. 11 and 12.

1879-80. ^Dioonites abietinus, Hosius and von Marck, Palseontographica, vol. xxvi. p. 213, pi. xliv. fig. 199.

Dioonites Buchianus, Fontaine, Potomac Flora, p. 182, pis. Ixviii.�Ixxiv. Dioonites Buchianus var. angustifolius, Fontaine, ibid. p. 185, pis.nbsp;Ixvii., Ixviii., and Ixxi.

Dioonites Buchianus var. obtusifolius, Fontaine, ibid. p. 184, pi. clxviii. fig. 3.

vol. Ivii. pp. 46 and 49, pi. ii. figs. 1 and 2 ; pi. iii. ; pi. v. fig. 2. Zamiophyllum Naumanni, Nathorst, ibid. p. 47, pi. v. fig. 1.nbsp;Zamiophyllum Buchianum, Yokoyama, Joum. Coll. Sci. Japan, vol. vii.nbsp;pt. iii. p. 223, pi. XX. fig. 1; pi. xxiii. fig. 6; pi. xxvii. fig. 5�, b;nbsp;pi. xxviii. figs. 1 and 2.

Zamiophyllum Buchianum var. angustifolia, Yokoyama, loc. cit. p. 224, pi. xxii. fig. 4 ; pi. xxv. fig. 5; pi. xxviii. figs. 8 and 9.nbsp;Zamiophyllum Naumanni, ibid. p. 225, pi. xxii. fig. 3 ; pi. xxvi.

Type. Portion of frond. Collection of Herr Hohenegger, Teschen.* Ettingshausen defined the species as follows ; � P. fronde pinnata,nbsp;pinnis circa 1-2 dm. longis, 4-7 mm. latis, alternis, linearibus,nbsp;�hbremotis, subangulo acuto adnatis, nervis creberrimis, tenuissimisnbsp;ihstructis; rhachide crassiuscula.�

' Possibly Zamites Milleri, Zigno, may prove to be identical with this species (Zigno, Flor. loss. Oolit. vol. ii. p. 40, pi. xxx. fig. 6; Hugh Miller, Testimonynbsp;�I the Rocks, p. 434, fig. 136).

80 at least 3-4 ft. in length, and compared with Pterophyllumnbsp;Uumboldti, Dunk.; the latter species, however, appears to be anbsp;typical Pteropliyllum species, and quite distinct generically from.nbsp;Ettingshausen�s type. Schenk reproduces the figure given bynbsp;Ettingshausen, and points out the fact that none of the pinnanbsp;apices are shown in the specimens. Very probably we may regardnbsp;Zamites Gopperti, Schenk, as Z. Buchiamis (Ett.), seen from thenbsp;lower surface; Schenk himself compared the former species withnbsp;Pterophyllum saxonioum, Reich.,' from Niedersohonen, and therenbsp;seems good reason to follow Fontaine in including P. saxonioum,nbsp;as figured by Ettingshausen, as synonymous with Z. Buohianum.nbsp;Hosius and von March ^ have figured a small fragment of a frond,nbsp;which they refer to P. saxonioum, but the specimen is too imperfect to admit of accurate identification. The same authorsnbsp;refer another specimen of cycadean frond to Bioonites alietinus,nbsp;which shows a distinct resemblance to the smaller forms of Zamitesnbsp;BucTiianus {of. Hosius and March, pi. xliv. fig. 199, and PI. III.nbsp;Fig. 1 of the present volume), but perhaps the similarity is hardlynbsp;sufficiently well marked to warrant the inclusion of the fossil innbsp;the synonym list of the present species, without the addition ofnbsp;a query. Some writers have preferred Bioonites to Pteropliyllumnbsp;for Ettingshausen�s species; Schimper and Fontaine both adoptnbsp;the former name. The Potomac flora has yielded numerousnbsp;examples of fronds which Fontaine refers to D. Buchianus-, henbsp;speaks of the species as � � one of the most widely diffused andnbsp;characteristic fossils of the Potomac flora.� � This author institutesnbsp;two varieties of Ettingshausen�s species�B. Buchianus, var.nbsp;oMusifolius and var. august ifolius. It might perhaps he advisable tonbsp;adopt Fontaine�s terms, and apply them to certain forms of thenbsp;species represented in the numerous examples from the Wealdennbsp;of Ecclesbonrne, hut there is the usual difficulty to be faced innbsp;drawing lines between one form and another. In looking at somenbsp;specimens we find the pinna apices are very distinctly acutelynbsp;tapered, and closely correspond with Fontaine�s B. Buchianus var.nbsp;angustifolius ; but in such a frond as V. 21737, although on thenbsp;whole the pinn� are tapered, yet some of the segments terminate

� Ettingshausen (A. 8), Sitz. k. Ak. Wiss. nbsp;nbsp;nbsp;vol. Iv. Abtli. i. pi. i.

ttiore obtusely, and lead us to specimens, such as V. 2227, in 'which -we have no longer the form characterized by graduallynbsp;tapered and pointed pinnae. A similar variation in the apices andnbsp;in the breadth of the pinnse is pointed out by Tokoyama' in thenbsp;�iapanese examples. The frond figured by Gdppert as Pterophyllumnbsp;^nrnallianum'^ shows a certain resemblance to some of the formsnbsp;nf the present species. Heer� has compared the present speciesnbsp;'With Zamia gloluliferus, Heer, from the Home beds of Greenland,nbsp;lieferenoe has already been made to Hathorst�s substitution of thenbsp;new genus Zamiophyllim for Miquel�s name Bioonites. He pointsnbsp;nnt that the species is from Urgonian strata, and not of Wealdennbsp;nge as Ettingshausen erroneously states. In referring to the basallynbsp;narrowed pinnse in the Japanese examples, Nathorst notes thatnbsp;the character is not apparent in the European specimens; we stillnbsp;�od, however, in the examples described from the Rufford Collectionnbsp;nbundant proof of this narrowing of the segments. The largestnbsp;pinna mentioned by this author has a length of 260 mm. andnbsp;n breadth of 4-6 mm.; in a pinna which is 3-5 mm. broad therenbsp;nre 12 veins, in one with a breadth of 6 mm. 17 veins, and 22nbsp;�''eins in pinnae 8 mm. broad. Hathorst compares Z. BucUanusnbsp;With the recent oycad Zamia media, Jacq. In describing a secondnbsp;�iapanese form, which he refers to another species, Z. Naumanni,nbsp;iie expresses the opinion that it may possibly represent the lowernbsp;Part of a frond of Z. Buchianus'^ �, it is compared with Zamitesnbsp;Schimp. {=Z. Gopperti, Schenk), from the Wemsdorfnbsp;i'eds. Tokoyama also figures under Hathorst�s species Z. Naumanninbsp;(Pl. xxii. fig. 3, and pl. xxvi.) a portion of a frond with pinnaenbsp;having a breadth up to 20 mm., but suggests that it may benbsp;Specifically identical with Z. Buehianus.

The English examples clearly demonstrate that the pinnse are attached to the surface of the rachis, and not laterally as Tathorstnbsp;inferred from his less perfect material. Z. Naumanni, Hath.,nbsp;in all probability identical with Z. BucUanus-, the specimensnbsp;if the former figured by Hathorst seem in some cases, e.g. pl. iii.nbsp;^nd pl. y. �g. 1^ siiow a surface attachment of the pinnae.

� (A. 3), FI. foss. Arct. vol. vi. pt. i. p. 12, pl. i. 1 (A. 3), Denksclir. k. Ak. 'Wiss. vol. ivii. p. 47.

In the specimen represented in pi. v. fig. 1 of Nathorst, there appear to he two elliptical soars about the middle of the frondnbsp;fragment; these agree very closely with similar soars in severalnbsp;of the Kutford examples, and represent the points of detachmentnbsp;of pinnae. The abundant examples in the National Collectionnbsp;enable us to extend and modify the existing definition of thenbsp;species. Among recent species Cerato%amia mexicana, Brong.,nbsp;Zamia oycadifolia, Jacq., Macrozamia Macleayi, Miq., and Encepha-lartos Lehmanni, Lehm., may be mentioned as possessing frondsnbsp;very similar in form to those of Z. Buchianus (Ett.).

Erond large, pinnate; raohis longitudinally striated, with a fairly broad median groove on the upper surface. Pinnaenbsp;alternate, opposite or subopposite, varying in length from 3-4 cm.nbsp;to 20 cm., and in breadth from 1'5 to 2 mm., attached obliquely tonbsp;the rachis, and slightly thickened and somewhat broadened at thenbsp;actual point of insertion; separated from the raohis by a distinctnbsp;absciss line, leaving an elliptical soar; generally narrowed towardsnbsp;the base, but in the narrower pinnae this reduction in breadthnbsp;decreases and is not nearly so evident; usually inclined at aboutnbsp;45� to the raohis, they may be almost at right angles to the frondnbsp;axis, and in the case of young fronds and the apical portionsnbsp;of larger ones, the pinnae are attached at a much more acute angle.nbsp;The distance between the pinnae varies considerably in differentnbsp;parts of the same leaf, and in leaves of different ages; apicesnbsp;generally tapering to a point, or more or less obtusely rounded.nbsp;Yeins numerous, parallel, and as a rule not very prominent.

V. 2120. PI. VIII.¹ ² Fig. 1 and PI. III. Fig. 5. This splendid specimen has a length of 7 7'5 cm.; broken at one point, andnbsp;somewhat displaced laterally. Breadth of pinnse varies from 2 tonbsp;9 mm. Attachment of the pinnae not shown in this example.nbsp;The broader and lower pinnae appear to taper towards the rachisnbsp;more than in many specimens. Tips of some of the pinnae clearlynbsp;seen, as in PI. III. Fig. 6 ; these agree very closely with thenbsp;apices of pinnae in V. 2123 (PI. III. Fig. 3). Certain parts ofnbsp;this frond agree exactly with V. 2925, etc. The apical portionnbsp;appears to be identical with Bioonites Buchianus var. angustifolia,nbsp;as figured by Fontaine in his pi. Ixx. fig. 2.² The lower portion

corresponds to that of V. 2123�, etc. Veins numerous, but not ^ell marked. Towards the upper end of the raobis there arenbsp;oiliptioal scars marking the original places of insertion of detachednbsp;segments. If V. 2120 and V. 2898 be compared, the most strikingnbsp;�difference seems to be the less tapered and narrowed bases of thenbsp;pinnse in the latter specimen. Ecolesbourne.nbsp;nbsp;nbsp;nbsp;Mufford Coll.

V. 2227. PI. III. Fig. 4. Large specimen, with unusually d'road pinnse; length of rachis 38 cm., width l�2cm.

due attachment of the segments distinctly shown, also the future of the base and the position of the pinnse on the axis,nbsp;j ctaee of the rachis marked with longitudinal strirn; aboutnbsp;ttm_ fj-Qjjjnbsp;nbsp;nbsp;nbsp;jjiargin of the rachis there is a distinct longi-

inally running line. Pinnse long, narrow, and slightly curved ; Cue row attached obliquely by tapering bases to the upper surfacenbsp;c the frond axis, alternate in position. In the lower part of thenbsp;^^oud the pinnae have a length of about 19 cm., and a breadthnbsp;. t�Scm.; their distal ends are bluntly acuminate, as shownnbsp;^u Pi. UP j'-g Very little, if any, indication of a basalnbsp;osity on the segments. There is only a slight difference innbsp;between the upper and lower pinnae. Cf. V. 2820. Eccles-

j V. 2820. A very similar specimen. Eachis 29cm. long; the ^Cgest pinna between 17 and 18 cm. in length and 1 cm. broad,nbsp;j cWards the upper end of the frond the pinnse are about 10 ornbsp;f cm. long; the apices are not shown, and therefore the actualnbsp;Cgth can only be approximately measured; breadth of thesenbsp;P^uuse about 8 mm.nbsp;nbsp;nbsp;nbsp;Ecolesbourne.nbsp;nbsp;nbsp;nbsp;Ruffordnbsp;nbsp;nbsp;nbsp;Coll.

V. 2125�. Eachis 21 cm. long. The upper portion closely Resembles Rioonites ahietinus as figured by Hosius and von Marck.nbsp;Ecolesbourne.nbsp;nbsp;nbsp;nbsp;Ruffordnbsp;nbsp;nbsp;nbsp;Coll.

V. 2128. Portion nbsp;nbsp;nbsp;ofnbsp;nbsp;nbsp;nbsp;anbsp;nbsp;nbsp;nbsp;rachisnbsp;nbsp;nbsp;nbsp;of a large frond; the bases of

the very distinctly and gradually tapering terminations of the 'w pmnjg �jyhich are preserved. Compare the apex of one ofnbsp;CSC pinnae, with those of the two shown in PI. III. Fig. 5

(V. 2120), and the less pointed form of V. 212Zc (PL III. Pig. 3). Ecolesbourne.nbsp;nbsp;nbsp;nbsp;RufforAnbsp;nbsp;nbsp;nbsp;Coll.

V. 2123lt;�. PI. III. Pig. 3. At the upper end of the specimen there is part of a rachis with five pinnae attached; the rest of thenbsp;slab shows several imperfect pinnae with gradually tapered tips;nbsp;the tips seem to he intermediate in form between those ofnbsp;V. 2227 (PI. III. Pig. 4) and V. 2363 (PI. III. Pig. 2),nbsp;V. 2123, etc. Ecolesbourne.nbsp;nbsp;nbsp;nbsp;Euffordnbsp;nbsp;nbsp;nbsp;Coll.

V. 2125J. PI. IV. Pig. 1. A young frond, 25cm. long; cf. V. 2123J, etc. Pinnas approximate, very gradually taperingnbsp;distally. The pinnae are broadest in the middle of the specimen,nbsp;and narrower towards either end; at the lower end the pinnaenbsp;are narrower and less closely arranged, agreeing with those ofnbsp;V. 2262 (PI. III. Pig. 1). Compare also V. 2898 and Pontaine�snbsp;pi. Ixxi. fig. 2. Ecolesbourne.nbsp;nbsp;nbsp;nbsp;RufforAnbsp;nbsp;nbsp;nbsp;Coll.

V. 2125c. PI. IV. Pig. 2. Eachis 33 cm. long. Pinnae alternate and opposite, rachis depressed, the points of attachmentnbsp;of segments clearly shown, apical portions not preserved ; broadestnbsp;pinna 7 mm., narrowest 5 mm. Venation distinct. Cf. V. 2123lt;;,nbsp;V. 2125a, etc. Ecclesbourne.nbsp;nbsp;nbsp;nbsp;RufforAnbsp;nbsp;nbsp;nbsp;Coll.

V. 2123a. PI. IV. Pigs. 4 and 5. 35 cm. long. Eachis well shown, the longitudinal irregular striations distinct; these probably indicate strands of hypodermal solerenchymatous tissue.nbsp;Tips of pinnae not seen, but bases very distinct; in Pigs. 4 and 5nbsp;the surface and oblique insertion is easily recognized, also in somenbsp;pinnm the very slightly broadened base. In the lower part ofnbsp;the frond the pinnae are nearly at right angles to the rachis, butnbsp;more oblique towards the upper end. Veins numerous and fine,nbsp;and in many pinnae clearly shown. Ecclesbourne. RufforA Coll.

V. 2925. PI. IV. Pig. 3. Specimen very similar to V. 2125a; hut pinnae farther apart, and in this respect identical with thenbsp;lower portion of V. 2123a. In addition to the larger example,nbsp;there is a very small fragment on one side of the slab, which isnbsp;evidently the tip of a frond (PI. IV. Pig. 3); in this the segmentsnbsp;are closer together and more oblique to the rachis than innbsp;V. 2125i, and more like V. 21233 and V. 21253. Ecclesbourne.

V. 2262. (PI. III. Pig. 1.) This agrees fairly closely with J^ioonitea aUetinus (G�pp.) as figured by Hosius and von March¹;nbsp;probably their specimen is not a true example of G�ppert�snbsp;species. Small piece of raohis with long and narrow pinnae ; modenbsp;of insertion well seen. Cf. 21255; breadth of pinnas about thenbsp;same, but in the present specimen the segments are farther apartnbsp;and more spreading; this difference, however, only applies to thenbsp;*ipper three-fourths of 21255; in the lower fonrth it is practicallynbsp;identical with V. 2262. This and some other specimens appear to

identical with B. Bmhianua var. anguatifolia as figured in the J^otomae Flora, and with some of Tokoyama�s specimens, e.g.nbsp;Pl- xxii. fig. 4, pi. XXV. fig. 5, and pi. xxviii. figs. 8 and 9.²nbsp;V. 2898, V. 1069, and the upper part of 2125a. Longestnbsp;pinna 7�5 cm., and 3 mm. broad. Ecclesbourne. Rujford Coll.

V. 720. Two specimens. Fragments of a young frond; pinn� V'ery obliquely set and approximate. Hastings. Bawaon Coll.

V. 1069. Small specimen, probably the lower part of a frond, tile pinn� being much narrower in the lower than in the uppernbsp;portion ; manner of attachment of segments distinct. Cf. V. 2262,nbsp;V. 2125�, V. 21235, etc. Ecclesbourne.nbsp;nbsp;nbsp;nbsp;Rujford Coll.

V. 2123. Very imperfect example. Bases of several pinn� show clearly the manner of attachment to the raohis; at the basenbsp;of some of the pinn� a wrinkled appearance suggests the existencenbsp;of a callosity. There is not quite the same gradual narrowing ofnbsp;the pinn� towards the frond axis as in V. 2227 and many othernbsp;specimens. Pinn� long, and tapered to an acuminate tip, as innbsp;V. 2363 (PI. III. Fig. 2), etc.

V. 21235. Eaohis about 27 cm. long; pinn� 4 mm. broad, alternate ^od approximate. Cf. V. 2898. The lower part of this specimennbsp;O-Ppears to be identical with the upper portion of V. 2125, also with

V. 2123(5. Part of a single pinn�, showing sharply pointed apex; �Venation not distinct. Ecclesbourne.nbsp;nbsp;nbsp;nbsp;Rujford Coll.

V. 2125. Rachis 19 cm. in length. Longest pinna 12 cm., gradually tapering towards the tip, which is not present. Thenbsp;long, narrow, and tapering segments of this specimen are exactlynbsp;the same as those in the smaller examples, V. 2262 and V. 1069.nbsp;Cf. Fontaine,¹ pi. Ixxiv. figs. 1-3, etc. Ecclesbourne.

V. 2373�. Broad rachis, 1'2 mm. at the lowest end, hut obviously flattened. Groove in the middle, and on each side of the groovenbsp;there are two elliptical scars marking the places to which segmentsnbsp;were attached; some of the segments appear to have a basalnbsp;callosity; pinnse opposite or subopposite. Cf. Nathorst�s figurenbsp;(PI. V. Fig. !�)'¹ of the rachis of what he calls Z. Naumanni.nbsp;Ecclesbourne.nbsp;nbsp;nbsp;nbsp;Ruffordnbsp;nbsp;nbsp;nbsp;Coll.

V. 2698. Very large specimen, but the details not well preserved. Rachis 43cm. long; pinnffi of the gradually taperednbsp;form, as in V. 2363, etc. The upper portion the same as V. 2898,nbsp;V. 2125^, etc. Breadth of uppermost pinna 1�5-2mm.; that ofnbsp;the lowest 1 cm. This appears to be a frond seen from the lowernbsp;surface, the pinn� being, therefore, apparently laterally attachednbsp;to the rachis.nbsp;nbsp;nbsp;nbsp;Ecclesbourne.nbsp;nbsp;nbsp;nbsp;Ruffordnbsp;nbsp;nbsp;nbsp;Coll.

V. 2898. A fine example of what is probably a partially expanded frond. Rachis about 20 cm. long. Some pinnse aboutnbsp;14cm. in length and 3-4mm. broad; towards the tip the pinnaenbsp;are crowded and imbricate, with a width of about 3 mm.; towardsnbsp;the base 6 mm.nbsp;nbsp;nbsp;nbsp;broad.nbsp;nbsp;nbsp;nbsp;Sussex.nbsp;nbsp;nbsp;nbsp;Bechle�nbsp;nbsp;nbsp;nbsp;Coll.

Cf. the tip of this and V. 2120, V. 2125J, etc. V. 3153, V. 3154, V. 3156, and V. 3157. Other specimens of this species.nbsp;Ecclesbourne.nbsp;nbsp;nbsp;nbsp;Ruffordnbsp;nbsp;nbsp;nbsp;Coll.

Zamites Carruthersi, sp. nov.

Frond pinnate; rachis longitudinally striate; pinnse alternate, attached by an oblique base to the outer part of the upper surfacenbsp;of the axis, almost at right angles to the rachis in a fully

�developed frond, linear or linear-elliptical, somewhat abruptly Darrowed towards the base, but slightly broadened at the actualnbsp;base of attachment, separate from the rachis by a well-markednbsp;articulation. Veins divergent from the base, hut for the mostnbsp;part parallel, and diverging at the tip of the segments, which isnbsp;bluntly rounded.

Id. the examination of the fossil fronds in the National Collection, and the comparison of them with the leaves ofnbsp;existing species, 1 have often been struck with the close correspondence between certain forms of Wealden fronds and speciesnbsp;of the genus Encephalartos. An inspection of a thoroughlynbsp;J�epresentative collection of the different forms of Eneephaldrtosnbsp;fronds, such as are to be found in the exceptionally goodnbsp;collection of living and herbarium specimens at the Hoyalnbsp;Cardens, Kew, tends to very considerably widen one�s conceptionnbsp;as to the characters of this recent genus of cyoads. Thenbsp;species generally met with in collections are those in whichnbsp;bbe pinnae are broad and more or less spiny; but the genusnbsp;^Deludes various other forms with pinnae of quite a differentnbsp;form, which often hear a striking resemblance to various fossilnbsp;fronds. As examples of the diversity of leaf form to be metnbsp;^ith in this genus, the few following species may be cited asnbsp;representatives of some of the forms assumed; E. Caffer, Miq.,nbsp;'^ith its stout oval pinnae, with or without marginal teeth,nbsp;pungens, Lehm., with long, narrow, and acuminately terminatednbsp;segments, E. cycadifolius, Lehm., and E. GJiellinchii, Lem. (PI.nbsp;XlII. Figs. 3�5), characterized by the long and narrow pinnae,nbsp;^hich form a striking contrast to the broader and better knownnbsp;Segments of E. horridus, Lehm., and other species. Some of thenbsp;species, e.g. E. Lehmanni, Lehm., E. cyeadifohus, Lehm., and othersnbsp;bave a close resemblance to some species of Cerato%amia and Zamia.

In the introductory remarks mention was made of the institution by Fontaine of a new genus, Encephalartopsis, for certain isolatednbsp;pinnae from the Potomac beds of North America; and snob a genusnbsp;�ight serve a useful purpose if founded on more satisfactorynbsp;tnaterial, but as at present defined it can have but little value.nbsp;Possibly the institution of such a genus, with a wide and modifiednbsp;definition, might prove a valuable addition to our list of fossilnbsp;genera, hut for the present it will perhaps be better to fall hacknbsp;en the old and comprehensive Zamites. I have ventured to

connect the name of Mr. Carruthers with the present species of cycadean frond, as a slight recognition of his valuable contributionsnbsp;to our knowledge of the fossil Cycadacete.

Some of the large detached pinnse figured by Pontaine as examples of his new species Zamites tenuinervis, agree fairly closelynbsp;with those of the present form, but in the Potomac plant thenbsp;venation appears to be coarser, and the bases of the segmentsnbsp;usually �abruptly subeordate� ; in the pinna shown in Fontaine�snbsp;pi. Ixxvi. fig. 7 the base seems much more like that in the Englishnbsp;specimen. The paucity and imperfect character of the Potomacnbsp;material, and the differences already alluded to, hardly warrantnbsp;the adoption of Fontaine�s name for the English forms. There is anbsp;close resemblance between the present specimens and some of thosenbsp;referred to Z. Buohianus, e.g. V. 2123, hut in the latter speciesnbsp;the longer and more gradually tapering pinnae are sufficientlynbsp;characteristic to distinguish the two forms. Among recent cycads,nbsp;Encephalartos longifolius, Lehm., is one of those which resemblenbsp;very closely in habit the fronds of Z. CarrutJiersi, As examplesnbsp;of other fossil fronds to be compared with this species, Zamitesnbsp;s, Schenk,' and Palmmmia recta, Tate,* may be mentioned.

In this specimen the manner of attachment of the pinnae is clearly shown; the line of separation being particularly distinctnbsp;at the base of the middle pinna of the portion of frond representednbsp;in Fig. 4, PI. VI. Eachis at least 1 cm. broad, and marked withnbsp;fine longitudinal lines. Venation very distinct, as in V. 2123(7.nbsp;Only a portion of the specimen shown in the figure ; rachisnbsp;21 cm. long. Ecolesbourne.nbsp;nbsp;nbsp;nbsp;Eufford Coll.

The pinnae are very like those with blunt apices which have been included in Z. Buchianws {e.g. V. 2227), but in the presentnbsp;specimen the base and manner of attachment of the pinnae constitute the special features. The form of base clearly seen innbsp;�ig. 3, and the blunt apex with the slightly divergent veins innbsp;Fig. 2. Eachis in this specimen 13 cm. in length, with portionsnbsp;of nine pinnae on one side. Ecclesbourne.nbsp;nbsp;nbsp;nbsp;Rufford Coll.

The specimens referred to as Otozamites Klipsteinii var. longi-Mia should he compared with Zamites Garruthersie.g. V. 2123�, ^� 2122, and V. 2122S ; hut in those and similar specimens thenbsp;pinnae have a more or less distinctly auriculate hase. V. 21233.nbsp;Single pinna. Cf. V. 2123(f. Eccleshourne.nbsp;nbsp;nbsp;nbsp;Rufford Coll.

Specimens of Doubtful Position.

V. 2742. PI. VI. Fig. 1. This may perhaps he the terminal portion of a frond of Z. Garruthersi, or possibly of Otozamitesnbsp;^�klipsteinii var. longifolia, hut it is very difficult to feel any greatnbsp;confidence in placing it in such a position. Eccleshourne.

V. 2744. Fig. 5. This specimen and V. 2743� suggest portions cf a frond very similar to Ctenis falcata, L. and H., hut nonbsp;indication of anastomosing veins has been detected in the pinnaenbsp;of these quot;Wealden examples. The portion of frond shown hasnbsp;^ length of 13'5cm.; one of the broadest pinnae is 7 mm. innbsp;fireadth, and is traversed by about ten veins (Fig. 5a). To somenbsp;extent the specimen reminds one, as regards general habit, of

Zamites Buchianus, but the venation and decurrent pinnse are distinctive features in the former, and the pinnae are morenbsp;oblique than in Ettingshausen�s species. Ecclesbourne.

V. 2275. Another terminal piece of frond, agreeing closely with V. 2743, and very possibly the same species, but thenbsp;specimen shows no details, being merely a brown stain on thenbsp;surface of a coarse grit. It is possible that these two specimensnbsp;may belong to the terminal portion of a frond of which thenbsp;older and larger segments are shown in PI. YII. Eigs. 1, 4, and 6nbsp;(V. 2122, V. 2126�). Ecclesbourne.nbsp;nbsp;nbsp;nbsp;Bufford Coll.

V. 2743fl!. In some respects not unlike Hathorst�s figure of Zamiophyllum Naumanni, a species of Japanese frond now referrednbsp;to Zamites Buchianus. The present specimen is in all probabilitynbsp;part of a frond seen from the under side. Cf. Ctenophyllum lati-foliwn, Eont., a plant which Eontaine^ refers for no very obviousnbsp;reason to Schimper�s genus Ctenophyllum. V. 3183. Fragment ofnbsp;the same form. Ecclesbourne.nbsp;nbsp;nbsp;nbsp;Bufford Coll.

In discussing the genus Nilssonia mention was made of Sohimper�s genus Anomozamites, which he instituted for certainnbsp;Pterophyllum-YiVe leaves possessing the following characteristics :nbsp;� Eolia speciosa, mediocria, vel parva, elongata - oblonga velnbsp;elongato-linealia, pinnatisecta, hie illic (juniora) integra, nervisnbsp;rhachi perpendicularibus, parallelis, simplicibus vel e basi dicho-tomis; pinnis insequalibus, rectangulis, membranaoeis vel tenui-coriaceis.�

No mention is made in this definition of the place of insertion of the segments, whether lateral or on the upper surface of thenbsp;leaf axis. Nathorst has since given special prominence to thenbsp;manner of attachment of the segments as the chief distinguishingnbsp;character between the present genus and Nilssonia; � the samenbsp;author has also instituted a new genus for the reception of

�^fgt;�0mo%amites-Yik6 leaves 'witli dichotomizing veins. quot;We must, then, somewhat modify Sohimper�s original diagnosis, and thenbsp;following may be adopted as a rough definition of this provisionalnbsp;genus Ammozamites; it is a slightly altered version of that innbsp;Zittel�s Handhuch,}

Frond comparatively small, linear or tongue-shaped, and usually divided into segments which present a more or less obviousnbsp;difference in size, separate or confluent at the base, attachednbsp;laterally to the rachis, and never entirely covering the uppernbsp;face of the frond axis; the segments bluntly rounded or truncatenbsp;distally; veins simple and parallel, generally at right angles tonbsp;fhe rachis.

The examples of this genus possess, as a rule, a characteristic habit which marks them off from the pinnate fronds of Ptero-P^yllum with their equal and longer segments. It is diificult innbsp;�onie cases to distinguish between the genera Anomozamites andnbsp;^ilssonia. In the former the segments are sometimes attachednbsp;to the rachis in such a manner as to suggest the surface insertionnbsp;of Nihsonia,, but there is always some part of the frond axisnbsp;opposed to view, whereas in Nilssonia the lamina appears to benbsp;Continuous from one side to the other. It is not easy in thenbsp;present instance to decide whether the genus Pteropliyllum ornbsp;�^nowiozamites is the more suitable; both are purely provisionalnbsp;genera, and it is not a matter of very great importance whichnbsp;form is adopted.

Anomozamites Lyellianus (Dunk.).

1846. Fterophyllum Lyellianum, Duiiker, Wealdenbildung, p. 14, pi. vi, figs. 1 and 2.

1848. nbsp;nbsp;nbsp;Pterophyllum Lyellianum^ Bronu, Index, pal. noTnencl. p. 1056.

1850. nbsp;nbsp;nbsp;Lterophyllum Lyellianum^ Unger, Gen. spec, plant, foss. p. 290.

1851. nbsp;nbsp;nbsp;Dioonites lyellianus, Miquel, Tijdacli. Wis. nat. Wet. iv. p. 205.

1852. nbsp;nbsp;nbsp;Plerophyllum Lyellianum, Ettingshausen, Abh. k.-k. geol. Eeiobs.

1870. nbsp;nbsp;nbsp;Dterophyllum Lyellianum, Schimper, Trait, pal. v�g. vol. ii. p. 137.

1871. nbsp;nbsp;nbsp;Dterophyllum Lyellianum, Schenk, Palaeontograpbica, vol. xix. p. 230,

� Pterophyllum fronde pinnata, pectiniformi, pinnis oppositis linearibus aeque distantibus, approximatis basi fere confluentibus,nbsp;apice obtusis, angulo recto adnatis, nervis iv. vel v. tenerrimis,nbsp;rhaohi plana subsulcata.�

The figure of Bunker�s type specimen is very much more suggestive of the genus Pterophyllum, than are the drawings given by Schenk, or that of the solitary specimen in the Rutford Collection.nbsp;Probably the original specimen is part of a frond seen from thenbsp;under surface, thus showing a particularly prominent raohis. Itnbsp;may be that the English fossil should be placed in a new species,nbsp;but the apparent differences which distinguish it from Bunker�snbsp;specimen may be merely such as are the result of a more fullynbsp;developed condition of frond in the latter case. Possibly Zamitesnbsp;mqualis, Bunk.,� should be included in this species. Schenk�snbsp;figure agrees more closely with the English specimen: this authornbsp;speaks of the segment as being attached to the upper surface ofnbsp;the raohis, and not laterally inserted ; if this were really the case,nbsp;the genus Pterophyllum as usually defined would be inapplicable.nbsp;Schenk�s example does not show the raohis sufficiently clearlynbsp;to definitely settle this point, but in all probability, as in ournbsp;specimen, there is a narrow line of axis separating the two rowsnbsp;of segments. A close inspection of Schenk�s figure enables usnbsp;to detect certain slight differences in the breadth of the pinnse,nbsp;similar to those in the Ecclesbourne specimen. It must benbsp;remarked, however, that there is very little difference in thenbsp;breadth of the several segments. There is a striking agreementnbsp;as regards general appearance and arrangement of the segments,nbsp;between the Wealden specimen and the Jurassic species Anomo-%amites Nilssoni. Bindley and Hutton^ figured this plant asnbsp;Pterophyllum Nilssoni (Phill.); from their figure it is not easy tonbsp;decide between Nihsonia or Ammozamites as the most suitablenbsp;genus, but an examination of several specimens of this speciesnbsp;from the Yorkshire coast, shows very clearly the characteristicsnbsp;of the latter genus. There are several examples of this form in

the Leckenby Collection, wbioh have been examined by Natborst ^�nd referred by him to Anomozamites.

In addition to the single specimen of Anomozamites Lyellianus in the British Museum, there is a somewhat larger example iu thenbsp;Museum of Practical Geology, Jermyn Street. In this latternbsp;specimen the pinnse are broader and have a more open arrange-nient. The lateral attachment is very clearly shown, and the fournbsp;or five veins in each segment are distinctly marked. Here andnbsp;there may be noticed slight differences in the breadth of thenbsp;Segments, which are arranged alternately towards the upper andnbsp;Wer ends of the specimen, but in a few cases the pinnae arenbsp;opposite. The Jermyn Street specimen is from the Wealden ofnbsp;tire near Hastings.

frobably a young leaf, showing clearly a gradual diminution rn the length of the segments towards either end of the rachis.nbsp;Manner of attachment and venations of the pinnm clearly preserved ; each segment appears to have four or five simple,nbsp;parallel, and distinctly marked veins as shown in Pig. 6�. The

segmeuts appear to be much thicker than in the similar Jurassic form A. Nilssoni, and their breadth is much more uniform than innbsp;the latter species.

A species of a somewhat analogous habit has recently been described by Fontaine under the name of Zamites Montanensis fromnbsp;the Montana Coal-field.¹ ² Ecclesbourne.nbsp;nbsp;nbsp;nbsp;Rujford Coll.

In 1875 Saporta proposed the term Cyeadolepis as a convenient generic designation for detached bud scales of cycadean fronds.nbsp;He defined it as follows : � Squam� coriacese basi dilatatee loconbsp;insertionis crassse facie interiori plus minusve conoavse nudseque,nbsp;facie autem dorsali convexiusculse, sursum elongat� lanceolato-acuminatee, extus ad utrumque latus tomento piloso donatse.�

A small number of fronds have been recorded in which larger or smaller basal scaly structures are preserved ; as a few examplesnbsp;of such, we have Zamites gigas, Morr., as figured by Saporta,²nbsp;Podozamites distans, Presl,² Otozamites sp.,² Podozamites lanceolatusnbsp;minor, Schenk,� etc. Among recent cycads we have, in additionnbsp;to the ordinary pinnate fronds, various forms of smaller scalenbsp;leaves; the latter are particularly well seen in Cgcas, where annbsp;old stem shows a clearly marked alternation of the persistentnbsp;basal portions of fronds alternating with the bases of scale leaves.nbsp;These scale leaves are true leaf structures, in which the greennbsp;assimilating portion of the phyllopodium has not been developed.

1 (A. 3), Proc. TT.S. Nat. Mus. vol. xv. p. 494, pi. Ixxxiv. fig. 4. Fontaine omits to mention that Dawson has described a fragment of cycadean frondnbsp;from the .Kootanie series of the Eocky Mountains as a new species, Zamitesnbsp;Montana, a name dangerously near to Z. Montanensis (see Dawson, Trans. Eoy.nbsp;Soc. Canada, 1885, section iv. p. 7, pi. i. fig- 6).

^ Ibid. pi. Ixxvi. figs. 3 and 4 [see also Zigno (A.), Flor. foss. Oolit. vol. ii. pis. XXIV. and xxxvi.].

In the genus Ceratozamia the expanded base of a frond shows two lateral stipule-like appendages, and these are also found innbsp;Ihe same position on the margin of the smaller scale leaves.' Innbsp;Some forms of Macrozamia, the surface of an old stem is entirelynbsp;enclosed in a thick armour of large persistent petiole bases withoutnbsp;^�ny accompanying scales. Many recent cycadean stems appearnbsp;to be covered by numerous scale-like structures of identical form;nbsp;It is by no means an easy task in many oases to distinguishnbsp;between the bases of true fronds and those of scale leaves, evennbsp;''^kere both forms of leaf are present. The variation in formnbsp;nnd size exhibited by the scale leaves of recent species, sufficientlynbsp;demonstrates the futility of attempting any exact generic ornbsp;specific discrimination in the case of the isolated fossil examples.nbsp;It is true we have in Dioon and Cycas fairly characteristic lanceolate scales, often clothed in a dense woolly covering; but a closenbsp;inspection of a tall stem of the latter genus reveals a markednbsp;difference in the scale leaves towards the apex of the stem, andnbsp;those in the older portions, where there are only the persistentnbsp;liioad bases adhering to the plant stem. In dealing with fossilnbsp;�nale leaves it will probably be wise to extend the definitions ofnbsp;Saporta�s genus and to include in it not merely the � elongatenbsp;lanceolate-acuminate� forms of bud scales, but also other formsnbsp;true scale leaves, as well as those structures which may benbsp;^ngarded as the persistent bases of petioles. This genus, usednbsp;a much more comprehensive sense, should afford a convenientnbsp;ttieans of grouping together those detached leaf structures, whichnbsp;frequently cannot be definitely referred to any particular genusnbsp;species. Such isolated plant members, in themselves, perhaps,nbsp;of little value, are worthy of record as contributions to thenbsp;t�aterial from which to build up a more complete history ofnbsp;fossil cycadean plants. Some of the numerous scales in thenbsp;HufEord Collection may, indeed, be referred to certain forms ofnbsp;�tems, and no doubt as our material is increased others maynbsp;ke recognized as portions of some well-defined genus or speciesnbsp;�f cycad.

In the male and female flowers of some recent species the �fetached scales bear a close resemblance to the sterile leafnbsp;structures of the stem; it will be well, therefore, to include

in the genus Cycaiolepis such scale leaves as afford no clear proof of their carpellary or antheriferous nature. As an example of suchnbsp;a resemblance, it may be noted that some of the smaller examplesnbsp;of the Ecclesbourne scales present a distinct agreement in shapenbsp;and size with the detached carpellary scales of such a form asnbsp;Macrozamia Byeri. Similarly the isolated leaves of the � bulbils � ^nbsp;of Cycas and other genera, erroneously compared by some withnbsp;the inflorescences known as Bennettites and Williamsonia, shouldnbsp;be included under this comprehensive genus; also the narrownbsp;lanceolate-acuminate and short broad leaves of the two latternbsp;genera.

The only forms included by Saporta in Cycadolepis, e.g. C. villoaa and C. hirta^ are narrow leaf-like structures similar to thenbsp;scales of Bioon or Cycas. Eliche and Bleicher � have adoptednbsp;Saporta�s name for a very imperfect fragment which seems tonbsp;be practically indeterminable. We may, perhaps, as a matter ofnbsp;convenience, and to avoid the obvious danger and inexpediencynbsp;of instituting several more or less meaningless specific names,nbsp;arrange the various detached scales under two main heads,nbsp;basing the distinction of the two sections on the general formnbsp;of the scales.

Cycadolepis.

Scale-like leaf structures of cycadean plants, varying considerably in form and including detached petiolar bases, bud scales, etc., also isolated carpellary or antheriferous scales which exhibitnbsp;no trace of ovules or pollen-sacs.

less linear-lanceolate form, broadest at the base and tapering gradually towards the apex.

with the broadest portion frequently nearer the distal than proximal end ; thick and fibrous structures.

� (A.), Bull. Soc. Sci. Nancy [2], vol. v. figs. 9-11, p. 76, * 8lt;ipu = spear.

1.�Dory- Cycadolepis.

The forma included in this section of Cycadolepis are practically �uch as Saporta describes in his diagnosis of the genus. Saporta�snbsp;examples are regarded as distinct species, hut it is surely unnecessary to institute elaborate specific definitions for such isolatednbsp;structures, and especially as the so-called species bear a distinctnbsp;Resemblance to one another. Both forms are from the Lowernbsp;^immeridgian of the province Ain. C. villosa is compared withnbsp;'-he scales of Stangeria, and C. hirta with those of Cgcas andnbsp;^ioon. It is by no means unlikely that both may belong tonbsp;^illiamsonia; similar scales figured by Fontaine in a specimennbsp;ef Williamsonia virginiensis, Font.,' from the Potomac beds, andnbsp;�oiHe of the Wealden Williamsonia scales both present a strongnbsp;likeness to the French specimens, and suggest the possibilitynbsp;generic identity. A specimen in the Rufford Collectionnbsp;(V. 2830), consisting of a collection of narrow acuminate hairynbsp;Scales, which is probably part of a cycadean stem, shows somenbsp;scales very similar to G. hirta. Sap. Feistmantel has figured anbsp;specimen of what he calls C. pilosa'^ from the Gondwana floranbsp;India which agrees very closely with Saporta�s examples,nbsp;also with the leaves of W. virginiensis, Font. The same authornbsp;�gures another form as Cycadolepis,^ which may perhaps benbsp;^�8arded as a cycadean scale. The specimen figured by Fathorstnbsp;Gycadospadix integer angusiior, Nath.,¹ ² suggests a form whichnbsp;^'ght be included in the present section of Dory-Cycadolepis.

2802. Single leaf; linear-lanceolate in form; 7'5 cm. long, ^ mm. broad, delicate hairs on each margin. Appears to benbsp;Identical with Saporta�s � species � C. villosa. Compare V. 2129,nbsp;'Which there are several similar scales aggregated together.nbsp;^�ry likely this specimen may be a detached leaf of TVtlliamsonia.nbsp;k'urther reference will be made to this form of scale in thenbsp;descriptions of cycadean stems and some of the specimens of

� nbsp;nbsp;nbsp;-r-,nbsp;nbsp;nbsp;nbsp;-1nbsp;nbsp;nbsp;nbsp;1nbsp;nbsp;nbsp;nbsp;T² .Ji ...7 rt-ii

* nbsp;nbsp;nbsp;Potomac Flora, pi. cxxxiii. figs. 6 and 7.nbsp;� Foss. FI. Gond. vol. ii. pi. vii. fig. 5.

V. 2129*. A scale or possibly petiole base with, numerous hair-like appendages ; similar to V. 2802. Cf. also Cycadolepisnbsp;hirta, Sap., and the Oto%amites petiole of PI. II. Fig. 4 of thenbsp;present volume. V. 2927. Part of another woolly or hairy scale.nbsp;Eccleshourne.nbsp;nbsp;nbsp;nbsp;Ruffurd Coll.

2.�Eury-Cycadolepis.

In this section are included several detached scales varying in shape from an almost orbicular or somewhat pentagonal form, suchnbsp;as V. 2699, represented in PI. VI. Fig. 6, to the larger and longernbsp;type as shown in PL V. Fig. 2. So far as I have been able tonbsp;discover, these forms have not been previously figured ; therenbsp;cannot be much doubt as to their original connection with somenbsp;form of cycadean stem, and indeed some of the specimens arenbsp;identical with the stout curved scales on such stems as those ofnbsp;Fittonia. In some cases the scales occur in very close association with stems, but in none are they found actually in place.nbsp;For the present, at any rate, it is better to describe some ofnbsp;the more characteristic forms, and to include them all undernbsp;Cycadolepis, suggesting at the same time the very probablenbsp;and indeed almost certain identity of some with the scales ofnbsp;Fittonia and other forms of stems. Hosius and von March havenbsp;described a specimen from the Gault of Ahaus (Westphalia), whichnbsp;they regard as probably made up of a few large petiole basesnbsp;belonging to some form of cycadean stem. One of thesenbsp;�petiole bases� has a length of 11cm., a breadth of 4�7 cm.,nbsp;and is 4 cm. in thickness. The generic name Megalo%amia isnbsp;proposed for this doubtful fossil, and the following definitionnbsp;is given by these authors:nbsp;nbsp;nbsp;nbsp;� Rhachidum basibus incrassatis

carnosis faloiformibus, costis quatuor longitudinalibus prseditis; costis marginalibus acutioribus, costa et dorsali et ventralinbsp;obtusiori.� ' Structures such as this diagnosis describes wouldnbsp;be legitimately included in the genus Cycadolepis, used in thenbsp;more comprehensive sense as suggested above.

The figure of this large example shows very well the general appearance of the longer forms; the surface shown in the drawingnbsp;strongly convex, and at the distal end somewhat suddenlynbsp;incurved. Some of the dark curved lines seen in the figure arenbsp;irregularly placed grooves suggesting the tracks of some smallnbsp;animal, which has slightly eaten into the hard fibrous substancenbsp;af the scale. Very similar markings or grooves have been noticednbsp;Grand�Eury ^ on a leaf of Cordaites, and described by him asnbsp;� galeries d�insecte.� Other lines and striations on the convexnbsp;�nrfaoe of the scale are probably due to a wrinkling of the leafnbsp;substance. The large petiole bases on an old stem of Macrozamianbsp;^ouglasi, Hill, in the Botanical Department of the British Museum,nbsp;i^sar a striking resemblance to this form of Cycadolepis. In thenbsp;recent scales there is the same tendency to terminate in thenbsp;Pointed angular fashion as seen in the figured specimen, andnbsp;in both there is a distinct narrowing towards the base of attach-ment. The convex under surface of the recent scales is coverednbsp;a thick down of hairs, and there is a similar wrinklednbsp;appearance to that of many of the fossil examples. Ecclesbourne.

2), 13-5 cm. in length, 7�5 cm. wide at the broadest part. *^6 convex surface is marked in places by reticulated lines andnbsp;^finklings ; towards the distal end the surface is curved gradually

^ Carruthers (1), pi. Ivi. fig. 1. (The original specimen is in the Museum Practical Geology, Jermyn Street.)

inwards. At the hase there is a well-defined semicircular area bounded hy a distinct line; this is probably an attachment scar.nbsp;Very similar to Macrozamia Bouglasi, Hill. Another specimennbsp;with the same registered number has a similar form, and showsnbsp;surface wrinklings. Eoclesbourne.nbsp;nbsp;nbsp;nbsp;Rufford Coll.

V. 2131. Several specimens : in some the curvature of the distal end is very pronounced.

V. 2131�. Smaller scales, about 5 cm. by 4 cm., closely resembling those of Fittonia and Buohlandia-, cf., e.g., Carruthers,'nbsp;pi. Ivi. fig. 1, and Saporta,^ pi. Ivii. fig. 1.

V. 21313. Scale of medium size; shows similarly convex surface and recurved apical portion, also distinct basal scar.nbsp;Ecclesbourne.nbsp;nbsp;nbsp;nbsp;Rufford Coll.

narrower basal end is bent sharply back, the opposite end shows a well-defined angular margin.

V. 2799. Shows the same kind of attachment surface at the base; the surface is marked by numerous dots and irregular linesnbsp;suggestive of insect ravages. Ecclesbourne.nbsp;nbsp;nbsp;nbsp;Rufford Coll.

V. 2749. Small scale showing distinct rectulate markings on the surface. On the same piece of rock there is an impressionnbsp;of a cycadean stem showing what appear to be the outlines ofnbsp;petiole bases; possibly this may be a badly preserved piece ofnbsp;a stem, to which the smaller scales were originally attached.nbsp;Ecclesbourne.nbsp;nbsp;nbsp;nbsp;Rufford Coll.

V. 2132. Similar specimen, but larger, and the stem impression Diore distinct. Eoclesbourne.nbsp;nbsp;nbsp;nbsp;Rufford, Coll.

V. 2913. Part of a well-preserved scale; shows very clearly the sharp angular contour; the general appearance is very similarnbsp;to that of a large recent cycadean scale. Cf. Macro%amia, sp.nbsp;h^eclesbourne.nbsp;nbsp;nbsp;nbsp;RujfoTd Coll,

Other specimens of similar Eury-Cycadolepis species: V. 2134, piece of Sphenopteris Fontainei, Sew., on the same rock; V. 2236,nbsp;2301, V. 2699(f, V. 2828, V. 2929. Ecclesbonrne.

V. 2800 and V. 2733. These specimens present rather a different appearance to that of most of the larger scales; thisnbsp;however, he due to folding over of the edges, of whichnbsp;there is distinct evidence. Some of the specimens of Cycadolepisnbsp;^re by no means unlike certain monocotyledons spathes, but therenbsp;be little doubt as to their cycadean nature. Eoclesbourne.

Fossil seeds are abundant in rocks of various ages, and in some ^S'Ses their excellent preservation enables us to study in detail thenbsp;structure of both testa and nucellus, and to refer them, withnbsp;considerable degree of certainty, to a particular class, family,nbsp;genus of plants. The superb illustrations in Brongniart�snbsp;posthumous work Recherches sur les graines fossiles silieijiies'-demonstrate in a striking manner, the excellent preservation ofnbsp;isolated gymnospermous seeds under certain favourable conditions ; but in spite of the perfection of the mineralized tissues,nbsp;1^ is scarcely ever possible to assign the detached seeds to theirnbsp;respective plants. In Mesozoic rocks seeds are by no means

uncommon, but their preservation is usually imperfect, and not such as to throw any appreciable light on their exact botanicalnbsp;position. The Wealden strata of England have as yet beennbsp;searched in vain, for any satisfactory indications of the existencenbsp;of angiospermous plants in the flora of that period, and this factnbsp;leads us to the assumption that most probably the Wealden seedsnbsp;are either coniferous or cycadaceous. There are, however, thenbsp;detached tubers of JSquisetites Burchardti, Dunk., and E. Yokoyamm,nbsp;Sew., which may easily be mistaken for seeds. In the first partnbsp;of this Catalogue ¹ ² doubt was expressed as to the nature of thenbsp;oval bodies described as seeds by Stokes and Webb, and Mantell,nbsp;and by some authors referred to EquiutiUs.

The name Carpolithes was proposed by Sternberg as a convenient and comprehensive genus for �Eruotus seminavi mono- vel dicoty-ledonea, solitaria, structura interna plane obliterata.� This termnbsp;has been adopted by many authors as a designation for isolatednbsp;seeds of doubtful position, and its use is in most cases morenbsp;appropriate than any term indicative of some special class or groupnbsp;of plants. In 1849 Pomel proposed Uloapermum,^ as a generic namenbsp;for fossil � fruits � resembling those of the recent Cycadacece, butnbsp;this term, like many others suggested by the same writer, hasnbsp;not been generally accepted. Schimper instituted the genusnbsp;Cycadinocarpus,^ for � Semina subglobosa, ovata vel oblonga,nbsp;quoad magnitudinem valde variantia, nunc parvula, nunc majoranbsp;volumenque Castane� attingentia; epidermide plus minus crassanbsp;instructa, laevia, hand raro compressione mutua angulosa, epi-derniide destituta solida, lignea, subl�via, striata, costata velnbsp;reticulata, basi insertionis cicatrice lata notata, apice minutenbsp;apiculata.� In 1875 Saporta substituted Cycadospermum, as anbsp;more fitting name for detached cyoadean seeds than Schimper�snbsp;genus Cycadinocarpvs, on the grounds of an implied misconceptionnbsp;of the exact morphological nature of the seeds of cycads. Thenbsp;genus is thus defined: �Semina e carpophyllis distracta postnbsp;maturationem in strata pervagata nunc majora nunc plus minusvenbsp;parvula, plerumque ovata ovatoque-oblonga hand raro compressione

�iiitua angulosa extus Isevia vel longitudinaliter striata costataque, ^asi semper rotundiore insertionis cicatrice notata apice autem plusnbsp;�inusve attenuata.�'

As regards the difference between fossil specimens of cycadean and Coniferous seeds, it would seem that we cannot trust to anynbsp;convenient method of distinguishing, in all cases, between thenbsp;two groups of plants. The large oval, or almost spherical seedsnbsp;cf certain cycads may generally he distinguished from the typicalnbsp;forms of coniferous seeds, hut in the latter group we have suchnbsp;forms as Oinkgo, Cephaloiaxm, and others, in which the sizenbsp;approaches more closely to that of the cycadean ovule, than tonbsp;ffie smaller seeds of such conifers as Larix, Pinus, and manynbsp;others. In many Coniferce the presence of a membranous wingnbsp;O�ffords a ready means of identification, at least as regards theirnbsp;reparation from Cyeaiacea, but the seeds of many conifers arenbsp;'quot;'ithout any winged appendage, and even in the case of wingednbsp;reeds, the thin membrane might readily become detached beforenbsp;fhe seed had been permanently enclosed in a mass of sediment.nbsp;-'Another obvious source of difficulty, worth referring to in thisnbsp;Connection, is the very great difference in size exhibited by thenbsp;reeds of the same plant at different stages of growth. The ripenbsp;reeds of such a genus as Cycm, preserved with the wrinklednbsp;J'cddish - brown outer coat intact, present a very differentnbsp;Appearance from those in which this coat has become detached,nbsp;thus exposing the perfectly smooth inner coat; and a still greaternbsp;Contrast is afforded by the more spherical kernel (nucellus), withnbsp;Its surface traversed by branching grooves marking the positionnbsp;cf Vascular bundles.¹ ² On the whole, it would seem advisable tonbsp;follow the example of Schenk in his Flora der GrenzscUcUen,^nbsp;And make use of the old term Carpolithes for gymnospermousnbsp;^fesozoic seeds. In certain cases the character of the seeds, ornbsp;their frequent juxtaposition with cycadean fronds, may enablenbsp;'lA to speak of them as cycadean with reasonable certainty;nbsp;^ten such is the case it will be well to give expression to ournbsp;accurate knowledge, either by adopting Saporta�s genus,

or, possibly the better plan, by adding the word cyeadean or Cycadacem as a descriptive epithet to Carpolithes. If some suchnbsp;course as this were generally followed, there would be less causenbsp;for the not altogether unwarranted criticisms, which students ofnbsp;recent plants are in the habit of passing on the misplacednbsp;dogmatism of palasobotanists. Our records of fossil plants oughtnbsp;surely to be sufficiently trustworthy, to be made use of bynbsp;botanists in compiling statistics of the geological history ofnbsp;any class or family of plants. It must be admitted that tonbsp;attempt a history of plant development or distribution in thenbsp;various epochs of the earth�s history, hy simply accepting asnbsp;reliable data the examples of fossil plants, or fragments ofnbsp;plants, described under the names of existing genera, or designated by terms plainly suggestive of botanical affinity, wouldnbsp;lead the too trustful student into hopeless error. Occasionallynbsp;a fossil seed may exhibit some definite and characteristic form,nbsp;for which some special specific term might be added, but innbsp;the majority of cases where the individual differences are merelynbsp;those of size or slight variation in shape the use of specificnbsp;terms is to be deprecated. In Pontaine�s Potomac Flora severalnbsp;seeds are recorded as species of Carpolithes, the genus beingnbsp;used in this instance for the �nut-like seeds of conifers.�'nbsp;Under the genus Cycadinocarpus the same author places � variousnbsp;horny seeds which resemble those of cyeadean plants more thannbsp;those of conifers.� ^ It is admitted by Pontaine that the correctnbsp;placing of these seeds is impossible; his species are founded innbsp;some cases on very slight differences in size and shape, and cannbsp;have but little taxonomic value. Saporta �� has instituted variousnbsp;species for the Prenoh Jurassic seeds referred to gymnosperms;nbsp;some of these show fairly well-marked characteristic features,nbsp;but in others it would be difficult to justify the adoption ofnbsp;specific designations. In a recent paper by Dawson, severalnbsp;gymnospermous seeds are wisely grouped together as examplesnbsp;of Carpolithes.^

Under certain circumstances, as suggested by Solms-Laubach III speaking of Cycadites (Cycas) Steendrupii, Heer,� it maynbsp;te legitimate to refer seeds and fruits to certain species of plants,nbsp;�'^en in the absence of any actual proof of organic connection;nbsp;^ut it can only be in exceptional cases where the associationnbsp;fronds and seeds renders such a course admissible. As annbsp;example of what appears to be an instance of a supposed con-liection, not sufficiently supported by facts, we may cite Heer�snbsp;Samites globuliferus,'^ where a frond occurs in association withnbsp;Seeds.

Carpolithee.

Fig. 7.�(V. 2130*.) CarpoUthes {Cycadaeem). a. and h. Nat. size. c. Portion of b slightly magnified.

This is a particularly well-preserved example of what must be �'egarded as a cycadean seed. It is impossible to refer it to anynbsp;particular genus, but in all probability it belongs to some other

form than Cycadites, if we assume that genus to have possessed seeds similar to those of the recent Cycas. Species of Maorozamianbsp;possess ovules closely resembling the present specimen. Thenbsp;seed has a length of I�Scm., and is I'lcm. broad. The mouldnbsp;from which the kernel (Fig. 7i) is readily removed, is lined withnbsp;a thin structure probably representing the integumentary portionnbsp;of the testa (Fig. 7�); between this and the matrix there isnbsp;a layer of coaly substance. The kernel may probably be regardednbsp;as a cast of the nucellus, with the impressions of the branchednbsp;vascular bundles clearly seen on its surface. The fossil figurednbsp;by Stokes and Webb as CarpoUthus Mantellij shows in thenbsp;enlarged drawing similar branched markings on the surface,nbsp;suggestive of vascular strands. It may be that Mantell�s specimen should be retained under its original genus and notnbsp;transferred to JSquisetites, but it is difficult to speak with anynbsp;certainty, at all events in the absence of the type specimen.nbsp;Ecclesbourne.nbsp;nbsp;nbsp;nbsp;Ruffurd Coll.

V. 2129, V. 2131, V. 2699. Large flattened and more or less spherical bodies showing coaly substance on the exposednbsp;surface; in V. 2131 the seed (?) is 4 cm. in breadth. It isnbsp;possible that some of them may be scales and not true seeds,nbsp;but their general appearance is not unlike that of some recentnbsp;cycadean ovules, e.g. species of Cycm. Ecclesbourne.

V. 2827fl!. Seed with pointed apex, not unlike Cycadeogperrmim olovatwm, Font.�² V. 2256. Cast of nucellus with remains ofnbsp;testa, similar to V. 2827fl!. V. 2700². Cast and mould ofnbsp;imperfect specimen. Ecclesbourne.nbsp;nbsp;nbsp;nbsp;Rufford Coll.

V. 3312. Probably a seed of some cycadean plant; it appears to have split partially open along the longest diameter. Ecclesbourne.nbsp;nbsp;nbsp;nbsp;Ruffurd Coll.

Carpolithes.

V. 2184. Part of a seed-like body. There are several small 'Circular holes on the surface of this and a few other specimens,nbsp;filled up with a fine brown dusty material, suggesting thenbsp;fiorings of some small animal. Similar examples are afForded bynbsp;V. 2739, V. 2918. Eoolesbourne.nbsp;nbsp;nbsp;nbsp;Rujford Coll.

The following specimens may also be included under Carpo-Uthes: V. 2130, V. 2130�, V. 21305, V. 2165, V. 2700, V. 2739*, V. 2826, and V. 28275.

Seed-like Bodies of Doubtful Position.

Tl. IX. Pig. 5 (V. 2796�). An oval body partially covered ''^'ith a chitinous-like coat ; the dark brown and brittlenbsp;substance which occurs over part of the specimen, suggests somenbsp;resemblance to the dried reddish coat of a Cyoas seed. Thenbsp;Central part does not show any signs of a nucellus or seednbsp;structure; it consists of an irregularly indented projecting portionnbsp;ci the rock. The external skin exhibits no cellular structurenbsp;render the microscope.

Compare Oolithes sphcericus, Carr. ; the figures given by Carruthers of this species, present a striking resemblance tonbsp;the present specimen and other similar forms in the EufEordnbsp;Collection. Buckman had previously identified these Stonesfieldnbsp;�late bodies as reptilian eggs, and Carruthers� examination ofnbsp;^fie same material leads him to accept this determination. Itnbsp;probable that whatever position be assigned to the Jurassicnbsp;dossils, it may with equal force be accepted for many of thenbsp;dealden seed-like bodies.

V. 2825. Flattened and subspherical body, with a hard shiny and dark brown skin more or less deeply indented. Cf.nbsp;Carruther�s figures of Oolithes, sp.

V. 2828. Specimens showing a similar hrown skin, enclosing a smooth central kernel.

V. 2817. Small specimen with smooth surface, showing at the two opposite ends of a diameter a number of very smallnbsp;rounded prominences; these are just visible, as small dots, tonbsp;the naked eye.

V. 2817rt. Small body, 5 mm. long, with smooth hrown coat, similar to V. 2796�, etc. V. 2165, fragment. Ecolesbourne.

FLORES.

In Solms-Laubaoh�s Fossil Botany^ we have a concise and critical resume of the various male and female cycadean flowersnbsp;described in palseobotanical literature prior to 1887. It willnbsp;he seen from this account, that our knowledge of the floralnbsp;structures of fossil cycadean plants is extremely meagre. Innbsp;the carpophylls of the recent Cycas, we have a well-markednbsp;and peculiar form of female flower which is readily distinguishednbsp;from the cone-like collection of carpophylls met with in othernbsp;genera; occasionally these Cycas forms of flowers have heennbsp;found in close association with the sterile fronds of Cyoadites,nbsp;and justify the conclusion that both structures formed parts ofnbsp;the same plant. In other cases, however, we are less fortunatenbsp;in the records of staminal or carpellary leaves, and there mustnbsp;he considerable hesitation in accepting several of the examplesnbsp;which have been described as true cycadean flowers.

It will be convenient to adopt Sohimper�s genus Androstrohus in speaking of a few Wealden specimens, of what appear to benbsp;male flowers of some genus of cycadean plant.

Sohimper has thus defined the genus: �Amenta cycadeacea ontherifera, oylindrica, e squamis imbricatis, latere posticonbsp;^otheras sessiles ferentibus efformata.�

A sufficient definition of Androstrohus, would be to speak of the genus as a convenient term to apply to such fossils asnbsp;J�esemble more or less closely the male flowers of recent cycads,nbsp;and which appear to belong to the CycadacetB.

The genus was founded on a specimen originally described by ^aporta from the Upper Bathonian of Etrochey as A. zamioides,nbsp;^at afterwards renamed A. Balduini^ after the discoverer of thenbsp;specimen, the latter specific term being considered more suitablenbsp;as not suggesting such a definite resemblance to a particular formnbsp;at cyoad. Saporta�s figures show the outline of several pollen-�ao8 between the spirally arranged staminal leaves, attachednbsp;apparently in the same manner as in recent species. Thisnbsp;author describes another and more imperfect specimen of a malenbsp;under the name Androstrobus {Zamiostrohus) Guerangeri; ¹ ²nbsp;same specimen having been previously referred to bynbsp;Brongniart � as an undoubted example of a male cycadeannbsp;flower. Saporta compares this fossil with the genus Bioon, but,nbsp;Solms-Laubach � has suggested, there seems to be but slendernbsp;8''ounds for such a comparison. The specimens described bynbsp;lathorst � and Heer � respectively as Androstrobus borealis andnbsp;Sibiricus, are far from satisfactory, and cannot be acceptednbsp;entirely trustworthy records of this genus. In addition to

� (A. 1), Flor. H�ganas, p. 49, pi. ii. figs. 12 and 13, and pi. ii. (Helsingborg) figs. 15 and 16.

the genus Frigia of Velenovsky,¹ ² reference may also be made to Zamites familiaris (Cord.) from the Lower Quader of Bohemia,nbsp;which Corda and Carruthers^ regard as an example of a malenbsp;cone; the figures of this form lend support to such an opinion.nbsp;The specimen described by Carruthers in his monograph onnbsp;cycadean stems as the � antheriferous cone of Buchlandia'�'amp;oesnbsp;not seem to me to afford any distinct evidence in favour of suchnbsp;a determination.

Androstrobus Natborsti, sp. nov.

Type. Specimens of imperfect flowers and detached staminal leaves. British Museum.

It is difficult to give a definition of this species which shall he in any sense complete. The following may serve to indicatenbsp;the most obvious features of this somewhat unusual form ofnbsp;cycadean cone.

Axis fairly stout, hearing spirally disposed and more or less triangular staminal scales; in section the scales have anbsp;hexagonal outline, in side view they show a broad base in closenbsp;contact with the axis of the flower, and measure about 1-1-5 cm.nbsp;in length, gradually tapered towards the apex, which is pointednbsp;or slightly rounded. On some of the staminal leaves there arenbsp;rows of regularly placed angular depressions, probably representing the impressions of pollen-sacs, borne towards the basalnbsp;or proximal end of each scale.

This specimen shows several fairly well-preserved scales of a male flower, length about 6'5 cm., breadth 4 cm. The centralnbsp;axis is not very clearly seen, hut there are indications here andnbsp;there of the points of attachment of the sporophylls. The axis

* nbsp;nbsp;nbsp;Corda in Eenss (A.), Verstein. bohm. Kreid. p. 86, pi. xlix. :nbsp;3 (3), p. 6.

appears to have been about 1 cm. in breadth, but it is difficult to estimate the dimensions with any accuracy. The surface ofnbsp;the scales is of a brown colour; the longest measures 1-5 cm. innbsp;^ direction at right angles to the floral axis; the surface is considerably wrinkled and bears obvious traces of having been foldednbsp;and crushed. Towards one end of the specimen the basal partnbsp;of a scale is seen in surface view, and on it are clearly preservednbsp;^hat are taken to be the outlines of pollen-sacs (Figs. 3 and 4).nbsp;These are in the form of small depressed areas radiating fromnbsp;the proximal portion of the scale surface; each depression isnbsp;hounded by a straight basal wall, and two slightly divergingnbsp;lateral walls, with two apical walls inclined to one another atnbsp;an angle of about 35�; from the apex there is a slight mediannbsp;ridge passing to the basal wall. In the upper row there arenbsp;about 14 of these pollen-sac impressions, and below these therenbsp;are the remains of a lower set of similar structures. Traces ofnbsp;Ihe pollen-sacs occur on some of the other scales, but less clearlynbsp;preserved. The striking regularity with which these impressionsnbsp;are arranged, is much more marked than in the pollen-sacs ofnbsp;I'ccent cycads. On the lower surface of a staminal leaf of Dioonnbsp;nr Eneephalartos, we find on the removal of the pollen-sacs anbsp;fairly distinct reticulate marking, but of much less regularity thannbsp;m the fossil. The angular outline of the sacs in the presentnbsp;specimens may be due, to some extent, to the mutual pressurenbsp;nf more or less oval structures, such as we have in the pollen-sacs of recent male flowers.

The tapered free ends of the scales are somewhat similar to file narrowed apices of the staminal leaves of species of Ence-phalartos', eg. the male flower of E. Altensteinii, Lehm., E.nbsp;Pungens, Lehm., etc. Of. Androsfrobus Balduini, Sap., pi. cxv.nbsp;figs. 2a and 2S.' Ecclesbourne.nbsp;nbsp;nbsp;nbsp;Eufford Coll.

7 cm. in length. This specimen shows several fairly well-preserved scales somewhat closely set on a central axis which Is narrower than that in V. 2701. Each staminal leaf presents

extending across the middle; if this be clearly seen in the scales on the right-hand side of the axis in PI. IX. Fig. 1, towardsnbsp;the upper margin of some of the scales there are clear indications of another projecting angle, e.g. the third from the bottomnbsp;on the right of the axis in Fig. 1. There is a close similaritynbsp;in form between the detached scale figured by Carruthers asnbsp;Araucarites PhiUipsii, Carr.,' and those of the present specimens.nbsp;The median line, as seen in the scales of Fig. 1, should be compared with the prominent lateral angles seen in the end viewnbsp;of the scales in PI. IX. Fig. 2. In one place on the surfacenbsp;of the argillaceous matrix, there were some fairly distinctnbsp;impressions of pollen-sacs. Ecclesbourne.nbsp;nbsp;nbsp;nbsp;Eufford Coll.

This shows several staminal leaves in end view, some being apparently in place and retaining the spiral arrangement. Therenbsp;can be little doubt as to the identity of these with V. 2810 andnbsp;V. 2701, the different appearance in the present example beingnbsp;due to the fact that here we have a view of the end, and in thenbsp;previous specimen, a view of the flattened sides of the scales.nbsp;The exposed ends show a central depression, and a distinctnbsp;hexagonal outline. The shape and general appearance of thenbsp;scales remind one of the staminal leaves in Zamia, sp.; but innbsp;the present specimen we are presumably looking at the basal, andnbsp;not the distal ends of the scales; the specimens V. 2810 andnbsp;V. 2701 show the much greater width of the base than thenbsp;apex. It may be, however, that in Fig. 2 (V. 2811) thenbsp;apices have been depressed, and we have a view of the apicalnbsp;rather than the basal parts of the scales.

V. 2236. A smaller specimen than V. 2701 and V. 2810; the remains of a central axis with a few well-preserved scalesnbsp;attached. Xo trace of pollen-sacs. Ecclesbourne. Eufford Coll.

Several -writers have called attention to the close resemblance Ween the cones of certain Coniferm and those of some speciesnbsp;�ycads. In attempting to determine the true nature of a fossilnbsp;�one, of which the internal structure is either very imperfect ornbsp;�tttirely wanting, we are met by the great difficulty of clearlynbsp;criminating between the female flowers of these two groups ofnbsp;Carruthers * has mentioned certain distinctive charactersnbsp;cyoadean cones which, he considers, should enable us tonbsp;tinguish them from the corresponding structures of conifers,nbsp;^ ^ the main differences which he. notes are such as can onlynbsp;recognized by the help of internal structure; he writes:nbsp;�Any difficulty' in determining the affinity of a cone by itsnbsp;External characters can easily be solved, as to whether it isnbsp;Coniferous, cycadean, or proteaceous, by a transverse section,nbsp;^hich Would show, if the structure is even a little preserved,nbsp;e form of the scale and the position of the seed.� ^ Unlessnbsp;� structure is fairly well preserved there is often no littlenbsp;�Hlhculty

in deciding in favour of one or other of the two �^ders of plants, Coni/ercB and Cycaiacem.

In view of the generally recognized difficulty of clearly Separating the cones of these plants, and of distinguishing somenbsp;Cones from small cycadean stems, there must be a certain amountnbsp;� hesitation in choosing the most suitable generic term for cone-^he fossils of doubtful affinity.

tndlicher^ proposed the name Zamiostrohus for a cone originally bUred by Lindley and Hutton as Zamia macrocepJmla ^; butnbsp;nrruthers has since shown that the original reference of thisnbsp;to the CycadacecB cannot be accepted, and it is now known

as Pinites maerocephala (L. and H.).� Owing to the erroneous inclusion of this specimen in the genus Zamiostrobus, and thenbsp;unwarranted application of the name to cones which are clearlynbsp;not cyoadean, Carruthers proposed the generic name Gycadeostrohusnbsp;as a more suitable designation for what are � supposed to be fruitsnbsp;of Cycadece.� In speaking of the cones figured by Carruthersnbsp;under this genus, Solms-Laubach¹ ² reasonably suggests that possiblynbsp;several of the fossils may be either small stems or true cones. Thenbsp;only certain cone he considers to be that figured as Cycadeostrohusnbsp;Brunonis, Carr., but this, he adds, �looks more like a cone ofnbsp;Araucaria than of Cycadem.quot; Having had an opportunity ofnbsp;examining Carruthers� type specimens, I must confess to a considerable amount of scepticism in accepting them as well-authenticated examples of cycadean fiowers.

In cases where it seems impossible to express oneself with any degree of certainty as to whether a specimen is a small stem ornbsp;cone, the better plan is probably to give expression to the doubtfulnbsp;afiinity by leaving the fossil unnamed, or by prefixing a query tonbsp;any name which it may have already received. The practice ofnbsp;replacing some of the older and more indefinite names of the oldernbsp;palaeobotanists by newer terms more expressive of definite botanicalnbsp;affinity, has not always marked an advance in accurate knowledge.nbsp;Such a name as Comtes does not, indeed, convey any particularnbsp;information to the mind of botanists as to the nature of thenbsp;fossils so designated, hut, on the other hand, Zamiostrobus ornbsp;Cycadeostrohus both definitely suggest either the male or femalenbsp;flowers of some form of cycad.

In the first volume of this Catalogue the term Algites^ was proposed as a useful generic designation for doubtful forms ofnbsp;fossil Algee, in preference to the more committal and frequentlynbsp;misleading names often made use of. Although such a coursenbsp;as this is, in one sense, rather retrogressive than progressive,nbsp;yet it would at all events minimise the chances of possible errornbsp;if we adopted the old name Conites for several of the conesnbsp;previously referred to the Cycadaoem on what appears to be too

often insufficient evidence. If we have distinct cycadean cones tgt;efore us, the name Cycaieostrobus vrould seem a suitable termnbsp;to apply to them. As in the case of the genus Carpolithes, wenbsp;�ay always give expression to any bias towards one or othernbsp;group of plants, by adding the word Cycadem or Conifera asnbsp;lt;iualifying epithets to the more comprehensive generic name.

I would suggest, then, the revival of the old genus Conites^ O'S a convenient generic name for cones of doubtful botanicalnbsp;offinity.

Conites elegans (Carr.).

1867. Cycadeostrohus elegans, Carruthers, Journ. Bot. vol. v. p. 7, pi. Ivii. fig. 9.

1870. nbsp;nbsp;nbsp;Zamiostrohus elegans, Schimper, Trait, pal. v�g. vol. ii. p. 203.nbsp;Zamiostrohus ovatus, Schimper, loe. cit. p. 203.

1871. nbsp;nbsp;nbsp;Zamiostrohus elegans, Schenk, Palmontographica, vol. xix. p. 228.nbsp;Zamiostrohus ovatus, Schenk, loc. cit. p. 228.

After an examination of the type specimens of Cycadeostrohus ^^^yans and G. ovatus which Carruthers has described, I havenbsp;I'entured to include both examples under one specific name. Thenbsp;Pyritized specimens do not appear to present any distinctivenbsp;characters which can be regarded as of specific value. Unfortunately the preservation is not such as to enable us to provenbsp;either cycadean or coniferous relationship. Carruthers speaks ofnbsp;Cycadeostrohus elegans as an � ovoid cone, truncate below; scalesnbsp;nearly as deep as they are wide,� ^ and of C. ovatus as an � ovatenbsp;none ; scales somewhat broader than deep.� � In the absence ofnbsp;structural characters it is impossible to give any more completenbsp;ihagnosis.

' The genus Strohilites was suggested in 1810 hy Schimper and Mougeot for �Certain cones from the Triassic beds of the Yosges.

40962. Journ. Bot. vol. v. 1867, pi. Ivii. fig. 9. One of the two specimens is the t3'pe of Carruthers� species, Cycadeostrohusnbsp;elegans. In the better specimen, as shown in the figure,nbsp;the pyritized cone has been more or less compressed ; atnbsp;the base there is a central depression or soar of attachment ofnbsp;a peduncle. The surface view of the scales suggests a wearingnbsp;down of their distal ends. The second example is less perfectnbsp;than the type specimen. Brook Point.nbsp;nbsp;nbsp;nbsp;Lady Ilasiinys Coll.

V. 2543. Two specimens, very friable. Cf. Carruthers� figure of Cycadeostrohus ovatus, Carr. Possibly Cycadeostrohus truncatus,nbsp;Cavr.,^ might also be included as a synonym of the presentnbsp;species. Brook.nbsp;nbsp;nbsp;nbsp;Presented hy A. Bendy, Esy., 1888.

V. 63. Imperfect pyritized specimens. These and V. 2543 are somewhat larger than the cone represented by Carruthers in Journ.nbsp;Bot. vol. V. pi. Ivii. fig. 9; but they are probably examples of thenbsp;same species. Brook.nbsp;nbsp;nbsp;nbsp;Purchased 1882.

V. 2853. Portion of a flattened cone ; apparently the same as 40962. Sussex.nbsp;nbsp;nbsp;nbsp;Bechles Coll.

TEUNCI.

In the introductory remarks^ on cycadean fronds it was suggested that the use of some more general term than that ofnbsp;CycadacecB, might prove advantageous in dealing with the remainsnbsp;of extinct cycad-like leaves. The chief reason for such a proposalnbsp;is to be found in the character of the floral structures of thenbsp;genus, for which Carruthers instituted the name of Bennettites.nbsp;This plant, as we have already shown, cannot well be includednbsp;in the class CycadacecB as at present defined for recent species;

the same necessity for a more comprehensive class designation is �(iually apparent in the case of cyoad-like stems. It would henbsp;impossible to so far extend the present limits of the Cycadacem, asnbsp;to incorporate under that term all the fossil stem structures innbsp;li^hich characteristic features of cycadean anatomical structurenbsp;have been recognized; but we must in any case clearly understand that such stems as Benneltites and others, although verynbsp;closely related to recent cycads in histological details, aie,nbsp;however, separated from living forms by certain peculiaritiesnbsp;m the morphology of their reproductive organs. The abovenbsp;heading, therefore, of Trunci does not exclude such stems asnbsp;^I'e known to be associated with a bennettitean form of floralnbsp;structure ; it must be taken in a more comprehensive sense thannbsp;merely including stem structures which agree in all essentialnbsp;features with living members of the Cycadem or Zatniem.

The study of cycadean stems has been raised to considerable importance by the fact of the preservation, in several instances,nbsp;cf more or less perfect internal structure in fossil specimens.

with fronds, so here again we are debarred from any complete '^inguoses of many fossil stems by the isolated occurrence of thenbsp;leaves and their supporting axes. We must for the presentnbsp;restrict ourselves to an investigation of facts as regards thenbsp;S-Datomy of stem structures; and, as in Bennettites, of thenbsp;8'Ooompanying floral shoots.

The early records of so-called cycadean stems in Palaeozoic rocks have already been referred to. It is often a matter ofnbsp;some considerable difliculty to confidently identify a structurelessnbsp;cast or impression of a cycadean trunk ; the imperfectly preservednbsp;�fenis of some forms of Sigillaria, lepidodendron, or Lepidofloyosnbsp;m�y simulate fairly closely the characteristic appearance ofnbsp;cycadean stems. In Grand�Eury�s recent monograph on thenbsp;Coal-field of Gard there is a figure of Lepidojloyoi la/ricinus,nbsp;^tcrnb.,' which may be reasonably compared to a stem of

cycadean plant, bearing lateral appendages suggestive of a hennettitean inflorescence. The tree fern genus Protopteris, withnbsp;leaf-trace bundle scars imperfectly shown or apparentlynbsp;absent, may be mistaken for a cycadean axis with its prominent

petiole bases. An example of such resemblance is afforded by a specimen figured by Hosius and von March as probably Proto-pteris punctata, Sternb.; the plant represented in their plate xliii.nbsp;fig. 186¹ ² might well be described as an imperfect cycadean stem.nbsp;Again, it is almost impossible in some eases to decide withnbsp;certainty between an imperfect cone and a small cycadean stem.nbsp;The fossil described by Lesquereux from Colorado as Zamiostrobusnbsp;mirahilis,^ is obviously a badly preserved stem with basal portionsnbsp;of petioles. An examination of such a stem as that of the livingnbsp;species of Cycas, is sufficient to demonstrate the difficultiesnbsp;attending our attempts to separate into specific forms fragmentsnbsp;of imperfect stems. The upper part of an old Cycas stem withnbsp;its bud scales still in place, presents a very different appearance tonbsp;the lower portion of the same axis, from which the scale leaves andnbsp;petiole bases have become detached, leaving clean-cut rhomboidalnbsp;soars. As a general rule we have a fairly easy task in identifyingnbsp;fossils as cycadean stems. The frond soars and scale leaves whichnbsp;clothe the woody axis afford a convenient distinguishing feature ;nbsp;but, on the other hand, it is important to keep in view the existencenbsp;of other forms of stems among recent cycads, in which the well-known covering of leaf bases is absent. In such plants as Zamianbsp;Loddigesii, Miq., and Z. Skinneri, IVarsz., the peculiar branchednbsp;stem, with its transversely elongated wrinklings and small knoblike protuberances, presents a totally different aspect to the trunksnbsp;of Cycas, Encephalartos, Pioon, and others. It may be notednbsp;in this connection that the Lower Greensand fossil which K�nignbsp;named Pracmna BensteAtii,^ and of which the National Collectionnbsp;contains several examples from the Kentish Eag of Maidstone,nbsp;and a few recently added by Mr. Eufford from the Ecclesbournenbsp;Wealden Beds, shows a striking resemblance to the stems of thenbsp;above-named forms of Zamia. The fossils have at all events nonbsp;claim to a generic name implying a monocotyledonous affinity.

quot;We cannot here undertake a descriptive account of the morphology of recent cycadean stems; but for information onnbsp;this head, reference may be made to the treatment of these

plants in Engler and Prantl�s Die natiirlichen pflamenfmnilien} ^aporta^ and Eenanlt� have also given some account of thenbsp;living cycads, and further details may be found in the writingsnbsp;Brongniart/ Miquel,� Eichard,� Karsten,� Carruthers,� Solms-Baubach, and others.

By far the greater number of known fossil stems have been found in Lower Cretaceous and Jurassic strata, and it is withnbsp;these Mesozoic examples that we are at present concerned. Innbsp;Brongniart�s Prodrome^ there is recorded but one example ofnbsp;^ cycadean stem ; this is the plant described by Buckland fromnbsp;the Portland dirt-bed, and for which the French author suggestednbsp;the name of MantelKa. The common Clathraria LyelU, Mant.,nbsp;included by Brongniart among the Monocotyledons. In thenbsp;tableau � we find several additions to the list of cycadean stems,nbsp;8nd among them is the interesting genus Medullosa of Cotta;nbsp;this Palmozoic plant has been subjected to a detailed investigationnbsp;hy Goppert and Stenzel,� Schenk,'^ and others, and we maynbsp;probably regard it as an extinct type of cycadean structure,nbsp;Using the term cycadean in a wide sense. In 1828 Buckland��nbsp;figured and described some large specimens of silicified stemsnbsp;from the Isle of Portland, and, with the concurrence of Eobertnbsp;Brown, instituted a new family, Cycadeoidem, for their reception.nbsp;Buckland fully recognized the close resemblance between thesenbsp;�petrified birds� nests� and the stems of certain cycads; butnbsp;u new family name was proposed on account of some peculiaritynbsp;Us regards the position and size of the rings of wood. In a

� Goppert and Stenzel.�In speaking of the MeAuUosem, Solms-Laubach remarks (Fossil Botany, p. 100) that �that in their anatomical structure theynbsp;show many points of resemblance to the Cycadeee, though they depart from them,nbsp;according to the most recent investigations, in some important particulars.�

Schenk. nbsp;nbsp;nbsp;��nbsp;nbsp;nbsp;nbsp;Bucklandnbsp;nbsp;nbsp;nbsp;(1).

later work,' this author has given some further description of the Portland fossils, and discusses the question of terminology;nbsp;the genus Cycadites being regarded by Brown as preferable tonbsp;Cycadeoidea, and the name of Mantellia, proposed by Brongniart,nbsp;is thought to be unsuitable, having been already used by Parkinsonnbsp;for a genus of fossil Zoophytes. At the present day it is unfortunately not always held that the use of a particular namenbsp;by palseozoologists, is a fatal objection to the adoption of the samenbsp;for a fossil plant. Among the new figures added by Buckland tonbsp;those given in his earlier paper, we find some drarvings of longitudinal sections of petioles and axillary buds; the latter have sincenbsp;been fully described by Carruthers as the inflorescence ofnbsp;Bennettites. In 1870 an important monograph appeared by thenbsp;latter author on Fossil cycadean stems from the Secondary roclcsnbsp;of Britain-,^ the memoir contains full reference to earlier recordsnbsp;of cycadean stems, and includes figures and descriptions of thenbsp;following new genera � Yatesia, Fittonia, Wtlliamsonia, andnbsp;Bennettites. Five years later, several additions were made bynbsp;Saporta ¹ ² to our knowledge of the stems of fossil cycads; henbsp;founded the genera Bolbopodium, Cylindropodium, Platylepis,nbsp;Clathropodium, and Cycadeomyelon. The numerous terms added ornbsp;substituted for those previously proposed by Carruthers havenbsp;involved the terminology of cycadean stem structures in somenbsp;confusion. In more recent years we have a valuable contributionnbsp;from Solms-Laubach and Capellini^ on the examples of bennettiteannbsp;stems preserved in Italian museums. These authors limit the usenbsp;of the term Bennettites to a single species, B. Qibsonianus, Carr.,nbsp;and in a still later preliminary paper by Lester Ward,� Carruthers�nbsp;genus is absorbed into the more comprehensive Cycadeoidea. Wenbsp;may look for an important monograph at an early date by Lesternbsp;Ward and Knowlton on the exceedingly fine series of Americannbsp;cycadean stems. In Dana�s Manual of Geology�' mention is

of some large stumps of cycads having been found near Saltimore, Maryland, and their age is spoken of by Tyson asnbsp;probably Upper Jurassic. Eontaine�s Potomao Flora'- containsnbsp;�' few photographs of these Maryland stems, and a splendidnbsp;Specimen has lately been received by the Botanical Departmentnbsp;the British Museum. The few facts we so far possess as tonbsp;these American stems lead us to expect a descriptive monograph

ne material so far collected from Upper Jurassic and Lower re aceous strata has already yielded valuable information withnbsp;Regard to the anatomy of the vegetative, and in some instancesnbsp;the reproductive, structures of Mesozoic oycadean plants. Tonbsp;rirther extend our knowledge of these various fossil species, a morenbsp;^'rtimate acquaintance with the several types of recent cycadsnbsp;ranch to be desired; and, as Solms-Laubach � points out, wenbsp;possess no detailed and modern account of the large tuberousnbsp;long a,go described by Buckland from the dirt-beds of

ortland.

arrtell in the �Wealden of Tilgate; the same fossil had been previously referred by Stokes and Webb to Clathraria^^ a termnbsp;proposed by Brongniart* in 1822 for certain forms of sigillariannbsp;MantelP was the first to give a description of these Tilgatenbsp;plants, but he proposed no name for them, merely pointing outnbsp;a probable affinity with the FuphorhiacecR, or possibly with thenbsp;arborescent ferns. Carruthers pays a tribute to the �remarkablenbsp;discrimination�� with which Presl recognized the oycadean nature

of the English fossils. In the Tableau'^ Brongniart includes Mantell�s plant in the Liliacece, noting at the same time its resemblance tonbsp;the stem of a cycad. Schimper^ retains Clathraria for plants ofnbsp;the type of Clathraria anomala, Stokes and Webb ((7. Lyelli,nbsp;Mant.); and Saporta,� who follows Carruthers in preferring thenbsp;name BucMandia to Clathraria, speaks of the plant figured bynbsp;Schenk as C. Lyelli as probably a species of Carruthers� genusnbsp;FiUonia. Hathorst, in his Floran ihd Fju/,* on the other hand,nbsp;includes under Clathraria two new species, but in the latter partnbsp;of the same work he substitutes BucMandia^ for Stokes andnbsp;Webb�s genus. Hathorst�s specimens are imperfect fragmentsnbsp;of stems with alternating series of narrower and crowded leaf-scars, and broader and more openly arranged leaf bases; henbsp;compares them with the stem of Cycas.

The separation of such conventional genera as Buohlandia and those proposed by Saporta, is often a matter of great difficulty,nbsp;and so long as we have only imperfect external or internal castsnbsp;to deal with, there must always be a certain amount of doubtnbsp;as to the existence of true generic and specific differences.nbsp;Carruthers thus defines the genus Buchlandia: �

�Trunk cylindrical, sometimes bifurcating, reticulate, with the scars of the bases of the leaves, which are arranged in alternatingnbsp;series of large and small scars, the large being placed on swellingsnbsp;and the small on constrictions of the stems. Androeeium a cone (?),nbsp;gynoecium a terminal crown of leaves bearing seeds on their somewhat altered margins.� �

The so-called male cone referred to in this definition was discovered in the same series of strata as those in whichnbsp;Buchlandia occurs; it is assigned to this genus on the strengthnbsp;of its occurrence in the same beds, and on account of a resemblancenbsp;which its scales present to the sporophylls of a male flower ofnbsp;the recent species of Cycas. In the absence of more satisfactorynbsp;evidence than is afforded by this single imperfect specimen, thenbsp;nature of which does not appear to he by any means established,nbsp;we are not in a position to include the male flower in a definitionnbsp;of Presl�s genus. The alternating swellings and constrictions

the stem, and the slight difference in the form and size of ''he leaf-soars, led Carruthers to draw a close parallel betweennbsp;^^(klandia and Cycas, and to infer the nature of the femalenbsp;flowers.* He writes:nbsp;nbsp;nbsp;nbsp;� If the interpretation I have given of

''he stem of BucUaniia he correct, and if there be good reason, from a morphological point of view, for connecting with it thenbsp;�eeds and male cone found in the same beds, we have a plantnbsp;^hich, in these known particulars, cannot be separated generioallynbsp;from Cycas.� � In addition to the original species, Buchlaniianbsp;*^^omala (S. and tV.), Carruthers proposes a second specific namenbsp;frr certain Wealden stems which he considers to be distinct fromnbsp;Stokes and Webb�s type. An examination of the British Museumnbsp;*Oaterial does not appear to favour this separation into twonbsp;flistinct forms, and I have ventured to incorporate both of thenbsp;Species into B. anomala. We cannot hope to separate suchnbsp;*tuperfect and structureless specimens into specific forms of anynbsp;teal value, seeing what marked variations in surface features wenbsp;expect to find in examples of cycadean stems clothed withnbsp;^ number of more or less decayed leaf bases. It may be notednbsp;'quot;hat the upper portion of the specimen of BuohJandia anomalanbsp;floUred by Carruthers in pi, liv. fig. 1, shows a close approximationnbsp;in form to that of some forms of the genus Fittonia.

1824. Clathraria anomala^ Stokes and Webb, Trans. Geol. Soc. [2] vol. i. p. 422, pis. xlv., xlvi. fig. 8; pi. xlvii. fig. iv.

1827. nbsp;nbsp;nbsp;Clathraria Lyelli, Mantell, Illust. Geol. Sussex, p. 52, pi. i. fig. 2;

fr83. Clathraria Lyelli, Mantell, Geol. S.E. England, p. 233, pi. i. figs, i., ii., and vi.

' This suggested resemblance to Cycas does not appear to me very close; ^ comparison of the fossil stem and its pith cast with the stem and pith castnbsp;Haerozamia, sp., reveals a striking similarity.

^ The specimen figured by Mantell (p. 293), Abth. i. p. 305, as Clathraria lyelli is a waterworu fragment of Benmtliles.

Clathraria Lyelli, Ettiugshausen, Abh. k.-k. geol. Eeichs. vol. i. Abth. iii. ISTo. 2, p. 25.

�52. Clathraria Lyelli, Bronn, Leth. geog. vol. ii. p. 63, pl. xxviii. Sg. 7. Clathraria Lyelli, Morris, Brit. Foss. p. 6.

BucHandia anomala, Carrnthers, Trans. Linn. Soc. vol. xxvi. p. 686, pl. liv. figs. 1-3.

Clathraria Lyelli, Schenk, Palseontographica, vol. xix. p. 227, pl. xxx. fig. 7.

Type. Pith casts and structureless casts of the cortical surface of stems. British Museum.

In 1822 Mantell gave a brief description of some fossil stems from Tilgate Porest for which he proposed no name, but suggestednbsp;that they might be allied to the Euphorliacem, or possibly to certainnbsp;arborescent ferns. Two years later Stokes and Webb proposednbsp;to include these fossils in the genus Clathraria, and gave themnbsp;the name of C. anomala; the surface-markings suggested to themnbsp;a resemblance to the recent Zamia and Cyoas. In the Geologynbsp;of the South-East of England Mantell claims priority for hisnbsp;name of Clathraria Lyelli; he notes the occurrence of imperfectnbsp;leaf fragments in association with the stems, and speaks of themnbsp;as linear-lanceolate in shape. Having mentioned such plants nsnbsp;he considers most closely allied to the Tilgate fossils, Mantellnbsp;adds : � the impressions of the petioles on the bark bear a greatnbsp;resemblance to those on the stems of Cycas revoluta andnbsp;C. circinalis.quot; In a later work '� the same writer speaks of thenbsp;axis, roots, leaves, and probably fruit of Clathraria as having beennbsp;discovered in close association or connection with one another.nbsp;He gives a woodcut showing some of the long linear-lanceolatenbsp;leaves attached to the stem, and remarks that impressions of suchnbsp;yucca-like leaves have often come under his notice. It isnbsp;unfortunate that none of the specimens of Clathraria afford anynbsp;evidence whatever as to the form of the leaves as described by

^lantell. In the younger stems Mantell notes that the internal (pith cast) cannot he distinguished from the outer cortical

Carruthers' gives the following detailed definition of BucTclandia �^nornala : � Scars of the leaves suhrhomhoidal, the lateral anglesnbsp;more or less truncate, inferior angle acute, the superior obtuse ornbsp;Somewhat rounded. The surface of the scar in some specimensnbsp;iQarhed with a triradiate ridge. The smaller scars oblong, withnbsp;^lunt lateral angles, obtuse inferior, and slightly rounded, almostnbsp;straight superior angle ; the scars equal in breadth to the largernbsp;Ones, hut not nearly so deep ; the cicatrix on the upper margin,nbsp;�fhe bases of the leaves are set somewhat obliquely on the stem,nbsp;their upper margin following the direction of the right-handnbsp;Spiral. Each series of leaves occupies a considerable length of thenbsp;stem. The phyllotaxy is represented by the fraction -A-.�

� Scars of the leaves rhomboidal, the lateral angles acute, the inferior and superior angles obtuse, the latter somewhat rounded.nbsp;The small scars equal in breadth to the large ones, and increasingnbsp;depth from the bottom of the constriction upwards. Eachnbsp;swelling of the stem bearing three or four series of leaves, thenbsp;Constricted portion much longer, and crowded with the smallernbsp;�cars, forming twelve or fourteen vertical series. The base of thenbsp;leaves set horizontally on the stem. The phyllotaxy is representednbsp;the fraction i%-.� * It will he noticed that the more acutenbsp;lateral angles of the leaf-scars and certain other slight differences,nbsp;Constitute the chief distinctive features of Buchlandia MantelU,nbsp;Carr. The close agreement between these two forms will henbsp;Pointed out in the following descriptive notes on the Britishnbsp;Museum specimens :�

In the original figures of Buchlandia anomala a specimen is shown with the cortical portion of the stem separated by somenbsp;coaly substance�no doubt the carbonized remains of the wood�nbsp;from a central sandstone cast; the latter, with its �interruptednbsp;loiigitudinal ridges,� being a oast of the pith cavity, and the ridgesnbsp;file impressions of spaces in the xylem cylinder which were

originally occupied by the cells of the primary medullary rays, lu speaking of the numerous specimens of these pith casts,nbsp;Carruthers admits the impossibility of referring them with anynbsp;certainty to their respective species of bucklandian stems, andnbsp;suggests that probably they may belong to three or four distinctnbsp;forms of the genus. Saporta^ has instituted a comprehensivenbsp;genus, Gyoadeomyelon, with the following definition: � Medullanbsp;centralis primum substantiae cellularis disperditione evanida, deinnbsp;sedimento cylindrum lignosum intus vacuum oumulorite substitulanbsp;et tunc post ligni oiroumfusi abolitionem cylindrum plenum plusnbsp;minusve compressum fasciculorum meatuumque impressionibusnbsp;superficialiter notatum efformans.�

In cases where we cannot be certain as to the relation between casts of the pith cavity and those of the cortical surface of a stem,nbsp;it will be convenient to make use of Saporta�s genus as a usefulnbsp;designation to express the absence of sufficient data for anynbsp;more accurate identification. In some specimens we have thenbsp;clathrarian pith oast enclosed in a bucklandian cortex, but innbsp;most cases the internal and external casts have been separated.nbsp;Although very probably, as Carruthers suggests, the detachednbsp;pith casts belong to more than one species of Bucklandia, yetnbsp;the very striking resemblance between those internal casts, whichnbsp;are still surrounded by the cortical surface, and the isolatednbsp;specimens is sufficient reason for the inclusion of such formsnbsp;under the present genus. Hosius and von der Marck'* havenbsp;described an Aptien fossil stem as Clathraria (?) galtiana, andnbsp;compare it with Clathraria Lyelli as figured by Schenk �; it is,nbsp;however, probably not identical with the English type, andnbsp;should perhaps be referred to the genus Fittonia.

8262. Figured by Stokes and Webb, Trans. Geol. Soc. [2] vol. i. pi. xlv. fig. 1. Mantell, Illust. Geol. Sussex, pi. ii. fig. 1.*nbsp;Carruthers, Trans. Linn. Soc. vol. xxvi. pi. liv. fig. 3.

From one end of the specimen the cast of the hollow pith projects beyond the encasing wood and cortex; it shows the

*� Mantell explains the repetition of Stokes and Webb�s plates in his �Illustrations of the Geology of Sussex� in a note on page 52 of that work.

characteristic ridges of Glathraria. The petiole scars have rounded or truncate lateral angles, hut those towards the lowernbsp;part of the specimen have their lateral angles more acute; thisnbsp;a� indicated in Carruthers� figure on the left of the pith cast.nbsp;Probably specifically identical with specimen V. 3309, figurednbsp;hy Carruthers as Bucklandia ManteUi. Tilgate Porest.

8358. Figured by Stokes and Webb, he. cit. pi. xlv. fig. 2. ^antell, he. cit. pi. ii. fig. 2.

The figure of this specimen does not do full justice to the details shown on the stem surface. If some of the soars benbsp;compared with the lower petiole bases of V. 3308 it will be foundnbsp;^o be a matter of some difficulty, not to say impossible, to pointnbsp;to any distinct difference between the two forms, the former ofnbsp;rrhich Carruthers speaks of as Bucklandia Mantelli, and the latternbsp;B. anomala.

A- second smaller specimen in the form of a slightly compressed hollow cast of the bark with weathered petiole bases, and C�! the upper portion bearing soars of scale leaves. Very similarnbsp;to Mantell�s pi. iii. fig. 4�. As shown in Mantell�s figure, therenbsp;a distinct projecting ridge above the petiole scar, and separatednbsp;t�y a depression from the main part of the petiole; this is pro-Pably due to the irregular or unequal weathering of scleren-c'lymatous and parenchymatous tissue. Cf. a stem of Cycas.nbsp;tilgate.

V. 3308. Figured by Mantell, Geol. S.E. England, pi. i. fig. 2, by the same author in Illust. Geol. Sussex, pi. i. fig. 2.

Here again the figures do not do justice to the specimen. In ^i'is stem the petiole soars are clearly preserved and larger thannbsp;the other examples of the same species; they are not so muchnbsp;obscured, as in many specimens, by the irregular surface ridgenbsp;ond projections which Carruthers regards as the remains of anbsp;gummy exudation on the original plant stem. Some of the morenbsp;prominent scales resemble the form which occurs in close association with V. 2749, and the lower portions of some of the largernbsp;Scales suggest a reticulate marking like that in V. 2749 and innbsp;^overal of the isolated scales. The lateral angles of the scars arenbsp;tor the most part rounded or obtuse, but in some the angles

are much more acute. The form of many of the leaf base soars points to a partially decayed petiole rather than a clean-cutnbsp;surface of a persistent corky base. Cf. Carruthers, loc. oit.nbsp;pi. liv. fig. 1 ; ef. also FUtonia. Tilgate Forest.

In this specimen the two kinds of soars are clearly shown, also the increased diameter of the axis where the large scarsnbsp;occur. The lateral angles of the scale scars are more obviouslynbsp;truncate than those of the petiole bases. Cf. 46644. Cuckfield.

In describing this specimen Carruthers speaks of a � triradiate ridge � on the surface of the scars, but I am utiable to recognizenbsp;any such character; the markings are probably the result ofnbsp;some secondary changes and cannot he regarded as an originalnbsp;character.

V. 713. This specimen shows the clathrarian pith cast, as in 8262, but much more clearly. The flattened internal cast hasnbsp;a length of 45 cm., and exhibits the usual surface featuresnbsp;characteristic of Clathraria Lydli. Part of it piojeots beyondnbsp;the surrounding cortical cast, but it is enclosed to some extentnbsp;by the remains of leaf bases, and between the cortical shell andnbsp;the pith there is a space about 5 mm. in width, here and therenbsp;filled with coal; this no doubt represents the portion of the stemnbsp;originally occupied by woody tissue. The pith case is flattenednbsp;and shows alternations of broader and narrower portions; thenbsp;surface markings, in the form of narrower and tapered ridges,nbsp;do not appear to be disposed with any regularity. The largenbsp;leaf bases are approximately 3 cm. in depth and 4'3 cm. in breadth.nbsp;Some of the leaf bases have projecting upper surfaces, as innbsp;V. 3308, suggesting partially decayed petioles, but in other partsnbsp;of the stem the scars are much flatter and more like the clean-cut bases of detached fronds.

V. 713a. 23 cm. in length. In the broader and lower part of the specimen the petiole bases are fairly large, and have muchnbsp;the same form as in V. 713.

V. 713i. Poor specimen. On one side an irregularly marked surface is exposed, which probably represents the impression ofnbsp;external portion of the wood. Hastings. Dawson Coll.

V- 2132^. Imperfect impression of a stem, anoe to V. 2132, but differs in having two well-marked formsnbsp;uf soars, and in the absence of ramenta. Part of a scale innbsp;close connection with the stem, identical with V. 2699, etc.

Rnfford informs me that he has frequently found large scales, such as V. 2799, in association with this form of stem;nbsp;tile base of V. 2799 might well have been in attachment withnbsp;scars like those of the present specimen.

2132c. A small specimen showing petiole scars: this may un example of Buchlandia, but it also closely resembles thenbsp;Senus Fittonia, and its precise nature must be a matter of uncertainty. Ecclesbourne.nbsp;nbsp;nbsp;nbsp;Rufford Coll.

. � 2749. Small piece of a stem with two or three fairly distinct Petiole scars, and some narrow and longer scars to which scalesnbsp;Uquot;ere probably attached. A single detached scale, with thenbsp;cuiate surface markings, occurs in close association with thenbsp;Cf. Carruthers, loo. oil. pi. liv. fig. 4 [Buchlandia MantelU).

Cycadeomyelon (Bucklandia), sp.

V. 2332. Probably a portion of a pith cast; but the larger medullary ray impressions suggest some other stem than thatnbsp;with which the ordinary Clathraria Lyelli is occasionally found.

V. 2804. Considerably flattened sandstone cast. The form of the medullary ray impression and of the leaf-trace bundle is morenbsp;distinctly shown than in most specimens. Schenk�s figure alsonbsp;shows this feature clearly. The form of the medullary ray castsnbsp;in quot;Weiss� genus Tylodendron is very similar to that of the corresponding structures in the present specimen. Good figures ofnbsp;Tylodendron will be found in a paper on that genus by Potoni�.'nbsp;Cf. also similar easts in Voltzia.

V. 2804�. Similar specimen, but less flattened. Part of a scale at one end of the oast. Ecclesbourne.nbsp;nbsp;nbsp;nbsp;Rufford Coll.

8268, nbsp;nbsp;nbsp;12333. Portions of pith casts; probably of the lowernbsp;portion of the stem axis. Tilgate Porest.

8264. Figured by Mantell, Illust. Geol. Sussex, pi. i.: 93 cm. in length; at broadest part the diameter is 13cm. The detailsnbsp;of surface characters not very clearly preserved. Tilgate Porest.

8269. nbsp;nbsp;nbsp;79 cm. long, diameter 13 cm. A large branched specimen,nbsp;showing at the summit an apparently dichotomous bifurcation. Innbsp;addition to the large branches, there are the scars of six or sevennbsp;smaller lateral branches, about 2cm. in length; the smallernbsp;scars are all on one side of the specimen and arranged in anbsp;fairly regular line. Cf. G�ppert�s figure of Cycas revoluta withnbsp;the numerous large and small branches.�* Tilgate Porest.

8274. 78 cm. long, 7'5om. in diameter. The breadth varies, but there is no regular alternation of narrower and thickernbsp;portions. Tilgate Porest.nbsp;nbsp;nbsp;nbsp;Mantell Coll-

V. 3307. A long specimen with distinct constrictions at fairly Regular intervals; the surface projections are rather broader thannbsp;^�1 many examples, and resemble those in V. 2332 ; they shownbsp;S' considerable variation in size, some being identical with thenbsp;visual clathrarian form of medullary ray cast, and others largernbsp;snd more prominent.

V. 3306. Here again there is a marked variation in the size of the ray impressions. A well-defined branch scar.

This generic name was proposed by Carruthers for an unusually Perfect specimen of a structureless cycadean stem, which wasnbsp;Pmbably obtained from the quot;W^ealden of Brook, in the Isle ofnbsp;''^ight. Mantell first figured and described this specimen asnbsp;^latlira/ria Lyelli in his Geological Excursions round the Isle ofnbsp;^ight *; a larger and more complete figure appears in Carruthers�nbsp;Monograph. The genus is thus defined:�

� Trunk short, obovate ; woody axis slender, enlarging upwards ; Cortical layer large. Scales and bases of the petioles large,nbsp;^^bricated, at first reflexed, then ascending.�

tlurruthers notes the absence of any fruit or foliage which can be referred to this form of stem, but he considers thenbsp;Resemblance to Encephalartos sufficiently distinct to justify himnbsp;assigning Fittonia to a position near to the living genus. Saportanbsp;^�'^opts the name, and points out the resemblance between Fittonianbsp;'^nd BucUaniia-, he calls attention, however, to certain points

of difference, and considers that Mantell�s specimen has been correctly made the type of a new genus. In Buchlandia thenbsp;stem appears to have been frequently branched, but no suchnbsp;habit is indicated in the specimens of Fittonia-, in the latternbsp;genus the increase in size both of the petiole and scale-leafnbsp;bases, and of the lower portion of the petioles as well as. thenbsp;scale leaves, with the subsequent disarticulation of the uppernbsp;part of the frond axis, constitute constant and characteristicnbsp;features. The pith is large and surrounded by a narrow zonenbsp;of wood. The scale leaves, associated here and there with thenbsp;bases of fronds, are distinguished from the latter by theirnbsp;thinner distal margins, and the absence of any distinct surfacenbsp;of articulation. As Saporta remarks, there is a striking resemblance between some forms of Fittonia and Buchlandia; thisnbsp;similarity has already been noted in the descriptions of somenbsp;of the Museum examples of the latter genus, and it is alsonbsp;clearly seen in Carruthers� type specimen in the Jermyn Streetnbsp;Museum of Practical Geology. It may be that the two generanbsp;are not really distinct, but merely represent different forms ofnbsp;preservation of very similar, if not identical, plants. In describingnbsp;the single English specimen of Fittonia, Mantell quotes thenbsp;opinion of Brongniart,^ to whom a drawing of the fossil wasnbsp;sent, that it is probably the upper portion of a clathrariannbsp;stem with persistent petioles. It may be more convenient, withnbsp;the evidence at present available, to retain both generic names,nbsp;and to make use of Fittonia as a useful designation for a certainnbsp;form of cycadean trunk.

Fittonia Ruflfordi, sp. nov.

The lower portion of the petioles persistent, showing a well-marked surface from which the upper part of the frond has been detached. The persistent and swollen bases are regularlynbsp;disposed in the stem surface, and apparently without anynbsp;alternation of petiole scars and scale-leaf scars.

18 cm. in length, 10 cm. broad. An impression of part of the eral surface of a cycadean stem, covered with a carbonaceousnbsp;^^yer. Tile petiole bases are shown with unusual definition, andnbsp;� form of the surface of articulation is particularly clearlynbsp;^8-iked. The swollen portion or cushion below the petiole scarnbsp;^fisents a fairly close resemblance to some forms of Sigillarianbsp;ardii^ Prong., with its similarly situated leaf-scar, whichnbsp;��rees closely in shape with that of Fittonia. The petiole basenbsp;cusbiQn has a length of I'Vcm. The form of the leaf-scarnbsp;practically identical with that of the cleanly-cut end of tlienbsp;ria of Otozamites Goppertianus as figured in PI. I. Pig. 2.nbsp;J- Fittonia squamata, Carr.,' and F. insignis, Sap.^; the presentnbsp;Species differs from these in the smaller size of the leaf base andnbsp;^ their much more regular and uniform arrangement on thenbsp;� �ra. Ecclesbourne.nbsp;nbsp;nbsp;nbsp;Rufford Coll.

2121

77 cm. long, 8-9 cm. broad. Here again the surface the stem has been carbonized, and the petiole bases presentnbsp;similar appearance to those in V. 2238, but are rather lessnbsp;Perfectly preserved. The comparatively long and narrow form ofnbsp;� leaf base is clearly shown. Ecclesbourne.nbsp;nbsp;nbsp;nbsp;Rufford Coll.

2244. 35 cm. long. Probably this may be referred to tonia Rujfordi, but the petiole bases are much less clearlynbsp;preserved. In all the specimens the characteristic feature is thenbsp;^�quot;eat length as compared with the breadth of the petiole bases,nbsp;l^cclesbourne.nbsp;nbsp;nbsp;nbsp;Rufford Coll

In 1855 the President of the Linnsean Society of that year, Robert Brown, exhibited a new form of cycadean stem whichnbsp;had been found by Saxby at Bonchuroh, in the Isle of Wight.'nbsp;The name Cycadites Saxhyanus was suggested by Brown for thisnbsp;new fossil, of which the two most striking features were thenbsp;elliptical outline of the stem as seen in transverse section, andnbsp;the presence of a bud in the axil of each leaf. Fifteen yearsnbsp;after the discovery of this stem, Carruthers ^ published a full andnbsp;scientific description of several examples of the same species.nbsp;As a result of careful investigation of the morphology of thesenbsp;specimens, Carruthers instituted the generic name Bennettitesnbsp;for the new form of stem, which he found could not be classednbsp;under any of the recognized subdivisions of the Order Cycadacea.nbsp;After speaking of the elliptical form of the axis and the presencenbsp;of buds in the axils of many of the leaves, Carruthers proceedsnbsp;to describe the anatomical structure of the stem, and calls specialnbsp;attention to the leaf-traces as affording another characteristicnbsp;feature of the genus. � In all the known members of the Ordernbsp;{Cycadacem),� ^ says Carruthers, �the leaf-traces arise in thenbsp;interior of the cylinder of wood, as bundles of small size, and,nbsp;passing through the meshes of the ligneous cylinder, and thennbsp;through the cortical parenchyma, as small distinct bundles, afternbsp;running for a short distance, at least in some genera, in anbsp;horizontal direction parallel to the periphery of the stem, theynbsp;pass in the petiole of the leaf.� In Bennettites, however, �thenbsp;vascular tissue for each leaf springs from the woody cylinder innbsp;a single large compact bundle, which as it passes outwards breaksnbsp;up into the different bundles required for the service of the leaf.�nbsp;This peculiar behaviour of the leaf-traces is also referred to

Carruthers� genus. The structure of the petiole bases and the rameuta-like scales has been fully described by Carruthers andnbsp;others. In the axils of some of the petiole bases which surroundnbsp;the woody axis of the stem, there occur the lateral branchesnbsp;^hioh constitute the most important and interesting feature ofnbsp;the genus. �These organs,� writes Carruthers, �are notnbsp;properly buds, for although they do not appear to have pushednbsp;themselves beyond the surfaces of the permanent bases of thenbsp;leaves, they are fully developed organs, and differ from thenbsp;Secondary axes of Mantellia, which are generally broken offnbsp;heyond the surfaces of the permanent bases of the petioles, andnbsp;�how there a woody cylinder agreeing in structure with thenbsp;principal axis of the plant. The secondary axis consists of anbsp;�rery short and slender stem, bearing a number of simple linearnbsp;acuminate leaves. These are the only foliar organs hithertonbsp;found connected with these fossils.� ¹ ² Each of the axillarynbsp;branches terminates in a subpyriform enlargement, bearing seeds;nbsp;fhis terminal portion with its seeds are fully described andnbsp;illustratednbsp;nbsp;nbsp;nbsp;by Carruthersnbsp;nbsp;nbsp;nbsp;andnbsp;nbsp;nbsp;nbsp;Solms-Laubach.nbsp;nbsp;nbsp;nbsp;Thenbsp;nbsp;nbsp;nbsp;former

author, in summing up the affinities of the genus, expresses Ibe opinion that � it must be considered to hold the samenbsp;relation to the other Cycadem, that Taxus, with its succulent,nbsp;cup-shaped pericarp does to the cone-bearing ConiferaT

In his nbsp;nbsp;nbsp;Einleitung innbsp;nbsp;nbsp;nbsp;dienbsp;nbsp;nbsp;nbsp;Paliiophytologienbsp;nbsp;nbsp;nbsp;Grafnbsp;nbsp;nbsp;nbsp;Solms

Confirms many of the characteristics of Bennettites as described by Carruthers, and by a careful examination of the Englishnbsp;material he is able to settle certain doubtful points, and to carrynbsp;a stage further our knowledge of this interesting type.² Anbsp;more detailed account was afterwards published b)�� Solms innbsp;Ibe Botanische Zeitung,^ and the article subsequently appearednbsp;lu English in the fifth volume of the A.nnals of Botany.^nbsp;I'Vithout entering at length into a histological description ofnbsp;bennettites, we may call attention to some of the more strikingnbsp;aud characteristic features. The structure and course of the

leaf-trace as described by Carruthers, and the analogy which he noted between the trellis-work form of the vascular cylindernbsp;and the bundle system of a fern stem are confirmed by Solms;nbsp;he adds, however, that a � closer examination will doubtlessnbsp;disclose a greater affinity with the course of the vascularnbsp;bundles in many ConiferThe pith and cortex are traversednbsp;by numerous and large gum-canals, and the stems are enclosednbsp;in the armour of leaf bases characteristic of recent cyoads.nbsp;Between the petioles of Bennettites and those of living cycadsnbsp;there is the closest agreement; the bases of the leaves arenbsp;separated from one another by a felt of ramenta-like outgrowthsnbsp;from the petiole surface. Occasionally the substance of the petiolenbsp;has rotted away, leaving deep cavities occupying the meshes innbsp;a network of intervening projecting ridges. In Bennettitesnbsp;Gihsonianus the structure of the lateral branches bearing thenbsp;remarkable form of fructification is very perfectly preserved,nbsp;and it is evident that we have to deal with a plant differingnbsp;in many important respects from the present types of cycads.nbsp;Eeferring to the cycadean character of the stem and leaf-stalknbsp;of Bennettites, Solms remarks: �We arrive at the surprisingnbsp;result that all the Jurassic and Heooomian stems which arenbsp;termed Cycas-stems, so far as anything is known of theirnbsp;structure, belong to Bennettites, and that not a single one ofnbsp;them has been proved to be a genuine stem of Cycades. Thisnbsp;further shows how precarious is the identification of fossil remainsnbsp;when it rests on superficial characters only.�¹ ² The form of thenbsp;lateral fructification branches may be summarized as follows:nbsp;They occasionally arise exactly over a leaf base, but in somenbsp;cases have been pushed somewhat to the side; as a rule itnbsp;appears impossible to �determine the precise relative positionnbsp;of the two kinds of organs.�² The flower-bearing shoots are,nbsp;at all events, not terminal structures, as in recent cycads. Eachnbsp;fertile branch is made up of several short internodes, and bearsnbsp;spirally-arranged lanceolate acuminate bracts; in structure thenbsp;bracts agree for the most part with the larger leaf bases of the

iMain stem. The apex of the fertile shoots has the form of a hemispherical parenchymatous cushion, from 'which are given offnbsp;number of closely crowded stalks united into a club-shapednbsp;group; between these stalks or seed-bearing cords there occurnbsp;smaller structures, the so-called interstitial organs. External tonbsp;these is a comparatively broad parenchymatous band of tissue,nbsp;which arises from a lower level on the cushion than thenbsp;cords and interstitial organs. Surrounding this homogeneousnbsp;peripheral tissue we have several lanceolate bracts encirclingnbsp;the entire fructification. Towards the upper surface of thenbsp;spadix, and near its periphery, there are numerous seeds situatednbsp;Jnst internal to the homogeneous tissue, which becomes rathernbsp;*nore strongly developed towards the apex of the fructification.nbsp;Each seed lies in a flask-shaped pit, and is borne on a longnbsp;stalk or cord; in some seeds the structure of the embryo withnbsp;radicle, plumule, and two cotyledons, may he clearlynbsp;recognized. Solms suggests that the homogeneous tissue whichnbsp;overtops the spadix and contains the seeds, has been formednbsp;hy the union of interstitial organs. A surface view of thenbsp;terminal portion of the fructification would present a numbernbsp;of areolse, probably raised to the form of pyramidal projections,nbsp;^nd between these areolse or distal ends of interstitial organs,nbsp;there would be seen the narrow openings leading to the flask-shaped seed cavities.

In his description of Bennettites, Solms compares its fructification with that described by Saporta and Marion as WilUamsonia ^orierei from the Oxfordian of Yaches-Noires. This Erenchnbsp;specimen, with its histological details preserved in unusualnbsp;Porfeotion, has recently been thoroughly examined and describednbsp;Prof. Lignier, of Caen. His description clearly shows thatnbsp;^^porta�s and Marion�s species must be referred to Bennettites,nbsp;^^d is closely allied to the English species of that genus.^ Innbsp;the main Lignier�s description agrees with those of Carruthersnbsp;Solms; there are, however, one or two points in whichnbsp;the French author�s account is slightly at variance with thatnbsp;the former writers. Bennettites Morierei has the form ofnbsp;^ detached ovoid fossil, with a length of 55 mm. and a breadth

of 35 mm. At the base a fractured surface reveals the existence of a slightly convex receptacle, from which is given off a compactnbsp;cluster of long peduncles, each of which hears at its apex annbsp;oval seed. The seed-bearing peduncles are surrounded by severalnbsp;involucral bracts closely applied to the surface of the fruit,nbsp;�fumerous thin lamellae occur in association with the seminiferousnbsp;peduncles; to these Lignier applies the term interseminal scales.nbsp;The seeds are arranged side by side close to the upper surfacenbsp;of the mass of peduncles and interseminal scales; the latternbsp;passing between and beyond the seeds, and their swollen distalnbsp;ends forming a protective covering to the blunt hemispherical apexnbsp;of the fruit. In surface view, the upper part of the specimennbsp;appears to be made up of a large number of small projectingnbsp;areas with polygonal bases and rounded summits. Here andnbsp;there the projections arrange themselves in the form of rosettesnbsp;round a small central cavity, marking the position of a seed.

As in Bennettites Oihsonianus the fruit is covered by involucral bracts, but we have the interesting suggestion of Lignier thatnbsp;these were not simply linear acuminate in form, but that thenbsp;portion of the bracts preserved is merely the petiolar part of anbsp;leaf structure of which the pinnate or flabellate lamina has beennbsp;detached. This inference is drawn from a detailed examinationnbsp;of the vascular strands traversing each bract. After comparingnbsp;and contrasting Bennettitece with the Cyeadacem and Conifera,nbsp;Lignier concludes that they represent �a family which has beennbsp;derived with the cycads from common ancestors, but not from thenbsp;cycads themselves. Of these common ancestors the two familiesnbsp;have preserved the form of the trunk, the structure of certainnbsp;tissues, the foliar origin of the ovule, etc. But whilst the cycadsnbsp;have retained a grouping of carpophylls on a single axis, andnbsp;have acquired special characters, such as the complication of thenbsp;leaf-trace and the lateral position of the ovules, the Bennettitetenbsp;have retained the simple leaf-trace and have acquired a terminalnbsp;position of the ovules, the reduction of the fertile axes to singlenbsp;carpophylls, the grouping of these fertile reduced axes, and thenbsp;modification of the neighbouring leaves.� *

America, Lester Ward^ transfers Carruthers� species of Bennettites to Buckland�s genus Cycadeoidea; he considers the fact of thenbsp;fruit being known in a single species, B. Oihsonianus, is not anbsp;suiScient reason for the retention of the genus. Solms-Laubach ^nbsp;had previously restricted Carruthers� term to the species innbsp;which the characters of the fructification are known, viz.nbsp;Bennettites Gihsonianus; this use of the term is probably thenbsp;niost convenient, and it would seem much better to retainnbsp;Carruthers� name for stems bearing the bennettitean inflorescencenbsp;than to include these under such a comprehensive and purelynbsp;provisional genus as Cycadeoidea.

Bennettites (Cycadeoidea) Saxbyanus (Brown).

1870. Bennettites Saxbyanus, Carruthers, Trans. Linn. Soc. vol. xxvi. p. 698. 1874. Bennettites Saxbyanus, Schimper, Trait, pal. v�g. vol. iii. p. 658.

1878. Bennettites Saxbyanus, Carruthers, in Dixon�s Geol. Sussex, p. 281. 1894. Cycadeoidea Saxbyana, Ward, Biol. Soc. Washington, vol. ix. p. 80.

The specific name of this plant was proposed by Eobert Brown m honour of Mr. Saxby, who found the first specimen nearnbsp;Eonchurch, in the Isle of quot;Wight. The following definition ofnbsp;the species was given by Carruthers :�

�Trunk elliptical, with large medulla, and thin, much interrupted woody cylinder, vascular bundles passing upwards and outwards and breaking up into two rows of small bundles, whichnbsp;ure parallel to the superior and inferior surfaces of the petiole;nbsp;section of petiole sub triangular.�

In accordance with the narrower use of the genus Bennettites, as suggested by Solms and others, the present species ought to benbsp;transferred to Buckland�s Cycadeoidea, but for the present it may

be better to retain Carruthers� generic name, and by tbe addition of Cycadeoidea to indicate our incomplete knowledge of thenbsp;fructification. Cycadeoidea is employed, therefore, in the widernbsp;sense of the term, including stems with a circular as well as annbsp;elliptical transverse section.

46628 (V. 3233^). Figured by Carruthers, Trans. Linn. Soc. vol. xxvi. 1870, pi. Ivii. fig. 3, as Bennettites Saxbyanus.

Length 21 cm.; at one end the specimen shows the transverse and polished section as represented in Carruthers� figure; at thenbsp;opposite end the stem has been considerably worn. The varyingnbsp;degrees of wearing in this specimen show remarkably well thenbsp;striking differences in appearance presented by the several surfacesnbsp;of cortical or woody tissue exposed to view. Carruthers callsnbsp;attention to the striking similarity � between these fossil stemsnbsp;and the caudex of a tree fern�¹ ²; the U-shaped petiolar bundlesnbsp;shown in section on part of the specimen present a distinctnbsp;resemblance to the corresponding leaf-traces in fern stems. Brooknbsp;Point, Isle of Wight.nbsp;nbsp;nbsp;nbsp;Saxly Coll.

The surface of the specimen shows fairly good sections of petioles with numerous vascular bundles; the portion representednbsp;in Carruthers� figure shows the inner face of the wood, and anbsp;radial section through the cortex in which the course of thenbsp;leaf-trace bundles is clearly seen. I am unable to detect anynbsp;difference between the present specimen and that bearing thenbsp;number 38360, which has been labelled by Carruthers Bennettitesnbsp;Gihsonianus. In his definition of B. Saxlyanus, Carruthers speaksnbsp;of the petioles as subtriangular in section, whilst those ofnbsp;B. Gihsonianus are described as subquadrangular. Each of thenbsp;above specimens shows both forms of petioles, and in other respectsnbsp;there appear to be no real differences.

The figure shows very clearly the form of the petiole bundles as seen in a tangential section of the cortex. The resemblance

between these bundles and the leaf-traces of Protopteris punctata, Sternh., is fairly close. Other portions of this specimen demonstrate very clearly the mode of origin and course of the petiolenbsp;bundles. The tangential section of the wood, close to the innernbsp;(pith) face, presents the same appearance as figured by Carruthersnbsp;(pi. Ivii. fig. 6).

The figured portion is the smaller half of a fine specimen in the Museum Collection. The surface is made up of triangular cavitiesnbsp;�which constitute the meshes in a network of projecting ridges ofnbsp;interpetiolar ramenta. A few inflorescences, or rather the cavitiesnbsp;originally occupied by inflorescences, are seen in surface view,nbsp;and also in the longitudinally cut face represented in Carruthers�nbsp;figure. The woolly or finely fibrous ramenta appear to he identicalnbsp;�with the corresponding interpetiolar structures in some specimensnbsp;in the Eufford Collection from Ecclesbourne {cf. Y. 3177, Y. 2349,nbsp;and Y. 2132). There is a distinct similarity between the slightlynbsp;raised base forming the floor of one of the inflorescence cavities,nbsp;and the rounded receptacle seen in some of the Ecclesbournenbsp;examples of isolated inflorescences {Bennettites [ Willianisonia~\nbsp;Carruthersi), e.g. Y. 2129^, etc. Brook Point.

38363. A waterworn specimen showing petiole base projecting Somewhat beyond the ramental network. Cf. Y. 3234 (pi. Ivii.nbsp;fig- 4) and 38360. Solms ^ refers to this and some other specimensnbsp;as no doubt examples of Bennettitece, but without traces of fructification. An old label on the specimen gives Tilgate as the locality;nbsp;this has been altered to Brook. Brook.nbsp;nbsp;nbsp;nbsp;Mantell Coll.

Cf. Bennettites Saxbyanus.

bases. Between the petioles there appears to have been a fairly wide space occupied by ramenta. A side view of the specimennbsp;shows the inward prolongation of some of the petioles, andnbsp;demonstrates that the so-called bases as seen on the surfacenbsp;simply represent the level in the leaf armour to which disorganization has extended. The inner face of the block is probablynbsp;the outer surface of the wood. Ho fructification seen. Eccles-bourne.nbsp;nbsp;nbsp;nbsp;Rufford Coll.

Bennettites Gibsonianus, Carr.

1870. Bennettites Oiisonianus, Carruthers, Trans. Linn. Soc. vol. xxvi. p. 700. 1874. Bennettites Gihsonianus, Schimper, Trait, pal. v�g. vol. iii. p. 559.

1890. nbsp;nbsp;nbsp;Bennettites Gibsonianus, Solms-Laubach, Bot. Zeit. 1890, p. 789, pis.

1891. nbsp;nbsp;nbsp;Bennettites Gibsonianus, Solms-Laubach, Annals Bot. vol. v. p. 419,

1894. Bennettites Gibsonianus, Lignier, M�m. Soc. Linn. Normandie, vol. xviii. p. 76.

Type. Portions of a large block showing internal structures. British Museum and Kew.

The specimens on which Carruthers founded this species are recorded by him as of Lower Greensand age, and Solms-Laubachnbsp;has since confirmed this determination.^ The examples of B.nbsp;Saxlyanus are regarded as Wealden, and were obtained from Brooknbsp;Point, in the Isle of Wight; B. Gibsonianus was found in Luccombnbsp;Chine, Isle of Wight, by Mr. Gibson. Some of the specimensnbsp;which Carruthers has referred to the present species are apparentlynbsp;from Brook, and from the same beds as B. Saxbyanus.

� Trunk compressed, elliptical, with small medulla, and a thick subcontinuous woody cylinder; vascular bundles passing almost

directly outwards and breaking up into a double series of small bundles, which are parallel to the superior and inferior surfaces ofnbsp;the petiole, except that a loop is sent down from the upper seriesnbsp;into the centre of the petiole. The section of the petiole is sub-luadrangular.�

Lignier has recently extended this definition to include the characteristic features of the infiorescence; he makes the followingnbsp;addition to the above diagnosis';�

� Fruit haut de 3 centim. Fracties involuerales lanc�ol�es, acumin�es, d�pourvues de limbe. Foils �caiUeux peu larges,nbsp;souvent ventrus, d cellules allong�es perpendiculairement auxnbsp;faces. Graines longues de 3 d 4 millim., larges de 1'2 d 2�5, nonnbsp;anguleuses (ou d peine anguleuses), d�pourvues d��pidermenbsp;rayonnant. Assise r�tioul�e sol�reuse et form�e de cellules quinbsp;ont 60 fi de longeur sur 50 ju de large.�

The following specimens are included here as somewhat doubtful examples of Bennettites Gibsonianus:�

38360. Figured by Mantell as Clathraria Lyellii, Medals of Creation, vol. i, 1854 (edit. 2), p. 163 ; Petrifactions and theirnbsp;Teachings, 1851, p. 46; Geological Excursions ronnd the Isle ofnbsp;TFight, 1854, p. 214.

The old label on this specimen gives Tilgate Forest as the locality, but Carruthers has substituted Brook Point. The preservation of this waterworn example is very different from thatnbsp;of the larger blocks of Bennettites. The surface, as figured bynbsp;^lantell, shows the characteristic ramental network with thenbsp;tti-eshes occupied by petiole bases. Towards the right-hand uppernbsp;Corner of the specimen there appears to be an inflorescence shownnbsp;on the waterworn surface. In the absence of any well-definednbsp;inflorescence, and in view of the striking similarity to, if notnbsp;identity with, V. 3234 (labelled by Carruthers B. Saxhyanus), itnbsp;ii'ill be better to regard the exact position of this specimen asnbsp;somewhat doubtful. Solms * quotes Mantell�s description of thisnbsp;fossil, and says he believes he recognizes the specimen in thenbsp;British Museum (Geological Department) Collection as one on whichnbsp;the word � Brook �� is written in ink on the upper surface. There

can be no doubt as to the original of Mantell�s figure; the fact of its being figured is recorded on the label, but there is no wordnbsp;�Brook� written on it. No doubt Solms is referring to somenbsp;other specimen; in another place the same author speaks ofnbsp;Mantel!�s figure as a very good representation of a specimen innbsp;the Geological Department.

38361 and 38362. Solms' refers to this specimen (cut into two pieces, bearing the above registered numbers) as being withoutnbsp;fructification. On the smaller piece there appears to be inflorescences shown in transverse and longitudinal section. The label onnbsp;one piece bears the name Clathraria Lyellii. Possibly this specimennbsp;should be referred to Bennettites Saxhyanus.

V. 3232. Another waterworn specimen cut into three pieces, one transverse slice and two larger portions. It would seemnbsp;impossible to definitely refer this imperfect example to one ornbsp;other species of Bennettites.

Bennettites (Cycadeoidea), sp.

V. 3177. The chief interest of this specimen is in the numerous casts of inflorescences which occur on the stem. The surfacenbsp;features cannot be accurately made out; the portions shown innbsp;PI. XV. Fig. 1 exhibit the conical cavities originally occupiednbsp;by the fertile axes, and the impressions of petioles and ramentalnbsp;tissue. The whole surface presents a somewhat waterwornnbsp;appearance, and instead of showing a surface view of the petiolenbsp;bases in the form of rhomboidal sections, it has the form ofnbsp;a worn-down surface with an oblique view of petiole casts andnbsp;ramental tissue.

In longitudinal section the fertile axes have the form of cavities narrowed towards the distal end; these cavities were no doubtnbsp;originally occupied by the fleshy terminations of inflorescences.nbsp;The wall of such a cavity (PI. XV. Fig. 3) shows clearly preserved

casts of the bases of crowded involucral bracts. The bracts themselves are represented partly by cavities and partly by carbonaceous Blatter; between the bracts the woolly or hair-like ramentanbsp;Bre distinctly shown, and these, as frequently happens in suchnbsp;tissue in cycadean stems, have been permeated by a mineralizingnbsp;solution, and so preserved. The bracts appear to be somewhatnbsp;expanded distally (PI. XV. Pig. 2), as described by Ligniernbsp;in Bennettites Morierei (Sap. and Mar.).' In PI. XV. Pig 3 annbsp;inflorescence oast is shown, of which the surface is covered bynbsp;B fine and well-marked reticulation; in the upper portion of thenbsp;cavity this reticulum has the form represented in Pig. 5, a fairlynbsp;regular network formed by thin plates projecting from the surface,nbsp;Bnd some of the meshes are partially filled by round or ovalnbsp;fiodies suggesting very small seeds; the meshes thus filled arenbsp;less angular than those without seed-like bodies. In Pig. 4 isnbsp;shown a wax cast of the inflorescence cavity of Pig. 3, and innbsp;^ig. 6 the small dot-like depressions on the surface of thenbsp;cast correspond to the small bodies in the meshes of Pig. 5.nbsp;towards the base of Pig. 4, the reticulum is rather larger, andnbsp;there are no dots in the more basal portion, owing to the absencenbsp;of any � seeds� in this part of the inflorescence (Pig. 7). In thenbsp;'niddle of one side of this retioulately marked cavity there is anbsp;narrow longitudinal ridge, probably representing a median groovenbsp;^n the inflorescence. The portion �f inflorescence figured in PI. X.nbsp;^Ig. 4, should be compared with the present specimen. Purthernbsp;�eference is made to this stem in the description of the detachednbsp;Specimens referred to a new species, Bennettites Carrutliersi. Cf.nbsp;1*1- X. Pigs. 1 and 4, and Saporta�s figures of WilUamsonia gigas,nbsp;llnrr., in the Pal. Pranq., vol. iv. pi. xiii. fig. 2, and pi. xiv.nbsp;Hgs. 1 and 2. Ecclesbonrne.nbsp;nbsp;nbsp;nbsp;Buford Coll.

V. 2896. Imperfectly preserved stem, 26 cm. in length, and Cm. in broadest part; leaf-stalks and ramenta shown, hut nonbsp;Inflorescence.nbsp;nbsp;nbsp;nbsp;Bechles Coll.

Bennettites (Williamsonia).

In the Rufford collection of quot;Wealden plants from the neighbourhood of Hastings, there are several specimens which must be assigned to the same position as the well-known Jurassic WilUam-sonia. For reasons which are stated more fully below, I havenbsp;referred these Wealden fossils to the genus Bennettites, and amnbsp;led to regard them as portions of the inflorescence of that plant.nbsp;Hitherto typical Williamsonias have not been recorded from anynbsp;Wealden or Lower Cretaceous rocks in England ; the importancenbsp;of the discovery is considerably increased by the fact that thenbsp;specimens appear to throw some new light upon the nature andnbsp;afiinity of this anomalous form of inflorescence. Before describingnbsp;the individual fossils in detail, it may be convenient to give anbsp;short summary of our present position with regard to the opinionsnbsp;of palseobotanists on the nature of Williamsonia.

In A Geological Survey of the Yorkshire Coast, by Young and Bird, published in 1822, there is a figure of a specimen fromnbsp;the ironstone of Saltwiok, which is spoken of as resembling tbenbsp;head of an artichoke {Cynara integrifolia), with the �coveringnbsp;or calyx consisting of numerous lanceolate and striated leaves.�^nbsp;Another figure in this work represents �a petrified nut of anbsp;singular kind �; these two fossils are examples of what wasnbsp;subsequently named Williamsonia?

Part of a oycadean frond from the � � Oolitic rocks of Scarborough � is figured by Bindley and Hutton, and named by them Zamiagigas^',nbsp;a few years later Williamson notes the occurrence with this formnbsp;of frond of � a remarkable fossil, apparently connected with thenbsp;fructification of a CycasB ^ In 1849 Tates� draws attentionnbsp;to the identity of Zamia gigas, L. and H., Zamia Mantelli, Brong.�nbsp;and Cycadites lanceolatus, Phill. He recognizes the difficulty ofnbsp;connecting the leaves and stem with the peculiar form of inflorescence associated with them, and while favouring the view

that Zamia gigm and the fructification are parts of the same plant, admits the absence of any actual proof. In the same yearnbsp;Mantell¹ ² figures a specimen of Williamsonia in his Medah ofnbsp;Creation, as the fruit of Zamites laneeolata, and describes thenbsp;lanceolate involuoral bracts as concealing the seeds of the inflorescence. In a short communication to the Yorkshire Philosophical Society, quot;Williamson gives a vertical section of a restorednbsp;inflorescence, and represents the ovoid body of the fructificationnbsp;as terminating in a cone.^ Brongniart calls attention in hisnbsp;TahUau^ to the similarity between the Scarborough fossils andnbsp;^ specimen described by Bucklandi from the Inferior Oolite ofnbsp;Charmouth, Dorset, under the name of Podoca/rya. This comparison is one which has, I believe, been justified by recentnbsp;investigations, and it is highly probable that Buckland�s specimennbsp;is a particularly well-preserved bennettitean inflorescence. It isnbsp;to be hoped that this valuable specimen may be rediscovered,�nbsp;snd subjected to a careful examination.

The name Podocarya was chosen by Buckland for this Oolitic fossil, on the suggestion of Eobert Brown. It is described asnbsp;chiefly resembling the inflorescence of the recent genus Pandanus,nbsp;nnd by most subsequent writers it is included in the Pandanaoece.nbsp;Saporta,� in his volume on Types proangiospermiques, reproducesnbsp;flnckland�s figures, and substitutes Williamsonia for Podocaryanbsp;the generic name; Unger� had previously named the speciesnbsp;after Buckland.

In a specimen figured by Leckenby in 1864� we have leaves Palmommia pecten (Lind.) in close association with a smallnbsp;form of Williamsonia, which Hathorst afterwards referred tonbsp;the new species Williamsonia Lechenbyi. Carruthers,� writing

� Pal. Fran9. vol. ir. p. 127, pi. ccxxxviii. figs. 1-3, and pi. ccxxxix. fig. 1.

nbsp;nbsp;nbsp;Leckenby (A.), Quart. Journ. Geol. Soc. vol. xx. p. 77, pi. ix. fig. 4a.nbsp;This specimen is in the quot;Woodwardian (Geological) Museum, Cambridge. Itnbsp;�Srees closely -with Williamson�s � carpellary disk� of Williamsonia gigas,nbsp;t'Mr., except in its smaller size.

in. 1867, refers to the opinions of Tates and Williamson, and adds; �I have examined numerous specimens of this fossilnbsp;(TFilliamsonia gigas) in the British Museum, hut have heennbsp;unable to determine anything satisfactorily in regard to thenbsp;precise structure of this anomalous fruit. It presents so manynbsp;peculiarities unknown in the fruit of any modern cycad, thatnbsp;for the present at least, and notwithstanding its Zamia-Vike leaves,nbsp;I must consider it a doubtful cycad.�

In the volume of the Linnean Society�s Transactions for 1870 we have the first exhaustive treatment of this Oolitic fossil.�nbsp;From his intimate knowledge, both of the fossils and theirnbsp;manner of occurrence in the rocks of the Yorkshire coast,nbsp;Williamson was peculiarly fitted to attack this difficult problem.nbsp;Whlliamson describes and figures what he regards as the stem ofnbsp;Zamites gigas, L. and H.; the surface is made up of broadnbsp;lozenge-shaped areas, and is compared with the trunk of a recentnbsp;Cycas. One example is referred to as � obviously the apexnbsp;of a stem with portions of seven or eight diverging fronds.�nbsp;The fronds are next described, also certain structures spokennbsp;of as the �squamous peduncles� of the fructification. Thenbsp;greater part of the paper is, however, devoted to a detailednbsp;examination of the �organs of fructification.� Surrounding thenbsp;ovoid inflorescence we have a number of linear bracts constitutingnbsp;an involucrum ; usually these involucral leaves have been brokennbsp;off towards the base, and immediately below the broken endsnbsp;there is exposed an annular area of �radiating cells.� Innbsp;pi. lii. fig. 7, Williamson represents a specimen in which thenbsp;bracts have been completely preserved and are continued to thenbsp;base of the fructification. The central part of the whole structurenbsp;is occupied by a pyriform cavity which contracts apioally, andnbsp;then expands into a funnel-shaped appendage. In one of thenbsp;figured specimens the form of this central cavity is clearly seen,nbsp;and from its being filled with carbonaceous matter, it is assumednbsp;that the axis was originally a solid structure. A very similarnbsp;appearance is presented by one of KufEord�s recently discoverednbsp;specimens.^ The ring of radiating cells seen at the base of mostnbsp;of the examples is considered to be the lowest margin of a layer

of cells 'wMoli �formed a cortical layer, arranged vertically upon and extending over the entire surface of the pyriform axis.�nbsp;In another type of specimen figured by Williamson, we havenbsp;a disk-shaped structure with a central depression, and split upnbsp;peripherally into several short and narrow segments; this isnbsp;named the � carpellary disk,� and towards the apex of each ofnbsp;these rays there are said to he two small pits on the lowernbsp;surface. These pits are regarded as the places of attachmentnbsp;of ovules; hut it is now generally agreed that there is no goodnbsp;evidence for the existence of well-defined depressions whichnbsp;Could be described as marking the position of seeds. The generalnbsp;Conclusion arrived at, and expressed by a restoration of the wholenbsp;plant, is that the two forms of fructification are probably thenbsp;male and female organs of Zamites gigas; the commoner ovoidnbsp;fossil being the male flower, with a pyriform axis originallynbsp;invested by a deciduous antheriferous tissue, and the femalenbsp;flower being represented by the much less abundant �carpellarynbsp;disk,� with the ovules inserted in pits towards the tips of thenbsp;star-like rays.

As a convenient provisional name for these anomalous structures Carruthers instituted the genus WilUamsonia, and placed it innbsp;^ new tribe�Williamsonea. The genus is thus defined:' �Stemnbsp;Cylindrical, elongated, marked with the equal-sized, tumid,nbsp;rhomboidal scars of the fallen leaves. Leaves ovate-lanceolatenbsp;cr linear acuminate, segments numerous, attached to the rachisnbsp;fly the central portion, with small free margins; veins numerous.nbsp;Parallel, at the base slightly diverging into the free margins.nbsp;flowers terminal, stamens surrounding a fleshy axis, ovules bornenbsp;cn the upper surface of an orbicular laciniate spadix.�

Carruthers thus expresses himself-with reference to Williamson�s ^ork: � He has introduced a clearer apprehension of the differentnbsp;forms of the supposed organs of reproduction, by the suggestionnbsp;that the two kinds represent the different sexes, and by thenbsp;discovery of a seed-bearing spadix.��* In the second volumenbsp;cf his Plantes Jurmsiques^ Saporta considers the problem ofnbsp;'WilUamsonia,^ and denies the existence of any satisfactory grounds

for the supposed connection between the fronds of Zamites gigas, the cycadean stem, and the floral structures; the last-mentioned henbsp;prefers to regard as some primitive form of a monocotyledonousnbsp;inflorescence, probably a pandanaceous type, analogous to Yuccites,nbsp;Podooarya, Jdolirion, etc.

In the Palmontologica Indica there have been described by Oldham and Morris,¹ ² and afterwards by Feistmantel,� variousnbsp;specimens of Williamsonia. The latter author figures certainnbsp;cyead-like stems found in association with the WilUamsonmnbsp;fossils as belonging to the plant which bore the WiUiamsonia,nbsp;inflorescence. Further additions to our knowledge of thenbsp;distribution of this fructification were made by Hathorst in 1880,�nbsp;and he put forward the opinion that WiUiamsonia should benbsp;placed with the BalanopJiorem. This idea he afterwards abandoned,nbsp;and in a later paper¹ upholds the view that WiUiamsonia wasnbsp;the inflorescence of a plant bearing cycadean fronds. He givesnbsp;a restoration of Anomozamites minor (Brong.) bearing in thenbsp;forks of a branched stem large flowers of the type WiUiamsonianbsp;angustifolia, Hath. The restored species presents an appearancenbsp;certainly more suggestive of some extinct form of plant than ofnbsp;anything at present in existence.� Hathorst associates his othernbsp;species, W. Lechenhyi, with Anomozamites Lindleyanus, Schimp.,nbsp;and connects W. gigas, Carr., with Zamites gigas, L. and H.,nbsp;adding that possibly Braun�s Weltrichia may he regarded as thenbsp;inflorescence of Otozamites. In another place ¹ the same writernbsp;refers to the likelihood of a connection or identity of Bennettitesnbsp;and WiUiamsonia.

In Zigno�s second volume of the Flora fossilis formationis Oolithicce,�' certain specimens from Italian rocks are describednbsp;under the new generic name of Blastolepis; the figures of thenbsp;two species of this genus, B. otozamitis and B. fdlcata, Zig., arenbsp;strikingly suggestive of Carruther�s genus. There can he littlenbsp;doubt that these specimens should be assigned to the genusnbsp;WiUiamsonia.

Feistmantel (2). nbsp;nbsp;nbsp;^ lathorst (3).nbsp;nbsp;nbsp;nbsp;¹ Ibid. (4).

� Since this was written, I have had an opportunity of examining Hathorst�s specimens, and can bear testimony to the accuracy of his description.

Ie a recent volume of the Plantes Jurassiques, Saporta ^ gives S' full account of the W�lliamsoma question, and discusses atnbsp;length such views as have heen advanced as to the nature of thenbsp;genus. Several of the Tates' specimens, now in the Parisnbsp;M^useum (Jardin des Plantes), and many others are figured innbsp;Saporta�s monograph. Some of these figures were originallynbsp;drawn for Brongniart, whose intention it was to publish anbsp;memoir on the subject. In addition to this very importantnbsp;contribution by Saporta, reference should be made to a papernbsp;ly him in conjunction with Marion in the Comptes Eendus fornbsp;1881,^ and also to the account by the same authors in L'Evolutionnbsp;r�gne vegetal^ The genus Williamsonia, with Yuccites,nbsp;Goniolina, Weltrichia, and others, is included in the special classnbsp;of � Types proangiospermiques.� � Ces types sont appel�s parnbsp;nous des Proangiospermes, paree, qu�ayant en r�alit� precede dansnbsp;1�ordre des temps les Angiospermes v�ritables et ne pouvant �trenbsp;classes m�thodiquement parmi ces derni�res, ils se distinguentnbsp;pourtant tres nettement de tous les v�g�taux pass�s en revuenbsp;jusqu�ici, et qu�ils s��oartent a la fois et des Cryptogames etnbsp;des Gymnospermes, n�ayant d�ailleurs de points de contactnbsp;nppr�oiables ni avec les Cycad�es, ni avec les Salisburi�es, encorenbsp;moins avec les Conif�res.� � As an introduction to the examinationnbsp;of the genus, Saporta writes:� �Avec les WilUamsonia nousnbsp;nbordons un des probl�mes les plus difiiciles, un des sujets lesnbsp;plus controverses, mais aussi les plus ourieux, peut-�tre m�menbsp;lo plus remarquable de tous ceux que nous off re P ensemble desnbsp;plantes jurassiques.� He agrees with Brongniart that Williamsonianbsp;probably generically identical with Buckland�s Podocarga.nbsp;Some of Williamson�s conclusions he does not accept; thenbsp;� carpellary disk � of that author, Saporta regards as a terminalnbsp;Expansion of the male spadix. One important point to notenbsp;m reference to Saporta�s conclusions is his interpretation of twonbsp;forms of specimens; that figured by Williamson in his pi. lii.

^ Ibid. (2), Les Phan�rogames, vol. i. p. 235. � Saporta, Pal. Franq. vol. iv. p. 63.

figs. 3-6 ^ is described as the male inflorescence, and another form, of which several figures are given in the Paleontologie Frangaise,nbsp;is regarded as the female inflorescence. Like Williamson, Saportanbsp;sees in the fibrous layer at the base of the common form ofnbsp;Williamsonia an antheriferous tissue ; the pyriform central axisnbsp;he describes as expanding distally into a large structure like thatnbsp;figured by the English writer as the carpellary disk; thisnbsp;terminal expansion appears to have been readily separated bynbsp;a natural surface of disarticulation from the rest of the axis.nbsp;In the whole inflorescence, according to Saporta and Marion,nbsp;we have a male involucre surrounding a conical axis with itsnbsp;base enclosed in a circular zone marked by radiating striae; thenbsp;external edge of this zone being occupied by a number of smallnbsp;compartments of an irregular hexagonal form, which seem tonbsp;correspond to pollen-sacs. The basal zone represents the sterilenbsp;and persistent portion of an androphore, which, when complete,nbsp;covered the whole of the conical receptacle with a layer ofnbsp;staminal appendages. In the female inflorescence the bracts arenbsp;somewhat shorter; the centre was occupied by a more or lessnbsp;globular axis, having its surface marked out into a numbernbsp;of compartments arranged in the form of facettes grouped innbsp;rosettes; the general appearance of the whole structure beingnbsp;very similar to that presented by Buckland�s Podocarya. Thenbsp;ovules were situated in subcortical cavities, which communicatednbsp;with the surface by small openings; the latter appearing as thenbsp;central points of groups of comparatively small meshes of thenbsp;superficial reticulum. In 1869 Mori�re described a petrifiednbsp;fruit from the Oxfordian beds of Vaches-Noires (Calvador),¹ ² andnbsp;this specimen is regarded by Saporta and Marion as throwingnbsp;considerable light on the nature of Williamsonia. The recentnbsp;examination of this specimen by Lignier ^ has already beennbsp;alluded to; his work has afforded us very important data withnbsp;regard to the connection between Bennettites and Williamsonia.

Saporta draws attention to a close resemblance between the terminal infundibuliform expansion of Williamsonia and the fossil

dealt with many years ago by Braun.¹ ² Braun�s genus Weltriehia �was instituted for a specimen discovered in the neighbourhoodnbsp;of Baireuth, and referred to as a new genus of the RUzanthece.nbsp;Schenk ^ expresses himself as very sceptical as to the correctnessnbsp;of Braun�s description of this fossil, and does not consider it ofnbsp;any scientific value.

In discussing the systematic position of Williamsonia and other genera, Schenk very reasonably calls attention to thenbsp;absence of any satisfactory evidence in favour of a pandanaoeousnbsp;alliance; he suggests that possibly Bennettites and Willianuonianbsp;should be classed together, but refers to the absence of histologicalnbsp;characters as a serious obstacle to any decided conclusion. Innbsp;Zittel�s Handhuch, Schenk � refers to Williammnia as the femalenbsp;inflorescence of Bennettites, ' a.'nh. refers to ISathorst�s and Solms�nbsp;Work in support of this opinion. Solms-Laubach � agrees withnbsp;Nathorst and Saporta amp; Marion as to the absence of trustworthynbsp;evidence of the connection between Zamia gigas and WilUamsonia.nbsp;After referring to Mori�re�s important specimen, he concludes : ^nbsp;� I have no doubt, therefore, that this specimen belongs tonbsp;Bennettites, but in saying this I have no intention of prejudgingnbsp;the question of its relation to WilUamsonia ; for it is still possiblenbsp;that further discoveries may show the fructifications of Bennettitesnbsp;and WilUamsonia both belong to similar stems resembling thenbsp;stems of CycadacecB, and confirm the opinion of Williamson andnbsp;Carruthers. But until the truth of these conjectures is ascertained.nbsp;We must be content to leave the relationship of WilUamsonianbsp;Undetermined.� In another place, Solms� draws attention to thenbsp;probability that we have long been familiar with the malenbsp;inflorescence of Bennettites in the fossil known as WilUamsonia.

In his account of Bennettites etrusca. Cap. and Solms,� Solms figures and describes certain boat-shaped sacs which he regardsnbsp;as pollen grains. He goes on to say that, if his interpretationnbsp;of these structures be correct, the inflorescence of Bennettites

� Note on Bennettites in Saporta�s Pal. Fran(;. vol. iv. p. 303. � Capellini and Solms, p. 202, pi. v. figs. 7 and 8.

apparently bore anthers and pollen grains in its early stage of development, and afterwards assumed the female condition. Thenbsp;figures of isolated �pollen grains� are not thoroughly convincing;nbsp;but if Solms� opinion be correct, we have a most importantnbsp;addition to our knowledge of this interesting genus.

My examination of the Wealden examples of Williamsonia leads me to support the view that this problematical fossil isnbsp;generically identical with Bennettite�, and so far as our evidencenbsp;goes, we are, I believe, justified in regarding the former genusnbsp;as a form of inflorescence of the same type as that which hasnbsp;been found in organic union with bennettitean stems. Thisnbsp;opinion is chiefly based on the Wealden forms, but if the suggestednbsp;relationship or identity of these with Bennettites be admitted, wenbsp;have a very strong case for including the Jurassic species in thenbsp;same category. A critical discussion of the Oolite specimens mustnbsp;be deferred until the French material has been studied; such anbsp;question will be best dealt with in a later volume devoted tonbsp;the Jurassic flora.

As regards the question of male and female inflorescences, I am unable, to recognize any sexual difference in the variousnbsp;examples from the Wealden beds, and there does not seem tonbsp;be any good reason for regarding the so-called male Williamsoniasnbsp;among the Jurassic specimens, as in any way proved to be ofnbsp;that nature. In comparing Williamsonia with Bennettites we havenbsp;to rely entirely on the female inflorescence of the latter plant,nbsp;and it would seem that so far as our present evidence goes, wenbsp;have more reason for speaking of Williamsonia as the femalenbsp;inflorescence. As to the nature of the male inflorescence wenbsp;are still without any very satisfactory evidence.

The following records have been made of Williamsonia, showing a fairly wide distribution; but probably some of these speciesnbsp;cannot well be retained as trustworthy examples of the genus.

Sweden (Scania). W. angustifolia, Nath.

Bornholm. W. Forchhmnmeri, Nath.

W. creiacea, Heer.

W. [Blmtolepis] otozamitis, Zig.

W. {Slastolepis) falcata, Zig.

TF. [Slastolepis) acuminata, Zig.

W. recentior, Daws.

Williamsonia ?

W. virginiensis, Font. bU elocata, Lesq.

W. ? Biesii, Hollick W. Slanfordi, Feist.

W. of. gigas, Carr.

W. microps, Feist. nbsp;nbsp;nbsp;,,nbsp;nbsp;nbsp;nbsp;�

The second English species, W. LecJcenbyi, is founded on a small rayed disk like the large � carpellary disk � of Williamson, andnbsp;is regarded hy Saporta as a lobed terminal expansion and not annbsp;Involucre as suggested hy Nathorst. Saporta considers it possiblenbsp;that the forms referred to this species may be simply a �morphological variation � � of W. gigas.

Buokland�s Podocarya is transferred hy Saporta to Williamsonia,^ and it is highly probable that this plant is a bennettitean inflorescence.

The French species W. Pougneti, Sap.,' is founded on an imperfect specimen which does not admit of any exact determination, and hardly justifies the institution of a new species. W.nbsp;Zeilleri, Sap.,� is also founded on a very poor specimen, and cannotnbsp;lie diagnosed with any exactness. W. minima, Sap.,� recentlynbsp;described from Portugal, is represented by an imperfect clusternbsp;of small bracts, and cannot be accepted as a satisfactory recordnbsp;of the genus.

Saporta has pointed out that Heer�s species, W. cretacea, from Greenland, shows many points of divergence from the typical

' A fossil figured by Sharpe as Asterophyllites ? from South Africa resembles an involucre of Williamsonia ; but Hooker�s description of the specimen is morenbsp;suggestive of such a genus as Schizoneura (Sharpe, Trans. Geol. Soc. vol. vii.nbsp;[^]) p. 227). I am indebted to Prof. Rupert Jones for calling my attention tonbsp;Sharpe�s figure.

^ Saporta, Pal. Franq. vol. iv. p. 167. nbsp;nbsp;nbsp;� lUd. p. 127.

� Ibid. p. 124, pi. ccxxxvii. ; see also Saporta and Marion (2), vol. i. p. 234.

� Ibid. p. 181, pi. ccxxxiv. fig. 3,

� Saporta (1), p. 105, pi. xix. fig. 9.

form; ¹ ² it may possibly be an inflorescence of some cycad-like stem, but is far from satisfactory.^ One of tbe forms figured bynbsp;Zigno ^ {Blastolepis aouminata) is named by Saporta ² W. Italica.nbsp;Dawson�s W. recentior is an exceedingly poor specimen andnbsp;hardly worthy of any name.

It is difficult in the case of some of the above fossils to precisely define their geological horizon. The Bornholm species,nbsp;W. Forohhammeri, Nath.,� is from beds which have recently beennbsp;shown by Bartholin to contain 46 species of plants, of whichnbsp;25 are regarded as Ehaetic, and about 15 as indicating a Lowernbsp;Oolitic age.

The fragment described by Dawson� as W. recentior is from the Canadian Middle Creek series, which is compared with thenbsp;Patoot series of Greenland and the Dakota group of the Westernnbsp;United States.

The nnnamed Williamsonia is from the Kootanie formation, which is correlated with the Neocomian of Europe.�

Eontaine�s Potomac series includes rocks differing somewhat widely in age, Jurassic and Cretaceous strata being incorporatednbsp;in one formation.

Williamsonia ? Riesii is probably a true Williamsonia with numerous and unusually narrow bracts; the author of the speciesnbsp;compares it with some composite flower.

As regards the Indian beds, it has long been a difficult problem to determine their exact geological position; the Eajmahal bedsnbsp;are spoken of by Oldham,� in the recent edition of the Manualnbsp;of Indian Geology, as Liassic, the Each (Cutch) or Umia bedsnbsp;being correlated with the Upper Oolite, but the precise horizonnbsp;can only be approximately stated.

� Heer, Flor. foss. Arct. vol. vi. p. 59, pi. xii. fig. 1; pi. xiii. fig. 9. [Since writing the above I have examined Heer�s type specimen of W. cretaceanbsp;in the Geological Museum in Copenhagen, and find the impression is verynbsp;indistinct and unsatisfactory.]

* nbsp;nbsp;nbsp;Saporta, loc. cit. p. 180, pi. ccli. fig. 3 ; and pi. cclii. fig. 4.

Bennettites CWilliamsonia) Carruthersi, sp. nov.

Type. Specimens in the British Museum from the Pairlight clays, Pairlight, near Hastings.

As a matter of convenience, a new specific name is adopt�d for several specimens in the Eulford Collection. Without a fullernbsp;knowledge of their anatomical structure it is impossible to givenbsp;an exact specific definition, but the general characters may henbsp;briefly expressed as follows :�

Inflorescence ovoid, surrounded by numerous linear bracts, enclosing a central axis (from which seed-bearing peduncles andnbsp;interseminal structures were given oft); between the involucralnbsp;bracts and the periphery of the spadix there was a regularnbsp;reticulum of projecting ingrowths marking out the surface ofnbsp;the inflorescence into small areolations.

In dealing with detached inflorescences or portions of such structures, it is impossible to clearly discriminate between differentnbsp;specific forms, as distinct from portions of the same species ornbsp;the same inflorescence in different stages of development. Thenbsp;most important features in the following specimens are thosenbsp;which serve to connect them, on the one hand with the typicalnbsp;Sennettites, and on the other with the Jurassic Williamsonia.nbsp;Some of the examples of these Wealden forms differ from thenbsp;majority in having short and broad bracts at the base of thenbsp;inflorescence; these we may speak of as B. (Williamsonia)nbsp;Carruthersi var. latifolius. I have ventured to associate the namenbsp;of Mr. Carruthers with the present species; it is to his worknbsp;that we are primarily indebted for our knowledge of Bennettites.

This specimen is one of the most important of those to be described. In general form and appearance it is very similar tonbsp;Bennettites Mori�rei as figured by Lignier,* except that in thenbsp;present specimen the linear bracts are distinctly shown, about

twelve in number, and, as occurs so commonly in the Jurassic quot;Williamsonias, they are broken off basally, leaving an annularnbsp;area immediately surrounding the base. This annular area showsnbsp;very distinctly numerous parallel longitudinal striations; these Inbsp;regard as corresponding to the so-called � antheriferous tissue� ofnbsp;some writers in Williamsonia gigas, Carr. At the base we havenbsp;a well-defined rim surrounding a central short and conical cavitynbsp;(Fig. 1J); a similar form of axis occupied by carbonaceous matternbsp;occurs in the specimen represented in Fig. 8. Compare also Saporta,nbsp;pi. xxvi. fig. 3. Length of V. 3177 6 cm., breadth 3-5 cm. Innbsp;Fig. \a the truncated base of one of the bracts is represented as seennbsp;from below; this shows a number of regularly placed projectionsnbsp;from the face of a bract, extending from the latter to the internalnbsp;fibrous structures. In some of the figures of the inflorescencenbsp;of Bennettites Gihsonianus, Carr., given by Solms-Laubach,' therenbsp;are similar internal projections represented in the inner face ofnbsp;� the �outer layer� of the fructification. In describing thenbsp;structure of the inflorescence, Solms writes: � Not unfrequentlynbsp;sharp and tolerably deep indentations penetrate from without intonbsp;this homogeneous external layer ; these indentations are coverednbsp;with the epidermis, and probably answer to the cross-sections ofnbsp;a superficial areolation of the entire fructification; they are particularly well and clearly seen near the base of the spadix innbsp;fig. xii. of pi. XXV.� The same indentations are seen in Solms�nbsp;pi. XXV. fig. 8, fig. 10, and fig. 11. These ingrowths are, Inbsp;believe, the structures seen in our Fig. la, PI. X. In the longitudinal section of Cycadeoiiea etrusca, Cap. and Solms,^ figurednbsp;by Capellini and Solms, pi. iv. fig. 1, we have the structuresnbsp;clearly represented, and again, on a larger scale, by Carruthers^nbsp;in B. Qibsonianus (pi. lx. fig. 3). In his description of the latternbsp;species, Solms speaks of the interstitial organs as becoming muchnbsp;more numerous towards the periphery of the spadix (pi. xxv.nbsp;figs. 8 and 11); Lignier also refers to this character, and in anbsp;figure of the peripheral region of the fructification (p. 31, fig. 2),nbsp;shows the superficial bracts on the outside, and internal to thesenbsp;much smaller interseminal bracts and atrophied seed-bearing

peduncles. The fine longitudinal lines seen on the surface of the specimen figured by Lignier on pi. v. figs. 55 and 56, andnbsp;described by him as the peduncles, bear a very close resemblancenbsp;to the fine lines traversing the basal annular area in our figurenbsp;(H. X. Fig. 1), and to the so-called antheriferous tissue ofnbsp;WilUamsonia gigas. It is, I believe, the interseminal structuresnbsp;or possibly atrophied peduncles that are seen in such specimens ofnbsp;Bennettites Garruthersi as are represented in PI. X. Fig. 1, PI. X.nbsp;Pig. 2, PL X. Fig. 3, and PI. XI. Fig. 4. The central conicalnbsp;cavity was originally occupied by a fleshy axis, on which werenbsp;borne the seed-bearing peduncles. Ecclesboume. Rufford Coll.

In this specimen we have the outer part of the basal portion of a larger form, or older inflorescence. In the centre there is anbsp;somewhat oval area with an uneven surface 2-3 cm. in diameter,nbsp;surrounded by a series of slightly raised structures; external tonbsp;this is a concave rim, with its surface marked by a well-definednbsp;reticulum of projecting ridges. The appearance of this saucer-likenbsp;rim suggests that there were originally numerous narrow bracts innbsp;close contact with it; the outline of these being indicated by thenbsp;shallow depressions and intervening ridges as shown in the figure.nbsp;This specimen approximately corresponds to the basal portion ofnbsp;V. 3177 (PI. X. Fig. 1), the external margin of the reticulatelynbsp;marked rim coinciding with the truncated bases of the involucralnbsp;bracts of such examples as V. 3177 (PI. X. Fig. 1) and V. 2129nbsp;(PI. X. Fig. 2). The network is probably formed by the samenbsp;regular indentations which are seen in Fig. 1�, and described bynbsp;Solms* as forming a regular areolation over the surface of thenbsp;spadix. Between the reticulately marked rim and the centralnbsp;floss there would be the continuous external layer of the peripheralnbsp;Zone. Cf. Solms, pi. xxv. figs. 4 and 7-12. Ecclesboume.

Here again we have the characteristic Williamsonia gigas base, somewhat larger than in V. 3177 (PI. X. Fig 1). The linear bractsnbsp;are unevenly broken, exposing the numerous fine, interseminal.

fibre-like structures. Probably in this specimen the whole inflorescence is partially expanded, and is not so conical in form as the more completely closed example shown in PI. X. Pig. 1.nbsp;Cf. W. gigas as figured by Williamson (pi. lii. fig. 3), W. Gagniereinbsp;as figured by Saporta' (pi. xxvi. figs. 1-3), etc. Ecclesbourne.

The remains of a fructification similar to but rather smaller than V. 3177 (PL X. Pig. 1). Occupying the centre is a well-defined hemispherical boss, 8 mm. in diameter, differing from thatnbsp;in PI. X. Pig. 15 in its more spherical form; its surface is coverednbsp;with small punctations. At the base of this boss there appears tonbsp;be a small projecting rim, and external to this an almost verticalnbsp;involucre of narrow bracts. Pesting on the inner face of thesenbsp;bracts is a fibrous material, consisting probably of the .samenbsp;slender structures as those seen in V. 3202 (PI. X. Pig. 2), andnbsp;like them being the small interseminal scales, or peduncles,nbsp;which make up the peripheral portion of the spadix. Comparenbsp;B. Gibsonianus, Carr.: in that species the central boss is morenbsp;cushion-like in form. At a distance of about two-thirds fromnbsp;the base the bracts are bent outwards, as shown in the figure.nbsp;Pcclesbourne.nbsp;nbsp;nbsp;nbsp;Bufford Coll.

A star-like cluster of broader bracts, closely resembling such a form as Williamsonia [Blastolepis] -acuminata, Zig. (^=W. Balica,nbsp;Sap.).^ The bracts are arranged in a close spiral, and do notnbsp;form a true whorl. In some there are a number of fine hairsnbsp;obliquely attached to the margin; these are shown on thenbsp;lower margin of the middle left-hand hraot in the figure. Thenbsp;hairy margin corresponds with that in Cycadolepis villosa, Sap.,¹ ²nbsp;and our V. 2802,^ etc. This specimen probably represents a seriesnbsp;of bracts which surrounded the base of an inflorescence. Pcclesbourne.nbsp;nbsp;nbsp;nbsp;Bufford Coll.

The bracts in this specimen are of large size, and less complete. In the centre we have the remains of the base of the axis tonbsp;which seed-bearing peduncles were originally attached; ef. V. 3201nbsp;(PI. X. Pig. 4). On one side of the central boss a portion of thenbsp;fibrous annular ring is preserved, showing the impressions of somenbsp;of the peripheral structures of the spadix; external to this wenbsp;have traces of a reticulate structure resembling that of V. 3201nbsp;(Pi. X. Pig. 4). Ecclesbourne.nbsp;nbsp;nbsp;nbsp;Jtujford Coll.

Portions of large and spreading bracts surrounding a slightly irregular conical cavity; on the inner face of this there arenbsp;numerous fine reticulations. Probably of the same type asnbsp;V. 3177 (PL X. Pig. 1), but much less complete. Ecclesbourne.

This specimen (Pig. 8) shows the clearly outlined central structure, of conical form, similar to the smaller conical cavity

in V. 3177 (PI. X. Pig. 1). The impression immediately below the inflorescence is probably not in organic connection with it.nbsp;Preservation imperfect. Pairlight.

V. 2913. Involucral bracts, surrounding a central boss. Towards the base of one of the bracts, and on the surface of thenbsp;rook, there are some traces of the reticulate structure usuallynbsp;met with in this position. Ecclesbourne.nbsp;nbsp;nbsp;nbsp;Rufford Coll.

V. 3172. Pairly broad involucral leaves surrounding a small central cavity. Cf. V. 3177 (PL X. Pig. 1). Ecclesbourne.

V. 2913a. Small specimen showing a portion of the annular zone,' with narrow irregular projecting ridges towards its innernbsp;margin, passing towards the outer edge into distinct reticulations :nbsp;cf. V. 3201 (PI. X. Pig. 4), etc. In the centre is a slightlynbsp;raised boss, separated by a circular groove from the annularnbsp;zone ; in the circular groove was probably situated the � continuous external layer of the peripheral zone,� as figured andnbsp;described by Solms.¹ ² Ecclesbourne.nbsp;nbsp;nbsp;nbsp;Rufford Coll.

V. 2801. This . specimen shows a central portion like that in V. 3201 (PI. X. Pig. 4); surrounding this there are irregularnbsp;radiating lines gradually passing towards the periphery intonbsp;a regular reticulation. Ecclesbourne.nbsp;nbsp;nbsp;nbsp;Rufford Coll-

V. 29133. A circle of narrow bracts surrounding a central area. Cf PI. XI. Pig. 1 (V. 2793).

V. 2129 g. Two specimens with an involucre of broader bracts surrounding a central boss. It is difficult to decide hownbsp;far the breadth of the bracts may constitute a specific difference.nbsp;These examples are somewhat intermediate between the specimensnbsp;referred to the variety latifolius, and the narrow leaved forms.nbsp;Cf. V. 2793S and V. 2129/, etc.

V. 2129. Small specimens showing a central cavity bordered by an annular zone of fibrous structures.

Bennettites (Williamsonia) Carruthersi, sp. nov� var. latifolius mihi.

Compare with a figure of Williamsonia gigas given by Oldham and Morris. In the centre is a depressed boss, as in V. 2793nbsp;(PI. XI. Fig. 1), V. 3201 (PI. X. Fig. 4), surrounded by a narrownbsp;�'ing, and external to this a rim of fibrous structures, about I'Scm.nbsp;broad. At a lower level than the fine radiating fibrous structuresnbsp;(peripheral interseminal scales or abortive peduncles), we find herenbsp;nnd there an impression of a reticulately marked surface; thisnbsp;reticulum is probably identical with that in V. 3201 (PI. X.

4), and in V. 3177 (PI. X. Fig. 1). External to the annular ring are the blunt and rounded tips of a few broad bracts. Thenbsp;Specimen consists of the detached basal portion of an inflorescencenbsp;�een from the inside, Ecclesbourne.nbsp;nbsp;nbsp;nbsp;Rufford Coll.

quot;Very similar to V. 2129 (PI. XI. Fig. 4); an involucre of short �nd broad bracts like those of the preceding specimen, but in thisnbsp;case seen from the outside. In places where portions of the bractsnbsp;have been removed, a fine reticulate structure is seen on the rock-�nrfaoe. In the centre of many of the polygonal meshes a slightnbsp;riragniflcation reveals the presence of a small black dot of coaly

substance. A mesh witb such a central dot presents some resemblance to the section of a peduncle as figured by Lignier/ the dot representing the central vascular bundle. Eoclesbourne.

? Bennettites.

V. 2304, V. 3174, V. 3176. Indistinct remains of carbonized hairy bracts; may possibly belong to Bennettites. Also V. 2830nbsp;and V. 2816. Eoclesbourne.nbsp;nbsp;nbsp;nbsp;Buffbrd Coll.

V. 3163 and V. 3164. A short axis, about 4 cm. in breadth, terminating in partially expanded bracts. The bracts and hairynbsp;ramenta closely resemble those of V. 3177 (p. 144), and suggestnbsp;the same plant. May possibly be an axis which bore anbsp;Williamsonia inflorescence. Ecclesbourne.nbsp;nbsp;nbsp;nbsp;Huffordnbsp;nbsp;nbsp;nbsp;Coll.

V. 2305, V. 2306. May be portions of similar stems, but cannot be accurately determined. Ecclesbourne.nbsp;nbsp;nbsp;nbsp;Buffordnbsp;nbsp;nbsp;nbsp;Coll.

Y. 2349. A mass of hairy or woolly structures evidently arranged round some axis; pressed togethernbsp;nbsp;nbsp;nbsp;asnbsp;nbsp;nbsp;nbsp;so manynbsp;nbsp;nbsp;nbsp;thin

laminae. Very similar to the hairy scale leaves, nbsp;nbsp;nbsp;etc., in thenbsp;nbsp;nbsp;nbsp;stem

V. 2930. Woolly scales or masses of ramental structures. Ecclesbourne.nbsp;nbsp;nbsp;nbsp;Buffordnbsp;nbsp;nbsp;nbsp;Coll.

This generic name was proposed by Carruthers for a form of cyoadean stem originally described as Cycadeoidea Tatesii, Carr.¹ ²nbsp;The genus is thus defined ;�

� Trunk cylindrical, of uniform thickness, and covered with the short persistent bases of the petioles; scars of the abortednbsp;leaves scattered among those of the true leaves. Androeciumnbsp;unknown; gynoeoium forming a cone, each carpophyll of whichnbsp;hears two reflexed ovules.�

The characters of the floral structures are based on certain cones of somewhat doubtful affinity, which have not been foundnbsp;in actual connection with Yatesia stems, and cannot, therefore,nbsp;be regarded as of much value. In some of the specimens referrednbsp;to this genus, the general arrangement of the tissues, and tonbsp;some extent the histological structure, have been preserved, hutnbsp;the latter are but very imperfectly known. If we examine thenbsp;figures of stems included in Yatesia, we shall find it a difficultnbsp;task to distinguish some of the examples figured by Carruthersnbsp;from stems referred to the genus BuoUandia. The form of thenbsp;leaf-scars in Bucklandia Milleriana, Carr.,'² agrees closely withnbsp;that in the Yatesia stems; and in Y. Joassiana, Carr.,� therenbsp;appear to be distinct indications of the transverse constrictionsnbsp;m the stem, as in many bucklandian stems; the form of thenbsp;leaf bases is also very similar in the two genera. Questions as tonbsp;fiifferences in age, stages of growth, and the manner of preservationnbsp;of the stems, render the discrimination and exact limitation ofnbsp;generic types exceedingly difficult, or even impossible; my impression is, that at all events in some of Carruthers� species therenbsp;are no satisfactory grounds for the application of two genericnbsp;names. It does not seem possible to draw any distinct line of

separation between some of the smaller stems of Buchlandia and the yatesian form of trunk. Schimper includes Yatesia Joassiana,nbsp;Carr,,' in the genus Clathraria, and expresses surprise that itnbsp;has been referred by Carruthers to another genus.

The specimens of Yatesia Morrisii, Carr., described by the author of the species, were obtained from the Lower Greensandnbsp;beds of Potton and Leighton Buzzard. One of the examples innbsp;the National Collection is described as being from the � Wealden�nbsp;beds of Leighton Buzzard. The exact age of fossils from thesenbsp;beds must be a matter of some uncertainty, owing to the factnbsp;of many of them being clearly derived forms; it may be noted,nbsp;however, that some of the plant fossils from Potton appear to benbsp;specifically identical with Wealden types.

Among the genera instituted by Saporta, that of Cylindropodmm includes some forms of stems which bear a striking resemblancenbsp;to Carruthers� species of Yatesia.

Yatesia Morrisii, Carr.

1870. Yatesia Morrisii, Carruthers, Trans. Linn. Soc. vol. xxvi. p. 688, pi. Iv. figs. 3-6.

Type. Stem with internal structure imperfectly preserved. Royal Agricultural College, Cirencester.

� Stem cylindrical, covered with the bases of the petioles, which are rhomboidal in form, and terminate in a tumid boss, the apexnbsp;of which is directed upwards. , The cellular axis is very large.nbsp;The pith has disappeared, except in one specimen, where sufficientnbsp;of it remains to show that it was permeated with vascularnbsp;bundles. The woody cylinder surrounding the pith, in thenbsp;specimen figured, consists of two rings (figs. 5 and 6) ; it isnbsp;everywhere pierced by medullary rays, which are often so largenbsp;as to break the continuity of the wood (fig. 6). The sides of thenbsp;wood-cells parallel to the medullary rays are covered with disks

in two or three rows (pi. lx. fig. 13). The inner surface of the Woody cylinder exhibits numerous narrow grooves, being thenbsp;meshes for the passage outwards of the vascular bundles tonbsp;the leaves. These meshes are larger and more regular on thenbsp;outer surface of the wood. A very thin layer of cortical tissuenbsp;separates the wood from the base of the petioles. The basesnbsp;of the petioles spring from this layer, at right angles to the stem.nbsp;Externally they present a rhomboid form, the horizontal diameternbsp;of which is but little more than the perpendicular.�

Although the Potton and Leighton Buzzard sands, in which the examples of this species have been found, are of Lowernbsp;Greensand age, the plant may be reasonably included in thenbsp;present list as possibly a member of the Wealden flora.

47029. Portion of a stem preserved in oxide of iron. The inner face of the wood is shown, with the elongated medullarynbsp;ray cavities; an impression of this face would present a similarnbsp;appearance to the medullary cast of Buchlandia. Pith large,nbsp;surrounded by a zone of wood l�4cm. in width, and consistingnbsp;of two concentric rings. The transverse section does not shownbsp;any well-marked medullary rays traversing the wood. Thenbsp;petiole bases fairly distinct; some of them terminate apicallynbsp;in a manner suggestive of a clean surface from which the frondnbsp;tas been detached. Leighton Buzzard.nbsp;nbsp;nbsp;nbsp;Morris Coll.

V. 221. Two smaller specimens. A distinct variation in the size of the leaf bases; of. V. 2610. Leighton Buzzard.

Cf. Yatesia Morrisii, Carr.

The following specimens in the Eeckles Collection, from the Wealden rocks of Sussex, while agreeing fairly closely withnbsp;Yatesia Morrisii, possess certain points of resemblance to thenbsp;genus BucMandia. The external characters correspond to somenbsp;extent with those in Saporta�s genus Cylindropodium,^ but the

specimens from the French beds are in a better state of preservation, and admit of a more complete diagnosis, than thenbsp;English examples.

Leaf bases very prominent and considerably waterworn (Eig. 9, a and h). At the upper end of the specimen, as seennbsp;in Eig. 9(7, the outer surface of the wood is exposed; in a viewnbsp;of the transverse section (Eig. 9r) there is seen to he a singlenbsp;ring of wood, with the bundles separated by broad primarynbsp;medullary rays, the spaces in the structureless cast representingnbsp;the xylem bundles. The small development of wood points tonbsp;a young stem, in which no second cambium had been formed.nbsp;Cf. Carruthers� figures of E. Morrisii (pi. Iv. figs. 3, 5, and 6).�nbsp;Sussex.nbsp;nbsp;nbsp;nbsp;Bechles Coll.

V. 2607. Two specimens. The leaf bases present for the most part a different appearance to those in the other specimens; theynbsp;are closer together and not separated by the deep grooves shownnbsp;in Fig. 9 (V. 26105). Some of the bases, however, project exactlynbsp;as in the preceding example, and there can be little doubt as tonbsp;their specific identity. There is a distinct disparity in the sizenbsp;of the petiole bases. Sussex.nbsp;nbsp;nbsp;nbsp;BecTdes Coll.

V. 2610. quot;Waterworn stem with leaf bases as in V. 26105. There is only one ring of wood, hut this is no doubt merelynbsp;a matter of age, and shows that the stem was younger than thatnbsp;of 47029 (Yatesia Morrisii).

V. 2610r. Two impressions of the worn surface of a stem, or possibly a large cone. Cf. V. 26105 (Fig. 9). Sussex.

V. 2612. An impression of the outer surface of a stem, or less probably of a cone. Cf. V. 2607, also V. 2749� {BuoTdandianbsp;anomala, Stokes and Webb).^ Sussex.nbsp;nbsp;nbsp;nbsp;BecTdes Coll.

Geological Magazine in 1886 by Starkie Gardner, we find the following statement; � The stems of TEndogenites erosa, so commonnbsp;in the Wealden and Heooomian, are now known to be cyeadeous,nbsp;and it is probable that the BraccBna-\\\.e stems from Tilgate Forestnbsp;and elsewhere, so often referred to by Mantell, are referable tonbsp;the same group.�¹ ² Endogenites erosa is now recognized as a fernnbsp;(Vol. I. p, 148); but the Brmcena-XTka stems are in all probability,nbsp;as Gardner suggests, cycadean. Unfortunately no reasons arenbsp;given for this opinion. In the Second Eeport of the Committeenbsp;on British Tertiary and Secondary Beds, Gardner writes; ² � We

are not able to speak with certainty regarding the supposed liliaceous or Draccma-Yiks, stems from the Wealden, so frequentlynbsp;mentioned by Mantell, since it is not easy now to identify thenbsp;particular specimens referred to by him.� Mantell ^ refers to thenbsp;fossil stems discovered by Bensted at Maidstone, as nearly relatednbsp;to Yucca or Draccena. On examining the large specimens of stemsnbsp;in the British Museum, from the Iguanodon quarry ^ at Maidstone,nbsp;I was struck by their resemblance in external characters to thenbsp;stem of such recent cycads as Zamia Loddigesii, Miq., Z. SMnneri,nbsp;quot;Warsz., and Z. pumila, L.

In 1868 Carruthers referred to the Maidstone fossils under the name Braeana Benstedtii of E�nig,� and expressed his opinion thatnbsp;they exhibit a closer resemblance to the stem of a Pandanus thannbsp;to that of a Bracana-, but he refers to certain specimens in thenbsp;British Museum which appear to show the remains of internalnbsp;woody tissue, and thinks it possible that a closer examinationnbsp;might not lend support to the comparison with either mono-cotyledonous genus. In one or two of the Maidstone Kentishnbsp;Bag stems, there are portions of what closely resembles woodynbsp;tissue showing well-marked rings of growth, but a section cutnbsp;from this wood-like material proves it to be simply a deposit ofnbsp;carbonate of lime formed in such a way as to closely simulate thenbsp;structure of wood. One of these so-called Bracma stems wasnbsp;figured and briefly described in the Geologist for 1862,^ but thenbsp;drawing does not give a very accurate idea of the specimen.nbsp;In a footnote to a paper by Bensted, Mackie ^ points outnbsp;the absence of any flgure or description of Braccma Benstedtiinbsp;by K�nig.� Morris� gives this name as Konig�s, but adds afternbsp;the author�s name �British Museum,� and gives no reference tonbsp;any published account. In attempting to trace the geologicalnbsp;history of Monocotyledons, we are confronted on every hand with

� K�nig was sometime Keeper in the Mineralogical Department of the British Museum.

exceedingly doubtful fossils quot;wliicli cannot be relied upon as satisfactory records; many of the supposed oldest monocotyledonousnbsp;plants have been shown to be either inorganic fossils, or to belongnbsp;to some other class of plants. These Maidstone Lower Greensandnbsp;stems, and the smaller quot;Wealden examples of what appear to benbsp;the same form of plant, do not afford any trustworthy evidencenbsp;of the existence of angiospermous plants at this horizon. Thenbsp;resemblance to Dracmna or Pandanus does not bear the test ofnbsp;any careful comparison with the recent genera; in the fossils wenbsp;have none of the regular transversely elongated leaf-scars sonbsp;characteristic of these living monocotyledons. The method ofnbsp;branching and the general surface characters are much more innbsp;harmony with certain species of the genus Zamia. It may perhapsnbsp;be advisable to institute a new generic name for this form ofnbsp;fossil stem, but for the present we are chiefly concerned with thenbsp;small Wealden examples, and need not introduce any new term.

This specimen is probably the impression of a stem at a point where branching is taking place; the surface is deeply andnbsp;irregularly wrinkled, and studded with round or oval prominences showing no regularity of disposition. On the surfacenbsp;of the specimen there is a small amount of carbonaceous matter,nbsp;which probably represents altered cortical tissue. A comparisonnbsp;of this specimen with Braocsna, and with Zamia Loddigesii ornbsp;Slcinneri, shows a much more striking resemblance to the latter �nbsp;genus. In these forms of Zamia, as in the fossil stems, therenbsp;IS not the characteristic armour of petiole bases, but a surfacenbsp;marked by transverse and irregular wrinklings, with here andnbsp;there small knob-like protuberances. There is some slightnbsp;resemblance to Saporta�s Changarniera inquirenda ^; but this isnbsp;described as a leaf, and not a stem structure. Very similar tonbsp;some of the Maidstone specimens. Eoclesbourne. Ruffurd Coll.

X portion of the specimen shown in the figure. The surface IS slightly convex, suggesting a small segment of a large stem,nbsp;fhe surface markings resemble those of V. 3162, but in this

case they assume the somewhat more definite form of transversely elongated elliptical areas. Were it not for the convexity of thenbsp;specimen, one might, perhaps, he inclined to regard it as a pithnbsp;cast showing impressions of medullary rays. It closely resemblesnbsp;the stems referred to as Draccena Benstedtii. A very thin layer ofnbsp;a mineral substance occurs on the surface of the stem. Eocles-hourne.nbsp;nbsp;nbsp;nbsp;Rufford Coll.

V. 2322. Part of a smaller stem than V. 3162; at one end it shows the same form of branching as in the figured specimennbsp;(PI. XII. Fig. 5). Just below the place of origin of the largenbsp;branch there is a smaller branch or lateral appendage. Interrupted transversely running ridges and numerous circular andnbsp;elliptical soars constitute the surface features. Ecoleshourne.

V. 2170. An impression of what seems to be the surface of a stem like V. 2322, etc. Ecoleshourne.nbsp;nbsp;nbsp;nbsp;Rujford Coll.

Trunci, etc. (incertse sedis).

V. 2807. Small specimen of some cylindrical structure; surface marked with irregular longitudinal ridges ; apparently crushed.nbsp;Mr. Eufford suggests that this may be the axis on which anbsp;williamsonian fructification was borne. Ecoleshourne.

V. 2259. Similar specimen, but more like some pith cast. Cf. Fittonia insignis, Sap. (Pal. Fran9. vol. ii. pi. Ivi.). Eccles-hourne.nbsp;nbsp;nbsp;nbsp;Rujford Coll.

V. 2133. This specimen shows several more or less rhomboidal scaly structures, which may be the petiole bases of a oyoadeannbsp;stem. Ecoleshourne.nbsp;nbsp;nbsp;nbsp;Rujford Coll.

V. 3237. Possibly the impression of a petiole base, showing what appear to he the impressions of vascular bundles, and othernbsp;smaller black spots, which may be gum canals. Ecoleshourne.

Specimens of Doubtful Position.

In the second volume of the Plantes Jwrassiqms, Saporta figures two specimens under the generic name Cycadoraohis,^ C. armata,'^nbsp;Sap., from the Lower Kimmeridgian, and C. abscissa,^ Sap., fromnbsp;the same horizon. The former species is represented hy a fairlynbsp;stout axis bearing four spinous recurved appendages, having thenbsp;appearance of rose thorns. This is not unnaturally comparednbsp;with the raohis of a cycadean petiole, in which, as in recentnbsp;species of Cycas and Pioon, the lower pinnae are reduced tonbsp;spiny processes.

� Rachides frondium foliolis destitutee vel etiam frondium partes inferae, petioli dictae, sive nudae sint, sive aculeis armatae, autnbsp;ad basin insertionis causa paullo dilatatam squamatis e tomentonbsp;piloso constantibus ad utrumque latus praeditae videantur.�

Such a provisional genus like that of Paehiopteris among ferns, is a useful institution, and the species Cycadorachis abscissa maynbsp;Well be included in it; but the discovery by Mr. Eufford ofnbsp;several specimens very similar to Saporta�s C. armata, negativesnbsp;the suggested relationship to a cycadean frond.

In the Ecclesbourne (Hastings) specimens there are large leaflike structures attached to the axis in the axils of the spines, and, without attempting to speak definitely as to the precisenbsp;nature of these two kinds of appendages, it would seem unwisenbsp;to retain a generic designation indicating a cycadean alliance.nbsp;Although it is held by some a wrong course to adopt, I propose tonbsp;substitute, in the case of Cycadorachis armata, Sap., and the almostnbsp;identical fossils from the English Wealden, a new generic name innbsp;place of that instituted by Saporta. To retain Saporta�s genus, with

2 nbsp;nbsp;nbsp;lUd. p. 196, pi. cxvii. fig. 1.nbsp;* Ihid. p. 198, pi. cxiv. fig. 3.

the recently discovered specimens before ns, would be practically equivalent to assigning the plant to a position which appears tonbsp;be entirely at variance with the facts. I propose, therefore, tonbsp;institute the new genus Withamia for these spiny axes withnbsp;leaf-like appendages, and in doing so to place on record somenbsp;slight recognition of the immensely important service whichnbsp;Witham of Lartington rendered to palseohotanical science. Thenbsp;Internal Structure of Fossil Vegetahles'- is widely known as anbsp;classic work marking the beginning of a new method of investigation; but so far as I am able to discover, the name ofnbsp;the author of this epoch-making book has not been made usenbsp;of as a genus of plants. We may define Withamia as follows;�nbsp;A woody axis bearing two rows of spiny appendages, in thenbsp;axils of which are borne flat leaf-like appendages.

There is a very strong likeness between Saporta�s species, C. armata, and the English specimens as regards' the axis andnbsp;recurved spines, but the absence of any leaf-like appendages innbsp;the former, and the difference in geological age, render it advisablenbsp;to adopt a new specific name for the present examples of thenbsp;genus. I have chosen as a specific designation the name of thenbsp;author who first described this form of fossil plant. The speciesnbsp;may be defined as follows:�

Axis having a breadth of about 1 cm., striated longitudinally, bearing stout recurved spines arranged laterally in two rows,nbsp;and at slightly irregular intervals. ^ In the axils of the spinousnbsp;processes there are attached more or less orbicular or obcuneatenbsp;leaf-like structures, having a distinct flabellate (Cyclopteris type)nbsp;venation.

^ For Saportaia on PI. II, substitute Withamia; the former name being too nearly identical with Saportcea (Fontaine and White, The Permian or Uppernbsp;Carboniferous Flora. Harrisburg, 1880).

A purely provisional genus like Withamia seems decidedly preferable for the present species, to one which in any waynbsp;implies a definite botanical position. It is by no means clearnbsp;how such a plant can well be included in the Cycadacew, andnbsp;We have no evidence of sufficient value to enable us to assignnbsp;the species to any other particular group. In the English quot;Wealdennbsp;rocks there has not so far been found any trustworthy recordnbsp;of an angiospermous plant. The Conifers and Ferns may benbsp;mentioned as possible groups in which to include this species',nbsp;but as yet we have not sufficient evidence to warrant the selectionnbsp;of a generic name, which would imply a connection with one ornbsp;other of these sub-classes.

In a letter written about a fortnight before his death, the Marquis of Saporta wrote to me at some length in answer tonbsp;on expression of doubt on my part as to the cycadean naturenbsp;of his Cycaiorachis armata, and our more perfect English species.nbsp;The following sentences are taken from his letter, written onnbsp;January 10th, 1895'; his words may be quoted in full; theynbsp;are valuable, not merely as giving the opinion of one so wellnbsp;qualified to speak on such a question, but as some of the lastnbsp;from the ready pen of this indefatigable and accomplished student.

�Je suis en effet ravi 1�apprendre que vous avez rencontr� dans votre Wealdien nne portion de fronde, encore munie denbsp;pinnules en place de mon Cycaiorachis armata (Pal. franq.nbsp;Cyoadees, p. 195, pi. oxvii. fig. 1). Point de doute relativementnbsp;^ I�etroite conformite de votre echantillon avec le mien les epinesnbsp;sont �gales des deux parts et distribu�es de la m�me faqon surnbsp;le rachis. Du Kimmeridgien au Wealdien la distance verticalenbsp;D-�est pas telle que la m�me esp�oe de Cycad�es n�ait pu senbsp;maintenir et r�paraitre sans changement appr�ciable. Le typenbsp;ost assur�ment curieux, et m�rite d�obtenir une denominationnbsp;g�n�rique. A votre place je donnerais �, ce type de Cycad�es lenbsp;Hom d.'Acanthozamites. Eemarquez d�abord, cher Monsieur, qu�ilnbsp;H�y a dans la pr�sence de ces �pines ac�r�es dispos�es le long dunbsp;rachis de la fronde rien d�insolite et pour en �tre persuade vousnbsp;H�avez qu�d consultor la figure 1, pl. xi. du volume des Cycad�esnbsp;de mes plantes Jurassiques. Cette figure represente la base d�une

fronde de Cycas avec les �pines, qui sauf la dimension plus petite sont pareilles �, celles du type fossile Wealdien et Kimmeridgien.nbsp;Seulement au lieu d�un Cycas ou d�un Dioon, comme je Ienbsp;pr�sumais, nous avons ici un type eteint dont les frondes portaient.nbsp;En m�me temps des �pines et �, leur aisselle vers milieu du rachisnbsp;des folioles, sans doute caduques dont Ie ressemblance avec cellesnbsp;du Sphenozandtes latifolius, Brong. (Pal. franq. Cycad�es, pl. cxvii. etnbsp;pis. cxii. and cxiii. pour les folioles de Sphenozamites), doit �tre prisenbsp;en consideration. II 1'agit seulement de d�cider si la presence denbsp;ces �pines constitue un caract�re g�n�rique ou seulement sp�oiflque,nbsp;puisque nous savons par l��chantillon de ma planclie cxii., que Ienbsp;rachis du Sphenozamites latifolius n��tait pas arm� d�appendicesnbsp;�pineux. Je crois que dans 1�incertitude on est fond� d reconnaitrenbsp;au moins dans cette particulit� 1�indice d�un sous-genre ou sectionnbsp;d part qu�il est naturel de designer par un denomination d part,nbsp;comme serait celle de Acanthozamites que je proposais plus haut ounbsp;toute autre d votre convenance.

Mon sentiment est done ici que Ie rapprochement avec les Fhyllocladus n�aurait aucune vraisemblance tandis que celui avecnbsp;les Cycad�es et les Sphenozamites en particulier dont �tre adopt�nbsp;comme Ie plus naturel.�

My reason for not adopting the genus Acanthozamites, as suggested hy Saporta, is that it suggests a cycadean affinitynbsp;�which is hardly supported by the nature of the specimens. Asnbsp;regards the spines, these in themselves are by no means opposednbsp;to a cycadean raohis, hut the structures in their axils seem tonbsp;me quite inconsistent with the morphological character of anynbsp;recent cycadean frond. Mr. Carruthers suggested to me that thenbsp;specimens show some resemblance to certain ferns, and expressednbsp;an opinion in favour of the Filicince as the most likely plantsnbsp;with which to compare the fossil species. We have various recentnbsp;fern fronds which are more or less spinous �; but the general habitnbsp;of the fossil form, the nature of the spines, their disposition onnbsp;the axis, and their definite relation to the leaf-like structuresnbsp;constitute important points of divergence from any living Filieina-

' An interesting form of fossil fern, Gleiehenia Smtonensis, is figured by Starkie Gardner from the Eocene plant-beds of Bournemouth, in which stronglynbsp;recurved climbing organs are preserved. [Gardner (2), p. 60.]

In answer to my question as to the probability of such a fern-afamp;nity, Mr. Baker, of the Kew Herbarium, wrote as follows; �I should not think this very curious fossil is likely to be anbsp;fern. Phyllocladus seems far more likely. But, of course, withoutnbsp;flower and fruit there can be no certainty. The climbing stemnbsp;and hooked prickles recall Calamus.quot; I had suggested thenbsp;possibility of the quot;Wealden plant being compared with the Hewnbsp;Zealand conifer, Phyllocladus,^ in the recent species of which wenbsp;have small scaly leaves subtending flattened cuneiform branchesnbsp;(phylloclades). If the leaves were modified into climbing-hooks,nbsp;�We should have a fairly close approximation to Witliamia, butnbsp;the evidence at hand does not allow of any great weight beingnbsp;attached to such a comparison. In a palm such as Calamus ornbsp;Pesmoncus we find somewhat analogous spines, but in thesenbsp;ttonocotyledonous plants, there is not the same relation as regardsnbsp;position on the axis between hooks and leaf segments as innbsp;l^ithamia. For the present, then, I propose to leave the positionnbsp;of Withamia an open question, in view of the difficulty of decidingnbsp;the morphological value of the stout recurved spines and leaf-likenbsp;appendages, and the insufficient evidence afforded by incompletenbsp;�vegetative structures.

Length of axis 12-2 cm., about 1 cm. broad, striated longitudinally. In the axil of each stout recurved hook there occurs a portion of a leaf-like appendage; these leaves or phyllocladesnbsp;are imperfectly preserved, but enough is seen to demonstrate thenbsp;flabellate venation, and to suggest a form similar to that ofnbsp;the detached �leaf� represented in PL II. Pig. 1 (V. 2195).nbsp;The markings shown on the surface of the uppermost left-handnbsp;spine are merely cracks, and not the remains of any originalnbsp;structure; the two highest spines are attached to the axis innbsp;a manner indicative of an alternate arrangement, the middle pairnbsp;are opposite, and the lowest subopposite. Ecclesbourne.

sessile; the veins are numerous and clearly shown. The irregularity of the margin is probably an original character; in some other examples of these �leaves� the margin is much more indented,nbsp;and the form of the �leaf� longer and narrower: cf. V. 2134�,nbsp;V. 2798, etc. There is a distinct similarity between this specimen and an unusually entire Ginkgo leaf. Saporta�s figure of anbsp;Sphenozamites latifolim'^ leaf agrees very closely with the Wealdennbsp;specimen; if this form of leaf were attached to Withamia (Cycado-rachis) armata it would make the resemblance between the Frenchnbsp;and English species still more striking. Ecclesbourne.

Axis 18 cm. in length, and 9 mm. broad. The longitudinal striations very clearly shown, also the stout nature of the hooks.nbsp;Very incomplete fragments of the flat appendages in the spinenbsp;axils. The spines are less regularly placed than in V. 2134nbsp;(PL II. Fig. 2) and farther apart. Cf. CyoadoracMs armata, Sap-Ecclesbourne.nbsp;nbsp;nbsp;nbsp;Ruffordnbsp;nbsp;nbsp;nbsp;Coll-

V. 2805. A short piece of an axis with two well-preserved recurved spines. Ho leaves shown.

V. 21343. Portion of a large leaf, apparently about 7-8 cm-in length. A good specimen of Sphenopteris Fontainei, Sew., on the same piece of rook.

V. 2923. Axis 43 cm. long, l-l�3cm. broad. Portions of four hooks seen on one side, smaller than those of 2134(2 (PI. V. Pig. 1).nbsp;The appearance of this specimen is suggestire of a hollow axis,nbsp;hut this is probably due to the preservation of the cortex apartnbsp;from the internal woody tissue. Eeolesbourne. Eufford Coll.

The specimens included under this genus are difficult to describe quot;with any completeness, on account of the fragmentary andnbsp;imperfect nature of the material. So far as I have been ablenbsp;to discover, it is impossible to include these fossils in any knownnbsp;genus; the above name is therefore proposed as a convenientnbsp;generic term, and one which does not imply any exact botanicalnbsp;position. The National Museum owes some of its valuablenbsp;examples of Mesozoic plants to the enthusiasm of the latenbsp;Mr. Beokles; I have therefore made use of his name as a genericnbsp;designation. As a specific name for the few examples referred tonbsp;the genus Becklesia, the term momaia may be adopted.

Axis comparatively broad, giving off (on one side ?) stout and stiff branches, attached to the axis of higher order in differentnbsp;positions, either laterally or on the surface, and at irregularnbsp;intervals. On one surface the branches show a number ofnbsp;Parallel longitudinal striations, and on the other surface a broadnbsp;median rib with a small groove on either side.

The specimens are, however, too imperfect to admit of any ^ntisfactory generic or specific diagnosis.

In his monograph on fossil cycadean stems, Carruthers makes n brief reference to a specimen found at Maidstone, which isnbsp;spoken of as possibly a bennettitean frond. This fossil bears

in some respects a distinct resemblance to the present species. Carrnthers thus describes the Maidstone specimen: �It is a verynbsp;large leaf, with numerous long linear segments, attached verynbsp;obliquely to the rachis. The segments are simple on the uppernbsp;part of the frond, hut the lower ones give off, at regular distances,nbsp;several long and slender ultimate segments.�' Mr. Carrnthersnbsp;afforded me an opportunity of examining these Lower Greensandnbsp;specimens, and suggested that the Wealden examples figured innbsp;PI. XIV. Pigs. 2 and 3, represent the lateral segments of thenbsp;Maidstone �frond� with their lateral long and slender segments.nbsp;Although there undoubtedly exists a distinct resemblance, yetnbsp;one cannot speak at all positively as to the identity of the twonbsp;sets of fossils. In the Wernsdorf flora of the northern Carpathians,nbsp;an abundant and characteristic species is that described bynbsp;Ettingshausen as Thuites Hoheneggeri, Ett.,^ and afterwards bynbsp;Schenk as Frenelopsis Hoheneggeri (Ett.).^ Of the specimens sonbsp;named, some of those figured by the latter author present a morenbsp;or less close resemblance to BecMesia anomala; this is especiallynbsp;the case with those represented in Schenk�s pi. v. figs. 1 and 2.nbsp;Frenelopsis was proposed by Schenk as a generic name for plantsnbsp;having a similar habit to the recent genus Frenela, and possessingnbsp;among other characters cylindrical articulated branches bearingnbsp;small scaly leaves. The majority of the Wernsdorf examples of thisnbsp;genus show these characters very clearly, hut those in pi. v. figs. 1nbsp;and 2 are apparently without them, and in some degree conformnbsp;to the present species. IVithout examining Schenk�s material,nbsp;it is impossible to speak definitely as to the exact nature of thesenbsp;particular examples; it may be that a difference in age or mannernbsp;of preservation, is sufficient to account for the apparent absencenbsp;of the articulations and small leaves. Schenk^ speaks of these twonbsp;specimens (pi. v. figs, 1 and 2) as older examples, in which thenbsp;leaves are only partially preserved. The chief point of contactnbsp;between the quot;Wealden fossils and those from the quot;Wernsdorf beds,nbsp;described as older portions of Frenelopsis Hoheneggeri, lies in the

^ Ettingshausen (A. 4), Abh. k.-k. geol. Eeichs. vol. i. Abth. iii. No-p. 26, pi. i. figs. 6 and 7.

long and stiff lateral branches. In the English specimens there are no signs of any articulations or of leaf structures; so thatnbsp;they cannot well be included in the genus Frenelopsis. Thenbsp;typical form of the genus is well illustrated by Ettingshausen�snbsp;figures, as also by those of Heer^ and Saporta.^ In Fontaine�snbsp;Potomac Flora,^ several specimens are referred to Schenk�s genus,nbsp;hut these have recently been transferred by Nathorst to a newnbsp;genus, Pseudofrenelopsis,^ on the ground that the American formsnbsp;have been incorrectly interpreted by Fontaine. As regards habit,nbsp;there is some slight resemblance between Beehlena and Camptopterisnbsp;spiralis, Hath.,^ from Bjuf. On the whole, however, Carruthers�nbsp;specimens offer the greatest similarity to the following fragments,nbsp;the nature of which must be left entirely unsettled.

In the portion of the specimen represented in the figure the characters of the lateral segments are fairly clearly shown. A smallnbsp;piece of the branch at the right-hand upper corner of the drawing,nbsp;shows the parallel striation and apparently woody nature of thenbsp;segments; a little below, this branch is crossed by another innbsp;which the broad median ridge may be seen. Most of the lateralnbsp;segments are flattened, and do not present such distinct surfacenbsp;features.

The flattened main axis fairly distinct, with the irregularly placed lateral branches. The third branch from the top does notnbsp;appear to arise laterally, but rather from the exposed face of thenbsp;firoader axis.

� Heer (A. 6), Seoc. Trab. Geol. Portugal, p. 21, pi. xii, figs. 3-7. (The specimens figured by Heer in his Flor. foss. Arct., and referred to Frenetopsis,nbsp;Me probably not true examples of this genus.)

Cf. Becklesia anomala, sp. nor.

The nature of this specimen is very doubtful, and its imperfect preservation does not allow of any accurate description. Lengthnbsp;46 cm., breadth about 1cm. From the central flattened axis anbsp;number of comparatively straight lateral branches are given offnbsp;at irregular intervals; many of these appendages are separatednbsp;from one another by about 1'�cm., and have a breadth of 3 cm.;nbsp;they are linear in form and of a uniform breadth; one branch, innbsp;which the tip is not shown, measures 13 cm. There appear to benbsp;a small number of parallel veins in each segment. In some casesnbsp;the branches appear to bifurcate close to the point of attachmentnbsp;to the central axis. Possibly we have here a larger specimen ofnbsp;Becklesia, but the occurrence of branches on both sides of the axisnbsp;makes it difficult to be at all certain as to specific or even genericnbsp;identity with the previous specimens. Among recent plants therenbsp;is a form of Macrozamia heteromera, M. heteromera var. glauca,nbsp;Moore,' in which the bifurcate pinnae bear a certain resemblancenbsp;to the Wealden fossil, but in the former the more regular disposition of the segments affords an important point of divergence.nbsp;Cf. Schenk�s Frenelopsis Holieneggeri (Ett.),^ as shown in pi. v.nbsp;figs. 1 and 2; also Camptopteris spiralis, Hath.�nbsp;nbsp;nbsp;nbsp;Beckles Coll-

V. 2608.

�Frons bipartita, bipinnata, rachide primaria, crassa, striata. Pinnie liberee, pinnatifidae, alternae, vel suboppositae. Pinnulaenbsp;coriaceae integerrimae, saepe basi angustatae, in racbides alatasnbsp;decurrentes. Nervi aequales pauoi, simplices, interdum furcati,nbsp;e rachide seriatim orti, ad apicem marginemve pinnularumnbsp;flabellatim. exourrentes. Sori rotundi, prominuli, sparsi. Capsulaenbsp;(Sporangia) ovato-globosae, sessiles, vel subsessiles, annulo completonbsp;cinctae. Filices elastioae, rachide crassa bipartita, facie Gleiobenia-cearum.�

Zigno�s genus is classed by Solms-Laubacb' with Wilssonia, Thinnfeliia, and others, which �have been shifted backwardsnbsp;and forwards by different authors from cycads to ferns, and fromnbsp;ferns to cycads.� In looking over the references to Dichopterunbsp;various writers, we find a considerable difference of opinion,nbsp;both as regards the necessity for such a generic designation, innbsp;distinction to the much older genus Pachypteris of Brongniart,nbsp;and as to the affinity of the plants described under this name.nbsp;Schimper, in the first volume of the Trait, pal. v�g.^ includesnbsp;Pichopteris vxAm Pachypteris, and remarks that it is �impossiblenbsp;to doubt the identity of the two genera �; in the third volume �nbsp;of the same work, he includes the former among the terns as annbsp;independent genus. Saporta includes some species of Bichopterisnbsp;in his genus Scleropteris ^; e.g. the two plants originally describednbsp;iiy Phillips from the Lower Sandstone and Shale of the Yorkshirenbsp;Coast as Sphenopteris lanceolata,^ Phill., and Neuropteris Icevigata,^nbsp;PhilL, but the nature of these species has been a matter of

much discussion and need not he considered here. Teistmantel ¹ ² describes certain plant remains from the Gondwana flora ofnbsp;India, which closely resemble Zigno�s species; he prefers to gonbsp;hack to Brongniart�s Pachypteris, and extends the original deflnitionnbsp;so as to make it embrace, not only plants with the ultimatenbsp;segments � enerviis vel uninerviis�^ hut those in which the veinsnbsp;are more numerous. In Schenk�s monograph on Die fossile Floranbsp;der QremscMcTiten .... a specimen is figured as Bichopterisnbsp;incisa, Schenk,� hut, as Beistmantel suggests,'² the characters donbsp;not seem to agree with Zigno�s genus. The larger and morenbsp;perfect specimens of Bichopteris figured by Zigno � would seem tonbsp;favour the inclusion of such plants among the Filicinlt;s; but, asnbsp;Schenk points out,� the fructification is too indistinct to be ofnbsp;any taxonomic value. It is safer, therefore, while expressingnbsp;a bias towards the pteridophytic nature of the genus, to speaknbsp;of it as occupying a somewhat doubtful position.

Dichopteris, sp. Cf. D. laevigata (Phill.).�

V. 3145. Part of a single pinna, showing the coriaceous ultimate segments without any distinct venation.

Cf. Bichopteris Visianica, Zig., B. Icevigata (Phill.), and Scleropteris Pomelii, Sap.� Ecclesbourne.nbsp;nbsp;nbsp;nbsp;Rufford Coll.

� = Neuropteris l(evigata, Phill., Fachypteris Icevigata (Phill.] Icevigata (Phill.).

Stem much branched, leaves usually small and simple. Flowers unisexual and without a perianth, plants monoecious or dioecious.

The past history of the Gmiferm is but imperfectly known, and, owing to peculiar difficulties connected with the determination ofnbsp;fossil forms, the evidence of palseobotany as to the developmentnbsp;and geological distribution of these plants, must he accepted withnbsp;the greatest caution. It would take us far beyond the limits ofnbsp;the present work to discuss at length the distribution in time ofnbsp;coniferous types. In the Palseozoic rocks there are various representatives of this Class, and we have an example, in such annbsp;extinct genus as Cordaites, of a synthetic type in which coniferousnbsp;characteristics are combined with certain structural features metnbsp;with in other Orders of gymnosperms. As a general rule, fossilnbsp;conifers are perhaps the most unsatisfactory plants with which thenbsp;palseohotanist has to deal: structureless and imperfectly preservednbsp;fragments of broken twigs, isolated cones, leaves or seeds, havenbsp;usually to be determined separately, and it is only in comparativelynbsp;rare instances that we are in a position to connect cones and vegetative branches. Coniferous wood, with its mineralized tissues morenbsp;or less well defined, is met with in rocks of nearly every age,nbsp;hut here, again, the stems or thick branches must be determined asnbsp;far as possible from histological structure alone, and without anynbsp;leafy twigs or reproductive organs. G�ppert,¹ Kraus,¹ Kleeberg,¹nbsp;Felix,¹ Schenk,^ Knowlton,� and others have attempted to devisenbsp;convenient methods of classifying and identifying fossil Coniferm bynbsp;means of the peculiarities of structure presented by the secondarynbsp;Wood and the distribution of resin ducts. For the most part,nbsp;however, fossil conifers are represented by structureless casts ornbsp;impressions of leafy branches, occasionally hearing characteristicnbsp;cones or other forms of reproductive organs.

In treating of the Cycadaoece, some general account was attempted of the difficulties and possible sources of error which ought to be

� Knowlton (A. 2), Bull. XI.S. Geol. Surv. No. 56, 1889. (See also G�ppeit and Menge, Die Flora des Bernsteins, vol. i., and Conwentz.)

kept in mind in the identification of fossil specimens. It may he useful to draw attention to similar difficulties in the case ofnbsp;Conifers, which have not always been observed by palseohotanicalnbsp;writers.

If we examine the external characters of older branches of recent conifers from which the leaves have been detached, it will be foundnbsp;impossible to institute on such a basis any useful classification. Itnbsp;happens, not infrequently, that the leaves and cortical tissuesnbsp;become readily detached from the surface of the wood, leavingnbsp;a smooth axis in place of the corticated branch or stem. A goodnbsp;example of this is afiForded by such specimens as those representednbsp;in PL XVII. Figs. 4-6. Occasionally we have to deal with pithnbsp;casts having the surface covered with lozenge-shaped prominences,nbsp;simulating elongated leaf bases. A good example of such anbsp;medullary cast is afforded by quot;Weiss� genus Tylodendron, of whichnbsp;the true nature was pointed out by Potoni� in 1887. Again, innbsp;some specimens of the Triassic VoUzia'^ we have smaller pith castsnbsp;of similar form. In his Introduetion to the Study of Paleontologicalnbsp;Botany, Balfour¹ ² calls attention to the unnecessary multiplication ofnbsp;fossil species, and illustrates the need for careful observation of thenbsp;characters of recent stems, by reference to the striking differencesnbsp;presented by a branch of Araucaria imhricata, Pav., when the barknbsp;is viewed intact, and after it has been more or less completelynbsp;stripped off the surface of the wood. In Araucaria Cunninghami,nbsp;Ait., we find equally striking contrasts between the youngernbsp;branches, with their stiff falcate leaves, the slightly older stems, onnbsp;which only the rhomboidal leaf bases are left, the smooth surface ofnbsp;the wood, from which overlying tissues are readily detached, andnbsp;finally, the surface features presented by a pith cast.

In the long needles of Pinus and the broad flabellately veined leaves of Ginkgo, we have sufficiently well-marked characters tonbsp;enable us in most cases to arrive at a generic determination. Innbsp;many instances, however, it is a hopeless task to attempt to foundnbsp;any accurate determination on leaves alone. Among recent generanbsp;we have a deciduous habit in such plants as Larix, Ginkgo, Taxodium

distichum, Eich.; also, to a certain extent, in Sequoia sempervirens, Endl., Thuja occidentalis, L., Lihoeedrus decurrens, Torr., etc.¹; butnbsp;in most species the leaves remain on the tree for more than onenbsp;year. Occasionally, the manner of occurrence of detached leaves ornbsp;leafy shoots in a fossil state may afford evidence of the existence ofnbsp;deciduous species. A. careful examination of branches of recentnbsp;conifers bearing vegetative leaves, enables us to realize the impossibility of relying for accurate determination or comparison onnbsp;such uncertain characters as leaf form or arrangement. The uni-veined leaves of Podocarpus in some forms of the genus, may benbsp;confused with the foliage of araucarian species, in which the veinsnbsp;are imperfectly preserved; in such a plant as Podoearpm andina,nbsp;Popp., the long narrow leaves agree closely with those of somenbsp;forms of Cephalotaxus, and the detached leaves of either bear a strongnbsp;resemblance to single pinnse of Cycas. A specimen of an unnamednbsp;species of Cephalotaxus in the British Museum possesses leavesnbsp;measuring 11 cm. in length and 4 mm. broad, a close approach tonbsp;the pinnae of Cycas species. The genus Agathis, e.g. A. Australis,nbsp;Sulisb., cannot be readily distinguished from some forms of Araucaria, if we have only the leaves to guide us. The large leaves ofnbsp;Agathis Pammara, Eich., and the broad pinnte of Podozamitesnbsp;cannot always be separated with certainty, at least in suchnbsp;specimens as do not show distinct venation. If we have not thenbsp;general habit of the tree, or characteristic difierences of colour tonbsp;help us, it is practically impossible to discriminate with accuracynbsp;between the leafy twigs of many recent genera. Without enteringnbsp;into any detailed comparison of living forms from this point ofnbsp;view, we may note the close agreement between Taxodium distichum,nbsp;Eich., Sequoia sempervirens, Endl., and Taxus haccata, L.; betweennbsp;Cryptomeria, sp., and Araucaria, sp.; between different genera ofnbsp;the Cupressinm, etc. Similar examples might be readily multiplied;nbsp;but an examination of the recent species will at once demonstratenbsp;the futility of attempting generic distinction on such data, and willnbsp;emphasize the unfortunate habit of some writers of applying tonbsp;fossil fragments the unaltered names of recent genera. Anothernbsp;pitfall as regards leaf form, is the fairly common occurrence ofnbsp;heterophylly among coniferous plants. Writing in 1803, Lambert

says�I must here observe a remarkable peculiarity belonging to tbe Coniferm of tbe Southern Hemisphere, which is, that while thenbsp;trees are young their leaves are long and divaricating, hut whennbsp;they become old enough to hear fruit, those leaves fall off, and arenbsp;succeeded by short scales, closely imbricated on the branches, sonbsp;that, seeing them in their different states, one could hardly supposenbsp;it possible that they could belong to the same species.� Thisnbsp;young form of leaf may be retained for some few years before thenbsp;adult foliage is developed, and thus present a possible source ofnbsp;error in the determination of fossil branches. In such a species asnbsp;Pinm pinea, L., we have the young leaves retained for some con-sidei�able time previous to the development of the needles and shortnbsp;shoots. An interesting case of this difference between the leavesnbsp;of young and adult plants was pointed out to me at the Royalnbsp;Gardens, Kew: the young plants of the new species of Widdring-tonia, W. Whytei, Rend.,¹ ² from Hyassa-Land, bear comparativelynbsp;long needles, in marked contrast to the small scale leaves of thenbsp;older tree. A striking instance of a similar kind is afforded bynbsp;some specimens of Araucaria excelsa, R. Br., in the Herbarium ofnbsp;the British Museum : there is a seedling with its long and spreadingnbsp;leaves, an older specimen with narrow and spreading leaves, andnbsp;another with the stiff leaves of the adult plant. In Daerydiumnbsp;Kirhii, E. Muell.,² from Hew Zealand, we find a marked differencenbsp;between the small and closely adpressed leaves, and the muchnbsp;larger and more spreading leaves of other branches; also, innbsp;D. elatum, Wall., B. Westlandioum, Kirk.,^ and other forms, therenbsp;is a striking disparity in the leaf form. In Podocarpus cupressina,^nbsp;R. Br., there is a decided difference between the young and oldnbsp;forms of leaves. In Athrotaxis selaginoides, Don.,� we have variousnbsp;forms of leaf, from the longer and more openly disposed to thenbsp;smaller and closer leaves. Among other species exhibiting similarnbsp;differences in the shape and size of the leaves, we may note thenbsp;well-known Juniperw CMnensia, L., Biota orientalis, End!., Juni-perus Bermudiana, L., Glyptostrobus heterophyllus, Endl., etc.

Examples of heteropliylly have been recorded among fossil forms� e.g., Volt%ia Jieterophylla, Brong.,� from the Bunter beds; and thenbsp;specimens of Sequoia Tuurnalii (Brong,), figured by Gardner¹ ² fromnbsp;the Bagshot beds of Bournemouth, show a considerable differencenbsp;in the form of the leaves. The microscopical examination of thenbsp;epidermal cell-outlines of fossil coniferous leaves has been successfully adopted in some instances, e.g. by Zeiller² in the case ofnbsp;Frenelopsis Hoheneggeri (Ett.), and by Schenk � in several instances.

In addition to the similarity of leaf form in different species and genera, and the heterophylly in the same species, it is importantnbsp;to note the common occurrence of more than one method of leafnbsp;arrangement in the same tree. Masters,� in his useful paper innbsp;the Journal of the Linnean Society for 1891, has drawn attention tonbsp;this variation in leaf arrangement among recent species of conifers.

In describing cyoadean flowers, it was pointed out how difficult it is in some cases to distinguish between the cones of cycads andnbsp;those of certain genera of conifers, when we have only externalnbsp;form to guide us. The seeds of Cephalotaxus, Ginkgo, Torreya, andnbsp;other genera may be easily mistaken for those of Cyeai and othernbsp;cycads. There is in many instances, the same difficulty in identifyingnbsp;the detached cones of recent conifers as in determining detachednbsp;leafy twigs. Schimper and Mougeot, recognizing the difficulty ofnbsp;discriminating between fossil cones, suggested the general genericnbsp;term Strohiliies,^ which they used in a somewhat similar sense tonbsp;that in which I have used the more comprehensive genus Comtes.

Hitherto the number of Coniferce recorded from English Wealden strata has been extremely small. In addition to isolated conesnbsp;described by Carruthers, Gardner, and others, we have only onenbsp;species represented by a leaf-bearing branch�Sphenolepiiium Kur-rianum (Dunk.). The Eufibrd Collection has enabled us to recognizenbsp;as British plants several of the species previously described fromnbsp;Germany and elsewhere, and to make several additions to the listnbsp;of Wealden Coniferce.

Sphenolepidium Kurrianum (Dunk.). Sphenolepidium Sternhergianum (Dunk.).nbsp;Sphenolepidium cf. S, subulatum (Heer).nbsp;Thuites, sp.

In the above list of Coniferm described in this volume, I have suggested the probable position of various species in the classification adopted in Engler and PrantFs recent work. It must,nbsp;however, be admitted that at present we cannot feel great confidence in the attempts to determine, even the approximate afiinitiesnbsp;of such provisional genera as Sphenolepidium, Fagiophyllum^ andnbsp;Brachyphyllum.

In the present instance this genus is used as a convenient designation for certain female cones which resemble, in theirnbsp;form and structure, those of the recent genus Araucaria.

Araucarites (Conites), sp. Cf. Conites elegans (Carr.) and Kaidocarpon rainor, Carr.

It has already been pointed out that the specimen figured by Carruthers as probably a male cone of BucTdandia has little or nonbsp;claim to be regarded as oycadean; it very closely resembles thenbsp;specimens figured in PL XII. Figs. 1 and 2, and like them should

probably be referred to the genus Araucarites. Another fossil �which is probably identical with these �W'ealden cones is thatnbsp;described by Carruthers as Kaidocarpon minor from the Pottonnbsp;beds of Bedfordshire; the type specimen ^ of this species in thenbsp;quot;Woodwardian Museum shows the general characters of a somewhatnbsp;waterworn female arauoarian cone. The same form of cone, butnbsp;one belonging to another species, is illustrated by a beautifulnbsp;specimen in the York Museum, which has been described andnbsp;figured by Carruthers from the Coralline Oolite of Malton, in Yorkshire, under the name of Araucarites Hudlestoni? Mr. Carruthersnbsp;tells me he is disposed to regard some of the fossils describednbsp;by him as monocotyledonous inflorescences, as more probablynbsp;arauoarian cones. It is proposed to discuss elsewhere, at greaternbsp;length, the value of several of the published records of supposednbsp;monocotyledonous plants from Jurassic and quot;Wealden strata.

In this specimen we have a view of the proximal ends of the scales, their broad and flattened form is clearly seen, also thenbsp;lozenge-shaped cavity in which the seeds were originally situated.nbsp;The form of the scales and the more or less globose form of thenbsp;cone, present a strong resemblance to the female strobili of speciesnbsp;of Araucaria; e.g., cf. the figure given by Martius in pi. cx. ofnbsp;his Flora Brasiliensis, with the specimens in our PI. XII. Figs.nbsp;1 and 2.

The stout central axis is clearly shown, with the spirally disposed points of insertion of the broad scales. Breadth of axisnbsp;about 4 cm.; the scales probably wider towards the distal end,nbsp;showing prominent lateral angles or wings with a slightly convexnbsp;and wrinkled upper and lower surface. In one part of thenbsp;specimen the impressions of the scale apices show a prominentnbsp;distal end with a central dot, as in V. 2148. Some of the scalesnbsp;show a clearly defined cavity, originally occupied by the smallnbsp;seeds which were narrowed towards the cone axis. There arenbsp;three small seeds which seem to have fallen out from this cone.nbsp;Ecclesbourne.nbsp;nbsp;nbsp;nbsp;Rufford Coll.

V. 2331. Two specimens, in the form of oblique sections, of a cone, probably belonging to the same species as V. 2180. Gf.nbsp;Carruthers,' pi. vi. figs. 1 and 9. Eoclesbourne. Rujford Coll.

V. 21805. Two specimens, showing an impression of the cone surface. In size, and as regards the form of the scales, verynbsp;similar to Carruthers� supposed bucklandian male cone, pi. liv.nbsp;fig. 6.^ Cf. V. 2148. Possibly this specimen should be referrednbsp;to another species; it is smaller than V. 2180 (PI. XII. Pig. 2).nbsp;Eoclesbourne.nbsp;nbsp;nbsp;nbsp;Ruffordnbsp;nbsp;nbsp;nbsp;Coll.

V. 2148. The impressions of the distal ends of the scales show a small central depression, corresponding to an umbo on thenbsp;scale apex. This and other examples should be compared withnbsp;Araucarites Pippingfordenaisnbsp;nbsp;nbsp;nbsp;(Ung.),nbsp;nbsp;nbsp;nbsp;the original of whichnbsp;nbsp;nbsp;nbsp;is in

the Museum Collection. nbsp;nbsp;nbsp;Eoclesbourne.nbsp;nbsp;nbsp;nbsp;Ruffordnbsp;nbsp;nbsp;nbsp;Coll.

V. 2265a. Scales clearly preserved, showing in side view a longitudinal depression, suggesting a shrinking of the seednbsp;cavity. V. 2265. A well-preserved, but smaller specimen.nbsp;Eoclesbourne.nbsp;nbsp;nbsp;nbsp;Ruffordnbsp;nbsp;nbsp;nbsp;Coll.

V. 2277. Cf. V. 2148. V. 3185. The thick central axis clearly shown, also impressions of scales in side view. Eoclesbourne.

V. 3173. Cone in cross section; scales with seeds clearly preserved. Other specimens referred to this form of Araucarites :

Araucarites, sp.

V. 2266 and V. 2280. Two specimens of small subspherical cones imperfectly preserved; of the same form as the precedingnbsp;examples, but considerably smaller. Eoclesbourne. Rufford Coll.

In dealing �with, detached and imperfect cones, in which the scales have a flattened form like those of Abies and certainnbsp;species of Finns, it is difficult, or indeed impossible, to arrive atnbsp;a very accurate generic determination. The use of Endlicher�snbsp;genus in a wide sense is, therefore, a matter of convenience, andnbsp;in most cases preferable to the application of the generic name ofnbsp;Finns to detached cones which cannot be referred with absolutenbsp;confidence to a narrowly defined recent genus.

�Eolia, amenta staminigera et strobili, diversis Pinuum speciebus similes.� The term is a convenient one to adopt, if we do notnbsp;confine its use within the limits of the genus Finns as defined innbsp;modern systematic works.

Several detached cones have been described by Carruthers, Gardner, and others from the Wealden rooks of England, undernbsp;the generic name Pinites. Their general character justifies thenbsp;choice of this genus, but an examination of several of the typenbsp;specimens lends no support to the existence of so many distinctnbsp;species as have been described. In the second report of thenbsp;Committee appointed for the purpose of reporting on the fossilnbsp;plants of the Tertiary and Secondary beds of the United Kingdom,nbsp;Btarkie Gardner figures and describes the following ne-w species:nbsp;Finites valdensis, P. Garruthersi, P. cylindroides, and P. Pottoni-ensis, from the Wealden rocks of the Isle of quot;Wight and thenbsp;Eower Greensand of Potton. The Potton specimens are probably of Wealden age. In the case of Pmites cylindroides, fromnbsp;Potton, Gardner describes the solitary specimen as being � innbsp;excellent condition, certainly not derived from any older bed.�nbsp;An inspection of the type specimen in the Woodwardiannbsp;Museum, Cambridge, leads me to unhesitatingly describe it asnbsp;distinctly worn and rolled, and imperfectly preserved. The figurenbsp;does not convey a very accurate idea of the actual fossil; the scalesnbsp;are very imperfect, and their half-moon form spoken of by thenbsp;author of the species, is almost certainly due to wearing, andnbsp;cannot, I believe, be accepted as an original character.

Pinites cylindroides may possibly be identical with P. Bunheri, Carr., and even with P. Carrutkersi, P. Pottomensis, sm.A P. mldensin-The type specimen on which the species P. Pottoniensis is foundednbsp;is too small to admit of any specific diagnosis, and does not appearnbsp;to me to have any claim to be regarded as specifically distinct fromnbsp;P. cylindroides. It must be admitted, however, that there isnbsp;considerable risk in attempting to discriminate between thesenbsp;imperfect and detached cones; but it is surely a mistake tonbsp;multiply the number of species without stronger evidence for thenbsp;existence of any real specific differences.

A careful revision of the cones of Mesozoic Ahietinem is very desirable; the number of species would no doubt be considerablynbsp;reduced.

In the Eufford Collection there are some specimens of fairly well-preserved cones attached to their branches, and in additionnbsp;to these, several isolated specimens in other Wealden collections,nbsp;which differ in their greater length from those preserved in thenbsp;position of growth. The larger detached cones I have referrednbsp;to Carruthers� species Pinites Bunheri; the others, with theirnbsp;branches and leaves, are placed in a new species, P. Solmsi.

Pinites Dunkeri, Carruthers.

1866. nbsp;nbsp;nbsp;Pinites Bunheri, Carruthers, Geol. Mag. vol. iii. p. 542, pi. xxi. figs. 1

1867. nbsp;nbsp;nbsp;Pinites Bunheri, Carruthers, Joum. Bot. vol. v. p. 14, pi. lix. figs. 1

The following definition is given by Carruthers for this species :� �Cone elongated cylindrical; scales broad, with a rounded andnbsp;thin apex; axis slender; seeds oval compressed.�

The largest cone referred to this species has a length of over 33 cm. and is 3 cm. hroad. Cones of a similar form have been

described by Velenovsky' under the name of Pinus longissima, having a length of 31cm. and 3 cm. broad. It is difficult tonbsp;decide in many instances between P. Bunkeri and P. Carruthersinbsp;as the most suitable species to which to refer the specimens.

46654. Portion of a long cone, with the scales partially expanded, showing some clearly preserved cavities from which the small oval seeds have fallen. Very similar to Pinites Carruthersi,nbsp;Gard. Several specimens with this registered number containingnbsp;iron pyrites, and very friable. Brook, I. Wight. Bowerhanh Coll.

Pinites Carruthersi, Gardner.

Gardner speaks of the type specimen as a long cylindrical cone with numerous persistent, leathery, imbricated scales, taperingnbsp;towards the base, with scales thicker than those of Pinites valiensis.nbsp;Gard., but thin at the edges, smooth, without a keel, and withnbsp;entire rounded margins.

This form of cone is very similar to that represented by the more perfect specimens which I have included under a new species,nbsp;Pinites Sohnsi. In P. Andrmi, Coem.,^ P. Coemansi, Heer,� andnbsp;other Mesozoic forms, the same type occurs; as a rule, however.nbsp;It is impossible to determine the precise affinities of these detachednbsp;examples.

V. 2611. PI. XX. Fig. 5. V. 2852.

Cones in a crumbling condition, partly preserved in iron pyrites. Larger than those of V. 2146 (PI. XIX. Fig. 1), butnbsp;it is possible that they both belong to the same species. Eccles-hourne.nbsp;nbsp;nbsp;nbsp;'nbsp;nbsp;nbsp;nbsp;Rufford Coll.

V. 2266. Two smaller specimens, probably belonging to this species. Ecclesbourne.nbsp;nbsp;nbsp;nbsp;Rufford Coll.

Pinites Solmsi, sp. dot.

Some of the specimens referred to this species, closely resemble Pinites Ca/rruthersi, Gard., hut in view of the much more perfectnbsp;nature of the Eufford material, and the doubtful identity ofnbsp;Gardner�s type, I have ventured to found a new species. Thenbsp;specific name Solmsi has been adopted as a slight record ofnbsp;Graf zu Solms-Laubach�s services to Mesozoic palacohotany. Thenbsp;species may be thus defined :�

Short lateral branches covered with well-marked elongated bases of the scale leaves, in the axil of which are borne the short shootsnbsp;with long needle-like leaves. Cones oblong in form, with broadnbsp;scales similar to those of the Strobus section of the recent genusnbsp;Pinus, or those of Picea and Abies.

In Pinites Carruthersi the scales have a similar form, hnt slightly larger, and with a more flattened thin upper border. Innbsp;a few specimens of this species the needles are in place, hut donbsp;not show the manner of attachment of the leaves with sufficientnbsp;clearness, to enable us to determine how many needles are bornenbsp;on each short shoot. It is possible, indeed, that the leaves arisenbsp;direct from the large branches, as in Abies and Picea, but thenbsp;form and size of the needles are much more in accordance withnbsp;the characters of the genus Pinus.

Portions of four unripe cones, apparently in place; possibly the three uppermost cones are in their natural position, and the lowernbsp;one displaced. The clearly marked impressions of the bracts shownbsp;their rounded outline very distinctly: cf. P. Carruthersi, Gard.,nbsp;and P. Andrmi, Coem., as figured by Gardner.^ Similar to Abietitesnbsp;elliptieus, Pont.,^ but smaller. The surface of the cones isnbsp;marked by a number of fine longitudinally running striations.nbsp;The branches are covered with well-preserved decurrent leafnbsp;'Cushions. There are no leaves in their position of growth, hutnbsp;several fragments of needles occur on the rook sin-face. Eccles-hourne.nbsp;nbsp;nbsp;nbsp;-nbsp;nbsp;nbsp;nbsp;Rujford Coll-

Here again the leaf cushions are clearly preserved; at the ends of the short lateral branches there are borne clusters of longnbsp;needles, but it is impossible to mate out with certainty the actualnbsp;leaf arrangement, or manner of attachment to the leaf-bearing axes.nbsp;A few of the leaves show an acuminate apex. The general habitnbsp;of the specimen is similar to that of Cedrus or Larix, but thenbsp;greater length of the branches and the form of the leaves offernbsp;a still stronger resemblance to Pinus. If we compare youngnbsp;branches of some species of Pinus with this and other specimensnbsp;we find a very close agreement. The portion of a cone below thenbsp;main branch probably belongs to this species. There is somenbsp;resemblance to Leptostrobus longifolius, Pont.� Ecclesbourne.

quot;With some of the scales in this specimen there appear to be associated narrow and pointed structures, similar to the seminiferous scales, and shorter and broader bract scales of such anbsp;form as Tsuga Pouglasii, Sab. This appearance is, however,nbsp;probably deceptive, and is the result of our seeing some ofnbsp;the bracts edgewise. There can be very little doubt as to thenbsp;identity of this cone with those in V. 2146 (PI. XIX. Pig. 1).nbsp;Ecclesbourne.nbsp;nbsp;nbsp;nbsp;Pufford Coll.

In this specimen we have two female cones which appear to be in place, and a branch continued above them, on the upper portionnbsp;of which there appear to be the remains of imj)erfectly preservednbsp;structures, which may possibly be male cones. There is not,nbsp;however, sufficient evidencenbsp;nbsp;nbsp;nbsp;onnbsp;nbsp;nbsp;nbsp;whichnbsp;nbsp;nbsp;nbsp;tonbsp;nbsp;nbsp;nbsp;foundnbsp;nbsp;nbsp;nbsp;anynbsp;nbsp;nbsp;nbsp;very definite

Part of a somewhat smaller cone, and a cluster of needles borne on a short lateral branch.nbsp;nbsp;nbsp;nbsp;Probablynbsp;nbsp;nbsp;nbsp;the same speciesnbsp;nbsp;nbsp;nbsp;as the larger

Part of a cone like that of V. 2147� (PI. XIX., Fig. 4), attached to a branch bearing the characteristic leaf cushions.nbsp;The difference between this specimen and V. 2146 (PI. XIX.nbsp;Fig. 1) is due to the scales being open in the present example.

V. 21475. Open cone, and portions of branches with well-preserved leaf cnshions. Cf. PI. XIX. Fig. 1. Eocleshoume.

V. 1069. Branches with leaf cushions fairly distinct ; cones imperfectly preserved. Somewhat similar to the specimens described in Volnme I. as rhizomes of Onychiopsis Mantelli (Brong.).'nbsp;Eoclesbonrne.nbsp;nbsp;nbsp;nbsp;Jtufford Coll.

V. 3167. In this specimen the leaf cushions are clearly shown, and the limits of annual growth are suggested by the closernbsp;arrangement of the cushions in certain parts of the branch.nbsp;Ecclesbourne.nbsp;nbsp;nbsp;nbsp;Rujford Coll.

V. 3165. Very small needles, like those of V. 2255 (PI. XVIII. Fig. 3). A fairly long branch with short leaf-hearing lateralnbsp;branches.

V. 2291. Possibly a different plant, but too imperfect to determine with any certainty. Ecclesbourne. nbsp;nbsp;nbsp;Eujford Coll.

Pinites, sp.

V. 2922. A single winged seed. Xumerous fragments of Onychiopsu Mantelli (Brong.) on the same piece of rock. Ecoles-bourne.nbsp;nbsp;nbsp;nbsp;Rufford Coll.

Pinites RufFordi, sp. nov.

V. 2304. A specimen of coniferous wood with, the minute structure clearly preserved, and showing the characters of thenbsp;genus Pinites. It is proposed to publish elsewhere a detailednbsp;description of the anatomy of this specimen of Pinites, undernbsp;the name of Pinites Rujfordi. The annual rings are very clearlynbsp;marked; resin ducts fairly numerous; the tracheids in radialnbsp;section show either a single row of bordered pits, or a doublenbsp;row having the arrangement characteristic of the genus Pinites.nbsp;Ecclesbourne.nbsp;nbsp;nbsp;nbsp;Rujford Coll.

Genus SPHENOLEPIDIUM, Heer.

The generic name Sphenolepis, proposed by Schenk in 1871,' was changed by Heer to Sphenolepidium, on account of the previous usenbsp;of the former name by Agassiz as a genus of fishes. Heer�s newnbsp;term is adopted by Schenk in his account of fossil Coniferm contributed to Zittel�s Mandhueh.^ The species of this genus have beennbsp;included by Schenk and others in the family Taxodiece, but Solms-Laubach^ considers that the botanical nature of these fossils is toonbsp;imperfectly known to admit of any precise localization amongnbsp;existing subdivisions of the Group Coniferm. Previous writers havenbsp;drawn attention to the resemblance of the Wealden species, Sphem-lepidium Kurrianum, to Athrotaxis, and Sequoia has also beennbsp;referred to as the nearest living genus. There is nothing in thenbsp;nature of the fossil cones of this genus, so far as I am able to judgenbsp;from the published figures, and an examination of fairly well-preserved English specimens, to stand in the way of a comparison withnbsp;these two living genera. As regards the leaf form and arrangement,nbsp;and the general habit of the fossil species, there is a very close

resemUance between Spheno�epidium, Kurrianum (Dunk.) and species of Aihrotaxis, e.g. A. laxifolia, Hook, and A. cupressoides,nbsp;Don.' Both Athrotaxu and Sequoia are placed by Eichler* in thenbsp;section Pinoidece-Ahietinem-TaxodUnm, and it would seem highlynbsp;probable that the Wealden form bears a close relationship to thesenbsp;recent genera, especially to Don�s genus Athrotaxu.

We may define Spheno�epidium as follows : Branches alternate, with spirally disposed and decurrent leaves, cones small, oblongnbsp;and spherical, borne on short lateral branches.

Spheno�epidium Kurrianum (Dunk.).

1839. nbsp;nbsp;nbsp;? Musettes imhricatuSy E-omer, Verstein. Ool. Geb. p. 9, pi. xvii. fig. \c.

1846. nbsp;nbsp;nbsp;Thuites (Cupressites?) Kurrianus^ Dunker, quot;Wealdenbildung, p. 20,

1850. nbsp;nbsp;nbsp;Widdringtonites Kurrianus, Unger, Gen. spec, plant, foss. p. 342.nbsp;Thuites Germariy Unger, loc. cit. p. 348.

1851. nbsp;nbsp;nbsp;Widdringtonites KurrianuSy Ettingsbausen, Abh. k.-k. geol. Eeiebs.

Araucarites Bimheriy Ettingsbausen, loc. cit. pi. ii. fig. 10; pi. ii. fig. 1. 1854. Thuites Kurrianus, Morris, Brit. foss. p. 24.

1870. nbsp;nbsp;nbsp;^Araucarites hamatuSy Trautscbold, Nouv. M�m. Soc. Nat. Moscou,

1871. nbsp;nbsp;nbsp;Sphenolepis Kurriana, Sebenk, Palfeontographica, vol. xix. p. 243,

1881. Sphenolepidium KurHanum, Heer, Secc. Trab. Geol. Portugal, p. 19, pl. x�. fig. 15; pl. xiii. liga. 15 and 85; pl. xviii. figs. 1-8.

1884. nbsp;nbsp;nbsp;Sphenolepidium Kurrianum, Schenk in Zittel�s Handbuch, p. 304, fig. 210.

1885. nbsp;nbsp;nbsp;? Sphemlepis Kurriana, Hosius and Von der Marck, Palaeontographica,

figs. 1-6 ; pl. cxxviii. figs. 1 and 7 ; pl. cxxix. figs. 1, 2, 4, 6, and 8 ; ? pl. cxxx. fig. 11; pl. cxxxi. fig. 4 ; pl. clxvii. fig. 2.

F Sphenolepidium Vxrginicum, Fontaine, loc. cit. p. 259, pl. cxxv. fig. 4, and pl. clxvi. fig. 6.nbsp;nbsp;nbsp;nbsp;.

? Athrotaxopsis expansa, Fontaine, loc. cit. p. 241, pl. cxxxv. figs. 15, 18, and 22.

1894. nbsp;nbsp;nbsp;Sphenolepidium Kurrianum, Saporta, Flor. foss. Port. p. 115, pl. xxii.

1895. nbsp;nbsp;nbsp;^Sphenolepidium Kurrianum, Kerner, Jahrb. k.-k. geol. Eeichs.

� Thuites ramulis erectis irregulariter pinnatis, compressiusculis utrimque subcarinatis, foliolis crassiusculis imbricatis irregulariternbsp;dispositis elongatis subflexuosis apice aoutis dorso carinatis sub-distantibus.�

The small fragment figured by Eomer as Mmcites inilricatus, Eiim., is probably identical with Sphenolepidium Kurrianumnbsp;(Dunk.); Schenk calls attention to this resemblance, but, not havingnbsp;seen the type specimen, hesitates to express any decided opinion.nbsp;Although there is a strong probability of E�mer�s specimen beingnbsp;a leafy twig of the present species, it would hardly be wise tonbsp;enforce the rule of priority as regards the specific designationnbsp;without more tru.stworthy data. The other fragment figured bynbsp;E�mer' as Muscites falcifolius, E�m., and compared by Dunkernbsp;with Sphenolepidium Kurrianum, is too small to identify withnbsp;certainty, and does not hear such a strong resemblance to Dunker�snbsp;species as does M imbricatus.

Ettinghausen�s species Widdringtonites Saidingeri is no doubt correctly included by Schenk in the present species. The specimensnbsp;figured by Ettingshausen as Araucariies curvifolius agree so closely

with. S. Kwrimium that there cannot he much doubt as to their specific identity. It has already been pointed out/ that one of thenbsp;specimens referred by Schenk to this species is no doubt a fertilenbsp;frond of Onychiopsis Mantelli (Brong.). Some of the numerousnbsp;fragments figured by Fontaine from the Potomac beds, as examplesnbsp;of S. Kurrianum, suggest a plant with a habit somewhat differentnbsp;to that of Dunker�s species. Without attempting to discuss thenbsp;exact nature of all Fontaine�s fragments, it is probably safe tonbsp;assert that the present species is represented in the Potomacnbsp;Flora. Fontaine�s specimens are all without cones, but thenbsp;small cones figured by him as Aihrotaxopais expansa, Font., maynbsp;in all probability be referred to S. Kurrianum. The specimennbsp;figured by this author as Sequoia gracilis, Heer,^ bears a decidednbsp;resemblance to the present species. The fragments of cone-bearingnbsp;twigs figured by Fontaine as a new species, 8. Virginicum, arenbsp;compared by him with 8. Kurrianum. I have included thesenbsp;specimens in the synonomy as probably identical with the presentnbsp;species. Saporta�s Portuguese examples of the species are fornbsp;the most part small fragments of twigs; but there can be littlenbsp;or no doubt as to their specific identity with the English plant.

The numerous specimens figured by Heer from the Lower Cretaceous rooks of Greenland as Cyparassidium gracile, Heer,^nbsp;agree so closely with 8. Kurrianum, that one feels tempted tonbsp;regard the two as identical. Heer notes the resemblance . asnbsp;regards leaf form and disposition between Cyparissidium, TFid-dringtonites, Glyptostrobus, Athrotaxis, and Sequoia-, but adds thatnbsp;the form of the cones in Cyparissidium is quite distinct fromnbsp;that in the other genera, and more allied to Cunningliamia.nbsp;Although there are some slight differences between the conesnbsp;figured by Heer and those of Sphenolepidium, the points ofnbsp;difference do not appear to be very wide. Ettingshausen figuresnbsp;some specimens from Niedersohoena under the name of Frenelitesnbsp;ReicMi, Ett.;'' these bear a strong resemblance to the presentnbsp;species, and it is difficult to determine on what grounds the

� Heer (A. 3), FI. foss. Arct. vol. vi. pi. i., and FI. foss. Arct. vol. iii. (2) pi. xix. etc.

comparison with Frene�a is made. Possibly Yokoyatna�s Japanese species, Cyparissidium (?) Japonicum,^ may be closely allied tonbsp;^S. Kurrianum, but the preservation of the specimens is toonbsp;imperfect to allow of any satisfactory comparison. We maynbsp;define the species as follows:�

Branching alternate; leaves ovate, acuminate, or triangular; keeled dorsally; cones small, borne on clusters of short slendernbsp;branches, globose or oblong; scales broad and short, thick, withnbsp;an elongated lozenge-shaped depression at the apex.

This specimen appears to be practically identical with that figured by Schenk in his pi. xxxviii. figs. 10 and 11.¹ ² Thenbsp;oblong cone and the broad scales with the elliptical transverselynbsp;elongated scars are well shown. The leaves of the fertilenbsp;branches have an elongated oval form, with acute tips, and arenbsp;closely adpressed to the stem. The cone is similar to that figurednbsp;by Heer in pi. xiv.,² but in his specimen the leaves are lessnbsp;adpressed to the branch. Compare also Fontaine�s 8. Virginicum;nbsp;this species may, however, be identical with 8. Sternbergianum.nbsp;Ecclesbourne.nbsp;nbsp;nbsp;nbsp;Rufford Coll.

This specimen shows very clearly the connection between the thicker and more slender branches. Compare the thicker portionnbsp;with V. 2316J. Ecclesbourne.nbsp;nbsp;nbsp;nbsp;Rufford Coll.

The clustered fertile branches and small cones represent a characteristic feature of the species, and may be compared withnbsp;those of such a recent plant as Athrotaxis laxifolia. Cf.

V. 2316tf. Part of a thicker unbranched stem with clearly preserved leaves. This specimen, if examined without reference

to the more slender branches of 8. Kurrianum, would probably be referred to Brachyphyllum, and it serves to illustrate the greatnbsp;difSoulty in attempting to discriminate between the variousnbsp;provisional genera of fossil conifers.

V. 2316(!. Fragment of a thick axis. Cf. Brachyphyllum ohemm, Heer. Ecclesbourne.nbsp;nbsp;nbsp;nbsp;B.uffurd Coll.

V. 2303. The adpressed leaves very clearly defined. The smaller branches appear to be identical with the fertile branch ofnbsp;V. 2313 (PI. XVII. Fig. 8). The long unbranched axis of thisnbsp;and other specimens suggests an open habit of branching.

V. 2303�. Probably the same species, but the leaves are rather less closely adpressed to the stem. Cf. Schenk, pi. xxxvii. fig. 5.�

V. 23035. A much branched specimen. Towards the lower part of the thickest branch the leaves are seen to be shorternbsp;and more crowded. Cf. also V. 2303, and WiMringtonitesnbsp;Uaidingeri, Ett., which Schenk has referred to 8. Kurrianum.

V. 2303c. Long branches with the leaves less clearly preserved. Ecclesbourne.nbsp;nbsp;nbsp;nbsp;Rufford Coll.

V. 2253. Portions of branches. The leaves showing longitudinal striations, as in V. 2750 (PI. XVII. Fig. 6). Ecclesbourne. nbsp;nbsp;nbsp;Rujford Coll.

V. 2285. Probably 8. Kurrianum. Ecclesbourne. Rujford Coll.

V. 2286. Cones with open scales, probably belonging to this species. V. 718. Fragments of branches. Ecclesbourne.

V. 2303�. Slender branches with leaves rather less adpressed to the stem than in some examples of the species. This formnbsp;suggests a passage to the more open leaves of 8. 8ternhergianum.nbsp;Cf. V. 2139� (PI. XVI. Fig. 5). Ecclesbourne. Rufford Coll.

? 8phenolepidium Kurrianum. V. 3343. Two large specimens,nbsp;from an axis 3 cm. in diameter.

Splienolepidium Sternbergianum (Dunk.).

1850. nbsp;nbsp;nbsp;Museites Sternbergianm, Unger, Gen. spec, plant, loss. p. 42.

1851. nbsp;nbsp;nbsp;Araucarites Dunkeri, Ettingshausen, Abh. k.-k. geol. Eeiebs. toI. i.

Arauearites curvifolius, Ettingsbausen, loc. cit. p. 28, pi. ii. figs. 11, 13, 14, 17-21.

1870. nbsp;nbsp;nbsp;Widdringtonites Dunkeri, Schimper, Trait, pal. y�g. vol. ii. p. 329.nbsp;Widdringtonites curvifolius, Sobimper, ibid. p. 329.

1871. nbsp;nbsp;nbsp;Sphenolepis Sternbergiana, Sebenk, Palaeontograpbica, vol. xix. p. 243,

1881. Sphenolepidium Sternbergianum, Heer, Secc. Trab. Geol. Portugal, p. 19, pi. xiii. fig. la; pi. xiv. figs. 2-8.

1884. nbsp;nbsp;nbsp;Sphenolepidium Sternbergianum, Sebenk in Zittel�s Handbucb, p. 304,

1885. nbsp;nbsp;nbsp;Sphenolepis Sternbergiana, Hosius and Von der Marck, Palaeontograpbica,

1889. Sphenolepidium Sternbergianum, Fontaine, Potomac Flora, p. 261, F pi. exxi. figs. 8, 10, and 11; pi. exxx. fig. 9.

1894. Sphenolepidium Sternbergianum, Saporta, Flor. loss. Port. p. 114, pi. xxii. figs. 1 and 2; p. 139, pi. xxvii. fig. 14; p. 193, pi. xxxiii.nbsp;fig. 13.

Dunker gives the following definition of the species:� �Museites oaule virgato subflexuoso, foliis bifariis imbricatisnbsp;patentibus ovato-lanoeolatis subfalcatis.�

Brongniart substituted Juniperites as a more appropriate generic name than Museites; and Ettingshausen renamed Dunker�s plantnbsp;Araucarites Tunlceri. The specimen figured in Ettingshausen�s

' These fragments, from the Heoeomian of T�nsberg, are very small and imperfect; fig. 206 is probably a fragment of S. Sternbergianum, but it wouldnbsp;be unwise to make any definite statements on such slender evidence.

pi. ii. fig. X. seems to be identical with what he called Widdring-tonites Maidingeri, and which we have included, following Schenk�s example, as a synonym of 8. Sternlergianum. Most of the specimens figured by Ettingshausen as Araucarites curvifoUus (Dunk.),nbsp;must no doubt be included in the present species, and are notnbsp;identical with Dunker�s Lycopodites curvifolius. Some of thenbsp;species, on the other hand, are the same as Dunker�s type,nbsp;and do not appear to belong to Sphenolepidium Sternbergianurn.nbsp;Eontaine�s small specimens agree with this species, but it isnbsp;impossible to discriminate with any degree of certainty betweennbsp;the numerous and very similar twigs, which he figures from thenbsp;Potomac beds under different specific names.

If we compare the figures of Sequoia ambigua, Heer,' with Sphenolepidium Sternbergianurn (Dunk.), we find a striking resemblance ; e.g., Heer, pi. xxi. fig. 3, and Schenk, pi. xvi.nbsp;figs. 3 and 4. It is exceedingly difficult to come to any satisfactory conclusion as to the specific identity of these variousnbsp;coniferous twigs; and the striking resemblance between thenbsp;branches of certain recent forms, should sufilciently demonstratenbsp;the futility of attempting to carry our comparisons too far innbsp;the case of fossil fragments. It would seem, however, that ifnbsp;Heer�s Greenland plant is not identical with the present species,nbsp;it is closely allied to it. S. Sternbergianurn differs from thenbsp;preceding species chiefly in the more open and linear leaves.nbsp;As regards the cones of the two species, it is not quite clear hownbsp;far Schenk�s description of a specific difference holds good; thenbsp;material from the English beds is hardly sufficiently well preservednbsp;to enable us to give any satisfactory detailed diagnosis.

V. 2311. PL XVI. Fig. 4.

An imperfectly preserved specimen showing two cones attached to short branches. The difference in the leaf form from that ofnbsp;S. Kurrianum is clearly shown. Ecclesbourne.nbsp;nbsp;nbsp;nbsp;Rufford Coll-

V. 2144. PI. XVI. Fig. 6.

Well-preserved branches with spreading and falcate leaves. Ecclesbourne.nbsp;nbsp;nbsp;nbsp;Rufford Coll.

Smaller speoimen, with, more crowded leaves, leaf form distinctly preserved. Compare with the upper part of V. 2139, in which the leaves are less closely arranged. Very similar to thenbsp;specimens figured by Schenk' in pi. xxxvii. figs. 3 and 4.

Mmcites Sternhergianus, Dunk. The thicker branch well preserved, showing clearly defined leaves, with the distal end free, and the broader basal portion adpressed to the axis. Severalnbsp;delicate branches shown at the upper end. Of. Schenk,¹ ² pi.nbsp;xxxviii. fig. 3. Ecolesbourne.nbsp;nbsp;nbsp;nbsp;Rufford Coll.

V. 2141. Probably the same species. Cf. V. 2144, also Ettingshausen�s figures of Araucarites ciirvifolim,^ some of whichnbsp;Schenk refers to (S'. Sternhergianum. Ecolesbourne. Rufford Coll.

V. 2289r. Imperfectly preserved, probably specifically identical with the above specimens. Compare the finer branches withnbsp;Dunker�s Lycopodites, pi. viii. fig. 8.^

V. 2289lt;7. Imperfect fragment showing the leaves in side and surface view.nbsp;nbsp;nbsp;nbsp;Ecclesbourne.nbsp;nbsp;nbsp;nbsp;Ruffordnbsp;nbsp;nbsp;nbsp;Coll.

V. 2290, V. 2315. Small curved and closely crowded leaves, probably identical with the form represented in Ettingshausen�snbsp;figures of Araucarites Bunheri. Cf. also V. 2139lt;?, etc. Possiblynbsp;this form should be regarded as a distinct species, but the evidencenbsp;is hardly sufficient to justify a separation from 8. Sternhergianum.nbsp;Ecclesbourne.nbsp;nbsp;nbsp;nbsp;Ruffordnbsp;nbsp;nbsp;nbsp;Coll.

V. 2919. The nbsp;nbsp;nbsp;leavesnbsp;nbsp;nbsp;nbsp;show very clearly the single mediannbsp;nbsp;nbsp;nbsp;vein.

Cf. Sphenolepidium (Sequoia) subulatum (Heer).

There are a few specimens in the Euiford Collection which differ from the typical S. Sternhergianum (Dunk.), in having morenbsp;closely arranged, longer and narrower leaves; they agree verynbsp;well with Heer�s Greenland species Sequoia sululata,^ and arenbsp;probably identical with this form, and with the specimens describednbsp;by Saporta from Portugal as Sequoia suhulata var. Lusitanica?'nbsp;Saporta�s variety appears to be hardly justified by the very smallnbsp;differences which he notes between the Portuguese and Arcticnbsp;forms. There would seem to be no satisfactory evidence tonbsp;warrant the use of the generic name of Sequoia.

Ettingshausen�s fig. 18, pi. ii.� \_Araucarites curvifolim (Dunk.)] bears a strong likeness to the specimen represented in our PI. XVI.nbsp;Eig. 3 (V. 2140). It is not improbable that some of the specimensnbsp;figured by Eontaine as Sequoia Reichenlaohii, Heer,'� are identicalnbsp;with the present species. Possibly the following specimens mightnbsp;reasonably be referred to as constituting a variety of Sphenolepidium Sternhergianum.

Branches fairly closely set. Leaves narrow, linear and keeled, slightly curved or almost straight. At first sight the branchesnbsp;appear to have a bilateral form as regards the leaf arrangement,nbsp;but this may be simply due to manner of preservation. Eccles-bourne.nbsp;nbsp;nbsp;nbsp;Mufford Coll.

V. 2093. Broken fragments. Leaves fairly distinct. Cf. S. Sternhergianum and Ettingshausen�s pi. ii. fig. 18. Eoclesbourne.

Sphenolepidium, sp.

V. 2281, V. 2283. Specimens of cones which cannot be referred with any certainty to a particular species of the genus.

' Heer, FI. loss. Arct. vol. iii. (2) p. 102, pla. xxvii.-xxix. 2 Saporta (1), p. 177, pi. xxxiii. figs. 7-12.

Thuites valdensis, sp. nov.

Althougli it is impossible to give a complete or entirely satisfactory definition of this species from the single specimen in the Bulford Collection, it may he convenient to adopt a new specificnbsp;term, as there seems to be little doubt that we have to donbsp;with fragments of a conifer different from any known species.nbsp;The characters may be summarized as follows:�

Branching opposite, leaves in whorls, adpressed to the stem, two in each whorl, keeled dorsally and with comparativelynbsp;blunt apices. .

Some of the specimens referred by Ettingshausen to Frendopsis Hoheneggeri (Ett.), and, in a less degree, a few of those figurednbsp;by Schenk under Ettingshausen�s specific name, show a distinctnbsp;resemblance to the present specimen.

Preservation good. The leaves and opposite branching are clearly seen. Cf. Frenelopds ramosissima, Font., Potomac Flora,nbsp;pi. xov.,' etc., also F. Hoheneggeri (Ett.).� In describing thenbsp;specimens subsequently referred by Schenk to the genus Frene-lopsis, Ettingshausen adopts the generic name Thuites in thenbsp;''''ide sense, as including various forms of Cupressinece; it wouldnbsp;probably he wiser to retain Thuites for such examples as thosenbsp;figured by Ettingshausen. In instituting the species T. Hohen-^ffgeri, Ettingshausen speaks of the branching as alternate, andnbsp;the leaves as � quadrifariam imbricatis � ; but the specimennbsp;shown in his pi. i. fig. 7 has clearly only two leaves in somenbsp;of the verticils; indeed, there is a striking similarity betweennbsp;this German specimen and the single example from the English

� Ettiugsbausen (A. 4), Abb. k.-k. geol. Eeichs. vol. i. Abth. iii. 1852, pi. i. figs. 6 and 7.

beds. Turning to Schenk�s description, we find that he notes the presence of rows of small tubercles on the branches as annbsp;additional character, and describes the leaves as decussate andnbsp;opposite, and not four in each whorl as stated by Ettingshausen.nbsp;It may be that under Frenelopsu Soheneggeri (Ett.) we havenbsp;more than one species; the younger fragments, such as thosenbsp;figured by Ettingshausen, and a few of those described by Schenk,nbsp;agree very closely with the English specimen of Thuites, butnbsp;the larger branches of Schenk may perhaps belong to anothernbsp;plant. Eoclesbourne.nbsp;nbsp;nbsp;nbsp;Rufford Coll-

Fontaine proposes this generic title for one of the most largely developed and characteristic Potomac plants. He compares tlienbsp;leaves with those of Podozamites, but is enabled by the large andnbsp;numerous specimens at his disposal to recognize distinct coniferousnbsp;features. He defines the genus as follows :�

�Trees or shrubs with leaves and branches spreading in one plane ; leaves varying much in size and shape, those towardsnbsp;the base of the twigs sometimes smaller than those higher up,nbsp;distichous mostly, or rarely subdistichous, opposite and persistent,nbsp;attached by a short, slightly twisted foot-stalk, usually to thenbsp;side of the twig, more rarely slightly within the margin onnbsp;the upper or under surface of the stem, either attenuated towardsnbsp;the base or abruptly rounded off there, at their ends acute ornbsp;sub-acute; nerves several, coalescing at base to form a foot-stalk,nbsp;forking immediately at the base or a short distance above, thennbsp;approximately parallel to near the tips of the leaves, wherenbsp;they are somewhat crowded together, but do not converge to anbsp;union, ending in or near the extremity.�

The geiius Podoco/rpm is divided by Eichler� into four sections, of which section i. is Nageia, formerly regarded as a distinctnbsp;genus.^ In this form of Podocarpws the leaves have numerousnbsp;veins, and not a single midrib as in other species of the samenbsp;genus.

Nageiopsis of. N, heterophylla, Tont.'

The few small fragments from the English heds hear a strong resemblance to the specimen figured by Eontaine as Nageiopsunbsp;heterophylla, and most probably belong to this species.

Compare Eontaine, pi. Ixxxvi. figs. 6 and 7, etc. There are several equal veins in each leaf which converge somewhat towardsnbsp;the apex; the leaves are gradually tapered distally, and towardsnbsp;the point of attachment. The actual manner of attachment tonbsp;the branch is not clearly shown. Near Hastings. Rufford Coll.

V. 2123, V. 2362.

Probably the same species, but much more imperfect. Ecoles-bourne. nbsp;nbsp;nbsp;Rufford Coll.

Coniferse incertse sedis.

Genus PAGIOPHYLLTJM, Heer.

Saporta^ includes in the tribe Araucarinece two genera, Paohy-phyllum and Araucaria; the former representing the extinct types, the latter the living species. Pachyphyllum was first institutednbsp;hy Pomel� as a section of his genus Moreama, including J/i hrev%-folia, Pom., as the typical species.

Saporta figures some examples of the genus in which portions of cones are preserved, and is thus able to give a fairly detailednbsp;�iiagnosis. He places Pachyphyllum close to Araucaria [Eutacta),nbsp;�^gathis, and Cunninghamia. The general appearance of the

brandies referred to this genus suggests an araucarian habit, and there is a decided probability that we may consider the fossilnbsp;forms as closely allied to the recent genus. As regards thenbsp;English specimens, in the absence of fossil cones we have nonbsp;very satisfactory evidence as to the relationship to modern forms.nbsp;Heer substituted Pagiophyllum for Pachyphyllum, on the groundnbsp;that the latter name had already been assigned to a genus ofnbsp;orchids. The following concise definition is given in Zittel�snbsp;Han�huch^ �. �Leaves spirally arranged, leathery, thick, triangular,nbsp;lanceolate; spreading or closely imbricate ; decurrent at the base.�nbsp;Solms-Laubachjustly considers Pagiophyllum a purely artificialnbsp;and provisional genus.

Pagiophyllum crassifolium (Schenk).

1871. Fachyphyllum crassifolium, Schenk, Palseontographica, vol. xix. p. 240, pi. xl. fig. 8.

1874. Fachyphyllum crassifolium, Schimper, Trait, pal. v�g. vol. iii. p. 570. 1884. ? Fachyphyllum crassifolium, Saporta, Pal. Fran9. vol. iii. p. 665,nbsp;pi. ccxxvi. fig. 1.

�Eolia in ramulo spiraliter disposita, trigona crassa conica falcata basi sessilia decurrentia.�

The specimens referred by Saporta to Schenk�s species were obtained from some limestone rocks of Upper Jurassic age, in thenbsp;neighbourhood of Grenoble ; it is by no means certain that theynbsp;are specifically identical with the Wealden type.

Leaves spirally arranged; sessile, with broad base, triangular, somewhat falcate, keeled on the dorsal surface; the leaf laminanbsp;marked by fine parallel lines. Branching alternate.

Well-preserved specimen; the thick, falcate, and keeled leaves are closely arranged on the branches, reminding one of Cryptomerianbsp;Japonica. This form is very similar to the smaller specimens,nbsp;compared quot;with L^jcopodites eurvifolim, Dunk., and referred tonbsp;Sphenolepidium Sternlergianum. Eocleshourne. Mufford Coll.

The left-hand fragment agrees closely with Sphenolepidium Sternbergianmn � the leaves seen edgewise appear to be narrow,nbsp;and show traces of a median keel. Cf. V. 2289.

Towards the upper end of the large branch the tips of the falcate leaves are distinctly preserved; in the lower part they arenbsp;seen in side view, and present a broader triangular appearance.

V. 2142 and V. 21425. Here the leaves appear to be narrower, but this is largely due to the fact that they are seen edgewise,nbsp;and not so directly as in V. 2747. Ecclesbourne. Rufford Coll.

V. 2747. Cf. V. 21i2a (PI. XVI. Fig. 2). The flattened leaves seen in this view do not show the falcate form so distinctly innbsp;other specimens. Ecclesbourne.nbsp;nbsp;nbsp;nbsp;Rufurd Coll.

Pagiophyllum, sp.

In this specimen the leaves are fairly well preserved; they appear to be broader than those of the specimens referred tonbsp;crasnfolium. The fragment is, however, too small and imperfect to admit of more exact determination. Ecclesbourne.

These fragments may possibly be portions of branches of the preceding species, but they are too fragmentary to determine withnbsp;accuracy.

Brongniart proposed this name for conifers with alternate leaves disposed in a spiral, short, fleshy, and inserted by a broad andnbsp;rhomboidal base. Sohimper ¹ extends this definition, and speaksnbsp;of the genus as differing in its characters from all living forms;nbsp;he points out the striking resemblance between old branches ofnbsp;Brachyphyllum, with the leaves in the form of hexagonal ornbsp;pentagonal cushions, and certain strohili of cyoads and conifers.nbsp;The surface features of a branch from which the leaves havenbsp;fallen resemble those of Lepidodendron. Saporta ¹ further extendsnbsp;Brongniart�s definition; and remarks that probably no genus ofnbsp;conifers has given rise to more confusion and uncertainty thannbsp;the present genus. It has been compared with several recentnbsp;genera, hut we cannot regard Brachyphyllum, with its numerousnbsp;species from various geological horizons, as more than a purelynbsp;provisional genus, the actual botanical position of which is verynbsp;uncertain; probably more than one family of Conifercc beingnbsp;represented by the forms referred to under this generic name.nbsp;Schenk� has drawn attention to the too comprehensive nature ofnbsp;the genus as used by Saporta, Heer, and others, hut does notnbsp;suggest any more precise definition of the generic characters.

As Fontaine remarks, there is a striking resemblance between some forms of Echinodrobus, Brachyphyllum, and Palmocyparis.nbsp;Saporta, in describing the characteristics of Echinostrobus, pointsnbsp;out that it difiers from Brachyphyllum in having the leaves lessnbsp;thick, more pointed, and less completely adnate to the stem. Wenbsp;may compare both of these fossil genera with certain species ofnbsp;the recent genus Athrotaxis. The thick fleshy leaves, with theirnbsp;broad rhomboidal bases and spiral arrangement, constitute thenbsp;leading features of this artificial genus of fossil conifers.

Brachyphyllum spinosum, sp. nov.

This specific term is proposed as a convenient designation for �what is probably a ne-w species of Brachyphyllum. It is possiblenbsp;that some of the Wealden specimens previously recorded fromnbsp;other localities, and referred to this genus, may be fragments ofnbsp;the present species; but of this there is no proof. Seeing thatnbsp;no fossils have been, so far, described in -which the characters ofnbsp;the large branches in the Bufford Collection are represented, wenbsp;must institute a new term. The species may be defined asnbsp;follows:�

Leaves fleshy, with a median keel on the convex surface; the surface may be finely striated; leaf-scars rhomboidal, contiguous,nbsp;and spirally arranged. Some of the short lateral branches havenbsp;a pointed thorn-like form, and are clothed with the characteristicnbsp;fleshy leaves with pointed apices. Two or three of the spinynbsp;branches arise approximately at the same level from the parentnbsp;axis.

The short, stiff, leaf-bearing thorns or pointed branches constitute the most prominent feature of the species. One may compare thenbsp;thicker branches, with their large rhomboidal leaf bases, with smallnbsp;cycadean stems or twigs of Lepidodendron. In the absence of anynbsp;reproductive organs, it is impossible to assign the species to anynbsp;definite position among the Coniferce. It is no easy task tonbsp;determine the connection, if any, between such large specimensnbsp;as that represented in PI. XVII. Fig. 1, and the numerousnbsp;smaller portions of branches. The specimen shown in PI. XVII.nbsp;Fig. 6 (V. 2135)' is regarded as specifically identical with PI.nbsp;XVII. Fig. 1 (V. 2746), the striking difference being probablynbsp;due to the destruction of the cortical tissues in the former case.nbsp;The small piece of branch represented in Fig. 5 (V. 2750), showsnbsp;very clearly the woody axis detached in the lower part of thenbsp;specimen from the cortical tissues and leaves.

This form of Brachyphyllum does not appear to have been hitherto recorded from rooks of Upper Jurassic or Lower Cretaceousnbsp;age, unless we may refer to this species some of the small twigsnbsp;described by Saporta, Heer, Fontaine, and others from Wealdennbsp;strata. The spinous branches and general habit of B. spinosumnbsp;distinguish it from previously figured examples with similar leavesnbsp;and leaf cushions. Yelenovsky^ figures some specimens fromnbsp;the Perucer beds of Bohemia under the name of Eehino�rdwnbsp;squamosus, Vel.: these have similar leaves to those of B. spinosum-,nbsp;but coniferous branches with this form of leaf are too abundant innbsp;Jurassic and Cretaceous rooks to admit of any strict comparison.nbsp;Brachyphyllum crassicaule, Font.,^ B. obesum, Heer,� and othersnbsp;may be referred to as similar in leaf form to the present species.nbsp;Among the numerous conifers figured by Saporta from Jurassicnbsp;strata, we have such species as B. nepos, Sap.,�� and B. Besmyersii^nbsp;(Brong.), and other forms which resemble the Wealden species innbsp;a greater or less degree. We may compare also Bagiophyllumnbsp;eirinicum,^ Sap. Possibly the latter genus might be a suitablenbsp;designation for the English specimens, but the distinction betweennbsp;Brachyphyllum and Bagiophyllum is in many cases by no meansnbsp;well marked, and neither term is more than a convenient genericnbsp;name which does not imply any precise botanical affinity.

V. 2746. PL XVII. Fig. 1.

The main axis of this large specimen has a breadth of 2 cm., and is covered with polygonal areas representing the impressionsnbsp;of large scale leaves, very similar to those in thicker branches ofnbsp;species of Athrotaxis, Thujopsis, etc., among recent genera. Thesenbsp;are the remains of short and broad leaves preserved in the formnbsp;of carbonaceous impressions on the matrix, immediately in contactnbsp;with the thick axis. The two lowest branches are 4 cm. apart;nbsp;their surface markings are identical with those on the main branch;nbsp;from each of the lateral branches there are given off short taperednbsp;stiff branches covered with similar leaf impressions. Each of

these small branches ends in a distinct spinous apex, and in the axils of some of them are seen the indistinct impressions of smallernbsp;leafy twigs. A still larger specimen, very similar to the above,nbsp;is in the possession of Mr. Eufford; it shows two branches, thenbsp;longest of which has a length of 42 cm., and gives off numerousnbsp;stiff, pointed branches like those shown in V. 2746. Thesenbsp;spinous branches appear to be given off approximately at thenbsp;same level, and probably there were three in each pseudo-whorl.nbsp;Some of the large leaves in these specimens show fine longitudinalnbsp;striations. Ecclesbourne.nbsp;nbsp;nbsp;nbsp;Rufford Coll.

This long and narrow branched axis I have referred to Brachy-phyllum spinosum, on the grounds that it represents a decorticated specimen of the same plant which is more perfectly preserved innbsp;V. 2746 (PL XVII. Fig. 1). The breadth is fairly uniform, aboutnbsp;5-6 mm.; the numerous short spinous branches are clearly marked,nbsp;and appear to have been given off in groups of twos or threes ; onenbsp;sees in some cases two branches lying lengthwise in the sandstone,nbsp;and the base of a third in the form of a round soar in the substance of the main branch. There are no signs of any leaf basesnbsp;or leaves in this example, the exposed surface of which probablynbsp;represents the face of the woody axis. In specimen V. 2750nbsp;(PI. XVII. Fig. 5) we have a good example of the markednbsp;difference in breadth and surface characters between the leafynbsp;branch and the decorticated woody axis. The thorn-like processesnbsp;in this specimen are regarded as the decorticated spinous branchesnbsp;of V. 2746 (PI. XVII. Fig. 1). Ecclesbourne. Rufford Coll.

At the upper end of the specimen we have what appears to be a pith cast, surrounded by a woody cylinder, in the form of anbsp;dusty substance, representing the remains of wood tissue. Thenbsp;figured portion shows two spinous branches and the base of anbsp;third; probably there may have been four such branches in eachnbsp;pseudo-whorl. It may be noted that the spinous appendages innbsp;this specimen and in V. 2135 (PI. XVII. Fig. 6), are more nearlynbsp;a^t right angles to the larger axis than in V. 2746 (PI. XVII.nbsp;I�ig- 1). Ecclesbourne.nbsp;nbsp;nbsp;nbsp;Rufford Coll.

This thicker branch resembles a small cycadean axis with well-marked bases of petioles. Probably it is a portion of B. spimsum. At the edges of the cast there are here and there the impressionsnbsp;of the narrow distal ends of scale leaves; also at one place therenbsp;is shown the point of attachment of a lateral branch. Lengthnbsp;15 cm.; diameter 1'5 cm. Ecolesbourne.nbsp;nbsp;nbsp;nbsp;Rufford Coll.

A small piece of a branch, showing clearly preserved leaves with longitudinal striations (5�). At each end there projects thenbsp;impression of the woody axis, or possibly of a large pith-cavity,nbsp;such as one finds in the genus Araucaria. It is difficult, not tonbsp;say impossible, to distinguish between fragments of this speciesnbsp;and those of B. ohesum.

V. 2746�. Axis 13 cm. long, showing the clearly defined outlines of short and broad leaves. Immediately above and below the point of attachment of a spinous branch, there are impressionsnbsp;of much more slender leafy twigs. Ecclesbourne. Rufford Cull.

? Brachyphyllum spinosum.

The broad leaves of this specimen show very clearly the fine striations similar to those already noted, and identical with thenbsp;surface characters represented by Saporta in species of Pagio-phyllum and Brachyphyllum.

Bradiyphyllum obesum, Heer.

1881. Brachyphyllum ohesum, Heer, Seco. Trab. Geol. Portugal, p. 20, pi. xvii. figs. 1-4.

1889. Brachyphyllum crassicaule, Pontaine, Potomac Flora, p. 221, pi. 100, fig. 4; pi. cix. figs. 1-7.

1894. Brachyphyllum ohesum, Saporta, Flor. foss. Port, pp. 112, 138, pi. xxi. figs. 1-7 ; pi. xxvii. figs. 7 and 8.

Brachyphyllum ohesifomne var. elongatum, Saporta, loc. cit. p. 176, pi. xxxi.__flg. 14.

Heer instituted this species from some fragments of coniferous branches from the Lower Cretaceous rooks of Almargem, Portugal ;nbsp;and defined it as follows :�

� Br. ramis alternis, ram nils numerosis, aggregatis, crassis, brevibus, apioe obtusis, foliis rhombeis, dense imbricatis, dorsonbsp;leviter striatis.�

In his recent monograph on the Portuguese flora, Saporta proposes a new specific name, obemforme, for some examples ofnbsp;Brachyphyllum, which he considers may be distinguished fromnbsp;Heer�s species by their more slender branches, which are less thickset, more elongated, and subdivided. He speaks of the differencenbsp;between the two forms as slight, and admits that one may benbsp;merely a variety of the other. If we compare Saporta�s figures withnbsp;those of B. obesum given by Heer, it must be admitted that thenbsp;grounds for a specific distinction are extremely slight, and we maynbsp;not unreasonably regard the two sets of specimens as specificallynbsp;identical. Another specimen figured by Saporta is spoken of asnbsp;B. obesiforme var. elongatum; this, again, appears to be too closelynbsp;allied to such examples of B. obesiforme as are figured on pi. xxxi.nbsp;figs. 12 and 13 to be entitled to a separate designation. The newnbsp;specific term instituted by Pontaine lor some Potomac specimens isnbsp;perhaps a somewhat unnecessary addition to specific nomenclature;nbsp;the author of the species does not, apparently, draw attention tonbsp;the very close resemblance between the fossils he describes asnbsp;B. crassicaule and those already figured by Heer and Saporta asnbsp;B. obesum. The figures in Pontaine�s pi. cix. may include morenbsp;than one specific form, but the evidence is too meagre to admitnbsp;of exact determination; figs. 4 and 5 resemble the form namednbsp;by Saporta B. confusum,^ but the difference between this speciesnbsp;and B. obesum is very small. The agreement between Pontaine�snbsp;pl. cix. fig. 1 (^B. crassicaule) and Heer�s pi. xvii. fig. 2, is verynbsp;striking: compare also Pontaine, pl. cix. fig. 2, Heer, pl. xvii.nbsp;fig. 3, and Saporta, pl. xxxi. fig. 12. Fontaine�s species, B.par-ceramosum^ comes very near to the specimen figured by Saporta asnbsp;B. obesiforme var. elongatum.

I believe we cannot sufficiently discriminate between the several specimens figured by the above authors, and the similar fragmentsnbsp;from the English Wealden strata, to arrive at any trustworthynbsp;specific distinctions or diagnoses of well-marked specific types.nbsp;To group together all these forms may be unwise, and it hasnbsp;already been pointed out that we find certain points of differencenbsp;between some of the specimens, which suggest either specificnbsp;distinctions or varieties of the same type. I venture, therefore,nbsp;to make use of Heer�s specific name B. oiesum in a somewhatnbsp;more comprehensive sense than has been adopted by Saporta.

Among recent conifers, e.g. Cupressus Bawsoniana, Pari., etc., we find a considerable difference in the appearance of the branchesnbsp;depending on the development of numerous or few lateral branchesnbsp;in the axils of the leaves ; we have the closely set lateral branchesnbsp;as in some forms of B. crassicaule, as figured by Fontaine, andnbsp;the more elongated branches without the closely set lateral shoots,nbsp;as in B. obesiforme var. elongatum, Sap. To unite such forms undernbsp;one name, especially in the absence of cone-bearing branches, cannbsp;hardly be regarded as an unwarranted extension of the limits ofnbsp;a fossil species of which our accurate knowledge is extremelynbsp;small. The Jurassic species Brachyphyllum gracile, Brong.,* appearsnbsp;almost identical with some forms of B. oiesum; such comparisonsnbsp;might, however, be considerably increased, but without leadingnbsp;to any satisfactory conclusions.

Many of the specimens referred to B. oiesum agree very closely with B. spinosum, sp. nov., and it is, I believe, almost impossiblenbsp;to feel much confidence in our attempts to distinguish betweennbsp;small specimens of plants of this particular form. Possibly itnbsp;would have been better to make use of Fontaine�s specific termnbsp;crassicaule for some of the following examples, and to havenbsp;included others under B. oiesum; but if we examine such a seriesnbsp;as is represented by the following specimens, the difficulty ofnbsp;accurate determination becomes apparent: V. 3348 (PI. XVII.nbsp;Fig. 9), V. 2137 (PI. XX. Fig. 1), V. 2137� (PI. XX. Fig. 2),nbsp;V. 2337 (PI. XX. Fig. 4).

fig. 14; also Fontaine�s Atlirotaxopsis expansa, pi. cxiii. figs. 5 and 6, and A. grandis, Font., pi. cxvi. figs. 1-4. This specimen does notnbsp;show the leaf form very clearly, hut the manner of branching isnbsp;well illustrated; similar to the smaller example (V. 3348) figurednbsp;in PI. XVII. Fig. 9. Ecoleshourne.nbsp;nbsp;nbsp;nbsp;Rufford Coll.

Leaves distinct; the habit agrees with that of the smaller specimen (V. 3348) shown in PI. XVII. Fig. 9. Cf. Saporta, pi. xxxi. fig. 14, and Fontaine, pi. cxi. fig. 7.

Leaves very indistinct. Cf. Braehyphyllum obesum, Heer, pi. xvii. fig. 2, and Saporta, pi. xxxiv. fig. 8 ; also �. crassloaule,nbsp;Font., pi. oxii. fig. 6. Ecoleshourne.nbsp;nbsp;nbsp;nbsp;Rufford Coll.

Leaves well shown, with faint indications of longitudinal striations. Similar to V. 2316�, etc. Eoclesbourne. Rufford Coll.

V. 2316�. Leaves agree closely with those of Sphenolepidium Kurrianum (Dunk.), hut the branching is apparently different.nbsp;Cf. V. 2316e ()S. Kurrianum). Eoclesbourne. Rufford Coll.

V. 2137�. Cf. Saporta, pi. xxxi. fig. 14, and Fontaine�s B. crassicaule, pi. cxi. fig. 7. Eoclesbourne.nbsp;nbsp;nbsp;nbsp;Rufford Coll.

V. 2292�. Cf. Saporta, pi. xxxiv. fig. 8. Branches in this specimen closer together and thicker. Similar habit to V. 2292.nbsp;Ecoleshourne.nbsp;nbsp;nbsp;nbsp;Rufford Coll.

V. 3312. Broad axis with two small branches; similar habit to that of V. 2292�. Cf. Fontaine, pi. cxi. fig. 7. Eoclesbourne.

Other specimens of branches or leaves referred to this species: V. 2257, V. 2303, V. 2303�, V. 2310, V. 2883,nbsp;V. 3188, V. 3192. Eoclesbourne.nbsp;nbsp;nbsp;nbsp;Rufford Coll.

On page 113 of tbe present volume I have suggested the revival of Sternberg�s genus Conites for cones of doubtful position, andnbsp;have included under this generic name two species previouslynbsp;described by Carruthers as examples of cycadean strobili. Fromnbsp;an examination of additional specimens in the Rutford Collection,nbsp;it would seem that the species Conites elegans should benbsp;referred to the genus Araucarites, as being, in all probability,nbsp;a female araucarian cone. Owing to the different manner ofnbsp;preservation of the Eufford specimens, it would, perhaps, benbsp;somewhat rash to speak of them as specifically identical withnbsp;Carruthers� cones from the Isle of Wight, but there is a strongnbsp;likelihood of their close relationship, if not specific identity.nbsp;Possibly the institution of a new species of Araucarites wouldnbsp;be the most convenient course to adopt in dealing with thesenbsp;new Wealden specimens; but the various strobili hitherto describednbsp;as species of Cgcadeostrolus, Araucarites, and KaiAocarpon requirenbsp;careful revision, and I prefer, therefore, to refrain from addingnbsp;a new specific name until the several forms have been morenbsp;thoroughly examined.

Conites armatus, sp. nov.

The single specimen to which I have assigned the above specific name is too imperfect to admit of any complete diagnosis, butnbsp;the very distinctly marked and characteristic spinous processesnbsp;render it convenient to have some descriptive designation, evennbsp;in the absence of those more important characters on which anbsp;satisfactory specific definition could be founded.

V. 2338. Details very obscure; the long recurved spines are in all probability the apical prolongation of the cone scales,nbsp;similar to those in the cones of such recent species as Araucarianbsp;Bidioilli, Hook., A. Coohi, Brown, Pinus Coulteri, Don, P.nbsp;Sahiniana, Dough, etc. Ecclesboume.nbsp;nbsp;nbsp;nbsp;Rufford Coll.

CONIFEROUS WOOD.

The occurrence of coniferous wood of Wealden age has long been known in the case of the so-called Pine-raft of Brook Point,nbsp;in the Isle of Wight. This seems to have been first observed bynbsp;Webster in 1811,� and was afterwards described by Mantell innbsp;1846^; the latter writer compares the numerous coniferous trunksnbsp;and associated fossils, with the rafts of drifted trees carriednbsp;down by the waters of the Mississippi. In addition to thenbsp;fossil wood, with its tissues more or less perfectly preserved innbsp;carbonate of lime, there are numerous deposits of lignite at variousnbsp;horizons in the Wealden strata, in which the lignitic materialnbsp;obviously consists of the wood of coniferous trees. In the Medalsnbsp;of Creation, Mantell writes : � �In the Wealden deposits of Sussex,nbsp;Kent, and Surrey, I have not observed a single fragment ofnbsp;coniferous wood.� More recently, in Dixon�s Geology of Sussex,nbsp;we find that the occurrence of wood similar to that of the recentnbsp;genus Finns is recorded, both in the form of brittle jet and asnbsp;mineralized fossil wood.�

In the British Museum Collection, there are several good specimens of lignite in which the characters of coniferous woodnbsp;are clearly seen, and numerous examples of wood with the tissuesnbsp;for the most part imperfectly preserved.

By far the most perfectly preserved specimen of coniferous wood is that previously mentioned as Pinites Ruffurdi, sp. nov., andnbsp;which I hope to describe in detail elsewhere. In addition tonbsp;this, the following specimens may be mentioned;�

V. 701. Specimens of lignite, or perhaps more accurately described as jet; the annual rings clearly seen at one end of the large block. Hastings.nbsp;nbsp;nbsp;nbsp;Dawson Coll.

V. 706. A fairly large piece of wood with patches of resinous material or amber traversing the mass longitudinally; some ofnbsp;these suggest the presence in the wood of large groups ofnbsp;parenchymatous tissue, such as Conwentz has described asnbsp;� abnormes Holz-parenchym � in the case of Pinus succiniferanbsp;(G�pp.) ¹ from the North German amber beds. The microscopicnbsp;structure is very imperfectly preserved. Ecoleshourne.

Specimens of wood showing little or no internal structure. Ecclesbourne.nbsp;nbsp;nbsp;nbsp;Bufford Coll.

quot;V. 2247 and 'V. 2247�. In the latter specimen the structure is fairly well preserved ; annual rings are distinct, hut less so thannbsp;in 'V�. 2304 {Pinites Rujfordi).

V. 2326. A specimen of wood which has nndergone comparatively little alteration. Ecclesbourne. nbsp;nbsp;nbsp;Bufford Coll.

[Spirangium Jugleri (Ett.). In the Bufford Collection there are several exceedingly well-preserved specimens of thisnbsp;fossil which merit careful examination; hut if we accept the viewnbsp;that they are the eggs of fishes and not plant structures, this isnbsp;not the place for any descriptive account of them. Eor information as to the nature of Spirangium reference may be made tonbsp;the following sources :�Ettingshausen, IJeher Palmohromelia, einnbsp;neues Eossiles Pflanzengeschlecht, Alh. le.-k. geol Reichs., vol. i.nbsp;Ahth. iii. p. 1 ; Schenk, Palmontographioa, vol. xix. 1871, p. 247 ;nbsp;Schenk, Die fossilen Pflantenreste (Schenk�s Handluch, vol. iv.nbsp;1888), p. 186; Nathorst, �fvers. hongl. Vetensh.-Ahad. Forhand.nbsp;1879, No. 3 ; Benault and Zeiller, Compt. Rend. vol. cvii. 1888,nbsp;p. 1022; Saporta, Pal. Franlt;^. vol. iv. 1891, p. 38; Seward,nbsp;A new British Carboniferous fossil, Naturalist, 1894, p. 233 ;nbsp;A. Hollick, Bemarks on a paper by Dean in the Trans. Newnbsp;Yorh Acad. Sci. vol. xiii. 1893, p. 115, etc.]

ADDENDA TO VOL. I.

Nathontia valdensis {= Lechenbya mldensis), gen. et sp. nov. Vol. I. p. 145, PI. VII. Pig. 5, and PI. IX. Pigs. 2 and 2a.

After the publication of the first volume of the Wealden Catalogue, it was pointed out to me hy Prof. Nathorst that hisnbsp;name had been previously made use of by Heer for certainnbsp;marattiaoeous ferns from the Cretaceous strata of Pattorfik, innbsp;Greenland.' This oversight on my part was corrected in a shortnbsp;note published in the Geological Magazine for 1894,* and thenbsp;generic name Leclcenhya suggested as a substitute for Nathorstia.

Phyllopteris acutifolia { = Sagenopteris aoutifolia). Vol. I. p. 143, PI. IX. Pig. 6.

The generic name Phyllopteris was chosen for certain small isolated leaflets, on account of the apparent absence of anynbsp;anastomosis of the lateral veins. The examination of morenbsp;perfect specimens has enabled me to detect true reticulate venation,nbsp;and to confirm Mr. Eufford�s opinion that the leaflets should benbsp;included in the genus Sagempteris.

In the synonymy of this species, I included a plant named hy Xathorst Weichselia erratica, as probably identical with the Englishnbsp;Wealden form. Prof. Nathorst* has expressed some doubt as tonbsp;the correctness of this suggestion, his species being found innbsp;a deposit which is probably of Tipper Cretaceous age; he addsnbsp;that there are no Wealden strata in Sweden.

In speaking of this species I referred to a plant described by I^elenovsky under the name of Thinnfeldia variahilis as probablynbsp;identical with S. Mantelli, and added, with regard to his use of the

genus Thinnfeldia, that he made no reference to the resemblance of the leaves to the genus Sagenopteris. Velenovsky�s errornbsp;was pointed out to him by Hathorst,' with the result that henbsp;afterwards acknowledged his mistake,^ a correction which I unfortunately overlooked.

CANADA.

Sir William Dawson has pointed out to me, that I appear to have done injustice to Canadian geologists in the brief noticenbsp;of the Kootanie plant beds in Yol. I. p. xxxi.^ It was notnbsp;my intention to give a complete historical sketch of thesenbsp;Cretaceous deposits, but I am glad to take the opportunity ofnbsp;calling attention to a paper by Dawson On the Correlation ofnbsp;early Cretaceous Floras in Canada and the United States, and onnbsp;some new plants of this period. In this contribution we have anbsp;summary of the work of Eichardson, Gr. M. Dawson, and others,nbsp;and a description of some new plants from the Kootanie formation of the Eocky Mountains. After giving a list of Kootanienbsp;species, Dawson discusses the age of the flora, and points outnbsp;that while some of the plants must be regarded as Jurassic, thenbsp;majority have a Lower Cretaceous facies. On the whole, henbsp;concludes that � the Kootanie flora belongs to the lowest portionnbsp;of the Cretaceous, and may he a little older than that of thenbsp;main part of the Potomac formation.� ^

Among the plants described by Dawson we And the following Wealden forms: Fquisetitea Lyelli, Mant., Pecopteris Browniana,nbsp;Dunk., Sphenolepidium, sp. The pinna figured as Pecopterisnbsp;Browniana is very imperfectly preserved, and does not affordnbsp;very satisfactory proof of the occurrence of Dunker�s Wealdennbsp;species in the Kootanie flora. In another form, Cladophlelisnbsp;falcata. Font., we have one of those ferns which it is verynbsp;diflioult to separate from the widely distributed C. Whitbyenstsnbsp;(Brong.) and C. Albertsii (Dunk.). The fragment figured as

2 Velenovsky (A. 2), Abh. k. b�hm. Ges. Wiss. vol. ii. 1888, p. 5. � Letter, Oct. 17, 1894.

Sphenopteru latiloha? Pont., I am disposed to regard as identical with 8. Fittoni, Sew.: cf. Vol. I. p. Ill, Pig. 11, and PI. VI.nbsp;Pig. 2. The specimens figured hy Dawson are for the mostnbsp;part very small and imperfectly preserved, and the task of determination has necessarily been extremely difficult.

AMERICA.

Prof. Lester quot;Ward writes to me as follows in reference to. the geological age of the Potomac formation�;�

�As yon may know, I have heen engaged for several years on our Potomac formation, and I have established the fact thatnbsp;it is by no means so simple a group as might be supposed fromnbsp;what has been thus far said about it. I have been able tonbsp;subdivide it into no less than six somewhat distinct horizons,nbsp;each of which is fairly well marked off stratigraphioally, andnbsp;has its own peculiar flora. The uppermost of these subdivisionsnbsp;embraces the well-known Amboy clays of Hew Jersey, whichnbsp;have yielded a very rich flora, a monograph of which had beennbsp;nearly completed by Dr. Hewberry at the time of his death,nbsp;and has been edited hy Dr. Arthur Holliok, and has now gonenbsp;to press to be published by the D.S. Geological Survey. Thenbsp;difference between the lowest and highest of these beds is verynbsp;great, and the flora is correspondingly different. I suppose thatnbsp;the series as a whole includes nearly the entire Lower Cretaceous.

� It is only with the older Potomac that your Wealden of England can properly be compared. It is in that that we havenbsp;the remarkable cycad trunks, so much like those of the Islenbsp;of Purbeck, which I suppose are really Jurassic; although, asnbsp;you show, they are most intimately connected with the overlyingnbsp;Wealden.�

PORTUGAL.

Reference was made in Volume I. (p. xxiii.), to Saporta�s important work on the fossil flora of Portugal, and a few notesnbsp;quot;Were added with regard to some of the more interesting species.

Since this was written a large monograph' has been published, in which the Portuguese floras receive a full and careful treatment;nbsp;this work is of considerable scientific importance, and its valuenbsp;is enhanced by the fact of its being the last important contributionnbsp;to paleeobotany by that indefatigable worker whose loss is so deeplynbsp;deplored.

Beginning with strata of infra-Liassic age, Saporta describes a number of plants from beds referred to the following geologicalnbsp;horizons ; Niveau de Sinemurien (Couches d, Gryphaa ohliqua),nbsp;Niveau du lusitanien, Niveaux du n�o-jurassique (Couches d Limanbsp;alternicostata, pt�roc�rien et portlandien), Niveaux infra-cretaciquesnbsp;(du valanginien a 1�aptien et de Purgonien a I�albien). In summingnbsp;up the characteristics and affinities of the neo-Jurassic floras,nbsp;Saporta draws attention to the occurrence of species characteristicnbsp;of the Corallian and Kimmeridgian, and associated with these henbsp;recognizes various Wealden types such as occur in Northernnbsp;Germany, the Carpathians, and in North America. To discuss atnbsp;length the numerous points suggested by this extremely valuablenbsp;memoir, would take us far beyond the limits of the present work,nbsp;but some portions of the monograph more immediately germanenbsp;to our present subject must bo briefly dealt with. In lookingnbsp;through the excellent plates, we find a large number of very smallnbsp;specimens referred to a great variety of species, and on examiningnbsp;the evidence on which many of the determinations are based, itnbsp;would seem that the number of specific names might well benbsp;considerably reduced.

Sphenopteris dissectifolia, Sap., p. 19, pi. iii. fig. 9; pi. viii. fig. 2; and pi. X. fig. 9.

Saporta notes a resemblance to the quot;Wealden species Rujfordia Gopperti (Dnnk.); we may also draw attention to a resemblancenbsp;with OnycMopsis elongata (Geyl.).

8. marginata, Sap., p. 20, pi. viii. fig. 6. While recognizing the striking agreement of this fern with OnycMopsis Mantelhnbsp;(Brong.), Saporta prefers to consider it a distinct type. It isnbsp;impossible to be quite certain as to the precise nature of the

specimen, but it is difficult to understand on wbat grounds it can be separated from such a form as 0. Mantelli.

After referring to the wide range and distribution of this fern, Saporta describes specimens of fertile fronds, which he comparesnbsp;with those of the recent species Davallia gihlerosa, Sw., andnbsp;B. conoinna, Schrad. The agreement of the fossil and recentnbsp;specimens, leads the author to refer the Wealden fern to the genusnbsp;Davallia, or at least to a sub-genus of the Dacalliem. The figurednbsp;examples of fertile pinnas are less perfect than those which Inbsp;have described in Volume I. (p. 50, PL III. Pigs. 2-4); and thenbsp;comparison made by Tokoyama in the case of OnycMopsis elongata,nbsp;and by myself as regards 0. Mantelli, with the recent genusnbsp;Onycliium, is, I believe, a much nearer approach to the truth thannbsp;if we adopt the conclusions of Saporta.

The following are a few of the numerous instances in which mere fragments are referred to specific types, or on which new speciesnbsp;are founded:�Sphenopteris trifida. Sap., p. 26, pi. x. fig. 20;nbsp;8. pedicellata. Sap., p. 26, pi. x. fig. 21; 8. minima. Sap., p. 26,nbsp;pi. xiv. fig. 16; 8. trape%oidea. Sap., p. 27, pi. xi. fig. la;nbsp;8. acutidens. Sap., p. 27, pi. x. fig. 14; Cladophlehis minor,nbsp;Sap., p. 30, pi. iv. fig. 14; C. obtusilola, Sap., p. 30, pi. xiv.nbsp;fig. 15. To carry the niceties of determination so far, and tonbsp;institute new species on almost microscopic fragments of pinnsenbsp;or pinnules, is, I venture to think, a retrograde rather than anbsp;progressive method in palseobotany. A large number of othernbsp;cases might be cited illustrating this method of determination,nbsp;but the above may serve as examples of this dangerous practice.nbsp;The specimens named 8. mioroclada. Sap., p. 23, pi. vi., maynbsp;be compared with 8. Fontainei, Sew., and those described asnbsp;8cleropteru sinuata. Sap. (p. 45, pis. vii. and viii,), with Sphenopteris Fittoni, Sew.

Any traces of angiospermous plants in strata of Jurassic age are of special interest. Saporta, in speaking of fossil angiosperms,nbsp;writes : � En d�pit du nombre restreint des esp�ces et de 1�extr�menbsp;raret� des �chantillons, la presence de v�g�taux angiospermiquesnbsp;dans la flore portugaise n�o-jurassique est cependant certaine.� �

The fragment referred to as Rhiamp;ocaulon vetus, Sap. (p. 57, pi. x. fig. 22), is very small, and in itself hardly satisfactory as annbsp;example of a monocotyledonous species. Other fragments ofnbsp;parallel-veined leaves are referred to five species of Poacites; itnbsp;may be that we have in them portions of monocotyledonous leaves,nbsp;but the specimens are so small and fragmentary, that one cannotnbsp;feel much confidence in them as trustworthy evidence in sonbsp;important a matter.

There is apparently no evidence that this fragment belongs to thenbsp;nbsp;nbsp;nbsp;Adiantum-. cf. Ruffordia Gopperti var. latifoUa (Yo\. 1.

PI. TI. Figs. 1 and \a). Similarly there does not appear to he sufficient reason for naming the fragment represented in pi. xvi.nbsp;fig. 14 Marattia minor, Sap. Oleandridium tenerum, Sap., p. 85,nbsp;pi. XV. fig. 3 ; pi. xvi. fig. 18, as Saporta �points out, closelynbsp;resembles Tceniupteris Reyrichii (Schenk). The leaf figured asnbsp;Glossozamites 'brevier, Sap., pi. xvi. fig. 32, may he a pinna ofnbsp;Otozamites Klipsteinii (Dunk.).

Saporta has drawn my attention to the resemblance of Cyclopitys Belgadoi, Sap. (p. 91), to the specimen figured by me in Vol. I.nbsp;(p. 19, PI. I. Fig. 7) as probably an equisetaceous sheath.

FTom these infra-Cretaceous strata we have several specimens referred to Rhizocaulon and Poacites, which may be portions ofnbsp;monocotyledonous leaves, but the leaf fragment spoken of asnbsp;Yuocites fractifolius. Sap., p. 110, pi. xix. fig. 20�, cannot benbsp;accepted as a certain monocotyledon. A few specimens arenbsp;described as probably dicotyledons, but the existence of suchnbsp;plants at this horizon is considered as still problematical.

Sp'henopteris eirealensis. Sap., p. 126, pis. xxiv. and xxv. : cf. Onychiopsis elongata (Geyl.). 8. cuneifida. Sap., p. 127,nbsp;pi. xxiii. fig. 5. This closely resembles the finely divided formnbsp;of Ruffordia Gopperti (Vol. I. PI. IV.).

An interesting series of specimens is described and figured under the name of Isoetes Choffati, Sap., p. 134, pis. xxiv., xxv.,nbsp;and xxvii. ; these fossils show a striking resemblance to thisnbsp;recent genus of vascular cryptogams. Among the specimens

figured as ProtorrMpis Choffaii, Sap., p. 144, pis. xxii., xxvi., and xxvii., there are some which suggest leaves very similarnbsp;to those described by Bartholin from Bornholm under the namenbsp;of Ilausmanma Forchhammeri, Barth.

Bphenopterh tenuijiua, Sap., p. 161, pi. xxviii. fig. 4. In all probability this is identical with Bujfordia Gopperti (Dunk.).nbsp;Saporta himself noted the close agreement; but he does notnbsp;accept my determination of the broad-leaved fronds figured innbsp;Vol. I. PI. VI. as a variety of Bunker�s species.

Adiantum eximium, Sap., p. 164, pi. xxviii. fig. 18; pi. xxxi. fig. 6. Probably identical with R. Gopperti var. latifolia.

In speaking of Cycadites Saportte, sp. nov. (p. 30), I have drawn attention to the very close similarity of the Portuguesenbsp;specimens named by Saporta Cycadites tenuisectus. The numerousnbsp;dicotyledonous leaves described by Saporta are of considerablenbsp;interest, but need not be dealt with here, as we have no tracenbsp;of angiospermous fossils in our IVealden strata, which occupynbsp;a lower horizon than the beds from which Saporta�s specimensnbsp;were obtained.

In taking a review of the whole flora described in his monograph, Saporta calls attention to the remarkable series of types which arenbsp;represented in the plant-bearing beds ranging from the Coralliannbsp;to the Cenomanian; the series is practically continuous, andnbsp;without any distinct break or hiatus in the succession of generanbsp;and species. The flora is compared with that of the Potomac bedsnbsp;of America; a close comparison of the two sets of plants' hasnbsp;also been recently instituted by Prof. Lester Ward.

JAPAN.

Tokoyama followed up his account of the Jurassic plants from Kaga, Hida, and Echizen, by a memoir on the Mesozoic plantsnbsp;from Kotzuke, Kii, Awa, and Tosa.�* Nathorst^ had previouslynbsp;described several species from some of these localities, and concluded that the plant-bearing strata should be classed as transition

beds between the Jurassic and Cretaceous systems. Yokoyama supports iN'athorst�s view, but prefers to regard the Japanese bedsnbsp;as corresponding to the whole Neocomian series, and to the Potomacnbsp;of America. The following species are mentioned in Yokoyama�snbsp;paper;�

Fodozamites lanceolatus, L. and H., var. minory Heer.

F. lanceolatus var. latifolia, Nath.

P. pusilluSy Velen.

Zamiophyllum Fuchianum (Ett.).

Z. Buchianum var. angtistifoliay Font. Z. Naumanniy Nath.

Glossozamites parvifoUuSy Yok. Nilssonia Johnstrupi, Heer.

N. Schaumburg amis y Dunk.

N. pterophylloidasy Yok.

Ftilophyllum cf. Cutchensey Morr. Cyparissidium (?) Japonicumy Yok.nbsp;Torreya vanusta, Yok.

Some of the specimens described as Fodozamites show a strong resemblance to Fontaine�s Nageiopsis; but it is no easy matternbsp;to decide between these two genera when we have to deal withnbsp;small specimens.

AUSTRIA.

Fritz Von Kemer � has recently described some fossil plants from the Island of Lesina, o�E the Austrian coast, in the Adriaticnbsp;Sea; the flora is regarded as Lower Cretaceous in age. Amongnbsp;the species recorded from these beds we have Dioonites cf.nbsp;Saxonicus (Reich.), p. 49, pi. iv. fig. 6: this species I havenbsp;included as a synonym of Zamites Buohianm (Ett.); hut it isnbsp;difficult to decide from the indistinct photograph of Kerner�snbsp;fragment how far his specimen may he regarded as identical withnbsp;the Wealden species. A coniferous twig figured by Kerner asnbsp;Fagiophyllum rigidum, Sap., resembles fairly closely some of thenbsp;English specimens of P. crassifolium, Schenk; cf. Kerner�s figure,nbsp;pi. iv. fig. 3, and my PI. XVI. Fig. 2.

A very small and imperfect specimen is figured as a piece of Sphenolepidium Kurrianum (Dunk.), (pi. iv. fig. 2); this does notnbsp;seem to be a typical example of the species, but it is impossiblenbsp;to accurately determine so small a fragment.

SPITZBERGEN.

In 1883 Rathorst' published a short account of some plant-beariug beds in Spitzbergen, -which had formerly been classed as Cretaceous, but which he preferred to consider as uppermost Jurassicnbsp;in age. Having lately had an opportunity, through the kindness ofnbsp;Professor Hathorst, of examining the Spitzbergen fossils, I amnbsp;disposed to think that we shall find in his forthcoming monographnbsp;on this flora a certain amount of evidence in favour of its beingnbsp;regarded as Wealden.

CONCLUSIONS.

Prior to the acquisition of the Rufford Collection by the British Museum, the following plants had been recorded from the Wealdennbsp;strata of England. Such alteration as 1 have suggested withnbsp;regard to any of th� species are noted in brackets.

Chara {~C. Knowltoni^ sp. nov.). ^quisetites LyelU, S. and quot;W.

Burchardti, Dunk.

Onychiopsis Mantelli (Brong.). Sphenopteris Fittoni, sp. nov. (~S.

Mantelliy Brong.).

Tempshya Schimperi, Corda. Tlt;^niopteris Beyrichii (Schenk).nbsp;TFeichselia Mantelli (Brong,).nbsp;Sphenolepidium Kurrianum (Dunk.).nbsp;Araucarite^ Bippingfordensis (Ung.).nbsp;Binites Bunkert^ Mant.

B. Carruthersi^ Gard.

B. valdensis. Gard.

Binites Mantelli, Carr.

B. nbsp;nbsp;nbsp;patens, Carr.

Cycadeostrobus elegans, \ [ = Conites Carr.nbsp;nbsp;nbsp;nbsp;Inbsp;nbsp;nbsp;nbsp;elegans

C. nbsp;nbsp;nbsp;ovatus, Carr. J (Carr.)].

G. tumidus, Carr. \^'==-Conites tumidus, (Carr.)].

B'wklandia anomala (S. and 'W�.). Fittonia squamata, Carr.

Yatesia Morrisii, Carr.

Bennetiites Saxhyanus (Brown). Bioonites Brongniarti (Mant.) (= Cyca-dites Brongniarti, Mant.).

Kilssonia Schaumhurgensis (Dunk.).

The following list includes the species described in Yolume !�, with the addition of those dealt with in Yolume II.

Specimen A. (Vol. I. p. xxxv. PI. I. Fig. 7). Specimen B. (Vol. I. p. xxxv. PI. I. Figs. 8 and 9).nbsp;t Withamia Saporta, Sew.nbsp;t Becklesia anomala, Sew.

In the accompanying table an attempt is made to sho'w the geographical range of such species as are not confined to thenbsp;liVealden rooks of England. The occurrence of the same speciesnbsp;in different regions is not regarded as necessarily proving homo-taxial strata. A more detailed consideration of the geologicalnbsp;correlation of the plant-bearing sti�ata, in �which Wealden typesnbsp;occur, will be undertaken elsewhere, as there are still a fewnbsp;species to be described from the English beds. Such plants asnbsp;Onychiopsis MantelU (Brong,), Matonidium Gopperti, Schenli,nbsp;Rujfordia Oopperti (Dunk.), are by no means confined to onenbsp;geological horizon; and it is possible that we have other species,nbsp;the range of which is not strictly limited to true Wealden strata.

In the following list are added the chief districts or localities in each country from which the plants have been obtained. Thenbsp;Dames of the principal authors of the several Wealden floras (ornbsp;floras containing Wealden species), are given below, with nunibersnbsp;referring to the bibliographies at the end of Part I. and Part II- ;nbsp;those in the former being printed with A. after the author�s name,nbsp;those in the latter with B.

England. Isle of AVight, Kent, Surrey, Sussex, and Bedfordshire. [Stokes and Webb (A.), Mantell (A. 1-7), Carruthers (A. 3 and 4),nbsp;(B. 1, 3, 6, and 8), Topley (A. 1.), Gardner (A. 1 and 2),nbsp;Bristow (A.), Peyton (A.), etc.]

Portugal. Cereal, Bellas, Torres-Vedras, etc. [Heer (A. 6), Saporta (B. 1).] Prance. Beauvais, [Brongniart (A. 4).]

Germany. Deister, Cluedlinburg, Tentoburgerwald, and other localities in N.AY. Germany. [Hunker (A. 2), Schenk (A. 2 and 4), Hosiusnbsp;and Von der Marck (A. 1 and 2).]

Austria. North Carpathians. [Ettingahausen (A. 4), Schenk (A. 3).]� Lesina I. [Kerner (B.).]

America. Virginia, Marjdand, Montana. [Fontaine (A. 2 and 3), Newberry (A. 1), Knowlton (A. 2).]

Japan. Kaga, Hida, Echizen, Kii, etc. [Yokoyama (A. 2) and (B.), Nathorst (A. 3).]

The questions of geological age suggested by the above table, will be discussed in a subsequent communication on the Wealdennbsp;floras, which it is intended to lay before the Geological Societynbsp;at an early date. With a view to discover if the manner ofnbsp;occurrence of the fossil plants, or the association of differentnbsp;genera and species, would afford any evidence as to the relativenbsp;positions of growth of the various floral types, I wrote to Mr.nbsp;Ilufibrd asking him to give me such information as his accuratenbsp;knowledge of the Hastings district* enabled him to contribute withnbsp;regard to this question. I cannot do better than reproduce thenbsp;main facts which he communicated to me. He writes that he isnbsp;Unable to discover any reliable data as to the relative altitudes atnbsp;which the plants grew, but adds the following useful informationnbsp;as to the occurrence of some of the characteristic species in thenbsp;different beds :�

(a) Fine clay ironstone; ferns well preserved, no cycads or conifers. This bed occupies about the same geological position as (5),nbsp;although separated from it horizontally; it contains Marchantites,nbsp;J^quisetites Burchardti, Onychiopsis Mantelli, Ruffordia O�pperti,nbsp;Sphenopteris Fittoni, and other ferns.

(J) Fine blue clay. The � cycad bed,� containing Zamites, Anomo%amites, Otozamites, Fittonia, etc.; ferns rare. A few yardsnbsp;farther on, and about the same horizon, there have been foundnbsp;coniferous twigs, etc.

{d) Blue clays, sometimes sandy. �Fern bed�; no cycads; occasionally-leaves of Pinites, also Onychiopsis Mantelli, Ruffordia,nbsp;Sphenopteris Ftttoni, Cladophlebis, etc.

(�) For the most part reddish ironstone, with some grey sandstone merging into sandy clay. About the same horizon as {ff); plants very abundant�Onychiopsis, Ruffordia, Sphenopteris, Clado-phlelis, Tmniopteris, Protopteris; also Dioonites Brongniarti, Zamites,nbsp;Otozamites, Benneitites (Williamsonia), Pagiophyllum, and othernbsp;conifers, etc. This bed is of somewhat coarser material than thenbsp;others, and contains a greater mixture of plants; Pagiophyllumnbsp;crassifolium and other conifers are very abundant.

In reading the above notes by Mr. Eufford, we find that with the exception of bed (�), in which the various classes of plants arenbsp;well represented, the ferns, cycads, and conifers are not usuallynbsp;intermixed. The partial or complete separation of ferns, cycads,nbsp;and conifers, may be due either to the nature of the plant material,nbsp;which might be sorted by the water by reason of some differencesnbsp;in weight, or to the relative adaptability to longer or shorternbsp;transport by water; or the result of the plants growing innbsp;different districts and at different elevations. The more delicatenbsp;fern fronds would probably be carried to a greater distancenbsp;than the heavier and larger pieces of cyoadean or coniferousnbsp;plants. On the other hand, the gymnosperms may well havenbsp;been more abundant on higher ground and in drier situationsnbsp;than the Filicince. Assuming the bed (e) to occupy the lowestnbsp;horizon in the series, it would appear that the material composing the sandstones and ironstones was laid down in somewhat

shallow water, and the fossils embedded in the strata were derived from a wide area, embracing localities rich in bothnbsp;ferns and gymnosperms. The petrological nature of (d) and thenbsp;absence of cycads suggest deeper water; while the fine blue claynbsp;of (J) may have been derived from rooks in an area characterizednbsp;by the predominance of cycads. It would, however, be difficult tonbsp;reconstruct the conditions of growth of the several plants withoutnbsp;a very careful examination of the rocks and their fossil contents,nbsp;and at best our conclusions would probably not possess any greatnbsp;scientific value.

In an old work by Unger,* we read that in the quot;Wealden period there were � small wet islands, covered with forests, inhabited bynbsp;the largest and most terrible monsters of the primitive world.nbsp;The atmosphere was filled with moist vapour, and carbon dioxidenbsp;exhalations favourable to the prodigious propagation of thenbsp;amphibian race, and to the development of ferns, cycads,nbsp;conifers, and some monocotyledons.� He goes on to say thatnbsp;�La triste sauvagerie de oet int�rieur de for�t est encore redoubl�enbsp;par celle de ses habitants, parmi lesquels Ie gigantesque Iguanodon

In a more recent monograph on the 'Wealden period. Schenk � confidently speaks of the climate as undoubtedly tropical, and refers to the occurrence of tree ferns, the abundance of cycads, and other factsnbsp;in support of this conclusion. It would be extremely difficult,nbsp;or indeed impossible, to give any approximate estimate of thenbsp;temperature in Northern and Central Europe during the Wealdennbsp;period. The general characters of the vegetation would certainlynbsp;seem to point to a tropical climate, and there can be little doubtnbsp;that the temperature was considerably higher than the Wealdennbsp;districts enjoy at the present day.

In discussing the climate of a past age, in which no living species of plants existed, and in attempting to make use of fossil plantsnbsp;as indices of climatal conditions, we have to bear in mind thenbsp;great danger of drawing conclusions from a comparison of extinctnbsp;and living forms. It is superfluous to point out the lesson sonbsp;clearly taught by recent plants, that closely allied species frequently

occur under very different conditions of temperature. In the present instance, the numerous species of cyoads naturally suggestnbsp;conditions similar to those most favourable or essential to thenbsp;living representatives of the CyeaAacea\ but we must remembernbsp;that the so-called cycadean fronds from Mesozoic rooks are nearlynbsp;always found apart from the stems and reproductive structures,nbsp;and we are still to a large extent in the dark as to the exactnbsp;nature and structure of these extinct cycadean plants.

Looking at the quot;Wealden plants collectively, we notice a very striking agreement with the flora of the underlying Jurassicnbsp;strata, and it would be difiicult to point to any well-marked ornbsp;essential difference between the plant-life of the two periods.nbsp;The evidence of palseobotany certainly favours the inclusion ofnbsp;the quot;Wealden rooks in the Jurassic series.

One of the most attractive and difSoult problems which is suggested to a botanist by such a flora as that of the Wealdennbsp;period, is the evolution of angiospermous plants. A reviewer hasnbsp;happily expressed this in the following words': �In the folds ofnbsp;the Wealden we imagine the secret of the evolution of angiospermsnbsp;must be locked. It is as if we stood at the mouth of a greatnbsp;river flowing from an unexplored interior, whose flotsam wenbsp;anxiously interrogated for clues as to the nature of the unknownnbsp;Hinterland; yet nothing reaches us from beyond the coast-belt,nbsp;which we have already explored.� Among the English speciesnbsp;there are none which can be regarded as the earliest angiosperms,nbsp;and we search in vain among the abundant samples of the Wealdennbsp;vegetation for any fragments of monocotyledonous or dicotyledonousnbsp;plants. In the Potomac beds of America, which include strata ofnbsp;Jurassic and Lower Cretaceous age, we have several undoubtednbsp;angiospermous species; and again, in the closely parallel series ofnbsp;Portuguese rocks, dicotyledons and monocotyledons are fairlynbsp;abundant. The true Wealden vegetation would seem to havenbsp;been without any examples of the highest class of plants, andnbsp;may be looked upon as the last of the Mesozoic floras in whichnbsp;the gymnosperms represented the limit of plant development.nbsp;One genus, however, carries us a few steps towards the nextnbsp;stage in botanical evolution ; the inflorescence of Bennettites marks

a distinct advance in the difierentiation of reproductive structures beyond the characteristic oycadean type. Were we in a positionnbsp;to speak of the anatomical structure, or to describe more fullynbsp;the reproductive organs, of Wealden plants, it might be that wenbsp;should be able to recognize a distinct foreshadowing of angio-spermous characters; unfortunately, however, the extreme scarcitynbsp;of mineralized plant tissues precludes any such treatment of plantnbsp;types. In spite of the somewhat disappointing nature of thenbsp;flora from the point of view of angiosperm development, wenbsp;may reasonably hope that a more detailed comparison of florasnbsp;possessing a Wealden facies will enable us to add something ofnbsp;Value to the history of plant evolution, and to the facts of plantnbsp;distribution.

LIST OF WOEKS QUOTED.

[With two or three exceptions the works quoted in Part I. are not repeated in the following list. The letter (A.) after authors� names in the footnotes of thenbsp;present volume signifies that the work referred to is in the bibliography ofnbsp;Part I. Under Feistmantel, Heer, and Zittel, I have given the dates of thenbsp;several parts of vols. i. and ii. of the Memoirs of the Geological Survey ofnbsp;India (second series), of the Flora fossilis Arctica, and of Schimper andnbsp;Schenk�s volume on fossil plants in Zittel�s Randbmh.']

Balfour, J. H. Introduction to the Study of Palaeontological Botany. Edinburgh, 1872.

Bartholin, C. T. Nogle i den hornholmske Juraformation forekom-mende Planteforsteninger. [jBof. Tidssk. vol. xix. 1894, p. 87.] Bennett, J. J. See Brown,, B.

Bensted, W. H. (1) On the Kentish Rag as exhibited in the Iguanodon Quarry at Maidstone. [Proa. Qeol. Assoc, vol. i. 1860, p. 57.]nbsp;Bensted, W. H. (2) Notes on the Geology of Maidstone. {Oeologist,nbsp;vol. V. 1862, p. 294.]

Berger, H. A. C. Die Versteinerungen der Fische und Pflanzen im Sandsteine der Coburger Gegend. Coburg, 1832.

Bertrand, C. E. and Renault, B. (2) Remarques sur les faisoeaux foliaires des Cycad�es actuelles. [Arch. Bot. Nord France,nbsp;Ann. 3, 1886, p. 232.]

Bornemann, J. 0. Ueber organische Reste Lettenkohleii Gruppe ThUringens. Leipzig, 1856.

Brongniart, A. (1) Observations sur les v�g�taux fossiles renferm�s dans les gres de Hoer en Scani�e. [Ann. Bei. Nat. vol. iv.

Brongniart, A. (2) Recherches sur l�organisation des tiges des Cycad�es. [Ann. Sci. Nat. vol. xvi. 1829, p. 389.]

Brongniart, A. (3) Sur la classification et la distribution des v�g�taux fossiles en g�n�ral. \_M�m. Mus. vol. viii. 1822, p. 209.]nbsp;Brongniart, A. (4) Note sur les v�g�taux fossiles de 1�oolite anbsp;Fong�res de Mamers. [Ann. Sci. Nat. vol. iv. 1825, p. 422.]nbsp;Brovm, R. \Proc. Linn. Soc. vol. ii. 1855, p. 130.]

Brown, B. and Bennett, J. J. Plantae Javanicae rariores. London, 1838-52.nbsp;nbsp;nbsp;nbsp;I

Buciland, W. (1) On the OycaicfeoicJecB, a family of Fossil Plants found in the Oolite Quarries of the Isle of Portland. [Oeol. Trans.nbsp;(2) vol. ii. 1828, p. 395.]

Buckland, W. (2) Geology and Mineralogy considered with reference to Natural Theology. London, 1858.

Capellini, O. and Solms-Laubach. I Tronchi di Bennettitee dei musei italiani. [Mem. R. Accad. Sci. inst. Bologna (5), vol. ii.nbsp;1891, p. 161.]

Carruthers, W. (1) On Fossil Cyoadean Stems from the Secondary Eocks of Britain. [Tram. Linn. Soc. vol. xxvi. 1870, p. 675.]nbsp;Carruthers, W. (2) On some supposed Vegetable Fossils. [Quart.

Carruthers, W. (3) On Gymnospermous Fruits from the Secondary Eocks of Britain. [Journ. Bot. vol. v. 1867, p. 1.]nbsp;Carruthers, W. (4) On some Fossil Coniferous Fruits. [Geol. Mag.nbsp;vol. iii. 1866, p. 534.]

Carruthers, W. (5) On some Coniferous Fruits from the Secondary Eocks of Britain. [Oeol. Mag. vol. vi. p. 1, 1869.]nbsp;Carruthers, IF. (6) On Cycadeoidea Tatesii, sp. nov., a Fossil Cycadeannbsp;Stem from the Potton Sands, Bedfordshire. [Geol. Mag.nbsp;vol. iv. 1867, p. 199.]

Carruthers, IF. (7) British Fossil Pandanaoeae. [Geol. Mag. vol. v. 1868, p. 153.]

Carruthers, W. (8) Description of a new species of Araucarites from the Coralline Oolite of Malton. [Appendix to a paper bynbsp;Blake and Hudleston on the Corallian Eocks of England.nbsp;Quart. Journ. Oeol. Soc. vol. xxxiii. p. 402, 1877.]

Dawson, J. IF. (1) On the Cretaceous and Tertiary Floras of British Columbia and the N.W. Territory. [Traris. R. Soc. Canada,nbsp;vol. i. 1882, sect. iv. p. 15.]

Dawson, J. W. (2) On the Correlation of the early Cretaceous Floras in Canada and the United States, and on some new Plantsnbsp;of this period. [Trans. R. Soc. Canada, vol. x. 1893, sect, iv.]

He Candolle, A. Prodromus systematis naturalis regni vegetabilis. Vol. xvi. p. 522. Paris, 1868.

De la Beche, H. T. Remarks on the Geology of the South Coast of England, from Bridport Harbour, Dorset, to Babbacombenbsp;Bay, Devon. [Tram. Oeol. Soc. vol. i. (2) 1824, p. 40.]

Bon, D. Description of the New Genera of the Natural Family of Plants called Coniferae. \Trans. Linn. Soc. vol. xviii. 1841,nbsp;p. 163.]

Eichwald, E. Lethsea Rossica, vols. i.-iii. Stuttgart, 1853-68. Endlicher, S. (1) Genera plantarum secundum ordines naturales dis-posita. Vindobonae, 1836-40.

Engler, A. and Prantl, K. Die natiirliohen Pflanzenfamilien. Lief. ii. Leipzig, 1889.

Ettingshausen, C. von. (1) Beitrage zur Kenntniss der fossilen Flora Neuseelands. [Benhchr. k. Akad. Wiss. vol. liii. 1887, p. 3.]nbsp;Ettingshausen, C. von. (2) Beitrage zur Kenntniss der Tertiarfloranbsp;Au.straliens. Zweite folge. [Ihid. vol. liii. 1886, p. 1.]nbsp;Ettingshausen, G. von. (3) Die fossile Flora von Looben in Steirmark.

Feistmantel, 0. (1) (a) Jurassic (Liassic) Flora of the Rajmahal Group in the Rajmahal Hills. Part ii. 1877.

(6) Jurassic (Liassic) Flora of the Rajmahal Group from the Golapili near Ellore S. Godavari. Pt. iii. 1877.

(amp;) Flora of the Jabalpur Group (U. Gondwanas) in the Son-narbeida Region. Pt. ii. 1877.

Feistmantel, 0. (2) Palaeontologische Beitrage. II. Ueber die Gattung Williamsonia, Carr., in India. [Palmontographica, supp. iii.nbsp;Lief. iii. 1877.]

Gardner, 8. nbsp;nbsp;nbsp;(1) On Mesozoic Angiosperms. [Oeol. Mag. vol. iii. 1886,

Filices, by Gardner and Ettingshausen, Paleeontographical Society, 1879-82. Vol. ii. Gymnospermw, by Gardner,nbsp;Palaeontographical Society, 1883-6.

Qeinitz, H. B. and Gutbier, A. von. Die Versteinerungen des Rothlie-genden und Zechstein Gebirges. Dresden and Leipzig, 1848.

O�ppert, H.R. (1) Ueber die fossilen Cycadeen �berhaupt, mit R�ck-sicht auf die in Schlesien vorkommender Arten. [ �ebericht der Arheiten und VeirWtiderungen der Schlesischen Oessel. f�rnbsp;Vart. Kultur. 1843, p. 114. Breslau, 1844.]

O�ppert, H. R. (2) Beitrage zur Kenntniss fossiler Cycadeen. [Neues Jahrh. 1866, p. 129.]

O�ppert, H. R. (3) Ueber die gegenwartigen Verhaltnisse der Palaon-tologie in Schlesien, so wie �ber fossile Cycadeen. [Schles. Oes. Denkschr. 1853, p. 251.]

O�ppert, H. R. (4) Ueber lebende und fossile Cycadeen. \_Zeitsch. geol. deutsch. Oes. vol. xvi. 1864, p. 173.]

O�ppert, H. R. (6) Monographie der fossilen Coniferen. [^Nat. Verhand. Holland. Maatschaf Haarlem. Leiden, 1850.]

O�ppert, H. R. and Stenzel, G. Die Medullosete. [PalmntograpMca, vol. xxviii. 1881, p. 113.]

Grand�Enry, M. G. G�ologie et pal�ontologie du bassin houiller du Gard. St. Etienne, 1890.

Grant, C. W. Memoir to illustrate a Geological Map of Cutch. [Trans. Oeol. Soc. vol. V. (2) 1840, p. 289.]

Guillard, � Note sur un v�g�tal fossile des terrains-houillers de Bive-de-Gier. [Ann. Sci. phys. nat. Agric. indust. Lyon, vol. ii-1839, p. 123.]

Vol. I. Zurich, 1868.�(i.) Die in Nordgr�nland, auf der Melville-Insel, im Banksland, am Mackenzie, im Island und in Spitz-bergen entdeckten fossilen Pflanzen. pp. 1-192 ; map and plates, i.-L.

(ii.) Mit. Einem Anhang �ber versteinerte Holzer der arctischen Zone. C. Cramer.

Vol. II. Winterthur, 1871.�(i.) Fossile Flora der Baren-Insel. [Kongl. Svensk. Vet.-Akad. Hand. vol. ix. No. 5, 1871,nbsp;pp. 3-51, pis. i.-xv.]

(ii.) Flora fossilis Alaskana ; Fossile Flora von Alaska. [K. Svensk. Vet.-Akad. Hand. vol. viii. No. 4, 1869, pp. 3-41,nbsp;pis. i.-x.]

(iii.) Die Miocene Flora und Fauna Spitzbergens. [Ibid. vol. viii. No. 7, 1870, pp. 3-98, pis. i.-xvi.]

being a description of the plants collected by Mr. Edward Whymper during the Summer of 1867. \^Phil. Tratis. 1869,nbsp;pp. 445-88, pis. xxxix.-lvi.]

Vol. III. Zurich, 1875.�(i.) Beitrage zur Steinkohlen-Elora der arotischen Zone. [K. Svensk. Vet.-Akad. Hand. vol. xii.nbsp;No. 3, 1874, pp. 3-11, pis. i.-vi.]

(ii.) Die Kreide-Elora der arctischen Zone. [Ibid. vol. xii. No. 6, 1874, pp. 3-140, pis. i.-xxxviii.]

(iii.) Nachtrage zur Miocenen Flora Gronlands. [Ihid. vol. xiii. No. 2, 1874, pp. 3-29, pis. i.-v.]

(iv.) Uebersicht der Miocenen Flora der arctischen Zone. Zurich, 1874 ; pp. 3-24.

Vol. IV. Zurich, 1877.�(i.) Beitrage zur fossilen Flora Spitz-bergens. [K. Svensk. Vet.-Akad. Hand. vol. xiv. No. 5, 1876, pp. 3-141, pis. i.-xxxii.]

(ii.) Beitrage zur Jura-Flora Ost Sibiriens und des Amur-landes. [M�m. Acad. Imp. Set. St. Pet. (7) vol. xxii. No. 12, 1876, pp. 1-122, pis. i.-xxxi.]

(iii.) Ueber die Pflanzen-Versteinerungen von Ando in Norwegen. Pp. 3-15, pis. i.-iii.

Vol. V. Zurich, 1878.� (i.) Die Miocene Flora des Qrinnell-Landes. Pp. 3-38, pis. i.-ix.

(ii.) Beitrage zur fossilen Flora Sibiriens und des Amur-landes. [M�m. Acad. Imp. Sci. St. Pet. (7) vol. xxv. No. 6. 1878, pp. 1-58, pis. i.-xv.]

(iii.) PrimitisD florae fossilis Sachalinensis. Miocene Flora der Insel Sachalin. [Ihid. vol. xxv. No. V, pp. 1-61, pis. i.-xv.]nbsp;(iv.) Ueber fossile Pflanzen von Novaja Semlya. [Ah Svensk:nbsp;Vet.-Akad. Hand. vol. xv. No. 3, 1878, pp. 3-6, pi. i.]

(v.) Beitrage zur Miocenen Flora von Sachalin. [Ihid. vol. XV. No. 4, 1878, pp. 3-11, pis. i.-iv.]

Th. i. Zurich, 1882.

(iii.) Nachtrage zur fossilen Flora Gronlands. [K. Svensk. Vet.-Akad. Hand. vol. xviii. No. 2, 1880, pp. 3-17, pis. i.-vi.]nbsp;(iv.) Beitrage zur Miocenen Flora von Nord-Canada.nbsp;Zurich, 1880 ; pp. 3-17, pis. i.-iii.

(v.) Untersuohung Uber fossile H�lzer aus der arotischen Zone. C. Schroeter, Zurich, 1880 ; pp. 3-38, pis. i.-iii.

Vol. VII.�Flor. loss. Grbn. Th. ii. Z�rioh, 1883 ; pp. 1-275, pis. xlviii.-cix., one map, and two photographic views asnbsp;frontispiece.

Hildebrand, F. Die Verbreitung der Coniferen. [Verh. Nat. Ver. prems. Rheinl. vol. xviii. 1861, p. 199.]

Hollick, A. Additions to the Palseobotany of the Cretaceous Formations on Staten Island. [Trans. NT. Acad. Sd. vol. xii. 1892-3.]

Karsten, H. Organographische Betraohtung der Zamia muricata, Willd. [Abh. k. Akad. MHss. 1856, p. 193.]

Kerner, F. von. Kreidepflanzen von Lesina. [Jahrh. k.-k. geol. Reichs. vol. xlv. 1895, Heft i.]

Kidston, R. On the Fruotifleation of some Ferns from the Carboniferous Formation. [Trans. R. Soc. Fdinb. vol. xxxiii. 1887, pt. i.]

Furr, J. Q. Beitriige zur fossilen Flora der Juraformation Wiirttem-bergs. Stuttgart, 1845.

Lemaire, C. L�illustration horticole. Journal special des Serres et des Jardins. Vol. xv. 1868.

Lesquereux, L. (1) Contributions to the Fossil Flora of the Western Territories. Pt. ii. : The Tertiary Flora. [Rep. D.S. Geol.nbsp;Swrv. Territ. vol. vii. 1878.]

Lesquereux, L. (2) Eecent Determinations of Fossil Plants from Kentucky, Louisiana, Oregon, California, Alaska, Greenland, etc., with descriptions of new species. Compiled and editednbsp;by Knowlton. [Proc. U.S. Nat. Mus. vol. xi. 1888, p. 11.]

Lignier, 0. V�g�taux fossiles de Normandie. Structure et affinites du Bennettiies Moriirei (Sap. and Mar.). [M�m. Soc. Linn.nbsp;Normandie, vol. xviii. 1894.]

Mackie, S. J. The �Dragon Tree� of the Kentish Rag. [Geologist, vol. V. 1862, p. 401.]

Mantell, G. (2) Notes on the Wealden Strata of the Isle of Wight, with an account of the bones of Iguanodon and othernbsp;Reptiles discovered at Brook Point and Sandown Bay.nbsp;[Quart. Journ. Geol. Soc. vol. ii. 1846, p. 91.]

Masters^ M. T. Eeview of some points in the Comparative Morphology, Anatomy, and Life-history of the Coniferoe. \Journ. Linn.nbsp;Soc. vol. xxvii. 1891, p. 226.]

Medlicott, H. B. and Blanford., W. T. Manual of the Geology of India.

Edit. ii. Edited by R. D. Oldham. Calcutta, 1893. Miquel, F. A. W. (1) Monographia Cycadearum. Utrecht, 1842.nbsp;Miquel, F. A. W. (2) Over de Rangschikking der fossiele Cycadeae.

\Tijdsch. Wis. Nat. Wet. vol. iv. 1851, p. 205.]

Xquel, F. A. W. (3) Ueber den Bau eines erwaohsenen Stammes von Cycas circinalis. [Unnoea, vol. ii. 1844, p. 125.]

Afoore, C. Notes on the genus Macrozamia. \Journ. Proa. B. Soc. N. S. W. vol. xviii. 1884, p. 115.]

Moriere, J. Note sur deux v�g�taux fossiles trouv�s dans le d�partement du Calvados. [M�m. Soc. Linn. Normandie, vol. xv. 1869.]

(1) Remarks upon the Recent and Fossil Cycadacem. [Annals, vol. vii. 1841, p. 110.]

(2) nbsp;nbsp;nbsp;See Grant, C. W.

(3) nbsp;nbsp;nbsp;See Sharpe.

(4) nbsp;nbsp;nbsp;See Oldham and Morris,nbsp;zu. Beitrage zur Petrefaotenkunde Bayreuth.nbsp;Bayreuth, 1843.

Nathorst, A. G. (1) Sur la presence du genre Dictyozamites, Oldham, dans les couches jurassiques de Bornholm. [�fv. k. Vet.-Akad. Fork. 1889, p. 96.]

Nathorst, A. G. (2) Beriittelse om en med underst�d af allmanna medel ut ford vetenskaplig resa till Soh�sveiz och Tyskland.nbsp;[Ihid. 1881, No. 1, p. 61.]

Nathorst, A. G. (3) Nagra anmarkningar om Williamsonia, Carruthers. [Ihid. 1880, No. 9, p. 33.]

Nathorst, A. G. (4) Nya anmarkningar om Williamsonia. [Ihid. 1888, No. 6, p. 359.]

Nathorst, A. G. (5) Beitrage zur Geologie und Palseontologie der Re-publik Mexico. Felix, J. and Lark, H. Th. ii. Heft i. Leipzig, 1893. Plants by Nathorst.

Nathorst, A. G. (6) Ueber die Wissensohaftlichen Resultate der letzten Schwedischen Expedition nach Spitzbergen. [Verh. k.-k.nbsp;geol. Reichs. 1883, No. 2, p. 25.]

Newberry, J. S. Rhsetic Plants from Honduras. [Amer. Journ. Sci. vol. xxxvi. 1888, p. 342.]

Nilsson, S. V. (1) Om Forsteningar aftryck af tropiska tradslag och deras blad, funne i ett Sandstenslager i Skane. [K. Svensk.nbsp;Vet.-Akad. Hand. 1820, p. 108.]

tillsammans ned kafsmusslor, Snachor m.m. i den Skanska Gr�nsands-kalken. [Ibid. 1824, p. 14.3.]

Oldham, T. and Morris, J. Fossil Flora of the Eajmalial Series in the Eajmahal Hills. [Mem. Oeol. Surv. India (2) vol.i. 1863, pt. i.]

Phillips, J. Illustrations of the Geology of Yorkshire. Pt. i. : The Yorkshire Coast. Edit. iii. London, 1875.

Pomel, A. Mat�riaux pour serrir a. la flore fossile des terrains jurassiques de la France. [Amtlicher Bericht �her die 25i�nbsp;Versammlung Oes. deutsch. Naturforsch. und Arzte in Aachen,nbsp;1847, p. 332. Aachen, 1849.]

Potoni�, H. (1) Ueber das Eothliegende des Thiiringer Waldes. [Abhand. k. prems. Oeol. Landesanst, Heft ix. 1893.]

Potoni�, H. (2) Die fossile Pflanzen-Gattung Tylodendron. [Jahrb. Oeol. Landesanst, 1887, p. 311.]

Renault, B. (1) Structure comparee de quelques tiges de la flore Oarbonif�re. [Nouv. Arch. Mus. Hist. Nat. vol. ii. (2) 1879,nbsp;p. 213.]

Renault, B. (2) Sur les Sphenozamites. [Compt. Rend. vol. xciii. 1881, p. 1165.]

Renault, B. and Zeiller, B. (1) Sur quelques Gycad�es houill�res. [Compt. Rend. vol. oii. 1886, p. 325.]

Renault, B. and Zeiller, B. (2) Etudes sur la terrain houiller de Commentry. Livre deux. [Bull. Soc. indust. min. (3)nbsp;vol. iv. livre ii. St. Etienne, 1890.]

Rendle, A. B. and others. The plants of Milanji, Nyasa-Land, collected by Mr. Alex. Whyte. [Trans. Linn. Soc. vol. iv. 1894,nbsp;pt. i. p. 60.]

Sandberger, F. Die Flora der oberen Steinkohlenformation im badischen Schwarzwalde. [ Verh. Nat. Ver. Carlsruhe,nbsp;vol. i. 1864, p. 30.]

Saporta, O. de. Flore fossile du Portugal. [Dirlct. trav. g�olog. Portugal, Lisbon, 1894.]

Saporta, 0. de and Marion, A. F. (1) Sur les genres Williamsonia, Carr., and Ooniolina, D�Orb. [Compt. Rend. v�l. xcii. 1881,nbsp;p. 1185.]

math.-phys. Cl. k. Sachs. Oes. Wiss. vol. xv. 1889, p. 523.] Schimper and Mougeot. Plantes fossiles du Gr�s Bigarr�. 1840.nbsp;Schmalhausen, J. Die Pflanzenreste der Artinskischen und Permischetinbsp;Ablagerangen in Osten des Europaisohen Eusslands.

Schoenlein, J. L. Abbildungen von fossilen Pflanzen aus dem Keuper Frankens. Wiesbaden, 1865.

Seward, A. C. (1) On the genus ilyeloxylon, Brong. [Annals Hot. vol. vii. 1893, No. xxv. p. 1.]

Seward, A. C. (2) Catalogue of the Mesozoic Plants in the Department of Geology (British Museum). The Wealden Flora, pt. i.nbsp;London, 1894.

Seward, A. G. (3) [Nature, vol. l. 1894, p. 594.] Some new facts with regard to Bennettites.

Sharpe, B. On the Secondary District of Portugal which lies on the North of the Tagus. Eemarks by J. Morris on Zamitesnbsp;gramineus var. mundce, Morr. [Quart. Journ. Oeol. Soo.nbsp;vol. vi. 1850, p. 135.]

Smith, J. Observations on a remarkable Cycadaceous Plant from Port Natal. [Hooker^s Journ. Bot. and Kew Gardennbsp;Miscellany, vol. vi. 1854, p. 88.]

Solms-Lauhach, Graf zu. (1) Ueber die Fructification von Bennettites Gihsonianws, Carr. [Bot. Zeit. 1890, p. 789.]nbsp;Solms-Lauhach, Graf zu. (2) On the Fructification of Bennettitesnbsp;Oihsonianus, Carr. [Annals. Bot. vol. v. 1890-91, p. 419.]nbsp;Solms-Lauhach, Graf zu. (3) Die Sprossfolge der Stangeria und dernbsp;�brigen Cyoadeen. [Bot. Zeit. 1890, p. 177.]

Sterzel, J. T. Die Flora des Eothliegenden im Plauensohen Grunde bei Dresden. [Ahh. math.-phys. Cl. k. Sachs. Oes. Wiss. vol. xix.nbsp;1893.]

Ward, L. F. Fossil Cycadean Trunks of North America, with a revision of the genus Cycadeoidea, Buckland. [Proc. Biol.nbsp;Soc. Washington, vol. ix. 1894, p. 75.]

Williamson, W. C. (1) On the Distribution of Fossil Eemains on the Yorkshire Coast, from the Lower Lias to the Bath Oolitenbsp;inclusive. [Trans. Oeol. Soc. (2) vol. v. 1840, p. 223.]

Williamson, W. C. (2) On the Scaly Vegetable Heads or Collars from Eunswiok Bay, supposed to belong to Zamia gigas-[Proc. Yorks. Phil. Soc. vol. i. 1847, p. 45.]

Williamson, W. O. (3) Contributions towards the History of Zamia gigas, L. and H. [Linn. Trans, vol. xxvi. 1870, p. 663.]

Yokoyama, M. Mesozoic Plants from Kozuke, KU, Awa, and Tosa. [Journ. Coll. Sci. Japan, vol. vii. pt. iii. 1894, p. 201.]

Young, O. and Bird, J. A Geological Survey of the Yorkshire Coast. Whitby, 1822.

Zeiller, R. (1) Bassin houiller et Permien d�Autun et d�Epinac. Paris, 1890.

Zeiller, R. (2) Notes sur la flore des couches permiennes de Trienbaoh (Alsace). [Bull. Soc. Ge'ol. France, vol. xxii. (3) 1894, p. 163.]

Zeiller, B. (3) Observations sur quelques cuticles fossiles. [Ann. Sci. Nat. 1882, p. 217.]

Zigno, A. (2) Monografia del genere Dichopteris nuovo genere di felce fossile. [M�m. instit. Veneto, vol. xii. 1864.]

Zittel, K. A. Handbuch der Palaeontologie. Abtheilung ii. : Palaeo-phytologie. Schimper, IF. P. and Schenk, A. Munich and

eipzig.	1890.
Lieferung i.		- - pp-	1-152 -	1879.
) gt;	ii.	quot; � ?)	153-232 -	1880.
	iii.	quot; nbsp;nbsp;nbsp;-nbsp;nbsp;nbsp;nbsp;jj	233-332 -	1884.
	iv.	quot; nbsp;nbsp;nbsp;quot;nbsp;nbsp;nbsp;nbsp;5 gt;	333-396 -	1885.
	V.	� nbsp;nbsp;nbsp;�nbsp;nbsp;nbsp;nbsp;J J	397-492 -	1887.
J?	vi.	� �	493-572 -	1888.
	vii.	� �	573-668 -	1889.
5?	viii.	� quot; gt; gt;	669-764 -	1889.
J)	ix.	quot; nbsp;nbsp;nbsp;�nbsp;nbsp;nbsp;nbsp;n	765-958 -	1890.

INDEX OE GEXEEA, SPECIES, ETC.

EXPLANATION OF PLATES.

Au. the figured specimens are preserved in the Eritish Museum (Natural History), their registered numbers being quoted innbsp;square brackets. The figures are drawn natural size, except innbsp;few cases where the enlargement is stated. With the exception of those represented on Plates XIII. and XIV., the figurednbsp;specimens are from the Rufford Collection.

Fig. 2. Otozamites Ooppertianus (Dunk.). The lower and middle portion of a frond. P. 70.nbsp;nbsp;nbsp;nbsp;[V. 2360.]

Fig. 3. Otozamites Klifsteinii (Dunk.). Terminal portion of a frond-P. 64. nbsp;nbsp;nbsp;[V. 2336.]

'■ ■ “..Sütf ;*»}» nbsp;nbsp;nbsp;'^- •■ -*-.nbsp;nbsp;nbsp;nbsp;/ ^ / ’■'lt;quot; ■-•

VVx. ^ . ’■ nbsp;nbsp;nbsp;^'.X,'*';;!’k r'nbsp;nbsp;nbsp;nbsp;. . .

B ‘ ' nbsp;nbsp;nbsp;'nbsp;nbsp;nbsp;nbsp;' ,nbsp;nbsp;nbsp;nbsp;‘ ,,.V‘‘^nbsp;nbsp;nbsp;nbsp;'i#-nbsp;nbsp;nbsp;nbsp;■'t i ■nbsp;nbsp;nbsp;nbsp;’,. f /nbsp;nbsp;nbsp;nbsp;^nbsp;nbsp;nbsp;nbsp;'nbsp;nbsp;nbsp;nbsp;' .

t ■ nbsp;nbsp;nbsp;- ‘V. *nbsp;nbsp;nbsp;nbsp;*'hnbsp;nbsp;nbsp;nbsp;■nbsp;nbsp;nbsp;nbsp;’■,'nbsp;nbsp;nbsp;nbsp;• •nbsp;nbsp;nbsp;nbsp;-nbsp;nbsp;nbsp;nbsp;'nbsp;nbsp;nbsp;nbsp;..--. ■'nbsp;nbsp;nbsp;nbsp;'nbsp;nbsp;nbsp;nbsp;4 j

■r- • \ nbsp;nbsp;nbsp;'V.,.;,’*;g:- * ;ü' -

'k -y'- nbsp;nbsp;nbsp;-gt;-.'^5Vc“-'-'^- '-tv-

; é/ S nbsp;nbsp;nbsp;-quot;-r- , quot; lt;fd

r- ' nbsp;nbsp;nbsp;:•'nbsp;nbsp;nbsp;nbsp;'• C'V. ^nbsp;nbsp;nbsp;nbsp;‘, ,■ -■

ï ' gt;-. nbsp;nbsp;nbsp;--V’ - if, .V

'■ nbsp;nbsp;nbsp;' ^ ■•' - •■•', ■. ■■;- •^'' '*nbsp;nbsp;nbsp;nbsp;■nbsp;nbsp;nbsp;nbsp;'’gt;. ^nbsp;nbsp;nbsp;nbsp;•’ -nbsp;nbsp;nbsp;nbsp;.nbsp;nbsp;nbsp;nbsp;-:■ ^.V

K nbsp;nbsp;nbsp;*v ^nbsp;nbsp;nbsp;nbsp;^ -nbsp;nbsp;nbsp;nbsp;**■•*.gt; ,-*1.

gt;. nbsp;nbsp;nbsp;•nbsp;nbsp;nbsp;nbsp;-‘/'f _nbsp;nbsp;nbsp;nbsp;- j 1nbsp;nbsp;nbsp;nbsp;*'j quot; Vnbsp;nbsp;nbsp;nbsp;•'Ï-'

Fig. 1. Withamia armata, gen. et sp. iiov. Single detached leaf, showing -well-marked flabellate venation. P. 177. [V. 2915.]

* Fig. 2. Withamia armata, gen. et sp. nov. Axis bearing recurved spines and imperfect leaves in their axils. P. 177. [V. 2134.]

Fig. 4. Otozamites, sp. Cf. 0. KUpsteinii (Dunk.). Basal portion of a young frond. P. 69.nbsp;nbsp;nbsp;nbsp;[V. 2734.]

* The name Saportaia was unfortunately printed on the plate before the previous use of Saportaia was discovered (p. 174).

1.2. Saporta.LSL,

Fig. 4. Zamites Buchianus (Ett.). Portion of a large frond, showing manner of attachment of a pinna. P. 83.nbsp;nbsp;nbsp;nbsp;[V. 3327.]

Fig. 6. Dioonites Dunkerianus (G�pp.). Portion of a frond, showing the form and manner of attachment of the pinnse. P. 45.

Fig. 7. Cycadites Saportce, sp. nov. Portion of a frond, showing the venation and arrangement of the pinnse. P. 34. [V. 2134a.]

PLATE IV.

Fie. 2. Portion of a large frond, showing the position of the pinnse in the rachis. P. 84.nbsp;nbsp;nbsp;nbsp;[V. 2126c.]

Fig. 3. Terminal portion of a very small frond. P. 84. nbsp;nbsp;nbsp;[V. 2926.]

Figs. 4 and 5. Portions of a large frond, showing manner of attachment of the pinnse. P. 84. nbsp;nbsp;nbsp;[V. 2123ct.]

PLATE V.

Fig. 1. Withamia armata, gen. et s^.noY. Axis with strongly recurved spines, with the basal portions of leaves in their axils. P. ITS.

Cycadolepis.

Figs. 2 and 3. Zamites Garruthersi, sp. nov. The apical and basal portions of a pinna. P. 88.nbsp;nbsp;nbsp;nbsp;[V. 2123c.]

Fig. 4. Zamites Garruthersi, sp. nov. Portion of a frond showing the manner of attachment and venation of the pinniB. P. 88.

In 5a part of a single pinna is slightly enlarged, showing the apex and single vein. P. 34.nbsp;nbsp;nbsp;nbsp;[V. 2797.]

Figs. 6 and 6a. Cycadolepis. A detached scale ; in 6a the incurved distal margin is represented. P. 99.nbsp;nbsp;nbsp;nbsp;[V. 2699.]

?]^s,l-4^.Zainites, Fig.s.5amp;,5a.Cyca(iites.l!'i�.6.Cycadolepis.

Fie. 1. Otozamites Klipsteinii (Dunk.), var. longifolius. Single pinna of the longer and narrower type. P. 68. (See also Fig. 6.)

Fig. 2. 0. Klipsteinii (Dunk.), var. superbus. Single pinna of the shorter and broader type, with lobed margin. P. 66.

[V. 2122a.]

Fig. 5. Part of a frond showing smaller pinnae, �f. PI. I. Fig. 4, and PI. I. Fig. 3. P. 67.nbsp;nbsp;nbsp;nbsp;[V. 2745.]

Fig. 6. 0. Klipsteinii var. longifolius. Single longer and narrower pinna. Gf. Fig. 1. P. 68.nbsp;nbsp;nbsp;nbsp;[V. 2122.]

Fig. 7. 0. Klipsteinii var. superbus. Shorter pinna showing distinctly auriculate base. P. 67.nbsp;nbsp;nbsp;nbsp;[V. 2740.]

Fig. 1. Zamites Buchianus (Ett.). Large frond, from a photograph by Mr. Gepp. One-sixth nat. size. P. 82.nbsp;nbsp;nbsp;nbsp;[V. 2120.]

Fig. 1. Androstrohm Nathorsti, sp. nov. Showing central axis and flattened sporophylls.nbsp;nbsp;nbsp;nbsp;P.nbsp;nbsp;nbsp;nbsp;111.nbsp;nbsp;nbsp;nbsp;[V.nbsp;nbsp;nbsp;nbsp;2810.]

Figs. 3 and 4. Two sporophylls, slightly enlarged, showing pollen-sac impressions. P. 110.nbsp;nbsp;nbsp;nbsp;[V.nbsp;nbsp;nbsp;nbsp;2701.]

Fig. 5. Seed-like body. Cf. Oolithes, Carruthers. P. 107. nbsp;nbsp;nbsp;[V. 2796�.]

Fig. 6. Fittonia Ruffordi, sp. nov. Portion of a stem with well-preserved petiole bases. nbsp;nbsp;nbsp;P.nbsp;nbsp;nbsp;nbsp;133.nbsp;nbsp;nbsp;nbsp;[V.nbsp;nbsp;nbsp;nbsp;2238.]

Fig. 7. Conites armatus, sp. nov. Badly preserved cone, showing the recurved spinous terminations of the scales. P. 222.

.1�4^.Androstrotus. Fig.6 .Fittonia. Fi^. /.Comtes

yVï'' nbsp;nbsp;nbsp;ijjiSS	? 'sE' ti





-r nbsp;nbsp;nbsp;’ * r '..t	* - ^ *

	V;.'


»■ ' nbsp;nbsp;nbsp;•■‘«.Ar Xnbsp;nbsp;nbsp;nbsp;,' /	■ ^-r,'

, .f ^i •' nbsp;nbsp;nbsp;■nbsp;nbsp;nbsp;nbsp;'nbsp;nbsp;nbsp;nbsp;■nbsp;nbsp;nbsp;nbsp;. /, • -'.inbsp;nbsp;nbsp;nbsp;’'•;# ' ' ■'t


	' ,' '■- 'i Z V 7

	' nbsp;nbsp;nbsp;quot;►» tft - J?'h , . ;■.¥' nbsp;nbsp;nbsp;^4nbsp;nbsp;nbsp;nbsp;•.»

	' nbsp;nbsp;nbsp;. ,vnbsp;nbsp;nbsp;nbsp;■'•nbsp;nbsp;nbsp;nbsp;.
«!■

	. .'r* ^ ' '.quot;
	„ nbsp;nbsp;nbsp;j -»nbsp;nbsp;nbsp;nbsp;»^nbsp;nbsp;nbsp;nbsp;^ , nbsp;nbsp;nbsp;gt;■■ -nbsp;nbsp;nbsp;nbsp;.nbsp;nbsp;nbsp;nbsp;4nbsp;nbsp;nbsp;nbsp;j
	gt; 'i-C ■..gt;,' ,

• . ♦ •

Kk*. nbsp;nbsp;nbsp;^nbsp;nbsp;nbsp;nbsp;^nbsp;nbsp;nbsp;nbsp;^•*1/4nbsp;nbsp;nbsp;nbsp;1nbsp;nbsp;nbsp;nbsp;*nbsp;nbsp;nbsp;nbsp;*'* •nbsp;nbsp;nbsp;nbsp;^ ^nbsp;nbsp;nbsp;nbsp;ynbsp;nbsp;nbsp;nbsp;•/nbsp;nbsp;nbsp;nbsp;fL ku*

«•• /‘Zi X- '^: -Vv.' -.r i'

Fig. 1. Unexpanded fructification showing external bracts, and in Fig. la the reticulate lamellar projections from the innernbsp;face of a bract. Fig. 16 represents the conical basal cavitynbsp;of the fructification. P. 157.nbsp;nbsp;nbsp;nbsp;[V. 3177.]

Fig. 3. A small fructification in longitudinal section, showing a spherical boss at the base, and a few involucral bracts withnbsp;the thread-like interstitial organs internal to the bracts.nbsp;P. 160.nbsp;nbsp;nbsp;nbsp;[V. 21296.]

Fig. 4. The basal portion of a longer and expanded fructification, in the centre the base of the central boss, surrounded by thenbsp;reticulately marked peripheral tissue. P. 159.nbsp;nbsp;nbsp;nbsp;[V. 3201.]

Fig. 1. Large expanded bracts below the base of a fructification, the base of the central boss, traces of interstitial organs, andnbsp;the reticulate peripheral tissue. P. 161.nbsp;nbsp;nbsp;nbsp;[V. 2793.]

Fig. 2. Bracts surrounding the conical cavity originally occupied by the central boss of a fructification. P. 161.nbsp;nbsp;nbsp;nbsp;[V. 2129dJ]

Fig. 3. Bennettites (Williamsoma) Carruthersi var. latifolius. Short and broad bracts seen from the under-side; at a lower levelnbsp;portions of the reticulations are shown. P. 163.nbsp;nbsp;nbsp;nbsp;[V. 2129/.]

Fig. 4. The base of the central boss, surrounded by expanded interstitial organs, and below these, in one place, somenbsp;of the reticulations are visible; and at a still lower levelnbsp;portions of short and broad bracts. P. 163.nbsp;nbsp;nbsp;nbsp;[V. 2129e.]

B.M.WEALD EIST PDAKTS .

Fig. 1. Araucarites (Conites), sp. Central axis of cone and imperfect scales. P. 191.nbsp;nbsp;nbsp;nbsp;[V. 3180a.]

Fig. 3. Nageiopsis, sp. A branched specimen with well-preserved leaves.nbsp;nbsp;nbsp;nbsp;P. 211.nbsp;nbsp;nbsp;nbsp;[V.nbsp;nbsp;nbsp;nbsp;3190.]

Fig. 4. nbsp;nbsp;nbsp;Cycadeannbsp;nbsp;nbsp;nbsp;trunk.nbsp;nbsp;nbsp;nbsp;P. 171.nbsp;nbsp;nbsp;nbsp;[V.nbsp;nbsp;nbsp;nbsp;2360.]

Fig. 5. nbsp;nbsp;nbsp;Cycadeannbsp;nbsp;nbsp;nbsp;trunknbsp;nbsp;nbsp;nbsp;showing branching. P. 171.nbsp;nbsp;nbsp;nbsp;[V.nbsp;nbsp;nbsp;nbsp;3162.]

Fig. 6. nbsp;nbsp;nbsp;Dichopteris, sp.nbsp;nbsp;nbsp;nbsp;P. 184.nbsp;nbsp;nbsp;nbsp;[V.nbsp;nbsp;nbsp;nbsp;3146.]

t' 5'\ ■ • gt; - nbsp;nbsp;nbsp;' ■ ^ -ïi ,nbsp;nbsp;nbsp;nbsp;,*■ -v ' ■. . ■■

lt; i;- v%-.' ■' V.'. nbsp;nbsp;nbsp;'. = ■' -f ■' i- V,--:

Fig. 1. Macrozamia heteromera, Moore. Single branched pinna. P. 5. (Royal Gardens, Kew.)

Fig. 3. Encephalartos Ohellinckii, Lem. Pp. 20, 22, 29, etc. (British Museum Herbarium.)

Fig. 5. Encephalartos Ohellinckii, Lem. Cross section of a pinna, showing the revolute edges. P. 29.

K.�s,3-6, Enceplialartos, Eigs ,1�2 , Macrozamia.

PLATE XIV.

Fig. 1. Cf. Beohlesia anornala, gen. et sp. nov. P. 182. nbsp;nbsp;nbsp;[V. 2608.]

Fig. 1. Surface view of stem, showing position of the inflorescences. Fig. 2. Bracts of a single inflorescence.

BenTLettites.

Pig. 1. Pagiophyllum erassifolium (Schenk). Branch with leaves well preserved. P. 213.nbsp;nbsp;nbsp;nbsp;[V. 2803.]

Fig. 2. Pagiophyllum erassifolium (Schenk). Preservation less perfect, and leaves more indistinct than in Fig. 1. P. 213.

Fig. 4. tiphenolepidium Sternbergianwm (Dunk.). Imperfect female cones, with expanded scales. P. 206.nbsp;nbsp;nbsp;nbsp;[V. 2311.]

Fig. 5. Sphenolepidium Sternhergianum (Dunk.). Branches showing spreading leaves. P. 207.nbsp;nbsp;nbsp;nbsp;[V. 2139ci.]

Fig. 6. Sphenolepidium Sternhergianum (Dunk.). Leaves and leaf bases clearly shown. P. 206.nbsp;nbsp;nbsp;nbsp;[V. 2144.]

Figs,l-2,Pagiopliyll^m, Hgs S-B,SpWoiep.d.^xm,

ff ■- r? ’-....IS-;'-•-f nbsp;nbsp;nbsp;'i --r

ült; ■ •' nbsp;nbsp;nbsp;gt; r '.: ■.'; '.e'V :■■nbsp;nbsp;nbsp;nbsp;■ -.a*,. v'- •i-''-'^-:

rf': nbsp;nbsp;nbsp;^nbsp;nbsp;nbsp;nbsp;^nbsp;nbsp;nbsp;nbsp;• .«rf---nbsp;nbsp;nbsp;nbsp;'••*nbsp;nbsp;nbsp;nbsp;*1,,nbsp;nbsp;nbsp;nbsp;-«‘fjnbsp;nbsp;nbsp;nbsp;1» .vim

Fig. 1. Brachyphyllum spinosum, sp. nov. Large branched specimen with leaves, leaf-scars, and thorn-like branches. P. 216.

Fig. 2. ? Brachyphyllum spinosum. Leaves slightly enlarged, showing the form and striate structure. P. 218.nbsp;nbsp;nbsp;nbsp;[V. 2296.]

Fig. 3. Brachyphyllum spinosum. Portion of a thick branch with leaf bases. P. 218.nbsp;nbsp;nbsp;nbsp;[V. 3180.]

Fig. 4. Brachyphyllum spinosum. Portion of a decorticated axis with three branches. P. 217.nbsp;nbsp;nbsp;nbsp;[V. 2240.]

Figs. 5 and 5a. Brachyphyllum spinosum. Branch showing decorticated axis. P. 218. nbsp;nbsp;nbsp;[V. 2760.]

Fig. 6. Brachyphyllum spinosum. One-third nat. size. Decorticated specimen showing the spinous branches. P. 217. [V. 2136.]

Figs. 8 and 8a. Sphenolepidium Kurrianum (Schenk). Single cone more perfectly preserved. P. 203.nbsp;nbsp;nbsp;nbsp;[V. 2213.]

Fig. 1. Sphenolepidium Kurrianum, (Schenk). Large specimen with well-preserved leaves. P. 203.nbsp;nbsp;nbsp;nbsp;[V. 2316c^.]

Fig. 2. Pinites Solmsi, sp. nov. Branches with leaf bases and long needles. P. 197.nbsp;nbsp;nbsp;nbsp;[V. 2169.]

Fig. 3. Pinites Solmsi, sp. nov. Female cone with short branch and narrower needles. P. 197.nbsp;nbsp;nbsp;nbsp;[V. 2255.]

TTl- T P T nbsp;nbsp;nbsp;Tnbsp;nbsp;nbsp;nbsp;� -i �nbsp;nbsp;nbsp;nbsp;T-inbsp;nbsp;nbsp;nbsp;VVCbt.i

t ig , i. bpheno iep 1 dmrn, Fi g s. 2 X, 3. Pmite s

B.M.WEAI^DEN PLANTS.

Fig. 1. Brachyphyllum ohesum. Heer. Twig showing manner of branching. P. 220.nbsp;nbsp;nbsp;nbsp;[V. 2137.]

Fig. 2. Brachyphyllum ohesum. Stouter branch with well-preserved leaves. P. 221.nbsp;nbsp;nbsp;nbsp;[V. 2137a.]

Fig. 6. Thuites valdensis, sp. nov. Twig with distinctly preserved branches and leaves. P. 209.nbsp;nbsp;nbsp;nbsp;[V. 2138.]

T�^igs. 1.2 (X;4', Brachypliyll-um. Big.3. Pagiophyll-axn. � o.PiTiites.nbsp;nbsp;nbsp;nbsp;B.Tkmtes.

Fig. 1. Branch with well-preserved leaf bases, and cones with unexpanded scales. P. 196. nbsp;nbsp;nbsp;'nbsp;nbsp;nbsp;nbsp;[V. 2146.]

Fig. 3. Branch bearing two female cones, and in the upper portion small indistinct structures. P. 197.nbsp;nbsp;nbsp;nbsp;[V. 2146c(.]

MESOZOIC PLANTS

BRITISH MUSEUM

the wealden floea.

PLATES I,-XX.

PRINTED BY ORDER OF THE TRUSTEES.

HEETFORD :

PREFACE

AUTHOR�S PREFACE.

NOTES.

EEEATA.

LIST OF FIGUEES IN THE TEXT.

Group SPERMAPHYTA (PHAYEROGAMIA).

Class GYMNOSPERM^.

1,�Cycadites Romeri, Schenk.

V. 2738. Fig. 1.

2.�Cycadites Saportse, sp. nov.

Dioonites Dunkerianus (G�ppert).

V. 3218. PI. III. Fig. 6.

Dioonites Brongniarti (Mant.).

Nilssonia Schaumburgensis (Dunk.).

* nbsp;nbsp;nbsp;(2), p. 82.

Genus OTOZAMITES, Braun.

Otozamites Klipsteinii (Dunk.).

O. Klipsteinii (Dunk.), var. superba mihi. v. 2745�. PI. I. Pig. 4.

O. Klipsteinii var. longifolia, mihi.

Otozaiuites sp. Cf. O. Klipsteinii (Dunk.).

Otozamites sp. Cf. O. Reibeiroanus, Heer.

Otozamites Goppertianus (Bunk.).

2 (1), p. 122.

Zamites Buchianus' (Ett.).

Zamites Carruthersi, sp. nov.

Specimens of Doubtful Position.

Anomozamites Lyellianus (Dunk.).

Cycadolepis.

1.�Dory- Cycadolepis.

2.�Eury-Cycadolepis.

r.

Carpolithee.

Carpolithes.

Seed-like Bodies of Doubtful Position.

FLORES.

Androstrobus Natborsti, sp. nov.

Conites elegans (Carr.).

TEUNCI.

ortland.

^nehlandi

Cycadeomyelon (Bucklandia), sp.

Fittonia Ruflfordi, sp. nov.

2121

Bennettites (Cycadeoidea) Saxbyanus (Brown).

Cf. Bennettites Saxbyanus.

Bennettites Gibsonianus, Carr.

Bennettites (Cycadeoidea), sp.

Bennettites (Williamsonia).

155

Bennettites CWilliamsonia) Carruthersi, sp. nov.

Bennettites (Williamsonia) Carruthersi, sp. nov� var. latifolius mihi.

? Bennettites.

Yatesia Morrisii, Carr.

Cf. Yatesia Morrisii, Carr.

Trunci, etc. (incertse sedis).

Specimens of Doubtful Position.

Cf. Becklesia anomala, sp. nor.

V. 2608.

Dichopteris, sp. Cf. D. laevigata (Phill.).�

Araucarites (Conites), sp. Cf. Conites elegans (Carr.) and Kaidocarpon rainor, Carr.

Araucarites, sp.

Pinites Dunkeri, Carruthers.

Pinites Carruthersi, Gardner.

V. 2611. PI. XX. Fig. 5. V. 2852.

Pinites Solmsi, sp. dot.

Pinites, sp.

Pinites RufFordi, sp. nov.

Genus SPHENOLEPIDIUM, Heer.

Spheno�epidium Kurrianum (Dunk.).

V. 2285. Probably 8. Kurrianum. Ecclesbourne. Rujford Coll.

Splienolepidium Sternbergianum (Dunk.).

V. 2311. PL XVI. Fig. 4.

V. 2144. PI. XVI. Fig. 6.

Cf. Sphenolepidium (Sequoia) subulatum (Heer).

'■ ■ “..Sütf ;»}» nbsp;nbsp;nbsp;'^- •■ --.nbsp;nbsp;nbsp;nbsp;/ ^ / ’■'lt;quot; ■-•