TER VERKRIJGING VAN DEN GRAAD VAN
DOCTOR IN DE WIS- EN NATUURKUNDE
AAN DE RIJKS-UNIVERSITEIT TE UTRECHT,
OP GEZAG VAN DEN RECTOR-MAGNIFICUS
Dr. H. BOLKESTEIN, HOOGLEERAAR IN DE
FACULTEIT DER LETTEREN EN WIJSBEGEERTE.
VOLGENS BESLUIT VAN DEN SENAAT DER
UNIVERSITEIT TEGEN DE BEDENKINGEN
VAN DE FACULTEIT DER WIS- EN NATUUR-
KUNDE TE VERDEDIGEN OP
MAANDAG 15 APRIL 1935
DES NAMIDDAGS TE 3 UUR
DOOR

Gaarne maak ik gebruik van de goede gewoonte, die den
promovendus gelegenheid biedt om allen, die tot zijn weten-
schappelijke vorming hebben bijgedragen, daarvoor zijn dank
te betuigen.

Hooggeleerde Kluyver, meer nog dan het verschijnen van
mijn proefschrift acht ik de tijd, welke ik in Uw laboratorium
heb doorgebracht de bekroning van mijn studie. Het valt mij
moeilijk mijn gevoelens jegens U onder woorden te brengen.
Gij waart het die mij in staat stelde het onderwerp mijner
keuze te bewerken voor mijn proefschrift. Dat Gij dit ver-
trouwen in mij zijt blijven stellen, zoo zelfs, dat Gij mij later
in Uw „stafquot; hebt opgenomen, heeft mij niet alleen met groote
dankbaarheid vervuld, maar heeft mij een zelfvertrouwen ge-
geven, dat naar mijn overtuiging onmisbaar is voor alle weten-
schappelijke arbeid. Ik acht het een voorrecht mij in de pret-
tige spheer van noesten arbeid, samenwerking en vriend-
schap, welke in Uw laboratorium heerscht, thuis gevoeld te
hebben en ik acht mij gelukkig mij in de groote familie van
U en Uwe leerlingen en oud-leerlingen opgenomen te weten.
Niet minder dankbaar ben ik voor de gastvrijheid welke ik
meermalen van U en Uw vrouw in Uw woning mocht onder-
vinden.

Onafscheidelijk verbonden met véle prettige herinneringen
aan het Delftsche laboratorium zijn de steun en de vriend-
schap, welke ik mocht ondervinden van U Zeergeleerde King-
ma Boltjes en Muller, van U, Mevrouw Kingma Boltjes en U,
Waarde Hoogerheide. Dank voor de vele goede zorgen welke
Gij mij ten deel hebt doen vallen.

Hoewel het mij nimmer heeft gespeten de laatste studie-
jaren in Delft te hebben doorgebracht, gedenk ik toch met
gevoelens van dankbaarheid de tijd dat ik studeerde aan de
Utrechtsche Universiteit. De veelzijdige opleiding die ik daar
heb genoten is een leerschool geweest, welke ik met gaarne
gemist zou hebben.

In de eerste plaats gaat hiervoor mijn dank uit naar U,
Hooggeleerde Went. Gij waart voor mij een voorbeeld in de
nauwgezetheid, welke Gij betrachtet bij het vervullen van al
datgeen wat Gij Uw plicht vondt. Uw voorbeeld was het dat

Uw leerlingen ertoe bracht om U slechts het beste te toonen
waartoe zij in staat waren en Uw voorbeeld dwong hen uiterst
critisch te staan, zoowel tegenover de onderzoekingen van
henzelve, als tegenover die van anderen.

Hooggeleerde Koningsberger, Hooggeachte Promotor, een
aangename taak kan het niet voor U zijn om als mijn promotor
op te treden, terwijl het proefschrift buiten U om is bewerkt.
Dat Gij mij toch hierin tegemoet zijt gekomen en met terzijde-
stelling van Uw eigen persoon eraan hebt medegewerkt om
dit proefschrift nog tijdig gereed te doen komen, stemt mij
tot groote dankbaarheid.

Hooggeleerde Westerdijk, Honing, Jordan, Kruyt, Nierstrasz
en PuUe, Zeergeleerde Burger, Hirsch, Schuurmans Stek-
hoven en Vonk, uw colleges en practica zijn voor mij een
integreerend bestanddeel geweest van mijn studie aan dé
Utrechtsche Universiteit. Het is mij niet mogelijk U anders
dan in één adem te noemen, want het onderwijs van een ieder
van U vulde dat van den ander aan en juist het geheel ervan
heb ik ondervonden als een kostelijk en afgerond bezit. Heb
dank voor het aandeel dat een ieder van U in mijn opleiding
heeft gehad.

Hooggeleerde le Cosquino de Bussy en Möhr, gij beiden ver-
staat de kunst door woord en beeld belangstelling te wekken
en gaande te houden voor de taak, die er in Indië weggelegd
is voor den bioloog en bodemkundige. Dat ik ook persoonlijk
zooveel belangstelling van U mocht ondervinden, heeft mij
zeer aan U verplicht.

Hooggeleerde Ornstein, waarde Mejuffrouw Eymers, waar-
de Vermeulen, dankbaar ben ik U voor de voorlichting en
ondersteuning, die ik van U ondervond, want zonder dat zou
dit proefschrift niet in dezen vorm tot stand gekomen zijn.

Een deel van de Utrechtsche Universiteit was voor mij het
U. S. C. en de U. S. R quot;Tritonquot;. De vriendschap daar onder-
vonden, is voor mij van onschatbare waarde.

Zeergeleerde vor der Rake, dat ik aan een deel van mijn
H.B.S.-tijd met genoegen terugdenk, dat heb ik te danken aan
U en aan de docenten van het Baarnsch Lyceum. Ook op deze
plaats wil ik mijn erkentelijkheid daarvoor betuigen.

Tenslotte wil ik niet nalaten hier te gewagen van de groote
medewerking, welke ik bij het bewerken van mijn proefschrift
ondervonden heb van het personeel van het Laboratorium
voor Mikrobiologie te Delft en van het Botanisch Laborato-
rium te Utrecht. Ik dank hen allen daarvoor zeer.

b.nbsp;Influence of preceding periods of darkness .

c.nbsp;„ „ temperature . ......

d.nbsp;„ „ light..........

e.nbsp;Substrate of the autofermentation.....

f.nbsp;Supposed sulphate reduction......

§ 3. Experiments with bacteria cultivated in an orga

nie medium .............54

§ 4. Behaviour of the Thiorhodaceae towards oxygen 61
§ 5. Summary ..............62

Chapter IV. The carbon dioxide assimilation of the Thiorho-
daceae.

§ 1. Introductory remarks..........64

b.nbsp;Bacteria from an organic culture medium ...nbsp;71
§ 3. The carbon dioxide assimilation with oxidizable

b.nbsp;Experiments with sulphite and thiosulphate . .nbsp;81
§ 4. The carbon dioxide assimilation with gaseous

§ 8. Experiments with carbon monoxide and methanenbsp;91
§ 9. The carbon dioxide assimilation with organic

The extensive and careful investigations of van Niex. (1931)'
on the coloured sulphur bacteria have thrown a new light
on the morphology, physiology and ecology of this group of
organisms. The chief result of his study is undoubtedly the
elucidation of the main principles underlying the metabolism
of these organisms. This is the more important, since previous
investigators held very divergent and partly conflicting opi-
nions regarding this subject. Van Niel succeeded in bringing
experimental proof for his theory that in purple and green
sulphur bacteria we are dealing with a group of organisms
which for their proliferation are dependent on a very special
type of photochemical carbon dioxide assimilation. Whilst
in photosynthesis of the ordinary green plants the carbon
dioxide reacts with water, in the case of the purple sulphur
bacteria the water is replaced by oxidizable sulphur com-
pounds or organic substances.

Van Niel and Muller (1931) have rightly emphasized the
great importance of a closer study of the photochemical car-
bon dioxide assimilation in the Thiorhodaceae^) for our con-
ception of photosynthesis in general.

The theoretical considerations of these authors made me
decide to study the process in question with the aid of the
methods .which had been so successfully applied by Warburg
and Negelein (1923) to the study of the photosynthesis of
unicellular chlorophyll-containing organisms (green algae)
(C. f. chapter IV).

Such a study seemed the more interesting, since the anae-
robic nature of the purple sulphur bacteria implies the ab-

As Thiorhodaceae or purple sulphur bacteria, or in the abbrevia-
ted form p. s. b., I will designate those Rhodobacteriales, which are
obligate anaerobes and which are characterized by a photosynthetic
autotrophic mode of life. Athiorhodaceae, or purple bacteria (p.b.),
are those Rhodobacteriales, which are facultative-anaerobes, obligate
heterotrophonts and which are obligately photosynthetic in an oxygen-
free medium, (van Niel and Muller 1931).

sence of a true respiration process, which in the study of the
green algae complicates the situation. Yet it appeared pro-
bable that in the Thiorhodaceae too some energy yielding pro-
cess of still unknown nature would proceed during periods
of darkness and therefore also a study of this point might be
deemed of interest (C. f. Chapter III).

Finally it seemed attractive to study the energy relations in
the photochemical carbon dioxide assimilation of the Thiorho-
daceae. The fact that on the one hand this process is closely
related to ordinary photosynthesis, but that on the other hand
other photochemically active pigments and different reaction
components are involved, seems to offer favourable condi-
tions for a deeper penetration into the mechanism of the ener-
gy transference in the photochemical processes of both green
plants and purple bacteria. As a first step in this direction a
determination of the number of light quanta required for the
reduction of one molecule of carbon dioxide appeared to be
of much importance, especially since theoretically it must be
considered possible to determine this number for different
wave lengths and whilst employing different hydrogen dona-
tors (C.f. Chapter V).

It may be remarked in advance that I had no opportunity
to exhaust completely the programme as outlined above.

It seems superfluous to give here a survey of the older lite-
rature on the metabolism of the coloured sulphur bacteria,
since a very thorough and critical discussion of it has recently
been given by van Niel in his publication mentioned above.

It will suffice to give here a brief review of the publications
which have appeared since van Niel's paper was published.
For the rest I will restrict myself to a short survey of the pre-
sent state of our knowledge regarding the points at issue in
the introductory remarks to each of the various chapters.

In the first place mention may be made of the book pu-
blished by Ellis in 1932 on the sulphur bacteria. However,
new points of view regarding the metabolism of the coloured
sulphur bacteria have not been opened in this monograph.
Neither is this the case in the publications by Ginsburg-
Karagitschewa (1932) on methods of cultivating Thiorho-

daceae and by von Deines (1933) on the nature of the sul-
phur stored in these organisms.

The important question of the behaviour of the Thiorhoda-
ceae in organic media, which was only more or less inciden-
tally struck by van Niel, has been more fully investigated by
Muller (I9331). The principal result of this study is undoub-
tedly that although these bacteria can thrive on various of
the most commonly accepted carbon sources, they obviously
are unable to utilize these compounds directly in the synthesis
of their cell substances. The Thiorhodaceae — and probably
also the Athiorhodaceae, when growing anaerobically in the
light in organic media — have the remarkable property of
utilizing the organic nutritional compounds only indirectly,
viz. as hydrogen donators in the photochemical reduction of
carbon dioxide.

Gaffron (1933, 1935) has published several investigations
on the metabolism of the Athiorhodaceae. This author too has
realized the great advantages of the application of the mano-
metric method for the study of this subject. Though one
should interpret the results of Gaffron with great reserve,
since he did not use pure cultures, but worked with impure
enrichment cultures, yet on the whole the results obtained
give support to the idea, that the metabolism of the Athio-
rhodaceae and that of the Thiorhodaceae in an organic me-
dium under anaerobic conditions in the light are fundamen-
tally the same. It must be emphasized, however, that in an-
other communication (1934) in which the metabolism of the
Thiorhodaceae is studied, Gaffron himself rejects this idea.
In this paper a quite new conception of the metabolism of
these bacteria has been given. Although my own investiga-
tion was already in full swing at the time when this paper
was published, I felt it necessary to give due attention to this
question. I have therefore repeated his experiments and sub-
jected them to a critical examination (c.f. pag. 49).

Finally important work has been published of late on the
chemical nature of the pigments of the purple bacteria.
Van Niel (1933) has published some preliminary results of
a study of the chemical composition of bacterio-erythrine,
the red pigment of the Rhodobacteriales. Van Niel's results

leave no doubt that this pigment belongs to the Carotinoids.
Noack and Schneider (1934), Schneider (1934) and
Fischer and Hasenkamp (1935) have made extensive stu-
dies of the chemistry of the green pigment of the purple bac-
teria. As was already assumed by other investigators, this
pigment appears to be closely related to chlorophyll and in
all probability the same will hold for bacterioviridine, the
pigment of the green sulphur bacteria.

In a preliminary communication (1934) some of the results
of my own work have already been published.

The majority of old and recent investigators of the physio-
logy of autotrophic organisms, with which I mean bacteria
as well as green plants, has probably underrated the dangers,
caused by the use of material, in which the organism to be
investigated is not in a pure condition.

Now I will not pretend that the results of the study of the
physiology of the green plants have been greatly impaired
by this. Nevertheless I think that a simple reference to the
publications of van Niel (1931) about the photosynthetic
autotrophic Thiorhodaceae and to that of Kengma Boltjes
(1934) about the chemosynthetic autotrophic nitrifying bacte-
ria will suffice to demonstrate the necessity of the purity of
the material employed in a physiological study of autotrophic
organisms also.

While an apparently wholly inorganic medium may be still
an excellent substrate for heterotrophic specialists like Bac.
oligocarbophilus and Hyphomicrobium vulgare, such a me-
dium allows a good development of even very common hete-
rotrophic bacteria, as soon as autotrophic organisms are thri-
ving or have thrived in it.

Many investigators in cultivating algae have been greatly
encumbered by the presence of bacteria in these cultures. So
one can easily imagine that in those cases where no pure, but
so-called quot;sound algaequot; were used for metabolic experiments,
these bacteria have contributed markedly to the respiration
observed. Yet definite statements regarding this point are
usually lacking.

A recent example of the confusion, which may arise from
neglecting the requirement of purity of the cell material em-

ployed, is to be found in the investigation of Gaffron (1934)
previously mentioned. I shall revert to this later on.

At the outset of my investigations the strains 1, 4, 7, 9, 12,
19 a b, and c employed by F. M. Muller were still available.
However, these strains had all been contaminated since, but
by applying the method described by van Niel (1931), it was
an easy task to purify them once more. The contamination
was so slight, that I could make use of shake cultures in pep-
tone agar (peptone quot;Poulencquot; 1%, NaCl 2%, 2% agar in tap-
water) thus profiting by the more rapid growth in organic
media, as already mentioned by van Niel (1931, p. 102).

The greater majority of my experiments was earned out,
however, with a strain (d), which was isolated by me from an
enrichment culture sent by H. Gafkron to the Microbiologi-
cal Laboratory, Delft. This bacterium, it is true, was present
in the enrichment culture as the prevailing organism, but for
its isolation it still proved to be necessary to employ the in-
organic medium, as described by van Niel (1931, p. 22). A
short description of this strain follows here.

Being obligately anaerobic and obligately photosynthetic,
it belongs to the group of the Thiorhodaceae. Under the con-
ditions of cultivation as employed by me, in casu in an inor-
ganic or organic medium with a pH of about 8,0 and a low
NagS concentration, mainly highly motile diplococcus forms
may be observed, though micrococcus forms are not rare.

Neither conglomerates of three or more cells nor the for-
mation of capsules or the production of mucus were ever ob-
served The cells have an average diameter of 1,5 In some
cases, specially in the organic media, they may be larger, up
to 4 M Under certain conditions the cells may contain one or
more highly refractive sulphur globules. The bacterium re-
minds one of the organisms described by van Niel as the

It proved impossible to bring this strain into a known genus
with the aid of one of the proposed systems of the Thiorho-
daceae. Considering the exceedingly unsatisfactory state of

The pure cultures obtained were kept as stab cultures in
BuRRi-tubes with organic agar media. The tubes were pro-
vided with a second plug of absorbent cotton-wool, with pot-
assium pyrogallate, and were rubber-stoppered. It appeared,
that the modification of the burri-method as indicated by
Ritter and Dorner (1932),viz. a substitution of the caustic
soda by sodium carbonate, offered no advantage in this case.
Thus placed in artificial light the cultures are easily kept
alive for one or two months.

The strains a, b, c, 9 and 19 were kept in the mentioned
peptone agar medium. The other ones, which did not grow
so well in this medium, were kept in a Na-malate medium.
This was prepared by adding 0,1% Na-malate, 0,1% NagSgOg
and the usual 2% agar to the standard salt-solution, i.e. NaCl
2%, (NH4)2S04 0,1%, K2HPO4 0,05%, MgS04 0,02%, in tap-
water.

Before the inoculation 0,005% NagS was added and the pH
was adjusted to 8,0 for strain d, to 8,5 for the strains i and 12
and to 8,7 for the strains 1 and 7. This was carried out by
adding sterile 10% solutions of H3PO4 or of NasCOg, using
the indicators cresol red and brom thymol blue. Of course
these media may also be used without agar; however, in using
the liquid medium in culture tubes, a weekly transfer is ne-
cessary. I have abandoned the use of glass-stoppered bottles
for keeping the pure cultures, because of the danger of conta-
minations entering between neck and stopper. The mentioned
substrates are to be preferred to inorganic media in that they
are more easily prepared, since NaHCOs, which requires a se-
parate sterilization by filtration, may be left out. Hence the
risk of contamination is smaller, whereas the growth is more
abundant. As a carbon source Na-malate was chosen by me,
since in consequence of its high quot;oxidation valuequot; 1,2 mol.
carbon dioxide is produced per mol. Na-malate taken up by
the bacteria (Muller 1933i, p. 150). As a result the pH of the
medium remains constant during a much longer time, which

Experience has taught me, that the addition of thiosulphate
has a favourable influence. Probably hereby the carbon
dioxide, originating from the Na-malate, is reduced again un-
der simultaneous formation of sulphuric acid and hence the
malic acid can be more fully converted into cell material.

The purity of the strains was controlled with each transfer
by inoculating peptone agar plates and by incubating these
during several days aerobically and anaerobically at a tem-
perature of 30° C. For the anaerobic cultures I used the con-
venient quot;anaerobic jarsquot; as devised by Mc Intosh and Fil-
des. These jars are in general use for the cultivation of anae-
robes in the Delft laboratory.

It was not only of paramount importance to find a strain
specially suitable for my experiments, but also to offer that
strain such favourable culture conditions, that 1 could daily
have a proper quantity of young and vigorous bacteria at
my disposal. The requirements for the material to be used
were fairly high. In the first place all cells should be in op-
timal conditions, specially for the experiments mentioned in
Chapter V. Cultures, in which a part of the bacteria would
have sunk to the bottom already or would have produced
mucus and formed cell conglomerates, were to be excluded
a priori and it was essential to select a strain which would
not, or only in a very slight degree, show these properties in
the culture medium employed.

In the second place the bacteria should have to thrive well
in a culture medium of a pH preferably below 8,0 (group 2
of van Niel 1931, p. 46). The reason for this is, that the pH
of the suspension liquid, which is to be used in the experi-
ments, may not differ much from that of the medium in which
the bacteria have been grown.

If one wants to measure changes in the carbon dioxide pres-
sure manometrically, the primary requirement is to use a
suspension liquid, which is at the most slightly alkaline. At
a pH gt; 8,0 one is too much encumbered by the so-called

quot;carbon dioxide retentionquot; (see p. 33). So the strains 1, k, 7
and 12, which were adapted to a pH of 8,5—8,7, had to be re-
jected for the majority of my experiments.

It appeared that of the remaining strains, strain d best com-
plied with the demands, at least when only the behaviour in
inorganic culture media was considered. Although the majo-
rity of the experiments were carried out with this strain d,
I have gained so much experience with the strains a, b, c, 9
and 19, that it seems justified to conclude that these strains
behave fundamentally in a completely similar way. As a mat-
ter of fact none but quantitative differences have appeared.

In order to obtain the material needed for the experiments,
bottles of 600 cm3 were inoculated from pure cultures. Like
the pure culture tubes the glass-stoppered bottles were incu-
bated in a light cabinet especially made for this purpose
(van Niel 1931). The temperature measured in the culture
bottles could be varied between 30o and 36° C. They were pla-
ced inside at a distance of 25 cm. from a 60—75 Watt electric
bulb. In some preliminary growth experiments, carried out
with strain a in peptone broth I could easily ascertain a diij-
tinct temperature optimum at 31 °C. 1 made these experiments
with a quot;temperature organquot; fundamentally similar to the ap-
paratus described by Ruinen (1933). It was placed before a
window facing the east.

In the light cabinet, however, my strains thrived best at
35° C. This is probably due to the greater quantity of light,
which is available there. Under these conditions the cultures
in suitable organic media attained their maximal development
within a week, whereas those in inorganic media reached it
in the second week.

The majority of the experiments was carried out with ma-
terial originating from an inorganic medium. In case other
media have been used, this has always been mentioned. The
inorganic medium, which finally proved to be the best for
niy purpose, was prepared by adding to the mentioned stan-
dard salt solution 0,1—0,15% Na2S203 and by just boiling up,
cooling and filtering it, before sterilization. In this way the
formation of a precipitate in sterilizing was prevented. Be-
fore using this medium it was freed from oxygen by boiling

it up, followed by a rapid cooling and by adding 0,005% NagS.
After adding 0,7—1% NaHCOs and adjusting the pH to 7,8
in the manner previously mentioned, the culture bottles
were filled almost to their necks and inoculated from culture
tubes with a liquid Na-malate medium, which served as a
link between the stab cultures and the cultures in the glass-
stoppered bottles. In order to exclude the air as much as pos-
sible, the bottles were as yet provided with a layer of pure
sterilized (120°C.) paraffine oil. In dry sterilization (200°C.)
of bottles with paraffine rests, substances may be formed,
which appear to be harmful to the bacteria; hence the used
bottles should be very thoroughly cleaned.

The NaHCOg was added by means of the apparatus as
shown in fig. 1, by adding a suitable quantity of a 7% solu-
tion of NaHCOs to the standard solution. The 7% NaHCOg

solution was sterilized with the aid of a Seitz-filter, which
previously had been thoroughly coated with tin. Preceding
this, carbon dioxide, sterilized by passing a plug of cotton-
wool, was led through the solution in the apparatus until the
pH was about 7,8. This was done because in the long run a
NaHCOg solution exposed to the air passes into a NagCOs so-
lution. As moreover a NaHCOg solution with a pH of 7,8 con-
tains much carbonic acid, the bacteria have more CO2 at their
disposal.

The concentrations of the carbon dioxide available, of the
NaaSaOs and of the NaaS were chosen in such a way that
the bacteria only temporarily contained sulphur. Just previous
to their maximal development, the cultures lost in most cases
the calcareous appearance and turned a dark red colour. On
microscopic examination it then appeared that nearly all cells
had lost the characteristic highly refractive sulphur globules.
This production and subsequent consumption of sulphur is
advantageous for three reasons.

In the first place it is conducive to an efficient absorption
by the bacterial pigments of the light used (see pag. 29),
which is of special importance for the experiments to be men-
tioned in Chapter V.

Secondly under these conditions it is excluded, that the
bacteria use their reserve sulphur for the carbon dioxide as-
similation which is important in those cases, where the assi-
milation with the aid of hydrogen donators added on purpose,

Finally the pH in the cultures is kept constant on account
of the formation of sulphuric acid in a quantity equivalent to
that of the alkali liberated from the NaHCOg. The advan-
tages thereof are that the crop per bottle is larger and that
the bacteria are accustomed to a pH which is suitable for the
use in the manometric method (see pag. 33). A bottle of 600
cm3 produced about 450 mms centrifuged bacteria with a dry
weight of about 120 mgr.

By using 0,2% NagSOg instead of 0,1% NaaSgOs, I tried to
prevent the formation of reserve sulphur. The growth in this
medium, however, was highly unsatisfactory.

sary to know the absorption coëfficients of nitrogen and car-
bon dioxide in the suspension liquid. Since these coëfficients
are unknown for a 2% salt solution, I have tried to adapt
strain d to a concentration of 0,5% NaCl. The growth herein,
however, was never as satisfactory as in a medium with 2%
NaCl and consequently I have made very little use of this
medium.

Furthermore it may be stated that an addition of f.i. Na-
malate to a quantity of 0,01% exerts a distinctly accelerating
influence on the growth of the bacteria and I availed myself
of this expedient, either when I wanted material within a
short time, or when I had to make transfers from old cultures.

The use of bacteria cultivated in wholly organic media,
involved a number of peculiar difficulties (see page 54). Con-
sidering that by the sterilizing and by the repeated boiling
of the media with (NH4) 2SO4 as a nitrogen source, a varying
and therefore unknown quantity of NH3 got lost, I investi-
gated whether Na-nitrate or urea were suitable to replace
the (NH4)2S04. For this purpose I made with these substan-
ces a so-called N-auxanogram according to Beijemnck (1889)
using as a substrate the previously mentioned malate-thiosul-
phate-medium without (NH4)2S04 in leached agar. The Petri
dishes were provided with potassium pyrogallate (see f. i.
Koch 1934) and after having been sealed, incubated in the
light cabinet. Strain d showed a pronounced preference to
(NH4)2S04, even when the agar was inoculated with bacteria
from a culture in which they had urea as the only nitrogen
and carbon source.

In those experiments, where I have restricted myself to
the study of the metabolism of the bacteria in the dark, I
have mostly used the single constant-volume type of mano-
meter, generally known as the warburg-type. All vessels
were provided wdth one or two side-bulbs. Useful indications
regarding details in the technique, the calibration of the appa-
ratus, the calculation of the constants, the shaking apparatus,
etc. are given by Dixon (1934) and by Krebs (1929).

All experiments, where besides the metabolism in the dark
also that in the light was studied, were carried out with two
differential manometers placed side by side. They were quite
similar to those which Warburg (1923) employed for his
experiments on the COa-assimilation of Chlorella. For further
technical data on these manometers and for indications con-
cerning their calibration I refer to the original description of
Warburg and Negelein (1923), to Warburg (1926) and to Gaf-
fron (1929).

The special requirements of the p.s.b. with respect to their
environment made it necessary to deviate in certain respects
from the technical procedures as applied by Warburg. Thus
I worked at different temperatures, with other suspension li-
quids and gases, etc. For some experiments with the differen-
tial manometers I used vessels which differed from the nor-
mal ones, in so far that they were provided with a side-bulb.
The two differential manometers which stood at my disposal
enabled me to make control or duplicate experiments. A
double kathetometer such as Warburg employed for the rea-
ding of his manometers, did not seem to me to be essential;
I did the reading with the naked eye. The menisci were illu-
minated by quot;Everreadiesquot;, which were fed by a bell trans-

former. For the experiments described in Chapter V, I used
silvered Warburg-vcsscIs. The silver coating was deposited
on the outside and in order to protect it against the harmful
effect of the water in the bath, it was not coppered as War-
burg recommends, but simply coated with an aluminium
varnish.

With those experiments, where the use of monochromatic
light was not required, the two WARBURG-vessels containing
the bacterial suspensions and the two compensation vessels
filled with suspension liquid were illuminated by a half sil-
vered 100 Watt Philips incandescent lamp, length of the fi-
lament 25 cm; distance from the vessels 5 cm. This lamp,
made watertight in the simple way shown in fig. 2, was placed
in the water bath in such a manner as to allow the two ves-
sels with bacteria to be illuminated with practically equal
intensity. This was controlled by a series of assimilation mea-
surements.

As the light of this lamp, when directly connected with 220
Volt cv), led to solarisation phenomena (Ursprung 1917), I
used a 75 Watt electric bulb as a resistance. The intensity of
the visible light of the 100 Watt lamp is then very low and a
rough comparison of it with the intensities of yellow, green

and blue light, which would lead to as high an assimi-
lation velocity (see Ch. IV) as caused by this lamp, makes
it probable that the assimilation observed was mainly due to
infra-red light. As a matter of fact a CUSO4 filter (6% aque-
ous solution of 1 cm thickness) diminished the assimilation
velocity to 10% of its original value.

The intensity of this light source was never consciously va-
ried, nor controlled or levelled. Fluctuations in the voltage in-
deed caused perceptible, though for these experiments no
troublesome, changes in the assimilation velocity.

Small changes in pressure occurred in the manometers im-
mediately after switching on and off the light. This is caused
by the fact that the suspension with bacteria absorbs more
light and hence becomes warmer (0,03—0,06°C.) than the sus-
pension liquid in the control vessel. These small fluctuations
were of no importance for my experiments, since they did
not last longer than a few minutes.

For the experiments with monochromatic light, described in
Chapter V, I used a quartz mercury lamp burning on 110 Volt
== or a 50 Watt Philora sodium lamp from Phflips, which
burnt on 220 Volt csd with the use of a suitable transformer
and a choke-coil. The light intensity of the mercury lamp was
not constant, but since I only used it in experiments in which
I did not make high demands upon the absolute value of the
results, I made no special arrangements for the control of
this intensity.

The sodium lamp, however, burnt very constantly without
any special arrangement, at least after the first hour had pas-
sed by. Fluctuations of 2%, inherent to the sodium lamp un-
der these conditions, were within the experimental error. I
isolated the yellow (578 m/^), the green (546 m/t) and the
blue (436 m/x) emission lines from the mercury light by
means of a CUSO4 filter and coloured glasses from Schott amp;
Gen., Jena, placed in a cuvette filled with water. Hereby the
indications of Nuernberghk (1933) were followed. There
further data can be found regarding the amount of the opti-
cal impurities still present in the filtered light. The sodium
light was applied without purification. According to an infor-
mation, kindly given by Prof. G. Holst of the Physical Labo-

ratory of the PniLiPS-works, the impurities present in the so-
dium light of this type of lamp amount to about 5% of the
total light intensity.

The assimilation effected by the impurities in the light of
the mercury lamp and of the sodium lamp was determined
with the aid of special dye solutions or coloured glasses which
absorbed the wave length selected for the experiments, but
not the impurities in question. Hereby I paid special atten-
tion to select such control filters through which the infra-red
light also passed.

In order to control the green light (546 m/x) I used a 1 cm
thick layer of 0,02% erythrosine in water. This absorbs the
green light almost wholly, but allows the red and infra-red
light to pass. The influence of impurities of the blue light I
could determine by means of the coloured glass OG2 of
ScHOTT which wholly absorbs the blue. Finally for the con-
trol of the yellow light (578 m^i and 589 m^i) I employed the
coloured glass BG 11 of Schott, which almost wholly absorbs
the yellow light and allows all the other wave lengths to pass.
These control filters were placed into the water bath in order
to prevent the loss of light by reflection. In this manner
I could determine the correction term which had to be intro-
duced for assimilation on false light.

With certain experiments it was necessary to diminish the
light intensity by a known amount. This was carried out with
blackened wire cloth screens with varying mesh and thickness
of wire. The brass wire cloth obtained from the „Metaaldraad-
weverij Dinxperloo, Hollandquot; was coppered and blackened,
strictly keeping to the prescription in Schürer-Waldheim
(1921, p. 286). By means of the spectral-pyrometer, still to be
described, the extinction coëfficients of these screens were de-
termined. As is known the absorption of the light by these
filters is independent of the wave length of the light.

As is shown in fig. 3, a homogenic beam of light was cast
on the bottom of a vessel with bacteria by means of the len-
ses Li and Lg. For this purpose the water bath was provided
with a window of plate glass, while a mirror (S2) placed un-
der the vessel with bacteria threw the light upwards. In order
to protect this mirror against deterioration, I had cemented

a glass plate against the silvered side with so-called quot;de Kho-
tinsky cementquot;i).

All determinations of the intensity of the monochromatic
light were carried out in absolute units (erg/cm2/sec.) with
the aid of a spectral-pyrometer, a somewhat simplified form
of the apparatus described by Eymers, Ornstein and Vermeu-
len (1932). I refer to this publication for the construction,

A cheaper substitute may be obtained by melting 3 parts of brown
shellac with 1—2 parts of Venetian turpentine.

manipulation and standardization of this apparatus. It was
constructed and standardized in the Physical Institute of the
University of Utrecht. I am very much indebted to Professor
L. S. Ornstein and his staff for their kind, manifold assistan-
ce and valuable advice. The apparatus having remained un-
employed for a long while at the time of the experiments de-
scribed in Chapter V, it seemed most advisable to have it com-
pared again with an absolutely standardized thermopile. This
was done in the Laboratory for Technical Physics at Delft with
satisfactory results. I am also greatly indebted to Professor
H. B. Dorgelo and his collaborators for their very kind as-
sistance in this matter.

For the experiments described in Chapter V it was neces-
sary to determine the quantity of light which was absorbed
by the bacterial suspension. This was carried out by the
method as used by Warburg (1923), viz. by determining
how much light fell on the bottom of the vessel filled with the
bacteria and how much of it was reflected again. The diffe-
rence is the light absorbed by the suspension, since the part
which is transmitted is very small and may be neglected pro-
vided dense bacterial suspensions are employed.

For the determination of the intensity of the light falling
on the bacteria I proceeded in the following way. I measured
by means of the spectral-pyrometer the intensity of the light
falling on the surface of a glass plate (MgQi, see fig. 3) which
had been quot;sootedquot; with magnesium oxide. This was placed
at an equal distance from a mirror Si as the vessel with the
bacteria. The intensity of the light on a second magnesium
oxide surface (Mg02) had previously been compared with
that of the light falling on a third magnesia surface located
in the air above the spot where normally the vessel with the
bacteria was placed. Thereby another calculated correction
was made for the extra reflection on the water-air boundary
introduced. In this manner once for all the correction terms
were determined which had to be introduced for the determi-
nation of the light intensity and from this moment on they we-
re all carried out on the plate MgQi. It may be stated by the
way, that the water in the thermostat was kept clear by re-
freshing it daily. It is evident that the mirror Si was inter-

posed only when the intensity of the light was determined.
This was carried out before and after each illumination pe-
riod and as I found the light intensity of the sodium lamp to
be very constant, it seemed allowable to calculate from these
data the quantity of light, which during the experiments had
fallen on the bacteria.

The part of the light reflected by the bacteria was deter-
mined with the spectral-pyrometer by comparing the intensi-
ty of the light, reflected by a magnesium oxide surface, with
that of the light reflected by a vessel with bacteria kept in the
same place (see Warburg i.e.). It is assumed thereby that
the magnesium oxide reflects all the light and furthermore
that this ideal white surface and the bacterial suspension have
an analogical type of reflection distribution. These assump-
tions will strictly taken not have been realized, but I may
safely say that they have been closely approximated. When
the light falls perpendicularly on the vessel with bacteria, the
light reflected by the suspension of course must be observed
at an angle with the perpendicular. In doing so one is not
troubled by the light reflected by the surface of the glass bot-
tom of the vessel. The size of this angle had, within reasona-
ble limits, no perceptible influence. Thus after each experi-
ment, the reflection of the bacteria in the warburg-vessel
was determined and also for this a correction was introduced.
An aqueous suspension of 15 mms sulphur-free bacteria per
cm3 reflects about 3% of the sodium light thrown on it, whe-
reas bacteria containing sulphur, under similar conditions,
can show a diffuse reflection of even 20% of the light.

The whole of the optical equipment and the water bath
were placed in a dark room improvised by means of heavy
curtains.

By the way I wish to draw the attention to the great possi-
bilities of application of the spectral-pyrometer in biology.
Particularly in studying the reflection and dispersion of the
light thrown on leaves (Seybold 1933) and other objects it
undoubtedly is a most useful instrument. The great simplicity
as well as the relatively great reliability of the once standard-
ized instrument make it, even in the hands of non-physicists,
an extremely convenient apparatus.

The thermostat consisted of a zinc cistern of 30X30X72 cm
filled with water and provided with a piece of plate glass,
inserted in one wall. The water was properly stirred and could
be kept at any temperature between 25—35° C. at 0,01 °C. con-
stant. The heating was done by means of ten heating tubes
connected in parallel, an arrangement which has been in use
for years in the Delft laboratory. The tubes which were U-
shaped were each of them 1.40 M. long (of which 1 M. sub-
merged) and had a resistance of 100 a. They consisted of a
copper tube, through which an asbestos-isolated commercial
resistance wire was drawn. The heating tubes were connected
with 50 Volt alternate current in order to exclude danger in
case of short-circuit. The thermoregulator was constructed ac-
cording to Wüst (1930). The electric current for this regula-
tor was obtained with a Philips rectifier 1016/1017.

The advantages of this uniform heating of the whole ther-
mostat and the accurate regulation of it by the mentioned
thermoregulator specially were of great profit in the experi-
ments with the single manometers. The simple way in which
the thermoregulator could be adjusted to special temperatu-
res was very convenient.

The gases or gas mixtures used in the warburg-vessels
greatly varied in different experiments. Except for the spe-
cial cases in which oxygen or hydrogen mixtures were em-
ployed, it was necessary to deprive them most rigorously of
all traces of hydrogen or oxygen. The Thiorhodaceae namely
are anaerobic and as will appear further on they also involve
hydrogen in their metabolism. For the said purpose all gases
always were led through red hot copper wool or wire cloth
(partly converted into copper oxide), put in a Pyrex glass
tube (compare fig. 4). A good device for the electrical heating
of the copper can be found in a publication of Kendall (1931).
The purified gas passed through a CaCla-tube and then was
led either to a washing-bottle with a suspension liquid which

was to be saturated with the gas, or to the WARBURG-vessels,
which had to be filled with it. In these arrangements copper
or glass tubes were employed, the number of taps and pieces
of rubber tubes being limited to a minimum. Where copper
and glass tubes had to be connected I used quot;de Khotinsky ce-
mentquot;, which is gas-tight. Data regarding the permeability of
various substances for oxygen are to be found with Hill
(1928).

In the course of the experiment nitrogen and hydrogen were
used and also mixtures of N2/CO2, N2/H2, N2/CO, N2/CH4, Ng/
H2/CO2 and H2/CO2. I had nitrogen, hydrogen andN2/5% CO2
in steel cylinders at my disposal. The other mixtures were
prepared either in two large calibrated communicating bott-
les partly filled with paraffinum liquidum, or by allowing
nitrogen or hydrogen to pass through a glass tube containing
glass beads, through which a NaHC03/Na2C03 mixture trick-
led (see van den Honert 1930). The pressure of carbon
dioxide in the atmosphere in equilibrium with these mixtures,
can be calculated with the formula given by Warburg (1919,
p. 230). Carbon monoxide and methane were prepared accor-
ding to the usual methods (c.f. Treadwell 1930).

The liquid used for the preparation of the suspensions of
the bacteria in the Warburg-vcsscIs of course had to be free
from all traces of oxygen. For that purpose, the water to be
used, after the addition of 2% NaCl, was boiled out in a strong
Pyrex flask and while still steaming the flask was closed with
a rubber stopper in order to prevent the entrance of air and
at once cooled under the water-tap. If needed inorganic and
organic salts were added and now the solution was brought
into equilibrium with the gas or gas mixture to be employed
in the experiment. This was done by bringing it into a washing
bottle (see fig. 4) and by allowing the gas, very finely divided
by means of a piece of rattan, to bubble through it during
30—60 min.

at the disposal of the bacteria, but wished to observe their
behaviour with respect to another gas, I used CO2 buffer
mixtures, which are solutions of Na-carbonate and Na-bicar-
bonate in a certain ratio. The corresponding buffered carbon
dioxide pressure can be calculated as indicated above. Both
calculation and experience show that the buffering capacity
of these solutions is considerable. The pH, it is true, is fairly
high, but the bacteria employed did not seem to be harmed
thereby.

In the cases where I wanted to test the behaviour of the
bacteria towards carbon dioxide itself I generally took nitro-
gen with 5% carbon dioxide as a gas phase and a suspension
liquid with about 0,5% Na-bicarbonate, which was previously
brought in equilibrium with the gas mixture. The addition
of Na-bicarbonate had to adjust the pH of the suspension
liquid to the pH to which the bacteria were adapted. The lat-
ter was determined colorimetrically before each experiment
on the clear culture medium after removing the bacteria by
centrifuging. The quantity of Na-bicarbonate, to be added to
the suspension liquid in order to reach the same pH, could
be calculated. I had namely previously determined which
quantity of Na-bicarbonate had to be added to an aqueous so-
lution of 2% NaCl being in equilibrium with a given nitrogen
carbon dioxide mixture, in order to obtain a definite pH. With
two of these data one is able to calculate the third (see Dixon
1934, p. 78).

These precautions as to the adjustment of the pH are essen-
tial, since the p.s.b. are especially very sensitive to a lower
pH. Bacteria, originating from a medium with a pH 8,0 show
a distinctly reduced assimilation when the pH of the suspen-
sion liquid is lower than 7,6. in most experiments I employed
a suspension liquid with a pH of 7,8.

In the beginning I have indeed doubted whether in using
such alkaline solutions, one would not be troubled by the re-
tention of the carbon dioxide and perhaps with reason be-
cause even in a solution of 0,5% Na-bicarbonate in equili-
brium with 5% carbon dioxide, Na-carbonate will still be pre-
sent. A calculation of this retention according to the principles
given by Warburg ^,1926, p. 110) shows, however, that it is to

The pH remained fairly well constant during the experi-
ment as only slight acid production took place and because
a bicarbonate-carbonic acid mixture is a pH buffer owing to
the property of compensating acid or alkali production by
driving out or taking up carbon dioxide.

The slight acid production by the bacteria in the dark, as
mentioned on page 41, consequently led to a carbon dioxide
formation not entirely to be neglected. When more than rela-
tive value is to be attached to the observed changes in the car-
bon dioxide pressure a production or consumption of acid has
to be taken into account. As is known (see f.i. Dixon 1934) one
cannot buffer the pH with other buffer mixtures without in-
creasing the carbon dioxide retention considerably, at least
at this pH.

Finally I still have to mention that the constants of the
different vessels have been calculated with the well-known
absorption coëfficients of gases in water. However, by the
addition of 2% NaCl the solubility of the gases will undoub-
tedly decrease. The absorption coëfficients for this solution
were not available. Those of carbon dioxide in 15% NaCl and
in RiNGER-solution at 38° C. are known. If one approximately
calculates the absorption coëfficiënt of carbon dioxide in 2%
NaCl from these data, then it appears that in consequence
of it the constants of the vessels are somewhat reduced. With
the differential manometers this reduction is about 2%, in
the other cases it is to be neglected.

After having described how the gas mixtures were prepa-
red and purified and how the suspension liquid was handled,
I will now deal with the treatment the bacteria had to un-
dergo before the experiment was started. In many cases it was
advisable for reasons to be mentioned later on, to put the
culture bottles with bacteria into an incubator of 35° C. in
the dark the day before. Then the content of the bottles was
centrifuged. In the beginning I was afraid of poisoning the
bacteria by the oxygen from the air and I therefore excluded
the air from the liquid in the centrifuge tubes by means of
paraffine oil. As appeared afterwards this was, however, su-
perfluous. The centrifuged bacteria were then suspended in

tlie liquid mentioned before and centrifuged again. Not un-
til the bacteria were washed out in this manner, were they
suspended again and put into the WARBUBG-vessels.

The gas or gas mixture to be used, was blown through the
manometer vessels before and after the bacterial suspension
was put into them. The manometers were then placed into the
thermostat and they were shaken for some five minutes. The
taps were just opened from time to time in order to allow
the gas (which had expanded with the rise of the tempera-
ture) to escape. Then all joints were worked in and made to
bind fairly firmly, and the shaking apparatus was set going
anew. After some ten minutes equilibrium was obtained and
now the experiments could be started.

It may be remarked, that in consequence of all these ope-
rations with non sterile glassware the bacterial suspensions
were no doubt contaminated. However, the short duration of
the experiments excludes any possibility of a perceptible in-
fluence of this contamination.

The quantity of bacteria in the vessels varied considerably.
A quantity of 34 cmS suspension solution with 50—450 mm»
bacteria came into the vessels of the differential manometers,
which had a capacity of about 55 cmS. For the vessels of the
single manometers with a capacity of 7—20 cm^, 1—4 cm^
bacterial suspension were used. The quantity of the bacteria
present was determined after the end of the experiment, by
centrifuging an aliquote part of the suspension in calibrated
so-called tromsoorff-tubes, during 10 minutes with 3000
revolutions per min. It appeared that working in this way 100
mm3 of bacteria had a dry-weight of about 27 mgr.

A question, arising immediately while studying the physio-
logy of the photosynthetic autotrophic organisms, is how they
keep alive during the nocturnal periods of darkness. Muller

and they certainly do not die or pass into a resting-stage over-
night. Neither can we expect that they will be able to survive
a period of several hours without some process taking place
inside their cells, which maintains the potential differences
in their physico-chemical structure.quot;

This question, which at that time could not yet be solved
on the basis of the available experimental data, is closely re-
lated to another question, viz. whether the Thiorhodaceae are
able to grow in the dark and, if so, under which conditions.

Van Niel (1931, p. 105) in a careful survey of the literatu-
re concerning this question, mentions two instances where co-
loured sulphur bacteria were reported to have grown in the
dark. Nadson should have been successful in cultivating his
green sulphur bacteria and Miyoshi his Thiorhodaceae in the

However, none of the later investigators of coloured sulphur
bacteria have succeeded in corroborating these observations.
Now it is possible to explain negative results in all those cases,
where one did not use pure cultures, by accepting an over-
growing of the sulphur bacteria by more rapidly growing he-
terotrophic organisms. This seems the more plausible, since
for growth experiments in the dark of course only organic
media can be applied.

pure cultures have been used. Van Niel (1931, p. 106) and
Muller (1933i, p. 140) have inoculated a number of pure cul-
tures of Thiorhodaceae in yeast water, peptone water and syn-
thetic glucose, fructose, lactate and pyruvate media. Even af-
ter a prolonged incubation at 25°C in the dark they could
not observe any growth, either under anaerobic, or under
aerobic conditions. However, they always found growth when
these culture bottles were exposed to the light.

This behaviour is in contrast with that of the Athiorho-
daceae, pure cultures of which grow abundantly in such me-
dia in the dark, as Molisch (1907) and van Niel (van Niel
and Muller 1931, p. 259) observed. The latter investigator
made the important observation, that contrary to what holds
for the growth in the light, the growth in the dark only can
take place in the presence of oxygen. The strictly anaerobic
character of the Thiorhodaceae should therefore be the rea-
son, why these organisms are fully unable to grow in the
dark.

How to conceive — in the light of the foregoing conside-
rations — the process or the processes, which maintain the
life of the Thiorhodaceae and Athiorhodaceae in periods of
darkness? As obviously neither the Thiorhodaceae, nor the
Athiorhodaceae are able to grow in the dark in media,
which are very suitable for the cultivation of bacteria in ge-
neral, van Niel (1931, p. 106) rejected for both groups the
idea of a fermentation process being the base of their me-
tabolism in periods of darkness. Since these organisms in na-
ture usually are found in a more or less complete anaerobic
environment, the idea of an oxidative dissimilation has to

remains an open questionquot;. On the other hand Müller (1933i,
p. 164) states in this connection: quot;However I think that it is
quite possible, that an organism can effect some special fer-
mentation process, without being able to synthesize its cell
material from the fermentation substrate, without the aid of
an external source of energy. Such a special fermentation
process might be the conversion of sugars into lactic acid. It

is quite conceivable that the p. s. b. are able to quot;maintainquot;
themselves on a lactic acid fermentation in the dark, but are
able to develop only in the light, because they need radiant
energy for some of the dehydrogenations, involved in the syn-
thesis of cell materialquot;.

In my opinion this conception is fully justified, since one
can find analogous cases in another group of photosynthetic
autotrophic organisms, namely among the green plants. Al-
though a great number of algae can grow in the dark on a
medium containing for instance glucose, there are also simply
organized algae, which in pure culture do not tolerate organic
matter at all, e.g. the blue algae. (Beijerinck 1898, 1901).
Now nobody will venture to conclude only from this, that the-
se blue algae are unable to maintain themselves for some time
on a dissimilatory conversion of their reserve food compounds.
As 9. m3.tt6r of fâct it will be shown tli3.t the suppositions of
Muller very nearly approximated the truth. He apparently did
not intend to say, that, if the Thiorhodaceae have a sufficient
quantity of a suitable substrate at their disposal, for instance
a carbohydrate, they should be able to maintain themselves
in the dark for an unlimited time. One should not forget that
also essential cell compounds will be consumed (maybe by
autolysis) and that a defective rebuilding of these substances
in the dark ultimately can cause the death of the bacteria.
This idea is supported by the experience that bacteria in an
organic medium cannot withstand longer periods of darkness
than a culture of the same organisms in an inorganic medium.
The contrary is even more likely to be true.

Summarizing we may say that on ground of negative or
positive results of some growth experiments some ideas re-
garding the metabolism of the Thiorhodaceae in the dark had
been given. However, a direct experimental proof of the cor-
rectness of these ideas was still lacking. Therefore it seemed
worth while to investigate this question.

mon inorganic medium, mentioned on p. 19, were centrifu-
ged, washed, and suspended in an oxygen-free solution of so-
dium chloride, either provided or not with sodium bicarbo-
nate, and then brought into the WARBUHG-vessels in an at-
mosphere of nitrogen or in a nitrogen carbon dioxide mixture,
in the dark an increase of pressure was always observed. Mo-
reover this increase in pressure proved to be constant for
many hours. All strains investigated by me showed the same
behaviour in this respect.

Evidently this increased pressure was the result of a pro-
duction of carbon dioxide, since it failed to turn up, when
alkali was put in a side-bulb of the manometer vessel, or when
the bacteria were suspended in one of the carbon dioxide buf-
fer mixtures mentioned previously.

Curve I. Carbon dioxide produc-
tion by bacteria of strain d, culti-
vated in the inorganic medium, sus-
pended in tap-water with 2% NaCl
and 0,5% NaHCOs, in equilibrium
with N2/7% CO2 at a temperature of
30°C.

Curve II. The same arrangement
except the presence of alkali in a
side-bulb of the manometer vessel.

Fig. 5 gives a graph of the course of such an experiment,
while in table 1 the data relating to a few experiments are
presented. The figures indicate the production of carbon
dioxide in mms gas of 0° and 760 mm pressure per hour by
a certain quantity of bacteria. These are the averages of mea-
surements, which usually were continued for some hours.

Bacteria of strain d, cultivated in the inorganic me-
dium, suspended in distilled water with 2% NaCl,
in equilibrium with N2, temperature 33 °C.

170 mm3 bacteria of strain d, cultivated in the or-
ganic medium, quot;starvedquot; during 20 hours at 36°C.,
suspended in:

a.nbsp;tap-water, 2% NaCl, with a 0,05 molar CO2
buffer mixture of 9 parts NaHCOs and 1 part

b.nbsp;tap-water, 2% NaCl, 0,4% NaHCOs, in equili-
brium with N2/5% CO2. (manometer I)

Conformable to experiment no. 3, except tempera-
ture of starvation period, which was SO'C., mano-
meter 1 with CO2 buffer mixture.
Manometer 3 with NaHCOs-solution.

1)nbsp;Brodie solution is the manometric fluid used and has the following composition: 23 gr. NaCl, 5 gr.
Na-choleinate, in 500 cm» water, with a trace of thymol. 10.000 mm Brodie is equivalent to 760 mm Hg.

2)nbsp;Differential manometer is abbreviated in all tables as D.M.
Single manometer (constant volume type) is abbreviated as S.M.

At the end of all experiments the bacteria were examinéd
under the microscope in order to verify whether they still
were in a good condition. As a rule the result of this exami-
nation was satisfactory, but sometimes the bacteria had ob-
viously come into contact with the caustic soda in the side-
bulb of the manometer vessel. Of course such experiments
were discarded.

Because the production of carbon dioxide, after having been
constant for some hours, shows a slight decrease rather than
an increase, it may be concluded that this production of car-
bon dioxide is a normal physiological process (quot;autofermenta-
tionquot;). Under these conditions it is tempting to identify this
process with the already mentioned fermentation process, sug-
gested by Muller (19331).

Further I have considered the question, whether the car-
bon dioxide evolved, was indeed produced by the bacteria, or
whether it was not, at least partly, due to the Na-bicarbonate
in the solution, from which it could have been expelled by
acid, formed in the autofermentation process. In the first
place acid production should have been reflected in a higher
increase in pressure due to carbon dioxide production by
bacteria suspended in a solution containing bicarbonate, as
compared with the carbon dioxide liberation by an equal
quantity of the same bacteria suspended in a solution free
from bicarbonate or carbonate. Furthermore one can also de-
tect a possible acid production by adding an excess of acid
to one suspension before and to a second (similar) suspension
after the experiment and determining the quantity of the car-
bon dioxide liberated in both cases. The results of the expe-
riments concerning this question are to be found in table 2.

From this it may be inferred, that as a matter of fact, a
development of acid takes place corresponding to about 30%
of the totally liberated carbon dioxide.

As a rule this acid production has not been taken into ac-
count, since in most experiments only relative values matte-
red. The assimilation velocity was always determined by
measuring the difference between the changes in pressure in
the dark and in the light and therefore the error caused by
evolution of carbon dioxide as a result of acid production.

Equal quantities of bacteria of strain
d, cultivated in the inorganic medium,
suspended in tap-water 2% NaCl, 0,5%
NaHCOs, in equilibrium with N2/5% CO2
at a temperature of 32°C., put in diffe-
rent manometers.

b.nbsp;tap-water 2% NaCl, 0,5% NaHCOs,
in equilibrium with N2/5% CO2. Ma-
nometer no. 3.

As will be expounded on page 93 this assumption is not qui-
te correct, because it appears, that these acids are assimilated
again in the light, at least to some extent. Such an assimi-
lation of acids will of course lead to a carbon dioxide uptake
in surplus to that due to the carbon dioxide consumption of
the cells.

In this and all following tables the data regarding carbon dioxi-
de production indicate the sum of the carbon dioxide produced
by the bacteria and the carbon dioxide driven out of the suspen-
sion liquid by acid formation.

However, a determination of the quantity of acid formed
or assimilated in the experiments with the differential mano-
meters presented insurmountable difficulties, since the quan-
tity of bicarbonate in the medium was too large to permit its
measurement in the manometers. On the other hand it was
impossible to lower the quantity of bicarbonate because the
pH and the concentration of carbon dioxide in the medium
were dependent on the bicarbonate concentration.

b.nbsp;The influence of a preceding period of darkness on the
rate of the production of carbon dioxide was in agreement
with the conception, that the phenomena described are due
to an autofermentation. In table 3 a number of experiments
is reported concerning the velocity of the autofermeutation
of bacteria from cultures, which had passed different periods
of darkness. In the table the figures relating to the carbon
dioxide production indicate the sum of the carbon dioxide
driven out of the solution by acid and of the carbon dioxide
produced directly by the bacteria.

From these data it can be concluded that in consequence of
a preceding darkness of 24 hours at 36° C., the autofermenta-
tion has fallen to 30% of its original value. The vitality of the
bacteria, judged by their motility had not decreased thereby
and I found in addition that also the assimilatory capacities
were only slightly diminished.

c.nbsp;The influence of the temperature on the velocity of the
autofermentation was very marked. As may be seen from
table 4, the value of the Qio was high. In this connection it
must be pointed out, that the temperatures used were not
higher than the temperature of the light cabinet in which the
bacteria were cultivated. Hence a harmful influence of these
temperatures is out of question.

d.nbsp;I was unable to make direct experiments about the in-
fluence of light on the autofermentation, since the assimila-
tion cannot be excluded.

Similar difficulties have been encountered by the investiga-
tors, who have studied the influence of light upon respiration
of green plants. In this connection I restrict myself to drawing
the attention to the reliable experiments of van der Paauw
(1932, p. 522). This investigator observed an increase of the

Bacteria of strain d, cultiva-
ted in the inorganic medium,
starved during 20 h. at 35°C.
suspended in tap-water 2%
NaCl, 0,5% NaHCOs, in equi-
librium with N2/5% CO2 at
various temperatures.

respiration of algae, under the influence of exposure to the
light from an ordinary electric bulb. He could prevent a strong
assimilation in this light, by absorbing all carbon dioxide by
alkali. The respiration was measured by the absorption of
oxygen.

Strictly speaking it remains doubtful whether the respi-
ration under conditions favouring a strong assimilation, i.e.
in the simultaneous presence of light and sufficient carbon
dioxide, is influenced in the same sense and to the same extent
by the exposure to light. Moreover it is not at all excluded
tliat the occurrence of the assimilation process in itself exerts
an influence on the respiration velocity (Müller 19331) 1).

For these reasons one certainly has to consider the possibi-
lity of a change in the autofermentation of the Thiorhodaceae
as a consequence of an exposure of these organisms to light.
Since we are dealing here with fermentation which cannot be
measured independently from the assimilation — as in the

1) In this connection it may be mentioned, that the Aihiorhodaceae are
the only photosynthetic organisms where the influence of Ught and of
the carbon dioxide assimilation on the respiration can be studied with-
out meeting complications caused by the liberation of oxygen in the
assimilation process.

case of respiration by oxygen uptake no direct experi-
ments to decide this question could be made. Yet in none of
my experiments concerning the assimilation of carbon dioxi-
de, I could observe any after-effect of exposure to light upon
the rate of autofermentation. The velocity of the carbon di-
oxide production attained its original value almost immedia-
tely after the end of the light periods. This is in contrast with
the experiences of van der Paauw (1932) with algae.

Hence it seems allowable to conclude that under the condi-
tions prevailing in my experiments the light did not exert
any influence upon the autofermentation of the bacteria and
that thus this process proceeded with unchanged velocity du-
ring the assimilation. At best the autofermentation will in-
crease, be it very slightly, in consequence of the small in-
crease in temperature caused by the illumination (vide p. 25).
e. The question arises which compound can be the substrate
of the autofermentation. The only observations which seem
appropriate to give a clue to the solution of this problem are
those of H. G. Derx, quoted by van Niel (1931, p. 106), who
found indications for the formation in the light of a reserve
product, which could be coloured violet by iodine. However,
in spite of repeated attempts, I was unable to corroborate the-
se observations.

Experiments made in order to investigate the possibility
of influencing the rate of the fermentation by adding organic
substances of known constitution, e.g. glucose, Na-lactate, Na-
butyrate and Na-malate yielded quite negative results (see
tables 5 and 6). Hence I could not get any indication regarding
the nature of the reserve substance by this indirect way either.
It is, however, not excluded that one will be able to detect an
influence of certain added substances upon the fermentation,
by using bacteria, which were previously exposed to a star-
vation period. It remains possible that the bacteria used in
my experiments did not react on these additions because they
had an excess of stored substrates for the autofermentation
process at their disposal.

In the meantime it is in my opinion not justified to con-
clude, either from a positive or from a negative result of the
addition of glucose upon the autofermentation, whether the

Experiments on the influence of organic substances and sulphates on the autofermentation.

reserve product which is fermented, is a carbohydrate or not.
For not rarely it happens, that one observes acceleration of the
respiration or of the fermentation of an organism, by adding
substances, which certainly cannot be substrates of these pro-
cesses. I cite in this connection only the examples of ethylene
(Mack and Livingstone 1933), cyanide in certain cases,
methylene blue, etc. Even if one observes a fermentation of
the added substance, it would mean only, that the Thiorho-
daceae are able to ferment that substance, but not that their
reserve substances are identical with it. On the other hand
it is possible not to find any influence of a certain substance,
which yet very probably is the substrate of the autofermen-
tation. So Watanabe (1933) observed no influence of the
addition of 1% glucose upon the respiration of Ulva, although
with other algae he could confirm the positive results found
by other investigators. In all probability the Ulva thallus used
has been rich in reserve products, but such considerations
have evidently not been taken into account by Watanabe.
f. In connection with the phenomenon of the autofermenta-
tion I must also draw attention to the publication of Gaffron
(1934). According to him the Thiorhodaceae should be able
to oxidize in the dark several organic substances with sulpha-
te as a hydrogen acceptor, which is then reduced to sulphide.
At least this is the explanation he offers for the increase in
carbon dioxide production, observed by him after addition of
glucose, Na-malate etc. together with sodium sulphate to the
medium. Gaffron even goes so far as to conclude that the pre-
sence of sulphates is indispensable for the use of organic
substances.

These conclusions seemed of so far-reaching importance
that I decided to repeat these experiments. I used in this case
both strain d — the same organism as used by Gaffron —
cultivated in the ordinary inorganic culture medium, and
strain 9, cultivated in the peptone medium. As the graphs in
fig. 6 and 7, and the experiments in tables 5 and 6, show un-
mistakably, the result of these experiments was fully negative.
Hence I have to reject the explanation Gaffron gave for the

Curve I. Carbon dioxide
production by bacteria of
strain a, cultivated in the
inorganic medium, suspen-
ded in distilled water with
2% NaCl pro analyse and
0,5% NaHCOs pro analyse, in
equilibrium with N2/7% CO2
at a temperature of 30 °C.
No S04-ions present.

Curve II. The same arran-
gement except the presence
of 0,2% Na2S04 and 0,3%
Na-malate in the suspension
liquid.

richment culture, from which I isolated strain d and as mo-
reover the method of cultivation was pretty well the same, it
seems impossible to explain the difference in results by a dif-
ference in the properties of the bacteria used. A more plausible
explanation 1 see in the fact, that Gaffron did not use pure
cultures. As mentioned before, heterotrophic organisms are
very well able to grow in an inorganic medium, provided that
autotrophic organisms are present. In this case one is tempted
to suppose, that the enrichment culture of Gaffron has been
infected with the sulphate reducing anaerobic Vibrio desul-
furicans and with Bacterium coli, which as a rule obsti-
nately accompanies the former organism (vide Baars, 1930).
The only guarantee Gaffron gives that in this case actually
the physiology of the Thiorhodaceae was studied, is to be
found in the following passage (Gaffron I.e. p. 449): „Da

Curve I. Carbon dioxide production by bacteria of strain d, cultivated
in the inorganic medium, suspended in distilled water with 2% NaCl pro
analyse, 0,5% NaHCOs pro analyse and 0,2% glucose pro analyse, in
equilibrium with N2/7% CO2 at a temperature of 30°C. No S04-ions
present.

in der genannten Nährlösung keine heterotrophen Organis-
men wachsen können, gelangt man mit etwas Sorgfalt rasch
zu Reinkulturen. Man überzeugt sich davon mit Hilfe des Mi-
kroskops, am besten unter Zusatz von Methylenblau, das nicht
nur die Bakterien gut färbt, sondern auch die lebhafte Bewe-
gung sistiertquot;. Even for a trained microbiologist such a cri-
terion is insufficient, especially as far as the very small Vibrio
desulfuricans is concerned.

A second, less important point of difference between the
results of Gaffron and my own concerns the very conside-
rable formation of acid, found by him. From Gaffron's data
it appears that in most cases. 70% of the formed carbon di-
oxide is expelled from the bicarbonate in the suspension by
the production of acids. I never could observe such a strong
acid production and probably the large production in Gaf-
fron's experiments is also due to the use of contaminated cul-
tures.

Finally I have to mention here a number of growth expe-
riments I made to test in another way the possibility of sul-
phate reduction by Thiorhodaceae. If this supposition would
be correct, one would expect that like the sulphate reducing
organisms, these bacteria should be able to live in the dark
on this fermentative dissimilation process. In order to test
this and also Gaffron's assumption that the Thiorhodaceae
should be unable to use organic substances in their metabo-
lism in the absence of sulphates the following experiments
were made. Strain d was inoculated in bottles of 60 cmS, filled
with the media mentioned in table 7, whilst these bottles were
incubated partly in the dark and partly in the light. In the
standard solution I had substituted in this case the sulphates
by chlorides. All organic and inorganic salts were pure and
had especially been tested on the absence of sulphates. In
the cultures, where no thiosulphate or sulphate had been
added, the only source of sulphur the bacteria had at their
disposal, was a little Na2S. This may have been partly oxi-
dized to sulphate.

The results of these experiments leave, however, no doubt
that the Thiorhodaceae are unable to grow with a sulphate
reduction as a base, either in the light, or in the dark. (Cf.

Dependence of growth of strain d on the presence of sulphate
and of organic substances with different quot;oxidation valuequot;
in the dark and in the light.

experiments 1, 2 and 3, table 7.). Moreover these experiments
prove that they are able to grow in the light with organic
compounds in media free from sulphates or bicarbonate, pro-
vided that the quot;oxidation-valuequot; of the compounds in question
is high enough (C.f. table 7 no. 5, 6 and 7 and further Muller
19331, p. 161.)

In my opinion the foregoing experimental results warrant
the conclusion that the above mentioned statements of
Gaffron regarding the metabolic activities of the Thiorho-
daceae are untenable.

Since I had found that the rate of growth and the yield
of bacteria in organic media were markedly higher than in
inorganic media I decided to test the suitability of the bac-
teria grown in the former media for my experiments.

Bacteria of one of the strains previously mentioned, were
cultivated in the peptone culture medium (see pag. 16), cen-
trifuged, washed and suspended in the ordinary inorganic
aqueous salt solution. If these suspensions were put in the
WARBURG-manometers with a nitrogen atmosphere I obser-
ved an increase in pressure, which calculated per cm3 cell
material, only amounted to about 10—20% of the increase in
pressure caused by bacteria of the same strain, when cultiva-
ted in the inorganic medium.

It appeared further that these quot;peptone bacteriaquot; on mi-
croscopic examination always showed big and quaint forms.
Besides, they showed a much reduced motility as compared
with bacteria of the same strain from inorganic media. Mo-
reover this motility very soon decreased and disappeared
completely after a relatively short stay in the manometer ves-
sels. Hence it must be concluded that these bacteria were ab-
normal in a morphological as well as in a physiological sense.

The reduced autofermentation — and as will be shown later
on also the low assimilatory activities — is obviously caused
by a reduced vitality in general, apparently due to the cul-
tivation of the bacteria in the organic medium.

Although the peptone medium differs considerably from the
medium preferred by the Thiorhodaceae in nature, such a
reduced vitality was not expected, since the growth in the
peptone medium is as a rule much more abundant than in
the inorganic culture media.

How remarkable this may be, still more astonishing was
the observation that the increase in pressure in the dark as a
consequence of the autofermentation, was not quantitatively
due to carbon dioxide production. It appeared namely that
in the presence of alkali in a side-bulb of the manometer
vessel, there still remained a measurable increase in pressure.

(fig. 8 and tables 8, 9 and 10). This can by no means be attri-
buted to HaS, since of course this gas would have also been
absorbed by the alkali. The idea of the unknown gas being
oxygen had also to be rejected, as it was not absorbed by pot-
assium pyrogallate.

Carbon dioxide and hydrogen production by bacteria from a peptone
culture medium.

Curve I. Changes in pressure effected by bacteria of strain a, cultiva-
ted in the peptone medium, suspended in distilled water with 2% NaCl,
in equilibrium with N2 at a temperature of 30 °C. CO2 production amoun-
ting to 20 mm® per hour when the H2 production is assumed to be
15 mm®/hour.

Curve II. The same arrangement as in I, except the presence of Na-
pyrogallate with excess alkali in a side-bulb of the manometer vessel.
Hydrogen production amounting to 15 mm® per hour.

Curve III. The same arrangement as in II, except the presence of Pd-
black in a second side-bulb of the manometer vessel. After two hours
the Pd is apparently poisoned by some metabolic product of the bacteria.

Experiments demonstrating the nature of the gas produced besides carbon dioxide by

However, when besides alkali, palladium black was put in
another side-bulb of the manometer vessel, no change of pres-
sure was detectable as can be seen from fig. 8 and in table 8.
Apparently the unknown gas was adsorbed by the palladium.
It must be remarked that in spite of the presence of Pd
in experiments like those represented by curve III in fig.
8, there always could be detected a slight increase in pres-
sure, when the experiments were continued too long. Pro-
bably in these cases the Pd was poisoned by volatile products
of the bacterial metabolism. By adding a little H2S, the palla-
dium was poisoned immediately and did not adsorb the un-
known gas anymore, (table 8 no. 3 and 4). Now it is a well-
known fact that palladium is a strong adsorbent for gaseous
hydrogen, but it also adsorbs considerable quantities of car-
bon monoxide and unsaturated hydrocarbons. However, these
last-mentioned gases are also absorbed by a solution of cupro
oxide in ammonia (Treadwell 1930), which was not the case
with the unknown gas. (table 8 no. 5.)

On the base of all this it seems safe to conclude that the
unknown gas in question was hydrogen. A further confirma-
tion of this conclusion is to be found in experiments like
those mentioned on page 71 and in fig. 10 and 11. There it
is demonstrated that the gas in question was assimilated again
in the light, which as is shown in Chapter IV, § 5, is in accor-
dance with the behaviour of the bacteria towards hydrogen.

Furthermore it was remarkable, that the ratio of hydrogen
and carbon dioxide produced varied a good deal, when dif-
ferent strains were tested. When using different cultures of
the same strain this ratio varied also, although to a far smaller
degree. The hydrogen production amounted to 8—30% of the
total gas production (table 9.). I got the impression, that the
larger and the stranger the bacteria were and the sooner their
motility in the warburg-vessels fell off, the more hydrogen
they produced. The minute bacteria of strain 9 ordinarily pro-
duced H2, amounting to 8 vol. % of the total gas production;
the large bacteria of strain a, however, gave off 30 vol. % H2.

The quite different behaviour of bacteria of a strain cul-
tivated in the peptone culture medium, as compared with bac-
teria of the same strain from the inorganic medium, as well

Ratio of carbon dioxide and hydrogen produced by bacteria from the peptone medium.

Bacteria of strain d, cultivated in the
peptone medium, suspended in tap-water
2% NaCl, in equilibrium with Nz, tem-
perature 30 °C.

Conformable to experiment no. 1.
Alkali present in side-bulb of manometer
no. IV and VI. Bacteria of strain 9.

Bacteria of strain c, cultivated in the
peptone medium, suspended in own cul-
ture medium, in equilibrium with Na at
a temperature of 30°C.
Alkali present in manometer no V.
Bacteria of strain a, cultivated in the
peptone medium, suspended in:

a.nbsp;tap-water 2% NaCl, 0,3% NaHCOs
in equilibrium with Nil5% CO2. Ma-
nometer no. 3.

b.nbsp;tap-water 2% NaCl, 0,05 molar CO2
buffer mixture of 9 parts NaHCOs
and 1 part NazCOs. Manometer no.1.

as the differences in the ratio CO2/H2 in the total gas produc-
tion of the different, in other aspects similar strains, also
speak for the idea that the quot;peptone bacteriaquot; must be consi-
dered as abnormal. In my opinion, the hydrogen production
must be considered from the same point of view; probably it
is connected with autolytic phenomena.

That the bacteria from the peptone culture medium must
be more liable to autolysis than bacteria from the inorganic
media, is clear, when their abnormal morphological habit and
their reduced vitality is taken into consideration. The cause
of the autolysis is not to be looked for in the transference of
the peptone bacteria into the synthetic solution, since they also
produced gaseous hydrogen after being suspended in their
own culture medium.

In my opinion the explanation must be sought either in the
vigorous shaking of the warburg-manometers, or in the
denseness of the bacterial suspension. For I never could de-
tect any hydrogen production in culture bottles, even after
their exposure to darkness during several days at a tempera-
ture of 35°C. In agreement with the above ideas I always
could detect ammonia in the suspension liquid after experi-
ments with bacteria from peptone cultures, this being in con-
trast to my experience with bacteria cultivated in an inorga-
nic medium.

However, not all organic culture media exerted such an
influence on my bacteria. Bacteria of the strains a, b or 9, cul-
tivated in a Na-malate-Na-thiosulphate medium for instance,
did not produce hydrogen in my experiments. Their micro-
scopic aspect was quite normal too. The influence of long pe-
riods of darkness on the autofermentation of bacteria from
peptone culture media, was much greater in comparison with
the effect of this treatment on the behaviour of bacteria from
inorganic media. After an exposure to a period of darkness
of twelve hours at a temperature of 35° C., most of the bac-
teria did not move anymore and the total gas production, as
Well as the assimilatory activity of the bacteria was as a rule
reduced very considerably.

Special experiments showed that the gas production of the
bacteria from the peptone medium could not be influenced

by the addition of organic compounds, (table 6.). In this re-
gard they behave like those from inorganic media.

Nor could I detect any increase of the gas production on
addition of Na-butyrate to suspensions of bacteria cultivated
in a Na-butyrate-Na-bicarbonate culture medium, as can be
concluded from the data in experiment No. 2 of table 5.

Anticipating the results mentioned in § 4 of Chapter IV, I
wish to state that in all probability the gaseous hydrogen, pro-
duced by the bacteria, is partly assimilated again. In this as-
similation process, the hydrogen serves as a hydrogen dona-
tor for the reduction of carbon dioxide. The change in pres-
sure detected in the manometers evidently results from
the difference between the rate of hydrogen production and
the rate of hydrogen assimilation. However, the uptake of
hydrogen will be very small in consequence of the very low
pressure of this gas in the experiments described above. This
may be concluded from the experiments mentioned on page
82, where it is shown, that even the presence of 1 % hydrogen
in the gas phase of the manometer vessels only had a relative-
ly small depressing effect on the total increase in pressure
per minute.

Finally I wish to draw the attention to the remark of
Gaffron in his publication on the metabolism of the Athio-
rhodaceae (1933, p. 7) „Irgend welche Stoffwechselprodukte
werden von den Purpurbakterien, wenigstens nach meinen
bisherigen Befunden, nicht ausgeschieden.quot;i) This statement
was not amended in a further publication of Gaffron on
this subject (1935). In the light of the results of my experi-
ments on the Thiorhodaceae, it seems rather improbable and
it would be advisable to reinvestigate the Athiorhodaceae in
this respect.2)

1) His remark, that Muller (1933 i) observed the same thing with the
Thiorhodaceae does not hold, since in this statement the production of
carbon dioxide in Mulleh's experiments is neglected.

In a private letter Mr. C. S. French of the California Institute of
Technology informed me that he observed a COa-production in the dark
with the Athiorhodaceae: Spirillum rub rum.

Although I did not intend to study the problem of possible
relations between the Thiorhodaceae and oxygen, 1 could
not avoid making some observations on this question. Dr. C. B.
van Niel had informed me in a private letter that he had ob-
served an oxygen consumption by suspensions of Thiorhoda-
ceae and this observation was always confirmed in my expe-
riments. In consequence in fermentation and assimilation ex-
periments the suspension liquid and the atmosphere in the
manometer vessels must always be thoroughly freed from
traces of oxygen. Especially the oxygen uptake of bacteria,
which had been cultivated in inorganic media, was very vi-
gorous.

I did not investigate the phenomenon of the oxygen con-
sumption in details supposing that this question was studied
by van Niel. So I do not know, whether the uptake either
is constant, decreases after some time, or even stops at all.
However, I have to draw the attention to the fact, that the
Thiorhodaceae are catalase-positive (van Niel and Muller
1931) and that I got the impression, that these bacteria are
not so strictly anaerobic as generally is supposed. As a matter
of fact I found, that strain a, inoculated in culture tubes,
nearly completely filled up with previously boiled out and
rapidly cooled peptone broth with 0,005% NagS, always sho-
wed abundant development, although the air had free access
to the media.

In addition one can use shake cultures in peptone agar for
the isolation of these bacteria, even without shutting them off
from the air by means of a rubber stopper, paraffine etc. Only
the upper part of the agar column did not contain any colonies
under these conditions.

Furthermore peptone agar plates with colonies of Thiorho-
daceae which had developed during an anaerobic incubation
in the light, could be exposed to the air without any harmful
influence on the bacteria, at least in so far as a microscopic
examination allows one to judge. Even after having been in
contact with the air during several days, all bacteria in such
colonies still showed the normal motility, provided the plates

I must mention, however, that since streak cultures on aero-
bic agar plates always yielded negative results in the light as
well as in the dark, it is obvious that the Thiorhodaceae are
not able to grow with an oxidative dissimilation as a base.
They even seem to die when exposed to the full oxygen tension
of the air. The facts mentioned above can be understood on the
assumption that the reducing capacity of a large number of
bacteria already present is able to stand the harmful effect
of the air, in contrast with the reducing capacity of one iso-
lated bacterium.

In view of this, I cannot recommend for the present to in-
terpret this oxygen consumption by suspensions of Thiorho-
daceae as a consequence of a normal respiration process.
Still it is not impossible that a certain restricted supply of
oxygen stimulates the metabolism and the development of
these bacteria. Even if this is the case, one should distinguish
between a favourable influence, caused by a direct participa-
tion of the oxygen in the metabolism and a possibly favou-
rable influence caused by a limited oxidation of the medium
surrounding the cells, which medium will be reduced under
the influence of the bacterial metabolism. Only when the first
alternative holds true, one should speak of respiration.

1.nbsp;The Thiorhodaceae in the dark as well as in the light,
maintain themselves with a fermentative dissimilation of
reserve food material, which process is accompanied by
a production of carbon dioxide and acids.

2.nbsp;The rate of this autofermentation highly depends on the
temperature and on a previous exposure to a period of
darkness.

3.nbsp;Under the conditions of my experiments in all probability
the illumination does not exert any influence on the auto-
fermentation.

4.nbsp;Neither the addition of any of the organic substances men-
tioned nor that of the inorganic substances tried, affect the
autofermentation.

5.nbsp;The Thiorhodaceae produce hydrogen in the dark under
certain conditions, provided they are cultivated in a pep-
tone medium.

6.nbsp;Oxygen is rapidly consumed by suspensions of these bac-
teria, but it remains open whether this process should be
termed respiration or not.

A few years ago only deficient and rather confused opinions
existed on the question, how Thiorhodaceae assimilate carbon
dioxide in a medium where they only have carbon dioxide
and oxidizable sulphur compounds at their disposal besides
the other necessary inorganic salts. In fact there was not yet
given a satisfactory explanation, why in such a medium the
presence of carbon dioxide and oxidizable sulphur compounds
as well as an adequate supply of light energy were essential
for the development of these organisms.

It was generally accepted that the p. s. b. were able to assi-
milate carbon dioxide photosynthetically in the same way as
is done by the green plants. The negative or at best very
doubtful results of all attempts to detect oxygen formation
by the p. s. b. were, however, a serious obstacle for this con-
ception. On the other hand it was suggested that the carbon
dioxide could also be assimilated chemosyntbetically at the
cost of the energy liberated in the oxidation of hydrogen sul-
phide to sulphuric acid. According to Buder (1919) the latter
process would occur with the aid of the oxygen produced in
the photochemical process. However, in this line of thought
the problem remained unsolved, why the Thiorhodaceae re-
quire both the presence of HgS and light simultaneously, i.
o.w., why they cannot reduce carbon dioxide either with the
energy of the HgS oxidation (colourless sulphur bacteria!) or
with light energy (green plants!) alone.

It was van Niel (1931), who, applying the general consi-
derations of Kluyver and Donker (1926, 1930) regarding
the unity in the biochemical activities of all living cells on the

problem in question, offered a fully satisfactory explanation.
His experimental results can be best interpreted on the base
of his theory, that the water used by the green plants as a
hydrogen donator in the hydrogénation of the carbon dioxide
cannot be used as such by the Thiorhodaceae, but that these
organisms in inorganic media require oxidizable sulphur com-
pounds for this purposei).

quot;It is shown that the metabolism of the purple and green
sulphur bacteria is a truly photosynthetic process, which
can best be considered as one of a number of possible
photosynthetic reactions of the general type: CO2 2 H2A =
CH2O H2O 2 A.quot;

Hence the Thiorhodaceae are able to build up the carbon
compounds necessary for their cell material, using one or more
of the following photosynthetic reactions as a base:
CO2 2 H2S = CH2O H2O 2 S
CO2 2 S(H20) = CH2O H2O 2 SO
C02 2S0(H20) = CH2O H2O 2 SO2
CO2 2S02(H20) = CH2O H2O 2SO3

Although as early as 1907 Molisch concluded: „dass die
Ernahrungsversuche mit Purpurbakterien uns mit einer neuen
Art von Photosynthèse bekannt gemacht haben, bei der orga-

van Niel (1931, p. 100), who suggested that such an assimi-
lation of organic substance has to be considered as an assi-
milation (reduction) of carbon dioxide, the organic com-
pounds acting solely as hydrogen donators.

Muller (19331) investigated the metabolism of the Thio-
rhodaceae in a number of organic media. The results of his
experiments were fully in favour of the conception given
and we may therefore conclude that in the equation: CO2

1) Of course one can also stick to the idea, that the assimilation
process itself is still accompanied by an oxygen production, but that
the presence of the oxidizable sulphur compounds in the culture me-
dium is indispensable for the removal of the oxygen, which would
exert a harmful influence on the bacteria. In my opinion there are no
arguments which speak for this less simple theory.

2 HaA = CH2O H2O 2 A. HgA and A may also represent
organic substances, both for the carbon dioxide assimilation
of the Thiorhodaceae and in all probability for that of the
Athiorhodaceae.

It seemed to be of great importance to investigate in how
far these ideas about the carbon dioxide assimilation of the
Thiorhodaceae, which were mainly based on the results of
analyses of outgrown cultures, could be confirmed and ex-
tended with the aid of a quite different technique. In the first
place it seemed desirable to study the velocity of the carbon
dioxide assimilation under various conditions.

As mentioned already in Chapter II, I used for all experi-
ments on assimilation the two differential manometers. Ex-
cept for the experiments in Chapter V always an incandes-
cent lamp was used as a light source. For further technical
details compare Chapter II.

When suspending Thiorhodaceae, cultivated in an inorganic
medium, in an isotonic solution of sodium chloride and 0,5%
Na-bicarbonate in previously boiled and rapidly cooled water,
and on bringing this suspension in manometer vessels with
an atmosphere of for instance pure nitrogen with 5% carbon
dioxide, I always could observe a decrease of pressure on il-
luminating the vessels. This is of course exactly the contrary
of what happens in the dark (compare Chapter III.)

This phenomenon can only be explained by assuming that
either nitrogen or carbon dioxide, or both, are assimilated in
the light. When I did not suspend the bacteria in a mixture
of bicarbonate and carbonic acid, but in a carbon dioxide
buffer mixture of Na-bicarbonate and Na-carbonate, I could
not observe any change in pressure either in the light, or
in the dark, although the bacteria remained quite healthy in
this solution, at least as far as may be judged from a micro-
scopic examination. Hence the conclusion seems justified that
the bacteria are able to assimilate carbon dioxide in the light,

although no hydrogen donators had been added to the me-
dium. When asking which substance served as a hydrogen
donator under the conditions of the experiment, we must
conclude that it cannot be water, since then an equivalent
amount of oxygen would have been produced. But in this
case no decrease of pressure would have been observed. Un-
der these circumstances the bacteria in the suspensions had
no other hydrogen atoms at their disposal than either those
from organic reserve food compounds possibly present in the
cells (1) or those from oxidizable inorganic reserve food com-
pounds (2) or those from oxidizable metabolic products (3).

From experiments like those represented graphically by
fig. 9 one must conclude that under the conditions mentioned,
the bacteria mainly used as hydrogen donators the organic
products of their metabolism. During these experiments the
same vessel with bacteria was exposed to intermittent pe-

riods of light and darkness. The arrows in the figure in-
dicate the moments when the light was switched on and
off. A closer examination of the graph learns that at the be-
ginning the uptake of carbon dioxide in the light is rapid
and slower later on. Evidently this phenomenon returns after
each period of darkness and apparently the assimilation ve-
locity remains high during a longer period, if the previous
period of darkness is longer. This behaviour is of course in-
compatible with the idea that the reserve food compounds
themselves are responsible for the assimilation observed (at
zero time in fig. 9 the experiment was already some time going
on.). The slower uptake of carbon dioxide (in fig. 9 at 0 b. 45,
2 b. 15 and 5 b.) was still considerable, viz. nearly as high as
the carbon dioxide production in the autofermentation. As has
been made plausible on page 44 we must accept namely that
the autofermentation goes on unchanged during the periods of
illumination!).

One might suggest, that the phenomenon described is a
symptom of quot;assimilatory inhibitionquot; (Ewart 1898), quot;Ermü-
dungquot; (Pantanelli 1904) or quot;Solarisationquot; (Ursprung 1917).
That this was out of question, and that the falling off of the
assimilation velocity was indeed caused solely by a lack of a
suitable hydrogen donator, may be concluded from the fact,
that if HgS or Na2S03 were added previously, the carbon
dioxide assimilation proceeded at a high rate during very long
periods. The only possible explanation is therefore that after a
period of darkness the bacteria consume the hydrogen dona-
tors, produced during that period.

In agreement with this conception was the fact, that the
velocity of the uptake of carbon dioxide in the absence of
added hydrogen donators — which I will call auto-assimila-
tion — decreased considerably when the bacteria had been
exposed to a period of darkness of some twenty hours at
35°C., before the experiment started^). Still the capacity of

1)nbsp;Numerous experiments like the one described above were perfor-
med, always with similar results. However, I will refrain from pu-
blishing more of them, since they do not lend themselves for a simple
tabular representation.

2)nbsp;The fermentation products formed during that period were of
course removed by the preceding centrifuging of the cultures.

assimilating carbon dioxide after addition of hydrogen dona-
tors was but little influenced by this treatment. Since it see-
med probable that such quot;starvedquot; bacteria would offer spe-
cial advantages for the study of the suitability of different
hydrogen donators, such bacteria have been used in most of
the following experiments.

The slow auto-assimilation which on continued illumination
follows the rapid one, will in all probability proceed at least
partly with the same hydrogen donators as those which are
responsible for the rapid assimilation. However, in the light
these donators do not accumulate as they do in the dark. It
seemed possible, that part of this rest-assimilation must be
ascribed to sulphur stored in the cells. As a matter of fact I
was much surprised to find that the part played by this stored
sulphur apparently was very small. I never could observe any
striking difference in the behaviour of bacteria of the same
strain containing or not containing sulphur. One must con-
clude therefore, that the sulphur present in the Thiorhodaceae
as a reserve food compound can only slowly be utilized.

From the experiment to be mentioned on page 77 it will
appear in addition that the oxidation of hydrogen sulphide
proceeds rapidly up to the level of sulphur, but that this stage
once being reached the assimilation velocity falls down con-
siderably. Evidently the hydratation and subsequent dehy-
drogenation of the sulphur hydrate is a slow process, limi-
ting the rate of assimilation under the conditions of the ex-
periments.

In raising the question which is the nature of the com-
pounds which accumulate during the periods of darkness and
which compounds can serve as hydrogen donators for the
auto-assimilation, it seems plausible to think of oxidizable
products of a carbohydrate fermentation.

Now on page 41 it was shown that during the autofer-
mentation acids were produced. The experiments mentioned
in table 14 however, showed that bacteria cultivated in the
ordinary inorganic medium could but use Na-malate, Na-lac-
tate and Na-butyrate as a hydrogen donator to a slight extent.
Hence it is improbable that the hydrogen donators, produced
during the autofermentation, would be organic acids, since the

auto-assimilation can proceed very rapidly. Only a small part
of the auto-assimilation can possibly be ascribed to these sub-
stances.

It must be possible in principle, to demonstrate a consump-
tion of acids as a result of the assimilation, since this will be
reflected during the assimilation in an increase of the carbon
dioxide present in the suspension as bicarbonate or as car-
bonic acid. It would be feasible to detect this in a duplicate
experiment by adding an excess of acid in one vessel before
and in the other after the assimilation, in the same way as
was done in the experiments mentioned sub 1 and 2 in ta-
ble 2.

However, in the dehydrogenation of the acids other acids
will be formed first. Only if the successive dehydrogenations
lead to the production of pyruvic acid, it is probable, that this
acid will be involved in the assimilatory metabolism of the
cells and hence will disappear from the medium (Muller
19331, p. 153).

Considering all this, it seems allowable to assume that the
assimilation of the acids produced by the bacteria, whenever
it would take place, will only be very slight. Hence it was
accepted that the uptake of carbon dioxide as calculated from
the difference between the changes of pressure per unit of
time during periods of light and of darkness, was quantita-
tively due to a carbon dioxide assimilation and not partly
to a conversion of carbonate into bicarbonate. This is an es-
sential point in the interpretation of the results of those expe-
riments in which an absolute value was attached to the
amount of carbon dioxide taken up and where no correction
for auto-assimilation could be applied.

Since the autofermentation of the Thiorhodaceae might be
a combined lactic acid and alcoholic fermentation, I have also
tried whether ethyl alcohol could be used as a hydrogen dona-
tor. Under the conditions of my experiments this compound is
unsuitable as such, as can be judged from the figures in table
14 no. 4. Acetaldehyde, however, can be used very well for
that purpose (table 14, no. 3) ; it may be expected that under
these conditions it will be oxidized to acetic acid. If acetalde-
hyde or some other aldehyde would be responsible for the

greater part of the auto-assimilation, in particular for the ra-
pid part of this process, this would imply that during the assi-
milation acid will be produced and hence carbon dioxide will
be expelled from the suspension. In consequence of this the
real assimilation would be larger than the apparent one. On
the other hand it seems extremely unlikely that substances as
reactive as aldehydes would be the final products of the auto-
fermentation.

If the conclusion of Gaffron (1934) that the Thiorhoda-
ceae are able to reduce sulphates to sulphides in the dark,
could have been confirmed, the auto-assimilation could have
been explained as an assimilation with the aid of those sul-
phides. As set forth earlier for this assumption no sup-
port whatever has been found. Moreover, it appeared that
the typical symptoms of auto-assimilation could be observed
as well, when there were no sulphate ions present in the sus-
pension, as is shown for instance in the experiment represen-
ted in fig. 9. Still it remains possible that the auto-assimilation
proceeds at the cost of organic or inorganic SH-compounds
produced indirectly, e.g. in the autolytic breakdown of pro-
teins.

It must be admitted that the unknown chemical nature of
the products of the metabolism in the dark and in particular
the unknown nature of the hydrogen donators active in the
auto-assimilation, involves a factor of uncertainty in the in-
terpretation of the experiments, especially of those described
in Chapter V.

In using bacteria from a peptone culture medium, which
as shown before, produce hydrogen in the dark, the auto-
assimilation is still more complicated, because there also will
be a carbon dioxide assimilation with gaseous hydrogen as a
donator. As will be shown in § 4 and 5 of this Chapter, the
Thiorhodaceae are capable of performing this, in the dark as
well as in the light. In the latter case, however, the process is
much accelerated.

When the bacteria are not illuminated and when they have
only the traces of hydrogen produced by themselves at their

disposal, the hydrogen uptake will be scarcely perceptible
and hence the production of hydrogen will predominate by
far.

In the light, however, this is no more the case. In agreement
with the considerations given there could be observed an in-
crease of the pressure in the manometer vessels in the dark
and a decrease in the light, even when these bacteria were sus-
pended in a carbon dioxide buffer mixture, this observation
being in contrast to what is observed when bacteria cultivated
in an inorganic medium are used. To give an idea of these phe-

Curve I. Changes in pres-
sure effected by bacteria of
strain a, cultivated in the
peptone medium, suspended
in tap-water with 1,8%
NaCl, 0,3% NaHCOs and
0,01% Na2S, in equilibrium
with N2/3% CO2 at 30°C.
Differential manometer No.
3.

Curve II. Changes in pres-
sure effected by an equal
quantity of the same bac-
teria as in I, suspended in
0,05 molar CO2 buffer mix-
ture of 9 parts NaHCOs to
1 part Na2C03, in equili-
brium with a N2/CO2 at-
mosphere at 30°C. Differen-
tial manometer No. 3.

Curve III. Changes in pres-
sure effected by another
sample of bacteria of strain
d, cultivated in the inorga-
nic culture medium, and sus-
pended in the same solution
as in II, at the same tempe-
rature. Differential mano-
meter No. 3.

nomena some type experiments are represented graphically
in fig. 10. The difference in behaviour between bacteria from
organic and those from inorganic media, when suspended in
a carbon dioxide buffer mixture, is strikingly demonstrated
by the curves II and III. That the former are able to assimi-
late carbon dioxide with HgS as a hydrogen donator as well,
appears from a comparison of the curves I and II in fig. 10.

In the foregoing paragraph it was shown that when using
bacteria cultivated in the inorganic medium, the auto-assi-
milation was the higher the longer the previous period of
darkness had lasted. Therefore it was expected that this would

fig. 11 and fig. 11a.
Changes in pressure effected by bacteria of strain a, cultivated in the
peptone medium, suspended in a 0,05 molar CO2 buffer mixture of 9
parts NaHCOs to 1 part Na2C0s in distilled water with 2% NaCl, in
equilibrium with a N2/CO2 atmosphere at SCC. Differential manometer
No. 3.

fig. 11a. Velocity of the H2 uptake in the first, second and third period
of illumination of the experiment in fig. 11, in dependence on the quan-
tity of H2, present in the vessel at the times in question.

hold as well for the assimilation of the gaseous hydrogen
which accumulates in the dark periods when bacteria from
peptone culture media are used. Such an experiment is re-
presented graphically in fig. 11.

The bacteria used were cultivated in peptone water, sus-
pended in a carbon dioxide buffer mixturei) and as soon as
possible the experiment was started. Hence I could assume
that there were but traces of hydrogen in the manometer ves-
sel at zero time, that is at the moment when equilibrium was
attained in the vessels. The bacteria were exposed alternati-
vely to light and darkness and apparently the assimilation
of a gas, different from carbon dioxide, increased according
to the length of the period of darkness.

This cannot be explained otherwise than by assuming, that
in the light the bacteria assimilate the gas they produce in
the periods of darkness. The hydrogen production will not be
influenced by the pressure of the hydrogen already present,
or at least not to any appreciable extent. The hydrogen assi-
milation, however, will be influenced very much by the hy-
drogen pressure, at least in the region of the low tensions in
question. The course of changes in tension in the second pe-
riod of exposure to the light, at the beginning of which only 60
mm3 gas had been produced, will have to be explained by
the fact that the assimilation of hydrogen by the illuminated
cells of the rather dense suspension, did not yet compensate
the production of hydrogen by all the cells together. In the
third period however 100 mm3 hydrogen had been produced
and now the hydrogen assimilation was high enough to sur-
pass the hydrogen production.

Fig. 11 o shows that under the conditions of the experi-
ment the hydrogen assimilation is approximately proportio-
nal to the hydrogen tension, as may be expected in view of the
principle of the limiting factors (Blackman, 1905).

In order to prevent confusion it must be remarked that at
the beginning of the experiment of fig. 10 curves I and II a
considerable amount of hydrogen already had been produ-

1) This means of course that in the experiment to be described, only
changes in tension of gases different from carbon dioxide could be
observed.

ced, because this time I had not started the observations im-
mediately after equilibrium had been attained.

Finally I wish to mention an experiment in which it was
shown that the assimilation of the hydrogen produced by the
bacteria themselves was more rapid, depending on the con-
centration of the carbon dioxide they had at their disposal.
The uptake of hydrogen could therefore be limited either by
the hydrogen tension or by that of the carbon dioxide. This
is shown by the experiment represented in table 10.

Equal quantities of bacteria of the same culture from the
peptone medium were suspended in tap-water with 2% NaCl
and in a carbon dioxide buffer mixture. In the first mentio-
ned case alkali was present in a side-bulb of the manometer
vessel, in order to absorb the carbon dioxide produced by the
bacteria. Hence in both cases only changes in tension of ga-
ses other than carbon dioxide could be measured. It appears
from the data in table 10, that the bacteria which had an
excess of carbon dioxide at their disposal, assimilated more
hydrogen per unit of time than the bacteria which obtained
a small quantity of carbon dioxide. This notwithstanding the

fact that the quantities of hydrogen available in both cases
must have been equal, since the same amounts of hydrogen
will have been produced per unit of time. That also in the
second case an absorption of hydrogen could still be obser-
ved must probably be ascribed to the fact, that the alkali in
the side-bulb was unable to absorb immediately all carbon
dioxide formed in the autofermentation process.

All this may be considered as a proof of the view that the
tension of the carbon dioxide limits the assimilation of the
gas produced by the bacteria. Therefore this gas must be a
hydrogen donator and hence this experiment can also be
considered in support of the view, expressed in Chapter III,
that the unknown gas produced by the quot;peptone bacteriaquot;
during the autofermentation, is hydrogen indeed.

I have called the phenomenon described in this paragraph :
quot;auto-assimilation of carbon dioxidequot;. This term is in so far
misleading, as it may give the impression that we are dealing
here with a quite new phenomenon. However, since the green
plants can assimilate the carbon dioxide and water formed
in their respiration process, we meet here exactly the same
situation. The only difference is that the Thiorhodaceae use
other hydrogen donators instead of water.

The foregoing also throws new light on the following state-
ments of Muller (19331, p. 165): quot;In these cultures practical-
ly no metabolic products other than relatively small amounts
of CO2 have been detectedquot; and quot;it has been shown that in
all probability the substrate is completely converted into cell
material and CO2, i.o.w. that the assimilation predominates in
the metabolism.quot; From the facts reported above it follows that
this predomination of the assimilation process should not be
seen as a suppression of the other metabolic processes, but
rather that the substances produced in these processes are for
the greater part reassimilated again.

The latest experiences of Gaffron (1935) show that the
Aihiorhodaceae too are characterized by a rapid auto-assimi-
lation, followed by a slow one and also in this respect there
appears to be a close resemblance between Aihiorhodaceae
and Thiorhodaceae.

For the study of the assimilation of carbon dioxide with
hydrogen donators specially added, I always used quot;starvedquot;
bacteria, because under these conditions one meets with fewer
difficulties since the auto-assimilation is so much decreased.
Yet it remained necessary to determine the extent of the auto-
assimilation under the same conditions in the control mano-
meter and to introduce a correction for the auto-assimilation
in the proper experiment.

Because the assimilatory capacity of bacteria from an or-
ganic medium is always considerably less than that of bac-
teria from an inorganic medium, for the greater majority of
these experiments only the latter bacteria were used. Other
than quantitative differences, however, could not be observed.
The bacteria nearly always were suspended in tap-water with
2% NaCl, in most cases with the addition of 0,5% sodium-
bicarbonate, in equilibrium with nitrogen containing 5% car-
bon dioxide.

a. Experiments with HgS. As may be derived from fig. 12,
the assimilation of carbon dioxide with hydrogen sulphide
as a donator proceeded at a high speed. It was constant for
a long time and apparently remained so until all HgS had
been consumed (at 1.30 in fig. 12), whereafter it fell off to
the level of the auto-assimilation, which was of the same
order as that in the control manometer without HgS. Only
in very low concentrations HgS was the limiting factor for
the assimilation of carbon dioxide. By determining the
amount of HaS addedi), it was possible to calculate how many
molecules of carbon dioxide were assimilated per molecule
of HgS.

In this connection it has to be remarked that for the calcu-
lation the changes in pressure during the periods of light were
supposed to be quantitatively due to the assimilation of car-
bon dioxide. Hence it was assumed that a possible assimila-

1) This amount was determined by volumetric measurement of a
saturated aqueous solution of known temperature.

■1/-- '0
V n
V
1/ gt; 0
V
n
/ 7
f 7
■}
}
»
1
n


s	L

f3C?

Curve I. C02'production and auto-assimilation of ± 300 mm® bacteria
of strain d, cultivated in the inorganic medium, exposed to a period of
darkness of 20 hours at 36°^, and suspended in tap-water with 2%
NaCl and 0,5% NaHCOs, in equilibrium Vitb N2/7% CO2 at SO'C. Diffe-
rential manometer No. 3.

Curve n. Same arrangement, except the addition of 0,6 cm® saturated
solution of H2S in water of 14°G.

tion of acids from the autofermentation could be neglected.
For a documentation of this assumption the reader may be
referred to p. 69.

The decrease in pressure caused by absorption of HgS could
also be neglected, as may be concluded from the following.
By calculating with the aid of the dissociation constant of H2S
(for the first H') and with the absorption coëfficient of
H2S in distilled water the H2S pressure in equilibrium with
water of pH 7,8, it appears that this pressure can have been
at most 3,5 X 10 '' atm. This involves that only 0,2% of the
changes in pressure observed have been due to an absorption
of HaS.

Considering, that for all these reasons the found average
quotient H2S/CO2 of 1,75 will still be on the low side, one may
conclude, that two HgS molecules are used for the reduction
of one CO2 molecule. As the equations on p. 65 show, one has
to assume therefore, that in the first instance the HgS was
only oxidized to the oxidation level of sulphur. Anyhow the
assimilation with sulphur as a donator was too slow to permit
its study by the manometric method. This is in agreement

Ï) These data are corrected for the carbon dioxide consumed by
auto-assimilation under the same conditions, measured in a control
manometer.

with the facts, mentioned on p. 69, concerning the auto-assimi-
lation of Thiorhodaceae either or not containing sulphur. Up
to the present the remarkable difference in oxidation velocity
of HgS and sulphur had escaped notice, since the earlier in-
vestigations were restricted chiefly to the chemical analysis
of out-grown cultures.

Carbon dioxide assimila-
tion NaaSOs as a hydrogen
donator, in comparison with
the auto-assimilation and
with the carbon dioxide as-
similation with HaS as a
hydrogen donator.

Curve I. CO2 production
and CO2 uptake of ± 230
mm® bacteria of strain a,
cultivated in the inorganic
medium, exposed to a pe-
riod of darkness of 20 hours
at 36 °C. and suspended in
tap-water with 2% NaCl
and 0,5% NaHCOs, in equih-
brium with N2/5% CO2 at
30°C. Velocity of CO2 assi-
milation 95 mm®/hour. Dif-
ferential manometer No. 1.

Curve n. Same arrange-
ment as in I, except the
addition of 0,2% Na2S03
7 H2O. Assimilation velocity
370 mm®/hour. Differential
manometer No. 1.

Curve III. Same arrange-
ment as in I, except the ad-
dition of an aqueous solu-
tion of 2500 ram® H2S. Assi-
milation velocity 1230 mm®/
hour. Differential manome-
ter No. 1.

of growth experiments with the outcome of different methods
of investigation is to be found in the observation of Gaffron
(1933) regarding the assimilation of carbon dioxide with HgS
as a donator by an Athiorhodaceae. Still as far as known the-
se organisms are unable to grow in an inorganic medium.

b. Experiments with sulphite and thiosulphate. The assi-
milation of carbon dioxide with an excess of NagSOs as a hy-
drogen donator proceeded at a well measurable rate, though
much slower than with HgS under the same conditions. Vide
fig. 13.

In contrast to what had been found with H2S, I could ob-
serve a distinct influence of the NagSOg concentration upon
the assimilation velocity. I first determined the assimilation
velocity with 0,05% Na2S03 7 H2O and after addition of mo-
re NagSOg the assimilation velocity was determined again,
all other conditions remaining unchanged. Thus by compa-
ring bacteria under similar circumstances, it became evident
that below 0,2% Na2S03 the assimilation velocity increased
with higher concentrations, but that above 0,3%, the sodium
sulphite exerted a harmful influence upon the bacteria which
may be due to the fact that they were not accustomed to
such high concentrations of sulphite.

I will not venture to decide, whether a NagSOg concentra-
tion above 0,2% is indeed no longer a limiting factor in the
assimilation process, as possibly the harmful influence of sul-
phite becomes already manifest at concentrations below 0,3%.

Since for low concentrations of sulphite the assimilation
velocity is very small, I refrained from determining the quo-
tient NagSOg/COa as has been done for H2S.

The carbon dioxide assimilation with sodium thiosulphate
proceeds nearly as quick as that with sodium sulphite under
similar conditions. Detailed experiments were not carried out.
Gaffron (1934) stated, that the sulphur from the sodium
thiosulphate is not oxidized further. The average of the quo-
tient NaaSOg/GOg, as determined by him, was 2,2. This author
gives as his opinion that polythionic acids will be formed out
of the NaaSgOg. More probable it seems to me, that in the first
instance the thiosulphate will be converted into Na2S04 and
sulphur.

The experiments with the Thiorhodaceae cultivated in pep-
tone water as reported in Chapter III § 3 and in § 2b of this
Chapter induced me to study the behaviour of these bacteria

Curve I. Change in pres-
sure effected by the CO2
and Ha production of ±
800 mm® bacteria of strain
a, cultivated in the peptone
medium, suspended in tap-
water with 2% NaCl and
0,3% NaHCOs, in equilibri-
um with N2/5% CO2 at a
temperature of 30°C. Diffe-
rential manometer No. 3.

Curve II. Change in pres-
sure effected by the Ha pro-
duction of an equal quantity
of the same bacteria as in I,
suspended in a 0,05 molar
COa buffer mixture of 9
parts NaHCOs to 1 part
NaaCOa in tap-water with
2% NaCl, in equilibrium with
a N2/CO2 atmosphere at
30°C. Differential manome-
ter No. 1.

Curve HI. Continuation of the
experiment represented by
curve I, after the introduc-
tion of 65% N2, 30% H2 and
5% CO2 in the manometer
vessel.

Curvel I'V. Continuation of the
experiment represented by
curve II, after the introduc-
tion of 70% N2 and 30% Ha
in the manometer vessel.

Restricting myself for the moment to what happens in the
dark, I wish to examine more closely the experiment represen-
ted in fig. 14. The course of this experiment was as follows.
Bacteria of strain a, cultivated in peptone water, were centri-
fuged in such a way that finally two equal quantities of bac-
teria were obtained. One half was suspended in a solution of
sodium bicarbonate in a manometer vessel with nitrogen con-
taining 5% carbon dioxide. The other half was suspended in
a carbon dioxide buffer mixture. The changes in pressure in
the two manometers may be seen from the lines I and II of
fig. 14; they do not show anything particular, as these results
are in agreement with the experiences mentioned in Chap-
ter III § 3. This experiment only served to determine, how
much hydrogen and how much carbon dioxide was given off
per unit of time. These quantities were 41 mm^ and 92 mms
per hour respectively.

Thereupon the manometers were taken out of the water
bath and now a mixture of 65% nitrogen, 30% hydrogen and
5% carbon dioxide was introduced in the manometer vessel
containing the bicarbonate solution. The other manometer
was filled with 70% nitrogen, 30% hydrogen. After having
been replaced into the water bath, they showed, when equi-
librium had been established, the changes' in pressure, which
are represented in the lines III and IV. These results leave no
doubt that hydrogen was absorbed in the dark. The differen-
ce in the changes in pressure per unit of time, indicated by
the curves III and IV, enabled me to calculate how much
hydrogen was absorbed per unit of time. In the correspon-
ding manometer namely, changes in pressure resulting from
changes in carbon dioxide pressure were excluded. The Hg-
uptake was 82 mm3 per hour.

If one substracts from the increase in pressure of curve I
the value, corresponding with the absorption of hydrogen of
82 mms per hour, then the dotted line V is obtained. Now, on
the understanding that the absorption of hydrogen in the two
manometers has been the samei) the difference between the

1) The only difference in the two experiments was a different pres-
sure of carbon dioxide and a different pH. A lowering of the H2-uptake
by a limiting CO2 supply in the case of line III still seems to be possible.

changes in pressure of curves V and III can only be explained
by an absorption of carbon dioxide. This calculated C02-up-
take attained the value of 35 mm3 per hour. The quotient of
the hydrogen and the carbon dioxide taken up is then 2,4.

It seems premature to conclude definitely from these few
experiments, which certainly are liable to objections, that per
molecule of carbon dioxide two molecules hydrogen are ab-
sorbed.

However, the experiments prove that a carbon dioxide and
hydrogen absorption by the Thiorhodaceae occurs in the dark.
Furthermore there are distinct indications, that at least two
hydrogen molecules are wanted for the reduction of one car-
bon dioxide molecule. This is in accordance with the general
formula for the carbon dioxide assimilation.

For further investigations in this direction it is, of course,
recommendable to use the non-hydrogen forming bacteria
from inorganic media. These moreover take up the hydrogen
much more rapidly and only by the use of very dilute sus-
pensions of bacteria and a high hydrogen pressure, it will be
possible to measure the velocity of hydrogen and carbon
dioxide uptake, without being encumbered by the slow sup-
ply of hydrogen as a consequence of the small solubility of
this gas in water.

Casually I studied with bacteria cultivated in peptone water
how quick the absorption of hydrogen was, if the pressure
of this gas was decreased. With 1% hydrogen in the atmos-
phere in the manometer vessels the hydrogen absorption was
only 20% of the hydrogen production. With 2% hydrogen it
was still only 40% and not until the hydrogen concentration
reached 4% it was high enough to surpass the hydrogen pro-
duction. Practically the gaseous hydrogen produced by the
bacteria themselves is not taken up again to any degree. How-
ever, this is no longer the case when the suspension is illu-
minated (Compare § 2h of this Chapter.)

COa-and Ha-assimilation in the darlc and in the light.
Curve I. Change in pressure effected by the Ha uptake of bacteria of
Sn d cultivated in the inorganic medium, suspended in 0,05 molar
COa buffef mixture of 9 parts NaHCOs to 1 part NaaCOs m tap-water
with 2% NaCl, in equilibrium with a Na/COa atmosphere at 30°C. Dif-
ferential manometer No. 3.nbsp;^ t, . i »
Curve H Change in pressure effected by COa and Ha uptake of ^
Sual c^JanHty i the Lme bacteria, suspended D«
NaCl and 0,5% NaHCOs, in equilibrium with N2/7% COa at 30 C. Dif-
ferential manometer no. 1.

had carbon dioxide and hydrogen at their disposal, I could
always observe a strong absorption of hydrogen as well as
of carbon dioxide. Fig. 15 can give an idea of this. I have
not determined the value of the quotient H2/CO2, since too
great a number of complicating factors were involved.

This quotient, however, has been determined by Gaffron
(1935) by measuring the hydrogen assimilation in the light
of suspensions of an Athiorhodaceae on addition of known
quantities of soda. Thus he obtained very satisfactory results,
amounting to an average of 2 for the quotient H2/CO2.

As the reduction of carbon dioxide with hydrogen under
these conditions is a reaction, which is characterized by a
decrease of free energy, and thus can proceed without the
aid of light energy, the acceleration by the illumination has
to be considered as a sensibilisation.

I consider the result of experiments like those which are
shown graphically in fig. 16 as a proof that also in this upta-
ke of hydrogen the carbon dioxide is the principal acceptor.

This experiment was quite the same as the one of table 10,
where it was demonstrated that for the assimilation of the
hydrogen formed by the bacteria carbon dioxide was neces-
sary. The only difference is that in the present series of expe-
riments about 30% hydrogen was present in the gas mixture
within the manometer vessels.

The results are clear, in so far that they prove once more
that carbon dioxide was the principal, if not the only, accep-
tor available to the bacteria in these circumstances. The fact
that in the vessel with alkali the illumination does not exert
any influence, in contrast with what holds for the vessel with
carbon dioxide, shows that in the former case the concentra-
tion of carbon dioxide was the limiting factor for the hydro-
gen assimilation.

In a few experiments of a preliminary character I investi-
gated whether special reducible substances could replace the
carbon dioxide as a hydrogen acceptor for gaseous hydrogen.
For this purpose I compared the velocity of absorption of

Curve I Change in pressure effected by the H2 uptake of bacteria of
strain a, cultivated in the peptone medium, suspended in tap-water with
2% NaCl in equilibrium with N2/30% H2 at 30°C. Alkali present m a
side-bulb of the manometer vessel. Differential manometer No. 3.

Curve II Change in pressure effected by the H2 uptake of an equal
quantity'of the same bacteria, suspended in a 0,05 molar CO2 buffer
Sure of 9 parts NaHCOs and 1 part Na2COs in tap-water with 2%
SaCl in equilibrium with a N2/CO2 atmosphere at 30°C. Differential

hydrogen in the dark and in the light by bacteria suspended
in tap-water with 2% NaCl, with the hydrogen uptake of an
equal quantity of the same bacteria suspended in the same
solution after addition of 0,1% Na-fumarate, or 0,05% Ca-
formate, or 0,1% NaNOg.

As may be seen from table 12, the results were completely
negative. Hence, as far as may be judged from these few ob-
servations, the Thiorhodaceae are unable to reduce under
these conditions the compounds in question with gaseous hy-
drogen. The slight decrease in pressure observed was appa-
rently due to an assimilation of hydrogen with the carbon
dioxide produced by the bacteria.

According to the experience of Gaffron (1935) the beha-
viour of an Aihiorhodaceae cultivated in yeast extract was
quite different.

It seemed interesting to ascertain whether the Thiorhoda-
ceae able to grow in an inorganic medium devoid of any
oxidizable sulphur compound, provided they have hydrogen
and carbon dioxide at their disposal. For this purpose tubes

containing standard solution to whicti 0,5% NaHCOg and
0,005% NagS had been added, were inoculated and placed in
an quot;anaerobic jarquot;. The jar was then filled with an atmosphe-
re of hydrogen with 5% carbon dioxide, free from oxygen, and
was placed before an electric bulb at 30° C.

Already after a few days definite signs of development of
the bacteria were present. As 1 was able to cultivate several
generations in this way, it is certainly out of question, that the
growth had taken place exclusively at the cost of the thiosul-
phate introduced with the first inoculation.

The 0,005% NagS was added to bring about a sufficiently
reduced state of the medium in the tubes and was necessary
to ensure a good start of the bacteria. If one calculates, how
much organic matter could be synthesized with the amount of
Na2S present per culture tube, then — assuming a complete
oxidation of the NagS tot Na2S04 — this appears to amount
to 0,5 mgr. at most. This organic matter corresponds with a
quantity of bacteria, which is not perceptible in that dilution,
of which I convinced myself by preparing such a suspension.
It may be added that I always could observe a definite de-
crease in pressure of the atmosphere in the anaerobic jar
after the bacteria had been growing in it for some time.

Further experiments will have to decide how much hydro-
gen and how much carbon dioxide is consumed in such growth
experiments, and whether also under these conditions carbon
dioxide really is the only acceptor which can be used by the
Thiorhodaceae.

In connection with the hydrogen and carbon dioxide assi-
milation performed by the Thiorhodaceae without the aid of
radiant energy, it seemed to be of interest to ascertain whether
it would also be possible to cultivate the bacteria under such
conditions. Organisms capable of growth at the expense of a
hydrogen and carbon dioxide mixture have already been
found before (see for instance Sohngen 1906, Stephénson
and Stickland 1933). To investigate this the anaerobic jars
with tubes containing young cultures of bacteria in a H2/CO2
atmosphere, were put in an ordinary (dark) incubator at a
temperature of 30° C. Without exception the bacteria ceased
to develop. Even cultures in a Na-malate-Na-thiosulphate me-

dium brought in the hydrogen carbon dioxide atmosphere,
refused to grow in the dark.

I did not investigate whether Athiorhodaceae too are able to
grow in a hydrogen carbon dioxide atmosphere in the light. If
so, this would be the first case in which these organisms can
be cultivated in an inorganic medium. Even when they would
require the presence of an organic substance under such con-
ditions in the dark, this would be a remarkable fact, as hither-
to Athiorhodaceae have never been cultivated in complete
darkness under anaerobic conditions. The only instance I
could find in the literature about the cultivation of Athiorho-
daceae in a hydrogen atmosphere, was the following statement
of Migula (1900) : „Nach meinen Versuchen gedeiht Spiril-
lum rubrum auf schrägem Agar auch in Wasserstoffat-
mosphäre vorzüglich, bildet aber ebenfalls keinen Farbstoffquot;.
Whether in this case the cultures were illuminated remains
an open question.

I have also made a casual observation regarding a possible
assimilation of hydrogen by green plants. Though Boussin-
gault (1868) and especially Ewart (1896) have found that
hydrogen and hydrogen carbon dioxide mixtures depressed
the rate of carbon dioxide assimilation of Chara and Elodea,
and even could inhibit this process, this might have been only
due to a lack of oxygen. Therefore I have tried whether
the green alga Stichococcus bacillaris could assimilate hy-
drogen in an atmosphere of air with 20% hydrogen, either
in the dark or in the light. For this purpose a pure cultu-
re of this alga was cultivated in daylight in the solution
used by Eilers (1926). The algae were suspended in a 0,05
molar carbon dioxide buffer mixture of 8 KHCO3 against 2
K2CO3.

The result was negative; green plants do not seem to be able
to derive from hydrogen gas the H-atoms necessary for the
reduction of carbon dioxide.

In connection herewith I will remark, that it would be inte-
resting to ascertain whether the green plants like the Rhodo-
bacteriales are able to involve organic acids into the photo-
chemical reduction of carbon dioxide.

wing cultures of green algae, organic compounds can disap-
pear from the nutritive medium in the course of time. This
may obviously be attributed to an oxidation of these com-
pounds with oxygen (respiration). The possibility that the
compounds in question will partly have acted as donators
for the photochemical carbon dioxide reduction, is however
not to be excluded. An indication for this could possibly be ob-
tained by determining in how far in such experiments the
assimilation quotient deviates from 1, provided the final pro-
duct of the assimilation remains a carbohydrate.

For the sake of completeness I will mention some experi-
ments carried out in order to establish whether carbon mono-
xide and methane are gases, which either in the dark or in the
light, could be utilized by the purple sulphur bacteria. For
this purpose the bacteria were suspended in the bicarbonate
solution and in a carbon dioxide buffer mixture, while the
gas phase in the manometer consisted of a mixture of nitrogen,
carbon dioxide and 30% of the gas to be studied (table 13).

Neither in the dark, nor in the light, could I observe anything
particular, whilst also the microscopic appearance of the bac-
teria at the end of the experiment was normal. Apparently
carbon monoxide and methane are completely inert gases for
the Thiorhodaceae, at least as far as can be judged from these
few observations.

Since the researches of Molisch (1907) on the Athiorhoda-
ceae and of van Niel (1931) on the Thiorhodaceae, it is
known that both groups of organisms can assimilate the orga-
nic substances offered to them. Muller (19331) has given ex-
perimental proof that the Thiorhodaceae also under these
conditions use carbon dioxide as an acceptor. The organic com-
pounds should undergo a series of dehydrogenations until
they are converted into pyruvic acid, acetaldehyde or carbon
dioxide. With a view to the great similarity in behaviour of
Thiorhodaceae and Athiorhodaceae Muller concluded that

Bacteria of strain a, cultivated in the peptone
medium, suspended in 0,05 molar CO2 buffer
mixture of 9 parts NaHCOs and 1 part NaaCOs
in tap-water 2% NaCl, at 30°C. Pd black pre-
sent in side-bulb.

Bacteria of strain d, cultivated in the inorga-
nic medium, starved during 20 h. at 36 °C.,
suspended in 0,05 molar CO2 buffer mixture
of 8 NaHCOs and 2 Na2C03 in tap-water with
2% NaCl at 30°C.

Bacteria of strain d, cultivated and treated
as the bacteria in experiment no. 2, suspen-
ded in:

a.nbsp;0,1 molar COs buffer mixture of 9 Na-
HCOs and Na2COs in tap-water 2%NaCl.
Manometer no. 1.

in all probability the metabolism of both groups of orga-
nisms in a suitable organic medium under anaerobic condi-
tions in the light will be the same.

The correctness of this assumption was then confirmed
by experiments of Gaffron (1933). In a later investigation
(1934), however, this author made observations, which led
him to'the following conclusion: (I.e., p. 447) „Der Unterschied
des Stoffwechsels von roten Schwefelbakterien (Thiocystis)
und Purpurbakterien (Rhodovibrio) ist sehr deutlich. Gibt
man zu einer Suspension von Rhodovibrio etwas buttersaures
Natrium und belichtet, so werden für jedes Molekül Butter-
säure 0,4 Moleküle Kohlensäure assimiliert, ausserdem wird
die Carboxylgruppe reduziert. Macht man denselben Versuch
mit Thiocystis, so geschieht gar nichts. Trotzdem wächst Thio-
cystis sehr gut in Medien, die statt Sulfid organische Verbin-
dungen enthalten. Die Erklärung hierfür liegt in einer Reak-
tion, die von der Kohlensäure-assimilation unabhängig ist
und'im Dunkeln (anaerob) abläuft. Während Rhodovibrio
Butyrat im Dunkeln unverändert lässt, benützt Thiocystis
die organische Substanz zur Reduktion von Sulfaten. Hier-
bei entstehen Sulfide, Kohlensäure und organische Säuren.
Die gebildeten Sulfide ermöglichen eine Assimilation von
Kohlensäure wenn die Bakterien belichtet werden.quot;

The experiments mentioned on p. 49 and p. 52 carried out
by me in order to test this point, justify the conclusion that
the explanation given by Gaffron of the difference in beha-
viour of Rhodovibrio and Thiocystis in his experiments, is in-
correct.

I had already made the observation, that the bacteria culti-
vated in peptone water indeed could use Na-malate as dona-
tor (table 14 No. 5 and 8), whereas the facts reported in § 2
of this Chapter show that the Thiorhodaceae very probably
are also able to use as hydrogen donators the organic products
formed in their dark metabolism. Hence I was very sceptical
of Gaffron's statement regarding the unsuitability of malate,
glucose, a.o., to serve as hydrogen donators for the Thiorhoda-
ceae.

When I repeated his experiments and used stram d, cul-
tivated in an inorganic medium, it appeared that the assimila-

tion with the organic donators mentioned was indeed very
weak as compared with the assimilation with HgS as do-
natori). Experiment no. 4 on table 14 shows that there is no
question of a carbon dioxide assimilation with ethyl alcohol,
since the assimilation was exactly equal to that in the control
manometer. Sodium malate and sodium lactate, however, were
assimilated unmistakebly, although very slowly (vide table 14
No. 1 and 2). Acetaldehyde caused a much stronger assimila-
tion (table 14 No. 3.)

However, far stronger assimilation of sodium malate could
be observed, when I used bacteria from the same strain d,
which were cultivated in peptone water (table 14 No. 5) or
in a Na-malate-bicarbonate medium (table 14 No. 6). It pro-
ved to be the same case with the assimilation of sodium bu-
tyrate by bacteria, cultivated in a sodium butyrate-bicarbonate

In these experiments concerning the influence of the quot;his-
toryquot; of the bacteria on the ability to use special organic
substances as donators, it appeared that this ability was to a
high degree dependent on the concentration in which these
donators were added. In other words, the supply of these
organic compounds to those parts of the cells in which the
photochemical reaction occurred, was a limiting factor in the
carbon dioxide assimilation, at least in those cells which were
under the most favourable conditions of illumination. One
therefore gets the impression, that the observed differences
are caused by changes in the permeability as a result of the

The differences stated by Gaffron (1934) between Athio-
rhodaceae and Thiorhodaceae thus appear to bear only a
quantitative character. Particularly if one realizes that this
investigator has compared Athiorhodaceae, cultivated in yeast
water, with Thiorhodaceae cultivated in an inorganic medium,
these differences lose their significance.

If one wishes to investigate the behaviour of Thiorhodaceae
towards certain compounds, one should use organisms which

1) In the beginning, viz. in my paper in 1934, I even believed that
under these conditions, the bacteria were unable to use organic hydrogen
donators.

are adapted to grow in a pure culture in media containing
the same compounds. Only then one can draw reliable conclu-
sions from a negative result of such an experiment.

I am fully aware that still many experiments must be per-
formed before a clear insight in these problems can be obtai-
ned. The experiments carried out only had an orientating
character and aimed chiefly at testing the correctness of the
assumption of Gaffron (1934) regarding the essential diffe-
rence between the Aihiorhodaceae and the Thiorhodaceae.

It is a well-known fact, that the nature of the metabolism of
organisms is reflected in the oxidation-reduction potential
occurring in the surrounding medium. For an extensive sur-
vey of the literature on this subject up to 1932 I refer IoElema
(1932). This investigator was the first, who successfully tried
to link up the nature of the predominating metabolic process

with the value of the potential. For further publications I
refer to: Elema, Kluyver and van Dalfsen (1934); Kluyver
and Hoogerheide (1934); Kingma Boltjes (1934); Lipmann

Oxidation-reduction potential in suspensions of bacteria in the light
and in the dark.

Bacteria of strain d, cultivated in the inorganic medium, suspended in
tap-water with 2% NaCl, 0,7% NaHCOs and 0,2% NaaSOs, in equilibrium
with N2/5% COa at 30°C, pH 8.1. Na/5% COa constantly passing through
the suspension. A 40 Watt electric bulb at a distance of 10 cm serving
as a light source.

Oxidation-reduction potential in suspensions of bacteria in the light
and in the dark.

Bacteria of strain d, cultivated in the inorganic medium, suspended
in tap-water with 2% NaCl, 0,3% NaHCOs, in equilibrium with N2/5%
CO2 at 30°C., pH 7,5. N2/5% CO2 constantly passing through the suspen-
sion. A 40 Watt electric bulb at a distance of 10 cm serving as a light
source.

Curve II. Same arrangement as in I, except the addition of 0,2%
NaaSOs to the medium.

I decided therefore to investigate, to which degree and in
which sense the redoxpotential in suspensions of Thiorhoda-
ceae is altered by the carbon dioxide assimilation, i.o.w. indi-
rectly by exposure of these suspensions to light. The results
of the experiments in question are represented in fig. 17 and
18 and in table 15. After being centrifuged and washed, the
bacteria were suspended in the ordinary solution of ± 0,5%
NaHCOs in tap-water with 2% sodium chloride, previously

brought in equilibrium with nitrogen containing 5% carbon
dioxide. As a hydrogen donator 0,2% NagSOg or 0,2% NagSsOs
was added, as these compounds, in contrast to H2S, do not di-
rectly influence the potential of the electrodes.

With the apparatus described by Elema (1932) I measured
the Eh in these suspensions under anaerobic conditions at a
temperature of 30°C. in the dark and in the light. During the
whole experiment an oxygen-free mixture of nitrogen con-
taining 5% carbon dioxide passed through the suspension. At
the end the pH was measured by means of the glass electrode
(Elema 1932). As a light source an ordinary electric bulb of
40 Watt, put at a distance of 10 cm from the bacterial suspen-
sion, was used.

The curves in fig. 17 and curve II of fig. 18 relate to suspen-
sions to which a hydrogen donator had been added, curve I
of fig. 18 represents the change of the E^ in the same suspen-
sion without a hydrogen donator. From the results it may be
concluded, that under the conditions of these experiments the
auto-assimilation and the carbon dioxide assimilation with the
added hydrogen donators induced an increase of the E^ of
± 60 m Volt, and 120 m Volt, respectively.

The rise brought about by the auto-assimilation pretty soon
changed into a considerable decrease during the exposure.
These facts correspond with the course of the auto-assimila-
tion as represented in fig. 9. After the short period of dark-
ness in fig. 18 from 4,15 until 4,32 the difference between
the E,, of the suspension of bacteria without and with hy-
drogen donator is still more striking. In consequence of the
shortness of the dark period the bacteria did not have at their
disposal products of autofermentation which could act as a
hydrogen donator.

Table 15 represents the results of experiments with bacte-
ria cultivated in a peptone medium. As might be inferred
from the property of these bacteria to produce hydrogen the
potential of these suspensions was much lower than in the
experiments with bacteria from an inorganic medium. It may
be remarked, that the measured Ej, was still much higher
than the E^ of the hydrogen electrode for atmospheric pres-
sure and for the pH of the medium. This may be due to the

fact that the hydrogen tensions, which occurred in the bacte-
rial suspensions, will have been extremely low, owing to the
nitrogen 5% carbon dioxide mixture, which was continuously
passing through the media. At these low tensions the E^ of
the hydrogen electrode is accordingly higher; under the con-
ditions of the experiment an E^ of 250 mV will correspond
to the potential of a hydrogen electrode at a hydrogen tension
of 10-66OO atmosphere. Moreover it must be emphasized that
the Eh in the suspension cannot be anything else but the
resultant of the different E^ characteristic of the various
oxido-reduction processes, which together constitute the auto-
fermentation, the assimilation and the hydrogen production.

In such a heterogeneous system like the cell-protoplasm,
it will be possible that these reactions proceed independent
of each other, and hence the Ej,, occurring in the surroun-
ding medium, is nothing else but a rough indication of the
total metabolism. In those cases, where one special metabolic
process, for instance alcoholic fermentation or denitrification,
is strongly predominating, more absolute significance can be
attached to the values of the Ej, observed.

In view of the changes in the E,, in aqueous solutions of
inorganic salts, observed by Pincussen (1934) as a result of
the exposure of these solutions to the light, it seemed advi-
sable to try whether this phenomenon could have interfered
in my experiments. However, in the absence of bacteria I
could not observe any change in the E^ in the suspension li-
quid on exposure to the light.

Finally the possibility that changes of the temperature of
the suspensions with bacteria, caused by the exposure to the
light, might have induced the changes in the E,, observed,
may practically be discarded. Firstly the changes of the E,,
on exposure were too rapid and secondly the observed diffe-
rence in the E^ of the suspensions with and without added
hydrogen donators cannot be accounted for.

In a way suspensions of Thiorhodaceae behave therefore as
a photo-half-cell, provided substrates for the assimilation are
present.

1.nbsp;The Thiorhodaceae are able to assimilate carbon dioxide
with special, still unknown products of their own dark
metabolism.

2.nbsp;The results of my manometric experiments are in agree-
ment with the theory of van Niel (1931) concerning the
assimilation of carbon dioxide with oxidizable sulpur com-
pounds.

3.nbsp;The manometric method is unsuitable to prove the hy-
drogen donator character of the sulphur stored in the cells
of the Thiorhodaceae or produced during the oxidation of
HgS or NagSaOg; the assimilation with sulphur evidently

4.nbsp;The Thiorhodaceae assimilate both hydrogen and carbon
dioxide in the light as well as in the dark; however, in
the former case with much higher velocity.

5.nbsp;The Thiorhodaceae are able to grow in the light in a
wholly inorganic medium, devoid of an appreciable quan-
tity of oxidizable sulphur compounds, provided they have
gaseous hydrogen at their disposal.

6.nbsp;The ability of the Rhodobacteriales to use organic sub-
stances as a hydrogen donator for the reduction of carbon
dioxide is to a high degree dependent on the nature of the
medium, in which the organisms were cultivated.

8.nbsp;The oxidation-reduction potential in suspensions of Thio-
rhodaceae increases considerably on exposure of these
suspensions to the light.

In this chapter I want to discuss the importance of a deter-
mination of the quantity of light required for the carbon
dioxide assimilation by the Thiorhodaceae. Moreover I shall
report about a series of experiments which I have undertaken
for these determinations. I want to emphasize in advance,
however, that these experiments have not led to ultimate re-
sults, they rather have a merely preliminary character. How-
ever, private circumstances forced me to bring my investiga-
tions to an end. Still it seems to me that it is justified to
publish the results obtained until now, since they may be
helpful to other investigators, who would feel inclined to aim
at a more definite solution of the problem in question.

On discussing the energy relations of the carbon dioxide as-
similation of the Thiorhodaceae it seems appropriate to give
first attention to what is known about these relations in the
corresponding process in green plants.

The thermodynamic side of this process has been very ela-
borately dealt with by Stern (1933), to whose survey may
be referred here. The view held by Stern, Briggs (1929),
wurmser (1929) and which is also found in the monographs
of Spoehr (1926) and of Stiles (1925) is as follows. The
actual photochemical process can be represented by the gene-
ral equation COg HgO CHgO Og. The increase in ther-
modynamic potential of this reaction is, when the actual con-
centrations of primary and final products in the cell are
taken into consideration, about 114.600 cal. i) (c.f. Stern i.e.)

This corresponds to 2,7 quanta of the wave length 660 mp,,
2,3 quanta of 578 mix and 1,8 quanta of 436 m/x (Stern I.e.
p. 370). With a view to this the authors mentioned are of opi-
nion, that one would expect to find three quanta of red or
yellow light and two quanta of blue light sufficient for the
reduction of one molecule of carbon dioxide by the green

reaction is very scant .The results of the older investiga-
tors endeavouring to arrive at a determination of this effi-
ciency (I refer for this to Spoehr 1926), have lost the greater
part of their interest. The reasons for this lack of reliable
data partly lie in the inaccuracy of the methods applied and
are for the rest due to a neglect of the necessity to control all
the factors which determine the rate of photosynthesis.

Warburg and Negelein (1923), however, approached the
said problem in a new and refined manner. Their investiga-
tion on the carbon dioxide assimilation in thick suspensions
of the alga Chlorella led to the conclusion, that 4,4 quanta of
red and yellow and 5,1 quanta of blue light are required to
attain the assimilation of one molecule of carbon dioxide.

Since the appearance of the fundamental publications of
Warburg and Negelein, other attempts to arrive at an ex-
perimental solution of the problem in question have been
Lde by WuRMSER (1923, 1924, 1926, 1929), Briggs (1929),
schmucker (1930) and Burns (1933). Whilst Schmucker
comes to results, which are quite identical with those of
Warburg and Negelein, the other authors found conside-
rably higher values. However, even the numbers of Warburg
and of Schmucker turn out higher (at least 50% for red and
yellow light and 150% for blue light) than might be expected
solely on the ground of the above mentioned energetic consi-
derations. This low efficiency, so they argue, may be accoun-
ted for by assuming that not all quanta absorbed need have
been photochemically active. Some quanta would have been

1) On the ground of measurements of the rH in assimilating cells,
Wurmser (1926) comes, it is true, for the carbon dioxide reduction
proper to only 43.800 cal. However, the starting point of his calculation
is too much hypothetical to attach much value to this conclusion.

lost as fluorescence energy and other quanta would have
been converted into heat, f. i. if the activated chlorophyll or
chlorophyll-carbonic acid complex loses its energy in colli-
sions with other than the chemically reactive molecules. Mo-
reover other coloured and colourless photochemically inacti-
ve substances in the cell, as f.i. the Carolines and protoplasm,
may have been responsible for the absorption of a portion
of the light.

Though thermodynamics can only tell us that a definite
photochemical reaction may take place under uptake of such
and so many quanta, the foregoing authors apparently take it
for granted, that the only condition to be fulfilled is to supply
the carbon dioxide molecule or the carbonic acid-chlorophyll-
complex with the amount of energy thermodynamically re-
quired.

It is characteristic of these considerations that therein
Einstein's law of photochemical equivalence, according to
which an equal number of quanta of various wave lengths
produces the same photochemical effect, is neglected.

Warburg and Negelein (I.e.) have concluded from the re-
sults of their investigations that Einstein's law holds also in
the case of the photochemical carbon dioxide reduction.
However, they did not find fully equal quantum numbers with
different wave lengths, but they ascribe the differences ob-
served to experimental complications and hence consider them
as being unimportant and only apparent. Whilst these authors
thus emphasize the probability of finding an equal number
of quanta with different wave lengths, they are on the other
hand relatively indifferent with respect to the absolute value
of this number. They only strive to attain as high an energy
efficiency of the carbon dioxide assimilation as possible.

This is due to the fact, that they do not try to visualize the
internal mechanism of the undoubtedly complex photosyn-
thetic process. We may conclude directly to this complexity
from the extreme improbability that the photochemically ac-
tive substance would be capable of gathering a number of
quanta, before the reaction CO2 HgO ^ CHgO -fOg pro-
ceeds. From a physical point of view namely, the degree of
probability for a molecule once activated by the uptake of

one quantum, to find ttie opportunity to absorb one or even
two other quanta in addition, is exceedingly small, conside-
ring the short lifetimes of activated molecules in general (see
f.i. Bonhoeffer and Harteck, 1933) i).

From this it seems quite indispensable to study the internal
mechanism of the carbon dioxide reduction. However, it is
superfluous to enter into a detailed discussion of the theories
existing thereabout. I refer to the monographs of Holluta
(1926) and of Spoehr (1926) and to the publication of Mul-
ler (19332). For the question in consideration it will suffice
to state that various of the theories proposed are characteri-
zed by the opinion that in reducing one carbon dioxide mole-
cule several separate endothermic reactions have to proceed
successively. This implies that in this line of thought several
photochemical processes have to follow one another before
a carbon dioxide molecule has been brought to the reduction
level of a carbohydrate.

This conception underlies the theories of van Niel and
Muller (1931) and of Shibata (1933). These investigators
assume that in the carbon dioxide reduction process more or
less stable intermediate stages are formed. Van Niel and
Muller have postulated formic acid as an intermediate pro-
duct, whereas Shibata apparently assumes, that four more
or less stable stages are passed through in the carbon dioxide
reductions). Which intermediate stages are involved, which
pigments are responsible for the photochemical reactions,
whether hydrogen atoms or carbon dioxide are activated the-
reby etc., is as yet of minor importance. The only essential
point for the moment is that the mentioned investigators ac-
cept that four so-called photochemical primary processes^) are

1)nbsp;This view is corroborated by the experimental result, that at least
with low light intensities, the velocity of the carbon dioxide assimilation
is directly proportional to the intensity (I) of the absorbed light. If a
simultaneous absorption of n quanta would occur, it would have been
proportional to Iquot; (James 1934).

2)nbsp;This assumption is to a certain extent supported by the recent
views regarding the rôle of free radicals in biochemical processes. C.f.
willstâtter and Haber (1931).

3)nbsp;With photochemical primary process is meant the initial process
wherein absorbed radiant energy is converted into chemical energy.

The fact, that Warburg and Negeleen (1923) and all
other authors previously mentioned have never found a smal-
ler number of quanta than just four to be sufficient for the
carbon dioxide reduction, may be considered a material sup-
port for the above conceptions regarding the internal mecha-
nism of the carbon dioxide reduction process. For as is known,
the Einstein law of photochemical equivalence requires light
absorption only to take place in quanta.

So if we assume the presence of four photochemical pri-
mary processes in the carbon dioxide reduction, then this ne-
cessarily means that just four quanta are absorbed in this pro-
cess, since, as was already mentioned, the idea of a simul-
taneous absorption of two or more quanta by the reacting
molecule is to be rejected.

On summarizing the foregoing considerations we may con-
clude, that if we only take into account the requirements of
thermodynamics, we must assume, that for the process of the
carbon dioxide assimilation by the green plants three, or for
special wave lengths two quanta, per molecule carbon dioxide
will suffice. On the contrary the investigators, who assume
that four photochemical primary processes are involved, have
to accept that this quantum number has to amount to four.
This means that an experimental determination of this number
offers a way to test experimentally the correctness of these
conceptions.

The trouble, however, is that the margin between the num-
bers of quanta postulated in both conceptions — three against
four quanta — is rather small, whilst an experimental deter-
mination of this number, as will appear from the following
paragraphs, is linked up with great difficulties and inaccura-
cies. Hence it is doubtful whether it is possible to come to a
definite answer when using green plants for the experiments.

For this reason it seemed highly important to study this
problem with the Thiorhodaceae. As has been set forth al-
ready in the Introduction, van Niel has shown that it is cha-
racteristic of the photochemical carbon dioxide reduction with
this group of bacteria, that not water, but oxidizable sulphur
compounds or organic compounds act as hydrogen donators

in this process. This involves that the increase of the free
energy in this case is much smaller than in the case of the
green plants, because in the latter the carbon dioxide re-
duction includes the splitting of water under liberation of
oxygen. In the case where hydrogen sulphide acts as a hy-
drogen donator, Stern calculates the increase of the free
energy of the reaction CO2 2 HsS CH2O HgO 2 S to
only a good 17.000 cal. under standard conditions. This implies
that in this case one quantum of visible or neighbouring ultra-
red light will already be amply sufficient to fulfil the ther-
modynamical requirements of this reaction. On the other hand
van Kiel's experimental proof that in Thiorhodaceae four
hydrogen atoms are involved in the photochemical assimila-
tion of one carbon dioxide molecule, makes it extremely pro-
bable that, here as well as in the carbon dioxide assimilation
of the green plants, four photochemical primary processes
will occur. It then necessarily follows, that in both processes
four quanta are required per molecule of carbon dioxide
assimilated.

When the experiments would show, that indeed in spite of
the very low energy requirements of the process, under no
conditions a lower quantum number than four can be found,
this might be considered as a strong indication, that in the
reduction of each carbon dioxide molecule four photochemical
primary processes are involved. This would mean, that prior
to the introduction of each hydrogen atom in the carbon di-
oxide molecule a photochemical primary process is required.
Of course such a result would strongly plead for an analo-
gous situation in the carbon dioxide assimilation of green
plantsi).

If on the other hand we should find, that less than four
quanta_f.i. one — would suffice, then this would mean in

1) Whether the hydrogen atoms of the donators, or the carbon di-
oxide and its subsequent reduction stages are activated, is another ques-
tion, which can be left out of consideration here. I refer in this connec-
tion to the different views of Noack (1926), of van Niel and Muller
(1931), of Muller (1933^), of Stoll (1932), of SmBATA (1933) and of
Dhar (1934).

all probability that the carbon dioxide molecule would be the
substrate of the photochemical primary process. The reduc-
tion of this molecule would then proceed in secondary con-
secutive reactions of non photochemical nature.

Furthermore I can point here to a second advantage of using
in particular the Thiorhodaceae for the determination of the
number of quanta required. As has been previously remarked,
the determination in question is attended with great experi-
mental difficulties, principally due to the fact that it is hardly
to realize that all the light falling on the bacteria acts photo-
chemically. If in the case where hydrogen sulphide is the dona-
tor we should find that no quantum number lower than four
can be reached, we. may still stick to the opinion that only one
quantum has been photochemically active and that the others
have been lost by absorption, by conversion into fluorescence
energy etc. The advantage of operating with purple bacteria
is, however, the fact that these organisms can use such greatly
different hydrogen donators. Among these there will be such
as to require, from an energetic point of view, the co-operation
of more than one quantum^). If now in using these donators
the number of quanta determined should likewise prove to be
four — or if at least this number would not change with the
nature of the different donators used — this would certainly
speak in favour of the assumption that in all cases four photo-
chemical primary processes are involved.

Finally it seemed that the determination of the number of
quanta might be of importance to decide which of the two
pigments of the Thiorhodaceae would be the photochemically
active one. It is generally assumed that it would be the green
pigment, named bacteriochlorine by Molisch (1907) and
bacteriochlorophyll by Noack and Schneider (1933). This
assumption is supported by the results of Schneider (1934)
and of Fischer and Hasenkamp (1935), who have shown
that this pigment is closely related to chlorophyll. Van Niel
and Muller (1931) are inclined to the view that the red

1) The carbon dioxide reduction, whereby S(H20) is converted in
three stages into h2so4, means an increase of free energy of not less
than 100.000 cal. One of these stages will therefore certainly require
more than one quantum of ultra-red light.

pigment too acts as a photocatalyst, namely in the ac-
tivation of the hydrogen atoms in the hydrogen donators.
Apparently these authors accept that this red pigment is acti-
vated directly by the light absorbed; Muller (19332), later
suggests that it may be activated indirectly by light absorbed
by the bacteriochlorophyll.

The absorption spectra of both pigments being known
(Molisch 1907, Buder 1919), a comparison of the quantum
number for different wave lengths probably will permit a
decision in these problems.

The considerations given above will suffice to show the
importance of a determination of the number of quanta invol-
ved in the carbon dioxide assimilation of the Thiorhodaceae,
if possible with different hydrogen donators and for different
wave lengths.

Having outlined in the preceding paragraph the problem
to be investigated, I wish to discuss the experimental methods

In the first place I will consider the methods formerly used
in studying this problem in the carbon dioxide assimilation of
the green plants and which of these methods would be most
suitable for my purpose. It is clear that one must determine
simultaneously, under conditions optimal for the utilization of
the light supplied, the number of molecules of carbon dioxide
absorbed by the bacteria and the number of quanta of a cer-
tain wave length absorbed by the photochemically active
agent.

Which conditions can now be considered to be quot;optimal for
the utilization of the light suppliedquot;?

As has already been remarked, the complete process of the
carbon dioxide assimilation must be seen as a chain of at
least three different groups of processes. Firstly the diffu-
sion of the carbon dioxide and of the hydrogen donators into
the cell and the adsorption of these by the photochemically
active surfaces, secondly one or more photochemical reac-

lions and finally a number of reactions in which the ultimate
products of the photochemical process undergo further con-
versions. Now the velocity of a chain process is mainly de-
termined by the slowest reaction participating in it. After
Blackman (1905) has applied this theory to the study of the
physiology of plants it is known as the quot;principle of limiting
factorsquot;. In particular the investigations on the carbon di-
oxide assimilation of the green plants by van den Honert
(1930), van der Paauw (1932), James (1934) and others, ha-
ve demonstrated that the photochemical part of the carbon di-
oxide assimilation process can be studied by investigating this
process as a whole, provided light is the quot;limiting factorquot;. This
means that the velocity of the photochemical process should
not be limited by a low rate of a preceding or of a subse-
quent physical or chemical process. These conditions are reali-
zed, if small changes of one of the external factors (except the
light) do not influence the rate of assimilation at all. Changes
in the intensity of light, however, ought to cause a proportional
change in the assimilation velocity. For instance slight varia-
tions of the carbon dioxide concentration, of the pH, of the
temperature, of the concentration of the hydrogen donators
etc. should not exert any influence. Moreover the conditions of
cultivation and the age of the organisms used must be optimal
since also quot;internal conditionsquot; can exert an influence upon
the velocity of the carbon dioxide assimilation (van der Paauw
1932). So Warburg and Negelein (1923) point out that it is
advantageous to use organisms that have been grown in feeble
light.

For the determination of the number of carbon dioxide mo-
lecules assimilated I followed the method of Warburg and
Negelein (1923) using the differential manometers previously
mentioned. Other methods for the determination of the carbon
dioxide assimilation mentioned in literature appeared to be
either too inaccurate, or not applicable in the case of cell
suspensions. It is evident that methods based on a determina-
tion of the quantity of carbohydrates, c.q. oxygen produced in
the assimilation process, could not be applied either.

It seems necessary to give here due attention to objections
raised against the method of Warburg and Negelein, as ap-

Van den Honert (1930) objected to the application of
cell suspensions because of the unequal illumination of the
cells, which is inevitable even when dilute suspensions are
used. Sound as this argument may be under certain condi-
tions, this unequal illumination does not interfere with the
study of the photochemical part of the process, since the
applied intensities of light have to be so small that even in the
best illuminated cells the light intensity is the limiting factor
in the carbon dioxide assimilation process. That it is possible
to fulfil this requirement follows directly from the observa-
tion that under suitable conditions the assimilation is indeed
proportional to the light intensity.

I also will mention two of the objections raised by Briggs
(1929) to the procedure of Warburg and Negelein. In the
first place Briggs is of opinion that the carbon dioxide as-
similation measured was too small in comparison with the
respiration, which predominated strongly. Briggs points out,
thai whenever the respiration would be influenced by light
(see van der Paauw 1932) this would lead to a relatively
great error in the calculation of carbon dioxide assimilation
from the difference between the decrease of pressure in the
light and in the dark. It seems difficult to overcome this ob-
jection, since the only way to increase the assimilation would
be the use of higher light intensities, but in all probability
this would accordingly have had a stronger influence on the
respiration too and no improvement would result. Moreover
this remedy would interfere with the necessity to use light
which still is in minimum. Since all points to the probability
that the autofermentation of the Thiorhodaceae is but very
little or not at all influenced by the light intensities used (see
page 44), I do not think this objection of any importance as
far as my own experiments are concerned.

Another of the objections made by Briggs bears upon the in-
termittent illumination (10 min. light, 10 min. darkness) as ap-
plied by Warburg and Negelein. The establishment of the gas
equilibrium between the suspension liquid and the gas phase
would be so slow that, notwithstanding the vigorous shaking
of the vessels, it would have levelled the observed differences

between the changes of pressure in the periods of light and of
darkness. As I felt this objection more or less justified I have
deviated on this point from the procedure of Warburg and
Negelein. I always prolonged the periods of light and dark-
ness to such an extent that the changes of pressure per unit
of time in such a period became constant.

I now come to the question how to determine the number of
quanta absorbed by the photochemically active agent. From
a perusal of the literature 1 concluded that here also the me-
thod of Warburg and Negelein (1923) was the most suitable
one. The methods applied by the greater majority of the other
investigators for the measurement of the light absorbed, could
not be used when cell suspensions instead of leaves or pieces
of thallus were to be employed.

In the method of Warburg and Negelein (1923) practi-
cally all the light falling on the cell suspension is absorbed
because very dense suspensions are used. If the part of the
light that is absorbed by the suspension liquid and by the pho-
tochemically inactive parts of the cells, may be neglected, the
quot;number of the quanta absorbed by the photochemicallly ac-
tive agentquot; may be assumed to be equal to the number of quan-
ta falling on the bacterial suspension. The amount of radiant
energy which is actually lost cannot be determined, either by
by experiment or by calculation; compare Briggs (1929) and
Seybold (1933). The latter estimated the fraction absorbed
by the colourless parts of green leaves at 10% of the whole
amount absorbed. This fraction will be considerably smaller
in the case of a suspension of green algae as this approxima-
tes a suspension of chloroplasts. A further postulate is, that
the fraction of the light that gets lost for the photochemical
process by its conversion into heat, or into fluorescence energy,
even after its absorption by the photochemically active agent,
may be neglected. In consequence of all this one will always
find a larger number of quanta per carbon dioxide molecule
than is indeed used for its reduction. From the results of
Warburg and Negelein it appears that this surplus has not
necessarily to be large; these investigators have obtained re-
sults which are 50% higher than thermodynamically required
and but 10% higher than is expected by some investigators

from a photochemical point of view (see p. 106). Even when
taking into account a loss of 50% or 100%, it remains possible
that for the carbon dioxide assimilation of the Thiorhodaceae
one might obtain numbers less than four. For the reasons set
forth in the foregoing paragraph this would already mean a
satisfactory result.

As has been mentioned previously 1 used bacteria which did
not contain sulphur, in order to make the error caused by the
dispersion of the light by the cells as small as possible. Never-
theless the quantity of dispersed light appeared to be rather
considerable. Consequently this amount has been measured in
the way described and taken into account in the calculations
made.

In the preceding paragraph already a survey has been given
of the factors to be taken into account if one wishes to study
the carbon dioxide assimilation of certain organisms under
conditions optimal for the most efficient utilization of radiant
energy. I will now deal with the consequence of the applica-
tion of these principles on the Thiorhodaceae.

In the first place it is desirable to use very young and vigo-
rous bacteria, whilst furthermore the carbon dioxide assimi-
lation has to be studied in conditions as much quot;physiologicalquot;
as possible, i.o.w. in a medium which is more or less identical
with the culture medium. Thus the following precautions were
always taken. The pH of the suspension liquid was adjusted
to that of the medium in the culture bottle from which
the bacteria were collected; the concentrations of NaCl and
NaHCOg used were made practically equal to those in the cul-
ture media, whilst the temperature in the experiment was the
same as that in the light cabinet in which the bacteria had
been grown.

As 1 had already observed, that a prolonged stay m the
WARBUBG-vessels had an unfavourable influence upon the
assimilatory ability of the bacteria, the duration of the expe-
riments was kept as short as possible.

culture media (c.f. pag. 54) it is self-evident that for the ex-
periments in this Chapter I only used bacteria cultivated in the
inorganic culture medium.

Whilst for the experiments mentioned in Chapter IV I could
use bacteria, which had been subjected to a quot;starvation
periodquot;, this was no longer allowed here, since the assimila-
tory capacity as a rule was appreciably diminished by this
treatment. The consequence of the use of fresh bacteria,
however, was, that the auto-assimilation began to play a much
more prominent part. This was still promoted by the fact
that I had to use dense suspensions of bacteria (± 13 mm^
bacteria per cm^). In consequence not only the auto-assimila-
tion came to the front, but also the carbon dioxide production
became extremely high. This was the more troublesome, sin-
ce the assimilation velocity had to be kept low. It was namely
imperative to apply low light intensities only (compare the
considerations on page 110). Preliminary experiments showed
that only with very low light intensities a region was found
in which there was a direct proportion between the light in-
tensity and the rate of assimilation. It may be remarked, that
in these circumstances photosynthesis does not manifest itself
by a decrease of pressure but by a slower rate of increase.

After having ascertained, that the use of lower temperatu-
res, as e.g. 25° C., did not alter the relation between the value
of the autofermentation and that of the assimilation in a
favourable sense, I performed all further experiments at 35°C..
This also was the temperature at which the bacteria were cul-
tivated.

With some experiments I first proved that lowering the
carbon dioxide concentration in the gaseous phase from 5%
to 3% did not have the least influence upon the rate of assi-
milation, provided the pH was kept constant by a suitable
lowering of the bicarbonate concentration. From this result
it may be concluded that in my experiments neither the con-
centration of the carbon dioxide nor that of the carbonate or
bicarbonate ions have been limiting factors.

The accumulation of unfavourable factors mentioned abo-
ve, forced me to give up the study of a number of questions
mentioned in § 2 of this Chapter. For instance it proved to

be impossible, to make the bacteria use exclusively the hy-
drogen donators added, since assimilation only on the dona-
tors produced in the autofermentation proceeded with the
same rate as after addition of hydrogen donators to the me-
dium. Even an addition of HaS did not cause the least acce-
leration of the assimilation velocity. Yet as has been previously
shown, H2S was the donator which caused a more rapid assi-
milation than whatever other oxidizable sulphur compound
or organic substance. Of course it remains possible, that the
bacteria will have used HgS besides the self-produced hydro-
gen donators because this compound was present in excess.
However, a definite answer as to the relative quantities of HgS
and of fermentation products used as donators cannot be gi-
ven.

In these circumstances I was forced to use the unknown
self-produced substances as hydrogen donators and to deter-
mine the number of quanta required for this reaction with
an unknown reactant. Yet I did not feel this a reason to give
up its determination. For a first approximation it seems of se-
condary importance from which donators the hydrogen atoms
necessary for the reduction of the carbon dioxide are obtained.
With a view to the efficient utilization of the radiant energy
during the auto-assimilation, which was apparent from the
impossibility to accelerate the carbon dioxide assimilation by
adding HgS, one may conclude that for the reduction of a
carbon dioxide molecule certainly no more quanta will be
necessary than if HgS had acted as a hydrogen donator. This
implies, that also with the auto-assimilation one quantum of
the visible or neighbouring infra-red light already will be
amply sufficient to meet the thermodynamic demands of this
reaction. In view of a further answering of the questions po-
sed, it is to be regretted that for the reason mentioned above,
I was incapable to test different hydrogen donators.

I wish to give a more detailed description of the experiments
than has hitherto been done with a view to the far-reaching
conclusions, which might be drawn from them.

Young cultures in the inorganic medium were centrifuged
and the bacteria washed and suspended in the ordinary oxy-
gen-free 2% salt solution with about 0,5% Na-bicarbonate, in

equilibrium with nitrogen containing 5% carbon dioxide. The
suspension came into a manometer vessel (vide p. 24) coated
with silver on the outside (bottom excepted) and containing
the same gas mixture. The filled manometer was placed into
the water bath. The velocity of the shaking of the manometer
was such that a permanent equilibrium between the suspension
and the gas phase was maintained, which could easily be con-
trolled by varying the speed of rotation. The number of turns
in my experiments was usually 500—600 per minute. As soon as
the increase of pressure per unit of time, due to the produc-
tion of carbon dioxide by autofermentation, had become con-
stant, I calculated how many mm^ carbon dioxide were produ-
ced per hour. Then the suspension was exposed to monochro-
matic light of known intensity and when the change in pressu-
re per unit of time had grown constant, (usually after 20 min.)
the increase in pressure per unit of time was determined again.
Usually the carbon dioxide production in the dark was de-
termined once more and thereupon the suspension was illu-
minated again, but now with a lower light intensity, in order
to control whether the light was indeed the limiting factor.

The quantity of carbon dioxide, which was assimilated per
unit of time and which is mentioned in table 16, thus was
calculated in the assumption, that the difference of the in-
crease of pressure per unit of time in the dark and in the
light quantitatively was due to the assimilation of carbon
dioxide. A possible formation or disappearance of acids was
thus neglected (see p. 69). Moreover a slight error has been
introduced in the calculation of the number of mm^ carbon
dioxide taken up, since vessel constants were used which had
been calculated in the assumption that the solubility of car-
bon dioxide in the suspension liquid did not differ from that
in distilled water.

Actually this is not quite correct, since the solubility of CO2
in water is reduced a little by the addition of 2% NaCl. As is
mentioned on p. 34, the error due to this cause can be esti-
mated to be 2%, i.o.w. the figures given for carbon dioxide
production or uptake are all 2% too high.

Finally a part of the carbon dioxide assimilation had to
be ascribed to quot;false lightquot;, not measured with the spectral

pyrometer. As explained on p. 26, this additional assimilation
was determined by means of special dye solutions or coloured
glasses. The sodium light used contained admixtures appa-
rently causing an assimilation, which amounted to ± 8% of

The necessary corrections being introduced, I knew, how
many molecules carbon dioxide were assimilated by the bac-
terial suspensions under conditions, which were optimal for
a most efficient utilization of the radiant energy.

It only remained to determine how many light quanta we-
re absorbed by the suspension per unit of time. I determi-
ned the intensity of the light, which fell upon the suspen-
sion, by means of the spectral pyrometer as described on p.
27 This enabled me to measure the intensity of the hght in
eres/cm2/sec. That part of the light, which was reflected by
the suspension, was determined after each experiment, using
the method, described on pag. 29. The light transmitted by
the bacteria could be safely neglected, of which 1 convinced
myself by means of the spectral pyrometer. Moreover in the
experiment itself the greater part of this light was reflected
again in the suspension by the silver coating.

As already mentioned before (p. 25) the determination of
the light intensity before and after the experiment, always
showed that the sodium lamp burned with a very constant
intensity. This, however, was not the case with the mercury
vapour lamp; variations of 5 to 10% in light intensity were
not rare. In connection with the very low photochemical ef-
ficiency of the monochromatic yellow, green and blue light
furnished bv the mercury lamp, it seemed superfluous to
take special' precautions to prevent these variations. Conse-
quently the data in table 16 No. 4-7 are less accurate than
those which refer to the intensity of the sodium light.

Assuming that the light had been constant during the en-
tire period of illumination and after the introduction of the
necessary corrections, I could calculate how many ergs had
been absorbed by the suspension per unit of time. This value
was divided by the number of ergs of one quant of the wave
length used, and thus the number of quanta absorbed per

Number of light quanta of different wave lengths required for the assimilation of one car-
bon dioxide molecule.

unit of time was obtainedi). This was divided by the number
Tmlcules of carbon dioxide assimilated per umt of time,
1 qurent gave the value of the number of quanta absor-
bed by the suspension per molecule carbon dioxxde ass.müa-

■V consideration of the results given in table 16 shows in
th; fSrUce that the number of quanta found to be neces-
ary for the reduction of one carbon dioxide molecule vanes
a good deal. It is impossible, that this should be caus d by
a different absorption of the light by

or by the quot;colourlessquot; parts of the protoplasm. In tha ca e
lo Wakburg and Negelkin (1923) ought to have lound with
the same method similar deviations in the assimila ion of
\ r^^inxide bv Chlorella. Obviously the cause of these
^«er^ncriiÏ inbsp;absorption by the bacterio-erythri-

ne the red pigment present in the bactena used
quot;V comparisL of the absorption spectra of thejreen an^
the red pigment of the Thiorhodaceae as given by Mousch
a907) and Buoer (1919) shows that the wave eng^s wi^
low assimilatory efficiency for the greater part will have

Reoarding the slight difference m wave length ot the so
Regarding me gnbsp;^^^^^ ^^ ^^^^ mercury lamp

urnrTs^gly large. It seems probable that the explanation of
hL ^ust be found in the following. In the region xn question
the b^cteriochlorophyll shows a very distinct absorption band,
^ rhTever ends abruptly between 589 and 578 m^. This
Wiés th^^^^^^^^nbsp;a Jul larger portion of the light will

TaCrbed by the red pigment, the more so, because this pig-
Ipnt has a strong absorption band m this region.

by the bacterio-erythrine is lost for the carbon dioxide assi-
milation, at least with the hydrogen donators at issue (e.g.
unknown metabolic products, H2S, NagSOg).

The number of quanta of the wave length 589 m/i needed
for the assimilation of one molecule of carbon dioxide by the
suspension, is many times higher than might be expected on ac-
count of exclusively thermodynamical or photochemical con-
siderations. However, it seems probable that also from this
wave length a considerable part has been absorbed by the red
pigment. In order to make an estimation of this, I extracted
in the simple way as indicated by Molisch (1907) the green
pigment of a given quantity of bacteria, viz. with absolute
ethyl alcohol. Thereupon I dissolved the red pigment from
the same bacteria in an equal quantity of chloroform. The
extinction coefficients of the green and of the red solution
for the wave length 589 m^u which I determined with the aid
of the spectral pyrometer, were in the ratio 1,2 : 1. As un-
doubtedly the separation of the two pigments has not been
complete, in reality this ratio will be higher. Furthermore the
absorption bands in solutions of alcohol and chloroform may
possibly be shifted a little as compared with the bands in the
living bacteria.

On the assumption that the light absorbed by the red pig-
ment is lost for the assimilation and accepting furthermore
that the ratio 1,2 : 1 can be applied for the conditions in the
cell also, the number of quanta absorbed by the green pigment
is 3,6—4,4. Considering the incomplete separation of the
pigments by the method of extraction used, the number of
quanta absorbed by the green pigments in the cell, will be
somewhat higher.

It seems to me, that these results point to the necessity of
four quanta per molecule carbon dioxide and certainly one
may conclude from it, that the number of one quantum per
molecule carbon dioxide, required thermodynamically, is en-
tirely insufficient.

Further experiments, however, will be necessary before
definite and reliable conclusions may be drawn. Then it will
be advisable to determine more accurately the absorption
spectra of both pigments of the purple bacteria. If possible

the assimilation experiments should be made with those wave
lengths, which only are absorbed by the bacteriochlorophyll.
As the absorption band of this pigment, near the D-line, is
the only one in the visible spectrum and since the red pig-
ment apparently has a not negligible share in the absorption
in this region, it is probable, that one will have to have re-
course to monochromatic infra-red light.i)

Furthermore it does not seem impossible that by using
other bacteria, e.g. Athiorhodaceae or other strains of Thio-
rhodaceae, one will not be hindered by the auto-assimilation
and the autofermentation to such an extent as in my experi-
ments. If so, different hydrogen donators could be applied.

Finally, judging from our — it is true, limited — know-
ledge of the metabolism of the green sulphur bacteria, it must
have great advantages, to study the problem under discus-
sion on these organisms. In this case one gets entirely rid of
the disturbing influence of the absorption of light by the bac-
terio-erythrine. Furthermore in all probability there will be
no question of auto-assimilation in this case, since up to the
present hydrogen sulphide has appeared to be the only sui-
table hydrogen donator for the carbon dioxide assimilation
of these organisms. It seems, however, difficult to cultivate
them.

In a preliminary communication (1934), I came to the con-
clusion that, in view of the good growth of Thiorhodaceae in
sodium light with sodium thiosulphate as a hydrogen donator,
the light absorbed by the bacterio-erythrine is lost for the
carbon dioxide assimilation of the Thiorhodaceae. This was
contradictory to the hypothesis of van Niel and Muller (1931)
according to which the activation of the bacterio-erythrine by
light absorption would be essential for the activation of the
hydrogen atoms in the donators.

My conclusion was based on the assumption, that sodium
light would practically only be absorbed by the bacterio-
cblorophyll. This conclusion seemed warranted by the ab-
sorption spectra of these pigments, as given by Molisch
(1907) and by Buder (1919). Since, however, we have seen
that the absorption of sodium light by the bacterio-erythrine
cannot be neglected at all, the mentioned growth experiments
fail to uphold the conclusion in question.

On p. 119, however, it is set forth that there are other reasons
why it seems extremely improbable that bacterio-erythrine
is active as a pbotocatalyst in the carbon dioxide assimilation

Of course it remains possible that — as was suggested by
Muller (19332) — an indirect activation of the bacterio-
erythrine takes place by radiant energy absorbed by the bac-
teriochlorophyll.

In this Chapter some theoretical considerations have been
given about the importance of the determination of the num-
ber of quanta in the carbon dioxide assimilation in general
and of this number in the carbon dioxide assimilation of the
Thiorhodaceae in particular. The experimental results obtai-
ned give strong indications that the light absorbed by the
bacterio-erythrine under the conditions of my experiments
is lost for the carbon dioxide assimilation. Furthermore it
seems probable that also for the carbon dioxide assimilation
with the most suitable hydrogen donators not less than four
quanta are required for the reduction of one molecule of car-
bon dioxide.

A study was made of ttie metabolism of the carbon dioxide
assimilation of the Thiorhodaceae or purple sulphur bacteria,
with the aid of the manometric method.

It appeared that these anaerobic bacteria can only main-
tain themselves in periods of darkness on a fermentation of
an_as yet unknown — reserve food substance. The produc-
tion of carbon dioxide and acids (under special conditions also
of hydrogen) could be demonstrated.

The conceptions of van Niel (1931) and Muller (1933i)
regarding the metabolism of purple sulphur bacteria in cer-
tain inorganic and organic media were fully corroborated by
the results of the manometric method. In addition it was
shown that as yet unknown products of the autofermentation
and also gaseous hydrogen can also act as hydrogen donators
for the carbon dioxide assimilation. Whereas carbon dioxide
assimilation with hydrogen as a donator occurred in the dark
as well as in the light, growth occurred only in the latter case.

The carbon dioxide assimilation, caused by the illumination
of bacterial suspensions, was reflected in the oxidation-reduc-
tion potential occurring in the medium.

Experiments with monochromatic light of known intensity
showed that the light, absorbed by the red pigment of the
purple sulphur bacteria, is of no use for the carbon dioxide
assimilation.

It has been found that more than one quantum — probably
four quanta — have to be absorbed by the green pigment for
the assimilation of each carbon dioxide molecule. The im-
portance of a further extension of these experiments for our
insight into the mechanism of the photochemical carbon di-
oxide reduction in general was stressed.

For a more extensive survey of the results obtained the rea-
der is referred to p. 62 and p. 101.

The investigations were carried out in the Microbiological
Laboratory, Technical University, Delft. I am very much in-
debted to Prof. Dr. Ir. A. J. Kluyver for his thorough and in-
dispensable help and criticism.

Baars, J. K., Over sulfaatreductie door bacteriën. Diss. Delft 1930.
Blackman, F. p., Ann. of Botany 19, 281. 1905.

Bonhoeffer K. F. u. p. Habteck, Grundlagen der Photochemie. 1933.
Boussingault, J. B., Agron., Chim. Agric. et Physiol., Vol. Paris,
1868.

Buder, J., Jahrb. f. wissensch. Botanik 58, 525. 1919.
Burns, G. R., Plant Physiology 8, 247 and 564. 1933.
Deines, 0. von, Naturwissenschaften 21, 873. 1933.
Dhar, N. R., Journ. Indian. Chem. Soc. 11, 145. 1934; Nature Ï34, 331.
1934.

Dixon, M., Manometric methods. Cambridge. 1934.
Eilers, H., Ree. tr. botan. Néerl. 23, 362. 1926.

Elema, B., De bepaling van de oxydatie-reductiepotentiaal in bacteriën-
cultures en hare beteekenis voor de stofwisseling. Diss. Delft 1932.

Ellis, D., Sulphur bacteria, a monograph. London. 1932.
Emerson, R., Journ. Gen. Physiol. 10, 469. 1927.
EwART, A. J., Journ. Linn. Soc. Bot. 31, 364, 554. 1896.
_, Ann. of Botany 12, 363, 415. 1898.

Eymers, Erl. J. G., L. S. ornstern u. D. Vermeulen, Zeitschr. f. Physik.
75, 575. 1932.

Fischer, H. und J. Hasenkamp, Liebig's Annalen, 515, 148. 1935.
Gaffron. H., Abderhalden's Hdb. Biol. Arb. Meth. Abt. XI, T. 4, H. 1.1929.

GrNSRURG-karagitschewa, I., Centralbl. f. Bakt. II, 86, 1. 1932.
Haber, F. u. R. WiLLSTäTTER, Ber. D. Chem. Ges. 64, 2844. 1931.
Hill, S. E., Science 67, 374. 1928.

Honert, T. H. van den, Ree. tr. botan. Nécrl. 27, 149. 1930.
James, W. O., The new phytologist 33, 8. 1934.
Kendall, E. C., Science 73, 394. 1931.

Kingma Boltjes, T. Y., Onderzoekingen over nitrificeerende bacteriën.
Diss. Delft. 1934.

Kluyver, A. J. u. H. J. L. Donker, Chem. d. Zelle u. Gew. 13, 134. 1926.
-, Arch. f. Mikrob. 1, 181. 1930.

Koch, F. E., Centralbl. f. Bakt. I Originale, 132, 358. 1934.
Krebs, H. A., in C. Oppenheimer, Die Fermente und ihre Wirkungen,
3, 635, 1929.

Lipmann, F., Bioch. Zeitschr. 268, 206 and 274, 329, 412. 1934.
Mack, W. B. and B. E. Livingstone, Botanical Gazette 94, 625. 1933.
Migula, W., System der Bakterien, II, 1900.

Pantanelli, E., Jahrb. f. wiss. Botan. 39, 167. 1904.
Paauw, F. van der, Ree. tr. botan. Néerl. 29, 497. 1932.
PiNcussEN, L., Bioch. Zeitschr. 272, 357. 1934.

Ritter, W. u. W. Dorner, Centralbl. f. Bakt. I Originale, 125, 379. 1932.
Roelofsen, P. A., Proceedings, Amsterdam, 37, 660. 1934.
Ruinen, J., Ree. tr. bot. Néerl. 30, 725. 1933.
Schmucker, Th., Jahrb. f. wiss. Botanik, 73, 824. 1930.
Schneider, E., Zeitschr. f. physiol. Chemie, 226, 221. 1934.
Schürer-Waldheim, M., Chem.-techn. Rezept-Taschenbuch. 1921.
Seybold, A., Planta 18, 479, 1933.
Shibata, K., Naturwissenschaften 21, 267. 1933.

onder den invloed van het organische leven. Diss. Delft 1906.
Spoehr, H. A., Photosynthesis. New York. 1926.
Stephenson, M. and L. S. Stickland, Bioch. Journ. 27, 1517. 1933.
Stern, K„ Pflanzenthermodynamik. Monogr. Gesamtgeb. Phys. Pfl. u.

Treadwell, F. P., Kurzes Lehrbuch d. analyt. Chemie. Bd. II. 1930.
Ursprung, A., Ber. D. bot. Ges. 35, 44. 1917.
Warburg, 0., Bioch. Zeitschr. 100, 230. 1919.

Watanabe, A., Acta phytoehimica 6, 315. 1932.
Wurmseb, R., Compt. Rend. Ac. Sc. Ï77, 644. 1923.

In de purperen zwavel bacteriën speelt zich bij voortduring
een zdfgisttag af; de eindproducten hiervan worden m het
licht weer geassimileerd.

De meening van GAFFRON. dat de Thiorhodaceae in staat
zijn in het donker sulfaten te reduceeren, is onjuist.

De assimilatie van een koolzuurmolecuul door de Thiorho-
daSae verefscht naar alle waarschijnlijkheid de absorptie van
vier hcht-quanten door het bacteriochlorophyll.

De rangschikking van de micellen in de celwanden van
Phycomyces Blakesleeanus geschiedt onder den invloed van
de protoplasmastrooming.

De onderzoekingen van WiLLSTATTER en ROHDEWALD
over de amylasen der leucocyten brengen een sterk ekmen
^In onzekerheid in de uitkomsten van een groot deel der tot
dusver gepubliceerde enzymologische onderzoekingen.
R. Willstatter und M. Rohdewald, Zeitschr. f. physiol. Chem. 221. 13, 1933.

Bii de bepaling der enzymatische activiteit dient men ook
rekening te houden met de oxydatie-reductiepotentiaal van

De micro-pedologische bestudeering van in den bodem
voorkomende ziekteverwekkende organismen is van grooter
belang voor de Phytopathologie dan een onderzoek van de
physiologie dier organismen in reincultuur.

De samenstelling van de recente flora van Zuid-Amerika,
Afrika Australië en Nieuw-Zeeland is tot nu toe slechts te
verklaren volgens het beginsel van de theorie van Wegener
over het ontstaan der continenten.

Het gehalte aan kiemen van Azotobacter chroococcum is
geen maatstaf voor de vruchtbaarheid van den bodem.

Naar alle waarschijnlijkheid wordt de cacao-fermentatie
ingeleid door Bacterium aerogenes.

Het met meer succes volgen van het hooger onderwijs in
de natuurwetenschappen door abituriënten van het gymnasium
in vergelijking met die der Hoogere Burgerscholen wettigt nog
niet de conclusie, dat het gymnasium de meest geschikte voor-
opleiding voor de natuurwetenschappelijke studie aan de
universiteiten biedt.

		h
		/
		/
	h


^- gt;	1
/
V
'1		n
n\-			houpâ

Exp. No,	Description of experimental conditions	DifFe- rential mano- meter	Starvation period	CO2 produc- tion in mm® per 1 cm® bacteria per hour.i)
1.	200 mm^ bacteria of strain d, cul- tivated in the inorganic medium, suspended in 2% NaCl, 0,5% Na- HCOs, in equilibrium with N2/7% CO2, temperature 30 °C.	1 3	no St. per. no St. per.	1300 1100
2.	Conformable to experiment no. 1. 260 mm^ bacteria p. manom. vessel	1 3	no St. per. no St. per.	1050 950
3.	Conformable to experiment no. 1. 170 mm^ bacteria p. manom. vessel	1 3	no st. per. no St. per.	1120 1120
4.	Conformable to experiment no. 1. 160 mm® bacteria p. manom. vessel	1	ISh.at 30°C	825
5.	Conformable to experiment no. 1. 180 mm® bacteria p. manom. vessel	1 3	ISh.at see	710
6.	Conformable to experiment no. 1. 290 mm^ bacteria p. manom. vessel	1	20h.at 35°C	475
7.	Conformable to experiment no. 1. 220 mm® bacteria p. manom. vessel	1	24h.at 35°C	410
8.	Conformable to experiment no. 1. 220 mm® bacteria p. manom. vessel	1	24h.at 35°C	390
9.	Conformable to experiment no. 1. 200 mm® bacteria p. manom. vessel	1 3	24h.at 36°C 24 h.at 36''C	360 360

			CO2 production in mm® per hour.
Experiment No.	Description of experimental conditions	Mano- meter	Without any addition	Added substance	After addi- tion of the mentioned
			Without any addition		substance
1.	Bacteria of strain d, cultivated in the inorga-	D. M.
	nic medium, suspended in distilled water with	3	400	—	—
	2% NaCl p.a. and 0,5% NaHCOs p.a., in equi- librium with N2/5% CO2, temperature 30°C.	I	—	0,3% glucose	400

2.	Bacteria of strain d, cultivated in the standard	D. M.
	solution with 0,1% Na-butyrate, 0,1% Na2	HI	270	__	—
	S2O3 and 0,7% NaHCOs, suspended in the	IV		0,2% Na-butyrate	270
	same solution as used in experiment No. 1.			0,2% Na-butyrate
3.	Conformable to experiment No. 1.	D. M.	1.50
	Conformable to experiment No. 1.	IV	1.50	—	-
		III	—	0,1% Na-lactate	155
4.	Conformable to experiment No. 1.	S. M.
	Conformable to experiment No. 1.	4	128	—	-
		1	_	0,3% Na-malate	125
				S 0,3% Na-malate ( ) 0,2% Na2S04 S	130
		u		S 0,3% Na-malate ( ) 0,2% Na2S04 S	130
5.	Conformable to experiment No. 1.	D. M.		0,2% Na-malate	270
	Conformable to experiment No. 1.	1	285	0,2% Na-malate	270
		3		0,2% Na-malate	245
		q		S 0,2% Na-malate; t 0,5% Na2S04 S	245
		0		S 0,2% Na-malate; t 0,5% Na2S04 S	245
6.	Conformable to experiment No. 1.	D. M.		0,2% glucose
	Conformable to experiment No. 1.	1	__	0,2% glucose	270
j				. 0,2% glucose ( j 0,5% Na2S04 S	260
1 1		1		. 0,2% glucose ( j 0,5% Na2S04 S	260

			Change in pressure in mm. Brodie per hour calcu-
			lated for equal volumina, caused by hydrogen
Exp. No.		Mano- meter	and carbon dioxide production.
Exp. No.	Description of experimental conditions	Mano- meter	Without any		With the addition
			Without any	Added substance	of the mentioned
			addition		substance
1.	Bacteria of strain 9, cultivated in the peptone medium suspended in distilled water with 2% NaCl p.a., 0,5% NaHCOs	D. M. 3	-1-28	—	—
	p.a., in equilibrium with N2/5% CO2, temperature 30 °C.	1	—	) 0.6% Na2S04 ^	30
	p.a., in equilibrium with N2/5% CO2, temperature 30 °C.			} 0,2% Na-malate S	30
2.	Conformable to experiment No. 1	S. M.
	Conformable to experiment No. 1	II	41	—
		V	-	0,3% Na-malate	40
		VI	-	SO,3% Na-malate lt;	39
				gt; 0,5% Na2S04 ^	39
3.	Conformable to experiment No. 1	S. M.
	Conformable to experiment No. 1	IV	31	—	—
		V	—	0,3% Na-butyrate	31
		VI		J 0,5% Na2S04 f.	32
				3% Na-butyrate'	32

		■
		//	/
		//
	//
	//
/
/
/
J			Minutes
0 10 1		0 c.	0

Strain d inoculated in 60 cm^ glass- stoppered bottles, completely fil- led up with a sulphate-free stan- dard salt solution with 0,005% NaaS, and with addition of the fol- lowing substances.	Incubated in the dark at a temperature of 30°C.	Incubated in the light ca- binet at a temperature of 30°C.
1. 0,1% Na-butyrate	—	—
2. 0,1% Na-butyrate and 0,1% Na-thiosulphate	--	-
3. 0,1% Na-butyrate and 0,1% Na-sulphate	—	—
4. 0,1% Na-butyrate and 0,3% Na-bicarbonate	—	-f
5. 0,1% Na-butyrate and 0,1% Na-malate	—	-t-
6. 0,1% Na-butyrate and 0,05% Na-formate	—
7. 0,1% Na-malate	—■

Exp. No.	Description of experimental conditions	Single mano- meter no.	Substances present in side- bulbs of the manometer vessel	Change in pressure in mm. Brodie per hour in equal volumina.
I.	Bacteria of strain a, cultivated in the	V	Alkali.	15
	peptone medium, suspended in tap-water	VI	Alkali.	13
	2% NaCl, in equilibrium witb N2, tem-	IV	Alkali and Pd black.	0
	perature 30 °C.			0
2.	Conformable to experiment no. 1.	IV	Na-pyrogallate, excess alkali.	13
		III	Na-pyrogallate, excess alkali.
			Pd black.	— 1
		I	Na-pyrogallate, excess alkali.
			Pd black.	— 1
3-	Conformable to experiment no. 1.	IV	Na-pyrogallate, excess alkali.	16
		VI	Na-pyrogallate, excess alkali
			and Pd black poisoned by
			HnS.	15
		I	Na-pyrogallate, excess alkali.
			Pd black.	0
4.	Conformable to experiment no. 1.	VI	Alkali.	14
		I	Alkali, Pd black, poisoned by
			H2S.	13
		III	Alkali, Pd black, poisoned by
			H2S.	13
5.	Conformable to experiment no. 1.	I	Alkali.	17
		V	Alkali.	17
		IV	Alkali, cupro oxide in ammo-
			nia.	18
		VI	Alkali, cupro oxide in ammo-
			nia.	16



1							s.

		ysô	m. 3t	m^ H 9 6	\PPOÙ '0 S à	fc/ceé '0 1.	f ÎO m
	A	•1			/
		r	-	/	r—	Fiç.
		1	1	! 1	1 1	1 1
/			1	! 1	1 1	1 1
/	1		1 1				1

Exp.	Description of experimental	D.M.	H2 produc- tion in the	H2 uptake in the light
No.	conditions	D.M.	dark in mm® per hour	in mm® per hour
1.	Bacteria of strain a, cultivated in the peptone medium, suspended in:
	a. tap-water 2% NaCl, in equi- librium with Na. Manometer no. I. Alkali present in side- bulb.	I	31	37
	b. tap-water 2% NaCl, with 0,05 molar CO2 buffer mixture of 9 parts NaHCOs and 1 part NaaCOs. Manometer no. II.	II	32,5	66
	Temperature 30° C. Light source incandescent lamp.

			Compound added.	CO2 assimilation in mm®		oBgo Squot;« 3 0 ® 3 » quot;
Exp. No.	Description of experimental conditions.	D.M.	either at the be- ginning, or in the second part of	per Control	hour. Total	— w clp Q'/! Sw i «8 1
			the experiment	auto-	assimi-	— w clp Q'/! Sw i «8 1
			the experiment	assimilation	lation	3. 0 2 Ko^j a
1.	Bacteria of strain d, cultivated in the inorganic me-	1	_	280	_	_
	dium, suspended in tap-water 2% NaCl, 0,5% Na- HCOs, in equilibrium with N2/7% CO2 at 35°C.	3	0,3% Na-malate	_	310	10
			0,3% Na-malate
2.	Conformable to experiment no. 1.	4	_	305	_	_
		3	0,3% Na-lactate	—	450	30
3.	Conformable to experiment no. 1.	3	0,1% acetaldehyde	—	440 i	70
		4	0,1% acetaldehyde later on	135	450 S	70
4.	Conformable to experiment no. 1.	3	0,1% ethyl alcohol later on	250	255	0
5.	Conformable to experiment no. 1, except the use of bacteria cultivated in the peptone medium.	3	__	160	_
		1	0,2% Na-malate	—	330	50
6.	Conformable to experiment no. 1, except the use of bacteria cultivated in a medium with, 0,2% Na-	3	0,1% Na-malate
			later on	270	475	45
	malate and 0,7% NaHCOs.
7.	Conformable to experiment no. 1, except the use of bacteria cultivated in a medium with 0,2% Na-buty- rate and 0,7% NaHCOs.	4	0,1% Na-butyrate later on	2.50	390	35
8.	Conformable to experiment no. 1, except the use of	1	0,2% Na-malate	160	530
	bacteria of strain 9, cultivated in the peptone me-		later on			70
	dium.	2	0,2% Na-malate later on	155	520	70
9.	Conformable to experiment no. 1.	3
	(Representing the ordinary assimilation velocity	1	H2S solution	300	1300	77
	with H2S as a donator.)

Exp. No.	Description of experimental conditions	Ejj in the dark on three electro- des in m. Volt.	in the light on the same electro- des in m. Volt.
1.	Bacteria of strain a, cultiva-
	ted in the peptone medium.
	suspended in tap-water with
	2% NaCl, 0,3% NaHCOs and
	0,2% NaaSOs, in equilibrium
	with N2/5% CO2, tempera-
	ture 30°C. pH 7,5. Light sour-
	ce was an electric bulb of	— 230	— 125
	40 Watt burning on a distan-	— 235	— 130
	ce of 10 cm from the suspen-	— 240	— 130
	sion.
2.	Conformable to experiment	— 225	— 140
	No. 1.	— 230	— 135
		— 235	— 135

Experiment No.	Description of experimental conditions.					Wave length in mn	Radiant energy ab- sorbed in erg/cni^/sec.	CO2 assimi- lation in mm® p. hour.	Number of quanta ab- sorbed per molec. CO2
1.	Bacteria of strain d, cultivated in the inorganic medium, at 35°C., suspended in tap-water 2% NaCl, 0,5% NaHCOs, in equilibrium with N2/5% CO2, temperature 35° C.					589	4050	320	7,3
2.	Conformable	to	experiment	No.	1.	589	4050 2250 1550	345 190 140	6,9 6,9 6,6
3.	Conformable	to	experiment	No.	1.	589	2200 1500	175 120	7,4 7,4
4.	Conformable	to	experiment	No.	1.	578	± 5000 ± 2500	140 70	±20 ± 20
5.	Conformable	to	experiment	No.	1.	578	± 6500	140	± 25
6.	Conformable	to	experiment	No.	1.	546	± 6000	60	± 55
7.	Conformable	to	experiment	No.	1.	546 436	± 6000 ± 3000	50 20	± 60 ± 60

			1	1		i /
					/	/
»				/	/
1			/	/		n
^		/	/
}	y
lt;						m

= 1			h				houm

1 . \ /
		jn


	\
	\
		, /JOUPS


N	X	\
	N	A



- i
1
■)\-

7—-

f)-

0
in			hour's

		A'
		n		N	/
				, \	1
				n


1	—-


\ \		\	1
i		t				hours