DESCENT OF THE PRIMATES

LECTURES DELIVERED ON THE OCCASION OF

THE SESQUICENTENNIAL CELEBKA�ION

OF PKINCETON UNIVERSI�Y

u v
A. A. W. HUBRECHT

PROFESSOR OF ZOOLOGY IN THE UNIVBBSITY OF UTRECHT

WITH ILLUSTHAHONS

NEW YORK
CHARLES SCRIBNEir

1897

Copyright, 1897,
BY CIIAULES SCKIUNER'S SONS.

JEntbetsttg

JOHK WlLSON AND SoN, CAMBRIDGE, U. S. A.

THE DESCENT OF THE PRIMATES

NEW FACTS AND OLD PROBLEMS

FOR a student of natural history, be h� a zo�lo-
gist or a botanist, the examination of the external
characters of animals or plants is generalij the
first step by which h� acquaints himself with the
objects of his study. Next comes the investiga-
tion of their internal fabric, of their organs and
of their tissues; finally, the yet more laborious
unravelling of their development. The slow and
numerous steps by which these organs and tissues
have come into being are always identical; they
lead on from a simple fecundated egg-cell to the
complexity of the adult animal or plant. In this
way embryology throws a welcome light upon
questions which anatomy al on e would not enable
him to solve. When once h� has thus become
thoroughly familiar with a considerable number
of facts concerning the particular object of his
researches, h� extends these to other objects, the
final goal of his life being to obtain a glimpse of
l

2 THE DESCENT OF THE PEIMATES

the plan on which living things are built, and of
the way in which they work.

In order to classify the collected facts methodi-
cally, h� looks out for some system or other
according to which a satisfactory arrangement of
the yaried forms of life may be made. More
than a hundred years ago Liunseus produced the
first really all-enibracing arrangement, in his
"Systema Naturse."

Up to 1859, however, all this seemed more or
less artificial; and although a difference was un-
doubtedly made between what was called an
artificial and what was looked upou as a natural
system, still the latter did not comrnend itself to
naturalist» as the expression of some great law
which is even now at work throughout nature,
but more as a cupboard in which the facts
happen to fit in together nicely; whereas in what
is called an artificial system they are, so to say,
heaped together according to size, color, or
number.

Since 1859 a fundamental change has come
over natural history. Thanks to the labors of
Charles Darwin, evolutiori is now as universally
acknowledged as is gravitation, and we have
come to look upon all systems of classification as
preliminary attempts definitely to establish that

THE DESCENT OF THE PBIMATES 3

most venerable tree of descent, that most real
and palpable pedigree, by which all things living
at this present moment are directly connected
with their ancestral forms, that existed in pre-
ceding geological epochs, as far as the earth's
past history reveals the presence upon it of living
beings.

Henceforth there can thus be no two competing
or rival systems, betweeu which biologists may
choose according to their convenience. There
is only one, and every consci�ntieus naturalist
should strive to disentangle, with all the tenacity
and accuracy h� is capable of, the complexity of
as many of its twigs or branches as may happen
to be within his reach.

Tf there is one fact which in the last twenty-
five years has become self-evident, it is the coin-
paratively insignificant place which the whole
number of living species, as we know them,
occupies in this immense tree of descent. Life
reaches down into the early stages of the earth's
youth, and there already we find it divided into
minor sterns and branches, the majority of which
do not send up off-shoots into the present period.
Fortunately, palseontology here steps in, and for
the animals or plants of bygone geological
epochs of which we shall forever miss the possi-

4 THE DESCENT OF THE PRIMATES

bility of studying the complete anatomy, or
the development, we yet dispose of skeletons,
teeth, shells, leaves, portions of bark, etc., by
which we are guided in determining their rela-
tion either to living beings or to other fossil
remains.

In no country has the discovery of important
fossils advanced at such a rapid rate as in
America, both north and south of the equator.
And it is to the undaunted zeal of your explorers,
and to the keen discrimination of your palaeon-
tologists, such as Leidy, Cope, Marsh, Scott,
Osborn, and others, that we owe most valuable
material, which is at the same time the firmest
foundation upon which Evolution can be es-
tablished.

It is just twenty years ago that Huxley, in the
three famous lectures which h� delivered at New
York, called this " the demonstrative evidence of
Evolution." And in those twenty years the
accumulation of new evidence, all tending in the
same direction, has never ceased. It has been
especially volumineus with respect to reptiles and
mammals. In this latter class new orders have
sprung up, new genera have had to be created by
the dozen, and certain skeletons have come to
light which must have belonged to what have

THE DESCENT OF THE PEIMATES 5

beeu very suggestively called collective, or syn-
thetic, types.

Collective types are such as will allow us to
pass by comparatively small and gradual changes
from them to two or more different types, which
in them may have found their starting-poiut.
Such collective types are not liniited to the fossil
fauna. Among living mammals of the higher type
(that is, after exclusion of Duckbills and Marsu-
pials) a very marked collective type is presented by
such animals as the hedgehog and its hairy rela-
tive, the Indian Gymnura. This has been firmly
established by no less an authority than Huxley,
who in 1880, in a celebrated paper, " On the
Application of the Laws of Evolution to the
Arrangement of the Mammalia," vindicated that
position for these two genera, and declared that
in them, even more than in other Insectivora, we
"possess the key to every peculiarity which is
met with in the Primates, the Carnivora, and the
Ungulata."

We shall in due time have to remind ourselves
of this momentous utterance of Huxley, and for
the moment will turn our attention to an animal
which could in no sense be looked upon as a col-
lective type, although some of its details suggest
its significance as an intermediate link between

6 THE DESCENT OF THE PRIMATES

genera otherwise widely separated. It is a smal]
mammal and has been excavated in the lower
Eocene of the United States, at least its skull,
jaws, and teeth. These remains offer certain
points of peculiar interest. Cope, to whom we
owe the first description of this fossil, gave it the
suggestive name of Anaptomorphus homunculus.
He thereby iutended to convey the expression of
the curious fact that, with respect to certain
peculiarities of its dentition, this small creature
rerninded him strongly of man and the higher
monkeys.

Anaptomorphus must have been about the size
of a squirrel, but whether it had a tail or not we
cannot at present say. It had big eyes and was
most probably a nocturnal animal of omnivorous
habits, whereas its brain capacity exceeded that
of any of the lower mammals of corresponding
size. There is, however, one genus of living
mammals with which its discoverer imrnediately
saw it to be closely related, namely, the rare and
quaint spectral Tarsius, of which the natives of
Sumatra, Banka, and Borneo stand in suspicious
dread because of its weird appearance. Eaffles
tells us that when the natives perceive a speci-
men on a tree near their rice fields, they aban-
don these and plant their rice elsewhere, being

THE DESCENT OF THE PEIMATES 7

firmly convinced that misfortune would be in
store for them or their families if they did not do
so. Cuming, who has observed live Tarsius iu the
Philippines, praises its particular cleanliness, and
remarks that when it is disturbed in its cage, it
clenches its teeth together and simultaneously
contracts its facial muscles in the same way as a
monkey would do. Certain peculiarities in the
structure of its legs enable it to accomplish long
jumps. In taking its food it sits down on its
hind-quarters, holding the morsel in its forepaws.
We have reason to believe that this description
of the habits and aspect of Tarsius would, to a
great extent, apply to the fossil genus Anapto-
morphus. And we will now further inquire in
what respect these two isolated genera might
prove useful to us in the determination of certain
points of maminalian affinities. I hope to be
able to make it clear to you that under certain
circumstances the value of such outlying and ap-
parently aberrant fornis, imperfect and few in
number as their rernains may be, can become
quite as decisive to us for the determination of
certain points of mammalian descent, as can, on
another occasion, a great number of fossil remaius,
such as those of the slow gradation by which in
the successive Tertiary deposits the gulf between,

8 THE DESCENT OF THE PEIMATES

say, the five-toed Eocene Condylarthra and the
modern horse may be bridged over.

In the case uuder consideration, the importance
of Anaptomorphus and Tarsius will be thoroughly
understood as soon as we call upon comparative
anatomy and ernbryology to furnish us with cer-
tain crucial facts, by which their exact position in
the mammalian system may be more definitely
determined.

In order to obtain material for studying the
embryology of Tarsius and other mammals, I un-
dertook some years ago a voyage to the Indian
Archipelago. I did not succeed, however, in pro-
curing one live specimen of Tarsius during my
seven months' stay in Java, Sumatra, and Borneo.
Nevertheless, I left behind me drawings and cle-
scriptions of the animal, fluids for the preserva-
tion of its uterus, and full instructions. I have
since been fortunate enough, thanks to the active
co-operation of friends and correspondents, to
obtain an unbroken developmental series of this
rare mammal. More than four hundred and
fifty stages, which range between the moment of
fecundation and that of birth, are already in my
possession ; it is upon these that the conclusions
which I will place before you are based.

Hitherto both Tarsius and Anaptomorphus

THE DESCENT OF THE PRIMATES 9

have ranked with the Lemuroids or Prosimiaa,
an order of mammals which has been looked
upon (as the German name of " Halbaffen" and
the Latin term " Prosimise " implies), as a sort of
half-way house between the lower mammals and
the ruonkeys and man. These Lemuroids are in
the present time restricted to the tropical forests
of the old world.

The island of Madagascar may be said to be a
regular Lemurian " reserve," � numerous species,
quite unknown elsewhere, being there very cu-
riously distributed, some being even restricted to
small districts of the central mountain chain or
of the coast forest.

The abundance of Lemuroids in Madagascar
has induced Haeckel to give to a supposed sub-
merged continent between Madagascar and con-
tinental India the name of Leinuria, and to fix
upon this hypothetical dry land as the original
starting-place whence the higher primates, �
monkeys and man, � slowly evolving out of
Lemuroids, have spread over the globe. We
will by and by see in how far this hypothesis
is confirmed or invalidated by what we know
at present.

In the past history of the earth the distribu-
tion of Lemurs was far less restricted than it is at

10 THK DESCENT OF THE PEIMATES

present. Both in Europe and in America the
Tertiary deposits contain very numerous remains
of species undoubtedly more or less closely allied
to the present Lemuroids, and we may look upon
this order as one which, like the Ungulates, has
ever so many more fossil than living representa-
tives, and which has in those former epochs
taken a inuch more prominent part in the consti-
tution of the mammalian fauna than it does now.
At the same time comparative anatomy shows us
that the Leinurian type of structure can have
been derived front one which need not have
been very distant from that of a collective type,
such as, for example, the Condylarthra, in which
not only Ungulate and Creodont, but also Meso-
dont (Lemurian) characteristics are represented.

At any rate the Lemurs are in no respect a
very specialized order of Mammals.

Now it is in this order, as I have just told you,
that both Tarsius and Anaptomorphus have been
placed by the systematists. With what right ?
we are bound to inquire. I, for one, would reply
to this question: With none at all; and I am
going to argue the case with you.

The only real resemblance is the opposable
thumb, which we find on the fore and hind limbs
both in the Lemurs and in Tarsius, as also the

THE DESCENT OF THE PRIMATES 11

flat nails to the fingers. These are however
replaced on two fingers of the hindfoot by hooked
claws, the Leniurs having only one fiuger thus
exceptionally provided.

On the contrary, a whole family (Arctopitheci)
of monkeys have a more considerable number of
claws instead of nails, and an opposable thumb
to only the hinder of the four extremities, so
that we see that even this most prominent point
of resemblance can claim only a restricted taxo-
nomical value. Systematists have undoubtedly
been led by the peculiar external aspect, perhaps
even by the nocturnal habits, to approach Tar-
sius so closely to the Lemuroids. At the same
time they have never failed to recognize it either
as " une esp�ce anormale" ¹ or as " a very aber-
rant form." ²

Indeed, in very many respects Tarsius does
not fit in with tha Lemurs at all. lts dentition
is much more archaic. lts upper and lower
incisors, especially the latter, as well as its
canine teeth, resemble more closely those of the
Insectivora than they do the modified and spe-
cialized incisors of the Lemurs. And its molar

¹ II. Schlegel, Museum d'Histoire naturelle des Pays-Bas,
Tome VII., Leyde, 1876, p. 331.

² Flower and Lydekker, Mammalia, Liviug aud Extinct,
London, 1891, p. 683.

12 THE DESCENT OP THE PEIMATES

teeth are most decidedly of a more primitive
type, both in the upper and in the lower jaw,
than they are in the Lemurs. The crowns of
the molars of Tarsius, as also those of Anapto-
morphus, whicli resenihle each other most closely,
conform to that type and to the initial variation
of it which Cope has first designated as the tritu-
bercular type. They are tri-cuspid, but the
middle cusp is uot in one line with the two outer
cusps. It lies at a certain distance inward, the
three cusps thus enclosing a triangle which is the
first indication of what in a more elahorate type
of inolar teeth will be the grinding surface. Now,
in the Lemurs this primitive arrangement is less
purely preserved, the true molars being mostly
quadri-tubercular. We find it, on the contrary,
rnost distinctly in those fossil precursors of the
Tertiary mammals that lived in the Mesozoic
period and to which Osborn has given the name
of Inscctivora primitiva, Numerous other points
of difference by which Tarsius is distinguished
from the Lemurs, could be enumerated, many of
them having undoubtedly a deeper significance
than might appear at first sight. We will, how-
ever, allude to them no further, but rather direct
our attention to a very remarkable divergence
between Tarsius and the Lemurs with which we

THE DESCENT OF THE PRIMATES 13

have only become aequaiuted in the past few
years.¹ This divergence concerns the mode of
attachment of the unborn anitnal to its mother
during the period of foetal life.

Lemurs from Madagascar and Lemurs from
India are found to be enveloped wh�e inside
their mother's womb in a closed sac that carries
all over its outer surface an immense number of
small kuobs and excrescences. If we would make
a comparison with a texture perhaps more famil-
iar to you we might say that the outer surface
of this sac, which entirely hides the young ani-
mal from our view, resembles Astrakhan fur. In
technical language it is called " villiferous." The
numerous separate little knobs, or villi, carry
extremely fine blood-vessels that directly com-
municate with the embryo's vascular system.
In this way the whole of the outer surface of
the sac may be considered as being eminently
fit for respiratory function or for nutritive ab-
sorption. The latter functions are actually in-
cumbent upon this villiferous surface, which fits
beautifully into corresponding little cavities by
which the internal surface of the maternal womb
is honey-combed. These two surfaces, the ma-

¹ A. A. W. Hubrecht, Spolia Nemoris; Quarterly Journal
of Microscopical Science, vol. xxxvi.,p. 77.

14 THE DESCENT OF THE PRIMATES

ternal and the embryonic, have, during pregnancy,
developed parallel to each other, this interlock-
ing having simultaneously advanced step by step.
Fine maternal blood-vessels are distributed every-
where close under the surf ace of these raaternal
cavities, and so the mother's blood which is laden
with fresh oxygen, thanks to the mother's breath-
ing, and with nutritive matter, thanks to the
mother's digestion, contains a full store of all
the necessaries which the embryo draws from it
by means of the peculiar arrangements on the
surface of the sac within which it is enclosed.

The attachment between the sac and the
raother is nevertheless quite superficial; they
remain permanently distinct and stand in the
same relation to each other as the hand does
to the glove which covers it, or as the rootlet
does to the damp soil into which it has pene-
trated. This latter comparison, taken from the
vegetable kingdom, is, however, in so far inac-
curate as it is often most difficult to uproot a
plant without tearing some of its rootlets,
whereas we may enucleate a Lemur-foetus out of
its mother's womb without as inuoh as tearing or
displacing even a single cell.

How entirely different these arrangements are
if we now come to consider the young Tarsius !

THE DESCENT OF THE PEIMATES 15

There is no interlocking of maternal and fetal
surfaces. There is no loose attachment between
two extensive vascular expansions. But there
is a very perfect, sharply defined, and compli-
cated organ by means of which the foetus
anchors itself, so to say, into the maternal tissue.
This organ is called the placenta. At the time
of its very early origin, when the young Tarsius
has only just started on its development, this
placenta is seen in its true nature as an embry-
onic proliferatixm. The small embryonic vesicle
may be said to scoop out a circular spot of the
mother's tissue and then and there to attach
itself most firmly, more firrnly than a leech or a
bloodhound, to the inner surface of its mother's
womb. The blood which under other circum-
stances would flow froin a wound thus made,
is carefully stored and conducted by the pro-
liferated embryonic tissues which after some
time succeed in establishing a very complex
spongelike, cavernous structure of purely embry-
onic derivation, in the cavities of which maternal
blood freely circulates. The solid meshwork, on
the other hand, eventually carries very fine and
very numerous embryonic blood-vessels, which
then find themselves bathed in maternal blood.
The way in which this result has been attained

16 THE DESCENT OF THE PRIMATES

is, as you will have understood, diarnetrically
different from what we have just discussed for
the Lemurs.

It should here be remarked that the Lemurine
arrangement is analogous to what we find in the
pig, the horse, and other Uiigulates, whereas
the Tarsius arrangement corresponds in varying
degree to what obtains in Insectivora, Rodents,
Bats, Monkeys, aud Man. We must thus come
to the conclusion that, with respect to its pla-
centation, Tarsius more closely resembles an In-
sectivore than it does a Lemur, a conclusion sim-
ilar to that whieh was derived from its dentition.
Now, with respect to another peculiarity in its
early developrnent, I am going to demonstrate to
you that Tarsius is more akin to a monkey than
a Lemur. And thus I may hope to justify the
conclusion which I have put forward that Tar-
sius is not a Lemur at all, that it should never
have been plac3d alongside of the Lemurs, but
that its position is somewhere between an
unknown type of Insectivores and our modern
monkeys and man.

The peculiarity to which I here allude is of a
somewhat abstruse nature, and I will not attempt
to initiate you into the details of it. Suffice it
to say, that human ernbryologists have noticed in

THE DESCENT OF THE PEIMATES 17

the very early human embryo a peculiar struc-
ture which the Germans have called the " Bauch-
stiel," or " Haftstiel" of the embryo. lts
homologue is found in no other order of Mam-
mals. Only lately it has been definitely settled
that the monkeys also have it,¹ but up to the
present time it was a distinguishing feature of
which the first origin was as yet quite obscure.
This " Haftstiel " or ventral stalk, as it has been
called in English, is a string of tissue connecting
the very young embryo with its envelope, and
differing in many respects from the so-called
umbilical cord which at a later period does the
same. In those Mammals that have a ventral
stalk, the embryo is suspended by it from the
very first, whereas in the others this suspension
is only secondarily brought about by means of
a special outgrowth which travels from the
embryo towards the envelope.

Tarsius throws full light on the obscure origin
of the ventral stalk, and at the same time reveals
itself to be possessed not only of this very Haft-
stiel, which is characteristic of man and
monkeys, but also to share with those two
another very striking peculiarity by which they

¹ E. Seleuka, Studi�n �ber Eutwickelungsgeschiehte der
Tiere, Heft V , Wiesbaden, 1892.
2

18 THE DESCENT OF THE PEIMATES

differ from all other Mammals, namely, the pos-
session of a diminutive yolk-sac which never
entirely fills the cavity of the embryonic vesicle.¹This Jatter point, which I am not going to elu-
cidate any further either, is of all the more
importance as it must keep pace with special
difl'erences of primary importance in the develop-
ment of the germlayers.

And so henceforth we are obliged to range
Tarsius with man and monkeys in one order
which may retairi Linnceus' adequate name of
Primates. We must at the same time recognize
that the facts here alluded to render it highly
improbable that many of the most important
characteristics of the Primates could ever have
been derived from arrangements such as we find
them in the Lemurs; so that the designation of
the latter order by the well-known name of
Prosimise is worse than misleading, because dis-
tinctly false.

A delicate question has yet to be solved with
respect to the fossil genus Anaptomorphus, �
delicate in so far as I can understand your
smiling at the idea that I could give you any
details of the foatal development of a small

¹ A. A. W. Hubrecht, Die Keimblase von Tarsius. Fest-
schrift f�r Gegenbaur, Leipzig, 1896.

THE DESCENT OF THE PEIMATES 19

inonkey of the Eocene period, of which up to
now we possess only oue imperfect skull, and of
which no human eye will ever see an embryo!
And still, if you consider the whole of the rea-
soning as it has here been given, you will agree
that it would be difficult to admit that these
intricate peculiarities which Tarsius shares with
the monkeys, to the exclusion of all other known
mammals, should not also have been possessed
by a fossil genus which resembles Tarsius so
very closely, and which by its dentition ap-
proaches closer yet to man and the anthropoid
apes.

The moment you admit, as I expeet you will
be willing to do, that Anaptomorphus has not
considerably differed from Tarsius with respect
to its embryology, then the order of the Pri-
mates, between which and that of the Lemurs
we have been accumulating anatomical and em-
bryological divergence, also becomes severed from
it in geological time. If as early as the lower
Eocene period, animals have existed that were
possessed of those peculiarities by which the
Primates are eminently distinguished (and it
should be well understood that of late true
monkeys have also been discovered in the older
Tertiary of South America), then it would be

20 THE DESCENT OF THE PKIMATES

obviously impossible to place the Lemurs of the
Tertiary period anywhere on the line of ascent
of the Primates. The moment we have traced
the Primates as far as the earliest Tertiary they
can only be connected genetically to ancestors of
the Secondary period. And we are prevented
from assuming that these Mesozoic ancestors
were in any way Lemur-like, because both anat-
omy and embryology point most distinctly in
another direction, namely, in that of the Insec-
tivora, I would here remind you of Huxley's
verdict pronounced sixteen years ago, to which
I have already alluded, that among the Insec-
tivora, the spiny and the hairy hedge-hog (Erina-
ceus and Gyntnura) represent the most central
type. Curiously enough, these two offer in their
embryonic development certain particulars with
which we can connect the divergences of the
Primates much more easily than with the Leru-
urine development. J

The formation of the human placenta, of the
human amnion, of the human decidua reflexa, is
foreshadowed in the hedgehog's development¹in such a way as to strongly support the views
concerning the pedigree of the Primates here

¹ A. A. W. Hubrecht. The Placentation of Krinaceus euro-
ptcus; Quarterly Journal of Microsc. Science, vol. xxx. 1889.

THE DESCENT OF THE PBIMATES 21

advocated. And so the outcome of all these
considerations may be said to h� a definite sim-
plification of the line of descent of man and the
higher monkeys. We need no longer be puzzled
at that rapid increase in complication of the
all-important placentary arrangement which we
were bound to admit as long as we accorded
to the Lemurs any place among our direct
ancestry.

We may now safely say that these complica-
tions are of an ever so much more remote anti-
quity, and that the insectivorous predecessor of
the Eocene Primates may in inany respects have
been a further differentiated mammal than its
contemporary, the ancestor of the Tertiary
Lemurs.

The tiny little Tarsius has thus shown us that
by judiciously converging anatomical, embryo-
logical, and palseontological sidelights into one
focus, we may sometimes succeed in clearing up
genetic relationships that would otherwise re-
main hopelessly intricate or vaguely coufusing.

We must now try to turn to more general ac-
count what we have here established, remem-
bering that in so doing we are starting on a
hypothetical track that leads us somewhat fur-

22 THE DESCENT OF THE PEIMATES

ther away from the landmarks of observable
facts to which we have up to now held fast most
conscientiously. Firstly, then, I would call your
attention to the probability that man and the
anthropoid apes may be only more distantly
allied to the non-anthropoid old-world monkeys,
at all events less closely than is at present gener-
ally admitted. In respect to details of denti-
tion Anaptomorphus points rather to the An-
thropoidea than to the Catarhine monkeys. So
does Homunculus patagonicus, one of Ameghino's
fossil Cebidse, whose dentition, to quote Osborn's
words, is " as advanced in reduction as that of
man." Secondly, certain Insectivora seem to
realize the archetype of the placentation of man
and the anthropoids, whereas the placentation
of the old-world monkeys, as far as it is known,
would more easily compare to what we find in
Tarsius, there being no decidua reflexa, which is
so essential for the formation of that very
peculiar type of discoid placenta that is com-
mon to Erinaceus, the Anthropomorphse, and
Man.

On these grounds I would not feel justified
in contradictiug a hypothetical view, if one of
you might be found willing to propound it,
according to which a direct ancestor of the

THE DESCENT OF THE PEIMATES 23

anthropoids and man, differing from Simise
Catarhinse, Platyrhinse, and Tarsidffi, must have
existed throughout the Tertiaries, and must have
directly sprung from a Mesozoic insectivorous
aucestor, s mail in size, but already more or less
erect in posture, provided with a spacious brain
cavity, with a decidua reflexa, and with a dis-
coid placenta of the Erinacean type of develop-
ment. Novr, in suggestiug the existence of this
unknown intermediate form, you would not be
overdrawing the amount which is booked to the
credit of scientific speculation in the bank of
probability. As to the smallness in size of
Mesozoic Trituberculata, palseontology not only
gives ample evidence, but it distinctly does not
encourage any other assumption. With regard
to a spacious brain cavity, it should be remem-
bered that among the South American monkeys
certain living genera, by the relative size of
their brain, outstrip the Anthropoidea and man
himself. Already in 1844 Geoffroy places Chry-
sothrix "au premier rang entre tous et a c�t�
de l'homme m�me, si ce n'est au-dessus, par la
masse proportionnelle de leur cerveau," at the
same time drawing attention to the fact that
the brain convolutions are very much less de-
veloped, these convolutions being to a great

24 THE DESCENT OF THE PRIMATES

extent correlated, as Flower reminds us,¹ with
the absolute bulk of the body.

And as to the erect posture, which is gen-
erally looked upon as being the monopoly of
man, the anthropoid apes having it in only a
very imperfect degree, we are in no way obliged
to follow the general belief that this has been
a comparatively late acquirement of our ances-
tors! Nor that they must needs first have
passed through a stage similar to the actual
stage of one of our living anthropoid apes.² The
earliest origin of the erect posture may most
reasonably be moved backwards in geological
time if we are mindful of the following two
facts: First, that the occipital l'oramen of cer-
tain American monkeys (Cebidse and Hapalida�)
is placed ever so much more below the skull
than is the case in many of the anthropoid
apes. We noticed the same in Tarsius and
Anaptomorphus. Now, Tarsius is generally found
in the erect posture, with which this positiou
of the foramen magnum is undoubtedly corre-

¹ Flower and Lydekker, Mammals, Living aud Extinct,
London, 1892, p. 705.

² Cf. Dubois, " Pithecaiitliropus crectus, eine menschenahn-
liche Uebergangsform." His conclusions were discussed and
dissented from by me in the Dutch Review, " De Gids," for
April, 1896.

THE DESCENT OF THE PBIMATES 25

lated, the balancing of the skull and brain on
the spinal column being thereby facilitated.
Many of the lower Primates thus realize con-
ditions highly favorable for the adoption of the
erect posture. Secondly, we should remember
that this erect posture is not even restricted to
the Primates, as we find among the Leinurs
the genus Propithecus which, when it has come
down from a tree, walks about on its hind legs,
even without resting its arms on the ground
as do the Gibbons. This important peculiarity
of Propithecus, figured by Milne-Edwards and
Grandidier in 1875, was already known to
Flaccourt not less than two hundred years ago.
To ascribe the same to the remotest Csnozoic,
or even to the Mesozoic, ancestors of man is
then not in itself irreconcilable with observa-
tion. As to the placental characteristics of this
hypothetical intermediate stage, I should think
that the considerable degree to which already now
the extremes (Erinaceus and Homo) resemble
each other in certain respects, justify us in accept-
ing them as here indicated. It would, neverthe-
less, be worth while to inquire if in any of the
living genera of American, monkeys, a decidua
reflexa and a discoid placenta of Erinacean type
of development might also be shown to exist.

26 THE DESCENT OF THE PRIMATES

That certain newly discovered facts, here re-
peatedly alluded to, tend to justify us in some-
what loosening the human pedigree from that
of the existing monkeys, may perhaps induce
more timid minds (the number of which has
not been decreasing of late) no longer to shrink
froru extending the doctrine of evolution to the
prehistorie development of man himself. Au-
thentic data proving the existence of refined
civilization and of highly developed art recede
backwards into an ever-increasing number of
prehistorie centuries as the archasologists extend
their researches in different parts of the globe.
And it should at the same time not be forgot-
ten that Huxley, more than thirty years ago,
expressed himself in an almost prophetic way
in a seatence which might serve as a motto for
this lecture, when h� said: " If any form of the
doctrine of progressive devolopment is correct,
we must extend by long epochs the most lib-
eral estimate that has yet been made of the
antiquity of man."

We may furthermore ask in how far the rea-
soning which we have applied to the order of
Primates, to their affinities and to their devel-
opment in geological time, will in any way con-

THE DESCENT OF THE PRIMATES 27

tribute to the solution of certain questions of
primary importance concerning the relationship
of the three great subdivisions of the Mammalia
to each other and to the lower classes of Verte-
brate animals. It seems to me that the usual
way of looking upon these three subdivisions,
the Duckbills, the Marsupials, and the Placental
Mammals, as a real and historical sequence, in
which the first, having so many reptilian affini-
ities and standing lowest, gave rise in their turn
to Marsupials, which later on again became modi-
fied into Placentalia, is not in accordance with
their true relationship.

I am in no way starting a new idea in lodging
a protest against the theory of the linear descent
of the Mammalian subgroups. Huxley was per-
haps the first to ventilate the same question.
Three years ago Osborn shook this traditional
arrangement to its foundations, by very cogent
reasoning based on palseontological research, in
his address, "On the Rise of the Mammalia in
North America." There was, as h� expresses it,
" not a succession, but a unity of ancestry of the
Monotremes, Marsupials, and Placentals." Still,
the number of those who do not concur in this
conclusion is very considerable, and not likely to
be diminished for the next few years. Semon has

28 THE DESCENT OF THE PEIMATES

gone to Australia purposely to study the develop-
ment of the Ornithodelphia, an inquiry which
many years ago had been commenced, but never
been terminated, by an English embryologist.
And the principal reason why so much interest
is feit in the development of these very animals
is, that they and not any others are expected to
give us the clue to many points at present insuf-
ficieutly understood in the embryonic history of
the placental mammals.

In this respect I would wish to choose a posi-
tion more or less diametrically opposed to that of
Semon.¹ The Monotremes cannot reasonably be
expected to teach us anything concerning the ear-
liest phases through which the Placentals have
passed. And too long has the conclusion re-
mained unchallenged that, because the Mono-
tremes have been shown to lay eggs of the
Sauropsidan type, the ancestors of the Placentals
must have passed through a stage in which they
necessarily reproduced themselves in the same
manner, certain details of the embryonic sac of
the Mammalia undoubtedly favoring this view.
A reptilian ancestor to the Mammalia has thus
found more favor than an amphibian one; and

¹ See Semon's Lecture in the Report of the Leyden Interna-
tional Zo�logical Congress of 1895, p. 295, footnote.

THE DESCENT OF THE PKIMATES 29

even Osboru accepts the term Sauromammalia in
exchange for Huxley's Hypotheria, thereby im-
plying that certain distinctive characters of those
Proto-mainmalia, which can hardly have been
other than sucli as belong to their reproductive
and developmental arrangements, were distinctly
Saurian.

With respect to this important question I feel
inclined to side rather with Huxley �� whose
brilliant vindication of the amphibian character
of his Hypotheria¹ is a perfect model of morpho-
logical argument � than with the views last-
mentioned, and I will attempt to explain my
reasons for so doing. The differences between
Keptilia and Amphibia, as exhibited by living
representatives of the two classes, are nowhere so
obvious as in the fact that the former possess an
important transitory embryouic structure kuown
as the amnion, whereas the latter do not. In
going back into the Mesozoic and Palaeozoic
periods we are led to presume that such extinct
orders as the Pterosauria, Dinosauria, etc., shared
these developmental features with the modern
reptiles. In fact the large eggs which some of
the latter are supposed to have laid are preserved
in certain Museums. But when we come to yet

¹ Proc. Zool. Society, 1880, p. 649.

30 THE DESCENT OF THE PRIMATES

older orders, that fiourished in the earliest Meso-
zoic, and in the Palaeozoic period, such as the
Theromorpha, we ought to pause before affirming
that they too were already Amniota.

On the other hand, the Palseozoic Stego-
cephala, which are classed with the Amphibians,
might for that reason be said to be as yet de-
prived of an amnion. Nothing however prevents
us from assuming that in the period of the earth's
history in which this class flourished, the earliest
traces of this embryonic structure first originated.
In fact, its appearance is understood to have been
largely influenced by the formidable changes of
habit and of structure which must have come
about at the time the aquatic Vertebrates grad-
ually adopted, first, a semiterrestrial (amphibian)
existence, and then became specialized in differ-
ent classes of terrestrial animals. The majority
of the aquatic Vertebrates may then, as now, have
been oviparous. With the change towards a ter-
restrial existence, the eggs may yet for a long
time have continued to be deposited in the water
by many of them. Others may have adopted the
most various devices for the hatching and the
protection of their eggs, as is still faintly echoed
in those not nuinerous genera of frogs and toads
which carry and batch their eggs, now on their

THE DESCENT OF THE PEIMATES 31

backs, now on their bellies, sometimes in special
pouches, sometimes in carefully constructed nests.

Others again may have developed much larger
eggs, with an albumen layer and a shell. There
can be no doubt on the other hand that a nuinber
of them have retained their eggs in the maternal
oviducts and have hatched them there, forming a
viviparous section.

In fact, there is really not one cogent reason
which ivould prevent us from deriving arrange-
ments as we find them in placental mammals
directly from viviparous amphibian ancestors.
The spherical embryonic vesicle with the en-
closed umbilical sac and with the embryonic area
spread out flat on the top, is not necessarily a deri-
vate of a preceding one in which this umbilical
vesicle enclosed an enormous quantity of fluid
yolk substance. This spherical extension of the
vesicle may also have been reached more directly.
At all events, this possibility has certainly no less
a claim to our careful consideration. Be it well
uriderstood, however, that I do not commit myself
to professing that I feel sure that it really has in
placental mammals developed thus differently.
What I wanted to point out is, that the genera-
tive adaptations baing so varied even amongst
living amphibians, they must have been so on

32 THE DESCENT OF THE PEIMATES

an extensive scale among their much ruore
numerous Palseozoic ancestors. For the present
I hold it to be at the least premature to pin our
faith to the one eventuality which those who
argue the necessity of an intermediate saurian
stage between an arnphibian anamniotic ancestor
and a mammalian amniotic descendant would
wish us to adopt.

Moreover, as the first steps in the phylogenetic
development of amnion and allantois are once
for all out of the reach of direct observation, we
must be guided solely by speculative argument.
And in that case we have a right to exact of those
who feel convinced that a megalecithal saurian
ancestor comss in somewhere in the pedigree of
the placental mammals, that they give us a
plausible hypothesis by which we can explain
the origin of the amuion. As yet they have
utterly failed in this respect, and that most con-
scientious and painstaking embryologist, Professor
Mmot, of Boston, frankly concluded in 1893, that
hardly anything " definite is known as to the
evolution or phylogenetic origin of the amnion."

About a year ago a new hypothesis on the
origin of the amuion was propounded, * in which

¹ A. A. W. Hubrecllt, Die Phylogenie des Aranions und die
Bedeutung des Trophoblast, Verhandel. Kon. Akademie v.
Wetenschappen. Amsterdam, 1895.

THE DESCENT OF THE PKIMATES 33

no place is left for an intermediate saurian link,
but accordiug to which the placental mammals
are connected directly with unknown amphibian
aucestors, leaving the Monotremes, the Marsu-
pials, and the Sauropsida to come in for lateral
connections � as yet wholly unknown � with
that more archaic and direct line of descent.
This hypothesis is essentially based upon ob-
servations that have of late been accumulating
concerning the actual mode of development of
the amnion in different orders of placental
mammals. The development of the amnion in
mammals is often said to conform with that
which we observe in the chick. This may hold
good for the sheep, for the rabbit, and for many
other placental mammals, but it certainly does
not for man, for the flying fox (Pteropus), for
the hedgehog, or for the guinea-pig. In all these
representatives of four different orders of mam-
mals, the origin of the amnion totally differs
from the traditional process just alluded to,
although the latter mode is nevertheless met
with in other members of those same four orders.
Up to the present time embrvologists have been
in the habit of looking upon the traditional
process present in the chick as the typical one
from which it ought to be possible to derive

34 THE DESCENT OF THE PEIMATES

the other modes of development in sorne way
or other, as special adaptations that have indeed
to be explained, but that are of no primary
importance.

The reason why this point has been so little
in dispute may partly be ascribed to the force
of habit, partly to the consideration that a mode
of amnion formation which is prevalent among
all birds and reptiles, must in the nature of
things be a more ancient and a more primitive
mode. All the more this appeared to be the
correct view, as the majority of the mammals
do conform with Sauropsida in the way in which
they form their amnion.

And so it requires a certain amount of deter-
mination to single out the exceptional cases
which we encounter amongst a small minority
of mammals, and to pretend that in them a
more primitive arrangement is preserved.

I will try to explain to you in a few words
which are the chief points at issue in these con-
flicting views. Let me first, then, remind you
that the amnion is a membrane, continuous with
the body wall and stretching hindwards so as to
form a closed sac over the embryo's back. In
it the young animal is enveloped as in a pro-
tective watercushion. Indeed, the amnion is un-

THE DESCENT OF THE PRIMATES 35

deniably a protective apparatus, which is com-
paratively more spacious iu the younger and
more delicate stages than in the older embryos.

Such a protective water-sac is evidently of
more paramount significance to an embryo that
resides inside its mother's generative organs �
where it is exposed to various pressures, peri-
staltic and otherwise � than to one which is
already protected either by a thick layer of
albumen or by a hard shell, or by both. And
so it seems more reasonable to look for its very
earliest origin rather among viviparous than
among oviparous animals. Take, for example,
the shark's and ray's eggs, with their black,
horny egg-case, the fluid albuminous contents,
and the yolk, on the top of which the young
undergoes its successive developmental changes.
We observe a close resemblance with a saurop-
sidan egg, the shell of which is not always hard
and calcareous, but in many cases of leathery
consistency (snakes, crocodiles). Mechanically
speaking, the eggs of both these classes of ver-
tebrates are similarly conditioned, and still
the sharks and rays never possess any amnion,
whereas the crocodiles, turtles, snakes, and birds
do. We may evidently not seek the starting-
point for the amnion formation in any peculiar

36 THE DESCENT OF THE PRIMATES

relation of the embryo either to a big yolk or
to a hard shell.

Most of the hypothetical explanations hitherto
proposed have, however, moved upon that basis,
with what success we have heard Minot affirm.
They do not teil us how the amnion can pos-
sibly have developed phylogenetically. In the
traditional cases we see it arise as a fold, which
slowly and gradually encloses the embryo. And
only when the closure has become final is the
amnion effectim. What, then, were here the in-
cipient stages?

All this appears in another light when we
trust to the exceptional cases above rnentioned
to guide us in determining the phylogeny of the
amnion. We can then start from the much more
reasonable basis that the amnion at its earliest
appearance must have leen a closed sao. Only
on this supposition can it be understood that it
was of high selective significance from the very
first. It was modified only gradually,� one of
the modifications being this, that it did not any
longer arise so very early in individual develop-
ment, but only later by means of the folds
alluded to. Now let me emphasize that at the
present day we do find this very same develop-
ment of the amnion as a closed sac in the

THE DESGENT OF THE PRIMATES 37

above-named representatives of four different
orders of mammals : Cavia (guinea-pig), Erina-
ceus (hedgehog), Pteropus (bat, flying-fox), Homo
(man). This will suffice to convince you that
as far as placental mammals go, the statement
that the amnion arises as an upward-growing
fold, which finally encapsules the embryo, does
not find universal application. There is indeed
an equal chance of the other mode of formation
having been the original one. My own choice
is fixed upon the latter hypothesis because in
the Amphibia, from which I suppose the earliest
placental mammals to have been derived, we
find arrangements that appear to explain the
earliest origin of the amnion in the way liere
advocated. There is, moreover, no difficulty in
tracing both umbilical and allantoidean pla-
centation to a disposition of parts such as we
encounter in the Amphibia. It would, however,
lead me too far if I should attempt to take you
over the whole ground covered by this hypoth-
esis, and it is more than time to turn back to
the realm of facts.

The facts to which I wish to cal l your atten-
tion, and which are confirmatory of the views
here developed, have been brought to light by
different observers, at different times. They all

38 THE DESCENT OF THE PEIMATES

tend to emphasize the possibility of a more
direct comparison between mammals and Ain-
phibia than between mammals and Sauropsida.
In this way Klaatsch¹ calls attention to the close
relations existing between the intestinal arteries
of mammals and the most primitive arrange-
ments of these vessels among amphibians. Else-
where h� declares that the mammals must be
connected with very primitive forms that have
already diverged from the common stem of the
Chordata below the point of divergence of the
amphibians now living. Howes² makes a direct
comparison between the amphibian epiglottis and
that of the mammals. Eabl³ states that the
formation of the heart is accomplished in the
same way in amphibians and mammals. Mau-
rer⁴ comes to the conclusion that with respect
to the epidermal sense-organs and the hairs, the
mammals diverge considerably from the Saurop-
sida, whereas the connection with the Amphibia
seems to be all the more close.

In the definite settling of this question

¹ H. Klaatsch, Zur Morphologie der Mesenterialbilrlungen
am Parmcaiial der Wirbelthiere. Morphol. Jahrb., Bel. 18,
§643.

² G. Howes, Proceed. Zool. Society of London, 1887, p. 50.
^a C. Eabl, Morphologisches Jahrbuch, Bd. 12, p. 273.

⁴ F. Maurer, Morphologisches Jahrbuch, Bd. 18.

THE DESCENT OF THE PKIMATES 39

tology will, of course, have a most influential
voice. We must hope for new discoveries to fill
up the immense gaps which our knowledge of the
Mesozoic and Palseozoic Vertebrata yet contains.
And in respect to that we cannot say but that the
last decades have surpassed our expectations.
Still, it should not be forgotten that even when
all the fossils from those remote periods were
brought to light and were spread out before us,
they would yet remain perfectly mute with
respect to the details of the embryonic develop-
ment of the animals of which they had formed
part, so that on this head even posterity will
have to be satisfied with speculative considera-
tions.

Leaving these for what they are, we may con-
clude by recognizing that Tarsius has taught us
several things: Firstly, to attach more value
than has hitherto been done to the inferences
which can be drawn from certain embryonic
phenomena for classificatory purposes; secondly,
to entertain a certain amount of healthy scep-
ticism with respect to the traditional tables of
mammalian descent. The genera known to us
very rarely converge towards known predecessors
as we go backwards in geological time; their
respective genealogies run much more parallel

40 THE DESCENT OF THE PKIMATES

to each other, the point of meeting being thus
continually transported further backwards to-
wards yet older geological strata. Thirdly, the
necessity of inquiring into the early embryonic
details and placentation of every known genus
of lusectivora and of Primates is imposed upon
us, � the Insectivora being especially instructive
by the suggestive divergences which they offer in
their numerous types of placentation; the Pri-
mates being more especially important with re-
gard to human development.

Seleuka has just made a commencement with
the ape-tribe, but nevertheless our acquaintance
with their development is as yet only very frag-
me.ntary. For a pati�nt explorer there is yet a
very extensive field, and the monkeys of the New
World, as they are sornewhat circumlocutionally
called, are certainly the first on our list of em-
bryological desiderata. I earnestly appeal to
your " scientists," � or, if as a European I might
be allowed to coin an Americanisin, I would
rather say, to your " forshers,"¹ �� to institute that
investigation without delay, even though it is not
in the United States that the material can be
obtained. However, the solidarity of the con-

¹ Perhaps the term " researchere," here and there used to
supplant the word " seientists," might be preferable.

THE DESCENT OF THE PRIMATES 41

tinent as such, being so distinctly insisted upon,
I presume I may express this wish even in a
latitude as high as that of Princeton.

I trust that you will kindly account for my
readiness in formulating this desire by my con-
fessing that in the last few weeks I have con-
tracted the somewhat awkward habit of believing
that the expression of a wish is in this country
the surest and shortest way towards the rapid
realization of it.



					THE

DESCENT OF THE PRIMATES

	LECTURES DELIVERED ON THE OCCASION OF THE SESQUICENTENNIAL CELEBKA�ION OF PKINCETON UNIVERSI�Y

		u v A. A. W. HUBRECHT PROFESSOR OF ZOOLOGY IN THE UNIVBBSITY OF UTRECHT

				WITH ILLUSTHAHONS

			NEW YORK CHARLES SCRIBNEir 1897



			Copyright, 1897, BY CIIAULES SCKIUNER'S SONS.

		JEntbetsttg JOHK WlLSON AND SoN, CAMBRIDGE, U. S. A.



			THE DESCENT OF THE PRIMATES

		NEW FACTS AND OLD PROBLEMS FOR a student of natural history, be h� a zo�lo- gist or a botanist, the examination of the external characters of animals or plants is generalij the first step by which h� acquaints himself with the objects of his study. Next comes the investiga- tion of their internal fabric, of their organs and of their tissues; finally, the yet more laborious unravelling of their development. The slow and numerous steps by which these organs and tissues have come into being are always identical; they lead on from a simple fecundated egg-cell to the complexity of the adult animal or plant. In this way embryology throws a welcome light upon questions which anatomy al on e would not enable him to solve. When once h� has thus become thoroughly familiar with a considerable number of facts concerning the particular object of his researches, h� extends these to other objects, the final goal of his life being to obtain a glimpse of l



		2 THE DESCENT OF THE PEIMATES the plan on which living things are built, and of the way in which they work. In order to classify the collected facts methodi- cally, h� looks out for some system or other according to which a satisfactory arrangement of the yaried forms of life may be made. More than a hundred years ago Liunseus produced the first really all-enibracing arrangement, in his "Systema Naturse." Up to 1859, however, all this seemed more or less artificial; and although a difference was un- doubtedly made between what was called an artificial and what was looked upou as a natural system, still the latter did not comrnend itself to naturalist» as the expression of some great law which is even now at work throughout nature, but more as a cupboard in which the facts happen to fit in together nicely; whereas in what is called an artificial system they are, so to say, heaped together according to size, color, or number. Since 1859 a fundamental change has come over natural history. Thanks to the labors of Charles Darwin, evolutiori is now as universally acknowledged as is gravitation, and we have come to look upon all systems of classification as preliminary attempts definitely to establish that



		THE DESCENT OF THE PBIMATES 3 most venerable tree of descent, that most real and palpable pedigree, by which all things living at this present moment are directly connected with their ancestral forms, that existed in pre- ceding geological epochs, as far as the earth's past history reveals the presence upon it of living beings. Henceforth there can thus be no two competing or rival systems, betweeu which biologists may choose according to their convenience. There is only one, and every consci�ntieus naturalist should strive to disentangle, with all the tenacity and accuracy h� is capable of, the complexity of as many of its twigs or branches as may happen to be within his reach. Tf there is one fact which in the last twenty- five years has become self-evident, it is the coin- paratively insignificant place which the whole number of living species, as we know them, occupies in this immense tree of descent. Life reaches down into the early stages of the earth's youth, and there already we find it divided into minor sterns and branches, the majority of which do not send up off-shoots into the present period. Fortunately, palseontology here steps in, and for the animals or plants of bygone geological epochs of which we shall forever miss the possi-



		4 THE DESCENT OF THE PRIMATES bility of studying the complete anatomy, or the development, we yet dispose of skeletons, teeth, shells, leaves, portions of bark, etc., by which we are guided in determining their rela- tion either to living beings or to other fossil remains. In no country has the discovery of important fossils advanced at such a rapid rate as in America, both north and south of the equator. And it is to the undaunted zeal of your explorers, and to the keen discrimination of your palaeon- tologists, such as Leidy, Cope, Marsh, Scott, Osborn, and others, that we owe most valuable material, which is at the same time the firmest foundation upon which Evolution can be es- tablished. It is just twenty years ago that Huxley, in the three famous lectures which h� delivered at New York, called this " the demonstrative evidence of Evolution." And in those twenty years the accumulation of new evidence, all tending in the same direction, has never ceased. It has been especially volumineus with respect to reptiles and mammals. In this latter class new orders have sprung up, new genera have had to be created by the dozen, and certain skeletons have come to light which must have belonged to what have



		THE DESCENT OF THE PEIMATES 5 beeu very suggestively called collective, or syn- thetic, types. Collective types are such as will allow us to pass by comparatively small and gradual changes from them to two or more different types, which in them may have found their starting-poiut. Such collective types are not liniited to the fossil fauna. Among living mammals of the higher type (that is, after exclusion of Duckbills and Marsu- pials) a very marked collective type is presented by such animals as the hedgehog and its hairy rela- tive, the Indian Gymnura. This has been firmly established by no less an authority than Huxley, who in 1880, in a celebrated paper, " On the Application of the Laws of Evolution to the Arrangement of the Mammalia," vindicated that position for these two genera, and declared that in them, even more than in other Insectivora, we "possess the key to every peculiarity which is met with in the Primates, the Carnivora, and the Ungulata." We shall in due time have to remind ourselves of this momentous utterance of Huxley, and for the moment will turn our attention to an animal which could in no sense be looked upon as a col- lective type, although some of its details suggest its significance as an intermediate link between



		6 THE DESCENT OF THE PRIMATES genera otherwise widely separated. It is a smal] mammal and has been excavated in the lower Eocene of the United States, at least its skull, jaws, and teeth. These remains offer certain points of peculiar interest. Cope, to whom we owe the first description of this fossil, gave it the suggestive name of Anaptomorphus homunculus. He thereby iutended to convey the expression of the curious fact that, with respect to certain peculiarities of its dentition, this small creature rerninded him strongly of man and the higher monkeys. Anaptomorphus must have been about the size of a squirrel, but whether it had a tail or not we cannot at present say. It had big eyes and was most probably a nocturnal animal of omnivorous habits, whereas its brain capacity exceeded that of any of the lower mammals of corresponding size. There is, however, one genus of living mammals with which its discoverer imrnediately saw it to be closely related, namely, the rare and quaint spectral Tarsius, of which the natives of Sumatra, Banka, and Borneo stand in suspicious dread because of its weird appearance. Eaffles tells us that when the natives perceive a speci- men on a tree near their rice fields, they aban- don these and plant their rice elsewhere, being



		THE DESCENT OF THE PEIMATES 7 firmly convinced that misfortune would be in store for them or their families if they did not do so. Cuming, who has observed live Tarsius iu the Philippines, praises its particular cleanliness, and remarks that when it is disturbed in its cage, it clenches its teeth together and simultaneously contracts its facial muscles in the same way as a monkey would do. Certain peculiarities in the structure of its legs enable it to accomplish long jumps. In taking its food it sits down on its hind-quarters, holding the morsel in its forepaws. We have reason to believe that this description of the habits and aspect of Tarsius would, to a great extent, apply to the fossil genus Anapto- morphus. And we will now further inquire in what respect these two isolated genera might prove useful to us in the determination of certain points of maminalian affinities. I hope to be able to make it clear to you that under certain circumstances the value of such outlying and ap- parently aberrant fornis, imperfect and few in number as their rernains may be, can become quite as decisive to us for the determination of certain points of mammalian descent, as can, on another occasion, a great number of fossil remaius, such as those of the slow gradation by which in the successive Tertiary deposits the gulf between,



		8 THE DESCENT OF THE PEIMATES say, the five-toed Eocene Condylarthra and the modern horse may be bridged over. In the case uuder consideration, the importance of Anaptomorphus and Tarsius will be thoroughly understood as soon as we call upon comparative anatomy and ernbryology to furnish us with cer- tain crucial facts, by which their exact position in the mammalian system may be more definitely determined. In order to obtain material for studying the embryology of Tarsius and other mammals, I un- dertook some years ago a voyage to the Indian Archipelago. I did not succeed, however, in pro- curing one live specimen of Tarsius during my seven months' stay in Java, Sumatra, and Borneo. Nevertheless, I left behind me drawings and cle- scriptions of the animal, fluids for the preserva- tion of its uterus, and full instructions. I have since been fortunate enough, thanks to the active co-operation of friends and correspondents, to obtain an unbroken developmental series of this rare mammal. More than four hundred and fifty stages, which range between the moment of fecundation and that of birth, are already in my possession ; it is upon these that the conclusions which I will place before you are based. Hitherto both Tarsius and Anaptomorphus



		THE DESCENT OF THE PRIMATES 9 have ranked with the Lemuroids or Prosimiaa, an order of mammals which has been looked upon (as the German name of " Halbaffen" and the Latin term " Prosimise " implies), as a sort of half-way house between the lower mammals and the ruonkeys and man. These Lemuroids are in the present time restricted to the tropical forests of the old world. The island of Madagascar may be said to be a regular Lemurian " reserve," � numerous species, quite unknown elsewhere, being there very cu- riously distributed, some being even restricted to small districts of the central mountain chain or of the coast forest. The abundance of Lemuroids in Madagascar has induced Haeckel to give to a supposed sub- merged continent between Madagascar and con- tinental India the name of Leinuria, and to fix upon this hypothetical dry land as the original starting-place whence the higher primates, � monkeys and man, � slowly evolving out of Lemuroids, have spread over the globe. We will by and by see in how far this hypothesis is confirmed or invalidated by what we know at present. In the past history of the earth the distribu- tion of Lemurs was far less restricted than it is at



		10 THK DESCENT OF THE PEIMATES present. Both in Europe and in America the Tertiary deposits contain very numerous remains of species undoubtedly more or less closely allied to the present Lemuroids, and we may look upon this order as one which, like the Ungulates, has ever so many more fossil than living representa- tives, and which has in those former epochs taken a inuch more prominent part in the consti- tution of the mammalian fauna than it does now. At the same time comparative anatomy shows us that the Leinurian type of structure can have been derived front one which need not have been very distant from that of a collective type, such as, for example, the Condylarthra, in which not only Ungulate and Creodont, but also Meso- dont (Lemurian) characteristics are represented. At any rate the Lemurs are in no respect a very specialized order of Mammals. Now it is in this order, as I have just told you, that both Tarsius and Anaptomorphus have been placed by the systematists. With what right ? we are bound to inquire. I, for one, would reply to this question: With none at all; and I am going to argue the case with you. The only real resemblance is the opposable thumb, which we find on the fore and hind limbs both in the Lemurs and in Tarsius, as also the



		THE DESCENT OF THE PRIMATES 11 flat nails to the fingers. These are however replaced on two fingers of the hindfoot by hooked claws, the Leniurs having only one fiuger thus exceptionally provided. On the contrary, a whole family (Arctopitheci) of monkeys have a more considerable number of claws instead of nails, and an opposable thumb to only the hinder of the four extremities, so that we see that even this most prominent point of resemblance can claim only a restricted taxo- nomical value. Systematists have undoubtedly been led by the peculiar external aspect, perhaps even by the nocturnal habits, to approach Tar- sius so closely to the Lemuroids. At the same time they have never failed to recognize it either as " une esp�ce anormale" ¹ or as " a very aber- rant form." ² Indeed, in very many respects Tarsius does not fit in with tha Lemurs at all. lts dentition is much more archaic. lts upper and lower incisors, especially the latter, as well as its canine teeth, resemble more closely those of the Insectivora than they do the modified and spe- cialized incisors of the Lemurs. And its molar ¹ II. Schlegel, Museum d'Histoire naturelle des Pays-Bas, Tome VII., Leyde, 1876, p. 331. ² Flower and Lydekker, Mammalia, Liviug aud Extinct, London, 1891, p. 683.



		12 THE DESCENT OP THE PEIMATES teeth are most decidedly of a more primitive type, both in the upper and in the lower jaw, than they are in the Lemurs. The crowns of the molars of Tarsius, as also those of Anapto- morphus, whicli resenihle each other most closely, conform to that type and to the initial variation of it which Cope has first designated as the tritu- bercular type. They are tri-cuspid, but the middle cusp is uot in one line with the two outer cusps. It lies at a certain distance inward, the three cusps thus enclosing a triangle which is the first indication of what in a more elahorate type of inolar teeth will be the grinding surface. Now, in the Lemurs this primitive arrangement is less purely preserved, the true molars being mostly quadri-tubercular. We find it, on the contrary, rnost distinctly in those fossil precursors of the Tertiary mammals that lived in the Mesozoic period and to which Osborn has given the name of Inscctivora primitiva, Numerous other points of difference by which Tarsius is distinguished from the Lemurs, could be enumerated, many of them having undoubtedly a deeper significance than might appear at first sight. We will, how- ever, allude to them no further, but rather direct our attention to a very remarkable divergence between Tarsius and the Lemurs with which we



		THE DESCENT OF THE PRIMATES 13 have only become aequaiuted in the past few years.¹ This divergence concerns the mode of attachment of the unborn anitnal to its mother during the period of foetal life. Lemurs from Madagascar and Lemurs from India are found to be enveloped wh�e inside their mother's womb in a closed sac that carries all over its outer surface an immense number of small kuobs and excrescences. If we would make a comparison with a texture perhaps more famil- iar to you we might say that the outer surface of this sac, which entirely hides the young ani- mal from our view, resembles Astrakhan fur. In technical language it is called " villiferous." The numerous separate little knobs, or villi, carry extremely fine blood-vessels that directly com- municate with the embryo's vascular system. In this way the whole of the outer surface of the sac may be considered as being eminently fit for respiratory function or for nutritive ab- sorption. The latter functions are actually in- cumbent upon this villiferous surface, which fits beautifully into corresponding little cavities by which the internal surface of the maternal womb is honey-combed. These two surfaces, the ma- ¹ A. A. W. Hubrecht, Spolia Nemoris; Quarterly Journal of Microscopical Science, vol. xxxvi.,p. 77.



		14 THE DESCENT OF THE PRIMATES ternal and the embryonic, have, during pregnancy, developed parallel to each other, this interlock- ing having simultaneously advanced step by step. Fine maternal blood-vessels are distributed every- where close under the surf ace of these raaternal cavities, and so the mother's blood which is laden with fresh oxygen, thanks to the mother's breath- ing, and with nutritive matter, thanks to the mother's digestion, contains a full store of all the necessaries which the embryo draws from it by means of the peculiar arrangements on the surface of the sac within which it is enclosed. The attachment between the sac and the raother is nevertheless quite superficial; they remain permanently distinct and stand in the same relation to each other as the hand does to the glove which covers it, or as the rootlet does to the damp soil into which it has pene- trated. This latter comparison, taken from the vegetable kingdom, is, however, in so far inac- curate as it is often most difficult to uproot a plant without tearing some of its rootlets, whereas we may enucleate a Lemur-foetus out of its mother's womb without as inuoh as tearing or displacing even a single cell. How entirely different these arrangements are if we now come to consider the young Tarsius !



		THE DESCENT OF THE PEIMATES 15 There is no interlocking of maternal and fetal surfaces. There is no loose attachment between two extensive vascular expansions. But there is a very perfect, sharply defined, and compli- cated organ by means of which the foetus anchors itself, so to say, into the maternal tissue. This organ is called the placenta. At the time of its very early origin, when the young Tarsius has only just started on its development, this placenta is seen in its true nature as an embry- onic proliferatixm. The small embryonic vesicle may be said to scoop out a circular spot of the mother's tissue and then and there to attach itself most firmly, more firrnly than a leech or a bloodhound, to the inner surface of its mother's womb. The blood which under other circum- stances would flow froin a wound thus made, is carefully stored and conducted by the pro- liferated embryonic tissues which after some time succeed in establishing a very complex spongelike, cavernous structure of purely embry- onic derivation, in the cavities of which maternal blood freely circulates. The solid meshwork, on the other hand, eventually carries very fine and very numerous embryonic blood-vessels, which then find themselves bathed in maternal blood. The way in which this result has been attained



		16 THE DESCENT OF THE PRIMATES is, as you will have understood, diarnetrically different from what we have just discussed for the Lemurs. It should here be remarked that the Lemurine arrangement is analogous to what we find in the pig, the horse, and other Uiigulates, whereas the Tarsius arrangement corresponds in varying degree to what obtains in Insectivora, Rodents, Bats, Monkeys, aud Man. We must thus come to the conclusion that, with respect to its pla- centation, Tarsius more closely resembles an In- sectivore than it does a Lemur, a conclusion sim- ilar to that whieh was derived from its dentition. Now, with respect to another peculiarity in its early developrnent, I am going to demonstrate to you that Tarsius is more akin to a monkey than a Lemur. And thus I may hope to justify the conclusion which I have put forward that Tar- sius is not a Lemur at all, that it should never have been plac3d alongside of the Lemurs, but that its position is somewhere between an unknown type of Insectivores and our modern monkeys and man. The peculiarity to which I here allude is of a somewhat abstruse nature, and I will not attempt to initiate you into the details of it. Suffice it to say, that human ernbryologists have noticed in



		THE DESCENT OF THE PEIMATES 17 the very early human embryo a peculiar struc- ture which the Germans have called the " Bauch- stiel," or " Haftstiel" of the embryo. lts homologue is found in no other order of Mam- mals. Only lately it has been definitely settled that the monkeys also have it,¹ but up to the present time it was a distinguishing feature of which the first origin was as yet quite obscure. This " Haftstiel " or ventral stalk, as it has been called in English, is a string of tissue connecting the very young embryo with its envelope, and differing in many respects from the so-called umbilical cord which at a later period does the same. In those Mammals that have a ventral stalk, the embryo is suspended by it from the very first, whereas in the others this suspension is only secondarily brought about by means of a special outgrowth which travels from the embryo towards the envelope. Tarsius throws full light on the obscure origin of the ventral stalk, and at the same time reveals itself to be possessed not only of this very Haft- stiel, which is characteristic of man and monkeys, but also to share with those two another very striking peculiarity by which they ¹ E. Seleuka, Studi�n �ber Eutwickelungsgeschiehte der Tiere, Heft V , Wiesbaden, 1892. 2



		18 THE DESCENT OF THE PEIMATES differ from all other Mammals, namely, the pos- session of a diminutive yolk-sac which never entirely fills the cavity of the embryonic vesicle.¹This Jatter point, which I am not going to elu- cidate any further either, is of all the more importance as it must keep pace with special difl'erences of primary importance in the develop- ment of the germlayers. And so henceforth we are obliged to range Tarsius with man and monkeys in one order which may retairi Linnceus' adequate name of Primates. We must at the same time recognize that the facts here alluded to render it highly improbable that many of the most important characteristics of the Primates could ever have been derived from arrangements such as we find them in the Lemurs; so that the designation of the latter order by the well-known name of Prosimise is worse than misleading, because dis- tinctly false. A delicate question has yet to be solved with respect to the fossil genus Anaptomorphus, � delicate in so far as I can understand your smiling at the idea that I could give you any details of the foatal development of a small ¹ A. A. W. Hubrecht, Die Keimblase von Tarsius. Fest- schrift f�r Gegenbaur, Leipzig, 1896.



		THE DESCENT OF THE PEIMATES 19 inonkey of the Eocene period, of which up to now we possess only oue imperfect skull, and of which no human eye will ever see an embryo! And still, if you consider the whole of the rea- soning as it has here been given, you will agree that it would be difficult to admit that these intricate peculiarities which Tarsius shares with the monkeys, to the exclusion of all other known mammals, should not also have been possessed by a fossil genus which resembles Tarsius so very closely, and which by its dentition ap- proaches closer yet to man and the anthropoid apes. The moment you admit, as I expeet you will be willing to do, that Anaptomorphus has not considerably differed from Tarsius with respect to its embryology, then the order of the Pri- mates, between which and that of the Lemurs we have been accumulating anatomical and em- bryological divergence, also becomes severed from it in geological time. If as early as the lower Eocene period, animals have existed that were possessed of those peculiarities by which the Primates are eminently distinguished (and it should be well understood that of late true monkeys have also been discovered in the older Tertiary of South America), then it would be



		20 THE DESCENT OF THE PKIMATES obviously impossible to place the Lemurs of the Tertiary period anywhere on the line of ascent of the Primates. The moment we have traced the Primates as far as the earliest Tertiary they can only be connected genetically to ancestors of the Secondary period. And we are prevented from assuming that these Mesozoic ancestors were in any way Lemur-like, because both anat- omy and embryology point most distinctly in another direction, namely, in that of the Insec- tivora, I would here remind you of Huxley's verdict pronounced sixteen years ago, to which I have already alluded, that among the Insec- tivora, the spiny and the hairy hedge-hog (Erina- ceus and Gyntnura) represent the most central type. Curiously enough, these two offer in their embryonic development certain particulars with which we can connect the divergences of the Primates much more easily than with the Leru- urine development. J The formation of the human placenta, of the human amnion, of the human decidua reflexa, is foreshadowed in the hedgehog's development¹in such a way as to strongly support the views concerning the pedigree of the Primates here ¹ A. A. W. Hubrecht. The Placentation of Krinaceus euro- ptcus; Quarterly Journal of Microsc. Science, vol. xxx. 1889.



		THE DESCENT OF THE PBIMATES 21 advocated. And so the outcome of all these considerations may be said to h� a definite sim- plification of the line of descent of man and the higher monkeys. We need no longer be puzzled at that rapid increase in complication of the all-important placentary arrangement which we were bound to admit as long as we accorded to the Lemurs any place among our direct ancestry. We may now safely say that these complica- tions are of an ever so much more remote anti- quity, and that the insectivorous predecessor of the Eocene Primates may in inany respects have been a further differentiated mammal than its contemporary, the ancestor of the Tertiary Lemurs. The tiny little Tarsius has thus shown us that by judiciously converging anatomical, embryo- logical, and palseontological sidelights into one focus, we may sometimes succeed in clearing up genetic relationships that would otherwise re- main hopelessly intricate or vaguely coufusing. We must now try to turn to more general ac- count what we have here established, remem- bering that in so doing we are starting on a hypothetical track that leads us somewhat fur-



		22 THE DESCENT OF THE PEIMATES ther away from the landmarks of observable facts to which we have up to now held fast most conscientiously. Firstly, then, I would call your attention to the probability that man and the anthropoid apes may be only more distantly allied to the non-anthropoid old-world monkeys, at all events less closely than is at present gener- ally admitted. In respect to details of denti- tion Anaptomorphus points rather to the An- thropoidea than to the Catarhine monkeys. So does Homunculus patagonicus, one of Ameghino's fossil Cebidse, whose dentition, to quote Osborn's words, is " as advanced in reduction as that of man." Secondly, certain Insectivora seem to realize the archetype of the placentation of man and the anthropoids, whereas the placentation of the old-world monkeys, as far as it is known, would more easily compare to what we find in Tarsius, there being no decidua reflexa, which is so essential for the formation of that very peculiar type of discoid placenta that is com- mon to Erinaceus, the Anthropomorphse, and Man. On these grounds I would not feel justified in contradictiug a hypothetical view, if one of you might be found willing to propound it, according to which a direct ancestor of the



		THE DESCENT OF THE PEIMATES 23 anthropoids and man, differing from Simise Catarhinse, Platyrhinse, and Tarsidffi, must have existed throughout the Tertiaries, and must have directly sprung from a Mesozoic insectivorous aucestor, s mail in size, but already more or less erect in posture, provided with a spacious brain cavity, with a decidua reflexa, and with a dis- coid placenta of the Erinacean type of develop- ment. Novr, in suggestiug the existence of this unknown intermediate form, you would not be overdrawing the amount which is booked to the credit of scientific speculation in the bank of probability. As to the smallness in size of Mesozoic Trituberculata, palseontology not only gives ample evidence, but it distinctly does not encourage any other assumption. With regard to a spacious brain cavity, it should be remem- bered that among the South American monkeys certain living genera, by the relative size of their brain, outstrip the Anthropoidea and man himself. Already in 1844 Geoffroy places Chry- sothrix "au premier rang entre tous et a c�t� de l'homme m�me, si ce n'est au-dessus, par la masse proportionnelle de leur cerveau," at the same time drawing attention to the fact that the brain convolutions are very much less de- veloped, these convolutions being to a great



		24 THE DESCENT OF THE PRIMATES extent correlated, as Flower reminds us,¹ with the absolute bulk of the body. And as to the erect posture, which is gen- erally looked upon as being the monopoly of man, the anthropoid apes having it in only a very imperfect degree, we are in no way obliged to follow the general belief that this has been a comparatively late acquirement of our ances- tors! Nor that they must needs first have passed through a stage similar to the actual stage of one of our living anthropoid apes.² The earliest origin of the erect posture may most reasonably be moved backwards in geological time if we are mindful of the following two facts: First, that the occipital l'oramen of cer- tain American monkeys (Cebidse and Hapalida�) is placed ever so much more below the skull than is the case in many of the anthropoid apes. We noticed the same in Tarsius and Anaptomorphus. Now, Tarsius is generally found in the erect posture, with which this positiou of the foramen magnum is undoubtedly corre- ¹ Flower and Lydekker, Mammals, Living aud Extinct, London, 1892, p. 705. ² Cf. Dubois, " Pithecaiitliropus crectus, eine menschenahn- liche Uebergangsform." His conclusions were discussed and dissented from by me in the Dutch Review, " De Gids," for April, 1896.



		THE DESCENT OF THE PBIMATES 25 lated, the balancing of the skull and brain on the spinal column being thereby facilitated. Many of the lower Primates thus realize con- ditions highly favorable for the adoption of the erect posture. Secondly, we should remember that this erect posture is not even restricted to the Primates, as we find among the Leinurs the genus Propithecus which, when it has come down from a tree, walks about on its hind legs, even without resting its arms on the ground as do the Gibbons. This important peculiarity of Propithecus, figured by Milne-Edwards and Grandidier in 1875, was already known to Flaccourt not less than two hundred years ago. To ascribe the same to the remotest Csnozoic, or even to the Mesozoic, ancestors of man is then not in itself irreconcilable with observa- tion. As to the placental characteristics of this hypothetical intermediate stage, I should think that the considerable degree to which already now the extremes (Erinaceus and Homo) resemble each other in certain respects, justify us in accept- ing them as here indicated. It would, neverthe- less, be worth while to inquire if in any of the living genera of American, monkeys, a decidua reflexa and a discoid placenta of Erinacean type of development might also be shown to exist.



		26 THE DESCENT OF THE PRIMATES That certain newly discovered facts, here re- peatedly alluded to, tend to justify us in some- what loosening the human pedigree from that of the existing monkeys, may perhaps induce more timid minds (the number of which has not been decreasing of late) no longer to shrink froru extending the doctrine of evolution to the prehistorie development of man himself. Au- thentic data proving the existence of refined civilization and of highly developed art recede backwards into an ever-increasing number of prehistorie centuries as the archasologists extend their researches in different parts of the globe. And it should at the same time not be forgot- ten that Huxley, more than thirty years ago, expressed himself in an almost prophetic way in a seatence which might serve as a motto for this lecture, when h� said: " If any form of the doctrine of progressive devolopment is correct, we must extend by long epochs the most lib- eral estimate that has yet been made of the antiquity of man." We may furthermore ask in how far the rea- soning which we have applied to the order of Primates, to their affinities and to their devel- opment in geological time, will in any way con-



		THE DESCENT OF THE PRIMATES 27 tribute to the solution of certain questions of primary importance concerning the relationship of the three great subdivisions of the Mammalia to each other and to the lower classes of Verte- brate animals. It seems to me that the usual way of looking upon these three subdivisions, the Duckbills, the Marsupials, and the Placental Mammals, as a real and historical sequence, in which the first, having so many reptilian affini- ities and standing lowest, gave rise in their turn to Marsupials, which later on again became modi- fied into Placentalia, is not in accordance with their true relationship. I am in no way starting a new idea in lodging a protest against the theory of the linear descent of the Mammalian subgroups. Huxley was per- haps the first to ventilate the same question. Three years ago Osborn shook this traditional arrangement to its foundations, by very cogent reasoning based on palseontological research, in his address, "On the Rise of the Mammalia in North America." There was, as h� expresses it, " not a succession, but a unity of ancestry of the Monotremes, Marsupials, and Placentals." Still, the number of those who do not concur in this conclusion is very considerable, and not likely to be diminished for the next few years. Semon has



		28 THE DESCENT OF THE PEIMATES gone to Australia purposely to study the develop- ment of the Ornithodelphia, an inquiry which many years ago had been commenced, but never been terminated, by an English embryologist. And the principal reason why so much interest is feit in the development of these very animals is, that they and not any others are expected to give us the clue to many points at present insuf- ficieutly understood in the embryonic history of the placental mammals. In this respect I would wish to choose a posi- tion more or less diametrically opposed to that of Semon.¹ The Monotremes cannot reasonably be expected to teach us anything concerning the ear- liest phases through which the Placentals have passed. And too long has the conclusion re- mained unchallenged that, because the Mono- tremes have been shown to lay eggs of the Sauropsidan type, the ancestors of the Placentals must have passed through a stage in which they necessarily reproduced themselves in the same manner, certain details of the embryonic sac of the Mammalia undoubtedly favoring this view. A reptilian ancestor to the Mammalia has thus found more favor than an amphibian one; and ¹ See Semon's Lecture in the Report of the Leyden Interna- tional Zo�logical Congress of 1895, p. 295, footnote.