3Ba^^^^^B^^^^^^^3B3E

Marine Biological Laboratory Library

Woods Hole, Mass.

PrcBented ty

Dr. C. B. Metz

3^^^E

SEX IN MICROORGANISMS

A sympos'miu presented on Deceviber 30, 1951

at the Philadelphia lueeting of the

American Association for the

Advancevieiit of Science

Editorial CoMiMixrEE

D. H. WENRICH,

Chainihw and Editor
IVEY F. LEWIS JOHN R. RAPER

A Publication of the

American Association for the Advancement of Science

1515 Massachusetts Avenue, N.W., Washington, 5, D.C.

1954

Copyright 1954 by

The American Association for the

Advancement of Science

Library of Congress Catalog Card Number 53-12747

PREFACE

In pi.ANMNG the syniposiuiii on which this volume is based, an
attempt was made to present the evidence for "sex" in the principal
groups of micr(K)rganisms, from the viruses through bacteria, func^i,
and unicellular algae to the protozoa. Since the material \vhich could
be presented in the short time afforded by the two sessions devoted
to the symposium was necessarily limited in scope, an effort has been
made in this volume to expand to some extent the ground covered at
the symposium itself.

Dr. Msconti discusses evidence for recombination of "genes" in
\iruses without claiming that the phenomena demonstrate sex for that
group. Dr. Lederberg and Dr. Tatum review genetic evidence for
"sex" in the bacteria, and Dr. Hutchinson and Dr. Stempen describe
cell fusions in certain bacteria. These cell fusions could provide an
opportunity for the recombinations of genes that have been indicated.
Next, Dr. Raper offers a comprehensive coverage of sex in the fungi
and the many variations in sex phenomena found in that group.

Passing on to the lower algae, Dr. Patrick describes syngamy in
diatoms and Dr. Lewin reviews the sexuality of other unicellular
algae, especially the flagellates.

Next is a review of sexual phenomena in the protozoa, with
sections on the Mastigophora, the Sarcodina, the Sporozoa, and the
Ciliophora. At the symposium. Dr. Cleveland reported on his studies
of sexual reproduction among the flagellates living in the gut of the
wood-feeding roach, Cryptocerciis pimctulatus, but he did not find
it possible to provide a paper for the volume. Consequently, a review
of his published papers on that subject follows the section devoted
to the Mastigophora.

After the section on sex in the CiHophora, Dr. Nanney sum-
marizes the status of mating type phenomena in Faramecium aiirelia
and then presents certain new^er interpretations of mating type behav-
ior that have evolved in Dr. Sonneborn's laboratory at Indiana Uni-
versity. This paper is followed by Dr. Metz's discussion of mating
type substances, with a comparison of the conditions found in Fara-
mecium and other ciliates wdth those found in the Metazoa.

At the end of the volume is a short section devoted to comments

iii

iv PREFACE

on the origin and evolution of sex, based primarily on conditions in
the Protozoa, but including questions about all of the microorganisms.
It is hoped that this volume will reveal to readers not only that
"sex" is widely distributed among microorganisms, but that many
groups appear to get along without sexual reproduction, at least
for indefinite periods of time, and that the details of sexual behavior
vary widely among and within the different groups of microscopic
organisms.

D. H. Wenrich

CONTENTS

Cicncric Reconibinarion in l^acrcrial \ iruses N. \^iscoNri 1

Sex in Bacteria: Genetic Studies, 1945-1952

Joshua Lederberg and E. L. Tatum 12

Sex in Bacteria: Evidence from Morphology

W. G. Hutchinson and Henry Spempen 29

Life Cycles, Sexuality, and Sexual Mechanisms in the Fungi

John R. Raper 42

Sexual Reproduction in Diatoms Ruth Pairick 82

Sex in Unicellular Algae Ralph A. Lewin 100

Sex in Protozoa: A Comparative Review D. H. Wenrich 1 34

Mating Type Determination in Faramechnn aurelia: A Study
in Cellular Heredity DAvm L. Nanney 266

Mating Substances and the Physiology of Fertilization in
Cihates Charles B, Meiz 284

Comments on the Origin and Evolution of "Sex"

D. H. Wenrich 335

Author Index 347

Subject Index 354

XK

%$L

Genetic Recombination
in Bacterial Viruses

N. VISCONTI, Carnegie Institution of Washington,
Cold Spring Harbor, New York

At first sight the life cycle of a bacteriophage particle seems simple.
It is adsorbed to a bacterium, and after a characteristic period of time
called the latent period, which is 22 minutes for T2, the bacterium
bursts open, releasing several hundreds of particles identical to the
one adsorbed at the beginning. To the scientist concerned with the
biological problem of self-duplication, this is interesting. In order to
discover how and from what these particles are formed, we must see
what is going on in the bacterium during the latent period. There has
not been developed a method to look inside the bacterium without
disturbing the process of phage formation. Suppose instead that the
bacterium is broken before spontaneous lysis. What is found then?
In 1942 at Vanderbilt University, Delbriick and Luria started
experiments along this line of thinking. The problem was to break
the bacterium without damaging the phage or whatever was to be
found inside. One effective agent was known: that agent was the
phage itself. Suppose a bacterium is infected with two phages, Tl
and T2. Tl has a latent period of 13 minutes, T2 of 22 minutes. The
two phages will start to grow together, but after 13 minutes the
progeny of Tl will supposedly break the bacterium open. What will
happen to the progeny of T2? With this experiment began a scries
of unexpected results w hich led to the discovery of recombination.
In the bacteria infected with both Tl and T2, nothing happened at
13 minutes; at 22 minutes the cells burst, but only T2 was found in
the lysed culture. If, instead of a mixture of Tl and T2, only T2 had
been used, the result would have been the same. The explanation is
that T2 excludes Tl. Both phages are adsorbed to the same bacterium,
but only one can grow. If, instead of infecting simultaneously with
Tl and T2, an advantage of 4 minutes is given to Tl, many bacteria
will yield Tl but not T2. The bacterial culture as a whole may yield

1

2 SEX IN MICROORGANISMS

both phages, but each burst coming from one bacterium will consist
entirely either of Tl or of T2. This effect was called mutual exclusion.
At that time Delbriick and Luria made another pecuhar observation.
If a single bacterium was infected with many identical phage parti-
cles instead of one, the latent period and the yield did not change.
This made it appear that actually only one particle took part in the
growth. The principle of mutual exclusion could thus be extended
to particles of the same strain. As soon as one particle begins to grow
in a bacterium, a reaction is started that results in the exclusion of all
other particles, even if they are already adsorbed. Quoting from Del-
briick: "The mutual exclusion effect is so novel that its explanation
calls for a bold hypothesis. We assume that the first virus which
penetrates the cell wall will make the cell wall impermeable to other
virus particles just as the fertilization of an egg by one spermatozoon
makes the tgg membrane impermeable to other spermatozoa."

In 1944 Luria found the first mutation in phage T2 (Luria,
1945). If a high concentration of bacteria is seeded on an agar plate,
the bacteria will grow in a thin continuous film on the surface. If a
bacteriophage is present on this surface, it will multiply by lysing the
bacteria, producing a colony of phage called a plaque, which appears
like a clear visible hole in the film of turbid bacteria covering the sur-
face of the agar. Strain B of the colon bacillus, sensitive to T2, can
mutate to B/2, resistant to T2. However, if a very large number of
T2 phage particles are plated on B/2, some plaques are obtained.
These plaques are due to T2/; mutants which lyse B/2. Therefore we
have two types of phage: T2 and T^2h, and two types of bacteria:
B and B/2. T2^ lyses both bacterial strains, T2 only B and not B/2.
Suppose we plate on B: T2 cannot be distinguished from Tlh. If
we plate on B/2, only T2/:) will give plaques. As a crucial test for
the theory of mutual exclusion, Luria infected the same bacteria first
with T2 and then with T2h; he found that T2h was excluded. He
concluded that exclusion exists also among closely related phages
like T2 and its mutant b. But the actual result of Luria's experiment
was due to a different phenomenon discovered by Dulbecco several
years later (1952) and called mutual exclusion between related
phages. This exclusion is due to the interval of time allowed between
the first and second infection, and not to exclusion in the previous
sense. The longer this interval of time, the more complete is the
exclusion.

GENETIC RECOMBINATION IN BACTERIAL VIRUSES 3

In 1945 another niurant was found by Hershcy at Washington
University (Hershcy, 1946). This mutant was called r. It gave
much larger plaques than the wild type r+, and it was directly recog-
nizable on the plate. Skeptical of mutual exclusion between two parti-
cles of the same strain, Hershey tried mixed infections with the r+
and r to analyze the burst of single bacteria. He used a technique
already worked out by Dclbriick. After infection, he plated the in-
fected bacteria before the burst. Thus, each plaque on the plate
represented the phage particles coming from one bacterium, because
tlic burst takes place in the agar and remains concentrated at one
point. If the principle of mutual exclusion was correct, all the plaques
should have been either r or f+. Hershey had no idea how mixed
plaques would look. What he was planning to do was to sample some
plaques at random and to analyze the population obtained from each
plaque. But the plaques were neither of the r nor of the r+ tyP^-
They were mottled, r and r+ growing together to give a type of
plaque very different from either one alone. The actual presence of
the two types in the same plaque was confirmed.

Mutual exclusion had been demonstrated in a very clear way for
morphologically and serologically unrelated strains like Tl and T2.
But, what would happen if similar strains were used, for instance T2
and T4, which have many features in common? Following Hershey's
lead, Delbriick found that T2 and T4 could grow together in the
same bacterium.

In one experiment, Delbriick used, instead of T2, a strain of T2r.
The original experiment of Hershey with mixed infected bacteria
could be repeated, not using T2r and T2r+, but using instead T2r
and T4r+. At this stage something unexpected occurred. The bac-
teria infected with T2r and T4r+, besides yielding both infecting
types, yielded also two new types, T4r and T2r+. Recombination in
phage had been discovered. Not only could two different phage
particles grow in the same bacterium, but they could also recombine
some of their characters. At the Cold Spring Harbor Symposium
for Quantitative Biology in 1946 Delbriick and Bailey announced
this result in their paper, "Induced Mutations in Bacterial Viruses."
They used the following scheme to represent the new phenomenon:

T4r+ > T4r (under the influence of T2r)

At the end of their paper Delbriick and Bailey made the follow-

4 SEX IN MICROORGANISMS

ing statement: "Perhaps one might dispute the propriety of calhng
the observed changes induced mutations. In some respects they look
more hke transfers or even exchanges of genetical materials." At
the same symposium, Hershey gave convincing evidence for the in-
dependent occurrence of h and r mutations in phage T2. The field
was open for genetic investigation. A two-factor cross, hr by h'^r'^,
could now be attempted.

In 1949 Hershey and Rotman published a paper in which they
described the main features of phage genetics. The principal finding
is the fact that in any two-factor cross, with equal multiplicity of
the two parents, recombinants are obtained in different frequencies
depending on the linkage relation between the two markers. Within
the two parental types and within two recombinant types the fre-
quencies are the same. What changes is the ratio of recombinants
to the total. Shortly after the discovery of the r mutant, Hershey did
some single-factor crosses to control the results of mixed infection.

The cross hr'^ x ^+r+ (original experiment of Luria) is a one-
factor cross, and obviously no recombinants can be found. Also hr x
/; + r is a single-factor cross. From his strain J? Hershey isolated an
r mutant (the frequency of the r mutation is quite high), and he
proceeded to cross hr by r. The amazing result of this simple experi-
ment was the discovery in the yield of a high percentage of f+.
As mentioned before, the hr was isolated from an r mutant in the
strain h. However, this strain can be prepared the other way round:
an /; mutant can be isolated from the r strain. Now we have again
two strains, hr and r. We repeat the cross as before; the result is
perfectly regular. No f+ is found. Conclusion: Different mutations
occurring at different loci give the same phenotype, r. Let us take
rl and r2. The following genotype will be obtained: flr2+, rl+r2
(parental types); rlrl, rl+r2+ (recombinant types). Only rl+f2 +
can be distinguished from the others. The existence of rlr2 can only
be demonstrated by doing separate crosses with rl and r2. In neither
case is the normal type found in the yield. All the r's that Hershey
isolated proved to be at different loci.

THE INTRACELLULAR LIFE CYCLE OF BACTERIOPHAGE

The first step in the growth cycle of a phage is the adsorption
of the phage particle to the bacterial surface. If the bacterial cell is

GENETIC RECOMBINATION IN BACTERIAL VIRUSES 5

artificially broken up soon after infection, no phage will be found.
F^or phage T2 this period, called the "eclipse period," lasts for about
10 minutes. From 10 minutes to 22 minutes mature phage particles
can be recovered from artificially lysed bacteria. Doermann (1948)
has shoM-n that mature phage accumulates in the bacterium linearly
Mith time.

Hershey and Chase (1952) have found that as soon as the
phage is adsorbed it releases its nucleic acid into the bacterial cell
\\hile most of the protein part of the phage remains outside stuck to
the bacterial surface. Only the nucleic acid part is necessary for the
subsequent formation of new phage. At this point we must infer a
mechanism by \\'hich a continuity is established between the infect-
ing phage particle and the progeny phage which starts appearing in
the bacterium 10 minutes after the infection. We assume that the
nucleic acid moiety of the phage forms a new entity which we call
vegetative phage. The vegetative phage grows in the bacterium,
forming a population of particles called the pool of the vegetative
phage. At a given moment vegetative phage particles are withdrawn
from this pool and transformed into mature phage particles. The proc-
ess of maturation which goes on linearly with time consists in the
phage's attainment of a protein coat formed of an external membrane
and a tail. When normal lysis occurs, a great many mature phage
particles are released and all immature particles are lost. Whereas
vegetative phage is the active phase of viral Hfe in the sense of growth,
mature phage is a dormant phase of the phage between two growth
cycles.

Doermann (1951) has shown that among the first mature phage
particles to appear in the cell, recombinants are already present. On
the other hand, Levinthal and Visconti (1953) have shown that by
delaying the lysis, the frequency of recombinants among the liberated
phage particles increases considerably.

TECHNIQUE OF PHAGE CROSSES

The viral mutations used as markers are either host-range mu-
tants or plaque type mutants. The first, indicated by h, are mutants
which can lyse bacteria resistant to the normal h'^ phage strain. The
plaque mutants indicated by r differ from the normal by the morpho-
logical character of the plaque.

6 SEX IN MICROORGANISMS

The procedure of making a cross between two different strains
consists in infecting a certain number of growing bacteria with two
kinds of virus. These bacteria will lyse, yielding new phage. We
refer to the infecting phage as the parent and to the phage which is
liberated upon lysis as the progeny. Different ratios of parental phage
types can be used to infect the bacteria, but the average procedure
is to use equal multiplicity, let us say 5 particles of one type and 5
particles of the other per single bacterium.

A suspension of bacteria in buffered saline, containing 10^ cells
per milliliter, is mixed with an equal volume of a suspension of phage
containing 10^ of each parental type. Under these conditions, adsorp-
tion of the phage to the bacteria occurs without starting the growth
cycle of the phage in the bacterial cell. After adsorption is completed,
growth of the phage is started by adding nutrient broth. At the end
of the latent period the bacteria will burst, yielding phage progeny.
This progeny is plated and the different types of plaques scored. The
ratios between the different types will give the result of the cross.
As the mating occurs not between two particles but between groups
of particles, a multiparental cross is possible if the bacterial cell is
infected with more than two parental types. Hershey for the first
time obtained results of triparental crosses, showing that some prog-
eny particles contain markers derived from all three parental types.

HETEROZYGOTES

The phage particles recovered in the yield of a cross are con-
sidered haploids. This statement is made for the sake of simplicity,
because Hershey and Chase (1951) have shown that 2 per cent of the
phage particles originating from a cross are heterozygous for any
character for which the two parents are different. Except for this 2
per cent, all the other particles give, on a subsequent infection, a pure
yield of either one parental type or the other.

At first sight one is tempted to interpret this finding by assuming
that a certain random fraction of the particles are diploids. In a two-
factor cross this leads to the prediction that heterozygosis with respect
to one of the two factors should be strongly correlated with heter-
ozygosis with respect to the other factor. In contradiction to this,
Hershey and Chase found zero correlation when the two factors are
unlinked. Only when the factors are closely linked does some correla-

GENETIC RF.COiMBINATION IN BACTERIAL VIRUSES 7

rion appear. One way of explaining this result would be to suppose
that small frequent duplications of the linkage structures occur: small
to account for the lack of correlation between distant factors, fre-
quent to account for the 2 per cent heterozygous at any given locus.
Practically every phage should carry some of these duplications.

THE LINKAGE SYSTEM

A linkage map in the usual sense of the term cannot be drawn
directly on the basis of recombination data in phage crosses. The
reason for this is that the phage cross is very different from a cross
between two organisms in the classical sense of Mendelian genet-
ics. An attempt can be made, nevertheless, to see if linkage exists, and
to demonstrate linear arrangement of the loci. The first attempt of
this kind was made by Hershey and Rotman with different mutants
of T2. In two-factor crosses the frequency of recombinants in the
total yield varied from 40 per cent downward. Three linkage groups
were found, each of which showed 40 per cent recombination with
the other two. As no higher value of recombination was obtained,
these groups were considered independent or unlinked. Lower values
of recombination between two loci indicate that they belong to the
same linkage group. Inside the linkage groups all kinds of recombina-
tion values can be found. If these characters are chosen rather closely
linked to each other by different combinations of three-factor crosses,
which one is in the middle can be determined. If the order is A, B,
C, the frequency of the ABC type will be much smaller in the cross
AC X B than in the crosses AB x C or A x BC. More complete and
elaborate data have been collected recently by Doermann and Hill
(1952) working with phage T4. From these data full evidence can
be obtained for the linearity of the markers on the different linkage
groups.

THE MECHANISM OF GENETIC RECOMBINATION
IN PHAGE

The interpretation of the data on phage recombination requires
some generalization of the idea of a cross. That the mixed infection of
a bacterium is not a straightforward analogue of a simple genetic
cross can be easily demonstrated by the following facts:

8 SEX IN MICROORGANISMS

1. In triparental crosses Hershey has shown that some prog-
eny particles contain markers derived from all three parental particles.

2, In a two-factor cross with unequal multiplicity of infection
(20 particles of parental type AB and one particle of parental type
ab) the progeny contains more recombinants of one type, say aB,
than of minority parents ab (Doermann, 1952).

As has already been mentioned, the progeny of a cross can be
examined at different times, either by inducing premature lysis or
by delaying the lysis. With closely linked markers, the fraction of
recombinants increases linearly with time between the moment of
appearance of the first mature phage and time 60 minutes after in-
fection. These two lines of evidence derived from premature lysis
and delayed lysis indicate that in the single growth cycle there is a
drift in the course of time toward genetical equilibrium.

Two different kinds of hypothesis can be considered at this
stage: (a) either the new phage is formed out of some kind of pool
of the parental characters, or (b) matings occur between vegetative
phage particles in such a way that the product of one segregation can
mate again with some other particle.

What we are actually considering is a choice between some com-
pletely novel type of recombination occurring during reproduction
of the phage and the classical mechanism of recombination by mating
between pairs as it might take place in clonal reproduction of vege-
tative phage. The first hypothesis was strongly influenced by the
phenomenon of multiplicity reactivation discovered by Luria in
1947. Bacteria infected with phage particles inactivated by ultraviolet
light multiply and produce viable phage offspring in a large propor-
tion of cases. Luria interpreted these results to mean that viable phage
particles could be formed by recombination even when all the paren-
tal particles were unviable as a result of one or more lethal mutations.
It is known now after more thorough investigation of this phenome-
non that reactivation is not due to a genetic mechanism. The second
hypothesis of the mating has been analyzed in great detail by Visconti
and Delbriick (1953). Their theory is based on the following assump-
tions: (1) prophage particles multiply and mate pairwise at random
inside the bacterial cell; (2) vegetative phage particles go on mating
and multiplying up to the time of lysis; (3) mature phage particles
which are accumulated in the bacterial cell do not mate. On the basis

GKNETIC RF.COiMRlNATION IN BACIFRI Al. \ IIIUSI'S c;

of the theory, predictions about the result of some crosses have been
made. Up to now these predictions have been verified.

The drift to\\'ard genetical equiHbrium can now be easily ex-
plained by assuming that mating among vegetative phage particles
goes on at a constant rate with time. The phage which matures at
any moment inside the bacterium is withdrawn from a population of
vegetative phage which has undergone a certain number of rounds
of mating. The later the process of maturation is stopped by delaying
the lysis, the higher will be the average number of matings and the
greater w^ill be the number of recombinants found. Two other impor-
tant facts are explained by this theory. One is the lack of correlation
bet\\'een recombinants. In a cross AB x ab, if the yield of a single
burst is examined, the number of recombinants Ab and aB should be
the same. A lack of correlation between the two frequencies means
that the process by which one recombinant is formed is independent
of the formation of the other one. This fact is accounted for by this
theory, as the maturation of a single vegetative phage into phage is an
event independent of the maturation of any other particle. The other
fact accounted for by this theory is an apparent negative interference,
which is that recombinants between some two markers show a higher
value of recombination between any other markers. It has already
been mentioned that, if the crude recombination data of a cross are
used to make a map, a linear arrangement of the markers belonging
to the same linkage group will be obtained. Let us suppose that there
are three unlinked factors. A, B, C. Considering any two of these
markers, we will get in a cross 60 per cent parental types and 40 per
cent recombinants. This fact can be explained as due to an incomplete
genetical equilibrium reached by the population of vegetative phage
at the moment of maturation. Suppose now that we compare a paren-
tal type to a recombinant, say AB to aB inside the class of recombi-
nants Be. Let the ratio be 1:1. Because of the random distribution
of the number of matings per particle and because any mating with a
similar type gives no possibility of recombination, we operate a
selection in favor of recombination when we choose particles which
have recombined for some other marker. When we apply the right
calculation, this apparent negative interference disappears if we con-
sider a single mating. Thus genetic maps can be made by calculating
the average number of matings. For normal lysis in phage T2 a value

10 SEX IN MICROORGANISMS

of five matings has been calculated. Recombination data obtained by
Doermann in phage T4 led to the conclusion that if we refer to a
single mating no interference can be demonstrated.

GENERAL CONCLUSIONS

Bacteriophage can be considered a haploid organism possessing
nothing analogous to sex. A theory has been postulated which ac-
counts for the recombination data known at present. This theory is
supported also by some biological data on the intracellular growth of
the bacteriophage. Some predictions of the theory have been verified.
One important factor is not explained by the theory: heterozygosis.
At the present moment phage looks rather unsuitable for genetic
research, because the progeny of a single mating cannot be examined.
We are in the situation of a geneticist who can examine his popula-
tion once in a while, say once in every five generations. On the other
hand, genetics of bacteriophage appears to be a powerful tool in the
attempt to explain the main biological features of virus reproduction.

REFERENCES

Delbriick, M., and S. E. Luria. 1942. Interference between bacterial viruses.

Arch. Biochem., 1, 111-135.
Delbriick, M., and W. T. Bailey. 1946. Induced mutations in bacterial viruses.

Cold Spring Harbor Symposia Quant. Biol., 11, 33-37.
Doermann, A. H. 1948. Intracellular growth of bacteriophage. Carnegie Ivst.

Wash. Yearbook, 47, 176-186.
. 1951. Intracellular phage growth as studied by premature lysis. Fed-
eration Froc, 10, 591-594.
Doermann, A. H., and M. B. Hill. 1953. Genetic structure of bacteriophage

T4 as described by recombination studies of factors influencing plaque

morphology. Genetics, 38, 79-90.
Dulbecco, R. 1952. Mutual exclusion between related phages. /. Bact., 63,

209-217.
Hershey, A. D. 1946. Spontaneous mutations in bacterial viruses. Cold Spriijg

Harbor Symposia Quant. Biol., 11, 61-11.
Hershey, A. D., and M. Chase. 1951. Genetic recombination and heterozygosis

in bacteriophage. Cold Spring Harbor SyiJiposia Quant. Biol., 16, 471-

479.
. 1952. Independent functions of viral protein and nucleic acid in

growth of bacteriophage. /. Ge72. Physiol., 36, 39-56.

GENETIC RECOMBINATION IN BACTERIAL VIRUSES H

Hershey, A. D., and R. Rotnian. 1948. Linkage among genes controlling in-
hibition of Ivsis in a bacterial virus. Proc. Natl. Acad. Sci. U. S., 34,
253-264.

. 1949. Genetic recombination between host range and plaque type

mutants of bacteriophage in single bacterial cells. Genetics, 34, 44-7 L

Levinthal, C, and N. \'^isconti. 1953. Growth and recombination in bacterial
viruses. Genetics., 38, 500-511.

Luria, S. E. 1945. Mutations of bacterial viruses affecting their host range.
Genetics, 30, 84-99.

. 1947. Reactivation of irradiated bacteriophage by transfer of self-

reproducing units. Froc. Natl. Acad. Sci. U. S., 33, 253-264.

Msconti, N., and M. Delbriick. 1953. The mechanism of genetic recombination
in phage. Genetics, 38, 5-33.

Sex in Bacteria

Genetic Studies, 1945-1952

JOSHUA LEDERBERG,* Department of Genetics,
University of Wisconsin, Madison

E. L. TATUM, Department of Biological Sciences,
Stanford University, Stanford, California

For many years bacteria were considered biologically exceptional
organisms with no genes, nuclei, or sex, although the recognition of
their biochemical similarities to other forms of life constituted one of
the main foundations of comparative biochemistry. Over the last
decade evidence has accumulated which has led to the satisfying con-
clusion that bacteria are not biologically unique but possess genetic
and behavior systems more or less analogous to those of other forms,
including nuclei, genes, and in certain instances even true sexual
mechanisms for recombination of unit characters.

Historically, this change in our thinking in regard to bacteria
stems from the pioneer concepts of Lwoff (1938) and Knight (1936)
relating the nutritional requirements of mircooganisms to an evolu-
tionary loss of synthetic abilities. If such losses in microorganisms
were based on mutation and selection as required by modern concepts
of evolution, the capacities for synthesis of essential nutrilites in
microorganisms should be determined by genes, which should be
subject to mutation, as are most genes in other organisms. Such consid-
erations led Beadle and Tatum (1941) to the successful production
by irradiation of nutritionally deficient or biochemical mutants in the
heterothallic ascomycete Neiirospora, and to the establishment of the
genie basis of biochemical reactions leading to the synthesis of amino
acids and vitamins. This relation has since been amply substantiated
and extended by further work with Neiirospora and other sexual
microorganisms (cf. Tatum and Perkins, 1950).

* Paper No. 528 of the Department of Genetics. This work has been
supported by research grants (C-2157) from the National Cancer Institute,
National Institutes of Health, Public Health Service, and from the Research
Committee, Graduate School, University of Wisconsin, with funds provided
by the Wisconsin Alumni Research Foundation.

12

SEX IN BACTERIA— GF.NF.TIC STUDIES 1 3

The next Step in the cvokition of our concepts of bacterial ge-
netics was the experimental application of the Neiirospora techniques
to the production of biochemical mutants in these simpler organisms.
The first nutritionally deficient (auxotrophic) mutants were pro-
duced in 1944 by x-ray treatment of Escherichia coli and Acetobacter
vielanogemim (Gray and Tatum, 1944; Roepke et al., 1944). Sub-
sequently, auxotrophic mutants have been obtained in almost every
species of bacteria investigated (see Tatum, 1946; Tatum and Perkins,
1950), In addition to mutations to growth-factor dependence and
reverse mutation to growth-factor independence, other types of
mutant characters have been obtained, thoroughly investigated, and
proved extremely valuable. These include such characters as virus
resistance, antibiotic resistance, and capacity for sugar utilization.

The first auxotrophic mutants were obtained by the tedious and
laborious process of plating out irradiated cells on fully supplemented
medium, and then isolating individual colonies which were subse-
quently examined for failure to grow in the simple synthetic medium
adequate for the original stock (minimal medium). The specific
requirements of these deficient clones were then determined by sys-
tematic supplementation of the minimal medium with known growth-
factors.

Later improvements in techniques have eliminated much of the
labor involved in isolating and testing mutants of bacteria, particu-
larly E. coli. These include (cf. Lederberg 1950, 1951a) the layer-
plate technique in which only presumptive mutants are isolated for
further testing, and the penicilhn method in which non-mutants are
actively eliminated, leaving mutant cells which form colonies after
removal of the antibiotic and suitable supplementation of the medium.
A further modification of this method, using solid medium and peni-
cillinase, permits easy isolation of any desired type of auxotroph
(Adelberg and Myers, 1952), and the isolation and testing of strains
has been still further simplified by the replica plating method (Leder-
berg and Lederberg, 1952) which in a single operation permits trans-
ferring all colonies on a plate to a number of other plates with dif-
ferent supplements.

By all available criteria it is now generally accepted that most, if
not all, characteristics of bacteria are controlled by hereditary units,
and that these hereditary units in bacteria are analogous with genes
in classically sexual organisms in the independence and randomness

14 SEX IN MICROORGANISMS

of their mutation, the effect of physical and chemical agents on their
mutation frequency, and qualitatively in the types of biochemical
and enzymatic effects of their mutation (cf, Tatum and Perkins,
1950).

After establishment of this functional analogy between genes
of bacteria and those of sexual forms, the next step logically was to
ask if the analogy could be carried further, and if a mode of inheri-
tance of bacterial characters similar to the Mendelian process in
higher types could be detected. Although a number of workers had
looked for character recombination in bacteria, and other microor-
ganisms, even as early as 1908 (Browning, 1908; for other important
references cf. Tatum and Lederberg, 1947), in most cases the biolog-
ical materials available were not adequate for a definitive experimen-
tal test, although some suggestive evidence of a circumstantial nature
has been obtained. The definitive nature of auxotrophic mutants and
the relative ease of their isolation and diagnosis provided ideal material
for testing the possibility of recombination of hereditary units in
bacteria.

The independent occurrence and expression of auxotrophic
mutations in E. coli permitted building up multiple mutant stocks of
E. coli strain K-12 with several deficiencies by successive mutational
treatment (Tatum, 1945). In this way, for example, cultures Y-10,
requiring threonine, leucine, and thiamin, and 58-161, requiring biotin
and methionine, were obtained. For simplicity in considering its
capacity for synthesis of the factors concerned, strain 58-161 can
be represented as biotin— methionine— threonineH- leucineH- thia-
min+ (B— Af— T+ L+ Bi+), while strain Y-10 would similarly
be represented as B-f M-\- T— L— Bi— . In this representation, in
analogy with other organisms such as Nejirospora, the genes deter-
mining alternative characters (B+ and B— for example) are consid-
ered allelic.

Accordingly, a sexual process in a mixed culture of these two
strains would involve reshuffling the indicated alleles at these five loci.
If this were at random, any recombination might be expected, and
might have been looked for. However, recombination to give a
nutritionally independent type (prototroph) would be most easily
detected, since it alone would grow on minimal medium, whereas any
dependent types, including both parental strains, would not.

Experimental tests were carried out (Tatum and Lederberg,

SEX IN BACTERIA— GENETIC STUDIES

15

1947) by growing the two strains either separately or together in
complete media, then centrifiiging out the cells, washing repeatedly to
remove growth-factors, and plating mixtures of the washed cells into
minimal .agar. The results were striking in that about 100 colonies
developed for each 10^ cells examined, and on reisolation and purifica-
tion these maintained their prototrophic character. Similarly treated
single cultures of each strain gave no colonies on the minimal medium.
This would be expected on the basis of the low frequency of muta-
tion of each character to independence (ca. 1 in 10^ cells) since the
derivation of a prototroph from a triply deficient strain would then
occur with a frequency of 1 in 10-^ cells.

©^. Y-IO

^T-L-BrB+M+

I per 102' cells? /

I per lo'^ cells?
I PROTOTROPHa •

MIXTURE

T-L-B,-|B+M+ \

— H >- V

t+l+b,+ ;b-m-

T+L+B,+ B+M+
CO. I per lO^cells

58-161
T+L+B,+B-M-

FiG. 1. Diagrammatic representation of recombination in prototrophs.

The simplest explanation for these results therefore appeared to
be that gene recombination took place to give the prototroph B+
M-\- T-}- L-\- Bi-\-, as shown in Fig. 1. Other possibilities, such as
association of cells, or the formation of unsegregated diploid, or of
heterocaryotic cells were made unlikely by various experimental
tests which established the homogeneity and uniqueness of the derived
prototrophs (cf. Tatum and Lederberg, 1947).

The only alternative to a sexual recombination seemed therefore
a type of unilateral change by a non-cellular transforming principle
similar to that involved in induced changes of type in the pneumo-
coccus (Austrian, 1952). Two Hnes of evidence made this improbable.

16 SEX IN MICROORGANISMS

First, cell-free filtrates gave no prototrophs and direct contact of the
cells themselves seemed essential, as shown by growing the two types
separated by an ultra-fine sintered glass bacterial filter (Davis, 1950).
Second, the successful recovery of most of the possible recombination
types in later crosses involving a much greater variety of characters,
including resistance to antibiotics and viruses and sugar fermentation
characters, would necessitate simultaneous transformation in different
directions for different characters, and in both directions in different
cells.

Thus the results of experiments of the type described above are
satisfactorily explained only as resulting from a sexual mating process,
followed by reassortment or segregation of genetic material. Repeti-
tion and extension of these experiments in a considerable number of
laboratories during the past five years have amply confirmed the
reality of the essential phenomenon and the validity of this conclu-
sion.

These experiments have added considerable information about
environmental factors affecting the recombination process, and sup-
port the concept that direct cell contact is necessary for the sexual
process. Some of the strongest support comes from the demonstration
by Nelson (1951) that recombination behaves as a bimolecular
reaction, as if factors such as relative and absolute concentrations of
the two types of cells, which would affect the frequency of contact
of appropriate cells, similarly affect frequency of recombination. The
experiments of Davis (1950), showing the need for cell contact, and
in a more positive sense the production of a genetically diploid cell
(Lederberg, 1949) likewise support the postulated occurrence of a
cell to cell sexual process in E. coU K-12.

The intimate details of mating are still obscure. Owing to its
infrequency we have been discouraged (until very recently) from
any serious attempts to detect its morphological basis, and were
obliged to be content with genetic inferences. E. Klieneberger-Nobel,
of the Lister Institute, London, England, has made a most painstaking
study of mating cultures of E. coli K-12 (unpublished work, quoted
by her kind permission). Although she occasionally observed what
appeared to be stages in the abortive development of L-forms
(Klieneberger-Nobel, 1951), she was unable to correlate them in any
way with recombination. The only conclusion that is warranted is
that recombination in E. coli does not involve spectacular formations.

SFX IN BACTERIA— GENETIC STUDIES 17

such as have been observed and speculated about in many bacteria.
The possibility of a mating process involving a rapid conjugation and
separation of the parent cells, without the intervention of special
gamete or zygote structures, has not been excluded and is perhaps
most likely.

The recently reported work of Hayes may throw further light
on the conjugation process. Working with a single pair of K-12
stocks, Hayes (1952a) found that streptomycin treatment sterilized
one (58-161) without affecting its recombination potency, but com-
pletely abolished recombination potency of the other (W677), and
he postulated a unidirectional transfer of metabolically inert genie
material from 58-161 to W677. The results of later studies (Hayes,
1952b) using ultra-violet irradiation for cell inactivation and stimula-
tion of recombination (Clark et al., 1950) were consistent with this
hypothesis. He has suggested that the recombination may involve only
a limited transfer of genie material, through a process which may
not require the participation of two intact cells.

Granted the ability of E. coli K-12 to undergo a sexual recom-
bination of genetic characters analogous to that found in other
organisms, even if the morphological basis is still obscure, can the
analogy be carried further? What evidence exists bearing on a chro-
mosomal organization of the genes in E. coli} In all organisms that
have been adequately studied, the genes are arranged in a linear
order on chromosomes, whose distribution at cell division and during
the formation of gametes follows very precise laws. It is difficult, in
fact, to conceive of any other arrangement of the genes that would
permit their regular and orderly distribution to the products of each
cell division, without an uneconomical redundancy of the genetic
factors for different traits. But aside from these speculations, there
is considerable experimental evidence that the genes of E. coli are
organized in linear order on one or more chromosomes (Lederberg,
1947; Rothfels, 1952; Lederberg et al, 1951).

In the crosses mentioned so far, all the differences between the
parents are directly involved in the selection of recombinants, so that
we had no opportunity to investigate the segregation of factors whose
expression is not enforced by the selective method. A number of
mutant characters have been discovered, however, which are indif-
ferent to plating on minimal medium. They include differences in
the fermentation of various sugars, resistance to antibiotics, and resist-

18 SEX IN MICROORGANISMS

ance to phages. Such characters will be called unselected, since their
segregation is regulated by the internal mechanism of recombination
rather than the exigencies of the technique.

The first unselected marker to be used in our experiments was
resistance to phage Tl (Tatum and Lederberg, 1947). According to
conventions, resistance and sensitivity are symbolized as V/ and Vi%
respectively. Vi' is a specially convenient marker, as it can be pro-
duced in any stock by the selection of spontaneous mutants with Tl.
Before they can be used in these experiments, such stocks must be
carefully purified and, as for any marker, the stability and repro-
ducible scoring of the mutation must be verified. A variety of crosses
was carried out in which one parent was V/, the other Vi^. In each
case we found a segregation for this marker, i.e., some of the proto-
trophs displayed the V/ trait, from one parent, and others F/ from
the other. In control crosses, V/ x F/' gave only Vx% and V/ x
V/ gave only Vi'. Such a segregation in the first filial generation,
the f-1, indicates a haplobiontic life cycle, similar to that of many uni-
cellular organisms.

If an unselected marker were associated with one chromosome
independent of others carrying the selected, nutritional mutations, the
f-1 should show a mendehan ratio of 1:1. Different ratios were ob-
served for each of the markers tested, the first evidence of linkage.
The observed frequencies varied from one marker to another, and
with a given marker, from one parental combination to another. In a
given cross, however, the f-1 segregation ratios have been as repro-
ducible as in any genetic material, which is to say they are subject
mainly to sampling error.

A simple test of the significance of f-1 ratios can be made by
reverse crosses, whereby a marker is introduced first in one, then in
the other parent. For example, BM— V Z x TL— F/' is compared
with BM— F/ X TL— V/. About 70 per cent of the prototrophs
from the first cross are F/'". In the second cross, about 30 per cent
are V/, i.e., this ratio is inverted. The same result has been obtained
in many reverse crosses involving different parental lines, and different
markers and combinations of markers. It sho^^'s that the f-1 ratios have
nothing to do with the physiological effects of the markers, but that
they are due entirely to the mechanics of segregation. It also shows
that dominance plays no role, and more generally that a genetic parti-

SFX IN BACTFRIA— GENKTIC SI UOIFS 19

cle controlling each observed trait has been segregated, and is rep-
resented only once in the genotype of the recombinant cell.

By compounding elementary principles, genetic maps of E. coli
can be constructed from segregation data involving numerous unse-
lected markers (Cavalli, 1950; Newcombe and Nyholm, 1950; Roth-
fels, 1952; Lederberg et j/., 1951). By using other methods, the auxo-
troph mutations can be relieved of their burden of the selection of
recombinants, and thus handled as unselected markers also. About
half of the known markers of E. coli K-12 have been satisfactorily
located in a single linear linkage group. Other markers have displayed
a more confusing behavior which does not fit any scheme very
satisfactorily, but is probably a result of rather complex chromosomal
aberrations, for w^hich there is independent evidence from the study
of exceptional diploids (v. infra). It was long thought that E. coli had
onh' one chromosome, but more recent evidence points to at least
two, the segregation of which is not, however, entirely independent
for secondary reasons (Fried and Lederberg, 1952). Cytological
observations on haploid K-12 cultures have been interpreted by De
Lamater (1952) as signifying three chromosomes, but further work
is needed for the detailed concordance of genetic with cytological
findings. Cytological study of E. coli has so far been confined to
vegetative cells, whereas the genetic studies deal principally with
segregation at meiosis.

The difficulties, briefly mentioned, in the segregation of cer-
tain misbehaving markers might appear to be fatal to a straightforward
sexual interpretation of recombination except for the confirmatory
evidence provided by exceptional diploid cultures. In ordinary cross-
ings, the diploid condition has been inferred from its consequences of
recombination and segregation, but is not directly observed. In 1946-
47, many unsuccessful attempts were made to secure artificial diploids
with agents such as camphor, acenaphthene, colchicine, and heat
shocks, which have been used for other organisms (cf. Roper, 1952).
More recently, however, a mutation, Het, occurred in one of our
stocks which serves the same purpose (Lederberg et al., 1951). Little
is known of the action or generic transmission of Het, but when it is
present in one or both parents of a cross, several per cent of the
prototrophs prove to be persistent heterozygotes. These heterozygous
cultures continually segregate the alternative markers brought in

20

SEX IN MICROORGANISMS

by the parents. Thus diploid cells heterozygous for lactose fermenta-
tion, Lac-\-/Lac—, produce mosaic colonies on an indicator medium,
as shown in Fig. 2. The dark or "-]-" sectors consist of still heterozy-
gous cells, Lac-{-/Lac—, and of Lac-{- haploid segregants; the light
or "— " sectors are Lac— haploid segregants. On complete medium,
the faster-growing haploid cells soon outstrip the original diploids, but
segregation can be effectively prevented on a minimal medium owing
to the nutritional requirements of the haploid components. Single

Fig. 2. Segregating diploid culture plated on indicator agar, showing va-
riegated colonies.

cell studies (Zelle and Lederberg, 1951) have verified that genetic
factors from two parents have converged to a single hybrid cell,
the essence of sexuality.

Haploid and diploid cultures have been studied cytologically,
especially for comparisons of their nuclear structure, by means of
the Piekarski-Robinow technique (osmic fixation; HCl hydrolysis;
Giemsa stain; mount in Abopon). This method gives brilliant nuclear
preparations, but E. coli appears to be technically unsuitable for

SEX IN BACH RIA—Cil,Nl. TIC SIUDIL.S 21

unequivocal demonstration of mitotic figures, as claimed in the pio-
neering and provocative work of DeLamater and his associates
(1951). Although nuclear aggregates that are very suggestive of
mitotic metaphases and anaphases can be found with a brief search,
definitive interpretations of E. coli cytology depend for the most part
on the validity of the conclusions that have been drawn from techni-
cally superior material. It is difficult for a geneticist to imagine how
bacteria could get along without some sort of mitotic process, but its
details require critical and objective definition. The comparisons of
haploid and diploid E. coll have revealed consistent and unequivocal
differences, as shown in Figures 3 and 4 and elsewhere (Lederberg
et ah, 1951). The determination whether the diploids show a doubling
of the chromosome number is not yet subject to independent, objec-
tive verification.

The correlation of genetic heterozygosity with nuclear complex-
ity is only a small step in the direction of a bacterial cytogenetics.
It has been furthered by Witkin's studies, in which the segregation
of mutant genes during fission has been correlated with the nuclear
plurality of the bacterial cells at the time the mutations were induced
(Witkin, 1951). These observations do accord, however, with a chro-
mosomal theory of inheritance and sexuality in E. coli.

The work cited so far has been done with derivatives of a single
strain, K-12, of Escherichia coli. A few early attempts to duplicate
genetic recombination in other E. coli strains popular in genetic work
were quite unsuccessful. Cavalli and Heslot (1949) discovered a
culture in the British Type Culture Collection, NTCC 123, that was
fertile with K-12, but a special screening method had to be developed
before many new strains could be effectively studied (Lederberg,
1951a). Of nearly 2000 independent isolations of E. coli from various
sources, over fifty have proven to be cross-fertile with K-12, and so
far as has been tested, with each other. All of the new strains conform
to the type E. coli, except for an occasional minor deviation, but are
otherwise as heterogeneous as any sample of strains. They are serolog-
ically quite diverse: an immunogenetic study has been initiated which
has so far put the antigens of E. coli on the same basis as the mam-
malian blood groups.

One important reason for undertaking this study of new strains
was to investigate the sexual compatibility relations of E. coli. Until
recently, several lines of evidence conspired to substantiate the idea

22

SEX IN MICROORGANISMS

* irf," •

Fig. 3. E. coli K-12, haploid.

Fig. 4. £. co/i K-12, diploid.

This photograph and Fig. 3 are taken from Osmic-HCL-Giemsa preparations by
Miss E. R. Lively and are reproduced here at about 2,500 magnifications. Both include
some cells that appear to display mitotic figures and others that do not.

SKX IN BACTFRIA— GF.NETIC STUDIES 23

that E. coJi K-12 was homothallic. The crossahlc strains arc all de-
rived from a single pure culture. Several workers had suggested that
a niating-tvpe system might be obscured by mutations from one mat-
ing type to the other, as occurs in certain yeasts. However, this
hxpothesis \\as rejected because no segregation of mating preferences
was observed from heterozygous diploids, as would have been ex-
pected from a heterothallic mating. It has since been discovered that
a unique compatibility mechanism does operate in Escherichia coli
(Lederberg, Cavalli, and Lederberg, 1952). The involved history of
this discovery must be detailed elsewhere, and only the general
conclusions can be given here.

Wild type K-12 carries a hereditary factor, F+, which is re-
quired for mating. Similarly, most of the auxotrophic mutants of K-12
are f +, and therefore mutually compatible, but the much used line
descended from the threonine-leucine mutant, 679-680 (Tatum, 1945),
is F—. Because most of the other tester cultures are F+, however, the
F— "mutation" was not detected in earlier experiments. The empiri-
cal definition of F— is that crosses of two F— parents are completely
sterile, although comparable crosses in which one or both parents is
F-{- are productive. The self-incompatibility of F— has been detected
in two \vays: sublines of 679-680 are mutually incompatible, and new
occurrences of the F— "mutation" have been discovered which are
incompatible with 679-680. The F— "mutation" is given in quotation
marks because its inheritance and transmission set it apart from all
of the other markers so far studied.

All the progeny of crosses within strain K-12 are F4-, whether the
parents were F— x F+ or F+ x F+ [F— xF— cannot, of course, be
tested]. This was explained by the finding that F+ was contagious,
that is that growing F— cells in mixture with F-f- resulted in many
of the former (identified by other genetic markers) becoming per-
manently F+. As many as 50 per cent of the originally F— cells may
become F-f by this conditioning process within a few hours. "F+"
is therefore tentatively regarded as an infective, virus-like agent, but
this has not yet been confirmed by the isolation of "F-f" in a cell-free
preparation. The transmission of F+ is not accompanied by the trans-
fer of any other marker, so far as is known.

The virus-like properties of the compatibility factor have led to
some speculation on its relationship to a virus known to be present in
E. coli K-12, the latent bacteriophage X. This question has been

24 SEX IN MICROORGANISMS

Studied in some detail, and it can be asserted that X is not related in
any way to genetic recombination or to the sexual compatibility
mechanism. Non-lysogenic, i.e., X-free cultures, are fully compatible
in sexual recombination, and the transfer of F+ is independent of the
transfer of \ which, unlike F+, can readily be obtained in cell-free
filtrates. Genetic studies of lysogenicity (Lederberg and Lederberg,
1953) have, however, demonstrated a close relationship between this
latent virus and the chromosomes of the bacterial host.

The details of the compatibility system are being studied at the
present time. It has been noted that F-f- x F+ crosses tend to be con-
siderably less productive than comparable F-\- x F— . Many of the re-
sults are consistent with the concept of relative sexuality as noted in
many algae and fungi. That is to say, different cultures can be ar-
ranged in sequence of relative potencies, such that the productivity of
a cross will be related to the difference in potency of the two parents.
In E. coli K-12, the relative potency can be controlled both by envi-
ronmental variations, and by genotypic effects. Within K-12, differ-
ences in the F-\- agent itself have not been found. However, the F-h
state as conditioned by some other wild-type strains appears to be
unstable in K-12 cells, suggesting the possibility of genetic differences
in the presumed agent itself.

The genetic basis of the observed "mutations" to F— in strain
K-12 is not known, and these have not been experimentally repro-
ducible. Many wild type strains are F— (i.e., non-infective) but
retain their compatibility status, so that it is impossible to generalize
on the causes of sterility or compatibility in the species E. coli taken
as a whole.

In the absence of direct morphological evidence of sexual fu-
sions, the principal alternative explanation for genetic recombination
in E. coli has been "transformation" or "transduction." The biology
of bacterial transductions is not very well understood (Ephrussi-
Taylor, 1951; Austrian, 1952); in many ways it may be constructive
to regard them as a limited form of hybridization. The chief distinc-
tion between transduction and sexual recombination is that the former
seems to involve only a very small part of the whole genotype of the
bacterium at each transfer (as in the capsular transformations in the
pneumococcus), whereas sexual reproduction allows reassortment of
the entire genotype of each parent, as in E. coli. The former seems to
be correlated with an active unit that is morphologically and chemi-

SEX IN BAG ll'.RIA— GENETIC S lUDIES 25

cally much simpler than the intact cell; in E. colt, no unit other than
the cell has been shown to be active in recombination.

A search for recombination in Sahnonella typhimur'mm, a species
distantly related to E. coli, has led to the discovery of another mode
of transduction. In this system occasional bacteriophage particles ap-
pear to become fortuitously contaminated with genetic fragments
from the host cells on which they are grown, and to be able to trans-
duce these fragments to new cells which they may invade without
killing them. The fragments retain their activity, and somehow enter
the genetic organization of the new host. In this way, for example,
flagellar antigenic traits from S. ty phivmrium may be introduced into
cells of 5". typhi to give a hybrid serotype or species not previously
described (Zinder and Lederberg, 1952). The possibilities that sexual
recombination may occur in Sabnofiella or that a genetic transduc-
tion may be found in E. coli are not unlikely, in view of the taxo-
nomic relationship of these bacteria. It is from a superposition of these
phenomena in a single species that we may expect to learn the most
about each of them. At present, however, they appear to be quite
distinct.

To sum up the present status of our knowledge of sexuality in
E. coli, it has been shown that in mixed cultures under suitable condi-
tions a small but significant number of cells of certain strains of this
organism undergo a process of recombination of genes governing a
wide variety of characters. This process apparently involves a cell-to-
cell contact, and presumptively copulation, or conjugation, with
zygote formation. Analysis of the recombination products indicates
that the genes are present in linear order on one or more chromo-
somes. In essence, then, certain strains of E. coli, especially K-12, are
capable of a sexual process, analogous in so far as it has yet been
analyzed to that of other organisms.

It is to be hoped and expected that sexual phenomena will not
be limited to E. coli among the bacteria. It is likewise to be hoped
that the significance of suggestive cytological phenomena in bacteria
such as apparent conjugation tubes (DeLamater, 1951), star-body
formation (Braun and Elrod, 1946), filterable L-forms (KHeneberger-
Nobel, 1951) and large bodies (Dienes, 1946; Stempen and Hutchin-
son, 1951) can be further evaluated by genetic analysis involving the
recombination and tracing of suitable genetic markers.

The future also holds not only promise of correlation of genetic

26 SEX IN MICROORGANISMS

and morphological aspects of conjugation and meiosis in E. coli, but
the even more exciting prospects of discovering, elucidating, and
correlating different modifications of sexuality and transfer of genetic
material in microorganisms. With a clearer understanding of these
relationships, the bacteria may be expected to occupy an increasingly
important place in the study of the comparative biology of sex.

REFERENCES

Adelberg, E. A., and J. W. Myers. 1952. Selection of biochcmicaUy deficient
mutants of E. coli. Federation Froc, 11, 179.

Austrian, R. 1952. Bacterial transformation reactions. Bact. Revs., 16, 31-50.

Beadle, G. W., and E. L. Tatum. 1941. Genetic control of biochemical reac-
tions in Neurospora. Proc. Natl. Acad. Sci. U. S., 27, 499-506.

Braun, A. C, and R. P. Elrod. 1946. Stages in the life history of Phytoiiwiias
tmnejacieiis. J. Bact., 52, 695-702.

Browning, C. H. 1908. Chemotherapy in trypanosome-infections: an experi-
mental study. /. Path. Bact., 12, 166-190.'

Cavalli, L. L. 1950. La sessualita nei batteri. Boll. ist. sieroterap. inilan., 29,
281-289.

Cavalli, L. L., and H. Heslot. 1949. Recombination in bacteria: outcrossing
Escherichia coli K-12. Nature, 164, 1057.

Clark, J. B., et al. 1950. The stimulation of gene recombination in Escherichia
coli. J. Bact., 59, 375-379.

Davis, B. D. 1950. Nonfiltrability of the agents of genetic recombination in
Escherichia coli. J. Bact., 60,' 507-508.

DeLamater, E. D. 1951. A new cvtological basis for bacterial genetics. Cold
Spring Harbor Symposia Quant. Biol., 16, 381-412.

. 1952. The nuclear cytology of Escherichia coli during mitosis. (Ab-
stract.) Geyietics, 37, 576.

Dienes, L. 1946. Complex reproductive processes in bacteria. Cold Spring
Harbor Symposia Quant. Biol., 11, 51-59.

Ephrussi-Taylor, H. 1951. Genetic aspects of transformations of pneumococci.
Cold Spri7ig Harbor Symposia Quant. Biol., 16, 445-456.

Fried, P. J., and J. Lederberg. 1952. Linkage in E. coli K-12. (Abstract.)
Genetics, 37, 582.

Gray, C. H., and E. L. Tatum. 1944. X-ray induced growth factor require-
ments in bacteria. Proc. Natl. Acad. Sci. U. S., 30, 404-410.

Hayes, W. 1952a. Recombination in Bact. coli K-12: unidirectional transfer of
genetic material. Nature, 169, 118-119.

. 1952b. Genetic recombination in Bact. coli K-12: analysis of the

stimulating effect of ultra-violet light. Nature, 169, 1017-1018.

SIX IN HACTI- RIA— GFNF.TIC STUDIES 27

Klicncbcrger-Nohcl, 1". 1951. liltcrablc forms of bacteria. Btict. Revs., 15,

77-103.
Knight, B. C. J. G. 1936. Bacterial nutrition. i\lcd. Research Council (Brit.),

Spec. Rcpt. Scr. No. 210.
Ledcrberg, E. M., and J. Lederberg. 1953. Cienetic studies t)f lysogenicity in

Escherichia coli. Genetics, 38, 51-64.
Lederberg, J. 1947. Gene recombination and linked segregations in Escherichia

coli. Genetics, 32, 505-525.
. 1949. x'Xberrant heterozvgotes in Escherichia coli. Proc. Natl. Acad.

Sci., 35, 178-184.
. 1950. Isolation and characterization of biochemical mutants of bacteria.

Methods Med. Research, 3, 5-22.
. 1951a. Genetic experiments with bacteria. In Genetics in the 20th Cen-
tury. The Alacmillan Co., New York.
-. 1951b. Prevalence of Escherichia coli strains exhibiting genetic recom-

bination. Science, 114, 68-69.
Lederberg, J., L. L. Cavalli, and E. M. Lederberg. 1952. Sex compatibility in

Escherichia coli. Genetics, 37, 720-730.
Lederberg, J., and E. M. Lederberg. 1952. Rephca plating and indirect selec-
tion of bacterial mutants. /. Bact., 63, 399-406.
Lederberg, J., et al. 1951. Recombination analysis of bacterial heredity. Cold

Spring Harbor Symposia Quant. Biol., 16, 413-443.
Lwoff, A. 1938. Les facteurs de croissance pour les microorganismes. Ann.

inst. Pasteur, 61, 580-634.
Nelson, T. C. 1951. Kinetics of genetic recombination in Escherichia coli.

Genetics, 36, 162-175.
Newcombe, H. B., and M. H. Nyholm. 1950. Anomalous segregation in crosses

of Escherichia coli. Am. Naturalist, 84, 457-465.
Roepke, R. R., R. L. Libby, and M. H. Small. 1944. Mutation or variation of

Escherichia coli with respect to growth requirements. /. Bact., 48, 401-419.
Roper, J. A. 1952. Production of heterozygous diploids in filamentous fungi.

Experientia, 8, 14-15.
Rothfels, K. H. 1952. Gene linearity and negative interference in crosses of

Escherichia coli. Genetics, 37, 297-311.
Stempen, H., and W. G. Hutchinson. 1951. The formation and development

of large bodies in Proteus vulgaris OX-19. /. Bact., 61, 321-335.
Tatum, E. L. 1945. X-ray induced mutant strains of Escherichia coli. Proc.

Natl. Acad. Sci. U. S., 31, 215-219.
. 1946. Induced biochemical mutations in bacteria. Cold Spring Harbor

Symposia Quant. Biol, 11, 278-284.
Tatum, E. L., and J. Lederberg. 1947. Gene recombination in the bacterium

Escherichia coli. J. Bact., 53, 673-684.
Tatum, E. L. and D. D. Perkins. 1950. Genetics of microorganisms. Ann. Rev.

Microbiol., 4, 129-150.

28 ' •. SEX IN MICROORGANISMS

Witkin, E. M. 1951. Nuclear segregation and the delayed appearance of in-
duced mutants in Escherichia coli. Cold Spring Harbor Symposia Quant.
Biol., 16, 357-372.

Zelle, M. R., and J. Lederberg. 1951. Single cell isolations of diploid hetero-
zygous Escherichia coli. J. Bact., 61, 351-355.

Zinder, N. D., and J. Lederberg. 1952. Genetic exchange in Sahnonella. Ibid.,
64, 679-699.

Sex in Bacteria

Evidence from Morphology *

W. G. HUTCHINSON, Laboratory of Microbiology,
Department of Botan\', University of Pennsylvania,
Philadelphia, Pennsylvania

HENRY STEMPEN, Department of Bacteriology and Immunology,
Jefferson Medical College, Philadelphia, Pennsylvania

The assumption that bacteria, together with blue-green algae, repre-
sent unique types of organisms lacking any mode of sexual reproduc-
tion has for many years challenged investigators in their search for
such a phenomenon. Until recent years the methods of approach have
been largely cytological with little apparent effort to correlate cyto-
logical and genetic evidence in the same organism. Genetic evidence
for the existence of sexuality in bacteria is convincing (Lederberg,
1947; McElroy and Friedman, 1951), whereas the cytological evi-
dence thus far presented is at most only suggestive.

By analogy with known sexual organisms, the demonstration of
sexual reproduction in bacteria would require the fusion of gametes
or of cells morphologically similar to vegetative cells. Such cell fusion
could either be complete to form a single cell or incomplete as in
conjugation. After fusion, nuclear material from the participants
would be expected to fuse and later to segregate in preparation for
the ultimate repetition of the process.

Sexuality in bacteria has also been claimed to occur by the fusion
and segregation of nuclear material within single cells. Such a phe-
nomenon, as well as cell fusion, has been reported to precede spore
formation (as reviewed by Bisset, 1950, and Knaysi, 1951). On the
other hand, DeLamater and Hunter (1952) and Lamanna (1952)
discredit the occurrence of a sexual process in the formation of bac-
terial spores. Bisset (1950) has described "sexual vegetative reproduc-
tion" in which the nuclear material in vegetative cells fuses and is
later thought to segregate. Microcyst formation is likewise thought

* This work has been supported in part by the Philadelphia Lager Beer
Brewers' Association.

29

30 SEX IN AlICROORGANISMS

to involve a sexual process. What appears to be similar fusion of
nuclear material in several species of Bacillus and in Clostridium
perfringeiis has been induced by exposure of the organisms to low
concentrations of hydrogen peroxide (Cassel, 1951).

FUSION OF GAMETES

A sexual process in bacteria involving the fusion of male and
female gametes, "spermit" and "oit" respectively, has been described
by Enderlein (1925). It is claimed that aging bacteria give rise to
small bodies called gonidia, each of which becomes transformed into
a "gonit." These "gonits" do not reproduce as such on solid media,
but in certain liquid media they may convert themselves into the male
and female gametes. Enderlein maintains that the "spermit" morpho-
logically resembles a vertebrate sperm with an oval to circular head
and a connecting piece from which emerges a long, thin tail giving
active motility to this cell. The "oit" is a larger spherical form with
a flagellum inserted in a warty protuberance, although motility is
indistinct or slow. Union of "spermit" and "oit" occurs directly after
their formation and is supposedly followed by nuclear fusion.

These observations have apparently not been repeated nor more
convincingly documented by others. On the contrary, Henrici
(1928) has cast serious doubt on the validity of Enderlein's results.

FUSION OF BACTERIAL CELLS

Fusion of bacterial elements has been described as occurring by
a unique process in Azotohactev and a number of other bacteria
(Lohnis and Smith, 1916). Cells, spores, or gonidia in close proximity
reportedly undergo dissolution with a subsequent mixing of their
contents to form a "symplasm." Small granules later arise from the
"symplasm" and enlarge into coccoid or oval cells. This series of
events was postulated from the examination of fixed and stained
films in which it is, of course, impossible to follow any development.
Lohnis (1921) later attempted to study the formation of new bac-
terial cells from the "s\niiplasm" in hanging drop cultures, but with
unsatisfactory results.

In a number of instances spherical bodies found in bacterial cul-
tures have been reported to result from cell fusion. In Escherichia

SF.X IN BACTERIA— I'AIDINCI' 1"R()AI iMORFIIOLOGY 3]

L'oli (.Mellon, 1925) such spherical forms have been observed to one
side of the junction of two rods. Strands of w hat was considered to
be nuclear material were seen to extend into the round body from
the two rods. Although the nature of this material was not definitely
determined, the process was considered to be sexual in nature and
the round forms were called zygospores.

Stoughton (1932) observed a similar phenomenon in Bacterkn/i
Vhihiicearimi. His attempts to study subsequent development of the
zygospores in living preparations were only partially successful, be-
cause the multiplication of the ordinary rods was found to be so
rapid as soon to overgrow the particular cell under observation. Thus
most of the results were derived from the study of stained films.
Other spherical forms in B. inalvacearinn (Stoughton, 1929) may
arise as the result of budding from normal or slightly s\^ollen or oval
cells which occur in cultures aged for about 6 weeks.

The fusion of rods in pairs has also been described for a strain
o{ Bacteroides fimdulifoi'7jns (Smith, 1944). Organisms removed from
broth cultures at 3 -hour intervals were stained with Giemsa stain
without previous acid hydrolysis. During the first 6 hours the regular,
single rods contained granules which stained deep blue. At 9 hours
the rods had increased in length and were attached end to end in
pairs with the apposed ends swollen and filled with the deep blue-
staining material. Between 9 and 1 5 hours the swollen ends had appar-
ently fused so that the organisms appeared as long rods with a central
swelling. Between 18 and 21 hours the cultures consisted almost en-
tirely of large round bodies which presumably resulted from the
absorption of the rods into the central swelling. These round bodies
containing the deep blue-staining material in clumped or discrete
masses developed further by fractionating into ordinary rods, each
rod receiving a few granules of the deeply stained material. This
material was considered to represent a nuclear apparatus, although,
unfortunately, its nature was not experimentally characterized further
as, for instance, by the Feulgen reaction.

Utilizing a fixing and staining procedure of proved value in
cytological investigations of higher organisms, Lindegren and Mellon
(1933) postulated a possible sexual mechanism in the avian tubercle
bacillus. From aceto-carmine-stained preparations they described the
fusion of nuclei from adjoining coccoid cells with the subsequent
growth and division of the zygote.

32 SEX IN A4ICROORGANISMS

Using the technique so successfully employed by Robinow
(1942) in the demonstration of bacterial nuclei, Klieneberger-Nobel
(1949) has described cell and nuclear fusions in a number of different
bacteria. The process described was essentially as follows. Rods or
filaments break up into small cells designated primary cell units, each
of which consists of a nuclear granule surrounded by a thin layer of
cytoplasm. Fusion of two to many of the primary cell units occurs
when the cytoplasm of adjacent units coalesces. This is followed by
nuclear fusion which is said to involve the formation of "ramifica-
tions." Although the products of fusion, the L-bodies, are commonly
round, oval, or spindle-shaped, their size and shape may vary with
the number and arrangement of the cell units which fuse. These
L-bodies can develop further in an appropriate environment to pro-
duce normal-appearing rods. This type of process was reported to
occur in Bacteroides fzmdulifor7/?is, Streptobacilhis vwnilijorviis, and
a strain of Escherichia coli under normal cultural conditions; in Fro-
teus sp. under the influence of temperature changes; in E. coli-
vmtabile grown on nutrient agar containing lithium chloride; and in
E. coli-mutabile, Proteus sp., and Salmonella schottiniilleri grown on
penicillin-containing medium.

Klieneberger-Nobel based her interpretations on this particular
study on dead organisms. Certain of these interpretations are not in
agreement with the results of later studies made directly on living
cells. Stempen and Hutchinson (1951a) observed that when cell
fusion occurred in Proteus vulgaris OX- 19 only two cells were in-
volved in each fusion observed. When E. coli is exposed to non-lethal
concentrations of penicillin (Pulvertaft, 1952), aberrant forms re-
sembling the L-forms of Klieneberger-Nobel are produced; however,
all the aberrant forms noted developed as distortions of single rods
which did not divide.

CELL FUSION IN Proteus vulgaris OX- 19

Cell fusion has been demonstrated in Proteus vidgaris OX- 19 by
the direct observation of living cells in slide cultures (Stempen and
Hutchinson, 1951a). The fusion process most commonly begins with
the appearance of a budlike structure at the junction of a pair of rods
attached end to end (Fig. 8). This bud becomes larger as the contents
of the rods pass into it (Figs. 9 to 11). The rods are soon "absorbed"

SEX IN BACTERIA— EVIDENCE FROAl MORPHOLOGY 3 3

completely by the growing bud, which now assumes the appearance
of a round body, the so-called "large body" (Fig. 12). Figures 16 to
19 illustrate a fusion process in wliich the budlike structure appears
on one end of the pair of rods. In some cases, after a rod has divided,
the daughter cells fuse directly without bud formation to produce a
large body. In such instances it is clear that sister cells fuse. In the
other instances of fusion, the cells that fuse are already found touch-
ing end to end and it is impossible to determine whether these cells
represent daughter cells of a previous cell division or cells of different
origins. Cells that come in contact with each other as a result of
elongation or cro\^ ding during growth have never been seen to fuse.
It may be significant, however, that fusion through bud formation
has not been observed in known sister cells.

Fusion of rods in P. vulgaris OX- 19 occurs very infrequently.
Numerous fields must usually be followed before a single case of cell
fusion is observed. The exact frequency with which it does occur is
unknown, because cell fusion can be detected only by direct micro-
scopic observation and only a relatively few rods can be observed in
each microscopic field examined.

Large bodies in cultures of F. vulgaris OX- 19, however, are
fairly numerous. The majority of these forms arise not by fusion but
by the swelling or budding of single rods. The latter two methods of
large body formation are illustrated in Figs. 1 to 3 and 4 to 7.

The behavior of nuclear material (chromatinic bodies) within
the large bodies formed by cell fusion on the one hand and by bud-
ding or swelling of a single rod on the other hand would be inter-
esting to determine with certainty. At the present time some informa-
tion is available. In actively growing cells the areas corresponding in
position to the chromatinic bodies appear lighter than the rest of the
cell with the dark-phase contrast microscope, the reverse being true
with bright-phase contrast (Tulasne, 1949a; Stempen, 1950). When
this information is used in the examination of living organisms, the
chromatinic material in large bodies arising by cell fusion appears to
be in a compact mass immediately after fusion is complete. In other
cases where the large bodies do not arise by fusion, the interior of
these forms is composed of indistinct light and dark areas. This find-
ing suggests that the chromatinic material does not occur in a com-
pact mass but is more scattered throughout the interior of the large
body. Actually, in impression smears of organisms which had been

y^zw^/l»fi

29 '^ %^s 30 |jy**^%^|

Plate I. Figures 1 to 27 represent living cells in slide cultures (Stenipen
and Hutchinson, 1951a). Figures 28 to 31 represent cells from osniiuni-fixed
and fuchsin-stained preparations made from 6-hour-old cultures (Stempen and
Hutchinson, 1951b),

34

Sl-X IN BACTERIA— E\IDENCE ERO.M iMORPIIOLOGY 35

fixed, hydrolyzed, and stained by a method similar to that described
by Robinow (1942), one can find large bodies in which the chroma-
tinic material is in a compact mass (Fig. 28) and also those in which
the cliromatinic material is more distributed within the large body in
the form of granules, rods, or filaments (Figs. 29 to 31) (Stempen
and Hutchinson, 1951b), Any correlation between the results of the
two sets of observations must be considered suggestive only, for there
is no way of determining the mode of origin of the large bodies seen
in such stained films.

Regardless of whether the large bodies arise by the fusion of two
rods or from a single rod, they may undergo the same type of con-
tinued development. The large body may divide into two; each half
then divides, and this process is continued until the normal rod form
is restored (Figs. 12 to 15). In the restoration of the rod form from
the large bodies of Streptobacilhis nwiiUi^onms (Dienes, 1943) and
Bacteroides strains (Dienes and Smith, 1944), the rods are reported to
occur preformed in the large bodies; and Dienes (1946) states that
the process in Proteus is similar to that in Bacteroides.

On the other hand, large bodies of Proteus vidgaris OX- 19 may
develop in another way. The body ruptures; the contents appear as
a more or less homogeneous mass which is less refractile than the
body from which it arose. In a short time thereafter one or more

Figs. 1 to 3. Formation of a large body by the swelling of a single rod. The
culture is IV^, 3%, and VA hours old (x 2000).

Figs. 4 to 7. Formation of a large body by means of a lateral bud. The culture is
2, 2'/4, 2'/2, and 2% hours hours old (x 2000).

Figs. 8 to 15. Fusion of two rods to form a large body with the subsequent
division of the large body to produce rods. The culture is 1, l'/>, H4, 2, 2%, 4, 6/4,
and 714 hours old (x 2000).

Figs. 16 to 21. Fusion of two rods with the formation of a large body on one of
the pair of rods. The large body ruptures with the subsequent formation of a
microcolony of granular forms. The culture is 14, ^4, 1, Wi, IK, and 4 hours old
(X 2000).'

Figs. 22 to 24. Production of granular forms following the rupture of a large
body and an adjacent rod (x 2000).

Figs. 25 to 27. Production of granular forms associated with the outgrowth of
material from the large body. The large body later ruptures. The culture is 114, 2,
and 214 hours old (X 2000).

Fig. 28. Large body with chromatinic material concentrated in a single mass
(X 3100).

Figs. 29 and 30. Large bodies containing different amounts of chromatinic
material (X 3100).

Fig. 31. Large body with chromatinic material in a filamentous form (X 3100).

36 SEX IN MICROORGANISMS

coccoid, refractile, granular forms ranging from 0.2 to 0.5 micron
in diameter are suddenly evident at the periphery of the mass. The
number of the granular forms increases quite rapidly until a micro-
colony is produced (Figs. 22 to 24). In a few instances masses of
granular forms appear to extend out from the large body into the
medium (Figs. 25 to 27). This process is similar to that described for
B. fimdulifoTifiis (Dienes and Smith, 1944) in which the granular
forms (so-called Ll forms) grow out from the large body.

The granular forms of Proteus were not isolated in pure culture
by Stempen and Hutchinson. The failure to do this may be due to
the peculiarity of the strain employed, because Dienes (1949) and
Tulasne (1949b) have reported success. These granules are reportedly
capable of reverting to the rod form in subculture.

The similarity in behavior of large bodies regardless of how they
are formed indicates that a basic similarity exists among them. Be-
cause large bodies have been reported to occur in old cultures (Hen-
rici, 1925) or under the influence of injurious substances such as mer-
cury salts, lithium chloride, or penicillin (Dienes, 1946), for instance,
these forms have been considered by many to represent dead or
dying cells referred to as involution forms. That some of the large
bodies observed in cultures are degenerate forms may be true. That
this does not apply to all large bodies is shown by two principal
points. In P. vulgaris OX- 19, at least, large bodies are often formed
early during the growth cycle in a slide culture when conditions are
favorable for growth and division of the normal-appearing rods. Also,
the large bodies are capable of undergoing further development as
indicated above.

Whether or not large body formation by cell fusion represents
a sexual process cannot be stated with certainty. Determination of the
behavior of the nuclear material presents the problems outlined above.
A genetic approach to this problem is essential. If cell fusion does actu-
ally occur only between sister cells, it would be impossible to cross
contrasting strains.

CONJUGATION

Processes resembling conjugation liave been described for a
number of different bacteria. For descriptions of some of the earlier
claims, the reader is referred to the review by Lohnis (1921).

SI X IN BACTERIA— EVIDENCI-: FROAI MORPHOLOGY 37

Agrobiicterhmi tiivicjaciens, the causative agent of crown gall in
plants, produces star-shaped aggregates of cells in which the cells ap-
pear to radiate from a common center. Lohnis (1921) interpreted
such aggregates as conjunction. The term conjunction was used rather
than conjugation because frequently more than two cells unite and
there is no detectable sexual differentiation among them. The be-
havior of the nuclear material in such "stars" was studied with the
aid of the Feulgen reaction by Stapp (1942) and later by Braun and
EIrod (1946). These investigators suggest that nuclear material from
the component organisms fuses in the center of the aggregate; al-
though, as Braun and Elrod have pointed out, in many of the aggre-
gates the Feulgen-positive material remained confined to the cell but
concentrated in that part of the bacterium closest to the center of the
star. They, therefore, were reluctant to interpret the phenomenon as
conjugation because of the relatively few instances in which they
observed what appeared to be an actual fusion of nuclear material.

Using the same method of approach, the study of fixed and
stained cells, DeLamater (1951) described the process of conjugation
in Bacillus megaterimn. Vegetative cells, considered to be haploid,
form conjugation tubes which are attached to only one cell or form
a connection between the ends of two cells. At times a pair of tubes
was observed connecting the ends of two cells or a loop-like struc-
ture formed on the side of a chain of rods connecting two cells of the
chain. The extension of a conjugation tube from the end of one cell
into the side of another in a different chain was taken as evidence that
two cells of distinct origin could fuse. The simultaneous fusion of
three rods was also observed on one occasion. The nuclei are de-
scribed as migrating in both directions through these tubes so that
both of the cells are usually "diploidized." Subsequently, it is claimed,
the nuclei fuse and the cells enlarge and multiply vegetatively as dip-
loids.

DeLamater has observed conjugation tubes on blood agar base
plus 4 per cent human whole blood, on casein hydrolyzate medium,
and in all the various sugars in which secondary colonies form. It
was presumed that a relationship may exist between conjugation and
secondary colony formation. The results from one series of experi-
ments, however, show that, although secondary colonies can form on
media containing blood or serum from different animals, conjugation
tubes are formed only on media containing whole blood, plasma, or

38 SEX IN AIICROORGANISMS

serum from humans. It must be admitted that the time of samphng
may influence these results. Secondary colonies were formed on
media containing all of Cohn's blood fractions, but conjugation was
observed only with the combined fractions (I, II, III-O, III-l, 2, IV,
V, VI) and the fraction PGP (I, II, III), both of which contain
fraction III.

The occurrence of conjugation tubes in B. megaterhnn has been
challenged by Bisset (1952, 1953). It is pointed out that the rods of
this organism occur in chains and that shrunken rods are frequently
found joined at each end to their neighbors. These shrunken cells,
Bisset claims, have been mistaken for conjugation tubes whose in-
terior is presumed to be continuous with the two attached cells
because DeLamater's preparations fail to show the cross walls that
would be present.

As yet, serial photographs of the formation and behavior of con-
jugation tubes in living material have not been presented. If sufficient
optical path differences exist between the migrating nuclei and the
cytoplasm, it is theoretically possible to observe and photograph nu-
clear migration.

The genetic evidence of Hayes (1952) and of Lederberg and
co-workers (1952) implies that cytoplasmic fusion is probably not
involved when recombination occurs between certain mutants of the
K-12 strain of E. coli. Hayes showed that cells "killed" by strepto-
mycin were still capable of participating in recombination. He con-
sidered it probable that the viable cell extrudes genetic elements
which adhere to the cell wall. The genetic elements, being unaffected
by the streptomycin, can serve in recombination, whereas the killed
cell acts as a passive carrier.

CONCLUSIONS

It has been demonstrated beyond question from continuous ob-
servation of living material that cells of Proteus vulgaris OX- 19 may
fuse together in pairs. There is no final demonstration as yet that this
is a sexual process involving fusion and segregation of nuclear mate-
rial.

Less convincing are the numerous reports of cell and nuclear
fusion and of conjugation in several bacterial species, because the
reports are based upon fixed and stained material in which any se-

SEX IN BACTERIA— FAIDENCF. FROM iMORPHOLOGY 39

qucncc of cliangcs must he inferred by logic and analogy rather than
demonstrated by direct observation.

\Mien a film of fixed and stained cells is examined microscopi-
cally, one sees in each cell only one manifestation of the manifold
morphological transformations of which the organism may be capa-
ble. All the different stages may or may not be present in the same
film. Unless the behavior of living cells under as nearly the same
environmental conditions as possible is also studied microscopically,
tlie sequence of the stages is left to the imagination of the investigator.

Unfortunately, the bacteriologist with present-day methods is
definitely limited in the amount of detail which he can resolve in the
living cell. Therefore, fixed and stained cells must still be heavily
relied upon particularly for the differentiation of minute internal de-
tail. This, however, is no reason wdiy one should confine all morpho-
logical studies to dead cells when the behavior of cells as units in such
phenomena as ceil fusion and conjugation could be examined in the
living condition by the phase contrast microscope, an instrument well
adapted to the study of gross morphology of bacteria. By what better
way, techniques permitting, can one determine the behavior of indi-
vidual cells in a culture than by observing and photographing in
natural sequence the activities of these cells as they are actually per-
formed?

REFERENCES

Bisset, K. A. 1950. The Cytology and Life-History of Bacteria, pp. 76-80, 43-

48, 81-85. Williams and Wilkins Co., Baltimore.
. 1952. Spurious mitotic spindles and fusion tubes in bacteria. Nature,

169, 247.

1953. Do bacteria have mitotic spindles, fusion tubes and mitochondria?

/. Gen. Microbiol., 8, 50-57.
Braun, A. C, and R. P. Elrod. 1946. Stages in the life history of Phytoiiwnas

twiiefacievs. J. Bact., 52, 695-702.
Cassel, W. A. 1951. Apparent fusion of the chromatinic bodies in species of

Bacillus. J. Bact., 62, 514-515.
DeLamater, E. D. 1951. A new cvtological basis for bacterial genetics. Cold

Spring Harbor Symposia Quant. Biol., 16, 381-412.
DeLamater, E. D., and M. E. Hunter. 1952. The nuclear cvtology of sporula-

tion in Bacillus viegateriwn. J. Bact., 63, 13-21.
Dienes, L. 1943. Reproduction of bacteria from the large bodies of Strepto-

bacillus nioniliforTiiis. Proc. Soc. Exptl. Biol. Med., 53, 84-86.

40 SEX IN A4ICROORGANISMS

, 1946. Complex reproductive processes in bacteria. Cold Spring Harbor

Syjnposia Quant. Biol, 11, 51-59.

1949. The development of Proteus cultures in the presence of penicillin.

/. Bad., 57, 529-546.
Dienes, L., and W. E. Smith. 1944. The significance of pleomorphism in Bac-

teroides strains. /. Bact., 48, 125-154.
Enderlein, G. 1925. Bakterien-Cyclogejiie. pp. 110-129. Walter de Gruyter and

Co., Berlin and Leipzig.
Hayes, W. 1952. Recombination in Bact. coli K-12: uni-directional transfer of

genetic material. Nature, 169, 118-119.
Henrici, A. T. 1925. A statistical study of the forms of growth of the cholera

vibrio. /. Injections Diseases, 37, 75-81.
. 1928. Morphologic Variation and the Rate of Growth of Bacteria, pp.

5-6. Charles C Thomas, Springfield, 111.
Klieneberger-Nobel, E. 1949. Origin, development, and significance of L-forms

in bacterial cultures. /. Gen. Microbiol., 3, 434-443.
Knaysi, G. 1951. Elements of Bacterial Cytology. 2nd ed. pp. 237-239. Com-

stock Publishing Co., Inc., Ithaca, New York.
Lamanna, C. 1952. Symposium on the biology of bacterial spores. Part I. Bio-
logical role of spores. Bact. Revs., 16, 90-93.
Lederberg, J. 1947. Gene recombination and linked segregations in Escherichia

coli. Genetics, 32, 505-525.
Lederberg, J., L. L. CavalH, and E. M. Lederberg. 1952. Sex compatibility in

Escherichia coli. Ge77etics, 37, 720-730.
Lindegren, C. C, and R. R. Mellon. 1933. Nuclear phenomena suggesting a

sexual mechanism for the tubercle bacillus. Proc. Soc. Exptl. Biol. Med.,

30, 110-112.
Lohnis, F. 1921. Studies upon the life cycles of the bacteria. Mem. Natl. Acad.

Sci. U. S., 16, 1-252.
Lohnis, F., and N. R. Smith. 1916. Life cycles of the bacteria. /. Agr. Research,

6, 675-702.
McElroy, W. D., and S. Friedman. 1951. Gene recombination in luminous

bacteria. /. Bact., 62, 129-130.
Mellon, R. R. 1925. Studies in microbic heredity. I. Observations on a primitive

form of sexuality (zygospore formation) in the colon-typhoid group.

/. Bact., 10, 481-501.
Pulvertaft, R. J. V. 1952. The effect of antibiotics on growing cultures of

Bacterium coli. J. Path. Bact., 64, 75-89.
Robinow, C. F. 1942. A study of the nuclear apparatus of bacteria. Proc. Roy.

Soc. (London), B130, 299-324.
Smith, W. E. 1944. Observations indicating a sexual mode of reproduction in

a common bacterium (Bacteroides fundidiformis). J. Bact., 47, 417.
Stapp, C. 1942. Der Pflanzenkrebs und sein Erreger Pseudomonas tumefaciens.

XI. Zytologische Untersuchunger des bacteriellen Erregers. TLentr. Bakt.

Parasitenk., II, 105, 1-14.

SEX IN BACTERIA— r.VlDENCK FROM MOKPI lOI ,()( iV 41

Stempcn, H. 1950. Demonstration of the clironvatinic bodies of Escherichia

colt and Proteus vulgaris w itli the aid of the phase contrast microscope.

/. Bact., 60, 81-87.
Stempcn, H., and W. G. Hutchinson. 1951a. The formation and development

of large bodies in Proteus vulgaris OX-19. I. Bright phase contrast obser-
vations of living bacteria. /. Bact., 61, 321-335.
. 1951b. The formation and development of large bodies in Proteus

vulgaris OX-19. II. Comparative cytology of bacilli and large bodies. /.

Bact., 61, 337-344.
Stoughton, R. H. 1929. The morphology and cytology of Bacterium inalva-

cearuw, E. F. S. Proc. Roy. Soc. (London), BIOS, 469-484.
. 1932. The morphology and cytology of Bacterium malvaceartmi,

E. F. S. Part II. Reproduction and cell fusion. Proc. Roy. Soc. {London),

Bill, 46-52.
Tulasne, R. 1949a. Mise en evidence du noyau chez les bacteries vivantes grace

au dispositif a contraste de phase. Compt. rend. soc. biol., 229, 561-563.
. 1949b. Existence of L-forms in common bacteria and their possible

importance. Nature. 164, %16-%11.

Life Cycles, Sexuality,

and Sexual Mechanisms in the Fungi*

JOHN R. RAPER,t Department of Botany,
University of Chicago, Chicago, Illinois

The fungi were once characterized, with considerable justification,
as "a mutable and treacherous tribe." Probably no other character-
istic or activity of the fungi contributed so prominently to this epi-
thet as sex and the phenomena associated with sex. For the better part
of a century the problem of sex in fungi has received a great deal of
attention among students of the group, and a tremendous literature
has accumulated through the years. The problem, however, seems to
grow a trifle faster than does the solution, resulting in the interesting
situation, like that described by Lewis Carroll, of losing only little
ground by running very fast.

During the early decades of the century numerous scholarly pub-
lications summarized the existing information and integrated it into
the more comprehensive problem of sexuality in plants and animals.
The more notable of these works were Kniep's Die Sexnalitat der
niederen Fflanzen of 1928, Gaumann's Vergleichende Morphologie
der Pike of 1926, and Dodge's translation and revision of this work in
1928, Link's highly intellectual review of reproduction published the
following year, and, more recently, Hartmann's Die Sexitalitat, pub-
lished in 1943. The implications of sexuality in fungi, however, re-
main largely unknown to biologists in general.

Several fungi, each carefully chosen to combine a number of
specifically required characteristics, have recently been used as near-
ideal research tools for the elucidation of basic phenomena of uni-
versal biological importance. More extended use along these lines,
however, depends largely upon a greater awareness among biologists

* The compilation of material for and the preparation of this review have
been materially aided by a grant from the Dr. Wallace C. and Clara A. Abbott
Memorial Fund of the University of Chicago.

f Present address: Biological Laboratories, Harvard University.

42

LIFE CYCLES, SEXUALIl V, AND SEXUAL AlECIL\NISiMS 43

of rhc peculiar hcnctirs offered by tlic fungi because of the variety
of sexual and cultural characteristics which they possess. Equally
important, of course, is an awareness of the limitations to their use.

No comprehensive, up-to-date source of such information is now
available. This paper presents, in outline form, a comparative review
of sex in fungi which might serve as a preliminary sketch for a source
of this kind.

Essential sexual processes may be defined as those processes req-
uisite to and including the juxtaposition and fusion of compatible
nuclei and the subsequent sorting out of genetic factors in meiosis.
These processes impose a cyclic progression of which plasmogamy,
caryoo-amy, and meiosis are the irreducible cardinal events. The cycle,
however, may be basically varied in three different ways: (1) by
variations of the temporal relationships between the cardinal events
by the intercalation, at different stages, of essential processes of
growth; (2) by the imposition of genetic restrictions upon universal
compatibility; and (3) by variations in the mechanical means of ac-
complishing the cardinal events.

These three modes of variation determine three distinct facets of
sexuality, all separately definable but inextricably interrelated in the
living plant: (1) life cycle, (2) basic pattern of sexuality, and (3)
the sexual mechanism per se, respectively. Each facet is understand-
able only as a time-integrated and dynamic process. A detailed exam-
ination of each facet brings to light a number of facts w^hich are little
known but which are of considerable biological interest and are
essential to an appreciation of the broad implications of sexuality in
the fungi.

LIFE CYCLES

The fungi are commonly considered organisms which are essen-
tially haploid — perhaps with nuclear fusion occurring now and again
to give rise to a diploid phase which persists for a single nuclear gen-
eration. Although this is true of many species, particularly among the
more primitive groups, the regular occurrence of exceptions to this
simple pattern among the lower groups and the various complexities
of the life cycle patterns characteristic of the more highly evolved
forms make such a generalization meaningless. Life cycles among
fungi run the gamut from completely haploid at the one extreme to
completely diploid, minus the immediate products of meiosis, at the

44

SEX IN A'lICROORGANISMS

Other, and a unique nuclear association, known as the dicaryon,
greatly increases the range of life cycle variability.

Since the dicaryon effects important changes in the life cycle
and is peculiar to the fungi, it deserves a brief description and illus-
tration at this point (Fig. 1.) The essential portion of the sexual
process is initiated by the fusion of two sexual cells or organs, each
containing one or more haploid nuclei (N). This fusion has been
termed plasinogamy or cy to gamy. The nuclei provided by the fusing
elements may retain their individuality and become associated in one

Haploid (N) {^.^ (y
Plasmogamy '!>.

Dicaryon (8)

Conjugate

Division H °

Caryogamy
Diploid (D) V €

Fig. 1. Schematic representa-
tion of the initiation, multipHca-
tion, and termination of the di-
caryotic association of compatible
nuclei. The dicarvon occurs in the
higher Ascomycetes in the ascog-
enous hyphae and universally in
the Basidiomycetes in the "second-
ary" (dicaryotic) mycelium.

or more pairs, each pair known as a dicaryon (B). The dicaryon may
be propagated for a short or for an indefinite period of time by re-
peated, simultaneous mitotic divisions of its members, the division
figures of the two components commonly lying side by side. This
process is termed conjugate division. Fusion of the two associated
nuclei, or caryogamy, eventually occurs in terminal binucleate cells
to establish the diploid phase (D). The dicaryotic phase thus serves,
when present, to effect a temporal and spatial separation of plas-
mogamy and caryogamy. It also serves, because of the repeated
divisions of its component nuclei, to increase greatly the produc-
tivity per sexual fusion both in numbers and in possible genetic
recombinations.

LIFE CYCLES, SEXUALITY, AND SEXUAL MECHANISA1S

45

Seven basic types of life cycles can be clearly distinguished;
these are diagranimatically represented in Fig. 2 and are designated
by the letters A to G. Changes in nuclear phase are here considered
the cardinal events in the life cycle. These changes are indicated for
each cycle, progressing clockwise, by differences in shading.

L/FE CYCLES IN FUNGI

NUCLEAR
PHASE

Fiu. 2. Schematic comparison of life cycles in fungi.

In each cycle changes in nuclear phase are indicated, progressing clockwise, hy
changes in shading. The double vertical line at the top of the diagram represents
meiosis, and each of the two narrow sectors adjacent to the line represents a single
nuclear generation.

Asexual reproduction by spores or other specialized organs oc-
curs at least in certain species belonging to each type and with few
exceptions propagates the phase of the cycle from which the special-
ized reproductive cells are derived; the few exceptions will be men-
tioned later.

Asexual Cycle

Species apparently lacking any sexual expression or alternation
of nuclear phase are fairly common throughout the fungi and con-

46 SEX IN MICROORGANISMS

stitute a sizable proportion of all known species, approximately 20
per cent according to Bessey (1950). Because of the failure to ob-
serve rarely occurring sexual stages, the actual number of exclusively
asexual species must be somewhat less than reported, but it must still
be very large. The entire group known as the Fungi Imperfecti be-
longs here as well as numerous species which are clearly assignable
by morphological characteristics to various groups throughout the
perfect fungi, such as PenicilHwii iiotatniu, the producer of the drug
penicillin.

Certain of the benefits of sexuality are provided in many sexually
sterile species by the association of nuclei of different origins in heter-
ocaryotic mycelia, in which different genetic characters are ex-
pressed in much the same way as in dicaryotic mycelia or in diploid
organisms (Pontecorvo, 1946). Of particular interest in this connec-
tion is the recent demonstration by Pontecorvo of recombination in
low frequency of genetic factors in heterocaryotic fungal systems
similar to that shown by Lederberg and Tatum (1946) in bacteria.
The exact mechanism whereby such recombinations are achieved in
fungi has not been fully elucidated. It has been shown, however, that
there are formed occasional diploid nuclei, heterozygous for the
characters carried by the heterocaryotic components, and that these,
through a pseudo-meiotic rearrangement lacking reduction, produce
diploids which are homozygous for one or more characters and which
represent new genetic combinations (Pontecorvo and Roper, 1952;
Roper, 1952).

HAPLom Cycle

The predominant type of life cycle found in the Phycomycetes
and the more primitive Ascomycetes is completely haploid with
the exception of a single, diploid, nuclear generation, the fusion or
zygote nucleus. This type of life cycle is the simplest possible one
that allows for sexual fusion and the recombination of genetic fac-
tors and in all probability represents the primitive type from which
the more complicated cycles have evolved. The general correlation
between this type of cycle and the relative morphological simplicity
of the forms exhibiting it, not only in the fungi but also in the algae,
would tend to support this view.

LIFE CYCLES, SEXUALll Y, AND SEXUAL MECHANISMS 47

Haploid Cycle wuh Risiricikd Dicarvon

A prcdoniinnnrlv haploid cycle, which differs from the one dis-
cussed above by the separation in space and time of plasmogamy and
caryogamy, is characteristic of members of the higher Ascomycetes
such as Neiirospora. At the time of the fusion of the sexual cells or
organs one or more dicaryotic pairs of nuclei are formed, and these,
by repeated mitotic divisions in the ascogenous hyphae, provide
paired nuclei for a large number of ascal primordia within which
caryogamy and meiosis occur. The multiplication of associated nu-
clei, though often extensive, is nevertheless restricted both in time
and by the complete dependence of the ascogenous hyphae upon the
haploid mycelium. The nature of the dicaryotic phase here would
suggest for this type of cycle an evolutionary position intermediate
between the exclusively haploid cycle and the more complex cycles
to be found among the Basidiomycetes.

Haploid-Dicaryotic Cycle

The predominant life cycle in the Basidiomycetes, excluding
many of the smuts, differs from the cycle just discussed by the unre-
stricted and independent growth of the dicaryotic phase. Both the
haploid, or homocaryotic, phase and the dicaryotic phase are com-
pletely independent and capable of indefinite vegetative growth and
are terminated by dicaryotization and fruit body formation respec-
tively. The termination of each phase depends upon the achievement
of certain requirements which is, in each case, largely a matter of
chance. The cycle may therefore be considered to comprise two
roughly equivalent phases and terminate in a single diploid nuclear
generation, the fusion or definitive nucleus in the basidium.

A number of cases have been described among these forms in
which differentiated spores produced by the dicaryotic mycelium
re-establish the haploid phase (Brodie, 1931; Nobles, 1942). This oc-
curs through the separation of the members of conjugate pairs of
nuclei in the formation of uninucleate conidia or oidia. These spe-
cialized cells appear to attain their greatest effectiveness as fertilizing
(dicaryotizing) agents, although germination in low percentage does

48 SEX IN MICROORGANISMS

serve to sort out the original dicaryotic components into liaploid,
vegetative mycelia. Similar cells produced on haploid mycelia behave
in an identical manner.

Dicaryotic Cycle

The extreme development of the dicaryotic phase is exemplified
by the cycle in which the immediate products of meiosis, ascospores
or basidiospores, fuse to initiate the dicaryotic phase. Both haploid
and true diploid phases are thus reduced to single nuclear generations.
This type of cycle is occasionally seen in the yeasts (Guillermond,
1940) and is of common occurrence among the smuts (Kniep, 1926).

The distinction made here between the haploid-dicaryotic and
the dicaryotic cycles emphasizes the two extremes in what, in all
probability, is a continuous series. Chance juxtaposition of compatible
germinating spores of the haploid-dicaryotic type might result in the
typical dicaryotic cycle; on the other hand, the experimental pro-
longation of the haploid phase, as sprout mycelia in the smuts for
example, converts the typical dicaryotic cycle into the haploid-
dicaryotic.

Of particular interest in this and the haploid-dicaryotic cycle is
the failure of the dicaryon in many cases to constitute a physiological
summation of its haploid components. This phenomenon is reflected
in ( 1 ) the host specificity of the two phases in the heteroecious rusts
(for example, the haploid phase of the "black stem rust" of wheat is
an obligate parasite of the barberry, whereas the dicaryotic phase is
an obligate parasite of grasses); (2) the saprophytic habit of the
haploid phase versus the obligate parasitic habit of the dicaryotic
phase of many smuts (Christensen and Rodenhiser, 1940); and (3)
the complex pattern of fruiting requirements of the dicaryotic phase
of certain Hymenomycetes as compared to the nutritional require-
ments of the component homocaryons (Schopfer and Blumer, 1940).

Haploid-Diploid Cycle

The alternation of haploid and true diploid generations, a com-
mon type of cycle in the algae and in the higher plants, occurs among
fungi only in certain members of the aquatic phycomycetous order,
the Blastocladiales, with Allomyces the best known example (Couch,

LIFE CYCLFS, SFXUALH V, AND SEXUAL MECHANISAIS 49

1942; Emerson, 1941; Harder and Sorgel, 1938; Kniep, 1929, 1930).
The vegetative nivcclia of the t\\o generations are identical except
for the specialized reproductive organs which they bear.

Diploid Cycle

The cycle that is typical for the animal kingdom, completely
diploid except for the immediate products of meiosis, is found among
the fungi in a number of yeasts (Guillermond, 1940; Winge, 1935)
and perhaps in some members of the phycomycetous order, the Blas-
tocladiales (Couch, 1942; McCranie, 1942). The latter case consti-
tutes a slight variation of this cycle, as Wilson (1952) reports a single
mitotic division of the meiotic products in AUoviyces cystogemis
prior to gamete differentiation. A regular haploid phase of two
nuclear generations is unique among the fungi and, strictly, should
be considered yet another type of life cycle.

BASIC PATTERNS OF SEXUALITY

Blakeslee, in 1904, in the course of an investigation of zygospore
formation in the common "black bread mold," Rhizopiis nigricans,
first demonstrated "bisexuality" in the fungi. The term hetevothallisin
was introduced to designate the occurrence, within a given species,
of two kinds of individuals, each self-sterile and presumably differ-
ing in sexual sign, and the necessity of interaction between mycelia
of the two kinds to accomplish sexual reproduction. The term hovio-
thallisin was coined for the antithetical condition, the occurrence of
only a single kind of individual, self-fertile and sexually self-sufficient.

The original definitions of homothallism and heterothallism, un-
fortunately, however, were somewhat ambiguous. The derivation of
the term heterothallism implies differences of any sort between the
individuals required for sexual interaction, whereas the original defi-
nition strongly implied differences in sexual sign. That Blakeslee was
convinced of the sexual nature of the race differences, in spite of the
slight and inconstant morphological differences, is strongly indicated
by the work done by him and his associates during three decades
toward the definite identification of ( + ) and ( — ) as ? and $ re-
spectively (Satina and Blakeslee, 1928, 1929).

Blakeslee and other workers (1920) determined the pattern of

50 SEX IN AIICROORGANISMS

sexuality in most of the members of the Mucorales, the order of
fungi to which "black bread mold" belongs. It is of interest in the
present discussion that all species of this group having a sexual stage
were unambiguously divisible into heterothallic and homothallic
groups, and that in each heterothallic species no irregularities in re-
spect to sexual sign were encountered, although individual isolates
often varied widely in sexual potency.

In the half century that has elapsed since Blakeslee's first demon-
stration of obligatory intermycelial reaction for sexual reproduction,
similar conditions have been reported for some members of every
major group of fungi. The necessity for intermycelial reaction, how-
ever, is the only feature common to all cases: in some, sexual differ-
ences are clearly evident; in others, sexual differences equally clearly
do not account for the pattern of self-sterility and cross-fertility.
With the discovery of the several patterns of interactions, each differ-
ing in some important respect from that oricrinally described in the
Mucorales and termed heterothallism, a number of proposals have
been made either to differentiate, by appropriate terms, these cases
from heterothallism as originally defined or to redefine heterothallism
in a more broadly inclusive manner. The chief result of these efforts
has been to indicate the degree of prevailing confusion rather than
to contribute to a unified scheme of categorization which was rea-
sonably free of damning objections.

Whitehouse (1949), in a very comprehensive and excellent re-
view, has recently advanced a logical system of differentiation which
promises to clarify considerably the entire subject of sexuality in the
fungi. He retains, on rational grounds and with historical justification,
the term heterothallism to include all those cases in which inter-
mycelial reaction is a requisite for sexual fusion. Two major types of
heterothallism are distinguished: (1) viorphological beterothaUisni, to
include those cases in which the two interacting thalli differ by pro-
duction of morphologically dissimilar sexual organs or gametes which
are identifiable as i and 9, and (2) physiological heterothallism, to
include those cases in which the interacting thalli differ in mating
type, or compatibility, irrespective of the presence or absence of sex-
ual organs or differentiated gametes per se. Homothallism is retained
in the original sense: sexual fusion between elements of the same
thallus or, in unicellular ors^anisms, between individuals of the same

1, 11 I CYCli S, SIXUAl.n Y, AND SFXUAL AlKCHANISMS 51

clone. A new rcriii, secaiidiVy bo/z/othnllis///, is applied to self-fertile
hctcrocarvons. These will he discussed in detail later.

Inevitably, there exist a number of forms which fit uneasily into
a simple breakdown of this sort; in a group of organisms as varied as
the fungi, this situation almost necessarily follows any attempt at
categorization in respect to characteristics of the mature thalli. A
somewhat less ambio;uous system could be erected on the distinction
between phenotypic and genotvpic determination of sexual or mating
behavior or both. The two major groupings here would be based
upon the ability or inability of genetically identical nuclei (sister
nuclei, daughters of a single primary meiotic product) to participate
in sexual fusion. Such a distinction would roughly parallel that be-
tween homo- and heterothallism. \\'ide acceptance and common usage
of the homo-heterothallism concept, however, dictate its perpetuation
without radical change in spite of its intrinsic shortcomings. Recog-
nition of the pattern of segregation at meiosis as the chief, and fre-
quently the sole, factor in determining the ultimate sexual character
or mating behavior, or both, of the thallus, however, results in a far
clearer understanding of the homo-heterothallism concept.

Each mature thallus, at the stage in its development at which
sexual fusions occur, commonly contains nuclei of only a single kind;
that is, they are hovwcaryotic (a number of important exceptions to
this generalization will be considered later). These sexually mature
thalli thus represent the expressed potentialities determined at meiosis
and imparted to the spores that constitute the immediate products
of this process. Spores, and the thalli into which they develop, may
be divided into four types in respect to the segregation of determining
sexual or mating capacities; (1) segregation of sexual factors, (2)
segregation of incompatibility factors, (3) segregation of sexual and
incompatibility factors, and (4) segregation of neither sexual nor
incompatibility factors. In the simple cases under consideration, spores
of types 1, 2, and 3 give rise to thalli which are self-sterile but which
are cross-fertile in those combinations bringing together complemen-
tary sexual or incompatibility factors. Such forms are clearly hetero-
thallic. Spores of type 4, on the other hand, produce thalli of only
a single kind, all of which are self-fertile; such forms are homothallic.

A number of complicating phenomena tend to mitigate some-
what the simplicity of this picture. Foremost among these is the reg-

52 SEX IN MICROORGANISMS

ular association, initiated in spore formation in certain species, of two
kinds of nuclei of dissimilar sexual or incompatibility types in a
single thallus which is self-fertile. A form of this kind, in spite of its
segregative pattern and the necessity of genetically dissimilar nuclei
for sexual fusion, must be termed homothallic because of the self-
fertile nature of its thallus.

A second complication is the possibility of final determination of
sexual or mating behavior, in forms lacking this determination at
meiosis, by environmental factors during the development of the
thallus. A physiological differentiation of this sort between individual
cells or groups of cells within a single thallus constitutes typical
homothallic behavior; if the final differentiation involves different
thalli, however, it must be termed heterothallic because of the self-
sterile nature of the sexually mature thalli. Students of different
groups of fungi have shown somewhat less than ideal accord in their
integration of phenotypic determination with the homo-heterothallism
concept. The general acceptance of the concept of the clone, now
frequently ignored except in the study of unicellular forms, would
resolve the more important discrepancies in interpretation of these
phenomena.

A third mode of deviation from a strict dichotomy between
homo- and heterothallism may arise through mutations of factors
controlling mating behavior or modifying sexual expression.

These departures from strict homo- and heterothallism will be
considered later in connection with the detailed accounts of the vari-
ous patterns of mating behavior.

HOMOTHALLISM

Of the several distinct patterns of sexuality to be found among
the fungi, homothallism is the most common; it occurs in all major
groups and, with very few exceptions, in a majority of species within
each group. The critical differentiation of compatible elements is
intramycelial and may involve single cells or relatively large groups
of cells. The spatial relationship between differentiated elements of
the fusing pair is also variable. This variability may best be illustrated
by certain species in the aquatic phycomycetous order, the Saproleg-
niales: (1) <^ and 9 elements may together constitute a specialized

LIFE CYCLES, SEXUALITY, AND SEXUAL MECHANISMS 53

lateral hvphal branch, a stalked cxigoniuni with an anthcridial cell
either differentiated in the stalk or arising from it; (2) $ and 9 ele-
ments may arise from adjacent sections of main hyphae; and (3) $
and 9 elements may arise from different main hyphae, each main
hypha being differentiated in its entirety as c^ or as 9 (Coker, 1923).

Differentiation of sexual elements within a single thallus is usu-
ally reversible, either to the vegetative state or in some cases to sexual
organs of the opposite sign. The vegetative development of unfused
9 gametes of Alloinyces (Emerson, 1941) and the ability of the
differentiated sexual organs and even of isolated 9 gametes of homo-
thaUic species of Achlya to regenerate normal hermaphroditic plants
are typical examples of such reversibility (unpublished observations).
A more extensive reversibility, from sexual organs of one sign to
organs of the opposite sign, is fairly common in the homothallic water
molds. The production of antheridial hyphae from oogonia and the
occurrence of small oogonia intercalated in antheridial hyphae have
been observed in various species (Coker, 1923; Humphrey, 1892;
Maurizio, 1899). It has also been demonstrated in several homothallic
species that sexual hormones from strongly sexed plants caused oogo-
nial intials to redifferentiate and produce antheridial hyphae (Raper,
1950).

One further point in connection with true homothallism is of
general biological interest. Sexual fusion normally occurs between
elements carrying sister (genetically identical) nuclei. This would
imply, a priori, that most fungi are deprived of the benefits occurring
in the recombinations of genetic factors following sexual fusion be-
tween dissimilar elements. Two facts would tend to mitigate this dep-
rivation: (1) the separate histories, often extended, of the two sister
nuclei brought together in the sexual act allow considerable oppor-
tunity for the accumulation and recombination of minor differences
due to induced or spontaneous mutations (Pontecorvo, 1947, 1950;
Pontecorvo and Roper, 1952; Roper, 1952); and (2) juxtaposed
thalli having totally different origins allow for extensive cross breed-
ing and hybridization in forms wdth motile or non-motile differen-
tiated gametes (Emerson, 1941, 1950) and for occasional cross breed-
ing and even hybridization in forms lacking free gametes (Raper,
1950; Salvin, 1942). The extent to which either or both of these
phenomena might duplicate in nature the benefits of enforced cross

54 SEX IN MICROORGANISMS

breeding cannot be accurately assessed. The widespread occurrence
of homothallism in fungi, however, is eloquent testimony of the evo-
lutionary success of this pattern of sexuality among these forms.

Superficially it would appear that in homothallic species, typi-
cally uniting genetically identical nuclei, the usual sexual endeavor
approaches a total sacrifice of quality for quantity; the exceptional
cases which prevent the accomplishment of this biological absurdity
appear to provide sufficient variability to allow for necessary adapta-
tion and survival.

Heterothallism

Six basic patterns of sexuality have been called responsible for
heterothallism among fungi. Beyond the single requirement for het-
erothallism, that the sexual act involve two individuals, these several
patterns are quite distinct.

The distinctions between the various basic patterns imposing
intermycelial mating reactions have been recognized by many au-
thors, several of whom have proposed new terms for one or more of
the seemingly coordinate patterns to distinguish them from hetero-
thallism as originally defined. Some of these terms have been widely
accepted and now constitute useful components of our working vo-
cabulary; others have probably deserved the oblivion to which they
have been relegated. It is certainly not the purpose here to add to
this burden of awkward descriptive terms, but rather to differentiate
as concisely as possible between a number of patterns which are based
upon distinct genetic devices, are quite similar superficially, and which
accomplish a common purpose.

The basic segregative mechanisms responsible for the six differ-
ent heterothallic patterns are diagramed in Fig. 3. The order within
the comparative listing here is not intended to convey any intimations
of phylogenetic or evolutionary significance.

In typical heterothallic species the immediate products of meio-
sis, spores of one sort or another in most cases, differ among them-
selves in respect to either sexual sign, or incompatibility factors, or,
in one known case, both sexual sign and incompatibility specificity.

The necessary use here of both sexual and incompatibility factors
forces upon the reviewer the most unwelcome chore of attempting
to distinguish concisely between the two; the onerous fact that a

MM CVCIA S, SFXUALI rV, AND SIXUAL AlECHANISiMS

55

clear distinction is impossible at the present time unfortunately does
not excuse him from making the attempt. Sexual factors are those
genetic determiners which characteristically result in morphologically
distinguishable S and 9 plants, or A and 2 sexual organs or gametes
or both. The criteria for the designation of S and 9 organs or cells
or both are largely borrowed from mammalian reproductive processes
and include relative size, inclusion of reserve food materials, motility,

B/IS/C PATTERNS OF SEXUALITY IN FUNGI

'MEIOSIS

Spore Types:
Sex

Compatibility

Somotic

Copulation

Sexual Organs
or Cells

HOMO-
THALUCI

NUCLEAR
<=^=FUSION

(±)

d9

(-) (+)

cf9

{-) (+)

(-) {+)

HETEROTHALLIC

^^^^^^9

6^d^

-9*9

m

I I

A a

A A

9c5'd"9

12

9d'd'9
I I 11

A A a a

9^(59

A 0

21

ABAbabaB

Fig. 3. The genetic devices that underlie the seven distinct patterns of

sexuaHty in fungi.

All individuals of a species of tvpe O are alike and functionallv hermaphroditic;
individuals of species of types I to VI arc divided, by sexual or incompatibilitv
differences, into two or more distinct mating strains among which cross-breeding
is obligatory. Multiple allelic series at the incompatibility loci commonly occur in
types \" and M, and the number of distinct mating strains of a single species of
the latter type may be of the order of ten thousand.

and particularly the direction of nuclear migration (or transport) in
fertilization. Incompatibility factors, by contrast, are those extra-
sexual genetic determiners of mating capacity which operate either
in addition to or in the absence of sexual factors. The difficulty of a
sharp, universally applicable distinction between the two arises pri-
marily from the occasionally known occurrence of heterothallic spe-
cies in which demonstrable sexual differences exist in the absence of
morphological differentiation. The best known example is the com-

S6 SEX IN MICROORGANISMS

plex of interbreeding species of the green alga, Chlaviydoinonas,
which comprises heterogamous, anisogamous, and isogamous forms
(Moewus, 1950). In other cases where sexual sign cannot be tested
by cross-breeding with sexually differentiated forms, it is impossible
to make a certain distinction between sexual and incompatibility
control of mating behavior in the absence of clear morphological dif-
ferences. This difficulty will be apparent in the following description
of segregative patterns.

Single Alternate Sexual Factors. The simplest pattern of sexual
differentiation yields two classes of progeny, each of which is either
immediately distinguishable as .5 or 9 or bears differentiated ($ or ?
sexual organs or gametes respectively or both. Sexual dimorphism is
typically rigid in plants belonging to this category.

Relatively few groups of fungi contain species which unques-
tionably show this type of differentiation. Among the more primitive
monoflagellated aquatic fungi numerous species produce thalli which,
at maturity, are differentiated into single gametangia with clear mor-
phological distinction between S and ? (Couch, 1942; Harder and
Sorgel, 1938; Sparrow, 1943). Such forms are obviously heterothallic;
whether the differentiation of the indvidual as a S or as a ? is phe-
notypically or genotypically determined, however, remains uncertain
although intensive efforts have been made to resolve the problem
(Cantino and Hyatt, 1953; Emerson, 1950). Certain groups among
these primitive fungi, particularly Blastocladiella of the Blastocladi-
ales, constitute series grading from clear distinction between $ and
9 thalli to forms that show no morphological difference between the
two mating types (Stiiben, 1939). The sure distinction here between
sexual factors and incompatibility factors is not possible, but it would
seem to the reviewer, in disagreement with the views of Whitehouse,
that here as elsewhere a common pattern of sexuality most probably is
shared by the members of a closely related group.

Sexual dimorphism is also known among the members of a few
groups of Ascomycetes, in Stromatinia narcissi of the Discomycetes
(Drayton and Groves, 1952), in many species of the Laboulbeniales
(Benjamin and Shanor, 1950; Thaxter, 1908), and in Pericystis
(Clausen, 1921).

The heterothallic members of the Mucorales, to which "black
bread mold" belongs and in which heterothallism was first discovered,
have long been cited as the classic examples of sexual segregation

LIFE CYCLES, SEXUALITY, AND SEXUAL MECHANISMS SI

among the fungi. Recent work and re-examination of earlier studies,
lio\\cvcr, leave such an interpretation in some doubt; current views
rather favor the determination of mating type in these forms by
incompatibility factors (Whitehouse, 1949).

Multiple Altenmte Sexual Factors. A somewhat more compli-
cated type of sexual segregation than that immediately preceding in-
volves the determination of several sexual strains, each typically self-
sterile but cross-fertile with all others. A heterothallic species of this
tvpe constitutes a linear series of sexual strains, each of which, with
the exception of the t\^'o terminal strains, reacts as 5 or as 9 depend-
ing upon its position in the series relative to that of its mate; each
of the two terminal strains reacts in a single sexual capacity, as $
or as $ . This pattern of sexuality has been found in all heterothallic
species of the biflagellate, phycomycetous orders, Saprolegniales,
Leptomitales, and Peronosporales, which have been intensively in-
vestigated (Bishop, 1940; Bruyn, 1935, 1937; Couch, 1926; Leonian,
193l';Raper, 1940, 1947).

(S I 6 and/or 9 1 9

E87 »ll,l59—»89—>l07, 190-^88 — >I55 — >78.80. 184

Englond I Illinois 1

Fig. 4. In Achlya aDibisexiialis (a heterothallic species of type II) there are
numerous self-sterile, intergrade strains in addition to pure- $ and pure- 9
strains. The intergrade strains may be linearly arranged in respect to $ versus 9
potentialities, and each strain can react as 5 or 9 or both. (From Raper, 1947.)

The mating pattern of a number of strains of Achlya ambisex-
ualis best serves to illustrate this type of sexuality (Raper, 1947).
Ten isolates of this species, collected from northern Illinois in 1946,
when mated in all possible combinations, were found to belong to
six strains. These strains, each self-sterile but cross-fertile in all
combinations, could be linearly ordered, in respect to $ and 9
potentialities, as shown in Fig. 4. In this series each isolate reacted as 9
to those on its left and as $ to those on its right. A strong $ strain,
E87, collected the following year in England, reacted as $ to all six
strains from Illinois. Any intergrade mycelium is capable of reacting
as $ and 9 in different portions of its thallus when mated simultane-
ously with strong 9 and $ plants.

Segregation at meiosis has been observed for plants exhibiting
this type of sexuality in only a single species, Dictyuchus inoiiosponis,

58 SEX IN iMICROORGANISJVlS

by Couch (1926), who found various intergrades in addition to $
and 9 strains among the progeny of a c^ by 9 cross. These prelimi-
nary findings in conjunction \\'ith the many reports of multiple sexual
strains collected from nature strongly indicate the random recombina-
tion of multiple sexual factors during the meiotic process. Further
investigation of the segregative pattern, however, is needed to place
this type of sexuality on the sound experimental basis shared by other
patterns.

Hervhtphroditis'ui mith lucovipatibility Factors at a Single Locus.
This type of heterothallic differentiation produces two self-sterile
and cross-fertile strains, each morphologically and functionally her-
maphroditic, and depends upon the equal segregation at meiosis of
extrasexual determiners or incompatibility factors. Mycelia of each
of the two strains characteristically produce both i and 9 sexual
organs; sexual fusion is accomplished, however, only between the
c^ gametes or the 6 gametangia of one strain with the 9 elements
of the opposite and compatible strain (Dodge, 1932; Drayton, 1932,
1934; Lindegren, 1932; Shear and Dodge, 1927; Wilcox, 1928).

The majority of the heterothallic species of the Ascomycetes,
with the exception of the heterothallic yeasts, as well as many species
of the rusts of the Basidiomycetes, exhibit this basic pattern of sex-
uality. In many Ascomycetes, Neiirospora for example, the $ , or
fertilizing, element is characteristically the microconidium or sperma-
tium, the 9 , an ascogonium. In other Ascomycetes differentiated
gametes arc not formed, and fusion occurs bet\^'een morphologically
distinct gametangia, antheridia and ascogonia, respectively. In rusts
exhibiting this t\"pe of sexuality, spermatia bring about fertilization
when brought into contact with receptive elements of the compatible
mycelium, usually a specialized organ known as the "flexuous hypha"
(Buller, 1950; Craigie, 1927, 1931, 1942).

The differentiation of sexual cells in plants having this type of
heterothallism is phenotypic, and there are numerous cases in w'hich
differentiated sexual cells, spermatia or microconidia, of the Ascomy-
cetes particularly (Dodge, 1932), have been shown to be capable of
purely vegetatixc dc\clopment. Conversely, in a number of forms,
fertilization is accomplished w ith equal facility by microconidia, by
asexual macroconidia, or, for that matter, by any cell of the vegeta-
tive thallus (Backus, 1939), and in some the ability to produce dif-
ferentiated S cells appears to have been lost, the function of fertiliza-

LIFF. CYCLES, SEXUALFFV, AND SF.XUAL .MFCHAXISAIS 59

rion being completely assumed bv asexual or vegetative elements
(Dow ding, 1933; Dowding and Bullet, 1940).

Secondatv homothallism, seiisu \\'hitehouse, a phenomenon of
fairly common occurrence in this group and shared to a greater or
lesser degree by forms of other patterns in which incompatibility
factors constitute the critical determination of mating behavior, re-
sults from the regular inclusion in the spore of two nuclei carrying
opposed incompatibility factors (Ames, 1934; Dodge, 1927; Dowding,
1931; Dowding and BuUer, 1934; Sass, 1929). Binucleate spores of
this sort give rise to heterocaryotic mycelia which are self -fertile; in
some species, sexual organs are present and appear to be essential, in
other species sexual organs may be absent or greatly reduced and
apparently non-essential. In several cases, occasional irregularities
during spore production yield small, uninucleate spores, each of
\\hich develops into a self-sterile but cross-fertile mycelium which
behaves exactly as do the individual mycelia of heterothallic species
(Ames, 1934; Dowding, 1931). The initial binucleate condition thus
predetermines a composite heterocaryon the net reaction of which is
homothallic.

Alternate Sexual Factors ucith Incovipatibility Factors at a Single
Locus. A pattern of sexuahty involving the independent assortment
of sexual factors and incompatibility factors has been demonstrated
to date in only a single species, Hypoviyces solani f. cucurbitae, an
ascomycetous fungus. The segregation of four distinct mating strains,
$ of compatibility type A, 2 of A, $ of a, and 2 of a, constitutes
the basic pattern of sexuahty in this species (Hansen and Snyder,
1946). Four additional strains, however, appear among the progeny
of certain crosses with a frequency of ca. 25 per cent; these are her-
maphrodites and neuters of both incompatibility types. Cytological
investigations (Hirsch, 1949) have revealed the mechanism that
accounts for these anomalous strains. Frequent nondisjunction at
meiosis of the two homologous chromosomes carrying the sexual
factors yields two additional classes as regards sexual factors: one
containing both, the other containing neither. The incompatibility
alleles, segregating independently of the sexual factors and presum-
ably located on different chromosomes, combine with these tw^o
classes at random to yield the four observed strains. Thus a total of
eight different strains constitutes the array of distinct mating types in
this species. Matings of 2 x 5 , 62 x 5 , and 2 x 52 , of the proper

60 SEX IN MICROORGANISMS

incompatibility types of course, yield the various classes of progeny
predicted on the basis of the random assortment of the determining
factors at the two loci and non-disjunctive doubling of the chromo-
somes carrying the sexual factors.

It is expected that this complex type of sexuality, with or without
the non-disjunctive feature or other complications, will eventually be
found in other species.

hicompatibilhy Factors at a Single Locus. This and the suc-
ceeding pattern of sexuality involved no sexual factors and no dif-
ferentiated sexual organs. Mating is commonly reciprocal, and in
multicellular organisms any cell of the thallus is potentially capable
of donating a fertilizing nucleus and of accepting a fertilizing nucleus
from the mate. The term somatic copulation — "Somatogamie" (Ren-
ner, 1916) — has been applied to this type of sexual fusion.

Species displaying sexuality of this type can be subdivided into
two classes according to the number of alternate allelomorphs at the
incompatibility locus.

In species of certain groups a single pair of alleles determines
mating type; all individuals of each species therefore belong in one or
the other of two mating categories, commonly designated A and a.
The heterothallic yeasts are the best known examples of this pattern
(Winge, 1935, 1944; Winge and Laustsen, 1939), and a recent review
of the sexuality of the heterothallic smuts indicates basic control of
mating behavior in this group by a one-locus, single-allelic-pair
mechanism (Whitehouse, 1951).

Mutations at the incompatibility locus in certain yeasts (Linde-
gren and Lindegren, 1944; Winge, 1944) may be considered slight
deviations of this pattern and may possibly indicate the mode of
origin of the following pattern. Either A or a may occasionally
mutate to altered states which permit fusion and ascus production
within a single clone. The ascospores of such unions, however, have
low viability. This possibly reflects either a lack of equivalence be-
tween the mutated alleles and the originals or the expression of delete-
rious, semisterility factors in the homozygous condition (Catcheside,
1951).

What would appear to be a much more highly evolved pattern of
single locus control of mating behavior is characteristic of many
Basidiomycetes, exclusive of the rusts and smuts. In these forms a
very large number of completely equivalent alleles may be found in

IIFF CYCLES, SEXUALITY, AND SEXUAL MECHANISMS 61

various individuals at the single incompatibility locus, and mating
occurs readily between any two haploid strains which carry different
alleles (Brunswik, 1924; BuUer, 1924; Vandendries, 1923; White-
house, 1949). Thus in lieu of the single pair, A and a, these forms
each comprise an extended series of mating types which may best be
designated by A^,. A~, A\ A'', • • • , A". This pattern of mating-type
determination has been termed bipolar sexuality.

hicovipatibility Factors at Tido Loci. The final pattern of
obligatory^ interstrain mating behavior to be described is one found
only among the Basidiomycetes, exclusive of the rusts and related
groups. It involves mating-type determination by incompatibility
factors at t^'o loci; for example, the diploid condition may be desig-
nated A^A~B'B~, and independent assortment of these factors at
meiosis yields progeny of four mating types, A^B\ A^B^, A^B\ and
A^B^. Aiating occurs only in those combinations having different
alleles at both loci, for example, A'B' x A'B' and A'B' x A'B'
(Brunswik, 1924; Hanna, 1925; Kniep, 1920; Mounce, 1922, 1926).
This pattern of segregation was first described by Kniep about 1920
and was termed tetrapolar sexuality.

In tetrapolar species, as in the bipolar forms discussed earlier,
the total number of mating types in the population is increased tre-
mendously by the occurrence of multiple alleles at both loci (Kniep,
1922). The number of equivalent alleles at each locus, however, ap-
pears to be consistently different in two large groups of Basidiomy-
cetes, the Gasteromycetes, which includes the "puffballs," and so
on, and the Hymenomycetes, which includes the "mushrooms,"
"bracket fungi," and the like. In the former group about 10 alleles
at each locus has been indicated as the extent of the series (Fries, 1940,
1943), whereas in the latter group as many as 27 alleles at each locus
have been demonstrated (Brunswik, 1924; Fries and Janasson, 1941;
Kniep, 1922) and the minimal total number of alleles at each locus in
the population has been estimated to be of the order of 100 (White-
house, 1949). In all cases the alleles at each locus appear to be physi-
ologically equivalent. The device of multiple incompatibility factors
allows for almost complete outbreeding while maintaining inbreeding
at 50 per cent in bipolar forms and at 25 per cent in tetrapolar forms.

Kniep, in his original work on tetrapolarity, observed that "muta-
tions" occurred at the A and B loci in frequencies of about 2 per cent
and slightly less than 1 per cent, respectively (Kniep, 1923). These

62 SEX IN MICROORGANISMS

changes in the incompatibiUty alleles, repeatedly observed by various
workers, however, seemed to occur only in the germling mycelia
recently derived from basidiospores, never in wxll-established mycelia.
It has recently been shown beyond any reasonable doubt, by Papazian
(1951), that the A factor consists of a number of pseudo-alleles at
closely linked but distinct loci, that these act together as a physiolog-
ical unit, and that occasional recombination at meiosis yields new
factors. These data are compatible with a composite factor comprising
4 to 10 distinct loci, the upper limits of which range (> 6)
could account for the estimated factors in natural populations without
necessary recourse to multiple allelomorphic series at any locus
(Raper, 1953).

Practically all species of fungi w^iich have sexual cycles may be
definitely assigned to homothaUism or heterothallism. There are,
however, a few species that appear to occupy positions which are
intermediate between these two opposed conditions. The yeasts, men-
tioned above, which are basically heterothallic but which produce
frequent, low-viability mutants of the incompatibility factors may be
considered to bridge, to some extent, the gap between true hetero-
thallic and homothallic conditions. Mather (1940) has suggested the
term partial heterothallism for cases of this type.

In a few cases, furthermore, indeterminate patterns of sexuality
appear to be more closely allied with homothaUism. Outstanding
among such forms is the Ascomycete Glovterella cingjilata, a sexually
ambiguous species without peer. Edgerton (1914) described a strong
sexual interaction in this species between weakly self-fertile strains.
Subsequent and intensive work with G. cingiilata (Andes, 1941;
Edgerton, 1945; McGahen and Wheeler, 1951; Wheeler, 1950;
Wheeler and McGahen, 1952) has revealed an extremely complicated
pattern of sexuality which results from the interaction of numerous
genetic factors, some exhibiting high mutation rates. As currently in-
terpreted (IMieeler and McGahen, 1952), two loci, A and B, are con-
sidered primarily responsible for the basic sexual characteristics, with
some twenty other loci modifying the sexual reaction. Two mutant
states are known at each of the two primary sexual loci in addition
to the t^\ o wild-type alleles. Thus all combinations between the
three alleles at the two loci, ^+, A\ and A^ and B+, B\ and B\ deter-
mine nine distinct strains, each having a characteristic pattern of self-
sterility or self-fertility on the one hand and interstrain matings on

LIFE CYCLES, SEXUALITY, AND SEXUAL iMECHANISIMS

63

the other. The major characteristics of these several strains and the
complex pattern of interstrain matings are diagrammatically rep-
resented in Fig. 5. The sexual characteristics of the various strains may
be further modified by mutations at loci other than the primary
sexual loci, A and B, and two of these, F' and st^, have been described
in detail (Wheeler and McGahen, 1952). Each of these mutants im-
poses self-sterility upon each of the normally self-fertile strains but

SEXUALITY IN GLOMERELLA

A'B'

Introstraln Reaction
Circled- Selt-fertile
Not Circled- Self- sterile

Interstrain Reaction
^= Very Heavy
-= Heavy

— Light

— Weak, Uncertain

Fig. 5. Intrastrain fertility and interstrain reactions in Glo?nerella cingulata.

Three allelomorphs at each of two loci, A and B, define nine strains, each having
a distinct pattern of morphological and sexual characteristics. (Diagramed from
data of Wheeler and iVIcGahen, 1952.)

does not interfere with interstrain reactions provided that the two
mates carry neither of the mutations in common.

According to the definitions of homothallism and heterothallism
adopted here, Glovterella cingulata must be considered basically a
homothallic species since the self-sterile strains are in each case derived
through degenerative, mutative changes from self-fertile, wild-type
strains. The enhancement of sexual productivity in interstrain con-
trasts between self-fertile strains and the occurrence of self-sterile but
cross-fertile strains constitute a pattern that falls short of the totality
of self-sterility and obligatory cross-breeding of true heterothallism.

64 SEX IN MICROORGANISMS

Furthermore, the factors at the A and B loci appear to exert chiefly
a quantitative control over intrastrain fertility and interstrain reaction
rather than the qualitative control imposed by both sexual factors
and incompatibility factors in truly heterothallic species. They differ
from sexual factors in that they do not, in any case, determine uni-
sexual strains, and they differ from incompatibility factors in that
common factors do not in all cases prevent interstrain mating.

The pattern of sexuality in Gloinerella c'mguJata is therefore
basically different from any other known among the fungi. It is pos-
sible that here we have on display a species in the process of evolving
from homothallism to heterothallism, or vice versa, and to accept
Mather's concept of partial heterothallism may well be the best that
can be done at the present time toward integrating this pattern into
the general scheme of sexuality in the fungi.

SEXUAL MECHANISMS

The final aspect that must be considered to give a comprehensive
understanding of sex in fungi is the sexual mechanism, the mechanical
means by which compatible elements are brought together under
the conditions imposed by the particular life cycle and pattern of
sexuality involved.

The number of possible combinations of basic sexual mechanisms
and developmental histories to be found among fungi precludes the
consideration of all significant combinations. Let us rather list a few
possible variants at certain critical stages and demonstrate by simple
developmental histories the range of variety of specific overall pat-
terns.

Sexual mechanisms may be differentiated on the basis of mor-
phological differences at three critical points in the life cycle. These
points are (1) meiosis, (2) the physical union of compatible sexual
elements, and (3) the fusion of compatible nuclei.

(1) The immediate products of meiosis, with few possible ex-
ceptions among the fungi, are spores of various kinds, such as zoo-
spores, ascospores, and basidiospores.

(2) In spite of the almost endless variety of sexual apparatuses
among the fungi they may be considered to belong to four basic
types, first recognized by Kniep (1928). Each type comprises definite

LIFE CYCLFS, SFXUAl.ll V, AM) SI XUAF iMFCHANISIVIS

65

groups of plants, but such groupings have very little correlation with
the major phylogcnctic groupings. The four basic types are:

(a) Gametic copulatioii, in wliich the two elements brought
together in the sexual act comprise uninucleate, free gametes of
M'hicli both, one, or neither may be motile.

(b) Gainete-gainctangiciJ copulation, in which one fusing ele-
ment is a differentiated uninucleate gamete and the other is a differ-
entiated s^ametangium which produces no discrete, uninucleate
gametes. The differentiated gametes may be either 5 or 5 depending
upon the group.

V/ r .

Fig. 6. The four basic modes of sexual fusion in fungi.

(a) Gametic copulation, AUomyces arbiisciila. iMotile gametes originating in $
gametangium (above) and $ gametangium fuse to form a zygote that germinates
directly to produce a diploid plant (lower right), (b) Gamete-gametangial copulation,
Acblya aii?bisex7ialis. Uninucleate gametes, or eggs, in spherical 2 gametangium are
fertilized by $ nuclei transferred through tiny tubes from S gametangia; mature,
fertilized eggs shown below, (c) Gametangial copulation, Fhycornyces blakesleeamts.
A pair of multinucleate gametangia, produced at the tips of large, arched processes,
fuse to form a heavy-walled zygospore surrounded by spines, (d) Somatic copulation,
Scbizophylhmi coimnime (schematic). Two types of hyphal fusion, tip-to-tip and
tip-to-peg, are shown in two stages of development, in an early stage of mutual
chemotropic attraction at the left and shortly after fusion at the right. The exchange
of nuclei in somatic copulation is typically reciprocal, each mate fertilizing the
other.

(c) Gavietmigial copulation, in which both fusing sexual ele-
ments are differentiated as gametangia; one or many pairs of nuclei
may be involved, and the two gametangia may be differentiated as
$ and 9 or they may be morphologically indistinguishable.

(d) Soviatic copulation, in which fusion occurs between undif-
ferentiated vegetative cells; nuclear migration here is frequently
reciprocal, each mate fertilizing the other, and the two compatible
nuclei usually retain, once brought together, a dicaryotic association
for an indefinite period prior to nuclear fusion.

66

SEX IN MICROORGANISMS

The four basic types of sexual apparatuses, as exemplified in
four well-known representative species, are shown, in surface view,
in Fig. 6. Each of these types may be found in a wide array of mor-
phological variations, but the basic aspects of the sexual progression,
nuclear behavior, and so on are relatively constant in the variants of
each type.

MEIOSIS=

COPULATION, Gometic
PLASMOGAMY

CARYOGAMY.

Gametangia
^^ Primordia

Gametangia

Gomete-
Gometanqiol Gometdnqial

Somatic

Fig. 7. Summary diagram of developmental sexual histories in fungi. The
numbered lines trace developmental variations relating the critical events at
meiosis, sexual fusion, and nuclear fusion.

( 3 ) After the fusion of the two sexual elements there exist two
possibilities with regard to the subsequent activity of the paired
compatible nuclei: they may fuse immediately to establish the diploid
phase, or they may become associated in one or more pairs and
divide conjugately by mitosis to provide ultimately a large number of
paired nuclei which finally fuse to form the definitive nuclei in the
asci or basidia.

The developmental patterns relating these cardinal stages in the
sexual cycle may be represented by the various pathways indicated
in Fig. 7. The specific combination of events at the critical points in

LIFL. CYCLES, SEXUALITY, AND SEXUAL MECHANISAIS 57

rlic developmental cycle, meiosis, fusion of sexual elements, and
nuclear fusion, defines about as well as is possible the sexual mecha-
nism tor any given species. A few well-known forms will be used
liere to illustrate the range of possibilities and also to demonstrate
a shorthand system for designating the sexual mechanism and develop-
mental sequence.

The production of gametes as the immediate products of meiosis
may possibly occur in a very few species, members of the Blasto-
cladiales of the aquatic Phycomycetes. A single mitotic division, how-
ever, has been reported as interposed between meiosis and the differ-
entiation of gametes in the single species which has been cytologically
investigated (\\'ilson, 1952).

Most species, if not all, produce spores immediately after meiosis,
and the further developmental sequence is extremely variable. In a
number of cases the fusion of these differentiated spores constitute
the sexual act. Of common occurrence in the yeasts is the fusion of
ascospores in pairs while still in the ascus to reestablish the diploid
phase ((2 — 10)) (Winge and Laustsen, 1939) or, rarely, a dicaryon
( (2 — 1 0 — 1 3 ) ) (Guillermond, 1 940) ; a similar sexual fusion is known
in many smuts, in which sporidia, or basidiospores, fuse to establish
a stable dicaryon ((^1 — 13)) (Bauch, 1925; Kniep, 1926).

The spores give rise in other fungi to vegetative thalli or clones
of vegetative cells prior to sexual activity. Vegetative cells may par-
ticipate without any discernible sexual differentiation in either of two
ways. In clonal, unicellular forms, such as many of the haploid yeasts,
each individual cell is functionally a gamete, and fusion between such
cells may be considered a gametic copulatory process ((3 — 5 — 10))
(Guillermond, 1940). In a large number of extensively developed
mycelial forms, including the majority of the species of the Basidi-
omycetes, all vegetative cells of the thallus are capable of reciprocal
somatic copulation to initiate the dicaryon ((3 — 6 — 13)) (BuUer,
1924; Kniep, 1920, 1922).

Remaining fungi produce sexual organs or gametangia, and these
are almost invariably essential for sexual activity.

The entire vegetative thallus may be differentiated at maturity
into one or more gametangia, which may develop further in either
of two different ways. The gametangia may undergo internal dif-
ferentiation to produce uninucleate gametes which fuse in pairs, as
m Blastocladiella ((3—7—9—10)) (Couch, 1942; Harder and Sorgel,

68 SEX IN MICROORGANISMS

1938), or the gametangia may fuse without further differentiation, as
in numerous monoflagellate Phycomycetes, such as Siphoiiaria ((3 —
7—12)) (Karling, 1945; Wager, 1913).

In yet other forms the sexual activity is relegated to gametangia
which originate de novo as extra-vegetative structures. Three dif-
ferent patterns of further sexual development are found in these
forms: gametangia may produce gametes which fuse in pairs, as in
Allomyces ((3—8—9—10)) (Emerson, 1941; Kniep, 1929);
gametangia of one sexual sign may produce differentiated gametes
which react sexually with gametangia of the opposite sexual
sign, as in Achlya (Bary, 1881; Raper, 1939) ((3—8—9—11)) or
Neurospora (Backus, 1939; Shear and Dodge, 1927), and many rusts
((3—8—9—11—13)) (Buller, 1950; Craigie, 1942); the game-
tangia, morphologically differentiated in respect to sexual sign or not
depending upon the species, may fuse directly with one another, as
in Mucor and Rhizopus ((3—8—12)) (Blakeslee, 1904, 1920; Bur-
geff, 1924; Krafczyk, 1935) or Fyronejna (Claussen, 1912) and
Ascobohis (Dodge, 1920) ((3—8—12—13)). The developmental
pattern of Phycomycetes and most of the lower Ascomycetes (Hem-
iascomycetes) differs from that of the higher Ascomycetes (Eu-
ascomycetes) and Basidiomycetes following plasmogamy in that their
nuclei fuse immediately, whereas in the higher groups dicaryons are
regularly formed.

The dozen or so developmental sexual histories and sexual mecha-
nisms sketched here are the more common types encountered among
the fungi. Most forms fit comfortably in one or the other of these
patterns, but there are a number of cases that would be categorized
variously according to the preferred interpretation of structural and
behavioral characteristics.

Relatively little is known of the underlying physiological and
biochemical aspects of sexual development and sexual activity. It
has long been recognized that an intimate relationship exists between
nutritional requirements and metabolic processes on the one hand and
sexual differentiation and activity on the other (Coker, 1923; Dodge,
1920; Klebs, 1898, 1899, 1900; Molliard, 1903; Raper, 1952). The
knowledge of such relationship, however, has commonly been arrived
at quite empirically, and only in a few cases is there a glimmer of
the underlying mechanism.

Intraspecific chemical regulators of sexual processes, sexual hor-

LIFE CYCLES, SEXUALITY, AND SEXUAL MECHANISMS

69

nioncs, \\ere first demonstrated in the fungi, by Burgcff in 1924, in
Ahicor vnicedo, a close relative of "black bread mold." Since that
time sexual hormones have been demonstrated or postulated on good
experimental evidence in various groups of fungi exclusive of Basidi-
omycetes (Backus, 1939; Bishop, 1940; Krafczyk, 1935; Raper, 1939,
1940, 1951, 1952; Zickler, 1952). In only a single case, however, has
an understanding of the over-all role of hormones as the coordinating
agents in the sexual process been approached.

?

Vegetative Plant

I — A- Initiates-

The A-Complex

Production of
Antheridiol Hyphoe

\,

B-

/Mutually Augment
A- Initiates '

Production of

->

Attraction of
Antheridiol Hyphoe

Thigmotropic Response

A/
Delimitation of ^

Antheridia

Oogonio

Initials

d;

->

■>

Delimitation of
Oogonia

Differentiation of
Oospheres

-Fertilization

->

V

Maturation of
Oospores

Fig. 8. The hormonal mechanism that coordinates the sexual interaction
between male and female plants in heterothallic species of Achlya. Each line
designated by a letter indicates a specific hormone, its origin, and its specific
activity. (From Raper, 1951.)

70 SEX IN MICROORGANISMS

In two heterothallic species of Achlya, a common genus of the
aquatic Phycomycetes, it lias been demonstrated that the initiation
and coordination of each of a chain of interdependent reactions,
which together constitute the sexual process, depend upon one or
more specific chemical agents (Raper, 1940, 1951, 1952). The hor-
monal mechanism, as currently interpreted in these plants, is shown
in Fig. 8. Essentially, the mechanism consists of a minimum of seven
distinct hormones, four secreted by the $ and three by the 2 , which
induce and regulate a series of reactions alternately in the $ and 5 ;
each reaction is chemically dependent upon and quantitatively regu-
lated by the reaction immediately preceding. The entire sexual
process, with the exception of the physical transfer of $ nuclei in
the act of fertilization, has been shown to be coordinated in this
manner. None of the sexual hormones has been either isolated in
chemically pure form or identified.

CORRELATIONS OF LIFE CYCLES, SEXUALITY, AND
SEXUAL MECHANISMS

The three principal facets of sex in fungi having been examined
in some detail, it should now be possible to attempt some correlation
between them and to approach some sort of integrated picture of
the problem in its entirety. Such a correlation is attempted in Table
L In this table are shown the more frequent combinations of life
cycle, sexuality, and developmental sexual history, as well as examples
of these, chosen wherever possible, from those fungi which are rela-
tively well known to biologists other than mycologists. Patterns of
sexuality and developmental histories have been bracketed within
each type of life cycle; indication of the actual combinations which
are known to occur would serve only to obscure the important con-
clusions that may be drawn from this body of information.

The most striking fact that emerges here is also one of consider-
able significance, namely, no rigid and inclusive correlation exists be-
tween the various combinations of sexual features and the universally
accepted phylogenetic groupings. To illustrate this: homothallism,
possibly the most primitive of the various pattern of sexuality, occurs
in conjunction with all types of life cycles, with practically all devel-
opmental histories, and in every major grouping from the most
primitive Phycomycetes to the most highly evolved Basidiomycetes.

LIFE CYCLES, SEXUALITY, AND SEXUAL MECHANISMS

TABLE I

Combinations of Life Cycles, Patterns of Sexuality,
AND Sexual Mechanism Occurring in Fungi

71

Life
Cycle

Pattern of Sexual
Sexuality Mechanism

Example

A
B

Nont
0~

I

II

V,

<
>■

None
r3— 7— 9— 10

3—7—12

3—8—9—10

3—8—9—1 1

3—8—12
^3—5—10

Synckyti-ium
Siphonaria
Monoblepharis
Achlya, Dicty uchus
A'lucor, Eremascus
Zygosaccharomyces

C

0^

III

[3- 8— 12— 13
^ \3— 8— 9— 11— 13

Pyronema, Ascobolus
Neurospora, Hypomyces

IV.

D

0'

III

V
VI.

'3—5—10—13
>■ ] 3—8—9—11—13
[3—6—13

Saccharomyces sp.
Rusts

Gasteromycetes, rusts
Hymenomycetes, smuts

E

01

4—13

vj

Smuts

F

0>

I

r 3—8—9—10
< 3—7—9—10
[3—5—10

Allomyces
Blastocladiella
Saccharomyces sp.

G

V 2—10

Saccharomy codes

In spite of the lack of any paralleled progression from simple to
complex life cycles, patterns of sexuality, and sexual mechanisms and
morphological characteristics, there are certain tendencies which
are worthy of mention. There is a very loose correlation between
morphological specialization and each of the three major facets of
sexuality.

Life cycles, on the whole, become progressively more complex

72 SEX IN MICROORGANISMS

proceeding from primitive to more highly speciaHzed groups. The
haploid cycle predominates in the Phycomycetes, the haploid with
restricted dicaryon cycle in the Ascomycetes, and the haploid-dicar-
yotic and the dicaryotic cycles in the Basidiomycetes. The exceptions
to this generalization, however, are numerous, and, when considered
in respect to probable phylogenetic lines, they are more than a little
puzzling. Haploid-diploid and diploid cycles, those cycles which
would seem to be the most highly advanced of all, occur only in one
group of aquatic Phycomycetes and in a number of yeasts.

The pattern of sexuality in heterothallic species shows a similar
progression. The role of sexual factors as the critical determinants
of mating behavior is for the most part limited to the more primitive
forms, particularly the aquatic Phycomycetes, although there are
several cases of strict sexual dimorphism among the Ascomycetes. A
single pair of incompatibility factors at a single locus possibly occurs
in the more complex Phycomycetes, the Mucorales, is very common
among the Ascomycetes, and is frequently encountered in two large
groups of Basidiomycetes, the rusts and smuts. The essentiality of
differentiated sexual organs would seem to follow similar broad phy-
logenetic lines: they are present and functional in practically all
Phycomycetes and most Ascomycetes, except the yeasts, and absent
in the Basidiomycetes, except the rusts.

Multiple incompatibility allelism is known only among members
of the most highly evolved fungi, the Basidiomycetes, and is unques-
tionably the most efficient of all means to insure for those species
possessing it the maximal benefit to be derived from genetic recom-
bination.

This might suggest a sort of coupling of the culmination of
incompatibility control of mating behavior with a high degree of
morphological development, particularly in the tetrapolar species,
were it not for the fact that species which are obviously closely
related to such tetrapolars are strictly homothallic and get along quite
nicely with no restrictions imposed by incompatibility factors.

Sexual mechanisms and developmental histories are fairly con-
stant within groups at the level of orders. There is also a tendency, in
passing from primitive to highly evolved forms, to progress from
gametic copulation through the loss of gametic differentiation in one
sex or the other (gamete-gametangial copulation), to loss of gametic
differentiation in both sexes (gametangial copulation), to the loss of

LIFE CYCLES, SEXUALITY, AND SEXUAL MECHANISMS 73

sexual organs conipctelv and the ability of all vegetative cells to par-
ticipate in sexual fusions (somatic copulation). The developmental
histories of sexual aspects per se, furthermore, show a marked tend-
ency toward simplification, probably tiirough reduction, in most of
the more highly evolved groups.

\\'har, then, can rationally be said of the probable origin of the
array of sexually different types that exist in the fungi at the present
time? Any attempt to rationalize six different life cycles, homothal-
lism, six distinct types of heterothallism, and four basic sexual appara-
tuses in the particular combinations in which they exist rapidly runs
afoul of difficulties that appear to be insurmountable. This can be
illustrated by testing two antithetical propositions. To start from the
assumption, as many have, that homothallic forms having gametic
copulation represent the primitive type from which all else has been
derived must totally ignore the random distribution of homothallic
and heterothallic species in every group of the fungi. The various
patterns of heterothallism would necessarily have been independently
evolved, at the appropriate levels, from the main stem of homothallic
forms. How, then, is it possible to account for homothallic, bipolar,
and tetrapolar species in a single genus, such as Copmms, the members
of which are obviously closely related phylogenetically, except that
heterothallism, of two very precise types common throughout the
much larger group to which it belongs, the Hymenomycetes, be in-
dependently evolved in this genus? Essentially the same situation
obtains in every phylogenetic grouping and in toto constitutes a
compelling argument against the derivation of heterothallism from
homothallism. The alternative proposition, that the variously ex-
pressed forms of homothallism were derived from heterothallic ances-
tral forms, encounters equally serious difficulties. Most important of
these difficulties are the twin necessities of (1) the origin of the
various types of heterothallism from some one primitive heterothallic
type and (2) the independent origin of homothallism in each homo-
thallic species. The latter might conceivably occur in sufficient fre-
quency to account for the large number of homothallic forms, but
it would appear most unlikely. More serious is the difficulty of the
evolution of the various types of heterothallism from other heter-
othallic types; the various types would appear to be much more
closely related to the corresponding homothallic types within any
phylogenetic grouping than to each other. Nor would espousal of

74 SEX IN MICROORGANISMS

polyphyletic origin of the various groups make less difficult the
rationalization of the existing sexuality of the fungi. There would
remain the same difficulties mentioned above, only partially obscured
by the introduction of additional uncertainties.

The nearest approach to a biologically feasible system which
could account for the existing sexual complexity would reject both
of the simple propositions stated above, but would constitute a partial
synthesis of the two. The following suggestion would appear success-
fully to avoid the various objections to simpler derivation.

Homothallic forms, having gametic copulation, could well have
given rise to a primitive homothallic group within which occurred
the major evolutionary changes in life cycles and sexual mechanisms
— the progressions from predominantly haploid to predominantly
dicaryotic cycles and from gametic through gamete-gametangial and
gametangial to somatic copulation. Each distinct type of heterothal-
lism could have been independently evolved, at the appropriate level,
from this primitive homothallic stem. Once the several types of heter-
othallism had evolved, existing homothallic species could well have
been derived from them by relatively simple means. For example,
several kinds of chromosomal aberrations causing a slight dislocation
of the locus of a sexual or incompatibility factor could result, after
transfer of the factor to the homologous chromosome through cross-
ing-over, in genetically stable self-fertile individuals. Such self-fertile
individuals, being assured the immediate benefits of sexual reproduc-
tion and being genetically isolated, might well prosper and constitute
a significant factor in speciation in the fungi.

This scheme, unduly indirect at first glance, would account for
numerous awkward facts implicit in any simpler hypothesis. The
more important of these are: (1) the occurrence of homothallic and
heterothallic species within groups having unique features which must
have been evolved relatively recently, the genera Achlya and Copri-
nus for example; (2) the occurrence of the same precise types of
heterothallism throughout large groups embracing widely divergent,
morphological characters, the Hymenomycetes for example, with
tetrapolar and bipolar species, the latter possibly a transitional stage
between the former and homothallism, and the aquatic, biflagellate
Phycomycetes with their peculiar multiple-sexual-strain type of
heterothallism; (3) the lack of strictly heterothallic species that can
be reasonably interpreted as intermediate between two distinct types

I.H'I" CYCLES, SL-.XUALHV, and sexual iMLCHANlSAlS 75

of hcrcrothallisni. Species of indeterminate sexuality can be cited in
abundance, but thev can be more easily rationalized as intermediate
betw een the various types of heterothallism and homothallism.

The fungi, viewed from this particular bias, present many
admittedly puzzling features that offer, however, no recognized out-
right contradiction to the essential idea of homo-^hetero^'homothal-
lism. The lack of heterothallic species in a few large and long-estab-
lished groups — the AspergUhis-Pemcill'mm, etc., complex — is a case
in point. Here, ho\\ever, there is a majority of sexually sterile species,
and the entire group may well represent a vestige of the primitive
homothallic stem, well along its degenerative course toward unequiv-
ocal inclusion in the Fungi Imperfecti.

Further speculation along these lines for the present time, ho\t^-
ever intriguing, cannot produce a wholly satisfactory answer to the
problem. There is as yet insufficient information to permit the postu-
lation of a completely feasible system that would account for the
complicated sexual situation now existing in the fungi. For the time
being, we can only recognize the situation for what it is, and marvel.

SUMMARY

Sexual reproduction in fungi displays a tremendous range of
variability. Recognition of three distinct features is necessary ade-
quately to describe the role of sex in any single species. These facets
are: (1) the life cycle, in which the critical events are synonymous
with the initiation, the progression, and the termination of the essen-
tial sexual process; (2) the pattern of sexuality, which determines
self-fertility or self-sterility, and in the latter case the exact pattern of
inter-individual fertility; and (3) the sexual mechanism, the means
by which sexual fusion is accomplished within the restrictions
imposed by the life cycle and pattern of sexuality. Seven types of
life cycles, seven distinct patterns of sexuality, and about a dozen or
more basic kinds of sexual histories allow, in combination, a be\\'il-
dering array of distinct sexual types.

Although there is a very loose correlation between morpholog-
ical specialization and each of the three major facets of sexuality, no
rigid correlation appears to exist between phylogenetic groupings
and the various combinations of sexual features. A possible scheme to
rationalize the sexual situation as now existing rejects the simple der-

76 SEX IN MICROORGANISMS

ivation of heterothallism from homothallism, or vice versa, in favor
of an indirect derivation of homothallism from the various types of
heterothallism, each of which in turn was independently evolved
from primitive homothallic forms. Present understanding is insuffi-
cient, however, to permit the postulation of a completely feasible
system to account for the existing sexual complexity of the fungi.
The multiplicity of sexual types and forms which contribute to this
complexity, however, provides a wealth of material for those who
would seek exact specifications in the tools for the elucidation of
many basic phenomena of universal biological importance.

REFERENCES

Ames, L. M. 1934. Hermaphroditism involving self-sterility and cross-fertility

in the Ascomycete Pleiirage anserina. Mycologia, 26, 392-414.
Andes, J. O. 1941. Experiments on the inheritance of the "plus" and "minus"

characteristics in Glomerella cingulata. Bull. Torrey Botan. Club, 68,

609-614.
Backus, M. P. 1939. The mechanics of conidial fertilization in Neurospora

sitophila. Bull. Torrey Botan. Club, 66, 63-76.
Bary, A. de. 1881. Untersuchungen iiber Peronosporeen und Saprolegnieen

und die Grundlagen eines natiirlichen Systems der Pilze. Beitr. Morphol.

u. Physiol, der Pilze, IV.
Bauch, R. 1925. Untersuchungen iiber die Entwicklungsgeschichte und Sexual-

physiologie der Ustilago bromivora und U. grandis. Z. Botan., 17, 129-177.
Benjamin, R. K., and L. Shanor. 1950. The development of male and female

individuals in the dioecious species Laboulbenia jorniicannn. Ani. J.

Botany, 37, 471-476.
Bessey, E. A. 1950. Morphology and Taxonomy of Fungi. Blakiston, New

York.
Bishop, H. 1940. A study of sexuality in Saproniyces Reinschii. Mycologia,

32, 505-529.
Blakeslee, A. F. 1904. Sexual reproduction in the Mucorineae. Proc. Ant. Acad.

Arts Sci., 40, 205-319.

. 1920. Sexuality in Mucors. Science, n. s., 51, 375-382, 403-409.

Brodie, H. J. 1931. The oidia of Coprimts lagopus and their relation with in-
sects. Ann. Botany, (Lo^idon), 45, 315-344.
Brunswik, H. 1924. Untersuchungen iiber die Geschlechts- und Kernverhalt-

nisse bei der Hymenomyzetengattung Coprinus. Botan. Abhandl. K.

Goebel, 5, 1-152.
Bruyn, H. L. G. de. 1935. Heterothallism in Peronospora parasitica. Phyto-
pathology, 25, 8.
. 1937. Heterothallism in Peronospora parasitica. Genetica, 19, 553-558.

LIFK CYCLl.S, SEXUALH V, AND SF.XUAL MECHANISMS 77

Buller, A. H. R. 1924. Expcriniciirs on sex in mushrooms and toadstools.

Nature, 114, 826.
. 1950. Researches 0/1 Fungi, Vol. 7. Roy. Soc. Can., University of

Toronto Press, Toronto.
Burgeff, H. 1924. Untersuchungen iiber Sexualitat und Parasitismus bei Mu-

corineen. I. Botan. Abhandl. K. Goebel, 4, 5-135.
Cantino, E. C, and Al. T. Hyatt. 1953. Phenotypic "sex" determination in the

life history of a new species of Blastocladiella, B. E?nersonii. A?itonie van

Leejiivenhoek J. Microbiol. Serol., 19, 25-70.
Catcheside, D. G. 1951. The Genetics of Micro-Organisjns. Pitman, London.
Christensen, J. J., and H. A. Rodenhiser. 1940. Physiological specialization and

genetics of the smut fungi. Botan. Rev., 6, 389-425.
Claussen, P. 1912. Zur Entwicklungsgeschichte der Ascomyceten, Fyroneina

confluens. Z. Botan., 4, 1-64.
. 1921. Entvvicklungsgeschichtliche Untersuchungen iiber den Erreger

der als "Kaltbrut" bezeichneten Krankheit der Bienen. Arb. biol. Reichs-

anstalt. Land- u. Forstzvirtsch., 10, 467-521,
Coker, W. C. 1923. The Saprolegniaceae. University of North Carolina Press,

Chapel Hill.
Couch, J. N. 1926. Heterothallism in Dictyuchus, a genus of water molds. Ajin.

Botany {Loiidon), 40, 848-881.
Couch, J. N., and A. Whiff en. 1942. Observations on the genus Blastocladiella.

Am. J. Botany, 29, 582-591.
Craigie, J. H. 1927. Discovery of the function of pycnia of the rust fungi.

Nature, 120, 165-761.
. 1931. An experimental investigation of sex in the rust fungi. Phyto-
pathology, 21, 1001-1040.

1942. Heterothallism in the rust fungi and its significance. Trans. Roy.

Soc. Ca?!., 36, 19-40.
Dodge, B. O. 1920. Life history of Ascobolus jnagfiificus. Origin of the archi-

carp from two strains. Mycologia, 12, 115-134.
. 1927. Nuclear phenomena associated with heterothallism and homo-

thallism in the Ascomycete Neurospora. J. Agr. Research, 35, 289-

305.
. 1932. The non-sexual and sexual functions of microconidia of Neuro-

spora. Bull. Torrey Botan. Club, 59, 347-360.

Dowding, E. S. 1931. The sexuality of normal, giant, and dwarf spores of
Pleurage aiiserina. Ann. Botany (London), 45, 1-14.

. 1933. Gelasinospora, a new genus of Pyrenomycetes with pitted spores.

Can. J. Research, 9, 292-305.

Dowding, E. S., and A. H. R. Buller. 1940. Nuclear migrations in Gelasino-
spora. Mycologia, 32, 471-488.

Drayton, F, D. 1932. The sexual function of microconidia in certain As-
comycetes. Mycologia, 24, 345-348.

. 1934. The sexual mechanism of Sclerotinia gladioli. Mycologia, 26,

46-72.

78 SEX IN MICROORGANISMS

Drayton, F. D., and J. W. Groves. 1952. Stromatinia narcissi, a new sexually

dimorphic Discomycete. Mycologia, 44, 119-140.
Edgerton, C. W. 1914. Plus and minus strains in Gloinerella. Am. J. Botany, 1,

244-254.
Edgerton, C. W., S. J. P. Chilton, and G. B. Lucas. 1945. Genetics of Glom-

erella. II. Fertilization between strains. Am. J. Botany, 32, 115-118.
Emerson, R. 1941. An experimental study of the life cycles and taxonomy of

Allo?nyces. Lloydia, 4, 77-144.
• . 1950. Current trends of experimental research on the aquatic Phyco-

mycetes. Ann. Rev. Microbiol., 4, 169-200.
Fries, N. 1940. Researches into the multipolar sexuality of Cyatbus striatus.

Symbolae Botan. Upsalienses, 4, 1-39.
. 1943. Uber das Vorkommen von geographischen Rassen bei Crucibu-

Iwm vulgare. Arch. MikrobioL, 13, 182-190.
Fries, N., and L. Janasson. 1941. Uber die Interfertilitat verschiedener Stamme

von Polyporus abietinus. Svensk Botan. Tidskr., 35, 177-193.
Gaumann, E. 1926. Vergleichende Morphologic der Pilze. Fischer, Jena.
Gaumann, E., and C. W. Dodge. 1928. Comparative Morphology of Fungi.

McGraw-Hill, New York.
Guillermond, A. 1940. Sexuality, developmental cycle and phylogeny of yeasts.

Botan. Rev. 6, 1-24.
Hanna, W. F. 1925. The problem of sex in Coprinus lagopus. Ann. Botafiy,

39, 431-457.
Hansen, H. N., and W. C. Snyder. 1946. Inheritance of sex in fungi. Proc.

Natl. Acad. Sci. U. S., 32, 272-273.
Harder, R., and G. Sorgel. 1938. Uber einen neuen planoisogamen Phycomy-

ceten mit Generationswechsel und seine phylogenetische Bedeutung. Nach.

Ges. Wiss. Gdttinge7i N. F. Fachgruppen 6, Biologic, 3, 119-127.
Hartmann, M. 1943. Die Sexualitat. Fischer, Jena.
Hirsch, H. E. 1949. The cytogenetics of sex in Hypomyces solafii f. cuciirbitae.

Am. J. Botajiy, 36, 113-121.
Humphrey, J. E. 1892. The Saprolegniaceae of the United States with notes

on other species. Trans. A?n. Phil. Soc, 17, 63-148.
Karling, J. S. 1945. Brazilian Chytrids. VII. Observations relative to sexuality in

two new species of Siphonaria. Am. J. Botany, 32, 580-587.
Klebs, G. 1898. Zur Physiologie der Fortpflanzung einiger Pilze. I. Sporodinia

grandis. Jahrb. wiss. Botan., 32, 1-70.
. 1899. Zur Physiologie der Fortpflanzung einiger Pilze. II. Saproleg7iia

mixta. Jahrb. wiss. Botan., 33, 513-592.

1900. Zur Physiologie der Fortpflanzung einiger Pilze. III. Allgemeine

Betrachtungen. Jahrb. wiss. Botan., 35, 80-203.

Kniep, H. 1920. Uber morphologische und physiologische Geschlechtsdif-
ferenzierung (Untersuchungen an Basidiomyceten). Verhandl. Physik.-
mediz. Ges. Wiirzburg, 46, 1-18.

. 1922. Uber Geschlechtsbestimmung und Reduktionsteilung (Unter-
suchungen an Basidiomyceten). Verhandl. Physik.-mediz. Ges. Wurzbiirg,
47, 1-28.

I.III CYCLES, SEXUALirV, AND SEXUAL MECHANISMS 79

— . 1923. tJber erbliche Anderungen von Geschlechtsfaktoren bei Pilzen.

Z. iiid. Abst.- u. Vererbgsl., 31, 170-183.
— . 1926. ijber Artkreuzungcn bei Brandpilzen. Z. Pilzkde. N. F., 5, 217-

247.

— . 1928. Die Sexualitdt der niedereii Fjlanzen. Fischer, Jena.
— . 1929. AUomyces javaniciis, n. sp. ein anisogamer Piiyconiycet mit

planogameten. Ber. dent, botan. Ges., 47, 199-212.

1930. Uber den generationswechsel von AUomyces. Z. Botan., 22,

433-441.
Krafczyk, H. 1935. Die Bildung und Keimung der Zygosporen von Pilobolus

crystalUniis und sein heterokaryotisches Myzel. Beit. Biol. Pfianz., 23,

349-396.
Lederberg, J., and E. L. Tatum. 1946. Novel genotypes in mixed cultures of

biochemical mutants in bacteria. Cold Spring Harbor Symposia Quant.

Biol, 11, 113-114.
Leonian, L. H. 1931. Heterothallism in Phytophthora. Phytopathology, 21,

941-955.
Lindegren, C. C. 1932. The genetics of Neurospora. IL Segregation of sex

factors in the asci of N. crassa, N. sitophila, and N. tetrasperma. Bull.

Torrey Botan. Club, 59, 119-138.
Lindegren, C. C, and G. Lindegren. 1944. Instability of the mating type alleles

in Saccharomyces. Ann. Missouri Botan. Garden, 31, 203-218.
Link, G. K. K. 1929. Reproduction in Thallophytes, with special reference to

Fungi. Botan. Gaz., 88, 1-37.
McCranie, J. 1942. Sexuality in AUomyces cystogenus. Mycologia, 34, 209-

213.
AIcGahen, J. W., and H. E. Wheeler. 1951. Genetics of Glomerella. IX.

Perithecial development and plasmogamy. Am. J. Botany, 38, 610-617.
Mather, K. 1940. Outbreeding and separation of the sexes. Nature, 145, 484-

486.
Alaurizio, A. 1899. Beitrage zur Biologie der Saprolegnieen. Z. Fischer ei u.

Hilf Swiss, 7, [2], 1-66.
Moewus, F. 1950. Sexualitat und Sexualstoffe bei einem einzelligen Organismus.

Z. Sexualforsch., 1, 1-25.
Molliard, M. 1903. Role des bacteries dans la production des peritheces des

Ascobolus. Compt. rend., 136, 899-901.
Mounce, I. 1922. Homothallism and heterothallism in the genus Coprinus.

Trans. Brit. My col. Soc, 7, 256-269.
. 1926. A preliminary note on Fomes pinicola and F. adiposa — two

heterothallic species of wood-destroying fungi. Phytopathology, 16,

757-758.
Nobles, M. K. 1942. Secondary spores in Corticiwn effuscatum. Can. J. Re-
search, 20, 347-357.
Papazian, H. P. 1951. The incompatibility factors and a related gene in Schizo-

phyllum commune. Genetics, 36, 441-459.
Pontecorvo, G. 1946. Genetic systems based on heterokaryosis. Cold Spring

Harbor Symposia Quant. Biol., 11, 193-201.

80 SEX IN MICROORGANISMS

. 1949. Genetical techniques for self-fertile (homothallic) microorgan-
isms. Eighth Intern. Congr. Genet. (Stockhohn). 1948, 642-643.

1950. New fields in the biochemical genetics of microorganisms. Bio-

chem. Soc. Symposia (Cambridge, Eiig.), 4, 40-50.

Pontecorvo, G., and J. A. Roper. 1952. Genetic analysis without sexual repro-
duction by means of polyploidy in Aspergillus nidiilavs. J. Gen. Microbiol.,
6, vii.

Raper, J. R. 1939. Sexual hormones in Achlya. I. Indicative evidence for a
hormonal coordinating mechanism. A??/. J. Botany, 26, 639-650.

. 1940. Sexuality in Achlya ainbisexualis. Mycologia, 32, 710-727.

. 1940. Sexual hormones in Achlya. II. Distance reactions, conclusive

evidence for a hormonal coordinating mechanism. Am. J. Botany, 27,
162-173.

. 1947. On the distribution and sexuaHty of Achlya ambisexualis. Paper

read before Microbiological Section, Botanical Society of America, Chi-
cago, Dec. 1947. Abstract: Ajh. J. Botany, 34, 31a.

. 1950. Sexual hormones in Achlya. VII. The hormonal mechanism in

homothallic species. Botaji. Gaz., 112, 1-24.

. 1951. Sexual hormones in Achlya. Ani. Scientist, 39, 110-120.

. 1952. Chemical regulation of sexual processes in the Thallophytes.

Botan. Rev., 18, 447-545.

1953. Tetrapolar Sexuality. Quart. Rev. Biol, 28, 233-259.

Renner, O. 1916. Zur Terminologie des pflanzlichen Generationswechsel. Biol.

Zeritr. 36, 337-374.
Roper, J. A. 1952. Production of heterozygous diploids in filamentous fungi.

Experientia, 8, 1-4.
Salvin, S. B. 1942. A preliminary report on the intergeneric mating of Thrau-

stotheca clavata and Achlya flagellata. Am. J. Botany, 29, 674-676.
Sass, J. E. 1929. The cytological basis for homothallism and heterothallism in

the Agaricaceae. A7//. J. Botany, 16, 663-701.
Satina, S., and A. F. Blakeslee. 1928. Studies on biochemical differences between

sexes in Mucors. V. Proc. Natl. Acad. Sci. U. S., 14, 308-316.
. 1929. Criteria of male and female in bread molds (Mucors). Proc. Natl.

Acad. Sci. U. S., 15, 735-740.
Schopfer, W. H., and S. Blumer. 1940. Le pouvoir de synthese d'un facteur de

croissance par Schizophyllwn commune (haplontes et diplontes). Proto-

plasfna, 34, 524-532.
Shear, C. L., and B. O. Dodge. 1927. Life histories and heterothallism of the

red bread-mold fungi of the Monilia sitophila group. /. Agr. Research, 34,
1019-1042.
Sparrow, F. K. 1943. Aquatic Phycomycctes. University of Michigan Press,

Ann Arbor.
Stiiben, H. 1939. tJber Entwicklungsgeschichte und Ernahrungsphysiologie

eines neuen niederen Phycomyceten mit Generationswechsel. Planta, 30,

353-383.

LIFE CYCLES, SEXUALITY, AND SEXUAL MECHANISMS 81

Thaxtcr, R. 1908. Contribution toward a monograph of the Laboulbeniales.

II. Mew. Am. Acad. Arts Sci., 13, 217-469.
\'andendries, R. 1923. Nouvelle recherches sur la sexualite de Basidiomycetes.

Bull. soc. botan. beige, 56, 73-97.
Wager, H. 1913. The life cycle and cytology of Polyphagus Eiiglenae. Aiin.

^Botany {London), 27, 173-202.
Wheeler, H. E. 1950. Genetics of Glonierella. VIII. A genetic basis for the

occurrence of minus mutants. Ani. J. Botany, 37, 304-312.
Wheeler, H. E., and J. W. McGahen. 1952. Genetics of Glomerella. X. Genes

affecting sexual reproduction. Am. J. Botany, 39, 110-119.
Whitehouse, H. L. K. 1949. Multiple allelomorph heterothallism in fungi. Neiv

Pbytologist, 48, 212-244.
. 1949. Heterothallism and sex in the fungi. Biol. Revs. Can/bridge Phil.

Soc, 24, 411-447.
-. 1951. A survey of heterothallism in the Ustilaginales. Trans. Brit.

Mycol. Soc, 34, 340-355.
\Mlcox, M. S. 1928. The sexuality and arrangement of spores in the ascus of

Neurospora sitopbila. Mycologia, 20, 3-17.
Wilson, C. M. 1952. Meiosis in Alloinyces. Bull. Torrey Botan. Club, 79,

139-160.
Winge, O. 1935. On haplophase and diplophase in some Saccharomycetes.

Conipt. rend. lab. Carlsberg, 21, 77-113.
. 1944. On segregation and mutation in yeast. Compt. rend. lab. Carls-
berg, 24, 79-96.
Winge, O., and O. Laustsen. 1939. Saccharomy codes Ludivigii, a balanced

heterozygote. Compt. rend. lab. Carlsberg, 22, 357-370.
Zickler, H. 1952. Zur Entwicklungsgeschichte des Askomyceten Bombardia

lunata. Arch. Protistenk., 98, 1-70.

Sexual Reproduction
in Diatoms

RUTH PATRICK, Academy of Natural Sciences of Philadelphia,
Philadelphia, Pennsylvania

In diatoms reproduction is most frequently accomplished by vege-
tative division. Ordinary somatic mitosis occurs. This type of repro-
duction in a healthy culture of diatoms may occur several times
during a twenty-four-hour period. The rate is dependent on the kind
of diatom and the cultural conditions present.

Of much less frequent occurrence are auxospores. An auxospore
is a resting cell which usually develops from a zygote. This process
has been occasionally observed in many of the genera of diatoms but
in only a comparatively few species.

To date there seems to be little correlation between the type
of auxospore formation and the taxonomic relationship of the species.
Indeed, several different types of auxospore formation have been
observed in different varieties and forms of the same species.

Often auxospore formation seems to be the result of sexual
processes. However, in many cases, as Sonneborn (Calkins and Sum-
mers, 1941) has said, perhaps we should "abandon the concepts of
male and female in unicellular organisms and view sexual union as
brought about by copulation-conditioning factors."

The cause of auxospore formation seems to be a combination
of cell size and external environmental conditions. All species of dia-
toms are known to vary in length. According to Geitler (1932,
1935), auxospore formation occurs only when the cells of a taxon
are of a certain length, characteristic for each taxon. Often this
range in length is fairly wide, and again it is very narrow depending
on the taxon. This range is very constant for each taxon. If auxospore
formation does not occur when the cells are of the correct size, it
never occurs. The cells often become smaller and morphologically
quite changed, especially in culture.

Auxospore formation in diatoms has been a frequent subject of
research during the nineteenth and twentieth centuries. Although the

82

SEXUAL Rl PRODUC'lION IN DIA lO.MS 83

coniplcrc process has been observed only in a relativly few species, it
is parriallv known in a great many more. Because many species are
\ cr\- small and their chromosomes are numerous, detailed cytological
studies arc difficult. The best summaries of this work have been made
by Geitler (1932) and Fritsch (1935).

The diatoms may be divided into tw^o distinct taxonomic groups.
Methods of auxospore formation characteristic of each taxonomic
group \\ill be treated under the appropriate heading.

CENTRALES

The Centrales are usually regarded as the more primitive group
of diatoms because they are found in earlier geological strata than the
Pennales.

The most common type of reproduction is fission after mitosis
Microspores are also produced in this group, but as yet it has not been
proved that they are reproductive cells. They are formed by the
division of the protoplast. Each microspore has two flagella.

Other types of reproduction are those which result in auxospore
formation. Little is as yet known concerning the nuclear reorganiza-
tion that takes place in their formation.

The simplest type of auxospore formation is that which has been
reported in the genus Melosira. The two halves of the wall of the
cell are pushed apart by the protoplast. Over the protoplast is secreted
a slightly silicified pectic membrane called the perizonium. After a
lapse of time new valves and connecting bands are formed inside the
perizonium, and a new individual results. In Melosira mnmmdoides
the auxospore lies outside the theca of the parent. There is consider-
able difference of opinion about the nuclear phenomena that accom-
pany^ auxospore formation in this genus (Karsten, 1897; Geitler,
1932).

In Biddidphia mobiliensis (Bergon, 1907) cell division imme-
diately precedes auxospore formation. The two daughter protoplasts
escape from the parent cell and form a pair of spores. Little is known
about the nuclear behavior during this process.

In Chaetoceras cochlea (Fritsch, 1935) the auxospores arise
laterally on the parent cell by budding. A similar lateral formation
of auxospores also takes place in some species of Rhizosolenia (Schiitt,
1893).

84 SEX IN MICROORGANISMS

Iyengar and Subrahmanyan (1944) observed in Cyclotella me-
jieghiniana zygote formation and subsequent auxospore formation.
They considered this the result of autogamy and automixis.

Recently von Stosch (1950) has reported oogamy in Melosira
varians. He has observed the filaments to be of two types, narrow
male filaments and broader female filaments. In about 9 per cent of
the cases where the size of the filaments overlap they are found to be
monoecious.

The antheridial cells undergo reduction division. Two sperma-
tozoid mother cells bud off the main cytoplasm in the second meta-
phase and then divide to form four spermatozoids. Flagella have not
been seen, but it is presumed that the spermatozoids are flagellated.

The young oogonia resemble vegetative cells in shape, yet their
plastids and chromatophores are larger. Meiosis takes place in the
usual manner. In the first telophase one of the daughter nuclei grad-
ually aborts. In the second telophase one of the nuclei becomes pyc-
notic. The two "polar bodies" thus formed are gradually absorbed.

The oogonium swells, and a strip of naked protoplasm is exposed
between the margins of the epitheca and hypotheca. The sperma-
tozoid may enter the oogonium as early as anaphase I or as late as
the maturation of the t^g nucleus. Later the zygote is released and
swells to form a subglobose auxospore.

Geitler (1952e) has recently observed in Cyclotella sp. sexual
reproduction similar to that described for Melosira varians. There
seems to be no morphological difference in the filaments that form
the eggs and sperms. In spermatogenesis there occurs a first meiotic
and a second meiotic division which result in the formation of four
sperms. The sperm enters the oogonium by the time diakinesis takes
place. As a result of a first meiotic and second meiotic division of
the nucleus, one ^gg nucleus and two pycnotic nuclei are produced.
The sperm nucleus migrates during interkinesis from the peripheral
region of the ^gg to the center, where fusion with the tgg nucleus
takes place after the second meiotic division is complete. A metagamic
mitosis occurs between the formation of the first and second shells of
the cell formed in the germination of the auxospore. Although meta-
gamic mitosis has been observed many times in the Pennales, this is
the first time it has been observed in the Centrales.

SEXUAL Rl PRODUCTION IN DIATOMS 85

PENNALES

Auxospore formation is usually initiated by the coming together
of the mother cells. Of course, in cases of apomixis and automixis,
this may not occur. These cells are diplonts and usually considered
not to be sexually differentiated. However, in Naviciila balophila
(Subrahmanyan, 1945), Syjiedra iihia (Geitler, 1939), and Synedra
rwnpens var. fragilarioides (Geitler, 1952f), it has been reported that
one cell produces two passive gametes, and one produces two active
gametes. This might indicate that the two mother cells are sexually
differentiated.

These cells may be about the same size as in Rhokosphenia
curvata (Geitler, 1952a) or may be very unequal in length as in
Ewwtia arciis (Geitler, 1951b). Sometimes more than two cells come
together for auxospore formation, as in GoiJiphonema parvzdwn var.
viicvopus, Achnanthes lajiceolata, and Navicula seininulum (Geitler,
1932). In Ajiovweoneis exilis (Geitler, 1949b), Navicula radiosa
(Geitler, 1952d), and Synedra uhia (Geitler, 1939), several cells
often come together. Usually these cells are not sister cells, but in
Navicida semimdinn they may be (Geitler, 1932).

In most cases both cells are active and approach each other. In
GoiJiphonema parviduvi var. micropiis one cell is attached by a gelati-
nous stalk and only one is mobile (Geitler, 1932). These cells assume
various positions on contacting each other. The most common posi-
tion is for them to lie opposite and parallel with their girdle faces
in juxtaposition. The cells of Goijiphoiiema parvidinn var. iincropus
orient themselves so that the apical pole of one cell is opposite the
basal pole of the other. Owing to the curvature of the frustule of
Rhoicosphenia curvata, the cells may be in various positions (Geitler,
1952a).

Jelly is produced by both cells in varying quantities. Liebisch
(1929) considered this jelly part of the hydrated pectin membrane
of the cell. Other research indicates that it has a different origin. It
is evident that more work needs to be done on this point.

This jelly is usually homogeneous and varies in thickness accord-
ing to the kind of diatom. In Achnanthes longipes, Navicida didyma,
and Fleurosigina imbecida it is fairly soft, whereas in Frustulia rhom-
boides var. saxonica and Achnanthes laficeolata it is relatively stiff.

86 SEX IN MICROORGANISMS

Geitler (1932) thinks that tensions which develop in this jelly as a
result of its viscosity determine to some extent the movement of the
gametes and the position of the developing zygote and auxospore.

In many species a large quantity of jelly is produced and the
copulating cells are embedded in it. However, in other species the
jelly is only represented by the formation of a copulation tube or
tubes. In Ewiotia arcus and Ewiotia flexuosa (Geitler, 1951c) the
copulation tube is formed by papillae which are formed by each of
the two copulating cells.

The number and shape of the tube or tubes may vary. Usually
only one tube is formed; however, in Fnistulia rhoviboides var.
saxonica (Geitler, 1949b) two tubes are present. The tube may be
long and narrow as in Eimotia arais (Geitler, 1951b) or short and
narrow as in Nitzschia subtilis and Amphipleura pellucida (Geitler,
1932, 1952c). Usually the tube is formed at or near the middle of
the longitudinal axis. In Fnistulia rhomboides var, saxonica (Geitler,
1949b) one tube is found near each of the apices of the cell. In Eimo-
tia arcus and Ewwtia flexuosa (Geitler, 1951a,b) a tube may be
formed at either end of the cell or on the girdle face of the dorsal or
ventral side of the valve.

GAMETOGENESIS

So far as is known, meiosis occurs by means of two meiotic divi-
sions. The spindle lies in the pervalvar axis of the cell. Sometimes it
is tipped slightly to one side as in Aviphipleura pellucida (Geitler,
1952c). The prophase of the first meiotic division appears to be nor-
mal, but, owing to the high numbers of chromosomes and the small
size of the nucleus in diatoms, it is difficult to make out all the stages.
The two nuclei resulting from the first meiotic division are usually
normal. However, in Cocconeis one of the nuclei forms a polar body.
It is believed to be an aborted gamete. In Navicula seminulum (Geit-
ler, 1932), one nucleus becomes pycnotic and is ejected, whereas in
Navicula cryptocephala var. veneta one nucleus is reabsorbed (Geit-
ler, 1952f).

Cytokinesis usually follows the first meiotic division. Previous to
this the chloroplasts usually have divided. The cell membrane is de-
veloped in a plane parallel to the valves of the cell.

The second meiotic division usually follows cytokinesis. One of

SEXUAL Rl PRODUCnON IN DIATOMS 87

rhc two nuclei formed in this division usually degenerates. Sometimes
it is reabsorbed in the protoplasm as in Goinphouema parvulmn var,
iincropus (Geitler, 1932). In other instances it becomes pycnotic and
remains in the gamete as in Ejmotia arcus and Eunotia flexuosa (Geit-
ler, 1951a, b) or is cut off as in Aviphipleura pellucida. However, in
NiViiciila radiosa (Geitkr, 1952d) and Navicida cryptocephala var.
vcneta (Geitler, 1952f) both nuclei remain functional, so that each
gamete has two functional nuclei.

One or two functional gametes may develop in each cell. If two
gametes develop, they usually change position. Instead of lying paral-
lel to the valves in the position in which they are formed, they come
to lie one above the other when viewed from the apex of the cell.
In Navicula radiosa (Geitler, 195 2d) this change of position of the
gametes does not occur.

One notable exception to this type of gamete formation is that
found in Eunotia arcus and Eunotia flexuosa (Geitler, 1951a,b,c).
In these species there is a transverse differentiation of protoplasts.
In Eunotia arcus one chloroplast becomes very large, while in Eunotia
flexuosa they both move to the same side of the cell. The spindle of
the first meiotic division is formed so that one pole is close to the
epitheca. Thus a very unequal cell division takes place. The larger
cell develops into a gamete. The smaller cell forms what Geitler calls
a "remaining cell." Geitler thinks that this remaining cell, by affect-
ing the osmotic pressure of the cell, brings about the movement of
the gamete. After cytokinesis occurs, the second meiotic division
follows in both the gamete and the "remaining cell." One nucleus in
each degenerates.

Another exception is, for example, in Cymbella ventricosa var.
(Geitler, 1932) where parthenogenesis occurs. In such cases reduc-
tional division does not take place.

Sometimes it happens, as in Eunotia arcus (Geitler, 1951b), that
the cells resulting from meiosis develop shells and become vegetative
cells rather than gametes. This phenomenon has been observed several
times in diatoms.

The sex differentiation of gametes, if it occurs, takes place during
meiosis. Geitler (1932) thinks that the anisogamy recognized by the
difference in size and movement is more apparent than real. He states
that movement is due to tensions which develop in the cell. This is
well described for Aniphipleura pellucida (Geitler, 1952c). The first

SEX IN MICROORGANISMS

SEXUAL RI'PRODUCTION IN DIATOMS 89

giimcte to mature is the one that starts the movement. Carefully con-
trolled experiments are necessary to determine the true condition.

FUSION OF GAMETES

The fusion of gametes in diatoms is of three types. Isogamous
fusion is the result of equal movement of the gametes or of copula-
tion of the gametes with both of them in situ. Anisogamous fusion is
the result of unequal movement of the gametes. Autogamous fusion
is the result of the fusion of two gametes in the same mother cell.

Isogamous fusion may be of three types. In one type, as in
Amphora and Denticula (Geitler, 1932), the two gametes from each

Figures 1 to 12 and 16 to 29 are taken from various papers by Lothar Geitler.

Figures 13 to 15 and 20 to 22 after Fritsch.

Fig. 1. Goniphone7na panmlmu var. viicropus (Kiitz.) CI., two mother cells in
copulation jelly with two gametes in each cell (x 750, approx.).

Fig. 2, and 3. Go?>rpho?ie?Na parz'-iilmn var. viicropus (Kiitz.) CI., showing the
fusion of gametes (x 750, approx.).

Fig. 4. Go7nphone?>ia parviilmn var. micro pus (Kiitz.) CI., one zygote in each
mother cell (x 750, approx.).

Fig. 5. Spermatogenesis in Cyclotella sp., beginning of first meiotic division (x
900, approx.).

Fig. 6. Cyclotella sp., telophase of first meiotic division (x 900, approx.).

Fig. 7. Cyclotella sp., telophase of second meiotic division (x 900, approx.).

Fig. 8. Cyclotella sp., daughter protoplasts (x 900, approx.).

Fig. 9. Cyclotella sp., spermatozoa (x 900, approx.).

Fig. 10. Cyclotella sp., formation of oogonium (x 900, approx.).

Fig. 11. Cyclotella sp., mature egg cell with sperm nucleus (x 900, approx.).

Fig. 12. Cyclotella sp., zygote after fusion of sperm and egg nuclei (x900,
approx.).

Fig. 13. Cocconeis place nttila var. klinor aphis Geitler, gamete formation (x 500,
approx.).

Fig. 14. Cocconeis placentula var. klinorapbis Geitler, fusion of gametes ( X 500,
approx.).

Fig. 15. Cocconeis placefitula var. klinorapbis Geitler, auxospore formation (X
500, approx.).

Figs. 16 and 17. Govipbonenia constrictuni var. capitatuvi (Ehr.) V. H., two
gametes in each cell (x 400, approx.).

Fig. 18. Goinpbone'ina constrictuni var. capitatuvi (Ehr.) V. H., zygote resulting
from automixis, nuclei not fused (x 400, approx.).

Fig. 19. Govipbonema constrictum var. capitatuvi (Ehr.) V. H., zygote with
nucleus fused (x 400, approx.).

Figs. 20 and 21. Cocconeis placentula var. lineata (Ehr.) CI., nuclear reorganiza-
tion in two approximate cells ( X 500, approx.).

Fig. 22. Cocconeis placentula var. lineata (Ehr.) CI., auxospore resulting from
automixis (x 500, approx.).

90 SEX IN MICROORGANISMS

cell move into the jelly mass between the cells. In Rhopalodia gibba
(Klebahn, 1896) the jelly is restricted and appears as a bridge be-
tween the two cells. Two zygotes are formed in this bridge. This is
believed to be the most primitive type of auxospore formation.

Isogamous fusion may also take place within a copulation tube.
In this type the zygote is formed within the tube. This type of fusion
occurs in Eimotia arciis and Eimotia flexnosa (Geitler, 1951a,b).
In Navicula radiosa each half of each mother cell rotates through an
arc of 90 degrees and the two gametes copulate hi situ (Geitler,
1952d).

Anisogamous fusion occurs if one gamete is active and the other
is passive. As in isogamous fusion this may occur with or without a
copulation tube. In Gomphoneina parviiluni var. inicropus usually
four gametes are involved. One gamete migrates into the other mother
cell and fuses with the passive gamete. This stimulates the other
gamete to move out and into the first mother cell. As a result two
auxospores are formed. Of less frequent occurrence in Gomphoneina
pamihnn var. 7/iicropus is the production of only one zygote from
two gametes. Of rare occurrence is the production of one zygote
from three gametes (Geitler, 1932).

Of common occurrence in anisogamous fusion is the production
of a copulation tube. Depending on the species of diatom, one or two
tubes may be produced.

Usually one tube is produced, as in many species of Nitzschia
in which the gametes pass in succession through the tube. If two tubes
are present, as in Frnstulia rhoiiiboides \'ar. saxonica, the two fusions
may take place at the same time (Geitler, 1949b).

An unusual type of anisogamous fusion is that reported for
Navicula halopbila (Subrahmanyan, 1945), Syiiedra uhia (Geitler,
1935), and Syiiedra nmipens var. fragilarioides (Geitler, 1952f). In
these species two active gametes are formed in one riiother cell and
two passive gametes in the other mother cell. The resulting fusion
produces two auxospores in the same mother cell. No copulation tube
is formed.

Automixis is not common in diatoms. Several cases \\'hich need
further investigation indicate that this is the means of reproduction.
In no Ciise is the nuclear behavior thoroughly understood. What seems
to be a true case of autogamy is described for Aviphora norvhwii

SEXUAL RI PRODUCIION IN DIATOMS 91

(Gcirlcr, 1935), in w hich one auxosporc is formed from a single cell.
The protoplast contracts, and two nuclei, two nucleoli, and tw^o chro-
matophores are formed. The valves of tlic cell are spread apart, and
the protoplast is transformed into an auxospore. Later there is found
only a single nucleus with a nucleolus.

A modified type of autogamy occurs in AchnaJithes siibsessilis
and Govipboneina constrictiim var. capitata (Karsten, 1897, Geitler,
1952b). Within a single cell two gametes are formed which later fuse
to form a single protoplast which is transformed into an auxospore.

Parthenogenesis is a method of auxospore formation in Cocconeis
plciceiituhi var. I'meata (Geitler, 1932). The nucleus of the parent cell
goes through two divisions, which correspond to the two meiotic
divisions except that reduction in chromosome number does not oc-
cur. Polar bodies are formed. The protoplast then becomes trans-
formed into an auxospore. Parthenogenesis is also known to occur in
one of the varieties of Cyinbella ventricosa.

Asexual auxospore formation has been reported for Syiiedra
affinis (Karsten, 1897) and Rhabdonema arcuatwn. In these species
the mother cell divides by mitosis to form two daughter cells. These
protoplasts, instead of developing normal vegetative walls, become
auxospores. Further cytological investigation is needed to make sure
that this is truly asexual formation of auxospores (Fritsch, 1935).

Sometimes two types of auxospore formation occur within a
single mass of copulating cells. For instance, in Goiiiphonevia par-
viihiin var. micropus three cells come together. One cell forms an
auxospore by automixis and the other two produce auxospores by
heteromixis (Geitler, 1932).

The time interval for the fusion of gametes varies greatly. In
Navicula sevmiiihivj the fusion of gametes takes 2 to 3 minutes,
whereas in Ainphipleiira peUucida the process takes an hour (Geitler,
1932, 1952b).

DEVELOPMENT OF ZYGOTE AND AUXOSPORE FORMATION

On fusion of the gametes the zygote starts to develop. The fu-
sion of the nuclei is often delayed until the auxospore is developed.
In Navicula radiosa (Geitler, 195 2d), Navicula cryptocephala var.
veiieta (Geitler, 1952f) there are two pairs or four functional nuclei.

92 SEX IN MICROORGANISMS

During auxospore development one pair fuses and the other pair de-
generates.

The zygote may be found in various positions. As a result of
isogamy it is formed betu^een the mother cells. The polar axis of the
zygote is at right angles to that of the mother cells. As a result of
anisogamous fusion the zygote is first formed in the mother cell.
When two zygotes are formed, one is usually produced in each
mother cell. However, in Naviciila halophila (Subrahmanyan, 1945),
Synedra ulna (Geitler, 1939), and Syiiedra nimpens var. fragilarioides
(Geitler, 1952f) the two zygotes are produced in the same mother
cell. Likewise in automixis and parthenogenesis the zygote is first
formed in the mother cells. Later it migrates out of the mother cell.
Usually in anisogamous fusion the long axis of the auxospore is paral-
lel to that of the mother cell, whereas in isogamous reproduction the
long axis of the auxospore is perpendicular to the long axis of the
mother cells. However, if the jelly surrounding the copulating cells
is relatively thin, the auxospores may vary somewhat in position.
Geitler (1932) believes that this interesting correlation of the position
of the auxospore with type of reproduction is a result of tensions
developed within the jelly rather than a result of the type of gametes.

The zygote elongates in the formation of the auxospore. In this
process the zygote membrane often breaks and appears as caps on the
ends of the auxospore, as in Frustulia rhomboides var. saxonica (Geit-
ler, 1949b). In Anomoeoneis exilis the zygote membrane persists as
laminations over the poles of the auxospore (Geitler, 1949b). In
Nitzschia fonticola (Geitler, 1932) the zygote membrane is elastic
and does not break.

The perizonium, which is the auxospore membrane, develops un-
der the membrane of the zygote. It becomes weakly silicified. The
silicification starts at the center of the auxospore and develops out
toward the poles. The perizonium may develop a distinctive pattern of
markings or be smooth.

When the auxospore is mature, a nuclear division (metagamic
division) occurs. One of the resulting nuclei is pycnotic. This phe-
nomenon has been observed in various genera of the Pennales.

After a period of time the auxospore develops the shells typical
of the vegetative cell. The first shell to develop is the epitheca. It is
irregular in that it does not have a girdle band. Therefore the edges
of the valve bend over and the valve has a curved appearance. The

SIMM R I PRODUCTION IN DIATOMS 93

InporhccM, li()\\c\ cr, is noniuil in that it possesses girdle bands. Thus
(he first vegetative cell is not symmetrical in appearance, and there-
fore it differs from subsequent vegetative cells.

SUiMiMARY OF TVPIS OF AUXOSPORE FORMATION

\ arious authors ha\e made classifications of the different types
of auxospore formations. The classification given below is taken from
Geitler (1932) but modified to include the results of more recent
research.

Normal Type A

Two mother cells each produce two gametes, which copulate
in pairs to produce two auxospores (Figs. 1 to 4).

(1) The gametes are isogamous; the apical axes of the auxospores
are perpendicular to the apical axes of the mother cells. Amphi-
prora alata Kiitz. (?), Amphora coffaejonnis Ag., A. cymbelloides
Grun., A. ovalis Kiitz., A. ovalis var. pediciihis Kiitz., A. piisio CI.,
A. veneta Kiitz., Auricula hyalina Karst., Denticula vanheurckii
Brun., Epithemia argus Kiitz. (?), E. sorex Kiitz., E. mrgida (Ehr.)
Kiitz., E. zebra (Ehr.) Kiitz., E. zebra var. saxonica (Kiitz.) Grun.,
Navicnla radiosa Kiitz., Rhopalodia gibba (Ehr.) O. Miill., R. gibba
var. ventricosa (Ehr.) Grun.

(2) Each mother cell produces a wandering and a resting gam-
ete. The apical axes of the auxospores are parallel to those of the
mother cells. Achnanthes lanceolata Breb., A. vmiutissima Kiitz.,
Amphipleiira pelhicida Kiitz., A. rutilans (Trent.) CI. (?), Amphi-
prora alata Kiitz. (?), Anomoeoneis sculpta (Ehr.) Pfitz., A. serians
(Breb.) CI., Brebissonia boeckii (Ehr.) Grun., Cymbella affinis Kiitz.,
C. caespitosa var. pediculiis (Ehr.) Brun. (?), C. cistula (Hemp.)
Grun., C. cymbijormis (Kiitz.) Breb. (?), C. gastroides Kiitz., C.
helvetica Kiitz., C. lacustris (Ag.) CI., C. lanceolata (Ehr.) V. H.,
C. parva (W. Sm.) CI. (?), C. prostrata (Berk.) CI., C. simiatrensis
Hust., C. ventricosa Ag., C. ventricosa Ag. var. I and II, Frustidia
rhcmtboides var. saxonica (Rabh.) DeT., Goinphoiiema constrictum
Ehr., G. constrictum var. capitata (Ehr.) CI., G. geminatum (Lyngb.)
Ag., G. intricatiim Kiitz. (?), G. intricatinn var. dichotovnim (Kiitz.)
Grun. (?), G. longiceps Ehr. (?), G. olivaceinn (Lyngb.) Kiitz., G.

94 SEX IN MICROORGANISMS

parvulum var. ?mcropus (Kiitz.) CL, G. tenellwn Kiitz. (?), Libelliis
constrictus (Ehr.) DeT. (?), Navicida crucigera (W. Sm.) CI., N.
cuspidata var. avibigiia (Ehr.) CL, N. dire eta Ralfs., N. firma Kiitz.
(?), N. pygmaea Kiitz., N. ramossissima (Ag.) CL, N. scopzdorwn
Breb., N. subtilis (Greg.) Ralfs., N. viridula Kiitz., Niediuvi affijie
var. aviphirbyuch'iis (Ehr.) CL (?), Nitzschia hybrida Grun., N.
lofigissmm (Breb.) Ralfs., N. siginoidea (Ehr.) W. Sm., N. subtilis
Kiitz., Finnularia gibba Ehr., P. hemiptera (Kiitz.) CL, F. stauroptera
Grun. (?), P. viridis (Nitz.) Ehr., Rhoicosphenia ciirvata (Kiitz.)
Grun., Schizonevia laczistre Ag., Stauro7ieis phoenicenteron (Nitz.)
Ehr. (?).

(3) One mother cell produces two wandering gametes, and one
mother cell produces two passive gametes. The apical axes of the
auxospores are parallel to those of the mother cells. Navicula halo-
phila (Grun.) CL, Synedra ulna (Nitz.) Ehr., 5. ruvipens var. jragi-
larioides Grun.

(4) The gametes behave according to no rule; the auxospore
position varies. Achnantbes brevipes Ag., A. lanceolata Breb., A. hn-
giceps Ag., Navicula didyvm (Ehr.) Kiitz., N. fonticola Grun., N.
hybrida Herib. & Per., Nitzschia longissima (Breb.) Ralfs., Pleuro-
sigina nubecula W. Sm.

Normal Type B

Spermatozoa and an tgg cell are formed (Figs. 5 to 12).

( 1 ) An antheridial cell buds off two spermatozoid mother cells
each of which produces two spermatozoids. One oogonium produces
one tgg cell. A spermatozoid enters the tg^ and a zygote is formed.
Me I o sir a variaus Ag.

(2) Four spermatozoa are produced from an antheridial cell.
One oogonium produces one egg cell. A spermatozoon enters the
egg and a zygote is formed. Cyclotella sp.

Reduced Type A

Two mother cells each build one gamete; these fuse to form a
single auxospore (Figs. 13 to 15).

( 1 ) The gametes behave isogamously. Cocconeis pediculus Ehr.,
C. placentula Ehr., C. pi ac en tula var. klinor aphis Geitler, C. placen-

SEXUAL REPRODUCTION IN DIATOMS 95

tnlii var. tciiiiistrhita Cicirlcr, Cy mat u pleura solea (Brcb.) W. Sm.,
Ewiotia avciis I'.hr., E. pexuosa Kiir/., E. formica I^lir., E. pectrnalis
(Kiitz.) Rabli., Naviaila crytocepbala var. veneta (Kiitz.) Grun.,
Rboicosphema nirvata (Kiitz.) Grun., Si/rirella caprovii Breb., S.
spleiidida Ehr., 5. striatnla Turp.

(2) The gametes behave anisogaiiiously. Navicula seminiilimi
Grun. coccoiieis pediaihis Ehr., C placeutula Ehr., C. placeiitula var.
pseudolincata Geitler.

Reduced Type B

One mother cell develops an auxospore through automixis (Figs.
16 to 19).

( 1 ) Two gametes of one mother cell copulate with each other.
Achuaiitbes siihsessilis Kiitz., Cyclotella meneghiiiiana Kiitz, Gom-
pbonevia coustvictuvi var. c a pit at a (Ehr.) V. H., Form I.

(2) The sexual nuclei of a mother cell copulate. Ampbora nor-
iihwii Rabh., Bacillaria paradoxa Gmelin (?), Chaetoceras boreale
Baily, C deiisiivi CI., Gravmiatophora marina (Lyngb.) Kiitz., Libel-
lus cojistrictiis (Ehr.) DeT. (?), Navicula coustricta Grun. (?),
Nitzschia palea (Kiitz.) W. Sm.

Reduced Type C

The auxospore formation is apomictic (Figs, 20 to 22).

( 1 ) From one mother cell there develop through vegetative di-
vision two auxospores. Acbuautbes loiigipes Ag. (?), Bacillaria para-
doxa Gmelin (?), Coccoiieis pedicidiis Ehr., Libelliis constrictus
(Ehr.) DeT., Navicula constricta Grun. (?), Rbabdoiiema arcuatuin
(Lyngb.) Kiitz., Syiiedra affinis Kiitz., Tabellaria sp.

(2) From one mother cell (the mother cells may pair) there
develops one auxospore.

(a) Parthenogenetically. Bacillaria paradoxa Gmelin (?), Coc-
coiieis pedicuhis Ehr., C. placeutula Ehr., C. placentula var. klino-
rapbis Geitler, C. placeutula var. lifieata (Ehr.) CI., C, placentida var.
euglypta (Ehr.) CI., Cymatopleura elliptica (Breb.) W. Sm., C. solea
(Breb.) W. Sm., Cymbella cistula (Hemphr.) Kirchn. (?), C. swna-
treiisis Hust. (?), C. ventricosa Ag. var. I, Grainmatopbora mariiia
(Lyngb.) Kiitz. (?), Meridion circidare (Grev.) Ag., Navicida

96 SEX IN MICROORGANISMS

grevillii (Ag.) Heib., Nitzschia palea (Kiitz.) W. Sm. (?), Rhabdo-
nema adriaticinn (Lyngb.) Kiitz., Sjirirella gemma Ehr.

(b) Purely vegetatively. Bacillaria paradoxa Gmelin (?), Melo-
sira and other Centrales.

DISCUSSION AND CONCLUSIONS

In diatoms the most common method of reproduction is vegeta-
tive division by mitosis. However, there do occur other types of re-
production which have been discussed in this paper. As in the proto-
zoa, fungi, and other algae, it is hard to determine whether there is
true sexual differentiation. It is reproduction which occurs as a result
of meiosis and fusion. This type of reproduction has only rarely been
found in the Centrales. In the Pennales it has been found in numerous
species, but in most of them only rarely observed. This is particularly
apparent when one considers that there are several thousand species
of diatoms and that thousands of collections of diatoms have been
made.

It is interesting to consider how this reproduction pattern, of
commonly occurring asexual reproduction with infrequently occur-
ring sexual reproduction, affects the structure of diatom species, their
distribution, and the evolution of the group.

The common concept of the species is based upon populations
in which sexual reproduction is obligate and each individual has a
different genotype. Thus in a large population there is a great deal of
intergrading variation. In diatoms there are species composed of many
clones differing by sometimes small but disjunct variations. Each
clone consists of many individuals with the same genotype. It is per-
haps for this reason that we have commonly in diatom species so
many named varieties and forms. This difference in species structure
requires that any study of population structure based on random sam-
pling must be carefully planned with these facts in mind.

This type of reproduction may also contribute to the distribu-
tional pattern of this group. Most species of diatoms have wider dis-
tributional patterns than are commonly found in higher plants. When
colonization is attempted by an obligate sexually reproducing species,
it is necessary for at least two individuals of opposite sex to be living
successfully in a given environment. In an asexually reproducing form
one individual may establish a colony. Thus, as Stebbins (1950) has
pointed out, asexual reproduction may bring about more rapid coloni-

SEXUAL REPRODUCTION IN DIAIOAIS 97

/ntioii of a species. The fact that diatoms arc diploid rather than
haploid means that each individual may have genie flexibility for any
trivcn trait. This, coupled with asexual reproduction, no doubt is an
important factor in establishing tlie \\ide distribution patterns of the
species. Of course, many other factors contribute to producing the
broad distributional pattern of diatom species (Patrick, 1948). This
association of a wide distribution pattern of a species with asexual
reproduction is also found in other algae, protozoa, and fungi.

The effect of this reproduction pattern on the evolution of dia-
toms is hard to evaluate, for it seems to produce effects some of w^iich
Mould favor evolution while others would slow it down. Asexual re-
production by its very nature greatly reduces chromosomal change,
eliminates the accumulation of mutations which have occurred in var-
ious genotypes, and the recombination of genotypes. It is interesting
to note that in the Centrales, in which reproduction by fusion has
only rarely been observed, there are a fairly large percentage of spe-
cies which have remained constant since Miocene times.

In the Pennales, how'ever, the effect of asexual reproduction in
reducing the rate of change in the species may not be so great. In
this group "sexual reproduction" has been observed in a greater num-
ber of species. As Stebbins (1950) points out, in organisms with
short generations the number of genie recombinations per generation
can be reduced without affecting the flexibility in terms of the num-
ber of gene combinations available in a given unit of chronological
time. Thus, if "sexual reproduction" occurs often enough, the genie
flexibility \y'\\\ be preserved. The question then is how frequently does
"sexual reproduction" occur.

The structure of the diatom species offers a favorable condition
on which natural selection can operate. A species is composed of seg-
regated population units in the form of clones between w^iich gene
exchange through sexual reproduction occasionally takes place. It
must be remembered, however, that these segregated population units
consist of a single genotype rather than several genotypes, as is usually
the case in segregated population units of obligate sexual species.

The rate of evolution in diatoms undoubtedly is influenced by
these various factors. Because of the varying frequency of sexual re-
production in different species, these algae are an interesting group
for the study of some of these basic problems.

Note: Since this paper w^as written Geitler (1953a) has shown
that Denticula tenuis Kiitz. may produce auxospores by Normal

98 SEX IN MICROORGANISMS

Type A-1 or by Reduced Type B-1. He has also shown (1953b)
that in Cocconeis phcenwla var. teimistriata Geitler two metagamic
divisions occur. The first division is associated with the formation of
the first shell in the auxospore and the second one with the second
shell. Geitler and Mack (1953) point out that although in the genus
Cymbella there is variation in the relation of the position of the axis
of the auxospore and the position of the first shell to that of the
mother cell, the arrangement for many species is very definite. Nip-
kow (1953) states that he has seen asexual auxospore formation by
Reduced Type C-2-b in Fragilaria crotonensis Kitton but does not
describe the various steps in the process.

REFERENCES

Bergon, P. 1907. Les processus de division, de rajeunissement, de la cellule et de

sporulation chez le Biddiilphia inobilievsis Bailv. Bull. soc. Botaii. France,

54, 327-358.
Calkins, G. N., and F. M. Summers. 1941. Protozoa in Biological Research.

Columbia University Press, New York.
Fritsch, F. E. 1935. The Structure and Reproduction of the Algae. Vol. I.

The Macmillan Co., New York.
Geitler, L. 1928. Kopulation und Geschlechtsverteilung bei einer Nitzschia —

Art. Arch. Protistevk., 61, 419-442.
. 1932. Der Formwechsel der pennaten Diatomeen. Arch. Protistenk.,

78 [1], 1-226.
. 1935. Reproduction and life history in diatoms. Botan. Rev., 1 [5],

149-161.
. 1939. Gameten- und Auxosporenbildung von Synedra ulna im Ver-

gleich mit anderen pennaten Diatomeen. Planta, 30 [3], 551-567.
. 1940. Die Auxosporenbildung von Meridian circulare. Arch. Protis-
tenk., 94 [2], 288-295.
. 1949a. Die Auxosporenbildung von Nitzschia signwidea und die

Geschlechtsbestimmung bei den Diatomeen. Portugaliae Acta Biological,

series A, Vol. R. B. Goldschmidt, 79-87.
. 1949b. Beitrage zur Kenntnis der Auxosporenbildung pennater Dia-
tomeen. Oster. botan. Z., 96 [3-4], 467-473.
. 1951a. Pragame Plasmadifferenzierung und Kopulation von Eunotia

flexuosa (Diatomee). Osterr. botan. Z., 98 [4], 395-403.
. 1951b. Kopulation und Formwechsel von Eunotia arcus. Osterr. botan.

Z., 98 [3], 292-338.
. 1951c. ZelldifTerengierung bei der Gametenbildung und Ablauf der

Kopulation von Eunotia (Diatomee). Biol. 7.entr., 70, 9/10, 385-398.
. 1952a. Die Auxosporenbildung von Rhoicosphenia curvata. Osterr.

botan. Z., 99 [1], 78-89.

SEXUAL REPRODUCTION IN DIATOMS 99

— . 19521). Untersucliungeii iibcr Kopulation und Auxosporenbildung pen-
niitcr Diatonicen. I. Autoinixis bei GoviphoneiJia constrictinn var. capitata.
Ostcrr. hot an. Z., 99, 376-384.

— . 1952c. Untcrsuchungen iibcr Kopulation und Auxosporenbildung
pcnnatcr Diatomecn. II. Wander- und Ruheganicten bei Ainphipleura
pcllucida. Ostcrr. botau. Z., 99, 385-395.

— . 1952d. Untcrsuchungen iiber Kopulation und Auxosporenbildung pcn-
natcr Diatomecn. III. CJleichartigkcit dcr Goncnkerne und Verhaltcn des
Hctcrochroniatins bei Naviciila radiosa. Osterr. botan. Z., 99, 469-483.

— . 1952e. Ooganiic, Meiose und nietaganic Teilung bei der zentrischen
Diatomce Cyclotella. Osterr. botaji. Z., 99, 506-521.

— . 1952f. Untcrsuchungen iiber Kopulation und Auxosporenbildung pcn-
natcr Diatomecn. IV. Vicrkcrnige Zvgoten bei Navicida cryptocephala
var. veiietafa. V. Allogamie bei Sy^iedra rum pens var. fragillarioides.
Osterr. botan. Z., 99, 598-605.

— . 1953a. Allogamie und Autogamid bei der Diatomic Denticida tenuis
dun die Geschlcchtsbestimmung der Diatomecn. Osterr. botan. Z., 100,
331-353.

1953b. Das aufiutcn zvveicr obligater, metagamer mitoscn ohmc Zcll-

teilung wahrend der Bildung der Erstilingsschalen bei den Diatomecn.

Ber. dent, botan. Ges., 66, 222-228.
Geitlcr, L., and B. Mack. 1953. Die Ackscnlagcn der auxosporen und Erstlings-

zellcn bei der Diatomic Cynibella. Osterr. botan. Z., 100, 261-265.
Iyengar, O. P., and R. Subrahmanyan. 1944. On reduction, division and auxo-

spore formation in Cyclotella vieiieglmiiana. J. Indian Botaiiy, 23 [4],

125-153.
Karsten, G. 1897. Untcrsuchungen iibcr Diatomecn III. Flora 83, 203-222.
Klebahn, H. 1896. Bcitrage zur Kenntniss der Auxosporenbildung I. Rhopa-

lodia gibba (Ehr.) O. Miill. Jabrb. wiss. Botan., 29, 595-654.
Liebisch, W. 1929. Experimentclle und Kritischc Untcrsuchungen iiber die

Pektinmcmbran der Diatomecn iinter besondcrer Beriicksichtigung der

Auxosporenbildung und der Kratikularzustandc. Z. Botan., 22, 1-65.
Nipkow, F. 1953. Die Auxosporenbildung bei Fragilaria crotonefisis Kitton ini

Plankton des Ziirichsces. Schiveiz. z. Hydrologie, 15(2), 302-311.
Patrick, Ruth. 1948. Factors effecting the distribution of diatoms. Botan.

Rev., 14 [8], 473-524.
Schiitt, F. 1893. Wechselbezichungen zvvischen Morphologic, Biologic, En-

twicklungsgeschichtc und Svstematik der Diatomecn. Ber. dent, botan.

Ges., 11, 563-571.
Stebbins, G. J., Jr. 1950. Variation and Evolution in Plants. Columbia Uni-
versity Press, New York.
Subrahmanyan, R. 1945. On somatic division, reduction division, auxosporc

formation and sex differentiation in Navicula halophila (Grun.) CI. Cur-

rent Sci. (India), 14 [3], 75-77.
von Stosch, H. A. 1950. Oogamy in a centric diatom. Nature, 165 [4196],

531-532.

Sex in Unicellular Algae

RALPH A. LEWIN,* Osborn Botanical Laboratory,
Yale University, New Haven, Connecticut

At the outset it would be well to define the scope of this review.
AVaddington (1939) interprets "sex" in terms of sexual differentia-
tion, Sinnott and Dunn (1939) extend it to cover gametogenesis and
zygote formation, while Darlington (1937) considers sexual repro-
duction as embracing not only syngamy but also meiosis, as comple-
mentary processes essential for the completion of a sexual cycle. In
many protists there is no differentiation between cells capable of
acting as gametes; in only a few has the sexual cycle been followed
through genetically; while in Escherichia coli knowledge of almost
all aspects of sex but the purely genetic is for the most part inferen-
tial. For these reasons, in a field where information is so sparse and
fragmentary, the author hopes that a certain latitude of interpretation
will be allowed.

Smith (1951a) considers the class of algae to be divided into
seven divisions. Among these the Cyanophyta are non-sexual, the
Rhodophyta and Phaeophyta are virtually devoid of unicellular rep-
resentatives, whereas in the Chrysophyta, the Pyrrophyta, and the
Euglenophyta sexual reproduction is unknown in all but a few excep-
tional cases. These flagellates are further discussed by Wenrich (see
"Sex in Protozoa: A Comparative Review " in this volume). A search
for experimental material must be thus largely restricted to the dia-
toms and the green algae. It is perhaps regrettable that the ubiquitous
species of Chlorelk, Scenedesmus, and Stichococcns, so favorable in
other ways for laboratory study, have not been known to exhibit
sexual reproduction. The desmids offer promising material; but the
success of Pringsheim (1919) in controlling zygote formation and
germination in Cylindrocystis has apparently not been followed up.

* Present address: Maritime Regional Laboratory, National Research
Council, Halifax, N.S., Canada.

Drs. S. H. Hutner, J. C. Lewin, and L. Provasoli have offered many crit-
ical suggestions which have been considered in the preparation of this review.

100

SEX IN UNICELLULAR ALGAE 101

It is therefore \\ith the Volvocales that this review will be chiefly
concerned.

No attempt has been made to discuss the literature on this sub-
ject exhaustively here. The field has already been well surveyed by
Fritsch (1935), Moewus (1941), and Smith (1951a); see also Luyet
(1950). Particular attention will be given here to studies of cultures
and experiments under controlled conditions, where a measure of
reproducibility can be expected in the results. The most critical in-
vestigations are those which involve material of genetic uniformity
and known physiological background, in a constant physical and
chemical environment and in the absence of any other living organ-
isms.

SEXUAL DIFFERENTIATION

Vegetative Cells and Gametes

In a unicellular organism sexual differentiation may be taken to
indicate a difference between the two sexes or mating types; it may
also be considered to refer to the differentiation often found between
vegetative cells and gametes (see Moewus, 1933). In the simplest
case, as in many species of Chlmnydomonas (Klebs, 1896; Smith,
1950b), Polytoma (Pringsheim and Ondratschek, 1939), and Dima-
liella (Lerche, 1937), all haploid cells may be capable of sexual fusion.
In Chlanty domonas gymnogyne (Pascher, 1943) the sexually active
cells are morphologically identical with vegetative cells but appear to
differ in their more trembling mode of progression. However, in most
Chlamy domonas species and in related genera, for example, Chlo-
romovas (Korschikoff, 1926), Chlorogoniwn and Haentato coccus
(Schulze, 1927), Fhyllocardium (Korschikoff, 1927), and Brachio-
vmnas (Moewus, 1944), it may be observed that only the smaller
cells, those most recently liberated after cell division, take part in
mating. Each young cell may be said to pass through a gametic stage
and then to lose its sexual activity as it enlarges prior to asexual divi-
sion. In some such cases the cells may be naked during the gametic
phase and only develop cell walls later. Nayal (1933) has described
in Protosiphon the formation of two types of swarmers, rounded zoo-
spores and elongate facultative gametes, both capable of direct devel-
opment into vegetative cells.

102 SEX IN MICROORGANISMS

\n Chlamy dojnonas paupera (Pascher, 1931-32) differentiation is
carried a stage further. The species is homothallic and isogamous; the
gametes after liberation develop first a male sexual potentiality and
later, if not involved in sexual fusion, become female. Unmated gam-
etes in culture ultimately die vi^ithout growing into a form capable of
asexual reproduction (that is, they appear incapable of parthenogene-
sis), so that here one finds a dichotomy between obligate gametes and
asexual cells. A similar condition has been described in Tetraspora
(Geitler, 1931), in Stephanosphaera (Moewus, 1933), and in the mi-
crogametes of C. praecox (Pascher, 1943). (According to Moewus,
1941, in some algae the ability of gametes to reproduce asexually is
controlled by a single, partially sex-linked gene. In C. eiigametos, in
which all motile cells are potentially gametes, he claimed to have
obtained by x-irradiation numerous mutants in which the gametes
lacked this ability to divide vegetatively. This paradoxical situation
awaits explanation.) Since both gametes and vegetative cells are pro-
duced after haploid mitosis, it is unlikely that they could differ geneti-
cally; and in a case of this sort it would be of great theoretical inter-
est to determine what synthetic faculty or cytoplasmic moiety may
be lost in gametogenesis, thereby rendering the sex cells incapable of
further multiplication in the haploid condition.

Outbreeding Mechanisms

In sexually reproducing organisms the frequency of sibling mat-
ings may be reduced ( 1 ) by sexual dimorphism between the gametes
(by anisogamy or oogamy) or (2) by genetically inherent factors of
self-sterility, or intraclonal sexual incompatibility. These two out-
breeding mechanisms are not mutually exclusive, and homothallic as
well as heterothallic species may be found to exhibit isogamy, ani-
sogamy, or oogamy. Multipolar heterothallism has not been demon-
strated in any alga. Subdioecism or subheteroecism has been described
in Chlamydomo7ias and other genera by Aioewus (1934 et seq.).

Sexual Dimorphism. In some species {Chlamydomonas el on gat a
for example) the gametes may be of various sizes but capable of pair-
ing in all combinations — a condition referred to by Korschikoff
(1923) as ataktogamy. In others the gametes are morphologically
similar but exhibit what may be described as physiological anisogamy;
for example, C. gymnogyne, in which only one gamete in each pair

SEX IN UNICELLULAR ALGAE 103

regularly sheds its cell wall before cytogamy (Pascher, 1943), and
C. '/noennsii, in w hich only one gamete remains motile after pairing
(Lewin, 1950a). (See page 119.)

The condition in which the two types of gametes are morpho-
logically^ distinguishable is generally known as anisogamy or heter-
ogamy; wliere one gamete is non-motile at the time of fertilization
(and is thereby designated as female), the species is said to be ooga-
mous. Sexual differentiation of this sort has been discussed by Fritsch
(1935) and Smith (1950a, 1951a), who have pointed out that more
than one condition may exist within a single genus. Thus sexual re-
production has been observed in only about 10 to 20 per cent of the
described species of Chlaiuydoijiouas; of these over forty are isoga-
mous species, about eleven are anisogamous, and three show fairly
well-marked oogamy (Skuja, 1949). Fhyllo7nonas striata exhibits
marked anisogamy (Korschikoff, 1926); Chlorogo?iiufn oogaimim
(Pascher, 1931) and Carteria iyengarii (Ramanathan, 1942) are ooga-
mous. If we are to consider heterothallic species exhibiting anisogamy
or oogamy as derived from heterothallic isogamous forms, we may
envisage the evolution of the larger, or female, gamete from either
the plus or the mimis mating type, and we need not expect that in
all evolutionary lines the same mating type would have become the
female. This consideration can hardly be pursued, however, until
some of the physiological bases for heterothallism have been eluci-
dated.

HeterothaUism. There are few diploid unicellular algae other
than the pennate diatoms, and of the latter almost all species which
have been critically examined appear to be homothallic or monoecious
(Geitler, 1949). In the case of Navicula balophila, however, Subrah-
manyan (1946) has presented circumstantial evidence for genetically
controlled dioecism, a condition found in Fiiciis vesiculosus, for in-
stance, but almost unknown in the lower Protophyta. Sex in diatoms
is more fully discussed (in this volume) by Patrick.

Among haploid organisms, homothallism (monoecism or synoe-
cism) may be defined as the condition in which a complete sexual
cycle can take place within a single clone, and heterothallism (dioe-
cism or heteroecism) as that state in which two haploid clones of
different genotype — different genetic mating type — are required for
sexual reproduction.

A discussion of homothallism, specifically in combination with

104 SEX IN MICROORGANISMS

isogamy, would hardly be complete without some reference to the
heated controversy which raged between biologists at Berlin and
Prague during the 1920's and 1930's. Hartmann (for example, 1932,
1943) adopted the theoretical anthropomorphic concept, borrowed
perhaps from Aristotle (for example, 1910), that there can be no
sexual union without sexual differentiation. If compatible gametes are
morphologically and, as in homothallic species, genetically identical,
then, he postulated, there must nevertheless exist some invisible phys-
iological difference between them, and the strain is said to possess
"bisexual potency." On the other hand, Mainx (1933) and Czurda
(1933a) saw no reason for adopting this hypothesis, for which they
found no corroboration in their experimental observations on algae
and other organisms, and they freely accepted the fusion of identical
cells in syngamy, just as it occurs in the hyphal anastomoses of fungi,
or the formation of plasmodia in Myxomycetes.

And then, in the early 1930's, Franz Moewus, a student of Kniep
and Hartmann at the Kaiser Wilhelm Institute, reported experimental
support for the theory of bisexual potency from his investigations
of unicellular algae such as Frotosiphoii (1933, 1935a), Folytoma
(1937), and Chlamy domonas eugmnetos synoica (1938a). In a clonal
culture of such homothallic, isogamous forms, both "male" and "fe-
male" gametes are said to be produced; and, when pairing takes place,
unless precisely similar numbers of the two mating types are present,
there will of necessity remain a few residual gametes {Restgameten)
of the supernumerary sex. Suspensions of residual gametes from sev-
eral cultures, freed in some way from already paired cells, may be
tested for mating type by mixing them in combinations of two or
with heterothallic tester stocks and observing in which mixtures mat-
ing is reinitiated. Lerche (1937) reported preliminary observations on
Haematococcus, in which, using morphologically distinguishable red
and green clones, she was able to demonstrate the regularity of resid-
ual gamete fusions; and Moewus (1940a) carried out a similar experi-
ment, using the marker "eyeless," in Botrydiinn granulatwn. But
Pringsheim and Ondratschek (1939) in Prague sought in vain to con-
firm Moewus' assertions about residual gamete behavior in Folytoma
and Protosiphon, thereby lending further weight to the initial objec-
tions of Czurda (1933a). Pringsheim and Ondratschek pointed out
that residual gametes could hardly be expected to behave in the man-
ner described by Moewus in organisms where sexuality is claimed to

SI X IN UNICELLULAR ALGAE 1()5

l)c hirLjcK- controlled l)\' the presence of soluble "sex substances" in
the medium (Aloewus, 1938b, etc.) (see page 115). From the evolu-
tionar\- standpoint, it would seem that a homothallic isogamous spe-
cies, with non-genetic mating-typc differentiation within a single
clone, would be at a disadvantage, since such a mechanism would
prcsumabh' reduce the potential sexual fertility of the species, while
in no wise promoting outcrossing or bringing other compensatory ad-
vantages.

Another concept inherent in Hartmann's theory of sexual dif-
ferentiation is that of relative sexuality. Several examples of such
behavior in different genera of algae, such as Chlamydomonas, Poly-
tovia, and Protosipboii) , have been reported by Moewus, and in cer-
tain cases viable zygotes have been obtained by mating "strong"
gametes with "weak" gametes of the same sex. Extensive physiologi-
cal and genetic investigations have been described in many such cases,
and these have been reviewed in some detail by Moewus (1939a,
1941, 1950a), Chodat (1941), Smith (1951a), and others. Relative
sexuality has apparently not been established in any unicellular alga
investigated by other \\'orkers. No report of sex reversal in algae has
come to the notice of the present author.

ISOLATION OF SEXUAL STRAINS

The isolation of unicellular algae capable of sexual reproduction
under laboratory conditions has always presented difficulties. These
are attributable partly to the fact that the diplophase is rarely iden-
tifiable in nature, and therefore, in the heterothallic species, two hap-
loid clones of complementary mating type must first be isolated. It is
reasonable to assume that, in many species, reproduction is entirely
asexual; M^hereas in others the conditions favorable for eliciting a
sexual response, whether in homo- or heterothallic forms, are un-
known except for fragmentary and largely empirical data for a few-
species. Schreiber (1925) had no success in finding mating strains
among seventy Chlaviydovionas clones tested.

Aloewus (1931) succeeded in isolating mating types of a species
of Chlamydomonas which he accordingly named C. eiigametos, later
retaining the name despite its prior description as C. sphagnophila by
Pascher in 1930 (see Moewus 1934, 1935b). A number of sexually
compatible clones of Chlainy domonas , probably identical with C.

106 SEX IN MICROORGANISMS

eugmnetos, were isolated from various soil samples by Gerloff (1940),
who described the species as C uioenjoiisii. Apart from a single homo-
thallic strain from the Cameroons, all these isolates conformed to
regular heterothallism. In 1948 Provasoli isolated complementary
mating types of a Chlamy doiitonas, the description of which closely
accorded with that of C. nioeumsn Gerloff. It has been found (see
Hutner and Provasoli, 1951, and Smith, in litt., November 1951) that
these isolates mate in reciprocal combinations with strains of C. eii-
gametos obtained directly or indirectly from Moewus, and it therefore
seems very probable that the two names are synonymous, as Gerloff
(1940) had originally suggested.

Smith recognized the difficulty of finding sexually active strains
among haploid clones isolated at random from soil and accordingly
devised a procedure (1946, 1947) which yielded considerably bet-
ter results. Essentially, this consisted of the examination of freshly
flooded agar cultures for the formation of mating pairs, which would
indicate either homothallism or the presence of both mating types
of a heterothallic species, originating from a single diploid zygote.
He reported (1950b) the isolation by this means of fifteen hetero-
thallic, not interfertile, species of Chlmnydo7nonas, and (1951b) a
number of homothallic species. Using a modification of this method
involving a prior enrichment for zygospore-forming species, Lewin
(1951) isolated from ten soil samples three heterothallic and two
homothallic species.

HYBRIDIZATION

In 1916 Pascher reported the hybridization of two species of
Chlamy domo7ias and described the progeny emerging from the hy-
brid zygotes. Strehlow (1928-29) obtained zygotes by crossing C.
paradoxa {plus) with C botryoides {viimis), though he could not
effect the reciprocal cross, nor could he induce the zygotes to ger-
minate. A4oewus (1935b) reported success in crossing C. eugametos
with C. paupera, obtaining zygospores which showed a greater sensi-
tivity to chilling than did those of the parent types, and haploid
progeny exhibiting segregation of six morphological character pairs.

He went on to describe extensive experiments in hybridization
of Folytoma spp. (1937), and of other species of Chlamy dojnonas.
In 1940 he reported that he had been able to isolate from 500 soil

SEX IN UNICELLULAR ALGAE 107

samples a number of clones of five species including anisogamous and
oogamous forms capable of interspecific mating. If we are to accept
Pascher's classification (1927) of this heterogeneous genus, then these
interfertile species of Aloewus' include representatives of three differ-
ent subgenera (Smith, 1946).

Since doubt has been cast on Pascher's results by Hartmann
(1934), and on those of Aloewus by a number of other workers (see
page 126), the information available in this field can hardlv^ be de-
scribed as illuminating.

INDUCTION OF SEXUAL ACTIVITY

A considerable amount of attention has been focused on the
physiological conditions which may induce cells to become sexually
active or may promote gametogenesis. As yet, no clear picture has
emerged from the many investigations carried out on various organ-
isms by a number of workers, each employing a different approach
(see reviews by Czurda, 1933b, and Bold, 1942). In the case of het-
erothallic species, it is of course essential at the outset to have isolates
of both mating types; and this difficulty has undoubtedly contributed
to the failures reported in the past (for example, by Reichenow,
1909, for Haematococcus; cf. Schulze, 1927). A few of the factors
\\hich have been found to influence sexuality are discussed below.

Light

Freliminary Observations. Klebs (1896) found that, though
dim light or darkness favored the formation of gametes in Proto-
siphon, light tended to promote the process of copulation. He ob-
served that svvarmers, formed in a mineral nutrient medium, could
be reversibly stimulated to sexual activity either by dilution of the
medium or by illumination, and he postulated that the action of light
might be through the photosynthetic formation of organic com-
pounds, which combined with and neutralized some constituent (s) of
the medium inhibitory to copulation. However, he was not able to
identify this inhibitor \A'ith the nitrate, phosphate, potassium, or cal-
cium salt in the medium he used.

In Chlamydovionas media (Klebs, 1896) and C. eugavietos (Aloe-
wus, 1933), light appeared essential for sexual activity, which ceased

108 SEX IN MICROORGANISMS

completely in darkness, and Moewus attributed these effects to the
formation and destruction of "sex substances" (see page 115). In an
attempt to confirm some of Moewus' results, though unable at the
time to obtain cultures of C. eugametos, Smith (1946) examined the
sexual behavior of a number of Chlamy domonas species isolated from
Californian soils. He found that all three heterothallic species tested
were capable of mating in darkness, though the activity was appre-
ciably stimulated by light. Lerche (1937) reported that light was not
essential for the initiation of mating in Dunaliella, and Maher (1946-
47) showed that Frotosiphon was capable of completing its sexual
cycle in complete darkness.

Quality of Light. The experiments reported by Smith in 1946
were made on suspensions of cells which had been initially grown in
light and were subsequently darkened prior to being tested for mating
ability. In 1948, having isolated the facultatively heterotrophic Chla-
mydomonas reinhardi, he was able to extend his observations to cul-
tures which had been grown in darkness on nutrient media contain-
ing sodium acetate. He found that such cells, although motile, were
incapable of sexual activity unless they had been subjected to a pe-
riod of illumination. Red light (6150 to 5900 A) or blue light (4357
A) was tested in place of white light, and in all cases the sexual
clumping of cells could be induced. In this respect C. reinhardi ap-
parently differs from C. eugametos, which, according to Moewus
(1939b), can only receive sexual stimulation from light at the blue
end of the spectrum (4300 to 5000 A).

In his description of C. chlaviydogama, Bold (1949) stated that
copulation occurred only in illuminated cultures. Using the strains
isolated by Bold, Cadoret (1949) was able to confirm the observation
and to investigate the character of light responsible for the physio-
logical effect. His experiments indicated that for C. chlaviydogama,
unlike C. reinhardi, white light could not be replaced by either red
or blue Hght, but that only a combination of both e]ualities of light
would induce sexual clumping and pairing. Cadoret suggested that,
for the manifestation of sexual activity, light must be absorbed by
two distinct substances, perhaps green and orange in color respec-
tively, and that chlorophyll and a carotenoid may both be involved.
The role of carotenoids in the sexual life of C. eugametos has been
given some weight in the publications of Moewus and co-workers

SEX IN UNICELLULAR ALCAE

109

(sec Alocwus, 1939c, 1940c, 1950b, for example); the possibility that
chlorophyll may be involved is discussed briefly below.

Provasoli and Pintner (sec Hutner and Provasoli, 1951) found
tliat liolit was required for the mating of C. vwcunisii. Preliminary
experiments indicated that any wavelengths within the visible range
Mere acti\c in promoting sexual clumping and pairing, and they con-
cluded that at a sufficient intensity all qualities of light were stinui-

1 1 1 1 1

y\. CHLOROPHYLL ABSORPTION

300

-/ \ ^^^ ■

# \^^4 REACTIVATION OF ^^^ \

M \ \ '^*"'^"'<» ACTIVITY ^ X ^

200

~ M \m m 1 \ ~

z

2

lO 100

~ u \ m m ^T \

(0

a.

<

^^ 1 1 I 1 1 1

400

500

600

700

WAVELENGTH (m;))

Fic. 1. Action spectrum for sexual induction in C moeioiisii Gerloff.

latory. It has since been shown (Lewin, unpublished) that in this
species light is required for the sexual activation of both mating types,
and unless both gametes have received illumination no mating takes
place. Not all wavelengths are equally effective: an action spectrum
for sexual induction in C. 7iioewusii is presented in Fig. 1. It will be
observed that the form of this curve is not dissimilar to the absorp-
tion spectrum of chlorophyll. The suggestion is therefore made that
the action of light on mating in Chlamy domonas is mediated by the
chloroplast, and that sexual activity is promoted by some product or

no SEX IN MICROORGANISMS

State resulting directly or indirectly from photosynthesis. It is sig-
nificant that sexual activation and photosynthesis are inhibited to the
same extent by 7 X 10^^ M phenylurethan. The effect of light can-
not be replaced by raised oxygen tension in the medium. However,
some indications of the nature of the active product have been ob-
tained and are discussed below.

Period of Ilhmmiatiojj. Klebs (1896) observed that cells of
Chlamy dojnonas media capable of active mating in light lost their
sexual activity in darkness, under which condition he considered that
they returned to a vegetative phase. Since gametes only reappeared
in such cultures after a week of reillumination, it seemed likely that
"rejuvenation" of the culture by vegetative reproduction was needed
to restore sexual activity. Moewus (1933), on the other hand, noted
that cell suspensions of C eugametos which had been kept in dark-
ness could regain their sexual activity after only 30 to 45 minutes in
light, indicating that the same cells were concerned in the recovery.

Smith (1946) found that as short a period as 5 minutes of illu-
mination markedly stimulated sexual activity in C. mimitissima. In
his later investigations with C reinbardi, which can be grown ab
initio in the complete absence of light, he found that sexual activity
began to be demonstrable after 90 minutes of illumination in bright
light. Using the same species, Wendlandt (see Smith, 1951b) showed
that, at lower light intensities, longer periods of illumination were
required before the first appearance of sexuality; whereas, for a con-
stant period of illumination, lower light intensities induced a less in-
tense sexual response (as measured by the number of clumps of
mating cells per unit volume). Once illuminated, cultures of this
species retained their sexual activity for at least a week in darkness
(Smith, 1948).

Experiments with C. jnoewusii (Lewin, unpublished) have
shown that sexually active cell suspensions reversibly lose their activ-
ity in darkness and regain it in light, these changes taking place in
a matter of minutes, so that considerations of cell division do not
arise. The species is peculiarly well suited to studies of this mechan-
ism, since cell pairs, once united, remain active without further fusion
for some hours (see page 121), and their numbers provide a con-
venient direct assay of mating activity. By using this fact it has been
shown that cells, sexually inactivated by a period of darkness, may
completely regain activity after 10 to 30 minutes of illumination in

SFX IN UNICM.l.UI.AR AKGAF, HI

white liglu. The reactivation curve does not commence to rise at
the origin, but a minimum of V2 to '^^ minutes is necessary before
any restoration of sexual activity can be demonstrated. The length of
the lag period, the slope of the curve, and the height of the asymp-
tote are affected by a variety of factors including intensity and qual-
ity of light, temperature, and the length of the preceding period of
darkness to which the cells have been subjected.

Medium

Attempts have been made by many workers, using different or-
ganisms, to evolve media which would specifically elicit or promote
sexual activity. \^arious degrees of success have been achieved, and it
is not proposed to review the literature on the subject here. In gen-
eral, it has been found that dilution of the medium, or transfer of the
cells from agar to liquid culture, has a marked effect in promoting
gamete formation in a number of diverse algae. Lerche (1937) con-
cluded that in Dimaliella the effect was attributable to depletion of
nutrients (cf. Klebs, 1896; Schreiber, 1925; Strehlow, 1928-29), since
she found sexual activity to increase in aging cultures. In Cylindro-
cystis, an apparently homothallic desmid, copulation w^as found to
be regularly induced by the depletion of nitrogen in the medium
(Pringsheim, 1919). This does not appear to be the case, however,
in other algae; for example, Chlauiydoinojias pseudoparadoxa (Moe-
wus, 1933), Chlamydobotrys (Behlau, 1935), Tolytoma (Pringsheim
and Ondratschek, 1939). In Chlmnydomonas s^)^. (Klebs, 1896; Moe-
W'us, 1933; Lewin, 1949a), Brachiomonas (Moewus, 1944), Protosi-
phon (Bold, 1933; iMaher, 1946-47) and other genera, the empirical
device of flooding agar cultures with water or dilute mineral media
has generally been found to be the simplest and most effective method
for obtaining suspensions of sexually active cells.

Schulze (1927) found that gametogenesis in Chlorogoniinn
could be induced in 16 to 20 hours at pH values of 7.9 to 10.4. Maher
(1946-47) obtained good gamete formation in Protosipbon between
pH 4.0 and 9.0 but noted that at the lower values gamete motility
was reduced. In Chlamydonwnas inoemusii gamete formation is more
active in media of higher pH; pairing takes plac(^ within a pH range
of 6.0 to 10.0, being most active between 7.0 and 9.0. In Polytoma,
too, pairing was found to be most active at pH 8.0 (Pringsheim and

112 SEX IN MICROORGANISMS

Ondratschek, 1939). According to Moewus (1935c, 1950a), the sex
ratios among gametes formed by certain homothallic Protosipho?i
strains and by C. eiigametos alpina are markedly affected by pH.

Warming for short periods tends to stimulate gametogenesis in
Protosiphon (Klebs, 1896; Maher, 1946-47). It was found that aera-
tion promotes sexual activity in the heterotrophic flagellate Folytovia
(Moewus, 1933; Pringsheim and Ondratschek, 1939), but its action
in illuminated suspensions of photosynthetic algae is probably less
marked.

It has been found (Lewin, unpublished) that sexual clumping
and pairing in Chlamy dojnonas moewusii are dependent on the con-
centration of the Ca-' ion in the medium. Sexually active cells lose
their ability to mate after washing in distilled water but regain it
when CaCl2 is added. The optimum range of Ca • • for mating lies
around 3 to 30 ppm, being higher in media containing much phos-
phate or amino acids, which combine with a proportion of the free
cations. Mating is also inhibited when the Ca • • is removed by such
agents as citrate or oxalate. Inhibition by 0,02 M. citrate may be re-
versed by washing the cells in mineral media, or by the addition of
0,001 M CaCl2, and may be slightly alleviated by higher concentra-
tions of MgCls or SrCh, Since, once the cells have paired, citrate
does not reverse the mating reaction, it appears that the Ca • • is con-
cerned primarily in the specific agglutination which occurs between
the flagella of mating cells (see page 117), Loeb (1915) and Vasseur
(1949) showed that the presence of Ca- • was required for the agglu-
tination of spermatozoa and for fertilization in certain echinoderms.
Possibly this requirement is related to the free superficial amino
groups, which, as indicated by Metz and Donovan (1951), take an
active part in this process. The phenomenon may also be in some
ways compared with that of the adsorption of bacteriophage by host
bacteria (Puck and co-workers, 1950).

"Genetyllin"

To provide a material basis for discussion of the experiments
described in this section (Lewin, unpublished), it has been postulated
that in Chhmiydovionas moeuousii there is a specific agent or hormone
responsible for sexual activity. We shall refer to this hypothetical
agent as genetyllin. Genetyllin is formed within the cells under the

SFX IN UNICELLULAR ALGAE 1 1 3

action of light, the energy probably being absorbed by chlorophyll
(see page 109); and as a result sonic stimulus passes into the flagella,
rendering them temporarily susceptible to intersexual agglutination.
"When cells arc kept in darkness, they rapidly lose their ability to
mate, and we may attribute this to disappearance or decomposition
of genetyllin. This reaction follows a course exhibiting two phases,
both affected by temperature. In the first 0 to 1 5 minutes after dark-
ening there is no loss of sexual activity; the duration of this phase is
controlled largely by the period of pre-illumination. There then fol-
lows the second phase, during which the activity falls off exponen-
tially to zero; its duration ranges from 15 to 60 minutes or more,
depending on temperature. If we further postulate that this exponen-
tial loss of activity results from a unimolecular breakdown of gene-
t\llin, it can be deduced from such data that at 30° its half-life is
less than 1.5 minutes; at 10° it is 9 minutes; and at 5° it is about 22
minutes.

A problem immediately presents itself, namely, can this hypo-
thetical agent be replaced by any compound supplied in the medium?
A number of known substances (glutathione, cysteine, acetate, suc-
cinate, glucose, hexose diphosphate, phosphoglycerate, and adenosine
triphosphate) were subjected to a reasonably sensitive test (the dis-
placement of the genetyllin breakdown curve in darkness) without
any positive result, perhaps as a consequence of permeability barriers.
Unsuccessful attempts were also made by Provasoh (personal com-
munication) and by the present author to demonstrate the presence
of a solute, in filtrates from sexually active cells, capable of evoking
a sexual response in non-sexual cell suspensions. Smith (1946) re-
ported a similar lack of success in experiments conducted with other
species of Chlmnydomonas, notwithstanding the positive results
claimed for C. eugametos by Moewus (193 3) (see below).

It thus seemed possible that, at least in C. TiweToiisii, lack of suc-
cess in attempts to demonstrate the activity of solutes in the medium
might be attributable to the non-diffusible nature of genetyllin, and
this possibility was tested indirectly in the following way. It had
been found that cells subjected to more than 30 minutes of illumina-
tion retain full sexual activity for some minutes after transfer to dark-
ness. The longer the period of pre-illumination, the longer the period
of darkness during which full activity is retained (see page 110).
This suggested that, though genetyllin disappears at a constant rate

1 14 SEX IN MICROORGANISMS

in all cases, the cells may have built up, during an extended period of
illumination, a supply of the agent exceeding the threshold for 100
per cent gamete mating; and that it is only when the level of gene-
tyllin falls below this threshold that mating activity declines. Sus-
pensions of cells containing various amounts of reserve genetyllin
were therefore prepared (different pre-illumination periods being
used) and were mixed with suspensions of sexually inactive cells. In
no case was there evidence that activity could be transmitted to in-
active cells by this means; and one may conclude that genetyllin can
probably not be transmitted from cell to cell through the medium.

Genetyllin may prove to be as mythical as the Genetylhdes
themselves. The concept has been introduced, with some misgivings,
in an attempt to indicate a less orthodox approach to the subject of
sex substances than is provided by a search for soluble hormones.

Sex Substances

A few authors have claimed to have demonstrated the presence
of "sex substances" in culture media from sexually active cells. A
clear distinction should be drawn here between at least four different
actions attributed to such agents.

1. Positive chemotaxis, drawing gametes of one mating type or
sex toward their prospective partners (see page 116),

2. In heterothallic species, the induction of clumping in a clonal
suspension by treatment with filtrate from cells of opposite mating
type.

3. The sexual activation of non-sexual cells by treatment with
filtrate from sexually active gametes of the same mating type.

4. The induction of the formation of gametes which behave tem-
porarily as a single mating type by treating cells of a homothallic
strain ("phenotypically dioecious") with filtrate from the appropri-
ate mating type of a heterothallic species.

Sex substances of type (2) were reported in the supernatant
medium after centrifuging gamete suspensions of Tetraspora (Geit-
ler, 1931 ). In T. lubrica, which is regularly heterothallic and appro-
priately lubricous, Geitler found that a certain amount of intraclonal
clumping could be induced in cultures by the addition of concen-
trated "extracts," practically free from cells, prepared from suspen-

SFX IN UNICFLI.Ul.AR Al.CAF. 115

sions of opposite niaring r\pc. Unlike the noniuil scxiuil reaction,
such clumping was not followed l)\' the formation of gamete pairs.
Though he considered the possibility that this effect might be due
to cell breakdown products, he decided that this was unlikely. In
1933 and 1934, Aloewus reported a similar phenomenon in contami-
nated cultures of Cblaviydovwiicts engLinictos and in Protosiphov and
Stcpbiinospbiie)\i. He found that the acti\e agent did not diffuse into
cotton or agar, and evidence for its insoluble nature lay in the fact
that activity could be removed by filtration through membranes with
0.01 [}- mean pore size, though not by coarser membranes with 0.7 5 -[i-
pores. When such intraclonal mating responses were induced, Moe-
wus observed that the cells in each clump appeared to have aggre-
gated around a central point, which he suspected might be a bac-
terium, though he \\as unable to confirm this. It may be noted that
Pringsheim and Ondratschek (1939) were able to induce aggregation
of various flagellates by the addition of bacteria alone. (Moew^us also
reported that solutions containing active sex substance did not freeze
when cooled to —10°, but that, when the agent was removed by
filtration, the filtrate froze normally — a phenomenon inexplicable on
simple physicochemical grounds.) Lerche (1937) was able to demon-
strate sex substance activity in centrifugates from Dunaliella; but in
this case she found that filtration through a membrane with 0.75-!x
pores M'ould remove the active agent, again suggesting that large par-
ticles, rather than a solute, were responsible for the effect.

Jollos (1926) reported that filtrates from "strong" phis gametes
of Dasycladiis tended to augment the sexual response of "weak"
plus gametes, and to reverse that of weak vnmis gametes {mutatis
mutandis, strong mimis gametes behaved similarly) ; but he did not
exclude the possibility that such changes could be attributed to
non-specific alterations in the composition of the medium.

On the other hand. Smith (1946) and Hutner and Provasoli
(1951) were unable to demonstrate the presence of any soluble sex
substances in three species of Chlaiuydomonas and in C i/ioeuiisii
respectively. E. Hinreiner (Al. Starr, in litt., July and September
1952) found no flavones in culture filtrates from C. moewusii or C.
eugametos, despite the activity claimed for certain compounds of
this sort in C. eugametos (Aloewus, I950a,b). Attempts to demon-
strate the activity of specific water-soluble sex substances in Farame-

115 SEX IN MICROORGANISMS

cium, in which clumping is Hkewise associated with agglutination of
the ciha, have been equally unsuccessful (Metz, 1948).

In 1927 Schulze reported some type (3) activity in Chi or o-
gonium filtrates, possibly attributable to their depleted nutrients.
Apart from this uncertain case, sex substances of types (3) (gam-
ones*) and (4) (termones) have apparently been reported in unicel-
lular algae only by Moewus, whose published results contain a num-
ber of discrepant features. They have been reviewed extensively by
Thimann (1940), Chodat and de Siebenthal (1941), Murneek (1941),
Sonneborn (1941, 1942), Hartmann (1943), Lang (1944), Cook
(1945), Smith (1946, 1951b), Raper (1952), and by Moewus himself
(1941, 1950a,b). For various reasons it is considered unprofitable to
discuss them further here.

THE COURSE OF THE MATING PROCESS

Chemotaxis

Except in such aplanogametic algae as diatoms and desmids, and
rare exceptions like the strange Chlaiuydovwims sp. described by
Pascher (1918a), fertilization is accomplished in unicellular algae as
a result of the motility of flagellated gametes of one or both sexes.

In oogamous organisms, a common device promoting fertilization
is chemotaxis. The immobile eggs secrete soluble substances which
diffuse into the surrounding medium, and, by swimming up the dif-
fusion gradients, the antherozoids tend to approach the female
gametes. The sperm attrahent in Sphaeroplea eggs can be adsorbed
on cotton fibers (Pascher, 1931-32), but its chemical nature has re-
mained unknown. The discovery that the chemotactic agent in certain
fucoids is volatile and may be akin to hexane (Cook and Elvidge,
1951) is one of the most remarkable of recent contributions to this
field.

According to Pascher (1931-32), chemotaxis may also be in-
volved in the mating process of Chlamy domonas pauper a, which is
homothallic and isogamous. In this species gametes which come to rest
early act as females, and it is only after their swimming has been
arrested that other gametes, acting as males, bes^in to cluster around

* Moewus (1941) refers to the agents responsible for type (2) reactions
as ganiones.

SFX IN UNICFI.LULAR ALGAE 117

them; Pringshcini and Ondratschck (1939) consider rlieir arrest of
motility an important corollary of chemotaxis.

In species where both gametes remain actively motile, however,
a chemotactic mechanism Mould seem to be virtually useless. None
has been observed in Cbhrinydovionas vioewiisiiy for example (Ilutner
and Provasoh, 1951). In C. eugametos Moewus (1933) found no
evidence for chemotaxis when gametes of one mating type were
allo\\ed to swim near the mouth of a capillary tube filled with filtrate
from cells of the opposite mating type, or near cotton or agar soaked
in such a filtrate. Some years later (1939b) the same author reported
in this species positive chemotactic responses to gentiobiose and iden-
tified the naturally occurring tactic agents with the gamones, viz.,
proportional mixtures of cis- and trans-d^imtxhyX crocetin. His earlier
failure to demonstrate such agents he explained (1941) by the
fact that he had not at first appreciated the photolability of the sub-
stances concerned.

Clumping

When active gamete suspensions of C. inoenjoiisii are mixed, the
cells seem to show no appreciable change in behavior until, by ran-
dom contact, flagella attached to cells of complementary mating type
make contact and adhere (Gerloff, 1940; Lewin, 1952b). As in C.
eugauietos (Moewus, 1933), large clumps of 100 or more cells may
be quickly built up in dense suspensions, whereas in sparser cultures
the cells associate in smaller groups of as few as two or three cells.
Such clumps were first described by Berthold (1881) in gametes of
Ectocarpiis, and have been since observed in a wide variety of other
algae with motile gametes. Non-swimming (ultraviolet induced)
mutants of C. moenjousii (Lewin, 1952a) mate comparatively poorly
with normal swimming cells, as might be expected. In paralyzed
stocks of the homothallic C. dysosmos, or in mixtures of C. moeuousii
in which cells of both mating types are unable to swim, virtually no
mating takes place unless the suspensions are mechanically agitated to
bring compatible pairs into contact.

In Tetraspora (Geitler, 1931) and C. inoeixnisii (Hutner and
Provasoh, 1951), clumping has been observed between cells killed
by gentle heating, ultraviolet irradiation or other treatments, and
living cells of the complementary mating type.

118

SEX IN MICROORGANISMS

Plate I. Stages in copulation of ChlmnydoDWiuis 7/ioewusii Gerloff.
(Electron micrographs: Pd-shadowed, 7:1)

1, Agglutination of flagella (X 1500); 2, formation of protoplasmic! bridge and
dissociation of flagella (X2500); 3, protoplasmic bridge (X4000); 4, vis-a-vis pair
(X 3000).

The electron microscope was made available by kind permission of Dr. J. L.
Melnick, of the Section of Preventive Medicine, Yale University; and much valuable
assistance was received in its operation from Airs. J. C). Aleinliart and Dr. W.
Gaylord.

SEX IN UNICELLULAR ALGAE 119

Pairing

In the formation of mating clumps and in the subsequent pairing
of d^amctcs, the flaijella of C nioeiviisii play an important role (Lewin,
1952b). Mutant strains in which the flagella are lacking (Lewin,
1952a) are completely asexual. It is probable that these organelles are
equally essential in the sexual processes of other flagellated algae, as
are the cilia of PcViT/f/eciir/n and other ciliates (Jennings, 1939). By
their aid the mating cells are brought together and held until they pair
and cytogamy can be initiated. As soon as pairing has been accom-
plished, the agglutination of the flagella ceases, perhaps as a result of
a stimulus transmitted through the papillae and basal granules, and the
flagella extend free in the medium once more. The whole process
occupies less than 10 minutes (at 23°); typical stages are illustrated in
Plate I.

In most unicellular algae, perhaps all, the process of fusion com-
mences between the anterior ends of the mating cells. In those in
which the gametes are not naked cells, the wall appears to dissolve
first in this region, and a minute process may emerge, as in Folytovm
(Pringsheim and Ondratschek, 1939), Chlamydovwjias proboscigera
(Korschikoff, in Pascher, 1927), and C. eugametos (Moewus, 1933).
A thin connecting strand of material, probably protoplasm, has been
observed in ChJaviydovwnas and FhyUocardhnn (Korschikoff, 1927),
C. praecox (Pascher, 1943), C. chlaviydogavia (Bold, 1949), C. nioe-
wusii (Lewin, 1950a; see Plate I, 2-3), and others, connecting the
two cells; and ultimately this widens until the protoplasts fuse.

Although there are several records in the literature of gametes
which have begun to fuse at their posterior ends — for example, in
Frotosiphon (Klebs, 1896), Dimaliella (Cavara, 1906), Tetraspora
(Geitler, 1931), and in Carteria and Chlamydomoiias spp. (Mitra,
1950) — there is good reason to believe that in most, if not all, cases the
observed cell pairs were not indulging in syngamy but had resulted
from incomplete cell division, as pointed out by Lerche (1937). Such
twinned cells have been described or figured in Cbkviydovwints
brawni (Goroschankin, 1890), C. stellata (Gerloff, 1940), C. varia-
bilis (Behlau, 1939), C. iipsalieiisis (Skuja, 1949), C. moeivusii
(Lewin, 1952a), Haemato coccus (Reichenow, 1909), Chlorococcmn
(Bold, 1930), Frotosiphon (Nayal, 1933), and so on.

120 SEX IN MICROORGANISAIS

Planozygotes

Once pairing has been initiated, the course of events leading to
zygote formation differs according to species. In some the cells come
to rest almost at once, whereas in others there is a further period of
motility, possibly with the biological advantage of locating the zygo-
spore in a suitable resting place. In most cases this motile stage occurs
after cytogamy and perhaps also caryogamy, so that we are justified
in referring to it as a planozygote. The axes of the cells usually swing
round until the gametes are laterally appressed, and then not only do
the cells fuse, but also the two pairs of flagella become coordinated
into a single integrated motor apparatus, so that a quadriflagcllatc
planozygote is formed, resembling a cell of Carteria. In fact, Behlau
(1939) has shown that planozygotes of Chlamy domonas variabilis
were apparently described by Jacobsen as ''''Carteria ovata^'' and, fol-
lowing Korschikoff (1926), Behlau considers that ''''Chlorobotrys
gracilliina^' and ''''Tetradonta variabilis^'' are likewise diplophase cells
of Chla77iydobotrys gracilis and Chloromonas paradoxa respectively.

According to Teodoresco (1905) the planozygotes of Dwmliella
are biflagellate, one pair of flagella being shed after copulation. In
Carteria iyengarii, the planozygote, formed by the fusion of a naked,
non-motile egg cell with a quadriflagellate microgamete, is propelled
from some time by the flagella of the male cell (Ramanathan, 1942)
like that recorded in the phycomycete Monoblepbarella (Sparrow,
1939). Such planozygotes may remain actively swimming for some
hours in Protosiphon (Klebs, 1896), Fhyllocardiiim (Korschikoff,
1927), and Tetraspora (Klyver, 1929), or, in dim light, for as long as
1 to 3 weeks in Chlamy domonas paradoxa and Chlamy dobotrys stellata
(Gerloff, 1940; Strehlow, 1928-29) and Phylloinonas striata (Kor-
schikoff, 1926). In Raciborskiella, according to Wislouch (1924-25),
the planozygotes are even capable of diploid mitosis!

Vis-A-vis Pairs

An unusual exception to the general pattern of planozygote
formation is furnished by Chlamy domonas eiigajnetos (Moewus,
1933; Mitra, 1951) and C. moewnsii (Gerloff, 1940). In these forms,
after gamete pairing has been initiated, cytogamy is arrested for a
period, the cell pairs continuing to swim actively for some hours

SEX IN UNICELLULAR ALGAE 1 2 1

(sec Plate I, 4). (iMocwus and Gerloff erroneously refer to this stage
as a planozygote.) Since the cells in each pair are attached by their
anterior ends, along a common axis, the two pairs of flagella would
operate in conflict if both remained active; but here, at least in C.
Diocii-usli, another coordinating mechanism comes into play. One
of the cells, though morphologically indistinguishable from its partner,
ceases to beat its flagella, and the propulsion of the pair is thus left
entirely in the hands of its mate. Such pairs regularly progress in one
direction, always propelled by the same partner; they do not move
forward and backward, as described for C. eiigametos indica by Mitra
(1951). It has been shown (Lewin, 1950a) that this difl^erence in
behavior is genetically inherent in the two mating types and can be
considered to be controlled by a gene inseparably linked with the
"sex" locus. A homothallic species, recently isolated, exhibits similar
behavior of the vis-a-vis pairs, presenting a special problem of intra-
clonal diflterentiation which is now being studied. Like the plano-
zygotes of Frotosiphon (Bold, 1933; Maher, 1946-47), of C. variabilis
(Behlau, 1939), and of several other Chlamydovwnas spp. (Smith,
1950b), the gamete pairs of C. moenjonsii resemble free haploid cells
in being positively phototactic, though those of C. eugametos (Aloe-
wus, 1940c) are said to show a reversal of taxis following copulation.
Recent evidence indicates that, in C. uioenimsii, there may be
exchange of intracellular metabolites through the protoplasmic bridge
joining the two copulants, so that one might be justified in referring
to such pairs as a special case of heterocaryosis.

Cytogamy

In darkness, the gametes of C. moeimisii remain in vis-a-vis pairs
indefinitely, until death overtakes them. In dim light, a day or more
may pass before fusion takes place (cf. C. eugavietos; Moewus, 193 3);
while in bright light (300 to 500 foot-candles) cytogamy is con-
summated 4 to 8 hours after pairing, and at least 10 hours of addi-
tional illumination are required if a viable zygote is to result (Lewin,
1949b). In this respect C. vweii:usii differs from some organisms, in
which light energy has been shown not to be essential during the
process of fusion (for example, Frotosiphon; Maher, 1946-47), and
from others (for example, Naviciila halophila; Subrahmanyan, 1946)
in which light has been found actually to inhibit the process. There

122 SEX IN MICROORGANISMS

is some evidence that in Chlaiuydoinonas the wavelengths of light
concerned are those absorbed by the chloroplast, so that the energy
for cell fusion may be provided indirectly by photosynthesis.
The flagella are not normally shed (as stated by Gerloff, 1940: cf.
C. tipsaliensis; Skuja, 1949) but can be seen to be withdrawn or re-
sorbed, as in Tetraspora (Klyver, 1929), in Protosiphon (Bold, 1933),
and in C. engametos indica (Mitra, 1951) and C. chlaviydogama
(Bold, 1949), in the course of about 30 minutes. Half an hour later
the cells finally fuse, the actual fusion taking place quite rapidly in
this as in other species of Chlamydomonas.

The fate of the chloroplasts is usually unknown, though in Phyl-
lovwTias (Korschikoff, 1926) they are said to fuse. A study of the
genetics of plastogene mutants in such algae would undoubtedly
provide valuable information on this point.

In Chlorogoiihnn oogavnnn (Pascher, 1931) and Carteria iyejj-
garii (Ramanathan, 1942) the egg cells shed their walls before fertili-
zation. In those species where the gametic cells are furnished with a
cell wall, the statement has occasionally been made that the walls
themselves coalesce in cytogamy (for example, Chlamydomonas
brainiii, Goroschankin, 1890; C. eiigametos, Moewus, 1933; C. en-
gametos indica, Mitra, 1951; C. ///oeivusii, Gerloff, 1940; C. upsalien-
sis, Skuja, 1949); but it seems most likely that in all species with
walled gametes — as described in C. media (Klebs, 1896), C.
elongata,. (Korschikoff, 1923), C. panpera (Pascher, 1931-32), C. g}'?;/-
7iogy7je (Pascher, 1943), C. chlamydogama (Bold, 1949), and C. moe-
iV7isii (Lewin, 1952c) — the walls are actually shed before (Smith,
1950b) or during cytogamy. In C. moenxmsii and other species of
ChlafJiydomonas the fusing gamete protoplasts become invested in a
primary zygote membrane of characteristic staining properties
(Lewin, 1952c), and the formation of such a membrane appears to
be a common phenomenon among related algae.

The DiPLom Zygospore

Fusion of the nuclei is usually difficult to follow: in Chlamydo-
botrys and Protosiphon caryogamy may take place about 4 hours
after copulation (Strehlow, 1928-29; Bold, 1933). Meanwhile, within
the primary membrane, the zygote secretes a thick, highly imper-
meable wall, often characteristically patterned in relief (in Brachio-

SEX IN UNICELLULAR ALGAE 123

nioihis this wall in;iy appear niaiivc), and a mmihcr of other nictaholic
changes become apparent in the young zygospore, hi most species
starch is first accumulated and later replaced by oil, the pyrenoids
frajrment or become invisible, and in many cases the cells accumulate
haematochrome and become bright orange or red. In others they
remain green, as, for example, in Fhyllocardhnn (Korschikoff, 1927),
Chhiinydovwuas viediii (Klcbs, 1896), C.viovoica, (Strehlow, 1928-
29), Chlorococcwn (Starr, 1949), C. engametos (Moewus, 1933) and
C. vweivusii (Lewin, 1949b). Although in the haplophase Polytonia
ftisifonnis is competely apochlorotic, the zygospores have been de-
scribed as chlorophyllose (Korschikoff, 1926; Strehlow, 1928-29),
a unique feature that would present a number of interesting problems
to the experimental phycologist.

Smith (1950b) has pointed out that one may distinguish two types
of zygotes, those in which no growth takes place after cytogamy,
and those wherein photosynthesis and growth continue in the diplo-
phase, in the planozygote (for example, Phyllocardiimt, Korschikoff,
1927) or in the zygospore stage (for example, Tetraspora, Klyver,
1929, Chhmiydovwnas chlamydogama, Bold, 1949, and Chlorococ-
ciivi, Starr, 1949). The volume of each gamete in C. moeiviisii is about
200 cubic microns, while that of the mature zygospore approaches
12,000 cubic microns, indicating a thirty-fold increase in bulk in
approximately 60 hours. This rate of growth is comparable with that
of the haplophase, which is capable of five generations (doublings) in
a similar period (Lewin, 1952a) . Though there is no vegetative division
in the diplophase (except perhaps in Raciborskiella, Wislouch, 1924-
25), and we can hardly speak of an alternation of sexual generations
(cf. Behlau, 1939), the line dividing Chlmnydovwiias from such an
alga as Chlorochytrium (Chlorococcales) is thin. According to Kurs-
sanow and Schemakhanowa (1927), the vegetative cell of Chlorochy-
trium levmae is typically diploid and, like the zygote of Chiaviy-
domo7ias vweujiisii, is capable of enlargement but not of direct mitotic
division. By meiosis and succeeding mitoses, 256 swarmers are pro-
duced which are capable of copulation to form quadriflagellate plano-
zygotes; and these in turn settle down to an endophytic existence as
a new diplophase. Thus the sexual cycles of Chlaviydoniouas and of
Chlorochytrium are seen to differ in no essential respect, but merely
in their relative emphasis on the haploid or diploid phase.

124 SEX IN MICROORGANISMS

Zygospore Germination

The red, stellate zygospores of Frotosiphon can be induced to
germinate only after a period of some weeks of dormancy (Klebs,
1896). According to Nayal (1933), young green zygospores of one
strain of Frotosiphon, when formed under moist conditions, may
germinate soon after being transferred to fresh media, while older
red spores, especially if they have been subjected to heat or desicca-
tion, remain dormant for some days before being capable of germina-
tion. The zygospores of Folytovia germinate readily under favorable
conditions (Pringsheim and Ondratschek, 1939). However, in a
depressingly large number of studies on the sexual cycles of unicellu-
lar algae, the obstacle of zygospore germination has proved insur-
mountable. The walls of such spores are usually relatively thick, and
in some species are said to consist of three layers of different composi-
tion (Gerloff, 1940). Though actually only 1 to 2 [j. thick, they may
achieve a remarkable degree of impermeability to water and chemical
agents. Thus zygospores of Chlamydoiiioims vweivmii have been
known to survive after several days of immersion in absolute acetone
or after 4 years in air, though the zygotes of the related C. eiigametos
and those of C. viedia are killed by drying (Moewus, 1933, 1950c;
Klebs, 1896). Zygospores of a strain of Frotosiphon, isolated from
desert silt, were shown to remain viable after heating at 75° to 91°C
for 18 hours, or at 50° to 60° for 15 to 18 days (Nayal, 1933): while
spores of Folytovia were found to be capable of germinating after
contact with solid COn for 10 minutes, or after four weeks' desicca-
tion over CaClo followed by 10 minutes at 75° (Pringsheim and On-
dratschek, 1939). In Chlaviydovwnas engametos, Moewus has re-
ported (1946, 1950c) genetic variation in the resistance of zygospores
to high temperatures and osmotic pressures.

Within such impenetrable walls, nuclear changes are extremely
difficult to follow (Kater, 1929; Zimmermann, 1921) while we know
nothing of the underlying metabolic changes associated with the
dormant state. In our present profound ignorance, we have to resort
to empirical methods for inducing germination, and a number of such
studies have been carried out (see Bold, 1942). In CyVmdrocystis,
Pringsheim (1919) found that germination of zygospores — even
before maturation was complete — could be induced by transferring
them to fresh nutrient media. Lerche (1937) achieved a fair degree of

SRX IN UNICELLULAR ALGAE 125

success w irh /Agosporcs of D/i/mlicllii by subjecting them to a con-
trolled program of maturation, desiccation, and resuspension in media
of augmented salinity, Starr (1949) effected zygospore germination
in Chlorococcwni and Chlamydovioiias chlamydoga-nm by incubating
the spores for 48 hours at 37°. In many species of ChJamydomoiias
Smith (1950b) found that about 20 days were required for the ripen-
ing of the zygospores, which would then germinate regularly 1 to 2
days after being transferred to fresh media. On the other hand, Lewin
(1949b) Mas able to induce only sporadic germination of ripe C.
7noewusu zygospores in a fraction considerably lower than the
10 per cent achieved at room temperatures by Gerloff (1940), but he
found that spores in which full maturation had been arrested by star-
vation in darkness M'oiild germinate regularly. Moewus and Banerjee
(1951) suggested that germination may be naturally arrested by the
accumulation of some inhibitor such as cis-cmnmnic acid, which
can be removed from some zygotes by exhaustive washing in water.

The New Haplophase

In most unicellular algae, meiosis precedes zygote germination,
and the sexual cycle is completed by the liberation of haploid zoo-
spores. In desmids there is a tendency for two or three of the four
post-meiotic nuclei to degenerate, so that only one or two haploid
cells emerge from a zygospore on germination (Pothoff, 1928). It
appears likely that in Protosiphon, too, meiosis precedes zygospore
germination. Possibly only one post-meiotic nucleus survives and
multiplies in the emerging thallus, which would otherwise be poten-
tially heterocaryotic. But the zygotes of this alga are not very favor-
able material for cytological study, and despite several investigations
of its life cycle (for example, Bold, 1933; Maher, 1946-47) this
aspect remains obscure.

In species of Volvocales in w hich the zygote exhibits no grow th
after syngamy (see page 123), four zoospores are produced as a rule
(Smith, 1950b), often within an extruded vesicle; whereas in cases
where the zygotes enlarge in the days immediately following their
formation a larger number may be formed as a result of subsequent
mitoses. In DiinaUella (Lerche, 1937) and in Chlamydomonas moe-
ivusii, 4, 8, 16 (Gerloff, 1940), or 32 zoospores may be formed, the
number depending on the size of the zygote, and hence on the condi-

126 SEX IN MICROORGANISMS

tions of its formation. In C. eugametos, such zoospores are said to be
immediately capable of acting as gametes (Moewus, 1933).

A few attempts have been made to follow meiotic segregation
among the products of zygote germination. Pascher (1916) succeeded
in crossing two 1 0-chromosome species of Chlamy domonas and in
effecting some analysis of genetic characters among the progeny: it
is claimed (Pascher, 1918b) that these early experiments were the
first to establish directly the fact of Mendelian segregation in any
plant.

Clones may be grown from zoospores produced after zygotic
meiosis, and in such cultures sex (or mating type) has been shown to
segregate equally in heterothallic species of Gonium and Eudorina
(Schreiber, 1925), Chlorogonium (Schulze, 1927), Dwialiella
(Lerche, 1937), Chlamy domonas eugametos (Moewus, 1933), C. vari-
abilis (Behlau, 1939), C. reijihardi (Smith and Regnery 1950), and
C. moeivusii (Lewin, 1950b). However, the occasional formation
of homothallic and neuter strains, variously attributed to cytologically
demonstrable non-disjunction of sex chromosomes (Hartmann, 1934),
to crossing-over between sex loci, and to mutations affecting sex
determination and behavior, have been reported in a number of algae
investigated by Moewus (see reviews by Jennings, 1941; Beadle,
1945; and Sonneborn, 1951; also Moewus, 1944). Philip and Haldane
(1939) have criticized some of these results: Moewus' replies to
these criticisms (1941, 1943) are far from satisfactory.

In Dimaliella, Lerche (1937) observed that zygotes would oc-
casionally give rise on germination to only two zoospores, cytokinesis
apparently coinciding with the first meiotic division. By taking ad-
vantage of such exceptional cases, she was able to distinguish first
from second division segregations and thus to estimate the map dis-
tances of two unlinked loci, for mating type and for a morphological
marker from for/na ^'oblonga,^^ from their centromeres. Moewus
(1940b) claimed to have mapped forty-two characters on the ten
chromosomes which he observed in Chlamy domojias eugametos and
its allies, in which crossing-over invariably took place between whole
chromosomes, as he had found also to be the case in Protosiphon,
Tolytoma, and Brachiomonas. However, he illustrated (1936) chias-
mata which appear similar to those of other organisms and stated
(1944) that, in nature, crossing-over occurs in the four-strand stage.
Several serious criticisms of these reports can be put forward, as has

SEX IN UNICELLULAR ALGAE 127

been pointed out by Patau (1941) and liartc (194S). In C. //loeivusii
the genetic behavior of eight ultraviolet-induced mutant characters
and of mating type has been investigated (Lewin, 1953); and evi-
dence for linkage was obtained between t and /, two mutant charac-
ters respectively affecting cell division and flagellar activity, and
between "p-aminobenzoic-less" and the mating-type locus. J. Harts-
horne (personal communication) has studied the genetics of an ultra-
violet-induced mutant "eyeless," and of tw^o suppressed semi-lethals
found in wild stocks of C. reinbardi. In this species, as in C. //loeiviisii
and in virtually all other organisms which have been studied, crossing-
over takes place between chromatids, and two or four different
haploid genotypes can be obtained from a single zygote.

CONCLUSION

Some of the greatest strides in the field of microbiology have
been made when a number of workers, using different approaches
and techniques, have concentrated on a single organism, such as
Neurospora or Escherichia coli. It might not be rash to predict that
similar progress will result from a concerted attack on the control of
sexuality in a suitable selected unicellular alga. Many problems
should be solved in this way, and others will inevitably arise; for the
simplicity of microorganisms is more apparent than real.

REFERENCES

Aristotle. De getieratione ani?nalium {tra?is. 1910 by A. Piatt). Clarendon Press,

Oxford. (See specifically 731-732 and 762-764.)
Beadle, G. W. 1945. Biochemical genetics. Che?/!. Revs., 37, 15-96.
Behlau, J. 1935. Die Spondylomoraceen-Gattung Chlmnydobotrys. Beitr. Biol.

Pfimiz., 23, 125-166.
. 1939. Der Generationswechsel zwischen Chla'/iiydo7/ionas variabilis

Dangeard und Carteria ovata Jacobsen. Beitr. Biol. Pfianz., 26, 221-249.
Berthold, G. 1881. Die geschlechtliche Fortpflanzung der eigentlichen Phaeo-

sporeen. Mitt. Zoo/. Stat. Neapel, 2, 401-412.
Bold, H. C. 1930. Life historv and cell structure of Chlorococcum infiisiomiin.

Bull. Torrey Botan. Club, 57, 577-604.
.•1933. The life history and cytology of Protosipbofi botryoides. Bull.

Torrey Botan. Club, 60, 241-299.'
. 1942. The cultivation of algae. Botan. Rev., 8, 69-138.

128 SEX IN MICROORGANISMS

. 1949. The morphology of Cblaniydovwnas chlmiiydogama sp. nov.

Bull. Torre y Botan. Club, 76, 101-108.
Cadoret, R. 1949. Motihty and sexuality in a South American species of

Chlamydomonas. Unpublished thesis, Harvard University. 40 pp.
Cavara, F. 1906. Alcune osservazione sulla Dunaliella salina (Dun.) Teod.

delle saHne di Cagliari. ReJid. accad. sci. fis. e 7/mt. (Soc. nazl. sci. Napoli),

ser. 3, 12 (Quoted by Lerche).
Chodat, F., and R. de Siebenthal. 1941. La sexualite relative des Chlaniydo-

vionas: expose des experiences de Mm. F. Moewus et R. Kuhn. Bull. Soc.

Botmi. Geneve, ser. 2, 33, 1-37.
Cook, A. H. 1945. Algal pigments and their significance. Biol. Revs. Cambridge

Phil. Soc, 20, 115-132.
Cook, A. H. and J. A. Elvidge. 1951. Fertilization in the Fucaceae: investiga-
tions on the nature of the chemotactic substance produced by eggs of

Fucus serratus and F. vesiculosus. Proc. Roy. Soc. {London), B138, 97-114.
Czurda, V. 1933a. Uber einige Grundbegriffe der Sexualitatstheorie. Bota?i.

Centr. Beih., 50 [1], 196-210.
• . 1933b. Experimentelle Analyse der Kopulationsauslosenden Beding-

ungen bei Mikroorganismen. Botmi. Centr. Beih. 51 [1], 711-762.
Darlington, C. D. 1937. Recent advances i?i Cytology. Churchill, London.
Fritsch, F, E. 1935. The Structure and Reproduction of the Algae. University

Press, Cambridge.
Geitler, L. 1931. Untersuchungen iiber das sexuelle Verhalten von Tetraspora

lubrica. Biol. Xentr., 51, 173-187.
. 1949. Die Auxosporenbildung von Nitzschia sigmoidea und die Ge-

schlechtsbestimmung bei den Diatomeen. Portugaliae Acta Biol., ser. A,

79-87.
GerloflF, J. 1940. Beitrage zur Kenntnis der Variabilitat und Systematik der

Gattung Chlainydoinonas. Arch. Protistenk., 94, 311-502.
Goroschankin, J. 1890. Beitrage zur Kenntnis der Morphologic und Sys-
tematik der Chlamydomonaden. Bull. soc. jiaturalistes Moscou, N. S., 4,

498-520.
Harte, C. 1948. Das Crossing-over bei Chlamydomonas. Biol. 7.entr., 67,

504-510.
Hartmann, M. 1932. Neue Ergebnisse zum Befruchtungs- und Sexualitats-

problem. Naturwissejjschafteji, 20, 567-573.
. 1934. Beitrage zur Sexualitatstheorie: mit besonderer Beriicksichtigung

neuer Ergebnisse von Fr. Moewus. Sitzungsber. preuss. Akad. Wiss.

Phys.-math. KL, 379-400.

1943. Die Sexiialitdt. Fischer, Jena.

Hutner, S. H., and L. Provasoli, 1951. The phytoflagellates. In Lwoff, A. Bio-

che?mstry and Physiology of Protozoa, pp. 27-128. Academic Press, New

York.
Jennings, H. S. 1939. Genetics of Paramecium bnrsaria. I. Mating types and

groups, their interrelations and distribution: mating behaviour and self

sterility. Genetics, 24, 202-233.

SKX IN UNICl.I.LULAR AIXIAK 129

. 1941. Inheritance in proto/oa, in (^nlkiiis, (I. N., and F. M. Sunmers,

Protozoa in Biolosj^icctl Rcscwch, pp. 710-771. Columbia University Press,

New York.
Jollos, V. 1926. Untcrsuchungen iibcr die Sexualitiitsverhiiltnisse von Dasy-

cladiis chrvacfor/nis. Biol. 7.ej/tr., 46, 279-295.
Kater, J. M. 1929. Morphology and division of Cblaiiiydoiiionas, with reference

to the phylogen\' of the flagellate neuromotor system. Uiiiv. Calif.

(Berkeley) Pnbls.'Zool., 33, 125-168.
Klcbs, G. 1896. Die Bediii{^iiiigeii der b'ortpthtiiznng hei einigen Algen wid

Filzen. Fischer, Jena.
Klyver, F. D. 1929. Notes on tlie life history of Tetraspora gelatinosa (Vauch.)

Desr. Arch. Protistenk., 66, 290-296.
Korschikotf, A. A. 1923. Zur Morphologic des geschlechtlichen Processes bei

den \^olvocales. Arch. Russ. ProtistoL, 2, 179-194.

. 1926. Algological Notes II. Arch. Russ. ProtistoL, 5, 137-161.

. 1927. Phyllocardiinn coinplaiiatwin, a new Pol)blepharidacea. Arch.

Protistenk., 58, 441-449.
Kurssanow, L. J., and N. M. Schemakhanowa. 1927. Sur la succession des

phases nucleaires chez les algues vertes. Le cycle de developpement du

Chlorochytriwn Le77mae (Cohn). Arch. Riiss. ProtistoL, 6, 131-146.
Lang, A. 1944. Entwicklungsphysiologie. Fortschr. Botciii., 11, 268-317.
Lerche, AV. 1937. Untcrsuchungen iiber Entvvicklung und Fortpflanzung in der

Gattung Dunaliella. Arch. Protistenk., 88, 236-268.
Lcwin, R. A. 1949a. Genetics of Chlaniydonwnas — paving the way. Biol. Bull.,

97, 243-244.
. 1949b. Germination of zygospores in Chlaniydomonas. Nature, 164,

543-554.

. 1950a. Gamete behaviour in Chlaniydonwnas. Nature, 166, 76.

. 1950b. Genetics of ChUrniydoinonas nioeivusii Gerloff. Proc. Intern.

Botan. Congr. {Stockhohii), pp. 858-860.
. 1951. Isolation of sexual strains of Chlamydomonas. J. Gen, Microbiol.,

5, 926-929.
. 1952a. Ultraviolet induced mutations in Chlamydomonas moeivusii

Gerloff. /. Gen. Microbiol., 6, 233-248.
. 1952b. Studies on the flagella of algae. I. General observations on

Chlamydoinonas moeivtisii GerlofT. Biol. Bull., 102, 74-79.
. 1952c. The primary zygote membrane in Chlainydomonas moeivusii.

J. Gen. Microbiol., 6, 249-250.
. 1953. Genetics of Chlamydomonas moeivusii Gerloff. /. Genetics, 51,

543-560.
Loeb, J. 1915. On the nature of the conditions which determine or prevent

the entrance of the spermatozoon into the e^gg. Am. Naturalist, 49, 257-

285.
Luyet, B. J. 1950. A comparative study of the life cycles of lower protozoa

and protophyta. Biodyna?7iica, 6, 265-364.
Maher, M. L. 1946-7. The role of certain environmental factors in growth

130 SEX IN MICROORGANISMS

and reproduction of Protosiphon botryoides Klebs. Bull. Torrey Botan.

Chib, 73, 573-587; 74, 20-37, 156-179.
Mainx, F. 1933. Die Sexualitdt als problem der Genetik (Versuch eines kriti-

schen Vergleiches der wichtigsten Theorien). Fischer, Jena.
Metz, C. B. 1948. The nature and mode of action of the mating-type sub-
stances in Parainechan aurelia. Am. Naturalist, 82, 85.
Metz, C. B., and J. Donovan. 1951. Specific fertilizin agglutination of dead

sperm. Biol. Bull, 101, 202.
Mitra, A. K. 1950. A peculiar method of sexual reproduction in certain new

members of the Chlamvdomonadaceae. Hydrobiology, 2, 209-216.
. 1951. Certain new members of the Volvocales from Indian soils.

Phytomorphology, 1, 58-64.
Moewus, F. 1931. Neue Chlamydomonaden. Arch. Protistenk., 75, 284-296.
. 1933. Untersuchungen iiber die Sexualitat und Entwicklung von

Chlorophyceae. Arch. Protistenk., 80, 469-526.
. 1934. tJber Subheteroezie bei Cbla?Jiydo7iwiias eugaj/ietos. Arch.

Protistenk., 83, 98-109.
. 1935a. Die Vererbung des Geschlechts verschiedener Rassen von

Protosiphon botryoides. Arch. Protistenk., 86, 1-57.
. 1935b. ijber einige Volvocales aus dem Georgenfelder Moor (Erz-

gebirge). Natnrwiss. Ges. Isis, Dresden (1933-34), 45-51.
. 1935c. iJber den Einfluss ausserer Faktoren auf die Geschlechtsbestim-

mung bei Protosiphon. Biol. T^entr., 55, 293-309.
. 1936. Faktorenaustausch, insbesondere der Realisatoren bei Chlaviydo-

nionas-Kvtuzungtn. Ber. deut. botan. Ges., 54 (45)-(57).
. 1937. Methodik und Nachtrage zu den Kreuzungen zwischen Poly-

t07na-Arten und zwischen Protosiphoii-Kussen. Z. ind. Abst. Vererb., 73,

63-107.
. 1938a. Vererbung des Geschlechts bei Chlamydonionas eugainetos und

verwandten Arten. Biol. 2,entr., 58, 516-536.
. 1938b. Carotinoide als Sexualstoffe von Algen. Jahrb. wiss. Bota7i., 86,

753-783.
— . 1939a. Untersuchungen iiber die relative Sexualitat von Algen. Biol.

Zentr., 59, 40-58.
— . 1939b. tJber die Chemotaxis von Algengameten. Arch. Protistenk.,

92, 485-526.
— . 1939c. Carotinoide als Sexualstoffe von Algen. N aturivissenschaften, 27,

97-104.
— . 1940a. iiber die Sexualitat von Botrydittin granulatwii. Biol. "Lentr.,

60, 484-498.
— . 1940b. Die Analyse von 42 erblichen Eigenschaften der Chlaniydo-

n/onas eugai/ietos-Gruppe. 2. ind. Abst. Vererb., 78, 418-522.
— . 1940c. Carotinoid-Derivate, als geschlechtsbestimmende Stoffe von

Algen. Biol. Zentr., 60, 143-166.
— . 1941. Zur Sexualitat der niederen Organismen. I. Flagellaten und

Algen. Ergeb. Biol., 18, 257-356.

SF.X IN UNICELLULAR ALGAE 131

— . 1943. Statistische Auswertung einiger physiologischer und genetischer

\'"crsuchc an Protosipboii und Cblaf//ydo7//o7ias. Biol. Zentr., 63, 169-

203.
— . 1944. Koppclung und Austausch dcr Rcalisatorcn bci Bracbiovwiias.

Beitr. Biol. Pfhriiz., 27, 297-338.
— . 1946. (Regarding the resistance to temperature variations of zygotes of

Chhviiydomoihis.) FIAT Rcpt. 964, 1-11.
— . 1950a. Scxualitiit und Sexualstoffe bei einem einzelligen Organismus. Z.

Sexualforsch., 1, 1-25.
— . 1950b. Die Bedeutung von Farbstoffen bei den Sexual-processen der

Algen und Blutenpflanzen. Aiigew. Chevi., 62, 496-502.
-. 1950c. iJber ein Blastokolin der C hla?nydo7nonas-Zy goten. Z. Natur-

forsch., 5b, 196.
Moewus, F., and B. Banerjee. 1951. Uber die Wirkung von cw-Zimtsaure und

einigen isomeren Verbindungen auf Chla7/iydo7/ionas-Zygoten. Z. Natnr-

forsch., 6b, 270-273.
Murneek, A. E. 1941. Sexual reproduction and the carotenoid pigments in

plants. A7I/. Naturalist. 75, 614-620.
Naval, A. A. 1933. A desert Protosipho?i, P. botryoides (Kiitz.) Klebs var.

deserti. Ann. Bota7iy, 47, 787-798.
Pascher, A. 1916. tjber die Kreuzung einzelliger, haploider Organismen:

Chla7}iydo7no7ias. Ber. deut. botafi. Ges., 34, 11%-1^2.
. 1918a. Uber amoeboide Gameten, Amoebozygoten und diploide

Plasmodien bei einer Chlamydomonadine. Ber. deut. bota7i. Ges., 36,

352-359.
. 1918b. uber die Beziehung der Reduktionsteilung zur Mendelschen

Spaltung. Ber. deut. bota7i. Ges., 36, 163-168.
. 1927. Volvocales — Phytomonadinae. hi Pascher, A.: Die Susstvasser-

flora. Fischer, Jena.
. 1930. Neue Volvocalen (Polyblepharidinen, Chlamydomonadinen).

Arch. Protiste7ik., 69, 125-126.
. 1931. Uber einen neuen einzelligen und einkernigen Organismus mit

Eibefruchtung. Botaji. Ce7itr. Beib., 48, 466-480.
. 1931-32. Uber Gruppenbildung und "Geschlechtswechsel" bei den

Gameten einer Chlamydomonadine (Cbla7/iydo7n07ms pauper a). Jabrb.

iviss. Botaii., 75, 551-580.

1943. Beitrage zur Morphologic der ungeschlechtlichen und geschlecht-

lichen Vermehrung der Gattung Cblai)iydo7}ioiias. Bota7i. Ce7itr. Beib.,

62a, 197-220.
Patau, I. 1941. Eine statistische Bemerkung zu Moewus' Arbeit "Die analyse

von 42 erblichen Eigenschaften der Cbhrr/iydoinonas euga7//etos-Gruppe.''

Z. iTid. Abst. Vererb., 79, 317-319.
Philip, U., and J. B. S. Haldane. 1939. Relative sexuality in unicellular algae.

Nature, 143, 334.
Pothoff, H. 1928. Zur Phvsiologie und Entwicklungsgeschichte der Conju-

gaten. Ber. deut. bota7i. Ges., 46, 667-673.

132 SEX IN MICROORGANISMS

Pringsheim, E. G. 1919. Die Kultur der Desniidiaceen. Ber. deut. botan. Ges.,

36, 482-485.
Pringsheim, E. G., and K. Ondratschek. 1939. Geschlechtsvorgange bei Poly-

toina. Botan. Centr. Beih., 59A, 117-172.
Puck, T. T., A. Garen, and J. Cline. 1950. The mechanism of virus attachment

to host cells. I. The role of ions in the primary reaction. /. Exptl. Med.,

93, 65-88.
Ramanathan, K. R. 1942. On the oogamous sexual reproduction in a Carteria.

J. hidian Botan. Soc, 21, 129-135.
Raper, J. R. 1952. Chemical regulation of sexual processes in the thallophytes.

Botan. Rev., 18, 447-545.
Reichenow, E. 1909. Untersuchungen an Haeiiiatococcus pluvialis nebst Be-

merkungen iiber andere Flagellaten. Arb. Kaiserl. Gesimdh. 33, 1-45.
Schreiber, E. 1925. Zur Kenntnis der Physiologie und Sexualitat hoherer

Volvocales. 2. Botan., 17, 337-376.
Schulze, B. 1927. Zur Kenntnis einiger Volvocales. Arch. Protistenk., 58,

508-576.
Sinnott, E. W., and L. C. Dunn. 1939. Principles of Genetics. McGraw-Hill,

Nevi^ York,
Skuja, H. 1949. Drei falle von sexueller Reproduktion in der Gattung Chla/ny-

domonas Ehrb. Svensk Botan. Tidskr., 43, 586-607.
Smith, G. M. 1946. The nature of sexuality in Chlainydoiiwnas. Am. J. Botany,

33, 625-630.
. 1947. Methods of obtaining and culturing sexual strains of Cbhviiydo-

monas. (Unpublished.)

. 1948. Sexuality in Cblaviydonwnas. Science, 108, 680-681.

. 1950a. The Fresbtvater Algae of the United States. McGraw-Hill,

New York.
. 1950b. Sexuality, zygote formation and z\'gote germination in Chla/ny-

domonas. Proc. Intern. Botan. Congr. (Stockbohn). (In press.)

. 1951a. Manual of Phycology. Chronica Botanica, Waltham, Mass.

1951b. The sexual substances in algae. In Skoog, F., Plant Growth

Substances, pp. 315-328. University of Wisconsin Press, Madison.

Smith, G. M., and D. C. Regnery. 1950. Inheritance of sexuality in Cblaniydo-
monas Reinbardi. Proc. Natl. Acad. Sci. U. S., 36, 246-248.

Sonneborn, T. M. 1941. Sexuality in unicellular organisms. In Calkins, G. N.,
and F. M. Summers, Protozoa in Biological Research, pp. 666-709. Colum-
bia University Press, N. Y.

. 1942. Sex hormones in unicellular organisms. Cold Spring Harbor

Symposia Quant. Biol., 10, 111-125.
-. 1951. Some current problems of genetics in the light of investigations

on Chlamydomonas and Paramecium. Cold Spring Harbor Syinposia
Quant. Biol, 16, 483-503.
Sparrow, F. K. 1939. Monoblepbaris taylori, a remarkable soil fungus from
Trinidad. Mycologia, 31, 737-738.

SEX IN UNICELLULAR ALGAE 1 3 3

Starr, R. C. 1949. A method of effecting zygospore germination in certain

Chlorophxceae. Proc. Natl. Acad. Scl U. S., 35, 453-456.
Strehlow, K. 1928-29. Uber die Sexualitiit einigcr Volvocales. Z. Botan., 21,

625-692.
Subrahmanvan, R. 1946. On somatic division, reduction division, auxospore

formation and sex differentiation in Navicula halophila (Grun.) CI. /.

Indian Botan. Soc, 25, 239-266.
Teodoresco, E. C. 1905. Organisation et dcvcloppement du Dimaliella, nouveau

genre de Volvocacee-Polvblepharidee. Bota??. Ce7?tr. Beih., 18 [1]: 215-232.
Thimann, K. V. 1940. Sexual substances in the algae. Chronica Botan., 6, 31-32.
\'asseur, E. 1949. Effect of calcium ions on the agglutination in Strongylocen-

trotiis droebachiensis Miill. Arkiv Kemi, 1, 105-116.
Waddington, C. H. 1939. An Introduction to Modcr?! Genetics. Allen and

L nwin, London.
Wislouch, S. AL 1924-25. Beitragc zur Biologic und Entstehung vom Heil-

schlamm der Salinen der Krim. Acta Soc. Botan. Polofz., 2, 99-129.
Zimmermann, W. 1921. Zur Entwicklungsgeschichte und Zytologie von Vol-

vox. Jahrb. iviss. Botan., 60, 256-294.

Sex in Protozoa

A Comparatioe Reu/eiu

D. H. WENRICH, Zoological Laboratory,
University of Pennsylvania, Philadelphia

In planning the symposium, it seemed especially important to in-
clude the highly interesting sex phenomena in the flagellates living
in the intestine of the wood-feeding roach, Cryptocercus pimcudatuSy
being worked out by Cleveland, and the latest developments in our
knowledge of mating-type behavior in Faravieciwn. Since these pa-
pers could cover only a small fraction of the subject of sexual repro-
duction in the Protozoa, it seemed desirable to offer a review of
sexuality in the group as a whole. In preparing the material for pub-
lication, the coverage presented at the symposium has been expanded.
Unfortunately, Dr. Cleveland did not find it possible to provide a
discussion of his results for the volume; hence I have added a con-
densed review of his published papers on this subject to the discussion
of the Mastigophora.

In addition to such general works on Protozoa as those of Kent
(1880-82), Biitschli (1882-89), Minchin (1912), Wenyon (1926),
Calkins (1933), Kudo (1946), Grasse (1952-53), and Hall (1953),
the reviews of Jennings (1920), Belar (1926), Calkins and Summers
(1941), Hartmann (1943), Doflein and Reichenow (1949-53), and
Luyet (1950) are useful as references. For many of the major groups
of Protozoa our knowledge of sexuality is very scanty, but for others
there is an extensive literature. References to the literature in this
review are therefore selective rather than comprehensive. In general,
the sequences of classes and orders as given in Kudo's Protozoology
(1946) will be followed. Kudo divides the Phylum Protozoa into two
subphyla, the Plasmodroma and the Ciliophora. The Subphylum Plas-
modroma is separated into the classes Mastigophora, Sarcodina, and
Sporozoa, which will be discussed in that sequence.

134

SEX IN PROTOZOA 135

CLASS MASTIGOPHORA

Kudo divides the Mastigophora into two subclasses, the Phy-
toniasrigina, or plant flagellates, and the Zoomastigina, or animal
flagellates. The Phytomastigina include the orders Chrysomonadina,
Cryptonionadina, Phytomonadina, Euglenoidina, Chloromonadina,
and Dinoflagellata. The books by Fritsch (1935) and Smith (1950,
1951a), are useful in connection with this group.

Order Cbry soiiionadina

Evidence for sexual reproduction in this group is scanty and
incomplete. Schiller (1925) saw sixteen small cells inside a mother
shell of Calyptrosphaera sphaeroidea (Coccolithophoridae) and
thought they might be gametes, but recorded no cell fusions. The
same author (1926) reported fusion of isogametes produced by
division of the protoplast of Dinobryoii sertularia. The zygotes lost
their flagella and sank to the bottom, but no further development
was observed. Krieger (1930) noted that cysts of Dinobryon cyl'w-
dricwn and of D. divergens were binucleate. This was confirmed by
Geitler (1935), who also observed cyst formation. The nucleus
divided (without reduction) before the cyst walls were developed.
Germination of the cysts was not seen, but Geitler supposed that
the two sister nuclei united before germination, thus accomplishing
autogamy. Schwarz (1932) described gamete formation, syngamy,
postzygotic meiosis, and development of four new cells from the
zygote in Ochrosphaera neapolitana (Fig. A, 1 to 12), Lackey (1938)
occasionally found a globule of protoplasm between two loricae of
Chrysococcus spiralis (13) and less often for C. heinisphaerica, which
he thought might possibly indicate "conjugation." Skuja (1950)
reported isogamy with encystment of the zygote in Dinobryoii
borgei, and Mack (1951) found encysted zygotes with adhering
loricae of the uniting individuals in Chrysolykos planktonicus (15).

While there are thus strong indications of isogamy in this group
of flagellates, full details of the life histories and chromosome condi-
tions are not provided in any of the descriptions except that for
Ochrosphaera 7ieapolitana.

136

SEX IN MICROORGANISMS

Fig. a.

1-12, diagrammatic life cycle of Ocbrosphaera iieopolitana (Coccolithophoridae)
from Schwarz (1932), redrawn. Non-motile stages predominate. 1, zygote; 2, four
cells resulting from meiotic divisions; 3, vegetative cell, derivable from a gamete or
zygote; 4-6, division of vegetative cell; 4a, 6a, escape of "zoospores"; 7-10, develop-
ment of gametocytes and gametes; 11, escape of gametes; 12, fusion of gametes. 13,
apparent fusion of protoplasts of two individuals of Chrysococciis spiralis, from
Lackey (1938), redrawn. 14, vegetative stage; 15, zygote of Cbrysolykos planktoniciis
with adhering loricae, from Mack (1951), redrawn.

SEX IN PROIOZOA 137

Order Cryptovjonadina

No instances of sexual reproduction in this group have been
found.

Order Fby tovioiiadiim

This group inckides the chlaniydonionads and volvocids that are
discussed by Dr. Le^in in this volume. They will not be considered
here except to state that sexual reproduction is extensively represented
in the group.

Order Engleiioidma

Perhaps the most frequently referred to account of syngamy in
the euglenoids is that of Dobell (1908) for the colorless Copromonas
subtilis (Fig. B, 1 to 8.) Cytoplasmic fusion of ordinary individuals
is followed by two meiotic nuclear divisions after each of which one
daughter nucleus degenerates. The haploid pronuclei fuse to form a
diploid nucleus. The zygotes usually encyst but may not before
resuming normal vegetative activity. This life cycle was confirmed
by Berliner (1909) for C. viajor and by Woodcock (1916) for C.
niviinant'nnu, and I have seen such cell fusions in my own cultures in
which C. subtilis had developed. However, the absence of any chro-
mosome determinations or other mitotic details leaves the accounts
incomplete and therefore unsatisfactory.

Haase (1910) described syngamy between amoeboid gametes of
reduced size in Euglena sangiiiiiea, but the descriptions are not con-
vincing, and later students of this species including Alainx (1928)
and Giinther (1928) could not confirm the findings of Haase and
discounted her interpretations. Biecheler (1937) in France reported
fusion of vegetative cells of an undetermined species of Engleiia.

Krichenbauer (1937) found individuals of Phacjis pynnn with
two nuclei of different size, the smaller one being more heavily stained
(Fig. B, 9, 10). He also found stages indicating nuclear fusion (11,
12). In a cell stained with Feulgen, a large nucleus seemed to show
diakinesis (13). Tw^o nuclear divisions (14, 15) were followed by
division of the cell into four daughters. Krichenbauer saw four-part
divisions many times among living specimens of Fhacus caudata. He

138

SEX IN MICROORGANISMS

Fig. B.

1-8, syngamy of Copromonas siibtilis, from Dobell (1908), redrawn. 1, early stage of
cell fusion; 2-5, formation of "reduction" nuclei, further progress of fusion; 6,
nuclear fushion in zygote; 7, 8, encystment accompanied by completion of nuclear
fusion. 9-15, "sexual" stages in Fhaciis pyrmn, from Krichenbauer (1937), redrawn.
9-12, fusion of a larger with a smaller nucleus; 12, enlarged nucleus with two endo-
somes; 13, "zygote" with "diakinesis" stage of nucleus; 14, first "meiotic" division;
15, after second "meiotic" division.

SEX IN PRO 1 (T/OA 1 39

interpreted the evidence as indicating- autogamy followed by a two-
division nieiosis and completed by four-part cytoplasmic divisions.
Although suggestive, the evidence lacks completeness and alternative
interpretations can be made. None of the accounts of syngamy in
the euglcnoids is complete and satisfactory.

Order Cbloromonadina
No instances of syngamy in this group have been found.

Order Dinofiagellata

There is some evidence for syngamy in the dinoflagellates, drawn
mostly from studies on Noctihica (Fig. C, 1-5), certain free-living
peridinians, and a few^ parasitic species.

Cienkow'ski (1871a) described fusion of adult specimens of
Noctihica, but the evidence can readily be interpreted otherwise,
as can also that presented by Ishikawa (1891). Both cases appear
to be examples of fusion following incomplete binary fission. On the
other hand, evidence that minute sw^armers (3 to 5) may undergo
syngamic fusion is somewhat more convincing. Pratje (1921) saw
swarmers adhering in pairs and complete fusion of isolated pairs, but
did not see nuclear fusion. Hofker (1930) observed that swarmers
derived from a single individual did not fuse. When he isolated to-
gether a number of individuals undergoing swarmer formation (Fig.
C, 3), he found many active young specimens of Noctihica the next
morning and these developed into typical adults. He believed that the
young specimens had resulted from fusion of swarmers. Gross (1934)
also found that swarmers from a single parent did not unite, but
those from different parents did, accompanied by nuclear fusion.
However, he did not follow the development of the "zygotes" thus
formed. Thus wc have a highly suggestive series of observations in-
dicating syngamy of sw^armers in Noctihica, w^hich, however, lack
completion.

Apparently Joseph (1879) was the first to report cell fusion in
the Peridiniidae, having seen united pairs of Feridiniinn stygiwii.
Stein ( 1883 ) showed a number of illustrations which were interpreted
as indicating stages of "conjugation" for Gle7wdimwn piitrisciihis
(Stein), Heterocapsa triqueta Stein, and Amphidiminii laciistre Stein,

Fig. C.

1-5, Noctiluca ?niliaris, from Kuhn (1921), after Pratje (1921), redrawn. 1, vegetative
individual; 2, binary fission; 3, multiple fission producing "swarmers"; 4, 5, enlarged
face and edge views of "swarmers." 6-8, Cerathiin hiriuidinella, formation of "zygote,"
from Kuhn (1921), after Zederbauer (1904), redrawn. 9-12, four stages in the fusion
of gametes of Coccidinhmi vjesmli, a parasitic dinoflagcllate, from Chatton and
Biecheler (1936), redrawn.

140

SEX IN PROTOZOA

141

but in each case a rearrangement of sequences seems to indicate cell
division rather than cell fusion. Zederhauer (1904) showed association
of individuals of Ceratiwn himndinella in pairs with the extrusion and
fusion of protoplasts between the "parents" to produce a rounded
"zygospore" which encysted (Fig. C, 6 to 8). Development of the

Fig. D.

1-14, stages in sexual reproduction of Glenodiniwn lubiniensiforme, from Diwald
(1938), redrawn. 1-4, development of gametes; 5, escape of gametes; 6, 7, fusion of
gametes; 8, nuclear fusion; 9, encystment of zygote; 10, excystment of zygote; 11-12,
development of "swarmers"; 13, degeneration of two "swarmers"; 14, excysted

142 SEX IN MICROORGANISMS

"zygospore" was not followed. He called the process "copulation."
Entz (1907) found similar stages of this species in preserved material
from "Balaton-Sees." Hall (1925) interpreted binucleate cysts of this
species as possibly indicating previous cell fusion.

Perhaps the most interesting reports for this group are those of
Diwald (1938) for Glenodinium hibiniensifomie. The life history
includes the following: binary fission, with parent shells separating
along the girdle; formation of two vegetative "swarmers" inside the
parent shell; encystment followed by excystment and binary fission;
sexual reproduction with the following stages: (a) two successive
divisions inside the parent shell forming four gametes (Fig. D, 1 to 4)
which are released from the shell (5); (b) fusion of isogametes to
produce zygotes (6 to 8); (c) encystment of the zygotes (9); (d)
excystment of zygotes after at least 10 days' rest (10); (e) while still
immobile, and surrounded by a jelly-like covering, two successive
divisions to produce "swarmers" (11 to 13); (f) development of
flagella and resumption of vegetative hfe (14). Diwald believed that
meiosis took place during the two divisions following excystment of
the zygotes, but chromosomes counts were not given. He found that
gametes did not form for a certain time after excystment. Further-

TABLE I

Results of Mixing Gametes of 16 Clones of Glenodinium luhiniensiforme

IN All Combinations

"Copulation" indicated by -(-; no union indicated by — . (From Diwald, 1938)

2

4

12

3

5

6

7

8

9

10

11

13

14

15

16

2

_

_

—

+

+

+

+

+

+

+

+

+

+

+

+

4

-

-

-

+

+

+

+

+

+

+

+

+

+

+

+

12

—

-

—

+

+

+

+

+

+

+

+

+

+

+

+

3

+

+

+

—

—

—

—

—

—

—

—

—

—

—

—

5

+

+

+

-

—

-

-

—

—

—

—

—

—

—

—

6

+

+

+

—

-

—

—

—

—

—

—

—

-

-

—

7

+

+

+

-

—

—

—

—

—

—

—

—

—

—

—

8

+

+

+

—

—

—

—

—

—

—

—

—

-

-

—

9

+

+

+

-

—

—

—

—

—

—

—

—

—

—

-

10

+

+

+

—

—

—

—

—

-

—

—

—

-

—

—

11

+

+

+

-

—

—

—

—

—

—

—

-

—

-

-

13

+

+

+

—

—

—

—

—

—

—

—

—

—

—

-

14

+

+

+

-

-

—

—

—

—

—

—

-

—

—

—

15

+

+

+

—

—

—

—

—

—

—

—

—

—

—

—

16

+

+

+

—

—

—

—

—

—

—

—

—

—

—

—

SFX IN PROTOZOA I43

more, plus and minus clones could be established. Filtrates from plus
clones induced gamete formation in minus clones and vice versa.
Ciamctes from any one clone did not fuse with each other. When
mixtures of plus and minus clones were made, gametes from any
plus clone would unite with those of any minus clone and vice versa
(Table I). This demonstration of the formation of chemical attrac-
tants is a matter of great interest; it recalls the claims of iMoewus and
others for chemical attractants in Chlmuydovionas (see paper by
Lewin).

Duboscq and Collin (1910) described phenomena interpreted
as sexual reproduction in parasitic dinoflagellates living in some marine
tintinnoid ciliates. In his monograph on parasitic dinoflagellates, Chat-
ton (1920) called this parasite Duboscquella tintinnicola and stated
that "gametocytes" were produced from larger, vegetative stages by
successive divisions. Two divisions, probably meiotic, then produced
flagellated gametes which had something of the appearance of Oxyrrhis
viariiia. These fused in pairs, producing zygotes which rounded up
and lost their flagella. Further development was not followed. Later,
Chatton (1927) described meiotic gametogenesis in Paradiniimi
poiicheti, a parasite of the body cavity of larger copepods of the
genus Acostta. The flagellispores which developed were considered
to be haploid gametes, but fertilization was not observed. Chatton
and Biecheler (1936) described fusion of flagellispores (Fig. C, 9
to 12) formed by the parasitic dinoflagellate Coccidimum mesnili
which parasitizes another dinoflagellate, Crypotperidijmim. Thus
there seems to be very strong evidence for syngamy in this group.

Order RhizoiJiastigina

Kudo recognizes four orders in the Zoomastigina: Rhizomas-
tigina, Protomonadina, Polymastigina, and Hypermastigina.

In the Rhizomastigina we have the accounts of sexual reproduc-
tion in Mastigella vitrea and Mastigina setosa by Goldschmidt (1907).
"Chromidia" were said to give rise to "secondary nuclei," each of
which with a bit of cytoplasm became a "gamete." The gametes were
said to emerge from parents and fuse to form zygotes which grew into
adults. Belaf (1926) stated that these accounts were based on misin-
terpretations involving small fungoid parasites. However, Ivanic
(1936) has described "copulation" in Mastigina hylae, an amoebo-

144 SEX IN MICROORGANISMS

flagellate inhabiting the rectum of tadpoles. Fusion of cells was sup-
posed to be followed by nuclear fusion, but further development was
not reported. The evidence in this case is not convincing and is too
incomplete to establish syngamy.

Order Protonionadina

In this group one of the most important families is the Trypano-
somatidae, which consists of the so-called haemoflagellates. They are
uniflagellate and have a characteristic kinetoplast (sometimes errone-
ously called a parabasal body) in addition to the nucleus. An
enormous amount of research has been conducted on this group,
especially the trypanosomes, yet no complete account of syngamy
has been produced.

The early interpretations of sexuality in haemoflagellates by
Schaudinn (1904), Prowazek (1904a, 1905), Moore and Breinl
(1908), Moore, Breinl, and Hindle (1908), Baldrey (1909), Schil-
ling (1910), and Lebedeff (1910), based largely on polymorphism,
have mostly been discredited. Doflein (1910) saw the wide variety of
morphological types in cultures of Trypanosoma rotatorium but
cautioned against calling these "male," "female," and "indifferent"
without convincing evidence. In 1926, Wenyon reviewed the descrip-
tions of sexuality in Trypanosomatidae as reported in the literature up
to that time and concluded that in no case had authors submitted suffi-
cient evidence to support the interpretations of sexuality. Since then,
additional descriptions of phenomena interpreted as sexual have ap-
peared from time to time, especially during the past decade.

Elkeles (1944) reported possible sexual phenomena for Try-
panosojna cruzi as had Chagas (1927) and Muniz (1927) some years
earlier. Vanderplank (1944) described two types of individuals for
T. congolense and T. rhodesience. In the latter species, one type had
two paired and two unpaired chromosomes (72=6), and the other
had two paired and one unpaired chromosome {n=^S). In T. congo-
lense one type had three pairs of chromosomes and the other had
three paired and one unpaired chromosome. It was supposed that
meiosis took place, the unpaired chromosomes acting like the sex
chromosomes of other animals, and that gametes so formed would
fuse with other haploid cells. Later (1947) the same author exhibited
demonstrations showings nuclear divisions with three chromosomes

SEX IN PROTOZOA 145

supposed ro be ganionts or gametes of T. congolense. Also, "latent
bodies" were said to form after sexual union, and these contained one
or two nuclei and six chromosomes.

Roskin and Schischliaie\\'a (1928) described mitoses with only
three chromosomes, not only in T. congolense but also in T. equiper-
dinn, T. briicci and T. ''surra,''' and Wolcott (1952) found only three
chromosomes in T. leivisi. In many protozoan nuclei the prophase
chromatids often become rather widely separated so that the chro-
mosome number appears to be double that shown in the metaphase
and anaphase. Thus the condition of two paired and two unpaired
chromosomes reported by \^anderplank may well be a prophase stage
show ing three pairs of chromatids, two of which are somewhat more
separated than the others, and the combination of two pairs and one
unpaired might well be a case where the chromatids were visible for
two of the chromosomes and not well separated for the third.

Fiennes (1945) also reported sexual phenomena for Trypano-
soma congolense. Developmental forms were found in sections and
smears from the skin of infected cattle. T. congolense was stimulated
to "conjugate" by adding a drop of 0.3% sodium chloride to a drop
of mouse blood containing the flagellates; micro- and macrogametes
w ere said to develop from sexually mature forms and fuse within
10 minutes. A "microgamete" penetrated the translucent posterior
region of a "magrogamete," the posterior end entering first. The
"zygote" thus formed was about 20 v- long and motile, with try-
panosome shape, but contraction of the body produced an "oocyst"
about 10 [i. in diameter. The "oocyst" appeared to form "sporoblasts,"
probably eight in number, which divided into many "sporozoites."
The fate of these "sporozoites" was uncertain. Trypanosomes in
mouse blood treated with weak salt solution showed three different
forms: (1) immature sexual or asexual stages which showed no
changes except swelling; (2) mature sexual stages which united as
described; (3) infective (?) forms which were changed into a variety
of developmental stages, round, stumpy, elongate, and so on, usually
found in the tsetse fly. The predicted confirmatory papers have not
been seen.

Fairbairn, Culwick, and Gee (1946) reported syngamy in Try-
panosoma rhodesience and T. simiae. Metacvclic forms of T. rhode-
siense occurred in two distinct types with significantly different
mean lengths. These two types produced three blood types, long.

146 SEX IN MICROORGANISMS

short and intermediate (cf. three types mentioned by Fiennes for T.
congolense). All forms were said to have six chromosomes (possibly
three pairs of chromatids). In a suspension in Ringer 's-glucose solu-
tion, the following events usually took place within 20 minutes: two
trypanosomes approached and intertwined, their centers making con-
tact but their ends free. After a short time the anterior end of each
fused with the posterior end of its mate, producing a fusiform body
with two nuclei and a flagellum at each end. After a time, these fusi-
form bodies divided at the middle, producing two trypanosomes.
Sometimes before the fusiform body divided one partner appeared to
degenerate and the other might or might not free itself. Instances were
found of more than two flagellates fused together. Similar conditions
were observed on stained slides. Sometimes head-to-head fusions
also took place. In T. siviiae identical processes were seen. The
authors stated that in blood trypanosomes syngamy could be induced
by almost any marked and sudden change of environment. Each of
the three forms, long, intermediate, or short, could fuse with itself
or with either of the other two. That is, sex in T. rhodesieuse was
not obviously related to morphology. (This statement is interesting
since most other authors have made differences in morphology the
basis of sex differentiation.) The authors considered that syngamy
was an adaptive process in which two flagellates associated to ex-
change three chromosomes. (Such an exchange of a haploid set of
chromosomes would correspond to conjugation in the ciliates.) The
authors noted that Hoare (1936) had reported somewhat similar
phenomena in T. congolense in blood films and interpreted the ap-
pearances as autoagglutination or agglomeration. Agglutination is
also suggested by the lack of distinctions between uniting forms
noted above. Cytological evidence for the "exchange of chromo-
somes" was not presented.

In 1951, Culwick, Fairbairn, and Culwick reported that when
they inoculated a mixture of T. rhodesieuse and T. briicei into rats
the resulting infections were different from the parent strains. Cycli-
cal transmission of this infection by Glossina morsitans did not restore
the morphology seen in the parental types. Hybridization was con-
sidered an explanation, and similar results were obtained when T.
ganibiejise and T. hnicei were the "parent" species. However, the
notorious polymorphism of all three of these species might well cast
doubt on the interpretation given.

SIX IN PROIOZOA 147

(^rlicr nicnibcrs of rhc Trvpniiosoniiuidac have been credited
with sexLiaUtw Fki (1908) described "male" and "fenvale" stages in
a leptomonad flagellate from the gut of Melophagiis ovinus. Although
he saw no "copulation," certain stages were interpreted as being
"ookinetes."

Adie (1921) reported "conjugation" between free flagellates of
Leisbviaiiia douovani in gut cells of the bedbug, Civiex kctular'ms,
which had been fed on spleen juice from a case of kala azar. Patton
(1922) stated that he had conflrmed the intracellular stages reported
bv Adie, but \\'enyon (1926) remarked that the bugs died after
feeding on the spleen juice and were then placed in an incubator at
27 "C. The developmental stages were found about 36 hours after-
wards. He thought that the observed development was more like that
in a culture than that in a vector and was skeptical of the interpreta-
tions given.

Although there appears to be a considerable amount of evidence
for cell fusions in members of the Trypanosomatidae, it must be con-
fessed that the observations reported are usually capable of inter-
pretations other than "sexual," and in no case are the so-called "sex-
ual" phenomena complete with satisfactory chromosome counts and
meiosis.

In the somewhat related Cryptobiidae there are several accounts
of sexual reproduction. The most convincing is that of Belaf (1916),
who repudiated his interpretation in 1926.

Possibly the earliest accounts of cell fusion in free-living mem-
bers of the Protomonadina were those of Dallinger and Drysdale for
several small unnamed flagellates (1873a,b; 1874a,b), in addition to
one with the characters of Folytovia (1874c), and the "calycine
monad" (1875), which apparently was Tetramitus rostratus. Cell
fusion was followed by encystment. At excystment, large numbers
of small granules wxre released which, according to the authors, grew
more or less rapidly to the adult condition. A\ hile evidence for cell
fusion might be accepted, the formation of large numbers of minute
granules which grow into adult flagellates has not been confirmed.

Additional accounts of fusion of ordinary cells followed by en-
cystment have been given for a number of protomonads, for exam-
ple, for Monas vivipara (Prow^azek, 1903); Bodo lacertae (Prowazek,
1904b); Moims ternio (Martin, 1912); Helkesi77iastix faecicola (Fig.
E, 1 to 6) (Woodcock and Lapage, 1915); Cercomoiias lovgicatida

148

SEX IN MICROORGANISMS

(7 to 12); and Spirovwnas angiistata; with some evidence for Hetero-
inita globosa (Woodcock, 1916); and ProwazekeJh lacertae (24-28)
(Wenyon, 1920). Woodcock (1916) also described fusion of indi-
viduals of Spiromorias angiistata within a cyst (13 to 18), and Ale-
xeieff (1925) described fusion of active flagellates of Alphay/wnas sp.

Fig. E. Cell fusion in protomonad flagellates, all redrawn.

1-4, fusion of vegetative cells. 5, 6, encystment of zygote, of Helkeshnastix faecicola,
from Woodcock and Lapage (1915). 7-12, cell fusion followed by encystment of
zygote of Cercomofias longicauda, from Woodcock (1916). 13-18, fusion of cells
after encystment of Spiroino7ias angiistata, from Woodcock (1916). 19, 20, vegeta-
tive cells. 21-23, fusion of cells of Alphavionas sp. (? — Spironionas angustata) from
Alexeieff (1925). 24-28, cell fusion and encystment of Proivazekella ( = Proteromonas)
lacertae, from Wenyon (1920).

SEX IN PROTOZOA 149

(19 to 23). Prown/.ck's '"Bodo^' laccrtctc is probably identical with
Wcnvon's Vro^vctzekclhi Jaccrtac, now called Vrotcroinoiias hicertae,
and Alexeieff's Alphav/ajhis sp. has been identified by Woodcock
(1921) as the Spirouionas angiistiita described by the latter in 1916.
Prowazck interpreted his evidence as indicating reduction divisions
followed by fusion of nuclei, but Wenyon (1920) does not support
that interpretation, although he does show cell and nuclear fusion
(Fig. E, 24 to 28). It is probable that the reduction nuclei of Prowa-
zek \\ere the paranuclear bodies (formerly called parabasal bodies)
characteristic of this flagellate. Prowazek (1904b) also described "au-
togamy" in cysts of ''Bodo'' hicertae, but this is probably a misinter-
pretation, as indicated by Wenyon (1920).

Dangeard (1910) described "autogamy" for Anthophysa vege-
tans. He believed that a nuclear division took place in weakened cells
which were unable to complete cell division. These binucleate ani-
mals encysted, a clear area appeared in the middle, later to disappear
and to be followed by nuclear fusion. Dangeard was not able to
follow the process of excystation.

Order Folymastigma

A half century ago some protozoologists were ready to accept
incomplete evidence as indicating sexuality; for example, for Tricho-
vioiias intestmalis (=T. hoviinis) and Lavjblia intest mails {=Giardia
laviblia) by Schaudinn (1903), and for Trichomastix lacertae, Tri-
chomonas ''intestmalis'''' from the rat, and HexamitJis ''intestinalis'^
from Testudo graeca by Prowazek (1904b). In some cases develop-
mental stages of Blastocystis were confused with stages in the life
history of the flagellates.

The account of cell fusion followed by encystment and the later
emergence of minute sporules for Tetrainitiis rostratiis by Dallinger
and Drysdale has already been mentioned. Bunting (1926) worked
out the life history of this flagellate, which showed that it could
transform completely into an amoeboid stage that could encyst. Only
amoebae encysted, and only amoebae came out of the cysts. Amoe-
boid stages readily transformed into flagellate stages and vice versa.
Under appropriate conditions either the amoeboid or the flagellate
stage could persist indefinitely. Bunting watched fusion of living
flagellates under the microscope (Fig. F, 1 to 7), and stained prepara-

150

SEX IN A/IICROORGANISMS

lO'SO

Fig. F.

1-15, Tetramitiis rostratus from Bunting (1926); 1-7, fusion of living flagellates; 8-11,
fusion of animals of unequal size, stained slide; 12-15, fusion of flagellates of equal
size, stained slide. Note multiple nuclei suggesting meiotic phenomena.

SI X IN PRoro/oA 151

rions of such tusiii<4 iiulix idiials showed nuclear conditions somewhat
like those reported l)\- Dohell (1908) for Copromonas siibtilis (H to
15). I lowcver. Bunting did not claim that a process of syngamy had
taken place. These processes of cell fusion probably are significant,
and further studies arc needed with that possibility in mind.

The onh' satisfactory accounts of syngamy in members of the
Polymastigina are those of (Cleveland for the flae^ellates living in the
gut of the wood-eating roach, Cry ptocerciis pmictiilatus. The same
may be said for the Hypermastigina. A review of Cleveland's papers
follows.

HORiMONE-INDUCED SEXUALITY IN ANIMAL FLAGELLATES

One of the most interesting developments in the field of sexual
reproduction in Protozoa is the occurrence of syngamy in the flagel-
lates that inhabit the gut of the wood-eating roach. Cry ptocerciis
pinictulatus. These conditions have been brought to our attention by
Cleveland who, with his associates, published an extensive paper on
the morphology, taxonomy and host-parasite relations of this group
of flagellates in 1934.

More recently Cleveland has begun a series of papers revealing
the details of "sex" in these flagellates. In two preliminary papers
(1947a,b) this author noted that the flagellate fauna of the roach
consists of some twenty-five species distributed into two orders, eight
families, and twelve genera, ranging from relatively simple polymas-
tigotes to extremely complex hypermastigotes, all of which exhibit
some form of sexual behavior at the molting period of the host. Be-
tween molts, no sexuality occurs. In termites, closely related flagel-
lates do not exhibit any sexual activity. At each molt of termites, the
flagellates are lost and must be regained from other members of the
termite colony by proctodaeal feeding.

Cleveland believes that the sexual activities of the flagellates in
Cry ptocerciis are induced by some direct effect of the molting hor-
mone of the host, rather than by an indirect effect through such
agencies as a decrease in food supply, increase in carbon dioxide, or
decrease in oxygen, all of which occur during molting.

One remarkable feature of this behavior is that different flagel-
lates follow different patterns of sexual activity, both in relation to
the process of molting itself and in relation to the various ways in

152 SEX IN MICROORGANISMS

GAMETOSENESIS

SEX IN PROTOZOA I53

w hich niciosis, gamete fornuuion, and syngani)' can be accomplished.
These variations allow for an analysis of the process of meiosis and
an evaluation of the differences between meiosis and vegetative mito-
sis. Diploidv and polvploid\' mav arise tiirough chromosome duplica-
tion without centriole duplication as \\ ell as through syngamy. The
series of papers on indixidual genera and species gives tiie important
details.

In preparing these re\iews of Cleveland's published papers, I
have, to a large extent, adopted my own phraseology. In doing so
there has been the risk of failing to convey the exact meanings that
Cleveland had in mind. It is to be understood, therefore, that I am
assuming full responsibility for the statements made in these reviews.
Naturally, limitations of space have made it necessary to omit a great
manv important aspects of Cleveland's work; hence the reader is
urged to read the original papers in order to appreciate their full im-
portance.

Folyniastigotes. In three separate papers Cleveland described
the sexual behavior of members of the genera Oxyvwiias (1950a),
Siiccinobdciilus (1950b), and Notila (1950c), all belonging to the
Polymastigina. These three genera are closely related, each having
four anterior flagella, intranuclear centrioles, flat, rather broad axo-
styles, and a relatively large number of chromosomes. In Oxymojias
(Fig. G, 1 to 4) and Saccmobacuhis (5 to 6) the sexual cycles are
very similar. In each the animals are haploid, and gametogenesis con-
sists of the division of a gametocyte (1) (which before the division
cannot be distinguished from an ordinary agamont) into two gam-
etes which are only slightly dissimilar in size. During this division in
Oxymonas doroaxostyhis the nucleus leaves the nuclear sleeve, and
this sleeve with the old flagella and axostyle degenerates, new flagella
and new axostyles being developed by the daughter centrioles for

Fig. G.

1-4, Oxyvionas doroaxostyhis, from Cleveland (1950a). 1, Gametocyte, prophase,
nucleus still associated with extranuclear organelles, axostyle (heavy, black body),
flagella, and nuclear sleeve shown; 2, gametocyte, metaphase, old flagella, and nuclear
sleeve have disappeared, old axostyle disintegrating, each centriole producing a new
axostyle; 3, gametes fusing, anterior end of one joining posterior end of the other;
4, axostyles are fused for half or more of their length and nuclei are touching. 5,
6, Saccinobaciihis avibloaxostylus, from Cleveland (1950b). 5, axostyles of gametes
beginning to fuse bringing the pronuclei close together; 6, anaphase of zygotic
meiosis, old axostyles nearly disintegrated, new axostyles about half grown.

154

SEX IN MICROORGANISMS

GAMET06ENESIS

SEX IN PROTOZOA 155

the dauQ^hter cells whicli ;irc gaiiicrcs (2). In Sacct7iobaciiliis ambloax-
ostylus the parental axosrxle is nor discarded until the mctaphasc of
nuclear division, and the parental flagella are retained, two going to
each daughter, two new ones being developed from new centrioles
arising by division of parental centrioles, a new axostyle growing out
from each new centriole.

In Oxv7no/his, two gametes meet and fuse (3), first the cyto-
plasm, then the axostyles (4), and finally the nuclei. Since gameto-
genesis begins 6 to 7 days before molting of the host, and meiosis,
which is zygotic, does not begin until approximately 1 day after
molting, the zygote stage lasts at least 6 days with no loss of extra-
nuclear organelles. In Sacciiwbaciihis, after cytoplasmic fusion (5),
nuclear fusion is delayed 2 or 3 days during which the cells become
spherical and remain rather inactive until the host molts several days
later. Within 8 to 12 hours after molting of the roach, the zygote
becomes more active, and, some 20 to 24 hours after molting, meiosis
(6) begins, to be completed in about 24 hours.

In the usual two-division meiosis the chromosomes are duplicated
during the first division, whereas the centromeres are not. In the
second division the centromeres are duplicated but not the chromo-
somes. In both these flagellates meiosis is accomplished by a single
nuclear division durino- which neither chromosomes nor centromeres
are duplicated. Only slight pairing takes place, and the chromosomes
move to the poles as monads, as seen in the second meiotic division
of most other animals. Random segregation takes place. Cytoplasmic
division of the zygote completes the formation of two haploid aga-
metes.

In Oxymonas nana essentially similar phenomena take place, but
in this species the zygote encysts. Later, excystation and division into
haploid agametes occur.

Fig. H. Notila proteus, from Cleveland (1950c).

1, agamont with nucleus, nucleoli, four flagella, axostyle, and axostylar granules; 2,
gametocyte shortly before cytoplasmic division, each nucleus with four flagella and
a half-grown axostyle; 3, axostyles of fused gametes beginning to fuse at their poste-
rior ends; 4, chromosomes of male and female nuclei have regained their major
coils preparatory to meiosis, male nucleus has become dissociated from fused
axostyles and is moving away; 5, male nucleus has moved to the posterior end of
the cell continuing its development, old fused axostyles still connected with female
nucleus which has two new axostyles and eight flagella; 6, early anaphase of male
nucleus, the female nucleus at this stage presents the same picture; 7, each male nu-
cleus is fusing with a female nucleus.

156

SEX IN A4ICROORGANISMS

SEX IN PROIOZOA I57

Between molts of the host the diploid Notila (Fig. H, 1) (Cleve-
huul, 195()c) divides by mitosis with twenty-eight chromosomes.
About 7 d;ivs before molting of the roacli occurs, this flagellate be-
comes a gametocNtc without an\- obvious morphological changes. In
a single division somewhat more rapid than usual (2), this jrametocvte
produces a male and a female diploid "gamete," the latter being some-
w hat larger. Each "gamete" possesses four flagella, an axostyle, and
two intranuclear centriolcs. These "gametes" increase in size, then
fuse, first the cytoplasm (3), then the axostyles (4), but not the
nuclei. In this condition the organisms remain for 7 to 8 days. Then
the male nucleus becomes detached from its axostyle, moving to
another part of the cytoplasm but retaining for a time its four fla-
gella and two centrioles.

Next meiosis occurs. This is nearly synchronous in the two
nuclei (5), There is no duplication of chromosomes in connection
with this division; hence reduction is accomplished by one division.
Two male and two female haploid nuclei, each with fourteen chro-
mosomes, are produced. Soon after meiosis is completed, fusion of
the pronuclei begins. iV male pronucleus ahvays fuses with a female
pronucleus, thus forming two diploid nuclei (7). Cytoplasmic divi-
sion of this "double zygote" produces two diploid asexual cells. There
are, of course, occasional deviations from the account just given.

Hypeniiastigotes. The first hypermastigote flagellate for which
a sexual cycle was described by Cleveland (1949) was Trichonyin-
pha. All the species of this genus go through the same cycle simul-
taneously, and all are haploid with twenty-four chromosomes during
periods between molts of the host.

Gametogenesis begins about 3 days before the host molts. At
molting time gametocytes are encysted, and nuclear division has taken
place but not cytoplasmic division. After molting, development of

Fig. T. Trichonympha, from Cleveland (1949).

1, a gametocyte in early stage of encystation; 2, detail of anterior end, duplication
of chromosomes has occurred, note difference in staining in nucleus; 3, vertical
view of metaphase-anaphase, most of male chromatids going to one pole and most
of female chromatids to the other, interconnections between light and dark groups
result from homologous pairing; 4, gametocyst of smaller species, gametes have more
space in which to move, are about ready to excyst; 5, male gamete is almost
halfway in the female gamete and has already become slightly smaller owing to some
of its cytoplasm having been passed to female.

158 SEX IN MICROORGANISA4S

gametes proceeds and excystation takes place within 15 to 20 hours;
within 4 to 6 hours, after excystation, fertilization begins. Excysta-
tion, maturing of gametes, and fertilization may continue until 35
to 40 hours, but in most cases are completed in 24 to 30 hours, after
molting. Some of the cysts are egested just before or just after the
host molts. The same development takes place in cysts remaining in
the original host as in those egested and taken up by a newly hatched
nymphal host.

After a haploid gametocyte encysts (Fig. I, 1), it produces two
unlike gametes. The first step is the production of two unlike daugh-
ter chromosomes by each chromosome in the nucleus. These daughter
chromosomes differ in staining capacity in Heidenhain's haematoxy-
lin, the male chromosomes staining somewhat more darkly than the
female (2). Separation begins during prophase of division, and a spe-
cial type of union insures that the male group of chromosomes sepa-
rates from the female group on the mitotic spindle (3). After cyto-
plasmic division, which produces two gametes (4), only slightly, if
at all, different in size, cytoplasmic differentiation of gametes takes
place. This may become initiated before excystation but usually be-
gins afterwards.

At the posterior end of the female gamete, a ring of deeply
stainable granules is gradually formed. The clear area within this
ring may be everted as a cone, or retracted. It is apparently attractive
to the male gametes, in which dark granules are rather uniformly
distributed. A male gamete becomes attached to the fertilization cone
and follows this as it is withdrawn into the body of the female gamete
(5). Penetration by the male gamete is rapid, after which the fertili-
zation ring disappears and the cytoplasmic organelles of the male
gradually degenerate. The male pronucleus then approaches and fuses
with the female pronucleus, thus producing a zygote. By two typical
meiotic divisions, each accompanied by cytoplasmic divisions, hap-
loid agamonts are produced.

There are a good many variations from the typical series of
events just described. A small percentage of individuals that remain
in the roach do not encyst or go through any sexual development
whatever. During gametogenesis the two resulting cells may be in-
completely sexed, becoming gynandromorphs. Such cells attempt fer-
tilization but usually fail to complete it. In a few cases two male
gametes may succeed in penetrating a female gamete, and their nuclei

SEX IN PROTOZOA 159

may even fuse wirli rlic fciir.ilc nucleus. Sometimes chains of gametes
result from incomplete fusion of gvnandromorpiis. A female gamete
may be found at the anterior end of such a chain and a male gamete
at the posterior end. Rarelv^ autogamy niav take place within a cyst.

Eiicovwiiyvipha shows some morphological similarities to 7V/-
chovyuiphit, although placed in a separate family 1)V Cleveland. The
sexual cycle of E. i/nla (Cleveland, 1950d) is similar to that of
Tricboiiyvipha in many respects, but there are characteristic differ-
ences.

The asexual animal (Fig. J, 1) is haploid wnth about fifty chro-
mosomes. It begins its sexual activities about 4 or 5 days before the
host molts by becoming a gametocyte without obvious morphological
change. This divides into a male and a female gamete by a single
nuclear and cytoplasmic division. As a rule, the male gamete is con-
siderably smaller than the female and its cytoplasm stains more darkly
because it contains many stainable granules. There is no special fertil-
ization area as in Trichonyuipha, males becoming attached to and
entering almost any part of the female gamete. At the point of con-
tact, the cytoplasm of the female begins to soften to permit the en-
trance of the male (2). However, the body of the male gamete does
not enter completely into that of the female. When half or more of
the male gamete has entered, its anterior end turns posteriorly and
the male organelles begin to degenerate. A4ost of the flagella and asso-
ciated structures are dissolved, but the rostrum, sometimes with the
axostyles and rostral flagella, projects posteriorly (3) and is pinched
off and discarded.

The freed male pronucleus moves to and fuses with that of the
female, thus producing a zygote (4). Further development is sus-
pended for about 4 days, then chromosome duplication occurs, but
not the centromeres. As the achromatic figure is developed by the
centrioles (those of the female gamete), pairing of chromosomes oc-
curs and tetrads are formed and divided (5), the daughter dyads going
to opposite poles. Cytoplasmic division completes the first meiotic
division. The second meiotic division, of the usual sort, produces hap-
loid vegetative individuals, which multiply by asexual mitoses as do
other similar flagellates.

Sometimes two male gametes will fuse with the same female.
There is no encystment at any part of the cycle.

Leptospkonympha, the next hypermastigote to be considered by

160

SEX IN MICROORGANISMS

SEX IN PROTOZOA 161

Clevchind (1 95 la), is placed in the Spirotrichonymphidae. It has a
rosrral area at the anterior end, and two long spiral flagellar bands
reaching nearly to the posterior end of the cell. L. uoachiila (Fig. K,
1) has an axostyle, but L. eupora lacks this organelle.

A vegetative individual of L. wachula becomes a gametocyte by
loss of the axostyle and the long flagellar bands and a rearrangement
of the flagella in the rostral region (2). Also, the cytoplasm becomes
denser, becoming filled with many small dark granules. Thus the
gametocyte presents a decidedly different appearance from that of
vegetative cells.

Gametogenesis is initiated by duplication of the ten chromo-
somes, each of which produces sister chromatids. This occurs between
the sixth and fifth days before the host molts. Some of the chromatids
are darker than others, but the contrast is not so marked as in Tricho-
iiyvipba. However, as in the latter genus, the chromosomes become
arranged into tw^o groups which separate from each other in the
ensuing gametogenic mitosis. That is, a male set separates from a
female set (3). Cytoplasmic division follows very quickly and may
take place before nuclear separation is complete, and even sometimes
so early that an anucleate cell is produced. An anucleate male was
seen to attach itself to a female gamete.

The organization of the gametes is like that of the gametocytes
except that the female has more cytoplasmic granules which tend to
congregate near the posterior end of the cell, and this area alone is
attractive to the male gamete, as in Trichoiiympha. By the time a
male gamete has pushed half way into the cytoplasm of the female
(4), the two cytoplasms begin to fuse and the male gamete progresses
no farther. The cytoplasmic organelles of the male gradually disin-
tegrate while the male pronucleus migrates to and fuses w^ith that of
the female. During fusion of the gamete nuclei, the chromosomes are
tightly coiled (5, 6) and remain so during the ensuing meiotic divi-
sion. Thus the chromosomes are not duplicated and pairing is slight.
The undivided homologues are segregated on the spindle that is

Fig. J. EiicoDwnyiupha hnla, from Cleveland (1950d).

1, an agamont seen partially in optical section; 2, stage in union of gametes; 3,
rostrum of male gamete has rotated 180 degrees, rostrum of female gamete should
extend anteriorly but has been bent in fixation; 4, male pronucleus in contact with
female pronucleus, male rostrum has been discarded, axostyles omitted; 5, metaphase
of first meiotic division, with about 50 tetrads, organelles other than centrioles,
achromatic figure, and nucleus, omitted.

162

SEX IN MICROORGANISMS

one-division meioiis

SEX IN PROTOZOA 163

formed, ;ind thus reduction is accomplished by a single nuclear divi-
sion (7). Ihc new asexual cells begin to develop flagellar bands, but
division is rapid for 2 or 3 days, so that the flagellar bands may not
attain full length until after this period of rapid division has ended.

Urhiynipbii is a hypermastigote with two anterior groups of fla-
gella symmetrically placed opposite each other. Besides the flagella
there are numerous parabasal bodies and very slender axostyles on
eacii side (Fig. L, 1). It has a diploid nucleus. Its sexual cycle has
recently been described by Cleveland (1951b).

In ordinary mitosis no organelles are lost. The two centrioles
elongate and produce a spindle at their distal ends. The two flagellar
areas separate and go to the daughter cells, a new group of anterior
flagella and other organelles developing to match the group passed on
from the parent. The sixteen chromosomes divide much as in other
hypermastigotes.

iMeiosis, which is pregametic, begins 5 to 6 days before the roach
molts and takes less than a day for completion, in most cases only a
few hours. The achromatic figure forms and functions much as it
does in ordinary mitosis, but the other organelles behave differently.
The flagellated areas with most of the parabasals and axostyles move
away from the nucleus with which the nuclear sleeve and achromatic
figure remain (Fig. L, 2), The old flagellar apparatus remains for
some time, undergoing various forms of disruption, and does not
disappear until after the new flagellated areas are formed in the zy-
gote. Meanwhile the nucleus undergoes a single meiotic division, the
chromosomes pairing and segregating without dividing, producing
haploid pronuclei (3). The cell body does not divide, but autogamous
pronuclear fusion takes place very soon after the haploid nuclei are
formed.

During autogamous nuclear fusion, one centriole, with a rem-
nant of the nuclear sleeve, moves away and degenerates. After nuclear

Fig. K. Leptospirovyi/ipha zvachula, inm-i Cleveland (1951a).

1, agamont; 2, late stage in formation of gametocvte, spiraling portion of flagellar
bands in process of disintegration; 3, early anaphase of gametocyte, each group of
sister chromatids forms a partial ring and rings lie close together, those of one
ring slightly darker than those of the other; 4, male gamete entering female gamete,
the male is losing its extranuclear organelles; 5, flattening of juxtaposed membranes
of fusing pronuclei; 6, nuclear fusion completed; 7, anaphase of one-division zygotic
meiosis, cell has increased greatly in size.

164

SEX IN MICROORGANISMS

■>cenfrioles

SEX IN PROTOZOA 165

fusion the remaining ccntriolc divides, and from each of the daughter
centriolcs a new half of the flagellar apparatus, along with axostyles,
parabasals, and nuclear sleeve, develops. Even after the new flagellar
apparatus develops, the old flagellar areas are still persisting. This con-
dition lasts for about 2 days, and then "pseudoencystation" takes place
w itii a rounding up of the cell but without the formation of a cyst
membrane. At this time the old flagellar apparatus completely disap-
pears. This rounded-up condition lasts for about 2 days, then, shortly
after the host molts, the organisms resume normal activity.

During these events a good many irregularities have been noted.
For example, when one of the centrioles fails to degenerate during
autogamic nuclear fusion, two complete sets of flagella and associated
structures develop. These tend to pull the cell into two parts, with
or without pulling the nucleus into two parts also. Anuclear and
partially nucleate cells die.

Sometimes, instead of meiosis and autogamy, endomitosis may
take place producing a 4N or even an 8N nucleus. Usually centrioles
are also duplicated, such duplication leading to the production of
extra sets of flagella. These cells do not survive. Occasionally, instead
of a meiotic nuclear division, an ordinary mitosis takes place without
cell division. The two diploid nuclei so produced fuse to produce a
4iV nucleus. This raises the important question why pronuclei fuse.
Perhaps the gametogenic nuclear divisions produce physiological dif-
ferences in the nuclei which result in their fusion. Sometimes pro-
nuclei are cut into two parts by the spindle. In such cases the two
parts of the same nucleus do not fuse with each other but with the
other pronucleus, or with parts of it, if it has also been cleaved into
parts. Another type of anomaly is division of the centrioles, in which
case new sets of extranuclear organelles are produced but the nucleus
fails to divide.

The very simple type of sexual cycle in Urinympha is consid-
ered by Cleveland in relation to other cycles from an evolutionary

Fig. L. Urinympha talea, from Cleveland (1951b).

1, diploid agamont in early mitotic prophase, showing two groups of flagella, para-
basals (pb), axostyles (ax), etc.; 2, entire anaphase of meiotic cell; eight chromosomes
are going to each pole; nucleus in center, with greatly disrupted flagellar areas,
parabasals and axostyles to the right; halves of nuclear sleeve almost separated; 3,
entire meiotic cell, discarded flagella, parabasals, and axostyles above the centrally
placed pronuclei.

166

SEX IN MICROORGANISMS

SEX IN PROTOZOA 167

point of view . This aspect of the subject is discussed in the paper
"Coninicnts on the Origin aiid I'Aokition of Sex."

In its morphology Rhyiicbonyinpba (Cleveland, 1952) is inter-
mediate l)et\\een Urinynipba and Bcirbiihrnympha. Its sexual cycle
shows similarities to that of both these genera but has features pecul-
iar to itself.

Sexual activities of diploid Rbyiicbonynipba tarda (Fig. M, 1)
begin about 9 days before the host molts and end 10 to 12 hours after
molting. First there is a meiotic nuclear division (2, 3) accompanied
by cytoplasmic division. In the cells thus produced a second meiotic
nuclear division (4, 5) takes place without cell division. The two
haploid pronuclei thus formed fuse in a process of autogamy (6 to 9).
This series of events takes about 2 days. The resulting cell, which
Cleveland hesitates to call a zygote, increases considerably in size for
about 2 days, then undergoes a reorganization during which the old
set of extranuclear organelles gradually degenerates while a new set
develops from two centrioles. Pseudoencystation, which lasts about 3
days, follows. During this period the flagellate may be passed along
to ne\\'lv hatched and uninfected hosts. After some 12 to 15 hours
the flagellates resume their activity, and a period of rapid cell divi-
sions ensues.

In Rbyncbonyvipba the size of the chromosomes {2N = 20)
varies considerably, but all have median or submedian centromeres,
in contrast to those of Urinympba, which are similar in size and have
terminal centromeres. In the first meiotic division, homologous chro-
mosomes pair only at their distal ends, so that no tetrads in the ordi-
nary sense are produced and crossing-over is thus practically im-
possible. Lateral joinings of non-homologous chromosomes produce
groupings on each side of the equatorial region, as described for
TricboiiyiJipba and Leptospironyvipba (2, 3). In the second meiotic
division the behavior of the chromosomes is very similar to that in
the first, except that there is no duplication, there being a minimum

Fig. M. Rhynchonympha tarda, from Cleveland (1952).

1, entire animal, showing organization; 2, metaphase or early anaphase of first meiotic
division, non-homologous chromosomes joined into two rings, homologous chromo-
somes connected at the equatorial plane; 3, details of stage shown in 2; 4, metaphase of
second meiotic division, two rings of non-homologous chromosomes, as in 2; 5,
details of stage shown in 4; 6, stage in fusion of pronuclei, organelles attached to
female nucleus; 7-9, successive stages of pronuclear fusion; 10, encysted animal.

158 SEX IN MICROORGANISMS

of contact between homologous chromosomes, while the non-homo-
logues join to form into two rings which pass intact to opposite poles
of the spindle (4, 5 ) .

During mitosis, after separation of the daughter groups of chro-
mosomes, in late anaphase, the middle portion of the elongated nu-
clear membrane dissolves and releases the contained granules into the
cytoplasm. Cleveland states his belief that such a partial breakdown
of the nuclear membrane may be much more prevalent in Protozoa
than previous statements would indicate.

The first meiotic division is like an ordinary cell division in that
the parental group of extranuclear organelles is separated into its
constituent halves, one of which passes to each daughter cell. A new
half set or group develops from each daughter centriole as the parent
centrioles function in nuclear division. After the second meiotic
division, the half set of extra nuclear organelles associated with the
male pronucleus degenerates, leaving the other set to function in
the usual manner (6). About 4 days after fusion of the pronuclei
the older complement of organelles gradually degenerates while an
entirely new group develops from the two centrioles which have re-
mained in contact with the nucleus. This is the process of reorganiza-
tion, mentioned above; it involves neither nuclear nor cytoplasmic
division.

There are a good many variations from the series of events de-
scribed above. For example, instead of pseudoencystation, sometimes
true cysts are formed (10) in which there is no development.

Certain of the more important aspects of the sexual cycles of
these flagellates living in the gut of Cryptocercus are shown in Table
II. There it will be seen that, of the three polymastigote flagellates,
Oxymonas and Saccinobaciilus are haploid and that postzygotic meio-
sis is accomplished in one division. Notila is diploid, but again the
pregamic meiosis involves only one division.

In the hypermastigote group, Leptospironympha, Trichonym-
pha, and Eiicomonympha are haploid. The postzygotic meiosis of
Leptospironympha is accomplished by a single division, whereas in
Trichonyrnpha and Eucomonyjnpha there are two meiotic divisions.
The diploid Urinympha achieves pregametic meiosis in a single divi-
sion and autogamy takes place, whereas the other diploid, Rhyjicho-
iiy/npha, requires two pregametic meiotic divisions to produce haploid
gametes which fuse in autogamy. Not only are these diversities in

SEX IN PROl OZOA

169

TABLE 11

SlMMAKV, EsrEClALLV OF MlCIOTlC TlIENOMENA, IN THE FlAGELI.ATES IN THE (iUT

OF THE Wood-Feeding Roach, Cryiocercus pundulalus, as Re\ealed Thus

Far by Cleveland

Nunil)crs in parentheses in column 2 refer to chromosomes; those in column 3 refer to numbers
of meiotic divisions. Further explanations in the text.

Genera

Adults

Meiosis

Autogamy

Polymastigina

Oxymonas

Haploid (25)

Postzygotic (1)

—

Saccitiohaculus

Haploid (30)

Postzygotic (1)

—

Xotila

Diploid (28)

Pregametic (1)

—

Hypermastigina

Leptospironympha

Haploid (10)

Postzygotic (1)

—

Trichonympha

Haploid (24)

Postzygotic (2)

—

Eucomonympha

Haploid (50)

Postzygotic (2)

—

Urinympha

Diploid (16)

Pregametic (1)

+

Rhynclwnympha

Diploid (20)

Pregametic (2)

+

sexual behavior very striking, but they are also interesting because
rhev occur only under the influence of the molting hormone of the
host and the time relations to the molting event are varied in the
different flasfellates.

CLASS SARCODINA

Syngamy has been reported for a good many representatives of
this class, especially for the more highly evolved Mycetozoa, Fora-
minifera, Heliozoa, and Radiolaria. The Mycetozoa, on the border
line bet\A'een the Protozoa and the Fungi, have been considered by
Dr. Raper in his paper on the fungi.

Order Frotcoviyxa

Much diversity occurs among the organisms in this group, but
generally they have slender, branching, and anastomosing pseudo-
podia. Cell fusion has been reported for a few members such as
Niiclearia sivtplexhy Dangeard (1886) and Aratari (1889), for Vam-
pyrella spp. by Dangeard (1886), and for Chlaviydomyxa moii-
tana by Penard (1904a), but nuclear details were not given.

170

SEX IN MICROORGANISMS

SLX IN PROTOZOA 171

Order Am o chin a

AmoniT the accounts of sexual reproduction in amoeboid forms,
that for SiippiiJCii {Amoeba) diploidea (Fig. N, 1 to 6) appears to be
the most acceptable. First described by I lartmann and Niigler (1908)
and Nagler (1909), the life cycle is about as follows. This species is
normally binucleate, the nuclei lying close together and dividing si-
multaneously at each binary fission (1, 2). Under certain conditions,
two of these binucleate animals come together and develop a common
cyst membrane about them (3,4). In each amoeba the two nuclei fuse
in a long-delayed fertilization karyogamy. The cytosomes next fuse,
then each diploid nucleus undergoes a "reduction" division, one
daughter nucleus degenerating. A second "reduction" division pro-
duces a haploid nucleus and a second reduction body (5). These two
nuclei remain as those of the vegetative animal which emerges from
the cyst (6).

The general features of this life cycle were confirmed by De-
schiens (193 3), who added a process of schizogony, and partially
confirmed by Kropp (1939). In none of the accounts are there clear
indications of chromosome numbers.

In addition to his study of Sappinea diploidea, Nagler (1909)
also reported evidence for autogamy for Amoeba f rose hi and Amoeba
alba, but the accounts are not convincing. Other descriptions of sex-
ual reproduction in smaller amoeba are those for Avweba mimita
(Popoff, 1911), in which the formation of "gamete" nuclei out of
chromidia was reported, and for Amoeba mira (Glaser, 1912), in
which two maturation divisions were described without syngamy
having been seen.

For Amoeba protens (Chaos diffineiis) and its close relatives
there are numerous accounts indicating complex life cycles, often

Fig. N.

1-6, Sappinea (Amoeba) diploidea, from Kuhn (1926), after Hartmann and Nagler
(1908), redrawn. 1, binucleate vegetative amoeba; 2, binary fission; 3, association of
two vegetative amoebae; 4, encystment of such a pair; 5, fusion of cytoplasms and
meiotic nuclear division; 6, binucleate vegetative amoeba emerged from the cyst. 7-11,
Euglypha sctitigera, from Penard (1938), redrawn. 7, two normal individuals attached
at oral areas; 8, new fusion body beginning to emerge; 9, new larger individual de-
rived from fusion of the two original animals; 10, encystment of "zygote"; 11, older
cyst.

172 SEX IN MICROORGANISMS

including formation of secondary nuclei from chromidia, and the
development of amoeboid or flagellated "gametes," and the like. Some
of the more recent accounts are those of Hausmann (1920), Hulpieu
and Hopkins (1927), and Jones (1928). But Johnson (1930), who
reviewed previous work and repeated some of the earlier experiments,
concluded that Amoeba proteus reproduces exclusively by binary
fission, and that internal parasites, introduction of small amoebae
which pass through ordinary filter paper, and mycetozoa derivable
from airborne spores could readily explain the great variety of life
cycles reported.

The formation of amoeboid or flagellated "swarmers" has been
described for the large multinucleated amoeba, Pelomyxa palnstris,
by Greef (1874), Korotneff (1879), Veley (1905), and Bott (1907),
(Bott reported fusions of gametes), but Schirch (1914) declared that
this amoeba reproduces only by binary fission accompanied by divi-
sion of all the nuclei, or by a process of budding, and that previous
accounts of sexual reproduction involved parasites. Wilber (1947)
stated that "'Pelojriyxa''^ carolinensis {Chaos chaos) and similar multi-
nucleated amoebae reproduce only by plasmotomy. Parasites may
have been involved in the account of the formation and fusion of
flagellated "gametes" from a large multinucleated marine amoeba
reported by Schepatieff (1910).

Older accounts of sexuality in endozoic amoebae, such as that
for Entavweba colt by Schaudinn (1903), which was "confirmed"
by numerous other authors, are no longer credited. Sexuality in Eii-
damoeba blattae has been described or assumed by several authors
(for example, Mercier, 1909, 1910; Elmassian, 1909; and Morris,
1936) but has not been confirmed. Lucas (1927) separated the amoe-
bae of diff^erent sizes in the roaches (previously fitted into one life
cycle) into three different kinds: the largest was Endainoeba blattae
(Biitschli, 1878); a smaller one was named Entavweba thomsoni; and
a still smaller one was called Endolimax blattae. Meglitsch (1940)
made an extensive study of the nucleus and of nuclear division in
Endamoeba blattae but found no evidence of sexuality.

Thus we have no convincing evidence of sexuality in amoebae
except that for Sappinea diploidea, and even that is incomplete.

SEX IN PROTOZOA 173

Order Testacea

The rendcMicy for testate rhizopods to display plasmogamy, or
the fusion of the cytoplasms of two or more individuals, has been
noted by many investigators, from the earlier observers, such as
Biitschli (1875) for Arcella vulgaris, Gabriel (1876) for ''Troglo-
dytes'" {=Chhr//iydophrys), and Leidy (1879) for Euglypha alveo-
lata, Arcella vulgaris, Difflugia lobostovia, and other species, down to
recent times.

Biitschli (1875) saw small amoebae in the shells of two of three
specimens of Arcella which had their cytoplasms fused. He believed
that these small amoebae were offspring rather than parasites, as did
Hertwig (1899) and his followers, Elpatiewsky (1907), Swarczew-
sky (1908), and others. Swarczewsky, for example, described a life
cycle for Arcella vulgaris which included, besides the usual binary
fission, the formation of two types of amoebulae (in different indi-
viduals) with nuclei derived from the chromidia of the parents. These
were supposed to be macro- and microgametes which fused to form
spherical cells that developed into adults. In addition, chromidiogamy
was described in which the cytoplasm of two parents fused and their
nuclei degenerated. The chromidia were supposed to fuse or mingle
and then give rise to secondary nuclei which became enclosed in
parental protoplasm to produce a brood of amoeboid stages; these
developed through a Nuclearia-Yike stage to the adult condition. In
still other cases, after the formation of secondary nuclei out of the
chromidia, a process of schizogony followed which produced small
amoeboid individuals either without the parent emerging or after
the emergence of the parent. Each such "offspring" developed
through a Nuclearia-\ikG stage to the adult.

Dangeard (1910) reported that he had seen no other form of
reproduction in Arcella besides binary fission. He stated that there
is only one kind of nucleus, denying that secondary nuclei formed
from chromidia, but he thought that uninuclearity might result from
nuclear fusion. Like others, he saw numerous examples of plas-
mogamy involving various numbers of individuals. He declared that
the pseudopodiospores of other authors were parasitic Nuclearia and
that the previously reported gamogony was the union of gametes of
Nuclearia that he had previously described for Nuclearia si?nplex.

The idea that nuclei can arise from the so-called chromidia has

174 SEX IN MICROORGANISMS

largely been abandoned in recent times. The presence of small amoe-
bae in the shells of Arcella is not too uncommon (I have observed
them myself), but there seems to be no real evidence that they have
any part in the life cycle of Arcella. Definite evidence of syngamy in
this genus appears to be lacking.

Plasmogamy has been reported for certain other members of the
Arcelhdae. Although Gabriel (1876) and Schaudinn (1903) reported
cell fusion in Chlmnydophrys stercorea, the well-known process of
budding and colony formation in this species might be misinterpreted.
Martin (1912) cultivated this species for a year without seeing any
signs of sexuality. However, Belaf (1921) reported plasmogamy for
Chlamydophrys minor and C. scbaudinni, which included nuclear
fusion, but he stated that karyogamy resulted in death of the cells.

Penard (1902) merely mentions finding a pair of individuals
of Fyxidicula cyinbahmi in "conjugation," as did also Cash and
Hopkinson (1909) for Cryptodiffliigia ovi^ormis. However, Ivanic
(1935a) reported complete "copulation," including nuclear fusion,
for CochUopodhiin digitatiivt. "Copulae" with food reserves were
able to undergo further development, but those without such reserves
could not. After a rest period, during which the food reserves were
used up, each "copula" divided into two cells in the parent shell.
These emerged in amoeboid form and each produced a new shell.

Plasmogamy, or the fusion of cytoplasms, has frequently been
reported for species of Difflugia, for example by Leidy (1879),
JickeH (1884), Verworn (1888, 1890), Rhumbler (1898), Penard
(1902), Zuelzer (1904), Cash and Hopkinson (1905), Edmondson
(1906), Goette (1916), PatefT (1926) and Dangeard (1937). In some
cases the contents of one shell migrated into the other one, after
which nuclear fusion might occur (Dangeard), but usually was not
observed (Rhumbler).

Rhumbler (1898) stated that he had seen hundreds of cases of
plasmogamy in Difflngia lobostovia, sometimes involving three indi-
viduals and occasionally four. During the previous several years he
had stained many pairs but found no unusual nuclear conditions. On
the other hand, Zuelzer (1904) reported not only plasmogamy in the
multinucleate D. iirceolata, which had nothing to do with reproduc-
tion, but also "copulation" and "conjugation." Plasmogamy lasted
for as short a time as 2 hours, but more often for 2 or 3 days, and
exceptionally for 8 to 14 days. Plasmogamy seemed to be more fre-

SI"X IN PROIOZOA 175

cjucnr under poor cultural conditions and in warmer weather. In
"copulation," two individuals first underwent plasuioganiy; then the
contents of one shell flowed into the other. The chroniidial material
of the two then mingled, hut karyogamy was not observed. Encyst-
ment followed, but before encystment "chromidia" emerged from
the nuclei, most of which degenerated. This process occurred regu-
larly in late w inter. "Conjugation" was similar to "copulation" except
that three individuals were involved. In time, new secondary nuclei
were said to form from the chromidia in the cyst.

A few other genera of the Difflugiidae are represented in the
records of "copulation" and "conjugation" for example, Ceiitropyxis
iiculcata for which Rhumbler (1895) reported "conjugation" with-
out karyogamy, and Schaudinn (1903) reported "copulation" fol-
lowed by encystment.

"Copulation" and "conjugation" have also been recorded for
species of Eiiglypha and its relatives. Blochmann (1887) isolated a
pair of Eiiglypha aheolata. Tw^o days later the contents of these two
had combined to form a third and larger shell. Reukauf (1912) made
a similar observation for the same species, but encystment followed,
as it did after "copulation" in an unnamed species described by
x\werinzew^ (1906) and in Eiiglypha sciitigera by Penard (1938)
(Fig. N, 7 to 11). These independent observations agree so well as
to details that complete cell fusion seems to be indicated.

Other species in the Euglyphidae for which "copulation" or
"conjugation" has been reported are Cyphoderia (Rhumbler, 1895;
Cash, Wailes, and Hopkinson, 1915); Trinevm lineare (Penard,
1902); Nebela collaris (Cash and Hopkinson, 1905); Nehela and
Assuliva (Awerinzew% 1906); Trbieiim eiichelys (Cash, Wailes, and
Hopkinson, 1915).

Although the above-cited records show that cytoplasmic fusion
is a common phenomenon among the Testacea and that complete
fusion of two cells including karyogamy may take place, in no case
has there been a demonstration of haploid-diploid sequences.

Order Foraminijera

Early students of the Foraminifera noted that in many of the
polythalamous species there were two types of shells: one with a
larger first chamber, or prolocujum, designated megalospheric, and

176

SEX IN MICROORGANISMS

■■.'•.\ o

Fig. O. Selected stages of the life cycles of (a) Spirillina vivipara, (b) Dis-
corhis patellifor?ms, and (c) Polystomella crispa, from Myers (1938), redrawn.

Stages: 1, microsphaeric adult; 2, formation of megalosphaeric offspring; 3, asso-
ciation of megalosphaeric gamonts (a, b), no association (c); 4, production of
gametes which may be amoeboid (a) or flagellated (b, c); 5, new generation of
microsphaeric young produced by syngamy.

SEX IN PROIOZOA 177

the other w ith a smaller proloculiim, called microspheric. It was also
noted that microspheric individuals were commonly multinucleate
and that megalospheric ones were uninucleate. Many of the general
features of the life cycles including alternation of sexual with asexual
generations were worked out by such authors as Lister (1895),
Schaudinn (1903) and Winter (1907), but apparently Myers (1935,
1936) was the first to describe a complete life cycle with cytological
details. He found in FateUina corriigata and Spirillina vivipara (Fig.
O, a, 1 to 5): that all nuclei result from mitotic division of other
nuclei, instead of being formed from chromidia as postulated by
Schaudinn (1903); that gametes are amoeboid in these species (a, 4);
that gamete formation is preceded by the association of two or more
gamonts inside a cyst or temporary brood chamber (a, 3); that gam-
ete formation is accompanied by a reduction in chromosome number;
and that microspheric agamonts also surround themselves with a
temporary cyst before producing young megalospheric gamonts by
multiple fission (a, 2). Later Myers w^ent to England and was able
to corroborate the earlier accounts of life histories involving flagel-
lated gametes.

As shown by Myers (1938), Die or bis patelUjonnis (Fig. O, b,
1 to 5 ) has a life cycle somewhat intermediate between that of Spiril-
lina vivipara (a) and Folystoinella crispa (c). The microspheric adult
(b, 1) gives rise to megalospheric offspring by a process of multiple
fission as shown by the other two species (b, 2). When these megalo-
spheric gamonts are mature, they associate in pairs with their ventral
surfaces in contact (b, 3). Between these parents a brood chamber is
formed in which flagellated gametes are produced (b, 4). These unite
in pairs, thus producing zygotes which develop into microspheric
young. These new individuals undergo growth and development to
a stage with several chambers before being released from the brood
chamber (b, 5). In Folystoiuella crispa there is no association of
gamonts before the formation of large numbers of flagellated gametes
(c, 4), so that fertilization is more a matter of chance than in the
other species illustrated.

Foyn (1937) studied nuclear conditions in the microspheric
stages of Discorbijia vilardeboana. The adult shells had from seven-
teen to twenty-one chambers, and there was much variation in the
numbers of nuclei, their sizes, and their distribution among the
chambers. Schizogony was preceded by two mitotic divisions. In

178

SEX IN MICROORGANISMS

»V» . ■■'■

\':'

X'

""'■^^fe

•■2- ■*.

6 \,. '■•>-.---**''^ ^J^

■-Vvi-;.,^.^,..,,,.,^.*'

,<'(Vi^'V><''Vi'iiv^:'';^.;jK;»3.^',i*v*'

\

'•i^iASS^ligJJIig:

isffisr-

SIX ix PRoro/oA 179

sonic examples of rlic lirsr of these di\ isions the prophase and nicta-
phasc chromosomes had rhe appearance of tetrads. Foyn stated that
the number of cliromosomes varied between ten and twenty, but the
drawings showino- the tetrads lia\ e the smaller numbers. Although
Foyn suggests tiiat these appearances, which resemble meiotic divi-
sions, may be the result of fixation, it is possible rliat a true meiosis
is indicated as claimed by Le Calvez (1946, 1950).

Le Calvez (1946), in new studies of Discorbis vilardeboaua with
flagellated gametes and of rateUina corniircita with amoeboid gam-
etes, declared that chromosome reduction rakes place during the last
two nuclear divisions preceding the formation of gamonts by mul-
tiple fission (schizogony), the entire gamont generation, therefore,
being haploid. In a still more recent paper, Le Calvez (1950) reiter-
ated his conclusions about ratelliva cornigata and Discorhis vilarde-
boana and added a study of association of gamonts of Discorbis iiicdi-
terraiieiisis. In the last species association takes place among gamonts
derived from a single microspheric "schizont" (agamont). Tw^o sex-
ual tendencies wxre evenly distributed among the gamonts, and Le
Calvez believed that chromosome reduction played a decisive role
in determining this sexual segregation. He believed that gamonts pro-
duce chemical attractants which bring together the gamonts of oppo-
site sex which are morphologically indistinguishable. Some kinds of
Foraminifera, he stated, like Discorbis orbicidaris and Entosolemis
Tnar<jriiiata show^ only a succession of mononucleate generations which
may be regarded as haploid parthenogamonts. If attractants are pro-
duced by these animals, they appear at two stages, one to bring
gamonts together, and the other to attract the gametes together.

Order Heliozoa

Sexual reproduction seems to be well established for a few species
of Heliozoa. Cell fusion, or plastogamy, is a fairly common phenome-

FiG. P. Syngamy in Actiiwphrys sol, from Kuhn (1926) after Belaf (1923),

redrawn.

1, mitotic division of an individual with temporary cyst membrane; 2, two game-
tocytes resulting from this division, "bouquet" stage of nuclei; 3, early tetrad stage,
left (male) individual slightly more advanced than right (female); 4, first meiotic
division; 5, second meiotic division, first "polar bodies" above; 6, initiation of fertili-
zation by pseudopodial formation by male (left); 7, fusion of nuclei in zygote; 8,
encystment of zygote.

180 SEX IN MICROORGANISMS

non for Actinophry s sol and certain other Heliozoa, as stated, for
example, by Penard (1904b). According to Schaudinn (1896), after
plastogamy, individuals of Actinophry s sol separate again without
nuclear fusion and without loss of pseudopodia. In addition, Schau-
dinn described a process of syngamy, involving first the fusion of two
individuals followed by withdrawal of pseudopodia and encystment
in a common gelatinous membrane; then a second folded membrane
formed about each individual. After this a "reduction" division took
place in each cyst, one daughter nucleus migrating to the periphery
and condensing into a "polar body." Then fusion of the two haploid
cells, including karyogamy, took place. Later the zygote nucleus
divided mitotically and two daughter cysts were produced, from each
of which a new vegetative individual emerged.

Schaudinn's account, while similar in general features, varied in
detail from the later accounts of Distaso (1908) and Belaf (1923).
Distaso described encystment of a single individual followed by
division within the cyst producing gametocytes, then two matura-
tion divisions, each producing a "polar body." Fusion of the haploid
isogametes followed. He called the process autogamy but failed to
give details about the chromosomes. Belaf's more complete account
(Fig. P) agreed in all general features with that of Distaso, but he
counted forty-four chromosomes in vegetative divisions, twenty-two
tetrads in the first maturation, and twenty-two dyads in the second
maturation division. After maturation short pseudopodia appeared at
one end of one of the gametes to initiate the fusion into a diploid
zygote (Fig. P, 6 to 8). The zygote formed a more permanent cyst
wall, and after a rest period emerged and assumed the active state.

It may be noted that in one gametocyte the progress of matura-
tion was slightly more rapid than in the other (3 to 5). Since the
gamete so produced took the initiative in bringing about fertilization,
it is regarded as the male.

Sexual reproduction in Actinophry s sol is essentially similar to
that described for Actinosphaeriwn eichhorni by Hertwig (1898).
Under certain conditions, such as starvation after being well-fed, mul-
tinucleate adults withdrew their pseudopodia and encysted, produc-
ing the "mother cyst." About 95 per cent of the contained nuclei
degenerated. The cytoplasm within the "mother cyst" then divided
into a series of "primary cysts," each containing one of the surviving
nuclei. (This primary cyst is comparable to the initial cyst of Ac-

SEX IN PROTOZOA 181

tinophrys sol.) Each "primary cyst" divided into two "secondary
cysts" ^^•ithin Mhich each nucleus gave off two "polar bodies," one
after the other. The pairs of secondary cysts tlien fused inside the
primary cyst membrane, thus producing a "conjugation cyst" which
included the zygote. The zygote eventually emerged as an active
individual.

In addition to asexual fission and budding, Zuelzer (1909) in-
cluded "gamogonie" in the life history of the marine heliozoon, Wag-
nereUa borealis. "Gamogonie" was of rare occurrence and involved
the formation of "flagellispores" which were presumed to be gametes.
There has apparently been no confirmation of this account, so that
it remains questionable. Sexuality has thus been fully demonstrated
only for Actinopbrys sol and Actinosphaerhnii eichhoriii among the
Heliozoa.

Order Radiohrria

This group consists of marine members of the Sarcodina, which
have radiating pseudopodia like those of the Heliozoa but differ from
the latter group by the possession of a membrane, the central cap-
sule, between the inner medullary region (endoplasm) and the outer
cortical region (ectoplasm).

Most of the Radiolaria have rigid siliceous skeletons, but one
group has skeletons of strontium sulfate. Some kinds have no skele-
tons, and others have loose spicules or crystals. Some are colonial in
organization. Kudo recognizes four suborders: Actipylea, Peripylea,
Monopylea, and Tripylea.

Some Radiolaria reproduce by binary fission or by budding or
even by multiple fission. These processes occur among skeletonless
and colonial forms. Some of the Actipylea and Tripylea also repro-
duce by binary fission. However, the most characteristic method of
reproduction is supposed to be the formation of large numbers of
flagellated swarmspores. These are of two types: isospores, usually
round, oval, pear- or spindle-shaped, each with a "crystal," which
Hertwig stated was organic rather than inorganic, and with one, two,
or three flagella (Fig. Q, 2, 3); and anisospores, usually without crys-
tals, often bean- or kidney-shaped, with two flagella and appearing
in two different sizes in the same species, or even in the same indi-
vidual (4, 5). The isospores are believed to represent asexual off-

182

SEX IN MICROORGANISMS

Fig. Q.

Life-cycle stages of various Radiolaria: 1-5 from Brandt (1885); 6-17 from Brandt
(1905)', 18-23 from Kuhn (1926), after Huth (1913); 24-28 from Schewiakoff (1926),
all redrawn. 1, developing microspores (left) and macrospores (right) in Sphae-
rozoum punctatum; 2, 3, isospores of Collozomn fulvuni; 4, 5, macrospore and
microspore of Collozown ineniie (2-5 from living material); 6, 7, macrospores, 8,
9, microspores of Sphaerozoum punctatum (from living material); 10-23, Thalas-
sicolla nucleatnm: 10, 11, macrospores; 12, 13, microspores; 14, 15, isospores killed
with iodine; 16, 17, from living material; 18-20, early to late stages in Schlauchkern-
serie, 21, higher magnification showing mitoses in stage shown in 20, note increase
in size with growth of Scblmicl^kernserie; 22, microspore from Schlauchkernserie;
23, macrospore from Spindelkernserie; 24, "gamete" of Fhyllostaurus cuspidatus; 25,
"gamete" of Fleuraspis costatus; 26, "gamete," 27, 28, "zygotes" of Acanthostaums
pnrpurasceiis.

SEX IN PRCriOZOA 183

spring-, \\hilc the anisosporcs ;irc thought to he gametes. However,
there is nuich confusion in the Hterature on the subject.

Although the Radiolaria had been the objects of study from the
rirst half of the nineteenth century onward, Cienkowski (1871b)
and I Icrtwig (1876) were among the first to report the two kinds of
swarmers and Haeckcl (1887), in his Challenger Report, indicated
his belief that sponilation was the common method of reproduction
in Radiolaria.

Brandt (1885) studied the Sphaerozoon group (colonial, Peri-
p\Iea) at Naples and recognized, besides the s\^'armers, young vege-
tative, young reproductive, old vegetative, and old reproductive
stages. The last stage produced the isospores and anisospores. Some-
what diverse conditions were found even within the same genus. For
example, for CoUozowii he found both isospores (Fig. Q, 2, 3) and
anisospores in C inervie (4, 5) and C. piilvuluvi, but only isospores
in C. pelctgicuvi and only anisospores in C. hertwigi. Similar differ-
ences occurred in the genus Sphaerozowii (6 to 9). Different condi-
tions prevailed in different groups. For example, in the Collosphae-
rida microspores and macrospores developed in different individuals,
the "spores" being spindle-shaped and sometimes having a crystal,
while in the Sphaeroida macro- and microspores developed in a single
individual (1), the "spores" being bean-shaped and wdth or without
a crystal; isospores generally were spindle-shaped and had crystals. In
all cases except the acanthometrids (which belong to the Actipylea)
the "spores" had two flagella. Like his predecessors, Brandt was un-
able to follow the development of swarmers to the young vegetative
stages.

Enriques (1919) also studied the colonial Radiolaria, finding
"zoospores" in three species of Collozoimt, two species of Sphaero-
zoiim, and one species of CoUosphaera. He also reported macro- and
microspores (anisospores) in two species CoUozoimi and one species
of Spbaerozoiim, but saw no fusions among the ffagellispores.

In 1905 Brandt found in ThalassicoUa jjucleata (Peripylea) that
some individuals developed only isospores and others only anisospores.
The isospores (Fig. Q, 14 to 17) in life w^re rather wedge-shaped or
spindle-shaped, but always with two flagella, not one, as stated by
Hertwig. The anisospores differed considerably in size, the macro-
spores being 16 to 17 [i- long and the microspores only 8 to 10 [j. long.
Both kinds had rounded ends and slanting furrow^s near the middle.

184

SEX IN MICROORGANISMS

SEX IN PROTOZOA 185

giving tlicin the appearance of certain dinoflagellatcs, like Gymno-
d'ni'mvt (10 to 13). In 1885 he had seen similar spores in Sphaerozoum
pimctiitimi (6 to 9). The nuclei of the two kinds were also different.
He believed that the two kinds of anisospores were produced by the
same parent. They emerged separately, then mingled together. He
never saw any pairing, but he did not put flagellispores from different
jiarents together.

Huth (1913) made an extensive study of ThalassicoUa, mostly
T. spuviida and T. micleata. He described two different series of
developments, the Schlauchkernserie , which led to the formation of
microspores (Fig. Q, 18 to 22), and the Spiudelkernserie, which re-
sulted in macrospores (23). During development of the spindle series
there was a great increase in the size of the animals and the size of
the central capsule. The latter increased approximately a hundred-
fold in volume and the nucleus grew proportionately. During the
later stages the nucleus became lobulated, then broke up into spindle-
shaped areas, each with one or more nuclei. Further nuclear divisions
resulted in large numbers of nuclei, but the spindle-shaped areas dis-
appeared, as did the old nucleus itself. The very numerous small
nuclei became rather uniformly distributed throughout the capsule
and eventually, each with a small amount of cytoplasm, became a
macrospore (23).

In the Schlauchkernsene a similar growth of the animals took
place (Fig. Q, 18 to 20). Groups of small nucleus-like bodies formed
in the large nucleus, these groups being surrounded by a membrane,
giving them a tubular appearance. These tubes grew through the
nuclear membrane out into the surrounding endoplasm (18). As they
grew, they split up (19) and eventually occupied most of the space
in the central capsule (20). In these "tubes" mitoses showing a rela-
tively small number of chromosomes seemed to be continually taking
place (21). Eventually microspores (22) were formed.

Fig. R. Stages of "sporulation" in the radiolarian, Aulacmitha scolymantha,
from Kuhn (1926), after Borgert (1909), redrawn.

1, early stage of dissolution of "primary" nucleus, small chromatin bodies ("sec-
ondary nuclei") appearing in endoplasm; 2, endoplasm full of "secondary" nuclei,
mere remnant of "primary" nucleus; 3, "secondary" nuclei when first seen in
endoplasm; 4, "resting stage" of "secondary" nuclei; 5, mitoses of "secondary"
nuclei; 6, break-up of cytoplasm into "spheres" or various sizes; 7, one of the
"spheres" enlarged to show many nuclei.

185 SEX IN MICROORGANISMS

Borgert (1900) made a study of the division stages of Aula-
cantha scolyviantha (Tripylea). In the mitoses he estimated about
1000 chromosomes. (Haecker, later, in 1906, estimated 1500 to 1600
chromosomes in Castanidhim variable). In 1909 Borgert extended his
study of nuclear division and described a quite different type of
development. In this, chromatin migrated out of the nucleus as small
units (Fig. R, 1, 2); these later became small nuclei which divided
mitotically with about 10 to 12 chromosomes (3 to 5). The central
capsule disappeared, and the cytoplasm became separated into many
"spheres" (6) of various sizes each containing many small nuclei (7).
Development of these nuclei was not followed, but Borgert believed
that the macro- and microspores were gametes and fused to produce
zygotes. No fusions were observed.

It will be recalled that Brandt (1905) described anisospores that
resembled small dinoflagellates. Chatton (1920), in his monograph on
parasitic dinoflagellates, reviewed the literature dealing with sporu-
lation in Radiolaria and concluded that the anisospores were, in most
cases, parasitic dinoflagellates. He noted that anisospores were bean-
or kidney-shaped with a constriction or furrow at the equator from
which two flagella arose, and they had no "crystals." He believed
that the anisospores described by Brandt and the Schlaiichkemserie
flagellispores described by Huth were parasites. The facts that Huth's
Schlauchkernserie began development in the nucleus, then pushed out
into the surrounding endoplasm, and that they were at all times sur-
rounded by their own membrane, and that the contained nuclei ap-
peared to be in continuous division (Fig. Q, 18 to 21) indicated char-
acters corresponding to those of species of Syndinhmt that Chatton
had found parasitic in copepods. Although Chatton was unable to
observe any fusions among the "anisospores" of Syndiniii/n, he and
Biecheler (1936) later observed fusion between dinospores of slightly
different size in Coccidinmm mesnili, another parasitic dinoflagellate
(Fig. C, 9 to 12). Chatton was quite willing to admit that isospore
development and that in Huth's Spindelkernserie were part of the
life history of the Radiolaria.

If Chatton's contentions are correct, we are left without any
sexual phenomena in the Radiolaria in the accounts reviewed up to
this point, since isospores were not thought to be gametes. However,
when we come to the study of Actipylea (Acantharia), the group
with skeletons of strontium sulfate, the story is somewhat different.

SI X IN PROTOZOA 1H7

Schewiakoff (1926), in liis monographic study of "Acanrharia"
collected in the (ailt of Naples, described the development and emis-
sion of large numbers of flagellispores from various species. In one
case he estimated the number produced at one swarming to be 17,000.
These "swarmers" resembled in most respects the isospores described
for other groups by previous authors (Fig. Q, 24, 25). They ranged
in size from 2 [j. to 10 [j.. Certain species produced swarmers with only
one flagellum, others with two unequal flagella, and still others with
two equal flagella. Schewiakoff believed these to be gametes but did
not actually observe fusion. He did, however, find among ordinary
biflagellated swarmers some of double size with four flagella and two
nuclei and others with four flagella and only one large nucleus. This
series (Fig, Q, 26 to 28) was thought to show gamete fusion to form
zygotes. He \vas unable to observe any development of these "zy-
gotes."

x\lthough I have not examined all the papers dealing with the
life history of the Radiolaria, of those reviewed Schew^akoff's ob-
servation of "double" flagellispores appears to be our only direct
evidence for sexuality in the Radiolaria. However, if this interpre-
tation is correct, it may be that the isospores of other Radiolaria may
be gametes and that fusion fails to take place between gametes pro-
duced bv the same parent, as seems to be true in the dinoflagellates.
Surely this whole subject of sexuality in the Radiolaria is in need of
extensive reinvestigation.

CLASS SPOROZOA

The literature on reproduction in this parasitic and spore-bearing
group of Protozoa up to about 1930 has been monographically
treated by Naville (1931), who provides a much more complete re-
view than can be attempted here. Older reviews are those of Biitschli
(1882-89), of Balbiani (1884), and of Labbe (1899).

Kudo segregates the Sporozoa into the subclasses Telosporidia,
Cnidosporidia, and Acnidosporidia. In the Telosporidia and Cnido-
sporidia sexual reproduction is well-known; it usually alternates with
various forms of asexual reproduction. The Subclass Telosporidia
consists of three orders, Gregarinida, Coccidia, and Haemosporidia,
with intermediate groups linking them together.

In the Schizogregarinaria, Coccidia, and Haemosporidia, asexual

188 SEX IN MICROORGANISMS

reproduction is commonly by multiple fission (binary fission in
Babesia). In the Eugregarinina, increase in numbers does not take
place during the trophic period but is provided for by the production
of large numbers of gametes by gamonts encysted in pairs, and also
by multiplication within the oocyst or spore membranes.

Order Gregarmida

In this group, both isogamy and anisogamy have been described,
but in anisogamy there is usually not the extreme divergence between
the gametes found in the Coccidia and Haemosporidia. However, the
male gamete may be relatively slender (Fig. T, 54; Leger and
Duboscq, 1903), and it sometimes has a flagellum (Leger and Du-
boscq, 1909; Goodrich, 1949).

The life history of a gregarine begins when a host swallows one
or more spores, each typically containing eight minute sporozoites.
When released in the host's digestive tract, the sporozoites usually
penetrate into the epithelium; there they may come to rest and begin
to grow, or continue on into the coelome or into some coelomic
organ. If the early stages of development are intracellular, a tropho-
zoite soon outgrows its cellular environment and emerges into the
digestive tract (or other cavity), retaining an anchorage to the epi-
thelium by means of the epimerite or mucron.

In many of the cephaline gregarines, some, at least, of the tro-
phozoites early become detached and join with another one which
may or may not remain attached to the host's tissues (Fig. S, 7, 8).
In this association in pairs, or syzygy, the anterior end of the satellite
(male) is usually attached to the posterior end of the primite (female)

Sometimes two or more satellites may be attached to one primite,
or there may be a chain of individuals attached one behind the other
(Fig. S, 4). In many cases (the so-called solitary species) pairing
does not take place until just before encystment. In some gregarines
attachment is at the anterior end of both members. This may occur in
cephaline species, for example Stylorhynchns ohlongatiis (Fig. T,
49; Leger, 1904), or in acephalines, for example Kalpidorhynchiis
arenicolae (50; Cunningham, 1907). Pairing may also be lateral,
either with "heads" together as in Cystobk chiridotae (51; Dogiel,

SEX IN PRO lOZOA

189

Fig. S. Life cycle of NejTiatopsis ostrearuni from Prytherch (1940), redrawn.

1, escape of sporozoite in intestine of crab; 2, attachment to intestinal epithelium;
3, development of epimerite and septum; 4, precocious syzygy of young trophozoites;
5, reattachment to intestinal wall; 6, more mature trophozoite; 7, syzygy of a pair of
trophozoites; 8, full-grown pair of trophozoites (gamonts), attachment to intestinal
wall; 9, encystment of gamonts forming gametocyst; 10, rupture of gametocyst
releasing clusters of sporozoites (gymnospores); 11, single gymnospore discharged
into sea water, ready to enter an oyster gill; 12, engulfment of gymnospore by
oyster phagocyte, separation of sporozoites; 13-16, growth of sporozoites, each
becoming enclosed in a sporocyst; 17, fully developed spore.

190

SEX IN MICROORGANISMS

1906) or "heads" in opposite directions as in Lecyth'mni thalassemae
(52; Adackinnon and Ray, 1931).

When mature, each pair rounds up and becomes encysted (Fig.
S, 9). In this gametocyst each gamont produces a relatively large
number of gametes after a series of nuclear divisions. In anisogamy
the uniting gametes are from the different parents, and it is assumed
that the same is true in isogamy. Each zygote secretes a sporocyst
or oocyst membrane within which eight sporozoites are commonly
formed.

Both pregametic and postzygotic meiosis have been recorded for

TABLE III

Meiosis in Eugregarines
(Sequences in each group are chronological.)

Diploid-

haploid

Genus and Species

Meiosis

Number

Author

A. Acephalina

Monocystis rostrata

Gametic

8-4

Mulsow, 1911

" agilis

a

8-4

Bastin, 1919

Diplocyslis schncidcri

Zygotic

6-3

Jameson, 1920

Monocystis sp.

Gametic

10-5

Calkins and Bowling,
1926

Urospora lagidis

((

4-2

Naville, 1927a

Monocvslis sp. A

((

8^

1927c

" B

((

8^

1927c

« Q

((

4-2

1927c

" mrazeki

Zygotic

8-4

Hahn, 1929

Monocystella arndti

((

22-11

Valkanov, 1935

Apolocyslis elongata

((

8-4

Phillips and

Mackinnon, 1946

B. Cephalina

Gregarina ovata

Gametic

8-4

Paehler, 1904

a tt

((

8-4

Schnitzler, 1905

Echiiiomcra hispida

ii

10-5

Schellack, 1907

Gregariiia ovata

Zygotic

Schellack, 1912

Stenophora juli

Gametic

Tregouboff, 1914

Monoductiis luiiatiis

Zygotic

4-2

Ray and Chakravartv,
1933

Hyalosporiiia camhalopsisae

u

4-2

Chakravartv, 1935

Zygosoma globosum

(I

12-6

Noble, 1938

Actinoccphalus parvus

11

8-4

Weschenfelder, 1938

Stylocephalns loiigicollis

((

8-4

Grell, 1940

Gregarina blattarum

(I

6-3

Sprague, 1941

SEX IN PROIOZOA ]91

gregarines. Most cases of prcganicric mciosis have been reported for
members of the Acephalina (Table III), while most members of the
Cephalina sliow post/.vtjotic meiosis, especially in more recent reports
(Table 111).

For the acephaline gregarines various authors including iVlulsow
(1911), Calkins and Bowling (1926), and Naville (1927a, c) have
reported pregamctic meiosis, while Jameson (1920) and others de-
scribed post/.vgotic mciosis. Figure T (1 to 48) shows selected illu-
strations from Naville's (1927a) paper on Urospora lagidis. Naville
illustrated mitoses for the last four divisions in gametogenesis. Figure
T shows mitoses for both female (1 to 8) and male (9 to 16) gamonts
for the second stage where four chromosomes divide and separate.
In the third stage (17 to 32), onlv two chromosomes go to each pole
in the anaphases. The fourth stage (not illustrated here) also shows
two chromosomes going to each pole. Thus there are two meiotic
divisions according to Naville. Zygote formation and development
of the eight sporozoites within it are also illustrated (33 to 48). Four
chromosomes appear in the mitoses. The evidence for postzygotic
meiosis offered by Jameson (1920) seems equally clear. Grell (1940)
after a critical review of the literature thought that probably all greg-
arines had postzygotic meiosis.

If it is true that some gregarines exhibit pregametic and others
postzygotic meiosis, then it may make little difference whether the
trophic existence is under the control of a haploid or a diploid set
of chromosomes. Since, however, the more highly evolved cephaline
gregarines (Table III) and Coccidia (Table IV) generally have post-
zygotic meiosis, these groups are apparently under no handicap by
living a haploid existence, the zygote being the only diploid stage in
the life cycle. Why should the more primitive (?) acephaline greg-
arines show both types of cycler If the gregarines and Coccidia
have evolved from the flagellates, as has been suggested by a number
of authors, for example BiitschU (1882-89), Minchin (1912), Alexei-
eff (1912), and Cleveland (1949), one wonders if the diploid and
haploid condition in some gregarines represents a carry-over from
diploid and haploid flagellated ancestors. In the group of flagellates
living in Cryptocerais, both haploid and diploid cycles are found.
This may indicate that neither type of cycle has become predominant
in the flagellates, although such evidence as we have indicates that
free-living flagellates are usually haploid. Since diploidy could arise

192

SEX IN MICROORGANISMS

(® (^.

m

I

Mm\

(m

i^ .o'lP ,;vw' .2v3

m m

15\^ 16

SKX I\ IM<()I()ZOA 193

by the simple process of cndoiiiirosis, or rhc failure of a zygote to
undergo meiosis, perhaps the difference between these two types of
cycle is not of outstanding- importance in the evolution of these

Order Coccidia

In typical Coccidia there is pronounced anisogamy, but in the
Suborder I'.imeridea the macro- and microgametes are generally
formed independently of each other, the small flagellated microgam-
etes swimming to and fertilizing the large non-motile macrogametes.
In the Suborder Adeleidea the gamonts become associated before the
gametes arc produced. Although some of the older accounts of game-
togcnesis in the Coccidia indicated pregametic meiosis, nearly all
recent accounts show postzygotic meiosis, so that the vegetative
stages are haplonts (Table IV).

Figure U shows diagrammatically the life cycle of Adel'ma
derovis, a typical member of the Adeleidea, as described by Hauschka
(1943). The zygote (1) nucleus undergoes a meiotic division (2, 3),
reducing the chromosome number from twenty to ten. A second
di\'ision produces four nuclei, two of which proceed to the next
division in advance of the other two (4). Other nuclear divisions
result in eight to fourteen, usually twelve, nuclei. The cytoplasm then
divides into uninucleated sporoblasts which surround themselves
with sporocysts to produce spores, each with two sporozoites (5, 6).

Fig. T.

1-48, stages in gametogenesis and sporogenesis of Urospora lagidis from Naville
(1927a). 1-8, stages in mitosis of female and 9-16 of male gamonts, second period of
development (note four chromosomes dividing and passing to each pole); in 3,
the eight chromosomes are explained as resulting from a precocious prophasic
separation of chromatids; 17-24, female, 25-32, male, stages in mitosis of third period
of development (two chromosomes pass to each pole); 33-48, development of
zygote and sporozoites; 33, fusion of gamete nuclei in zygote; 34, resting stage of
zygote nucleus; 35-40, first mitosis of zygote; 41, binucleate stage; 42-44, second
mitosis in zygote; 45, tetranucleate condition; 46, 47, eight nuclei following third
mitosis; 48, small abortive zygote. 49, "head" to "head" pairing of Stylorhynchus
oblofigatus, from Leger ( 1904). 50, anterior end pairing of Kalpidorhyncbiis arenicolae,
from Cunningham (1907). 51, lateral pairing, both "heads" together, in Cystobia
chiridotae, from Dogiel (1906). 52, lateral pairing, "heads" toward "tails" of Lecy-
thiufn thalassemae, from Mackinnon and Ray (1931). 53, macrogamete, 54, micro-
gamete, of Pterocepbahts nobilis, from Leger and Duboscq (1903). (All illustrations
redrawn.)

194

SEX IN MICROORGANISMS

16

LIFE CYCLE OF
ADELINA DERONIS

stage

dura

ion in days

7-9 incl.

1

9-11

A

12-16

3

17-20

3-^

1 -3

3

5-6

2-3

totol 16 ~ 18 (Jays
duralion of side-cnain (no. 15) is unolelermined

SEX IN PROTOZOA

TABLE IV

Meiosis in Coccidia

(All these reports show zygotic meiosis.)

195

Species

Meiosis

Diploid-
haploid
Number

Author

Reichenow, 1921

;-io— 4-5

Greiner, 1921

12—6

Dobell, 1925

8—4

Naville, 1927b

10—5

Wedekind, 1927

16—8

Yarwood, 1937

14—7

Mackinnon and Ray, 1937

8 4

Nabih, 1938

20—10

Hauschka, 1943

Karyolysis sp., sp.

Haemogregarinidae
AdeU'o ovahi

Adeleidae
AggrcgiUa ehcrthi

Aggregatidae
Klossia helicis

Adeleidae
Barrouxia sp.

Eimeriidae
Adelina cryptocerci

Adeleidae
Ovivora thalessemae

Aggregatidae
Klossia loosei

Adeleidae
Adelina deronis

Adeleidae

Zygotic

On being taken into the digestive tract of a host, the sporozoites are
released (7) and they penetrate the gut wall (8) into the coelome
where they enter coelomic cells. The first generation of merozoites
(10 to 12) greatly increases the infection. The schizonts of the second
generation of merozoites (13, 14) are already differentiated as to
sex, about twice as many smaller merozoites being produced from a
male schizont than larger merozoites formed from female schizonts.
Young macrogamonts penetrate host cells and are joined by micro-

FiG. U. Diagrammatic outline of life cycle of AdeVma deronis,
from Hauschka (1943).

1, zygote, chromosomes undergoing synapsis; 2, metaphase, meiotic division of
zygote; 3, telophase, first zygotic division; 4, nuclear divisions in zygote, two divid-
ing faster than the other two; 5, sporogony, young spores; 6, fully developed spores;
7, release of sporozoites in host gut lumen; 8, penetration of gut wall by sporozoites;
9, 10, young schizonts; 11, 12, merozoites, first generation; 13, schizonts, second
generation; 14, 16, merozoites, second generation; 15, aberrant form of schizogony; 17,
young gamonts in host cell; 18, nearly full-grown gamonts; 19, nuclear division in
microgamont; 20, one microgamete penetrating macrogamete.

196

SEX IN MICROORGANISMS

Fig. V.

1-18, outline of life cycle of Aggregata ebertbi, from Dobell (1925). 1, fertilized
zygote; 2, first nuclear division in zygote; 3, later nuclear divisions in zygote; 4,
formation of sporoblasts; 5, mature spore; 6, release of sporozoites in digestive tract
of crab; 7-10, growth of schizont and formation of merozoites; 11, 12, passage of
merozoites to cuttlefish; 13-15, development of microgametes; 16-18, development of
macrogamete; 18, fertilization. 19-26, sporulation of oocyst of Chagasella sp., from
Gibbs (1944). 19, first nuclear division in young oocyst; 20, binucleate stage in
oocyst; 21, one nucleus divides; 22, the three nuclei divide as cytoplasm divides into
three parts; 23, three young spores; 24, 25, further development of spores; 26,
mature spore with four sporozoites. (All illustrations redrawn.)

SEX IN PROTOZOA 197

gamonts (17). Both increase in size (18) until rhcy arc mature; then
four microganictes are produced by the microgamont (19). One of
these fertihzcs the niacroganiete (20) to produce the zygote (1). In
the Kinieridca, ganionts develop independently, as in Air^regata
chert hi (Fig. \'/l2 to 18).

Order Haemosporidia

Typical members of this group, as found in the Plasmodiidae and
Haemoproteidae, have life cycles similar to those of the Eimeridea
among the Coccidia, except that two hosts are involved and sporocysts
are not formed. In vertebrate hosts asexual reproduction by schi-
zogony takes place, and gametocytes (gamonts) are formed which do
not proceed further in their development until taken into the digestive
tract of the proper blood-sucking invertebrate host, in which sexual
reproduction and sporozoite formation take place. In the Babesiidae,
or at least in the genus Babesia, asexual reproduction may be by bin-
ary fission instead of multiple fission, and the sexual stages occur in
ticks. There does not seem to be any satisfactory account of meiosis
or chromosome cycles in the Haemosporidia.

Subclass Cnidosporidia

In this subclass the sporozoites (sporoplasms) are amoeboid, as
are often the trophic stages; thus the group shows affinities with the
Sarcodina. The spores contain polar capsules in each of which there
is a coiled filament which is discharged upon appropriate stimulation.
In some of the Adicrosporidia and in the Helicosporidia the coiled
filament is not enclosed in a capsule. Kudo divides the Cnidosporidia
into the orders iMyxosporidia, Actinomyxidia, Microsporidia, and
Helicosporidia.

Order Myxosporidia

Figure W provides a diagrammatic representation of the life
cycle of Ceratoviyxa blemmis (Noble, 1941). A binucleate sporo-
plasm (1) emerges from the spore which has entered the digestive
tract of a new individual host {Hypsoblennius gilberti). Its haploid
nuclei fuse and the zygote (2) migrates to the gall bladder of the

198

SEX IN MICROORGANISMS

Fig. W. Life cycle of Ceratoviyxa blennius, from Noble (1941), redrawn.

1, zygote, released from sporocyst in gut of host {Hypsoblemmts gilberti); 2-5,
growth accompanied by nuclear divisions with four chromosomes; 6-10, further
growth and asexual reproduction; 11-17, sporognoy, each pansporoblast (11-13)
producing two spores (14-16); 14, meiosis, producing two nuclei each with two
chromosomes.

SFX IN PROroZOA

199

host, wlicrc rlic rest of the development takes phice. Cirowth (2 to
5 ) is accompanied by nuclear divisions showing four chromosomes
(3). Further grow tli mav he accompanied by asexual reproduction
(6 to 10) or the production of spores (11 to 17). Kach trophic
individual becomes a pansporoblast (11 to 14) and produces two
spores (15, 16). All nuclear divisions show four chromosomes except
that (14), in which meiosis takes place, producing the gamete nuclei,
each with two chromosomes. The two haploid nuclei, show^n with
marginal endosomes, remain in the sporoplasm (15 to 17) while the
other nuclei, belonging to the "somatic" cells which produce the
valves of the sporocyst and the polar capules, eventually disappear.

Fig. X.

1-4, diagrammatic sequences of mitoses within a single sporoblast of a myxosporidian
(cf. Fig. W), from Noble (1944). iMeiosis in one nucleus at third division. 5-9,
nuclear history in sporogony of Thelohania legeri, from Kudo (1924a), including
nuclear fusion (9). 10-13, stages in sporogony of Haplosporidmvi IhnnodriU, from
Granata (1914); 10, gametes; 11, 12, fusion of gametes; 13, resulting spores. (All
illustrations redrawn.)

200 SEX IN MICROORGANISMS

Figure X (1 to 4) shows diagrammatically the sequence of
mitoses within a single sporoblast, according to Noble (1944). At
the end of nuclear division in each sporoblast there are six nuclei,
two of which are sister haploid nuclei, the other four belonging to
the somatic cells as explained above. George vitch (1935, 1936)
reached a similar conclusion in regard to the life cycles of Myxospori-
dia in general, with the exception that he believed that meiosis takes
place in two nuclei during the final nuclear divisions, one haploid
product of each division degenerating like a polar body. If the inter-
pretations of Georgevitch and Noble truly represent the life cycles
of the Myxosporidia, then the cycles are relatively simple. However,
there is much disagreement in the extensive literature on the group,
but only one more example will be cited.

Naville has contributed a series of reports on this group besides
providing a general review in his monograph (1931). In his study
oi Sphaeromyxa sabrazesi (1930a) along with other Myxosporidia, he
described meiosis in cells segregated from the general protoplasm of
the trophozoite. There are two kinds, larger (female) and smaller
(male). By two meiotic divisions, the larger cell produces two "polar
bodies" and the smaller gives rise to four equal small haploid cells.
A smaller haploid cell unites with a larger one without nuclear fusion
to constitute a pansporoblast which produces two spores. Thus all
nuclei in the sporoblast are haploid according to this interpretation.
In other respects the process of spore formation is similar to that of
Ceratomyxa blennius (Fig. W.) More detailed studies are needed to
reevaluate these more complicated life cycles, described by many
authors, in the light of the interpretations of Georgevitch (1935,
1936) and of Noble (1944).

Order Actinoiny xidia

The spores of members of this order are somewhat like those
of the Myxosporidia but have ternary symmetry with three sporocyst
valves and three polar capsules. There are a number of descriptions
of life cycles, but only one will be mentioned. According to Naville
(1930b), in Giiyenotia sphaeriilosa early trophic stages are always
binucleate but the origin of the nuclei is obscure. After growth and
nuclear division, two large rounded cells are surrounded by a proto-
plasmic membrane containing two nuclei, which may increase to

SEX IN PROTOZOA 201

four. Tlic two larger cells are male and female gamonts each of which
is destined to give rise to eight gametes, the male line developing
faster than the female. The first division of the gamont is "gonial,"
sho\\ing the diploid number of four chromosomes. During the sec-
ond division, which is heterotypic, reduction to two chromosomes
takes place. The third di\'ision is homeotypic, again, with hut two
chromosomes. 1 hus eioht marcrogamctes and eight slightly smaller
microgametes are formed. These fuse in pairs to produce eight zy-
gotes, each of which becomes a sporont and develops into a spore.

Order Microsporidia

In this group the spores are small, and in the majority there is
only one polar capsule or polar filament. The small size of many of
these organisms makes the study of their development difficult. Most
accounts in the literature are incomplete.

In Thelohania legeri (Kudo, 1924a) young sporonts develop
as shown in Fig. X (5 to 9). A quadrinucleate stage (5), formed
after schizogony, divides into two binucleate cells (6). These undergo
division with both nuclei dividing at the same time (7, 8). In the
four binucleate cells so produced (8), nuclear fusion takes place
(9). The uninucleate sporonts thus formed proceed to the formation
of spores. In several other accounts Kudo has described autogamy in
connection with the development of the sporont, for example in
Thelohania opacita (1924b), Sternpella (Thelohania) inagjia (1925),
and Nosema aedis (1930), all parasitic in mosquitoes. Aleiosis is not
clearly indicated.

Debaiseaux (1928) postulated two somewhat different patterns
of development, one for the polysporous and the other for the mono-
sporous Microsporidia. For the former, vegetative development may
be by binary fission or by plasmodium formation, but eventually a
special binucleate stage is reached. This divides by a peculiar type
of division to produce two binucleate cells within each of which
nuclear fusion takes place to produce a zygote. This becomes a spo-
ront and forms a spore (see Fig. X, 5 to 9). For the monosporous
forms a similar vegetative development ends likewise in a special
binucleate stage which undergoes the special type of division pro-
ducing two binucleate cells. In these cells nuclear fusion does not take
place until after spore formation. In Plistophora chironoiiii, Weiser

202 SEX IN MICROORGANISMS

(1943) found a diploid number of about twelve chromosomes with a
meiotic division preceding sporogony. Just when fusion of gametes
occurred was not determined. Much more study is needed to elucidate
the problems of syngamy in the Microsporidia.

Order Helicosporidia

This order consists of one species, HeUcosporidiiim parasiticmn,
the life cycle of which has not been fully worked out.

Subclass Acnidosporidia

In this subclass, as presented by Kudo, there are two orders, the
Sarcosporidia and the Haplosporidia. No sexual stages have been
determined for the Sarcosporidia.

Order Haplosporidia

This group shows some similarities to certain members of the
Microsporidia, but the spores are simple, not having any polar cap-
sules or polar filaments. Accounts of the life cycles are usually incom-
plete. For Ichthyosporidiimj giganteim?, Swarczewsky (1914) de-
scribed a life cycle which included meiosis and fertilization within
the sporobkist during development of the spores. This would presum-
ably involve pregametic meiosis. A similar life cycle is claimed by
the same author for Ichthy osporidiinn herfwigi and for Pleistophora
periplanetae. Granata (1914) has described gametogenesis in Haplo-
sporidiimi Uvmodrili. The gamete fusion stages are shown in Fig. X
(10 to 13). Gametes are formed (10) and fuse (11, 12), producing
zygotes which become spores (13). Although these life histories are
not so complete as desired, at least they show evidence for the forma-
tion and fusion of gametes.

Comments on Sexuality in the Sporozoa

Certain topics in connection with syngamy in the Sporozoa merit
separate discussion. Two of these are (1) sex determination and sex
differentiation, and (2) the effect of the host on life cycles.

SEX IN PRorozoA 203

6V.V Dctcnniihitioii ii/ni Sex Di^Jcrcnt'hnioii

Tclosporid'ia. 1 here is a considerable body of literature which
tends to show that sex determination in the I elosporidia takes place
soon after zygote formation, possibly during the first zygotic nuclear
division. Joyet-Lavergne's extensive publications show cytoplasmic
differences between male and female trophozoites and gamonts among
the I'ugregarinida, and especially among the Cephalina. It will be
recalled that in this group each zygote commonly gives rise to eight
sporozoites, each of which may grow into a gamont without further
division.

Using neutral red, Aliihl (1921) showed that, for Gregarina
cuneatu and G. polyiitorpha from the larvae of Tenebrio molitor, the
two members of a pair in syzygy stained differently, although there
was much variation when pairs were compared to each other. Joyet-
Lavergne (1926) obtained similar results with the same species and
\\ith Stein'mx oralis. Other vital dyes, which color the cytoplasm
directly, such as methyl blue, cresyl blue, Nile blue, and dahlia violet
were used. W \x\\ these the anterior member of the pair, the primite
(female), stained more intensely than the posterior member, the
satellite (male). Tests for pH showed little or no differences between
the sexes, a point confirmed by Gohre (1943). Tests for golgi appara-
tus and for fats also showed differences in the cytoplasm of the two
sexes. Sex differences in the gamonts either before or after encystment
were show'n by these various techniques when the gamonts themselves
were morphologically indistinguishable and when similar gametes
(isogametes) were produced by the gamonts. As Gohre (1943)
remarked, sex differences in gregarines become apparent very early,
and are probably genotypic in origin. Literature in this field up to
1930 is reviewed by Joyet-Lavergne (1931) in his book.

Certain conditions in the Coccidia are suggestive. In Adelina
deroiiis Hauschka (1943) found evidence of differential behavior
between the four nuclei derived from the zygote nucleus (Fig. U,
4), two nuclei proceeding to the next division faster than the other
two. This difference in division rate resulted in about twelve spo-
roblasts and tw^elve spores (6). It is possible that eight of these are
male and four female. In addition, Hauschka could recognize sex dif-
ferences in the second generation schizonts and their merozoites (14).

204 SEX IN MICROORGANISMS

There was also evidence that a single parasite, passed along from
parent to daughter during fission of the parent, was unable to carry
on the life cycle; that is, any individual parasite is male or female, but
not both. However, if a single spore containing two sporozoites can
give rise to both sexes, then the two sporozoites should be of different
sex. Otherwise a male spore and a female spore would have to be
taken into a new host to produce male and female gametes.

In the case of Chagasella sp., Gibbs (1944) showed that after the
first division of the zygote nucleus in the oocyst (Fig. V, 19, 20)
only one of the two daughter nuclei divided (21). This differential
division, by analogy with Adelina deronis, could indicate maleness for
the nucleus that divided and femaleness for the one that did not. Thus
sex determination would be accomplished at the first nuclear division
in the zygote. During further development the cytoplasm of the
oocyst divided into three parts while the nuclei also divided, produc-
ing three sporoblasts which became spores (22-23). Each spore
produced four sporozoites (24-26). According to the suggested
interpretation, two of the spores would be male and one female.
However, Gibbs stated that the oocyst membrane gradually disap-
peared so that the saHvary glands of the host (a hemipteran insect)
often contained many single spores. This suggests that the spore is
the infective stage.

The so-called spore of a gregarine, derived as it is directly from
a zygote, is equivalent to the oocyst of the Coccidia. It is interesting
that in gregarines generally, and in many of the Coccidia, the oocyst
is the infective stage. This should contain both sexes. When spores
develop in the oocyst, as in so many Coccidia, they might well be
male or female, but not both. Becker (1934) was able to maintain
infections of Eivieria in rats by inoculating a new host with a single
oocyst. Experiments are needed to see if a single spore of Emieria
(there are four in each oocyst) would be capable of maintaining the
life cycle if transferred alone to a new host. Why should the spore
not be an infective stage?

Perhaps nature offers such an experiment in the case of Aggre-
gata eberthi, a coccidian that requires two hosts to complete the life
cycle (Fig. V, 1 to 18). According to Dobell (1925) this species
undergoes gametogony, fertilization, and sporogony in the cuttlefish.
Sepia officinalis, while schizogony takes place in the crab, Portunus
depurator. The zygote (1), formed in the cuttle fish, does not pro-

SEX IN PROTOZOA 205

diice a resistant oocyst membrane as does that of most Coccidia. By
numerous divisions many nuclei are formed from the zygote nucleus,
the number varying with the size of tlic zygote. The cytoplasm then
divides into as many sporoblasts as there are nuclei (4). Sporocysts
next form about the sporoblasts which produce spores. Within the
spore cytoplasm the single nucleus divides, then only one of the
daughter nuclei divides, just as occurs in the oocyst of Chagasella
(21). Therefore, only three sporozoites are formed in each spore (5).
Since the spores are not confined by an oocyst membrane, they are
free to separate from each other. One would suppose that a single
spore would be able to continue the life cycle. When taken into a
crab, the schizogonic part of the cycle develops. When a cuttlefish
eats a crab, the merozoites from the crab continue the cycle in the
cuttlefish by developing into gamonts and gametes, which unite to
produce zygotes again (12 to 18). By analogy to Chagasella (19 to
26), we might suppose that sex determination takes place in the spore
in this case, rather than at the first postzygotic nuclear division. Pos-
sibly two of the three sporozoites in a spore are male and the other
female. If so, any spore could continue the life cycle in a new host.

If we could assume that any spore with two or more sporozoites
contains both male and female elements, then most of the spores of
gregarines and Coccidia would contain both sexes. However, in the
life cycle of Neviatopsis ostrearum (Prytherch, 1940), each spore
contains only one sporozoite (Fig. S, 17). To produce an infection
with new gamonts, such a sporozoite would have to be able to give
rise to both male and female elements; otherwise male spores and
female spores would both have to enter a new host in order to con-
tinue the life cycle. Perhaps the fact that sporozoites enter a new host
oyster in clusters (Fig. S, 11) would help to insure that both sexes
would be present in an infected oyster. We are thus faced with the
interesting problem of the mechanism of sex determination in this
group of parasites.

In typical Haemosporidia, as a rule, no sex difi^erences are visible
until the gametocyte (gamont) stage is reached. The gametocytes,
however, are distinguishable as to sex, and the gametes are strikingly
different, as in the parasite of malarial fever for example. Presumably
a mosquito must take up both kinds of gametocytes in order for fer-
tilization to take place in its digestive tract. After sporozoites are
formed and migrate to the salivary gland of the mosquito, they are

206 SEX IN MICROORGANISMS

ready to infect a new host. One wonders if a single sporozoite could
give rise to both kinds of gametocytes in a new vertebrate host. In the
case of Babesia (Dennis, 1932) the gametocytes are not morphologi-
cally differentiated in the vertebrate host. Just where and how sex
determination takes place in the Haemosporidia can only be guessed
at, but one might suppose that the mechanism would be like that of
the Coccidia.

Cnidosporidia. In the Myxosporidia the available evidence indi-
cates that the trophic stages are commonly diploid and that meiosis
takes place at some stage during sporogony. However, according to
Naville (1930a), meiosis takes place at a very early stage in sporogony
with evident differences between the sexes, whereas according to
Georgevitch (1935, 1936) and Noble (1944) all nuclei of a myxospo-
ridian are diploid except the gamete nuclei, which are formed at the j
last nuclear division during spore formation. Since these nuclei seem '
to be exactly alike (Fig. W, 15 to 17), there is no sex difference, and
therefore no sex determination. Furthermore, the interpretations of
Noble indicate that the two gamete nuclei, as sister nuclei, unite to
produce a diploid zygote in a process of autogamy. Autogamy is
likewise indicated for the Actinomyxidia, Microsporidia, and Hap-
losporidia as previously indicated. Just what biological advantages
are derived from such a process is difficult to imagine.

The Effect of the Host on Life Cycles

It is interesting to speculate about what determines the shift
from one stage of a life cycle to the next. In gregarines, for instance,
what induces gametocyst formation? In Coccidia, what causes the
cessation of schizogony and the onset of gametogony? In the Hae-
mosporidia, what determines that certain merozoites will become
gametocytes (gamonts) instead of schizonts? Are such changes of
activity due to some inherent cyclic tendencies, or do they occur in
response to conditions or stimuli provided by the host?

Cleveland's remarkable discovery of the inducement of sexuality
in the intestinal flagellates of the wood-feeding roach, Cryptocercus
pimctiilatus, by the molting hormone of the host naturally raises the
question whether parallels can be found elsewhere. Possibly similar
conditions have been described where different stages in the life cycle
are correlated with developmental stages of the host. For example,

SI X IN PROTOZOA 207

Nowlin (1922) reported that ;i grcgarinc {Schneideria sp.), found
in the flv Sc'huct copro^ihila goc^ no further in its development in the
hirval stage of the host than the fullv^ grown trophic stage (gamont);
in the pupa, the ganionts unite in pairs, become encysted, and begin
sporulation. In the adult, sporulation is completed. Whether any of
these changes in the life cycle are influenced by molting or other
hormones of the host is, of course, unknown at the present time, but
the correlation may have some well-defined host-relationship basis.
Shortt and Swaminath (1927) reported a somewhat similar relation-
ship between Movocystis inackki and another dipteran host, Fhle-
hotovnis argan'ipes. In this case sporulation did not begin until the
host became adult.

As shown by Wenyon (1911) and later by Ganapati and Tate
(1949), the development of the gregarine Lankesteria czilicis in mos-
quitoes shows similarities to that of the species cited above. Trophic
development takes place in the gut of the larva, and union of gamonts,
cyst formation, and sporogony take place in the malpighian tubules
of the pupa. Only spores are found in the adult.

Hentschel (1926) reported a correlation between the develop-
ment of the acephaline gregarine Gonospora varia, with the sexual
development of its host, the polychete worm, Andoninia (Terraw-
his) teinaciilata. The coelomic parasites live among the developing
genital products of the host and complete their life cycle at the time
of sexual maturity of the worm. Occasionally gregarines are present
in segments which do not contain gonads. In these segments the
parasites are unable to complete their full life cycle, remaining small
and apparently unable to produce spores. Hentschel suggested that
the host gonads may produce a hormone or other substance which is
essential to the growth of the gregarines. The same author (1930)
reported a somew^hat similar relationship between Gonospora areni-
colae and its host, Arenicola ecaudata, except that the correlation
was not so close as for G. varia and Audouinia tentacidata.

For those gregarines which have two hosts, as do members of
the family Porosporidae (including Neviatopsis, Fig. S), there is a
correlation between stages in the life cycle and the change of hosts;
this is true in other cases of two-host development, as in the coccidian
Families Aggregatidae and Haemogregarinidae and in the Order
Haemosporidia. It is fair to assume that, when any transmissible stage
in the cycle leaves one of the alternative hosts, transfer to the other

208 SEX IN MICROORGANISMS

host constitutes a necessary stimulus to further development. Yet the
well-known development of micro- and macrogametes of Haemos-
poridia from warm-blooded hosts when blood containing the game-
tocytes is drawn and reduced to "room" temperature suggests that
mere reduction in temperature induces gamete formation. However,
we have the definite circumstance, in these two-host parasites, that
transfer to an alternate host is required for the life cycle to continue.
Whether the host influences are in the nature of "hormones" or other
specific substances is yet to be determined. It is possible that the rela-
tion between the molting hormone of the wood-feeding roach and
the sexual activities of its contained flagellates is unique.

SUBPHYLUM CILIOPHORA

The Subphylum Ciliophora is divided into two classes, the
Cihata and the Suctoria. Class Ciliata contains Subclass Protociliata, in
which there are two or more similar nuclei in each animal, and Sub-
class Euciliata, in which each animal has one or more macronuclei
and one or more micronuclei. All the Protociliata are endozoic, mostly
in the large intestine of Amphibia. Sexual phenomena in cihates have
been reviewed by a number of writers, more recently by Diller
(1940a), Turner (1941), and Finley (1946).

Subclass Protociliata

According to Neresheimer (1907), the life cycle of Opalina
ranarum and O. dimidiata includes binary fission through most of
the year. In the spring, cysts are formed which pass into the water
of breeding pools to be taken up by a new generation of tadpoles,
in which sexual reproduction takes place. Similar uninucleate gametes
fuse in pairs to produce zygotes which become encysted. When the
zygotes excyst, they develop into adults in the new host. Metcalf
(1908) confirmed most of this life cycle for Opalina (Protoopalina)
caudata, O. (P.) intestinaUs and O. dimidiata, except that he found
the gametes to be different in size and failed to confirm the encyst-
ment of the zygote. Valkanov (1934) also found anisogamy. Figure
Y shows diagrammatically the life cycle of Opalina rananim as
illustrated by Prenant (1935) and based on the work of Zeller (1877)
and KonsulofF (1921). None of the authors mentioned above gave

SEX IN PROTOZOA

209

Fig. Y. Life cvcle of Opalina ranarum, from Prenant (1935), after Zeller
(1877) and Konsuloff (1921), redrawn.

1, encysted zygote in gut of tadpole; 2-6, development of zygote into multinucleate
adult; 7, binary fission of adult; 8, rapid fission leading to precystic stages; 9, cyst
eliminated into water to be ingested by tadpoles; 10, mononucleate gametes from
excysted individuals; 11, fusion of gametes to produce zygote.

satisfactory accounts of chromosome numbers, but Chen (1936, 1948)
showed that adults have a diploid set of paired chromosomes. Meiosis
has not been described.

Subclass Euciliata

Kudo divides this subclass into the Orders Holotricha, Spiro-
tricha, Chonotricha, and Peritricha. In these groups, as well as in the
Suctoria, sexual reproduction, with rare exceptions, takes the form
of conjugation, or some modification of that process. While there are
many variations in conjugation, these are not generally associated
with any particular order, except for the Chonotricha and Peritricha.
Consequently it seems best to depart from the strictly taxonomic

210 SEX IN MICROORGANISMS

sequences as previously followed. One may find about as many varia-
tions within the orders Holotricha and Spirotricha as there are be-
tween them.

In the process of conjugation, two individuals become attached
to each other and exchange gamete nuclei, after which they separate
and a new nuclear apparatus arises from the products of the fusion
nucleus, or syncaryon, in each exconjugant, while the old macronu-
leus gradually disappears.

Conjugation in Farainecimn caudatum

Conjugation as it occurs in Faramecium caudatum is usually con-
sidered typical. Accounts have been published by a number of
authors, for example by Gruber (1887) (under the name F. aurelia),
Maupas (1889), Calkins and Cull (1907), Dehorne (1920), Miiller
(1932), Penn (1937), Diller (1940b, 1950a) and Wichterman
(1953). The principal steps are shown diagrammatically in Fig. Z,
which is based on the accounts of Maupas and later authors.

Two individuals become attached to each other along their oral
surfaces ( 1 ) ; then the micronucleus in each, after a long period of
preparation involving the "crescent" stage (not illustrated), undergoes
the first pregametic (meiotic) division (2). The two nuclei thus
formed quickly go through the second pregametic (meiotic) division
to produce four haploid nuclei (3). Three of these nuclei begin to
degenerate while the fourth divides to form the pronuclei (4). The
pronuclei usually take a position near the oral region, where a cone-
shaped protrusion, the paroral cone (Diller, 1936), extends from
each conjugant toward the other (4, 5). Through these cones the
migratory pronucleus (male) of each conjugant passes into the mate
(5), where it fuses with the stationary pronucleus (female) to form
the fusion nucleus or syncaryon (6). The conjugants then separate
(7), and the further changes take place in the exconjugants. In each
exconjugant the macronucleus gradually becomes transformed into
a skein (7 to 9) which breaks up into small fragments (10, 11). Mean-
while, by three successive mitoses the syncaryon gives rise to eight
small nuclei (7 to 9). Four of these begin to grow and are known
as macronuclear anlagen. As growth of the anlagen proceeds, three
of the other nuclei grow slightly, then begin to degenerate and even-
tually disappear, while the eighth nucleus becomes the functional mi-

Fig. Z. Diagrammatic outline of conjugation in Paraiuechan caudatwn,

original.

1, pair of conjugants become attached along oral region; 2, first pregametic micro-
nuclear division; 3, second pregametic division; 4, degeneration of three products of
second division, fourth product dividing to form pronuclei, migratory pronuclei in
paroral cone; 5, passage of migratory pronuclei to other conjugant; 6, fusion of
pronuclei to form svncarya; 7, separation of conjugants, division of syncarya, begin-
ning of resolution of macronuclei; 8 second postzygotic nuclear division, early
skein stage of macronuclei; 9, third postzygotic nuclear division, late skein stage
of macronuclei; 10, in each exconjugant four products of syncaryon become macro-
nuclear anlagen, degeneration of three products, persistence of eighth product as
functional micronucleus; 11, first postconjugation cell division, two anlagen going
to each daughter; 12, second postconjugation division, each cell has a new macronu-
clcus and a new micronucleus derived from the syncaryon.

211

212

SEX IN MICROORGANISMS

SEX IN PROTOZOA 213

cronuclcus (10). Two cell divisions, each accompanied by a division
of rhc iiiicromiclcus, scr\c to segregate the four niacronuclear anlagen
of each exconjuganr into four descendants (11, 12). During these
changes the fragments of the old macronucleus gradually become
absorbed while the anlagen attain full size. At the end of these two
divisions, there are four new cells derived from each conjugant. Each
of these cells has a new macronucleus and niicronucleus derived from
the svncaryon (12),

Beginning with Biitschli (1876) and Maupas (1889), students of
conjugation have noted deviations from the typical series of stages.
Diller (1940b) has called attention to a number of the variations for
P. caiidatiim. One nucleus may degenerate at the end of the first pre-
gametic division, as happens regularly in P. biirsaria. Instead of three
products of the second pregametic division degenerating, tw o, three,
or four may begin the third pregametic division, although only one
normally completes it to form pronuclei. Among the products of the
syncaryon there may be various numbers of nuclei, up to sixteen,
with varying fates for the supernumerary nuclei. When the usual
eight products are produced, as a rule only four become macronuclear
anlagen, but as many as seven may do so. More recently Diller
(1950a) has reported an extra postzygotic division in the exconjugant
of F. cmidanmi. After the first division one of the daughter nuclei
degenerates, the other proceeding to the second division. Two more
divisions are necessary to produce the normal eight products derived
from the syncaryon.

Reconjugation between normal individuals and exconjugants has
been noted by many observers; the subject is reviewed by Diller
(1942), who showed many examples for P. caudatuvi. Figure AA
(2 to 4) shows three such reconjugating pairs. As pointed out by
Diller, other observers have reported this phenomenon, for example
Biitschh (1876) for P. piitmiimi (?), Doflein (1907) and Klitzke
(1914) for P. caiidatimt, Miiller (1932) for P. nmltimicronucleatum,
and Sonneborn (1936) for P. aurelia. This process has also been de-
scribed for other ciliates, for example by Enriques (1908) for

Fig. AA. Reconjugation in P. caudatum, after Diller (1942), redrawn.

1, normal exconjugant with four anlagen and four smaller nuclei, three of which
are degenerating; 2, 3, exconjugants united with normal animals, all showing crescent
stages indicating recent association; 4, normal with two dividing micronuclei united
with exconjugant having eight small dividing nuclei, four anlagen present.

214 SEX IN MICROORGANISMS

Chilodonella uncinatus, and by Collin (1912) for various suctorians.
Wichterman (1939, 1940) placed conjugating pairs of P. cauda-
tmn under a precision microcompressor which enabled him to observe
nuclear phenomena in living animals. Under these conditions he dis-
covered that in each conjugant self-fertilization (autogamy) took
place instead of cross fertilization. He called this process "cytogamy."
Cross fertilization probably occurs in most instances of conjugating
pairs, as shown by Diller (1950b).

Conjugation in Other Species of Parajneciujn

Different species of Paramecium show varying degrees of diver-
gence from the account just given for P. caudamm. P. biirsaria and
P. trichhim are like P. caudatnm in that each has a single, rather large
micronucleus, whereas P. aiirelia and P. calk'msi usually have two
smaller, more vesicular micronuclei, and P. imiltimicronucleatum,
P. njooodnifji, and P. poly car ywn commonly have four micronuclei of
the P. ajirelia type.

Although P. biirsaria has a micronucleus much like that of P.
caudatwn, the stages of conjugation are somewhat different. For
accounts of conjugation in this species we may refer to those of
Maupas (1889), Hamburger (1904), Chen (1940a, b, c, d; 1946a, b;
1951a, b) and Wichterman (1948a, 1953). According to these authors
each of the first two pregametic divisions are followed by degenera-
tion of one of the products. The remaining haploid nucleus divides
to provide the two pronuclei. According to Maupas and Hamburger,
after exchange of pronuclei and syncaryon formation, two divisions
provide four nuclei, two of which become macronuclei and two
micronuclei. One cell division segregates these into two animals each
with one nucleus of each kind. However, Chen (1946b, 1951a,b)
and Wichterman (1948a) found that after the first postzygotic nu-
clear division one nucleus degenerates, so that it takes two more
divisions to produce the four products which develop into two macro-
and two micronuclei. In this species the old macronucleus does not
fragment but undergoes gradual absorption in the exconjugant.

All these authors have recorded variations in the conjugation of
this species. Hamburger (1904) recorded passage of a pronucleus
from one conjugant to the other without receiving a pronucleus
from the partner, leading to one "haploid" and one "triploid" indi-

SI X IN PROTOZOA 215

\ idiiiil, a condition later reported l)\' (^hen (I94()a, 1), d) and W'icliter-
man (1946). Ilainbur^cr also found one conjugant with no micro-
nucleus, as did Chen (]94()a, b, d) later (Fig. AB, 1). Chen showed
that the aniicronucleate animal could receive a pronucleus (hcmi-
caryon) from its partner which would retain a "haploid" nucleus.
Development of such nuclei could be called parthenogenesis, Maupas
found an exconjugant with three macronuclear anlagen and three
micronuclei, and another with four anlagen and four micronuclei,
possibly the result of the failure of one nucleus to degenerate after
the first postzygotic division, and Hamburger found ten micronuclei
and six macronuclear anlagen in an exconjugant, indicating sixteen
products from the syncaryon.

Chen (1940a, b, d; 1951b) called attention to the great variation
in nuclear size, chromatin content, and chromosome numbers in
P. biirsctria, which indicates a high degree of polyploidy; and he
showed that animals with such different nuclei apparently conjugate
successfully. He (1940c; 1946a) also described cases of conjugation
between three animals. Usually, two assume the common parallel
position with oral areas together, while the third individual has its
anterior end attached to the posterior end of one member of the
pair. The two in the normal position undergo normal conjugation
while the other individual undergoes autogamy (Fig. AB, 2). Chen
(1951a) also reported that, when a double monster consisting of
incompletely separated daughter animals conjugated with normal
specimens, many kinds of association occurred but the most typical
was for a normal animal to conjugate with the anterior member of
such a tandem pair (Fig. AB, 3). In this case the sequence of events
was essentially the same as when three animals conjugated. The ante-
rior pair conjugated normally while the posterior member of the
tandem pair underwent autogamy. In a study of conjugation between
members belonging to two different varieties, Chen (1946b) found
that during the first 16 hours of association nuclear phenomena were
normal, but after that time various abnormalities appeared and the
participants failed to survive.

In another study Chen (1951b) determined the results of con-
jugation between an "old" strain which had been maintained in the
laboratory for about 20 years and a "young" strain which had been
cultured for about a year. In Fig. AB (4), it will be seen that the
numbers, sizes, and shapes of chromosomes are different in the two

216

SEX IN MICROORGANISMS

SEX IN PROTOZOA 217

conjugants. Conjugation appeared to be normal, but about 50 per
cent showed variations in nuclear number in the exconjugants, the
range being from 2 to 20. In many cases the "haploid" nuclei either
with or without exchange dc\eloped parthcnogenetically. Jennings
(1944) had shown that conjugation between an "old" and a "young"
clone or between two "old" clones resulted in high mortality among
exconjugants. Chen's study reveals some of the cytological conditions
in such crosses.

It may be noted in passing that conjugation of more than two
animals has frequently been reported in the literature. For example,
Maupas (1889) figured three individuals of Loxophylhivi fasciola
and three of L. obtusziin in conjugation, and four individuals of Spi-
rostomom teres. Recently Weisz (1950) reported multiple conjuga-
tion in BlepharisDia undulaus^ involving often three, occasionally
four, and rarely five individuals. Nuclear changes took place in all
associates, and autogamy was considered probable for some of the
individuals. EUiott and Nanney (1952) also reported frequent con-
jugation of three animals in Tetrcihyvieiin sp. The three animals were
all attached together in normal positions, and the evidence indicated
that tripolar fertilization took place.

Paramecium trichium is the smallest species in the genus and
normally has a single micronucleus of the F. caiidatinn type. Diller
(1948) reported many variations in the conjugation of this species
(Fig. AC). Most commonly the progress of events followed the
scheme already indicated for P. caiidatiim, except for the absence of
a crescent stage preceding the first pregametic division (Fig. AC,
column a). As shown in (a, 2), one nucleus may degenerate after
the first division, as regularly occurs in F. bwsaria. Autogamy (cy-
togamy, Wichterman) may take place instead of cross fertilization
(column b). After degeneration of one of the two products of the
first pregametic division, either cross fertilization (column c) or

Fig. AB. Stages in conjugation of P. bursaria, redrawn.

1, conjugation between normal and amicronucleate individual, from Chen (1940b);

2, conjugation of three animals, posterior animal undergoing autogamy, from Chen
(1946a); 3, conjugation between a double monster and a normal animal, all in syn-
caryon stage, note differences in chromosome picture of nuclei in the anterior pair,
indicating exchange, attached animal at left undergoing autogamy, from Chen
(1951a); 4, conjugation between old and young clones, prophase of first micronuclear
division showing differences in chromosomes, from Chen (1951b).

218

SEX IN MICROORGANISMS

Fig. AC.

Variations in conjugation of Paraviechivi trichiuni, from Diller (1948), redrawn;
macronuclei omitted; degenerating micronuclei black: column a, the more common
scheme which is much like that of P. cctiidatimi, with occasional degeneration of one
nucleus after pregametic division, shown in 2 and 3, right; column b, similar to
a, but with autogamy instead of cross fertilization; column c, third pregametic
division omitted; column d, like c with autogamy; column e, like d for 1 and 2,
but with single nucleus in each conjugant after second division— these pass to mates
to develop parthenogenetically; column f, like e but with parthcnogenetic develop-
ment without passage to mate.

SEX IN PROTOZOA 219

autogiiniy (colinn d) may occur. Ihc liaploid nuclei produced by the
two prcgametic divisions may fail to form pronuclei, but pass to their
mates (column e) or develop parthenogenetically (column f.) Some
of these variations arc illustrated in Fig. AD. Transfer of a single
nucleus into a mate with degenerating nuclei is shown at (1), leaving
one conjugant without a nucleus. At (2) a binucleate specimen is
conjugating with a uninucleate one. In (3) and (4) exchange of
macronuclear material is taking place. In (3) one pronucleus has
transferred without reciprocal exchange, producing a "haploid" in-
dividual at the left and a "triploid" at the right, as previously men-
tioned for P. bur sarin. Diller also saw conjugation between a normal
animal and the anterior member of a tandem pair, as described above
for P. bursaria. Later Diller (1949) described an abbreviated process
of conjugation in P. trichm-ni, where exchange of nuclei took place
after the first pregametic division. The products of the first division
proceed directly to reconstitute the new nuclear apparatus by syn-
karyon formation or parthenogenesis, or a combination of the two.
There is no degeneration of nuclei between divisions.

The account of conjugation in P. piitrijiwn by Doflein and
Reichenow^ (1949) follows closely the "standard" series of events in
P. trich'mm, as shown in Fig. AC, column a. As stated by Diller, it
is possible that Doflein's account refers to P. trichmm. The identity
of P. piitrinuvi is still in doubt (Wenrich, 1926).

For Paramecium aiirelia, with two small micronuclei, accounts
of conjugation have been supplied by a number of authors, especially
by Hertwig (1889), Maupas (1889), Diller (1936), and Sonneborn
(1947). There is general agreement that the two micronuclei go
through a crescent stage previous to the first pregametic division.
The four nuclei produced by this division quickly go through the
second division to produce eight haploid nuclei. According to Hert-
wig and Sonneborn seven of these nuclei degenerate while the remain-
ing one divides to produce the pronuclei. However, Diller stated that
a variable number, from two to five, of these eight nuclei at least
begin the third division, but only two of the resulting nuclei, appar-
ently those nearest the paroral cone, become pronuclei. As suggested
by Diller (1936) and Sonneborn (1951), the position in the cell
apparently determines which haploid nuclei become functional. Son-
neborn (1951) cites this as an example of cytoplasmic control over
nuclear activities. After exchange of pronuclei and syncaryon forma-
tion, the two conjugants separate and two mitotic divisions produce

220

SEX IN MICROORGANISMS

Fig. ad. Variations in conjugation of Parmnecium tricbiuvi.

1, 2, 4, from Diller (1948); 3 from Diller (1949), redrawn. 1, transfer of a single
gamete nucleus into a mate with degenerating nuclei, leaving one conjugant without
a micronucleus; 2, a binucleate animal conjugating with uninucleate one; 3, one
pronucleus has passed over without reciprocal exchange, producing a "haploid"
animal at left and a "triploid" one at right, intermingling of macronuclcar material;
4, two functional nuclei in each conjugant, large lobe of macronuclcar material
transferring from one to the other.

SEX IN PROTOZOA 221

tour nuclei troin the svncaryon in each exconjugant. Two of these
become macronuclear anlagen while the other two become functional
niicronuclei. Only one cytoplasmic division, accompanied by micro-
nuclear division, is required to segregate the two new macronuclei into
two daughter cells. Resolution of the old macronuclei into a skein
begins about the time of pronuclear formation. The skein later breaks
up into fragments \\'hich may persist to some extent until after estab-
Hshment of the new nuclear condition in the exconjugants.

Fiimviechnn viulthincromicleatinn usually has four micronuclei
of the same type as those of P. aurelia. At the beginning of conjuga-
tion, according to Landis (1925), all four micronuclei go through
the first division following crescent formation. All eight of the
daughter nuclei undergo the second pregametic division. Of the six-
teen nuclei so produced, usually four embark on the third division
while the other twelve degenerate. Only one nucleus completes the
third division to form the two pronuclei. Again, the location in the
cell seems to determine which nuclei will become functional. After
cross fertihzation, the conjugants separate and three divisions provide
eight nuclei from the syncaryon. Of these, most, probably seven,
degenerate. Further divisions provide two macronuclear anlagen and
two micronuclei. Four macronuclei are eventually produced, appar-
ently by division of the two first formed, while four micronuclei are
evident at the same time. By two successive cell divisions, accom-
panied by micronuclear divisions, the four new macronuclei are
segregated into four cells, each with one macronucleus and four
micronuclei. Miiller (1932) found that, although the conjugations as
described by Landis occurred in over 60 per cent of the animals,
there was a very wide range of variation, especially in the exconju-
gants, in which macronuclear anlagen varied from two to fourteen,
and micronuclei ranged up to twenty-eight. According to Koster
(1933) the fragments of the old macronucleus are distributed among
daughter cells during the six to eight divisions following conjugation
and are eventually taken bodily into the new macronuclei.

Presumably all species of Parmnechnn will undergo conjugation
under appropriate conditions, but no extended accounts have been
published for P. calkinsi, P. ivoodruffi, or P. polycaryum. When P.
calk'msi was first described and named by Woodruff (1921), he had
not been able to induce conjugation or any other type of nuclear
reorganization. Wenrich and Wang (1928) first reported conjugation
for this species but gave no details. Recently Wichterman (1948b)

222 SEX IN MICROORGANISAIS

reported two mating types and conjugation among four races of
P. calkinsi, but cytological details were not given. No accounts of
conjugation in F. poly cary 21m or of P. nuoodruffi have been found.
Gelei (1938) stated that conjugation occurs in Parameciwn nephrid-
iatum but gave no details.

Paramecium "Hybrids"

Various attempts have been made to hybridize different species
of ciliates. In view of the existence of mating types in Paramecium,
one would hardly expect successful hybridization to take place. Miil-
ler (1932) described conjugation between P. caudatum and P. mul-
timicromicleatwn, but none of the exconjugants survived. This was
also true of the crosses between P. caiidatum and P. aurelia obtained
by DeGaris (1935). Considering that P. caudatuvi has a different
type of micronucleus from those of P. aurelia and P. imdtiniicromi-
cleatum, these results are not surprising, although autogamy (cy-
togamy) or parthenogenesis might take place in such cross-species
attachments. It will be recalled that Chen (1946b) found that crosses
between two varieties of P. bursaria were also lethal.

Nuclear Reorganization Other than
Conjugation in Paramecium

''''Endojnixis''^

Woodruff was able to maintain a strain of P. aurelia for many
years and for thousands of generations by the daily isolation method
without conjugation occurring unless induced in side lines. However,
cell division rate underwent a decrease or "depression" about every
30 days. Erdmann studied the nuclear phenomena during these de-
pression periods and found that the old macronucleus broke up into
fragments and a new nuclear apparatus developed from products of
micronuclear divisions. There are some variations in the process but,
as described by Woodruff and Erdmann (1914), the main events are
as follows. AVhile the macronucleus is breaking up, two micronuclear
divisions produce eight small nuclei. Of these, all degenerate except
one or two. A cell division produces two animals, each with one mi-
cronucleus. Two micronuclear divisions give rise to four small nuclei,
two of which become macronuclear anlagen and the other two remain
as micronuclei. A cell division, accompanied by micronuclear division,
segregates the two anlagen into different cells, thus restoring the
normal complement of one macro- and two micronuclei. These

SEX IN PROTOZOA 223

authors ciupliusi/cd the absence of any process of syngainy in "endo-
niixis." In rlie same \ ear llerrwig (1914) described a process of
nuclear reorLiani/.arit)n in this species w hicli he called parthenogenesis.

"Endoniixis" has also been described for VavLrincciu'in Ctrudatuvi,
first very incompletely by Erdmann and Woodrutf (1916) and more
fully by Che j fee (1930) (Fig. AE, 13 to 22). According to Che j fee
the macronucleus begins to fragment while two micronuclear divi-
sions produce four small nuclei (13 to 16). Three of these degen-
erate while the fourth undergoes division (17, 18). After a second
micronuclear division (19) two nuclei become anlagen and the other
two remain as micronuclei (20). One cell division distributes one of
each type of nucleus to each of the two daughters (21, 22). The
similarity of the micronuclear behavior in the earlier stages to that in
conjugation is striking, and if a process of nuclear fusion should have
been overlooked this might prove to be autogamy.

Erdmann's (1925) account of "endomixis" in P. Inirsaria is very
incomplete, and apparently such a process is very rare in that species.
Stranghoner (1932) described "endomixis" in P. miiltmiicromiclea-
tinn involving eight to ten generations. Again the nuclear history is
similar to that of conjugation, especially in the earlier stages. "Endo-
mixis" has been reported to occur in Faramecium polycaryum by
\\'oodruff and Spencer (1923) and in F. nephridiatimt by Gelei
(1938), but without details.

Autogamy

Diller announced the discovery of autogamy in Paramecium
awelia in 1934 and gave a full account of it in 1936 (Fig. AE, 1 to
12). This process occurs in single individuals and parallels the events
described as "endomixis" by Woodruff and Erdmann (1914), except
that a crescent stage develops in preparation for the first pregametic
micronuclear division (2), and fusion of gamete nuclei takes place.
After the second micronuclear division (3), most of the micronuclei
degenerate, while a third division (5) produces two pronuclei (6)
which unite (7) in self-fertilization or autogamy. Meanwhile the
macronucleus passes through a series of skein stages (4 to 8) before
fragmenting into small pieces (8 to 10). Pronuclear fusion takes place
in the paroral cone which forms at the place where exchange of
pronuclei occurs in conjugation. From the syncaryon (8), four nu-
clei are produced by two divisions (9, 10). Two of the four grow
into anlagen (11), and a single cell division (12), accompanied by

224

SEX IN MICROORGANISMS

Fig. AE.

1-12, autogamy in Paramechmi aurelia, from Diller (1936); 13-22 "endomixis" in
P. caiidatum, from Chejfec (1930), all redrawn. 1, normal animal with two micro-
nuclei; 2, micronuclei in crescent stage; 3, second pregametic division, four nuclei
in metaphase; 4, eight products of second pregametic division, beginning resolution
of macronucleus; 5, six micronuclear products degenerate, two others divide; 6,
pronuclei in paroral cone, early skein of macronucleus; 7, fusion of pronuclei in
paroral cone, further resolution of macronucleus; 8, syncaryon in paroral cone; 9,
two products of division of syncaryon; 10, four products of second postzygotic
division; 11, two micronuclei and two anlagen; 12, products of first postzygotic cell

SEX IN PROTOZOA 225

niicroniiclciir dixision, segregates these into two daughter cells, each
with the normal equipment of one macronucleus and two micro-
nuclei (12). The similarity between this process and conjugation is
very close.

Sonneborn and his associates made extensive studies of the con-
ditions under which "endomixis" took place in Faramcciiivi aureVui.
x\fter his discovery of mating tvpes in this species (1937), Sonneborn
made genetic studies of the consequences of this type of nuclear re-
organization and found that in each case autogamy took place instead
of "endomixis" (Sonneborn, 1947). It remains to be seen how many
other cases of so-called "endomixis" may turn out to be autogamy or
parthenogenesis.

Other Forvis of Nuclear Reorganization

In addition to autogamy, Diller (1936) also described a series of
nuclear reorganizations of the macronucleus which did not involve
replacements from micronuclear sources. These reorganizations were
called hemixis and could take place according to several patterns.
Usually a partial or complete breakdown of the macronucleus occurs.
Larger fragments, when segregated into daughter cells by cell divi-
sions accompanied by micronuclear divisions, could develop into
functional macronuclei. When the entire macronucleus fragments
into balls, degeneration followed.

A slightly different form of nuclear reorganization was described
by Sonneborn (1947) for F. anrelia. When animals undergoing con-
jugation or autogamy were kept at a temperature of 38 °C after
fertilization and during subsequent development, the growth of the
macronuclear anlagen was arrested, not to be resumed until several
cell divisions had ensued (Fig. AF). Because the micronuclei were
less affected, cell division could continue. During these divisions some
animals were separated which lacked any macronuclear anlagen (3).
In these, one or more fragments of the old macronucleus began to
grow, and, as they became segregated by subsequent divisions, they
eventually developed into full-grown macronuclei (4 to 7). Sonne-
born called this process macronuclear regeneration.

division, each with one macronucleus and two micronuclei. 13, normal specimen of
P. caitdatJiui; 14, two micronuclei formed, macronucleus breaking up; 15, 16, three
and four micronuclei, fragments of macronucleus; 17, three micronuclear products
degenerating; 18, 19, two micronuclear divisions; 20, growth of two anlagen; 21, cell
division segregating one macronucleus and one micronucleus into each daughter;
22, normal animal.

<

bn

*4-i

U

(U

o

<D

u

JD

bD

m

<u

*-<

x:

o

o

3

c

^

^o

O

?

.^

D-

"o

"O _H

^O "O -C

r^

n

C

c

-t-

!5

!2

E

^

m

.>

bi:)

3

"O

IH

c

C

>*H

Ui

O

•"^

t^

O

u

ID

rt

S^

JJ

U

JJ

ii

^

r— ^

o

c

^

E

3

c

p

C

o

M

O

en

*->

o
n

u

s

o

^

E

iH

^

M-

>-l-l

r-*

rt

o

.2^

ba

^^

c

rt
bJj
u

"^
^

s

^

rt

^

«

ui

^

^>i

4->

bfl

W5

•^

3

J=

U

J2

<o

C

bfi

OJ

«

tiJ

O

3

X

E

S

rvj

L4

a!
13

3

E

c
o

M-l

CL,

o

^''

"rt

.S

4-1

>^

K^

E
o

C

^

o

bx)

c

c

'i-i

■^

o

a

o

n

4-1

rt bD i« ii

2 ^ g ^

3 ^ C "-I

u be .

bo rt 5:3

2 ^ ^

c -a

o -

'-' .a

c

^ o o

rt

3 d e t/'

.— o ^ 3

c -35 g' jj

y-l bi.3

a OJ ^ —

.s u <J «

So c .t:; n

^ V ^ C

,^ .2

226

SEX IN PROTOZOA 227

Nuclear Reorganizaiion in Other Ciliates

It would be reasonable to suppose that nuclear reorganization
comparable to that in species of Faraiiieciu'in would also be found in
other ciliates, and it is. Woodruff (1941) gave a review of these proc-
esses under the head "Endoniixis."

One approach to this question has been to separate animals after
they had become associated for conjugation. Calkins (1921) acci-
dentally separated a pair of conjugants of U role puis mobilis soon
after they had come together. The individuals were isolated and
found to undergo nuclear reorganization. Calkins cut off the anterior
ends of fused pairs of this species, yet the separated animals regen-
erated and underwent nuclear reorganization. When the cutting was
done in an earlier stage, Calkins assumed that autogamy took place,
but when he cut off the anterior ends while the migratory nuclei were
passing through that region, some other explanation was presumed
to apply. Cutting off the anterior ends of exconjugants just after
separation did not interfere with completion of nuclear reorganiza-
tion. Ilowaisky (1926) also stated that individuals of Stylonychia
my tills underwent complete nuclear reorganization when forcefully
separated after becoming associated for conjugation. The initial con-
tacts between conjugants were presumed to stimulate the nuclear
changes. On the other hand, Moore (1924) found no nuclear reor-
ganization in individuals of Spathidhim spatlmla when separated soon
after becoming associated for conjugation.

Nuclear reorganization in single animals in the active state has
been reported for a number of different ciliates. Heidenreich (1935)
described various patterns of reorganization in several kinds of endo-
zoic astomatous holotrichs during which the old macronucleus was
replaced by a new one from micronuclear sources. He referred to the
process as parthenogenesis. DaCunha and Muniz (1930) described
endomixis in Balantidhnn siviile, and I have seen a reorganization of
that type in a similar form of Balantidhivi.

For the Hypotricha, Klee (1926) reported nuclear reorganiza-
tion in Eiiplotes longipes; Ivanic (1929) described nuclear changes in
Euplotes char 071 and E. patella, and Kay (1946) for Oxytricha bifaria.
Horvath (1947, 1948a,b) produced amicronucleate races of Kahlia
simplex either by treatment \\\x\\ ultraviolet light or by excess prod-
ucts of bacterial metabolism. By the first method, micronuclei were

228 SEX IN MICROORGANISMS

destroyed or changed to macronuclei; by the second method, micro-
nuclei grew into macronuclei. Amicronucleate animals were said to
be capable of "endomixis" (presumably hemixis or macronuclear re-
generation) with a repetition every 8 to 10 days. The macronuclei
(normally two) broke up and new macronuclei developed from
some of the fragments. Even amicronucleate animals were said to
"conjugate," the "conjugants" being of different sizes. The joined
animals fused into one, the macronuclei became resolved into ribbons,
then fragmented, and new macronuclei regenerated from fragments
the size of micronuclei. This process cannot properly be called con-
jugation in the absence of micronuclei. When amicronucleate animals
conjugated with normals, there was also complete fusion of the asso-
ciates, and presumably autogamy or parthenogenesis took place.

Among the Peritricha, "endomixis" was described for a species
of Trichodina from tadpoles by Diller (1928). Later (1936) he
thought that exconjugant and hemictic stages might have been lumped
together as "endomixis," while Padnos and NigreUi (1942), after a
study of conjugation in Trichodina spheroidesi, thought that Diller's
"endomixis" could be reorganization in exconjugants. Faure-Fremiet
(1930) recorded nuclear changes under the term "endomixis" in
TjOotbmnniimi alterna7is, and Seshachar (1946) reported hemixis in
Epistylis plicatilis and E. anastatica. Willis (1948) recorded an appar-
ent endomixis in Lagenophrys tatter salli.

Nuclear reorganization seems to occur more commonly in cysts
than during the active state. In my experience, it is common for the
macronucleus to undergo fragmentation in cihates after encystment.
Some of the ciliates which have been reported as undergoing nuclear
reorganization in the cyst stage are Uroleptiis mobilis (Calkins,
1919); Spathidiinn spathula (Moore, 1924); Eiiplotes longipes (Klee,
1926); Chilodonella uncinatus (Ivanic, 1928, 1933a, 1935b); Vorti-
cella nelnilifera (Ivanic, 1929); Urostyla graiidis (Tittler, 1935); and
Paraclevelafidia simplex (Kidder, 1938). In the last-named species,
the reorganization was of an unusual type. The micronucleus divided
into two. The macronucleus underwent a partial breakdown with
about half of its chromatin being eliminated through the reservoir of
the contractile vacuole. The remaining half rounded up and joined
with one micronucleus to form a new macronucleus. This passed
through a "ball-of-yarn" stage like that in the exconjugants of Nycto-
therus cordijorviis and some other ciliates. The other product of

SEX IN PROTOZOA 229

niicroiuiclcar division hccnnic the new niicronucleus. Most of these
accounts are not siifiicicntly detailed to determine just what kind of
reorganization takes place. They could be "endomixis," hemixis, au-
togamy, or parthenogenesis. However, Corliss (1952a, b) has re-
cently described in detail an autogamous nuclear reorganization in
the cysts of Tetrahyviena rostrata. Apparently this is the first clear
account of autogamy in a ciliate cyst. Doubtless other cases of autog-
amy will be found.

Order Peritricha

Since the time of Ehrenberg (1838) it has been observed that the
members of the Peritricha (especially the Vorticellidae) conjugate in
a manner different from the typical pattern. One or more preliminary
cell divisions produce larger and smaller conjugants. Figure AG (17)
shows a microconjugant and a macroconjugant being formed by a
single division of a '"neutral" individual of Vorticella microstoma, as
described by Finley (1943). The microconjugant becomes detached,
becomes motile, and swims about until it finds a macroconjugant with
which it unites. Variations in the formation of microconjugants con-
sist of one, two, or three divisions producing two, four, or eight
microconjugants from the first smaller animal derived from a "neu-
tral" individual. Finley and Nicholas (1950) found that four micro-
conjugants were produced from the smaller product of the first "sex-
segregating" division of Rhabdostyla vernalis. According to Finley
(1943) the macroconjugant is attractive to the microconjugant for
about 2 hours. Presumably a micro- and macroconjugant from the
same parent could conjugate together. Several microconjugants may
attach to one macroconjugant (Fig. AG, 18).

The nuclear details are shown in Fig. AG (1 to 16) as illustrated
by Maupas (1889). Here it will be seen that there are three prelimi-
nary micronuclear divisions in the microconjugant but only two in
the macroconjugant (1 to 5). After these preliminary divisions one
nucleus in each conjugant survives and divides to produce pronuclei
(6 to 8), but only one of those in the macroconjugant fuses with one
from the microconjugant to form a syncaryon (9), By three divi-
sions, the fusion nucleus gives rise to eight nuclei (10 to 13) of which
sevTn enlarge as anlagen while one becomes the micronucleus (14).
Accompanied by micronuclear divisions, the anlagen are segregated

230

SEX IN MICROORGANISMS

SEX IN PROTOZOA 231

into single individuals. Most of the material in the niicroconjugant
enters the niacroconjugant, but a remnant (10) is cast off. In the case
of Opistbo/iectii hcnuevi^uyl, a free-swimming vorticeUid (Fig. AG
19), the niicroconjugant fused completely with the macroconjugant
(Rosenberg, 1940).

In the Urceolariidae, which are free-swimming commensals on
or in various hosts, the difference in size between conjugants is not
so marked as for the sedentary vorticellids. The most recent accounts
are those for Ti'ichodma spheroidesi (Padnos and Nigrelli, 1942) and
for Urceolaria synaptae (?) (Colwin, 1944). Padnos and Nigrelli
reported only two preliminary micronuclear divisions in the micro-
conjugant, whereas Colwin recorded three. Colwin found that ex-
change of pronuclei sometimes took place before nuclear material of
the niicroconjugant passed over into the macroconjugant, leaving a
remnant which was later discarded. In Trichodina spheroidesi the
usual postconjugant development took place, eight nuclei being de-
rived from the snycaryon, seven of which became macronuclei and
one the micronucleus. In Urceolaria syjiaptae, however, only four of
the eight products of the syncaryon became macronuclei, three de-
generated, and one became the micronucleus, following the pattern
for Faraniecium caiidatwn instead of that characteristic for vorti-
cellids.

According to Awerinzew (1936), in a species of Lagenophrys,
which lives in a lorica on the gill plates of a freshwater crab, Tel-

FiG. AG.

1-16, conjugation in Vorticella inoiiilata, from Maupas (1889). 1, niicroconjugant
attaching to macroconjugant; 2, first pregametic micronuclear division in niicro-
conjugant, break-up of macronucleus; 3, first pregametic division in macroconjugant,
second in niicroconjugant; 4, second pregametic division in macroconjugant, third in
niicroconjugant; 5, products of these divisions; 6, single nucleus left in each con-
jugant after others have degenerated; 7, 8, final pregametic division to produce
pronuclei; 9, passage of one pair of pronuclei into macroconjugant where syncaryon
is formed; 10, division of syncaryon, small remnant of microconjugant; 11, second
postzygotic division; 12, third postzygotic division; 13, eight products from syncar-
yon; 14, seven products become macronuclear anlagen, one becomes the micro-
nucleus; 15, 16, after first postconjugant cell division. Further divisions segregate
anlagen into single cells. 17, 18, Vorticella microstoma, from Finley (1943). 17, divi-
sion of a neutral individual into one micro- and one macroconjugant; 18, three
microconjugants attached to one macroconjugant. 19, conjugation of macro- and
microconjugants of Opistbonecta benneguyi, from Rosenberg (1940). 20, conjugants
of unequal size in Lada tanishi, from Aliyashita (1928), with unusual points of con-
tact. (All illustrations redrawn.)

232

SEX IN MICROORGANISMS

phusa, conjugation was preceded by an unequal division of an
ordinary individual into a larger macroconjugant and a smaller
"gametocyte," which then divided into two microconjugants. The
gametocyte was not ciliated, but the micronconjugants were. There
were two types of these with different ciliation and size. Only the
larger type was seen to function as a micronconjugant. The nuclear
phenomena were typical of the vorticellids.

Fig. ah.

1-4 from Dogiel (1925). 1-3, stages in Opisthotrichium jaims: 1, progamic division
producing unequal conjugants; 2, conjugation of macro- and microconjugants; 3,
conjugation of two macroconjugants; 4, conjugation of Cycloposthium bipahuatiim
showing differentiation of migratory pronucleus into a sperm with tail; 5-8, stages
in total fusion of conjugants of Spirochona gemviipara, from Plate (1886). (All illus-
trations redrawn.)

si'X IN PRcriozoA 233

Order Cbojwtricii

According- to Plate (1886) and Swarczcwsky (1928), Spiro-
cboihi of this cpi/.oic group undergoes conjugation by complete
fusion between the conjugants, as sho^n in Fig. AH (5 to 8). Nu-
clear details were nor fully worked out.

CLASS SUCTORIA

Conjugation among Suctoria, which are usually sedentary and
without cilia in the adult state, foUow^s the common pattern in regard
to nuclear behavior. There are a number of accounts, but the mono-
graph by Collin in 1912 provides a review of the literature published
up to that time.

A^'hen epidemics of conjugation occur, individuals of many spe-
cies form cytoplasmic "arms," sometimes as long as the animal itself,
which extend out from the body in various directions. If one of these
contacts a similar extension from another individual, they become
attached and pronuclei are exchanged across the "bridge" thus
formed.

Collin recognized two general types of conjugation in this group;
one is the common mode of exchange of pronuclei, separation, then
reconstitution of the nuclear apparatus from the syncaryon in each
exconjugant. In the other type, there is complete fusion with only
one "synconjugant" surviving. In this type the two individuals may
be equal in size ("isogamy") or unequal ("anisogamy").

In Tokophrya cyclopum (Fig. AI), complete fusion occurred.
A first micronuclear division (1 to 3) was followed by a second (4).
After fusion of the conjugants a division of the syncaryon produced
a single micronucleus and a new macronucleus (5, center) w^hile the
macronuclei began to degenerate (5, at ends). Reconjugation could
occur. In the case illustrated (6) a synconjugant (left) is shown con-
jugating w^ith another individual which is the product of a previous
reconjugation (right). The latter has two smaller and two larger old
macronuclei besides a new macronuclear anlage and three micro-
nuclei. In another illustration (not shown here) a larger synconjugant
was in conjugation with two normal conjugants. The Suctoria thus
show transitions between the type of conjugation common to the
Peritricha and the more typical condition of the other Euciliata.

Note: The literature dealing with conjugation is so extensive that a compre-
hensive discussion is out of the question. Instead, a few topics have been
selected for longer or shorter comments.

•2 2

o
U

B

o

a

'c
o
U

2i

a
^

_4>

MS

o-

C

c

rt

i-i

o

rt

S
O

■-l-l

o

i->

c

T3

'<u

<u

c^

C3

bX)

O

"o

Cu

O

3

C
O

O

■t->

in

"oj

o

C/)

O

*->

c

a!

rt

-

>>J=

6

r<^

,

M-i

bij

c

•^-^

JJ

_c

.2

u

4-1

c

•r

O

fcn

■-5 t; 1 g

.2 e

tJ CD

bJl O

. ? "

o ^

o -^

<u

rt

J=, O

O 3

^« ^^

^ o o

o

rt

3
C

O oj

o o

rt 3

e s

I

8 S

£i Si

_S a-

Cu-O

o o

-2 O

O '3

c o

J-i

3 r^
bfl ^ .

3 '

o

'So

O o

e §

O -

in -^

'3 .3^j2

s-i -IS
(u -r

234

SEX IN PROTOZOA 235

Differentiation of Conjugants

Attention has already been called to the regularity with which
conjugants of different sizes occur among members of the Peritricha
and sometimes also in the Suctoria. Dogiel (1925) believed that con-
jugants were generally different, and he offered many tables of meas-
urements, mostly in the Ophryoscolecidae, to support the concept.
Some later w riters have reported size differences between conjugants,
as, for example, in various species of Chilodonella (MacDougall, 1925,
1936), Cry ptochihivi echini (Russo, 1926, etc.), Lada tctnishi (Fig.
AG, 20; Miyashita, 1928), Fty cbostountm (=Lada) chattoiii (Studit-
sky, 1932), Concho phtbirius viytili (Kidder, 1933), and Entorrhi-
pidiiivi echini (Yagiu, 1940). According to Enriques (1908) conju-
gants of Chilodonella nncinatns did not differ in size until after
attachment, and he thought that the contact stimulated size differen-
tiation, which he called "hemisex." On the other hand, Ivanic (1933b)
denied any sex differences in C cuciillus. In Dogiel's (1925) review
some cases wqvq cited where differences between conjugants were
still more pronounced. In Opisthotrichiinn jamis, an endozoic ciliate,
a single preconjugant division produced a macroconjugant and a
microconjugant of different morphology (Fig. AH, 1). However,
Dogiel noted that not only would a micro- and a macroconjugant
unite (2), but that two macroconjugants might associate together (3).

Maupas (1889) believed that in most ciliates the conjugants were
smaller than non-con jugants and that this reduction in size resulted
from one or more special preconjugant divisions. Although this idea
has too many exceptions to be a vahd generalization, a considerable
number of examples can be found to support it. Special preconjugant
divisions have been described or postulated for a number of ciliates
including Dileptus gigas (Visscher, 1927), Balantidimn sp. from the
chimpanzee (Nelson, 1934), Nyctothenis cordi^ormis (Wichterman,
1937), and Fabrea salina (Ellis, 1937). According to Scott (1927),
conjugants of Balantidiiivi caviae were always smaller than non-
conjugants, and Bogdanowicz (1930) reported the same condition for
Loxodes striata. One wonders if these rapid fissions could be stimu-
lated by abundance of food, then inhibited by subsequent lack of
food, since decline of food after abundance has often been suggested
as a stimulus to conjugation. However, Giese (1938) found no regu-
lar decline in size before conjugation in Blepharisma undulans, al-

236 SEX IN MICROORGANISMS

though size of conjugants did depend on previous history, mostly
relating to nutritional conditions. Thus, in many ciliates conjugants
are smaller than non-conjugants but may be essentially of the same
size, and Pearl (1907), Jennings (1911), and others have claimed
that selective mating takes place, since, on the average, members of
pairs of conjugants are more alike in size than non-conjugants chosen
at random.

Pregametic Nuclear Phenomena

The number of pregametic mitoses in the Euciliata is relatively
constant. There are certain other features that are generally found.
The preparation for the first division (second in Euplotes) is a period
of long duration, usually with special configurations of the intra-
nuclear material. Species of Farameciwn commonly show a "crescent"
or "sickle" stage (Fig. AA, 3, at right) while many other ciliates
show a "parachute" or "candelabra" stage (Fig. AJ, 13; AL, 1).
Wichterman (1940) estimated that the time from first association of
conjugants to the first metaphase lasted lYi hours in Farameciwn
caudatwn, the total time of attachment being about 13 hours. Since
this first division is the first of the two meiotic or maturation divisions,
and the micronucleus increases greatly in size, it is generally assumed
that this size increase corresponds to the growth stage in metazoan
gametogenesis. Most authors report "reduction" in the second of
these divisions. In harmony with metazoan conditions, one would
expect this first division to involve synapsis of homologous chromo-
somes and the formation and division of a haploid number of tetrads
which would separate into dyads on the mitotic spindle. However,
ciliate chromosomes frequently do not show the pattern of behavior
that one finds in the typical metazoan first gametocyte nucleus. In
Euplotes patella (Turner, 1930) there are four pregametic nuclear
divisions after the conjugants have become joined. The first division
follows a regular cell division pattern and shows eight chromosomes
on the mitotic spindle (Fig. AJ, 1 to 8). The second division, then,
parallels the first pregametic division found in most other ciliates,
with a long-continued prophase (9 to 14) which involves a "para-
chute" stage (13). Instead of four tetrads appearing in the metaphase
of this division, about thirty-two granules, most of them in pairs,
make their appearance (15), and sixteen of them go to each pole (16).
During the following division (17 to 20), only eight chromatin
bodies show in the prophase (18), but in the metaphase four elon-

SEX IN PROTOZOA

237

24\:

Fig. AJ. Nuclear divisions in conjugation of Euplotes patella, from Turner

(1930), redrawn.

1-8, first (extra) pregametic micronuclear division showing eight chromosomes; 9-16,
second pregametic (first meiotic) division, "parachute" stage at 13, thirty-two gran-
ules in metaphase (15), sixteen going to each pole (16); 17-20, third pregametic
(second meiotic) division, "reduction" takes place; 21-24, fourth pregametic division
with four chromosomes dividing.

238 SEX IN MICROORGANISMS

gated pairs of chromatids appear (19) and four chromatids go to
each pole (20). This is the third nuclear division, which corresponds
to the second in other ciliates; it seems to be the "reduction" division.
In the fourth and last division (21 to 24) only four chromosomes
appear, and these divide to send four to each pole of the spindle (24).

The peculiar behavior of the chromatids in the first meiotic
division has caused much confusion about chromosome numbers in
meiosis. Turner considered the granules in the prophase of the first
meiotic division (15) to be chromomeres, and stated that they usually
are arranged in about eight groups. In the anaphase (16), the chroma-
tin is still in the form of chromomeres. In the next division four pairs
of chromatids appear (19). Presumably each pair is derived from
four chromomeres. If the chromatin bodies which appear during
the first maturation division are chromomeres, it is possible to fit the
whole series of divisions into a scheme comparable to that in the
Metazoa.

Gregory (1923) had still more difficulty with Oxytricha fallax,
since she counted twenty-four dumbbells in the first maturation meta-
phase, twelve going to each pole; in the second division twenty-four
dumbbells again appeared and twelve went to each pole; in the third
division twelve dumbbells appeared and twelve granules separated to
the poles. This case seems to involve double reduction or an extra
division of chromatids between the first and second divisions. How-
ever, Kay (1946), in Oxytricha bifaria, reported forty-eight granules
in the first maturation division, twenty-four going to each pole, and
twenty-four chromosomes in the second, twelve going to each pole.
She considered that reduction took place in the second division.

The tendency for protozoan chromosomes to appear as granules
in one stage of mitosis and as more solid strands at another is illus-
trated by conjugation stages in Nyctothenis cordijoriuis (Wichter-
man, 1937). Two of these stages are shown in Fig. AK (1, 2). In (1)
the anaphase in the conjugant at the left shows a group of granules
passing to each pole in the third pregametic division, which produces
the pronuclei. Metaphases of this division also show a similar number
of granules. In the right-hand conjugant of this figure, the telophase
of this division shows a "spireme." In the other figure (2) are seen
the fusing pronuclei where there seems to be a single thread-like
chromosome in each pronucleus. Noland (1927) found a parallel
situation in Metopus sigmoides. These figures illustrate some of the
difficulties involved in interpreting chromosome numbers in ciliates.

Fig. AK. Conjugating stages of Nyctotherus cordijorviis from Wichterman

(1937), redrawn.

1, anaphase stage of third pregamctic micronuclear division at left, with many small
chromosomes, telophase of this division at right; 2, conjugating pair showing fusion
of pronuclei, one chromosome in each pronucleus; note mingling of macronuclear
fragments at anterior ends of the conjugants in both figures.

239

240 SEX IN MICROORGANISMS

Recently these difficulties have been the subject of a special
study by Devide and Geitler (1947) and Devide (1951). In the ac-
count of Devide (1951) it is stated that the chromosomes of eucil-
iates in the earher literature are not and cannot be chromosomes, but
chromosome aggregates, the transverse division of which, so far as it
occurs, and the inconstancy of shape and number, do not present any
problem for modern karyology. The "true" chromosomes are said
to be visible chiefly during meioses, when they pair and form tetrads
seen in metaphases where dyads separate. The masking observed in
somatic, metagamic, and postmeiotic mitoses can also apply sometimes
to meioses.

Devide used aceto-carmine mostly and described micronuclear
divisions and meioses for Colpidhim compylum, Euplotes char on,
Oxytricha sp., Chilodonella dentata (= C. uncinata), and mitoses for
Opal'ma ranarnm and Cepedea dimidiata. His demonstration of typi-
cal tetrads in the first meiotic division of Colpidium compyhim and
Euplotes char on are impressive, but not so satisfying for Oxytrica,
Stylonychia, or Chilodonella, since clearly defined tetrads are not
indicated for these other species. Hence, even with Devide's tech-
nique, satisfactory details of meiosis were not always forthcoming.
The idea of chromosome aggregates is a useful one and would help
to explain such a situation as that seen in Nyctotherus, referred to
above (Fig. AK). In the prophase of the first meiotic division of
Stylonychia, Devide states that chromatic bodies number well over
100 and possibly 300 to 400. With such large and indefinite numbers,
it would be very difficult to work out meiotic counts, and one won-
ders if, after all, in such stages, the chromatic bodies are not chromo-
meres instead of chromosomes.

Another feature of the meiotic period of conjugation deserves
to be mentioned. As already noted, Diller showed transfer of macro-
nuclear material from one conjugant to another in Taraviechmt
trichium (Fig. AD, 3 to 4). In Nyctotherus cordifonnis (Wichter-
man, 1937) conjugants eventually became completely fused in the
anterior region so that, presumably, cytoplasm and macronuclear
fragments are free to transfer from one conjugant to the other (Fig.
AK). In those cases where conjugants fuse partially or completely,
as in Metopjis sigmoides, the Peritricha (Fig. AG, 7 to 9), the Chono-
tricha (Fig. AH, 5 to 8), and the Suctoria (Fig. AI), cytoplasmic
and macronuclear material of the two conjugants may become mixed

Fig. AL. Conjugation in Ajjoplophrya branchiarwn from Collin (1909),

redrawn.

1, "candelabra" stage of micronuclei; 2, metaphase of second pregametic micro-
nuclear division, macronuclei elongating; 3, prophase of third pregametic nuclear
division, three nuclei degenerating in each conjuganf, 4, pronuclei before exchange;
5, prophase of division of syncaryon, one macronucleus projecting into other con-
jugant; 6, telophase of division of syncaryon; 7, second postzygotic division, elon-
gated macronuclei stretching through both conjugants; 8, constriction of ma-
cronuclei, telophase of second postzygotic division; 9, exconjugant with old
half-macronuclei, three small nuclei, one enlarging to become an anlage; 10, exconju-
gant with rounded-up macronuclcar remnants, one macronucleus and one micronu-
cleus.

241

242 SEX IN MICROORGANISMS

together. In certain ciliates, however, conjugation is accompanied by
a regular exchange of portions of the macronucleus between the
conjugants, as described, for example, for Anoplophrya branchianim
(Collin, 1909). As shown in Fig. AL, each macronucleus becomes
much elongated and then projects itself into the cytoplasm of its
mate (5 to 8). The result is that each exconjugant contains half of
its own macronucleus and half of that of its partner (9). Summers
and Kidder (1936) found a similar condition in Anoplophrya or-
chestiae, and Macdougall (1936) in Chilodojiella labiata. Just why
certain ciliates should regularly exchange portions of their macro-
nuclei during conjugation would be difficult to explain.

Differentiation of Gametes

Often there is no obvious difference in the appearance of the
two pronuclei which form in a conjugant, but in many cases there
are more or less pronounced differences. Sometimes there is merely
a slight difference in size, as in Faraiuecmm caiidatiim (Maupas, 1889;
Calkins and Cull, 1907, and others), or there may be a special cyto-
plasmic area in connection with the migratory pronuclei as described
for Uropleptiis mobilis (Calkins, 1919). But perhaps the most ex-
treme case of gamete differentiation is that in Cycloposth'mm bipal-
viatinn (Dogiel, 1925) where the migratory pronucleus becomes
sperm-like with a long tail, breaking out of the endosarc into the
joined peristomal cavities and passing down the cv^topharynx to the
endosarc of its partner (Fig. AH, 4) . The differentiation of the male
gamete in this case is comparable to that in many Metazoa.

Different Patterns of Behavior in Conjugation

Some of the variations in the behavior of the nuclei during con-
jugation have been referred to in the previous pages. No attempt has
been made in this review to classify them. Dogiel (1925), in his
extensive review of conjugation, recognized five different types of
nuclear behavior leading to synkaryon formation and ten types of
nuclear reconstruction in exconjugants. Since there is often a great
deal of variation within any one species, as shown, for example, by
Diller (1948) for farmnechim trichhivi, any attempt to make use of
or expand Dogiel's classifications does not seem profitable.

SIX IN PROro/OA

!43

Faoi.uiion ok Q)NjL'(;Ari()N

\Mien one tries to compare conjugation with other forms of
svnijamv, one is struck by the fact that in typical examples most of
tlie potential gametes fail to develop. Since three pregamctic divisions
in each coniufrant seem to he the rule, potentially eight gametes could
be produced from each original micronuclcus, whereas normally only
two become functional. This suggests that bringing the gamonts (con-
jugants) togetlier reduces the hazards to potential gametes, since an
exchange of gametes can be effected with dangers of loss greatly
reduced. Conjugation might therefore be thought of as an evolu-
tionary product derived from more primitive ancestral conditions
where a brood of gametes was produced, somewhat like that in

-QD ) (ccg^ 13 <^^0 _ _

C> (k> O <D> "CD (2> <£> (S)

Fig. am. Scheme of life cycle of Dallasia frovtata including gamete brood
formation, from Calkins and Bowling (1928), redrawn.

1, 2, young "tailed" forms; 3-5, division of "tailed" forms; 6, preconjugant; 7, con-
jugation of "tailed" forms; 8, "boat" form, 9-11, divisions without growth, to pro-
duce gametes; 12, encystment of sister gametes; 13, 14, zygote formation; 15, 16,
excystment and development of new "tailed" forms.

244 SEX IN MICROORGANISMS

chlamydomonad flagellates (see paper by Lewin), or the formation
of uninucleate gametes in the Opalinidae.

Fortunately, we have in the unusual behavior of Dallasia fron-
tata (Calkins and Bowling, 1928), the development of just such a
brood of gametes (Fig. AM). In this case the cell body divides with
each nuclear division, leading to the formation of gametes. Isogametes
are formed which, however, unite with sister gametes in a process of
autogamy (called paedogamy by Calkins and Bowling) to produce
each zygote. Normal conjugation also occurs in this species (Calkins
and Bowling, 1929). In typical conjugation, nuclear divisions take
place without cell divisions; hence gamete development is telescoped
into the cell body of each conjugant.

Calkins and Bowling (1928, 1929) suggested that the peculiar
brood formation in Dallasia is related to "endomixis" and that both
might relate to ancestral gamete brood formation. Attention has
already been called to the situation in the vorticeUids, where an
unequal division of a "neutral" individual produces macro- and micro-
con jugants. In many cases, four or eight microcon jugants are pro-
duced. These divisions might also reflect the supposed ancestral gam-
ete brood formation pattern. Furthermore, it will be recalled that in
the microconjugant there is an extra micronuclear division, which
may indicate a definite trend toward multiple gamete formation. Just
why there should be manifested such an ancestral trend in the micro-
conjugant and not in the macroconjugant is an interesting question.
A survival value might be postulated since the microconjugant dies
if it does not find a mate, whereas the macroconjugant, after a few
hours of receptiveness toward microconjugants, reverts to a "neutral"
status, capable of continuing the line by normal binary fission (cf.
discussion by Finley, 1952). The extra micronuclear pregametic divi-
sion in each conjugant of Euplotes can be interpreted as a part of this
ancestral trend, as can also the one or more preconjugant cell divi-
sions that have been reported for a good many ciliates (see discus-
sion of differentiation of conjugants).

The relationship between the peculiar gametogeny of Dallasia
and "endomixis," suggested by Calkins and Bowling, is not so readily
discerned. However, since Diller (1936) and Sonneborn (review,
1947) have shown that "endomixis" in Faramecium aurelia is au-
togamy, and the so-called paedogamy of Dallasia is also seen to be
autogamy, then that aspect, at least, is common to both processes. But

SEX IN PROTOZOA 245

autogamy also occurs as a variation of conjugation, as previously
shown for various species of Faramecium. Furthermore, autogamy is
widely distributed among the Protozoa, seemingly being character-
istic for the Cnidosporidia and Haplosporidia, and occurring in Helio-
zoa, and possibly in the Radiolaria, and various animal flagellates
(Cleveland's studies). If ciliates have evolved from flagellates where
autogamy as well as union of gametes of diverse parentage occurs,
the appearance of autogamy in ciliates should not be surprising. Un-
fortunately, Calkins and Bowling found that the zygotes resulting
from autogamy were mostly non-viable. This might suggest that
conjugation, providing for cross fertilization, would have a greater
survival value than autogamy. One could not say the same for the
Cnidosporidia and Haplosporidia, in which autogamy is the rule.

In the preceding consideration of sexual phenomena in ciliated
Protozoa, no attempt has been made to describe the occurrence and
behavior of mating types, since discussions and interpretations of
mating-type phenomena are provided by Dr. Nanney and Dr. Metz
in the next two papers.

SUMMARY

Among the Phytomastigina, syngamy is common in the Phy-
tomonadina (see paper by Lewin). It is uncommon in the Chryso-
monadina, Euglenoidina and Dinoflagellata, and in these groups no
haploid-diploid cycles have been fully demonstrated except for the
chrysomonad, Ochrosphaera JieopoHtana.

In the Zoomastigina, cell fusions have been described for scat-
tered representatives of the Rhizomastigina, Protomonadina and Poly-
mastigina, but without adequate cytological details. According to
Cleveland, however, the flagellates that inhabit the gut of the wood-
feeding roach, Cryptocerciis pimctiilatiis, exhibit sexuality whenever
the host molts, and detailed cytological descriptions have been pro-
vided for certain species. Haploid flagellates undergo postzygotic
meiosis, some kinds with one, some with two divisions. Diploid flagel-
lates show pregametic meiosis which requires two divisions in some
species, and only one in others. Autogamy occurs in some species.
In Trichojiyinpha and some other haploid flagellates, each gameto-
cyte produces a male and a female gamete by a sex-differentiating
mitosis during which each parental chromosome gives rise to dif-

246 SEX IN MICROORGANISMS

ferently staining chromatids; the more darkly staining male chroma-
tids are segregated to one pole and the more lightly staining female
chromatids to the other pole by a special mitotic mechanism.

In the Sarcodina, cell fusions without adequate chromosome data
have been described for various members of the Proteomyxa, Amoe-
bina and Testacea. In the Foraminifera, syngamy is common, usually
alternating with asexual multiple fission. Diploid-haploid sequences
have been worked out for a few species. Gametes are amoeboid in
some species, flagellated in others. In the Heliozoa, meiosis and ap-
proximate isogamy have been reported for Actinosphaerium eich-
horni and for Actinophrys sol with full chromosome details described
for the latter species. Formation of flagellispores has been reported for
various kinds of Radiolaria by several authors, but satisfactory evi-
dence for syngamy has not been found.

In the Sporozoa, sexual reproduction appears to be general. In
the Gregarinida both isogamous and anisogamous unions and both
pregametic and postzygotic meioses have been described. In the
Coccidia and Haemosporidia, anisogamy prevails with postzygotic
meiosis the rule in the Coccidia. iMeiosis has not been satisfactorily
worked out for the Haemosporidia. Attention is called to evidence
indicating that sex determination takes place early in the development
of gregarines and coccidia, possibly at the first postzygotic nuclear
division. In certain cases, development of a parasite is seen to be
correlated with developmental stages of the host. In the Myxospo-
ridia, Actinomyxidia, Microsporidia and Haplosporidia, autogamy
seems to be the rule, with isogamy or slight anisogamy occurring.
No sexual reproduction has been reported for the Hehcosporidia or
Sarcosporidia.

In the Ciliophora, certain members of the Protociliata, all of
which are endozoic, are said to show isogamy or anisogamy of gam-
etes which are produced by fission of gamonts that excyst in new
hosts. In the Euciliata and Suctoria, sexual reproduction commonly
takes the form of conjugation during which mutual fertilization takes
place between the conjugants. Conjugants are often similar but may
be dissimilar in size and morphology. They frequently are smaller
than nonconjugants, the result, in some cases at least, of special pre-
conjugant divisions. In the Peritricha, conjugants are regularly dis-
similar in size, the microconjugant fusing partially or wholly with
the macroconjugant, which alone survives. In certain members of the

SEX IN PRO 1 OZOA 247

Clionotricha and of the Suctoria, complete fusion of conjugants takes
place to produce a single synconjugant. Other variations in conjuga-
tion include cytogamy (autogamy in each conjugant) and partheno-
genesis in each member of a pair. Single individuals may undergo
autogam\'. Nuclear reorganization without sexuality may also take
place as in endomixis, hemixis and macronuclear reorganization. For-
mation of a brood of small gametes by a single parent, as in Dallasia
froiitiitii (and other evidence), suggests that conjugation has evolved
from a brood-forming ancestral condition.

It is a pleasure to acknowledge my indebtedness to Dr. J. Russel Gabel
for his care and skill in redrawing illustrations from the literature. Acknowl-
edgment of sources is made in connection with the individual illustrations or
groups. The Wistar Institute kindlv granted permission to use or copy the
illustrations in Figs. F, G, H, I, J, K, L, M, S, U, W, AA, AB, AC, AD (1, 2,
4), AE (1 to 12), AK, and AM.

REFERENCES

Adie, H. A. 1921. Preliminary note on the development of the Leishman-

Donovan parasite in spleen juice and in the alimentary canal of Cijnex

lecndarius. bid. }. Med. Research, 9, 255-260.
Alexeieff, A. 1912. Le parasitisme des eugleniens et la phylogenie des sporo-

zoaires sensu strictu. Arch. zool. exptl. et gen., ser. 5, 10 (Notes et Rev.),

73-84.
. 1925. Notes sur quelques protistes coprocoles. Arch. Frotistenk., 50,

27-49.
Aratari, A. 1889. Morphologische und biologische Studien iiber Nuclearia

delicatula Cienk. Zool. Aiiz., 12, 408-416.
Awerinzew, S. 1906. Beitrage zur Kenntnis der Susswasserrhizopoden. Arch.

Frotistenk., 8, 112-119.
. 1936. Zur Biologic des Infusors Lagenophrys. Arch. Frotistenk., 87,

131-141.
Balbiani, G. 1884. Legons sur les sporozoaires. Doin, Paris.
Baldrey, F. S. H. 1909. Versuch und Beobachtungen iiber die Entwicklung

von Trypanosojna leivisi in der Rattenlaus Haeniatopiniis spimdosiis.

Arch. Frotistenk., 15, 326-332.
Bastin, A. 1919. Contribution a I'etude des gregarines monocystidees. Bull.

biol. France Belg., 53, 325-373.
Becker, E. R. 1934. Infection with a single coccidan oocvst and its significance.

A7n. Naturalist, 68, 571-574.
Belaf, K. 1916. Protozoen Studien II. Arch. Frotistenk., 36, 241-302.
. 1921. Untersuchungen iiber Thecamoben der Chlamydophrys Gruppe.

Arch. Frotistenk., 43, 287-354.

248 SEX IN MICROORGANISMS

. 1923. Untersuchungen an Actinophrys sol Ehrenberg. I. Die Morphol-

ogic des Formwechsels. Arch. Frotistenk., 46, 1-96.
-. 1926. Der Fonnwechsel der Frotistei7keriie. G. Fischer, Jena.

Berliner, E. 1909. Flagellaten-Studien. Arch. Frotiste?jk., 15, 297-325.
Biecheler, B. 1937. Sur I'esistence d'une copulation chez une euglene verte et

sur les conditions globale qui la determinent. Covipt. rend. soc. biol., 124,

1264-1266.
Blochmann, F, 1887. Zur Kenntnis der Fortpflanzung von Euglypha alveolata.

Morphol. Jahrb., 13, 173-183.
Bogdanowicz, A. 1930. Ueber die Konjugation von Loxodes striatus und

Loxodes rostnnn. TjOoI. Anz., 87, 209-222.
Borgert, A. 1900. Untersuchungen iiber die Fortpflanzung der tripyleen Radio-

larien, speciell von Aulacantha scolymantha. I. Zoo/. Jahrb., 14, 203-276.

. 1909. Aulacantha scolymantha. II. Arch. Frotistejik., 14, 134-263.

Bott, K. 1907. Ueber die Fortpflanzung von Felojnyxa palustris. Arch. Fro-
tistenk., 8, 120-158.
Brandt, K. 1885. Die Koloniebildended Radiolarien (Sphaerozoen) des Golfes

von Neapel. Monog. 13, Fauna & Flora des Golfes von Neapel., 1-276.
. 1905. Beitrage zur Kenntnis der Colliden. Arch. Frotistenk., 6, 245-

271.
Bunting, M. 1926. Studies on the life-cycle of Tetramitus rostratus Petty. /.

Morphol. and Fhysiol, 42, 23-81.
Biitschli, O. 1875. Zur Kenntnis der Fortpflanzung bei Arcella vtdgaris. Arch.

mikroskop. Anat., 11, 459-467.
. 1876. Studien iiber die ersten Enrwicklungsvorgange der Eizelle,

die Zellteilung und die Konjugation der Infusorien. Abhandl. Senckenb.

naturf. Ges., Frankfurt, 10, 213-452.
. 1878. Beitrage sur Kenntnis der Flagellaten und einiger verwandeten

Organismen. Z. wiss. Zool., 30, 273-281.
-. 1880-89. Der Frotozoa. Bd. 1, Bronn^s Klassen und Ord. des Tierreichs.

C. F. Winter, Leipzig.

Calkins, G. N. 1919. Uroleptus mobilis, Engelm. I. History of the nuclei dur-
ing division and conjugation. /. Exptl. Zoo/., 27, 293-357.

. 1921. Uroleptus mobilis, Engelm. IV. Effects of cutting during conju-
gation. /. Exptl. Zool, 34, 449-470.
-. 1933. Biology of the protozoa. Philadelphia, Lea & Febiger.

Calkins, G. N., and R. Bowling. 1926. Gametic meiosis in Monocystis. Biol.

Bull, 51, 385-399.
. 1928. Studies on Dallasia frontata Stokes. I. Polymorphism. Biol. Bjili,

55, 101-112.
. 1929. Studies on Dallasia frontata Stokes. II. Cytology, gametogeny and

conjugation. Arch. Frotistenk., 66, 11-32.
Calkins, G. N., and S. W. Cull. 1907. The conjugation of Farameciuni attrelia

(caudatum). Arch. Frotistenk., 10, 375-415.
Calkins, G. N., and F. M. Summers. 1941. Frotozoa in Biological Research.

Columbia University Press, New York.

SEX IN PROTOZOA 249

Cash, J., and J. Hopkinson. 1905. The British Freshwater Rhizopoda and

Heliozoa, Vol. 1. The Ray Society, London.
. 1909. The British Freshwater Rhizopoda and Heliozoa, Vol. 2. The

Ray Society, London.
Cash, J., G. H. W'ailcs, and J. Hopkinson. 1915. The British Freshwater

Rhizopoda and Heliozoa, Vol. 3. The Ray Society, London.
Chagas, C. 1927. Quelques aspects evolutifs du Trypanosoma cruzi dans

rinscct transniitteur. Conipt. rend. soc. biol., 97, 829-832.
Chakravarty, ]\L 1935. Studies on Sporozoa from Indian millipedes. IV. Life-
history of a cephaline gregarine, Hyalosporina ca/nbalopsisae, n. gen., n.

sp. Arch. Protistenk., 86, 211-218.
Chatton, E. 1920. Le peridiniens parasites. Morphologie, reproduction, etholo-

gie. Arch. zool. exptl. et gen., 59, 1-475.
. 1927. La gametogenese meiotic du flagelle Faradiniuin poiicheti.

Covipt. rend., 185, 553-555.
Chatton, E., and B. Biecheler. 1936. Documents nouveaux relatifs aux cocci-

dinides (dinoflagelles parasites). La sexualite du Coccidinium mesnili n.

sp. Covipt. rend., 203, 573-576.
Chejfec, M. 1930. Zur Kenntnis der Kernreorganisationsprozesse bei Farame-

ciiini caudatum. Arch. Protistenk., 70, 87-118.
Chen, T, T. 1936. Observations on mitosis in opalinids (Protozoa, Ciliata). I.

The behavior and individuality of chromosomes and their significance.

Froc. Natl. Acad. Sci. U. S., 22, 594-602.
. 1940a. Polyploidy and its origin in Faraineciinn. J. Heredity, 31, 175-

184.
. 1940b. Conjugation in Paramecium hursaria between animals with

diverse nuclear constitution. /. Heredity, 31, 185-196.
. 1940c. Conjugation of three animals in Paramecium hursaria. Froc.

Natl. Acad. Sci. U. S., 26, 231-238.
. 1940d. Conjugation in Paramecium hursaria between animals with very

different chromosome numbers and between animals with and without

nuclei. Froc. Natl. Acad. Sci. U. S., 26, 243-246.
. 1946a. Conjugation in Paramecium hursaria. I. Conjugation of three

animals. /. MorphoL, 78, 353-395.
. 1946b. Conjugation in Faramecimn hursaria. II. Nuclear phenomena in

lethal conjugation between varieties. /. MorphoL, 79, 125-262.
. 1948. Chromosomes in Opalinidae (Protozoa, Ciliata) with special ref-
erence to their behavior, morphology, individuahty, diploidy, haploidy,

and association with nucleoli. /. MorphoL, 83, 281-357.
. 1951a. Conjugation in Paramedian hursaria. III. Nuclear changes in

conjugation between double monsters and single animals. /. MorphoL,

88, 245-291.
-. 1951b. Conjugation in Paramecium hursaria. IV. Nuclear behavior in

conjugation between old and young clones. /. MorphoL, 88, 293-359.
Cienkowski, L. 1871a. Ueber Schwarmerbildung bei Noctiluca 7/iiliaris. Arch,
viikroskop. Anat., 7, 131-139.

250 SEX IN MICROORGANISMS

. 1871b. Ueber Schwarmerbildung bei Radiolarien. Arch, mikroskop.

Anat., 7, 372-381.

Cleveland, L. R. 1947a. Sex induced in tlie Protozoa of Cryptocercns by molt-
ing. Scie?2ce, 105, 16-17.

. 1947b. The origin and evolution of meiosis. Science, 105, 287-289.

. 1949. Hormone-induced sexual cycles of flagellates. I. Gametogenesis,

fertilization and meiosis in Trichoyiynipha. J. MorphoL, 85, 197-295.

. 1950a. Hormone-induced sexual cycles of flagellates. II. Gametogenesis,

fertilization and one-division meiosis in Oxyinonas. J. MorphoL, 86, 185-
214.

. 1950b. Hormone-induced sexual cycles of flagellates. III. Gametogen-
esis, fertilization and one-division meiosis in Saccinobaculus. J. MorphoL,
86, 215-228.

. 1950c. Hormone-induced sexual cycles of flagellates. IV. Meiosis after

syngamy and before nuclear fusion in Notila. J. MorphoL, 87, 317-348.

. 1950d. Hormone-induced sexual cycles of flagellates. V. Fertilization

in Euco?nonympha. J. MorphoL, 87, 349-368.

. 1951a. Hormone-induced sexual cycles of flagellates. VI. Gametogen-
esis, fertilization, meiosis, oocysts, and gametocysts in Leptospiroiiy//ipha.
J. MorphoL, 88, 194-244.

. 1951b. Hormone-induced sexual cycles of flagellates. VII. One-division

meiosis and autogamy without cell division in Urinyi)ipha. J. MorphoL,
88, 385-440.
-. 1952. Hormone-induced sexual cycles of flagellates. VIII. Meiosis in

Rhynchony7npha in one cytoplasmic and two nuclear divisions followed
by autogamy. /. MorphoL, 91, 269-323.

Cleveland, L. R., S. R. Hall, E. P. Sanders, and J. Collier. 1934. The wood-
feeding roach, Cryptocercns, its Protozoa, and the symbiosis between
Protozoa and roach. Mem. Am. Acad. Arts Sci., 17, 185-342.

Collin, B. 1909. La conjugaison d'Afwplophrya branchiaruDi (Stein) {A. cir-
culans Balbiani). Arch. zooL exptl. et gen., [5] 1, 345-388.

. 1912. Etude monographique sur les Acinetiens. Arch. zooL exptL et

gen., 51, 1-457.

Colwin, L. H. 1944. Binary fission and conjugation in Urceolaria synaptae (?)
Type II (Protozoa, Ciliata) with special reference to the nuclear phe-
nomena. /. MorphoL, 75, 203-249.

Corliss, J. O. 1952a. Nuclear reorganization (autogamy) in a species of Tetra-
hymena. Proc. Soc. ProtozooL, 3, 4.

. 1952b. Le cycle autogamique de Tetrahymena rostrata. Coiupt. re?id.,

235, 399-402.

Culwick, A. T., H. Fairbairn, and R. E. Culwick. 1951. The genetic relation-
ship of the polymorphic trypanosomes and its practical implications. Ann.
Trop. Med. ParasitoL, 45, 11-29.

Cunningham, J. T. 1907. On Kalpidorhynchus areuicolae, a new gregarine
parasite in Arenicola ecaudata. Arch. Protiste?ik., 10, 199-213.

Da Cunha, A. A'l., and J. Muniz. 1930. On the endomixis phenomenon in

SF.X IN PROTOZOA 251

cili;it;i of the ijcmis BaLtnt'uiUiiii, etc. Mem. i//st. OrLi\iliio (/rnz, 23, 1K9-

235.
Dallingcr, \\'. H., and J. Drwsdalc. lS73;i. Researches on tlic life hisroi\ of a

cerconionad; a lesson in biogenesis. Monthly Mlcroscop. ]., 10, 53-5S.
. 1873b. Further researches into the life history of monads. Monthly

Microscop. /., 10, 245-249.
. 1874a. Further researches into the life history of the monads. II.

Monthly Microscop. J., 11, 7-10.
. 1874b. Further researches into the life histor\' of the monads. II (con-
tinued). Monthly Microscop. J., 11, 69-72.
. 1874c. Continued researches into the life history of the monads.

Monthly Microscop. J., 12, 261-269.
. 1875. Further researches into the life history of the monads. Monthly

Microscop. }., 13, 185-197.
Dangeard, P. A. 1886. Researches sur les organisms inferieurs. Aim. sci. nat.,

ser. 7 (Bot.). 4, 241-341.
. 1910. Etudes sur le developpement des organisms inferieurs. Le

Botaniste, ser. 11, 1-311.
-. 1937. Memoire sur le Difflugia globiilosa Dujardin. Le Botaniste^ ser.

28, 229-274.
Debaiseaux, P. 1928. Etudes cytologiques sur quelques microsporidies. La

Cellule, 38, 387-452.
DeGaris, C. F. 1935. Lethal effects of conjugation between Parameciin/i

aurelia and double monsters of P. caudatinn. Am. Naturalist, 69, 87-91.
Dehorne, A. 1920. Contribution a I'etude comparee de Tappareil nucleaire des

infusoires cilies {Paraineciiau candatwn et Colpidiuvi truncatiivi), des

euglenes et des cyanophycees. Arch. zool. exptl. et gen., 60, 47-176.
Dennis, E. W. 1932. The life-cycle of Babesia bigeniina (Smith and Kilbourne)

of Texas cattle-tick fever in the tick, Mar gar opus annulatus (Say). Univ.

Calif. (Berkeley) Pubs. Zool., 36, 263-298.
Deschiens, R. 1933. Recherches sur le cycle evolutif et la culture d' Amoeba

diploidea Hartmann et Nagler (1908). Ann. inst. Pasteur, 50, 695-702.
Devide, Z. 1951. Chromosomes in ciliates. Bull, intern, acad. yougoslave,

Zagreb, n. s. 3, 75-114.
Devide, Z. and L. Geitler. 1947. Die Chromosomen de Ciliaten. Chroiiwsoiiia,

3, 110-136.
Diller, W. F. 1928. Binary fission and endomixis in the Trie hod ina from tad-
poles. /. Morphol. and Physiol., 46, 521-561.

. 1934. Autogamy in Paraineciwn aurelia. Science, 79, 57.

. 1936. Nuclear reorganization processes in Paranieciu7U aurelia, with a

description of autogamy and "hemixis." /. Morphol., 59, 11-67.
. 1940a. Nuclear Behavior and Reproduction in Ciliated Protozoa.

Cytology, Genetics and Evolution. University of Pennsylvania Press,

Philadelphia.
. 1940b. Nuclear variation in Paraiueciwii caudatmn. J. Morphol., 68,

605-633.

252 SEX IN MICROORGANISMS

. 1942. Reconjugation in Farainechmi caudatwn. J. MorphoL, 70, 229-

253.
. 1948. Nuclear behavior of Paraiiiechnii trie hi wn during conjugation.

/. MorphoL, 82, 1-52.
. 1949. An abbreviated conjugation process in Farmnechnn trichiuin.

Biol. Bull, 97, 331-343.
. 1950a. An extra postzygotic nuclear division in Parameciinii caudatwn.

Trails. Am. Microscop. Soc, 69, 309-316.
-. 1950b. Cytological evidence for pronuclear interchange in Faranieciinji

caudatwn. Tratis. Am. Microscop. Soc, 69, 317-323.
Distaso, A. 1908. Sui processi vegetativi e suU'incistadamento di Actinopbrys

sol. Arch. Protistenk., 12, 277-313.
Diwald, K. 1938. Die ungeschlechtliche un geschlechtliche Fortpflanzung von

GlenodifiiwH lubiniensiforme sp. nov. Flora, 132, 174-192.
Dobell, C. C. 1908. The structure and life history of Coprovioyias subtilis, n. g.,

n. sp., a contribution to our knowledge 6f the Flagellata. Quart. J.

Microscop. Sci., 52, 75-120.
. 1925. The life history and chromosome cycle of Aggregata eberthi

(Protozoa, Sporozoa, Coccidia). Farasitol., 17, 1-136.
Doflein, F. 1907. Beobachtungen und Ideen iiber die conjugation der Infusorien.

Sitzber. Ges. MorphoL Physiol., Miivchen, 23, 107-114.
. 1910. Studien sur Naturgeschichte der Protozoan, VI. Experimentelle

Studien iiber die Trypa^iosoma des Froschen. Arch. Protistenk., 19, 207-

231.
Doflein, F., and E. Reichenow. 1949-53. Lehrbuch der Protozoenkunde, 6th

ed. G. Fischer, Jena.
Dogiel, V. 1906. Beitrage sur Kenntnis der Gregarinen. 1. Cystobia chiridotae,

n. sp. Arch. Protistenk., 7, 106-130.
. 1925. Die Geschlechtsprozesse bei Infusorien (speciell bei den Ophryo-

scoleciden), neue Tatsachen und theoretische Erwiigungen. Arch. Pro-
tistenk., 50, 283-442.
Duboscq, O., and B. Collin. 1910. Sur la reproduction sexuee d'un protiste

parasite des tintinnides. CoTiipt. rend., 151, 340-341.
Edmondson, C. H. 1906. The Protozoa of Iowa. Froc. Davenport Acad. Sci.,

11, 1-124.
Ehrenberg, C. G. 1838. Die hifusionsthiere als vollkomniene Organismen. Voss,

Leipzig.
Elkeles, G. 1944. Investigaciones sobre la evolucion del Trypanosoma cruzi,

especialmente sobre la genesis de la forma tripanosomica. Bol. acad. nac.

cienc. Cordoba, 36, 330-408.
Elliott, A. M., and D. L. Nannev. 1952. Conjugation in Tetrahymejia. Science,

116, 33-34.
Ellis, J. M. 1937. The morphology, division and conjugation of the marsh

ciliate Fabrea salina Henneguy. U7iiv. Calif. (Berkeley) Pubs. Zoo/., 41,

343-388.

SFX IN PROTOZOA 253

Elniassian, Al. 1909. Sur VAinocha blattcic, morphologic, generation. Arch.

Protistciik., 16, 143-163.
Elpaticwsky, W. 1907. Zur Fortptlan/.ung von Arcclla vulgaris. Arch. Pro-

tisteiik., 10, 439-466.
Fnriqucs, P. 1908. Die conjugation und sexuellc differenzierung der Infusorien.

II. ^^'icdcrconjuganre und Heniisexe bci Chilodon. Arch. Protistcnk., 12,

213-276.
. 1919. Richcrche sui radiolari coloniali. R. coin, talassogr. ital., mem.

71, 1-177.
Entz, G. 1907. Uebcr dcr Organisationsvcrhaltnisse einiger Peridineen. Math.

iiatwiv. Ber. Ungarii., 25, 246-274.
Erdmann, R. 1925. Endoniixis bei Parauicciuiii bursaria. S. B. Ges. Nattir. Fr.

Bcrl. Jahrg., 1925, 24-27.
Erdmann, R., and L. L. Woodruff. 1916. The periodic reorganization process

in Paraviechwi caudatwii. J. Exptl. ZooL, 20, 59-97.
Fairbairn, H., A. F. Culwick, and F. L. Gee. 1946. A new approach to

Trv'panosomiasis. A'/m. Trop. Med. ParasitoL, 40, 421-452.
Faure-Fremiet, E. 1930. Growth and differentiation of the colonies of Zoo-

thanmion altenians. (Clap. & Lachm.). Biol. Bull., 58, 28-51.
Fiennes, R. N. 1945. A sexual reproduction cycle of T. congolense. Nature,

156, 390.
Finley, H. E. 1943. The conjugation of Vorticella microstoma. Trails. Am.

Microscop. Soc, 62, 97-121.
. 1946. Patterns of sexual reproductive cycles in ciliates. Biodynamica,

6, 33-71.

1952. Sexual differentiation in peritrichous ciliates. /. Morphol., 91,

569-606.
Finley, H. E., and P. A. Nicholas. 1950. Sexual differentiation in Rhabdostyla

vernalis. Proc. Natl. Acad. Sci. U. S., 36, 588-590.
Flu, P. C. 1908. tJber die Flagellaten im Darm von Melophagus ovinus. Arch.

Protistcnk., 12, 147-153.
Foyn, B. 1937. Foraminiferenstudien II. Zur Kenntnis der asexuellen Fort-

pflanzung und der Entwicklung der Gamonten von Discorbina vilarde-

boana d'Orbigny. Bergens Mus. Arbok. Natur. Rekke, 1937, 1-16.
Fritsch, F. E. 1935. The Structure and Reproduction of the Algae, Vol, 1. The

Macmillan Co., New York.
Gabriel, B. 1876. Untersuchungen iiber Morphologie, Zeugung und Entwick-
lung der Protozoen. Morphol. Jahrb., 1, 535-572.
Ganapati, P. N., and P. Tate. 1949. On the gregarine Lankesteria cjilicis

(Ross), 1898, from the mosquito Aedes (Finlaya) geniculatus (Olivier).

Parasitology, 39, 291-294.
Geitler, L. 1935. Uber zweikernige Cysten fon Dijiobryon diver gens. Oesterr.

botan. Z., 84, 282-286.
Gelei, J. von. 1938. Beitrag zur Ciliatenfauna der Umgebung von Szeged. VII.

Paramecium nephridiatum. Arch. Protistcnk., 91, 343-356.

254 SEX IN MICROORGANISMS

Georgevitch, J. 1935. Uber Diplo- und Haplophase im Entwicklungskreise

der Mvxosporidien. Arch. Protisteiik., 84, 419-428.
. 1936. Uber Diplo- und Haplophase im Entwicklungskreise des Myxo-

sporids Zschokkella rovignensis Nemeczek. Arch. Protistenk., 87, 151-154.
Gibbs, A. J. 1944. Chagasella sp. (Protozoa, Coccidia) found in the salivary

glands of Cenaeus carnifex Fabr. (Insecta, Hemiptera). S. African J. Med.

Set. {Biol. Sup pi.) 9, 55-62.
Giese, A. 1938. Size and conjugation in Blepharis7iia. Arch. Protistenk., 91,

125-134.
Glaser, H. 1912. Uber Kernteilung, Encystierung und Reifung von A^ioeba

viira. Arch. Protistenk., 27, 172-194.
Gohre, E. 1934. Untersuchungen iiber den plasmatischen Feinbau der Gregari-

nen mit besonderer Beriicksichtigung der Sexualitatsverhaltnisse. Arch.

Protistenk., 96, 295-324.
Goette, A. 1916. Uber den Lebenszyclus von Diffiugia lobostovia. Arch. Pro-
tistenk., 37, 93-138.
Goldschmidt, R. 1907. Lebensgeschichte der Mastigamoben Mastigella vitrea

und Mastigina setosa. Arch. Protistenk., Suppl. 1, 83-168.
Goodrich, H. P. 1949. Heliospora n. gen. and Rotiindula n. gen., gregarines of

Ganimanis pulex. Quart. J. Microscop. Sci., 90, 27-35.
Granata, L. 1914. Ricerche sul ciclo evolutivo de Haplosporidiiini liinnodrili

Granata. Arch. Protistenk., 35, 47-79.
Grasse, P.-P. 1952-53. Traite de Zoologie, I. Masson & Cie., Paris.
Greef, R. 1874. Peloviyxa palustris (Pelobins), ein amobenartiger Organismus

des siissen Wassers. Arch. Mikroscop. Anat., 10, 51-73.
Gregory, L. H. 1923. The conjugation of Oxytricha fallax. J. Morphol., 73,

555-581.
Greiner, J. 1921. Cytologische Untersuchungen der Gametenbildung und

Befruchtung von Adelea ovata (A. Schneider). Zool. Jahrb. Abt. Anat.,

42, 327-362.
Grell, K. G. 1940. Der Kernphasenwechsel von Stylocephalus (Stylorhynchus)

longicollis F. Stein. Arch. Protistenk., 94, 161-200.
Gross, F. 1934. Zur Biologic und Entwicklungsgeschichte von Noctiluca

miliar is. Arch. Protistenk., 83, 178-196.
Gruber, A. 1887. Die Conjugationsprozess bei Paraviecimn aurelia. Ber.

naturforsch. Ges. Freiburg, 2, 43-60.
Giinther, F. 1928. Uber den Bau und die Lebensweise der Euglenen besonders

der Arten E. terricola, geniculata, proxima, sanguinea und lucens nov. sp.

Arch. Protistenk., 60, 511-590.
Haase, G. 1910. Studien iiber Euglena sanguinea. Arch. Protistenk., 20, 47-59.
Haeckel, E. 1887. Rept. on Radiolaria collected by H. M. S. Challenger. Report

V. 18. Eyre & Spottiswoode, London.
Haecker, V. 1906. Zur Kenntnis der Castanelliden und Porospathiden. Arch.

Protistenk., 8, 52-65.
Hall, R. P. 1925. Mitosis in Ceratitmi hirundinella O. F. M. with notes on

SFX IN PROTOZOA 255

nuclear phenomena in enc\sted fornis and tlie question of sexual repro-
duction. Univ. Calif. (Berkeley) Viibs. Zool., 28, 29-64.
1953. Protozoology. Prentice-Hall, New York.

Hamburger, C. 1904. Die Konjugation von Paraniecitini bursaria. Arch. Pro-

tistcfik., 4, 199-239.
Hartmann, AI. 1943. Die Sexjialitat. G. Fischer, Jena.
Hartmann. AI., and K. Nagler. 1908. Copulation bei Avioeha diploidea mit

selbstandigbleibcn dcr gametcn kerne wiihrcnd des ganzen Lebenszvclus.

Sitzber. Ges. NLiturforsch. Fremuie, Berli//, 1, 112-125.
Hauschka, T. S. 1943. Life history and chromosome cycle of the coccidian

Adelina deronis. }. Morphol., 73, 529-573.
Hausman, L. A. 1920. A contribution to the life-history of Amoeba proteus

Leidy. Biol. Bull., 38, 340-352.
Heidenreich, E. 1935. Untersuchungen an parasitischen Ciliaten aus Wiirmern.

II. Kernveranderungen. Arch. Protistenk., 84, 393-414.
Hentschell, C. C. 1926. On the correlation of the life history of the acephaline

gregarine Gonospora with the sexual cycle of the host. Parasitology, 18,

137-143.
. 1930. On the correlation of the life-history of the acephaline gregarine,

Gonospora, with the sexual cycle of its host. II. Goiiospora (Kalpidor-

hynchiis) arenicolae. Parasitology, 22, 505-509.
Hertwig, R. 1876. Z//r Histologic der Radiolarien. Uiitersuchungeti iiber den

Bail und der Entivicklung der Sphaerozoide?i mid Thalassicolliden. W.

Engelmann, Leipzig.
. 1889. tJber die Conjugation der Infusorien. Abhandl. k. bayr. Akad.

Wiss. II. CI., 17, 151-233.
. 1898. Uber Kernteilung, Richtungskoperbildung und Befruchtung, von

Actinosphaeriwn eichhorni. Abhandl. k. bayr. Akad. Wiss. II. CI., 19,

633-734.
. 1899. Uber encystierung und Kernvermehrung bei Arcella vidgaris

Ehrb. Festschr. Knpffer.

1914. Uber Parthenogenesis der Infusorien und die Depressionzustiinde

der Protozoen. Biol. Zcfitr., 34, 557-581.
Hoare, C. A. 1936. Morphological and taxonomic studies on mammalian

trypanosomes. II. Trypanosoma simiae and acute porcine trypanosomiasis

in tropical Africa. Trajjs. Roy. Soc. Trap. Med. Hyg., 21, 619-645.
Hofker, I. 1930. Uber Noctiluca scintellans. Arch. Protistenk., 71, 57-78.
Horvath, J. 1947. The question of the equality of somatic and germ nuclei

in respect to heredity and survival, on the basis of studies in a soil

protozoon. Arch. Biol. Hung., Ser. II, 17, 193-202.
. 1948a. Experimentelle Hervorbringung mikronucleuslosen Ciliaten.

Arch. Biol. Himg., Ser. II, 18, 1-6.
. 1948b. Uber Konjugation von Individuen mit kompletter Kerngarnitur

und mikronucleuslosen Tieren einer Bodenciliatenart. Arch. Biol. Hung.,

Ser. II, 18, 7-10.

256 SEX IN MICROORGANISMS

Hulpieu, H. R., and D. L. Hopkins. 1927. Observations on the life-history of
A?noeba proteus. Biol. Bull., 52, 411-417.

Huth, W. 1913. Zur Entwicklungsgeschichte der Thalassicollen. Arch. Pro-
tistenk., 30, 1-124.

Ilowaisky, S. A. 1926. Uber die Kernprozesse des getrennten Conjugaten der
Stylonichia inytilus und Fara7necimn caudatum. Arch. Protistenk., 53,
243-252.

Ishikawa, C. 1891. Vorlaufige Mitteilungen iiber die Konjugationserscheinungen
bei den Noctiluceen. Zool. Anz., 14, 12-14.

Ivanic, M. 1928. Uber die mit den parthenogenetischen Reorganisationsprozes-
sen des Kernapparates verbundenen Vermehrungscysten von Chilodon
uncinatus. Arch. Protistenk., 61, 293-348.

. 1929. Uber die centrosomenahnlichen Gebilde bei der Grosskernteilung

und die promitotische Kleinkernteilung, nebst Bemerkungen iiber die
Chromosomenverhaltnisse bei einem Infusor {Euplotes patella Ehrbg.)
Arch. Protistenk., 66, 33-60.

. 1933a. Neue Beitrage sur Kenntnis der mit den Reorganisationsprozes-

sen des Kernapparates verbundenen Vermehrungsruhestadien von Chilo-
don uncinatus Ehrbg., etc. Arch. Protistenk., 79, 170-199.

. 1933b. Die conjugation von Chilodon cucullus Ehrb. Arch. Protistenk.,

79, 313-346.

. 1935a. Uber die "voUkommene Kopulation" bei Cochliopodium digi-

tatuni Penard, nebst Bemerkungen iiber die Kopulation bei Siisswasser-
thalamorphoren im allgemeinen. Biol. 'Lentr., 55, 225-245.

. 1935b. Uber das Vorkommen des "Sichelstadiums" bei den mit den

parthenogenetischen Reorganisationsprozessen des Kernapparates ver-
bundenen Vermehrungsruhestadien von Chilodon uncinatus Ehrbg. und
dessen Bedeutung. Arch. Protistenk., 85, 329-333.
-. 1936. Zur Kenntnis der Entwicklungsgeschichte bei Mastigina hylae

Frenzel. Arch. Protistenk., 87, 225-241.

Jameson, A. P. 1920. The chromosome cycle in Gregarines, with special ref-
erence to Diplocystis schneideri Kunstler. Quart. J. Microscop. Sci., 64,
207-266.

Jennings, H. S. 1911. Assortive mating, variability and inheritance of size in
the conjugation of Paramecium. /. Exptl. Zool., 11, 1-134.

. 1920. Life and death, heredity and evolution in unicellular organisms.

Badger, Boston.

. 1944. Paramecium bursaria. Life history. III. Repeated conjugation in

the same stock at different ages, with and without inbreeding, in relation
to mortality at conjugation. /. Exptl. Zool, 96, 243-273.

Jickeli, F. 1884. Uber Copulation von Difflugia globidosa. Zool. Anz., 7,
449-451.

Johnson, P. L. 1930. Reproduction in Amoeba proteus. Arch. Protiste7ik., 71,
463-498.

Jones, P. M. 1928. Life cycle of Amoeba proteus (Chaos diffluens) with spe-
cial reference to the sexual stage. Arch. Protistenk., 63, 322-332.

SI X IN PRoro/oA 257

Joseph, G. 1879. iJbcr CuH)ttcn-lnfusoricn. 7.o6\. Aiiz., 2, 114-118.
Jov^et-Lavergne, P. 1926. Rechcrclies sur Ic c\toplasnie des sporozoaircs. Arch.

aiiat. Diicroscop., 22, 1-128.

. 1931. La Pbysico-chcviic tic At scx/iiilitc. Berlin, Gebriider Borntrager.

Kay, M. W. 1946. Studies on Oxytricl?a hif\viii. III. Conjugation. Trans. Avi.

Microscop. Soc, 65, 132-148.
Kent, S. 1880-82. Manual of the hifiisoria. D. Bogue, London.
Kidder, G. W. 1933. Studies on CoiicophtJ?irhis TiiytUi De Morgan. II. Con-
jugation and nuclear reorganization. Arch. Protistenk., 79, 25-49.
. 1938. Nuclear reorganization without cell division in Faraclevelandia

simplex, an endoconiniensal cihate of the wood-feeding roach, Panesthia.

Arch. Protistenk., 91, 69-77.
Klee, E. E., 1926. Der Formwechsel ini Lebenskreis reiner Linien von Euplotes

longipes. Zoo/. Jahrb., Abt. allgevi. Xool., 42, 307-366.
Klitzke, M. 1914. Uber Wiederconjugaten bei Paramecium caudatum. Arch.

Protistenk., 33, 1-20.
Konsuloff, S. 1921. Untersuchungen iiber Opalina. Arch. Protistenk., 44, 285-

345.
Korotneff, A. 1879. Etudes sur les rhizopodes. Arch. zool. exptl. et gen., 8,

467-482.
Koster, W. 1933. Untersuchungen iiber Teilung und Conjugation bei Para-
median vndtiniicronncleatiini. Arch. Protistenk., 80, 410-433.
Krichenbauer, H. 1937. Beitrag zur Kenntnis der Morphologic und Entwick-

lungsgeschichte der Gattungen Englena und Phacus. Arch. Protistenk.,

90, 88-122.
Krieger, W. 1930. Untersuchungen iiber Plankton-Chrysophyceen. Botan.

Arch., 29, 257-329.
Kropp, B. 1939. Sexual and asexual nuclear division in a binucleate amoeba.

A7}at. Record, 73, (Suppl. 2), 33.
Kudo, R. 1924a. Studies on Microsporidia parasitic in mosquitoes. III. On

Thelohania legeri Hesse (= Th. illinoisensis Kudo). Arch. Protistenk., 49,

147-162.
. 1924b. Studies on Microsporidia parasitic in mosquitoes. VI. On the

development of Thelohaiiia opacita, a culicine parasite. /. Parasitol., 11,

84-89.
. 1925. Studies on Microsporidia parasitic in mosquitoes. V. Further

observations upon Stempellia (Thelohania) magna Kudo, parasitic in

Culex pipiens and C. territans. Biol. Bull., 48, 112-127.
. 1930. Studies on Microsporidia parasitic in mosquitoes. VIII. On a

microsporidian, Nosema aedis nov. spec, parasitic in a larva of Aedes

aegypti of Porto Rico. Arch. Protistenk., 69, 23-38.

. 1946. Protozoology (3d ed.) Charles C Thomas, Springfield, 111.

Kuhn, A. 1921. Morphologic der Tiere in Bildern. I. Protozoen, 1, Flagellaten.

Gebriider Borntrager, Berlin.
. 1926. Morphologic der Tiere in Bilder??. 2. Protozoen, 2, Rhizopoden.

Gebriider Borntrager, Berlin.

258 SEX IN MICROORGANISMS

Labbe, A. 1899. Sporozoa. Das Thierreich, Lief. 5. R. Friedlander und Sohn,

Berlin.
Lackey, J. B. 1938. Scioto River Valley forms of Chrysococciis. Am. Midland

Naturalist, 20, 619-623.
Landis, E. M. 1925. Conjugation of Paramedian 7/ntltiinicro7mcleata. Powers

and Mitchell. /. Morphol. and Physiol., 40, 111-167.
Lebedeff, W. 1910. Uber TrypanosoDia rotatorium Gruby. Festschr. 60 Ge-

burtstag R. Hertwig, 1, 397-430.
Le Calvez, J. 1946. Place de la reduction chromatique et alternance de phases

nucleaires dans le cycle des foraminiferes. Compt. rend., 222, 612-614.
. 1950. Recherches sur les foraminiferes. II. Place de la meiose et sexua-

lite. Arch. zool. exptl. et gen., 87, 211-244.
Leger, L. 1904. La reproduction sexuee chez les Stylorhynchus. Arch. Pro-

tistenk., 3, 303-357.
Leger, L., and O. Duboscq. 1903. La reproduction sexuee chez Pterocephalus.

Arch. zool. exptl. et gen., Ser. 4[1] Notes et Rev.), 141-147.
. 1909. Etudes sur la sexualite chez les gregarines. Arch. Protistenk., 17,

19-127.
Leidy, J. 1879. Freshwater rhizopods of North America. U. S. Geol. Survey,

12, 1-324.
Lister, J. J. 1895. Contributions to the life-history of the Foraminifera. Phil.

Trans. Roy. Soc, B186, 401-453.
Lucas, C. L. T. 1927. Two new species of amoeba found in the cockroaches

with notes on the cvsts of Nyctotherus oralis Leidv. Parasitology, 19,

223-235.
Luyet, B. J. 1950. A comparative study of the life cycles of lower Protozoa

and Protophyta. Biodynamica, 6, 265-364.
MacDougall, M. S. 1925. Cytological observations on gymnostomatous ciliates,

with a description of the maturation phenomena in diploid and tetraploid

forms of Chilodon uncinatus. Quart. J. Microscop. Sci., 69, 361-384.
. 1936. Etude cytologique de trois especes du genre Chilodo7iella Strand.

Morphologic, conjugaison, reorganisation. Bitll. hiol. Frarice Belg., 70,

308-331.
Mackinnon, D. L., and H. N. Ray. 1931. Observations on dicvstid gregarines

from marine worms. Quart. J. Microscop. Sci., 74, 439-466.
. 1937. A coccidian from the egg of Thalassema neptuni Gaertner.

Parasitology, 29, 457-468.
Mack, B. 1951. Morphologische und entwichlungsgeschichtliche Untersuch-

ungenen an Chrysophyceen. Oesterr. Botan. Z., 98, 249-279.
Mainx, F. 1928. Beitrage zur Morphologie und Physiologic der Eugleninen. I,

II. Arch. Protistenk., 60, 305-414.
Martin, C. H. 1912. A note on the Protozoa from sick soils, with some account

of the life-cycle of a flagellate monad. Proc. Roy. Soc. (Lojidon), B85,

393-400.
Maupas, E. 1889. Le rajeunissement karvogamique chez les cilies. Arch. zool.

exptl. et gen., 2 Sen, 7, 149-517.

SEX IN PROTOZOA 259

iVIeglitch, F. A. 1940. Cvtological observations on Eiidaniocba hlattac. III. Biol.

Moiiofir., 17, 1-147.
Mcrcicr, L. 1909. Le c\cle cvolutif d'Aiiiocba blattae Biitschli. Arch. Vrotls-

tc/ik., 16, 164-168. '
. 1910. Contribution a I'etudc de I'aniibe dc la blattc {Entamoeba blattae

Biitschli). Arch. Frotistenk., 20, 143-175.
Mctcalf, iM. iM. 1908. Opalhia. Arch. Protiste//k., 13, 195-375.
Alinchin, E. A. 1912. An Introduction to the Study of the Protozoa. Edward

Arnold, London.
iMivashita, Y. 1928. On a new parasitic ciliatc Lada tanishi n. sp., with pre-

Iiniinar\" notes on its heteroganiic conjugation. Japan J. Zoo/., 1, 205-218.
Moore, E. L. 1924. A further stud\' of the effects of conjugation and encyst-

ment in Spathidiuvi spat hula. J. Exptl. "Zool., 40, 217-230.
Moore, J. E. S., and A. Breinl. 1908. The life-history of Trypanosoma equiper-

duiu. Proc. Roy. Soc. (London), B80, 288-298.
Moore, J. E. S., A. Breinl, and E. Hindle. 1908. The life history of Trypanosoma

lezi-isi. Ann. Trap. Med. ParasitoL, 2, 197-220.
Alorris, S. 1936. Studies of Endamoeba blattae (Biitschli). /. Morphol., 59,

225-263.
Miihl, D. 1921. Beitrag zur Kenntnis der Morphologic und Physiologic der

Mehlwurmgregarinen. Arch. Protistenk., 43, 361-414.
Miiller, W. 1932. Cytologische und vergleichend-physiologische Untersuch-

ungen iiber Paramecitim vniltii}iicro7iucleatu7n und P. caudatuni. Zugleich

ein Versuch zur Kreuzung beider Arten. Arch. Protistenk., 78, 361-462.
Mulsow% K. 1911. tJber Fortpflanzungserscheinungen bei Monocystis rostrata

n. sp. Arch. Protistenk., 22, 20-55.
Muniz, J. 1927. Quelques formes interessantes trouvees dans les cultures de

Trypanosoma criizi. Coinpt. rend. soc. biol., 97, 821-823.
Myers, E. H. 1935. The life history of Patellina corrugata, a foraniinifer.

Scripps Inst. Oceanogr. Bull., Tech. Ser., 3, 355-392.
. 1936. The life cycle of Spirillina vivipara, with notes on morphogenesis,

systematics and distribution. /. Roy. Microscop. Soc, 56, 120-146.

1938. The present state of our knowledge concerning the life cycle

of the Foraminifera. Proc. Natl. Acad. Sci. U. S., 24, 10-17.
Nabih, A. 1938. Studien iiber die Gattung Klossia und beschreibung des

Lebenzyklus von Klossia loosi (nov. sp.). Arch. Protistenk., 91, 474-515.
Nagler, K. 1909. Entwicklungsgeschichtliche Studien iiber Amoben. Arch.

Protistenk., 15, 1-53.
Naville, A. 1927a. Le cycle chromosomique d^Urospora lagidis (De Saint

Joseph). Parasitology, 19, 100-138.
. 1927b. Recherches sur le cycle evolutif et chromosomique de Klossia

helicijia (A. Schneider). Arch. Protistenk., 57, 427-474.
. 1927c. Le cycle chromosomique et la meiose chez les Monocystis. Z.

v^iss. Biol. B. Zeitschr. Zellforsch. inikroskop. Anat., 6, 257-284.
. 1930a. Recherches sur la sexualite chez les Myxosporidies. Arch. Pro-
tistenk., 69, 327-400.

260 SEX IN MICROORGANISMS

. 1930b. Le cycle chromosomique d'une nouvelle actinomyxidie:

Guyenotia spbaendosa n. gen., n. sp. Quart. J. Microscop. Sci., 73, 547-
576.
-. 1931. Les Sporozoaires (cycle chromosomiques et sexualite). Mem.

soc. physique d^hist. nat. Geneve, 41, 1-223.
Nelson, E. C. 1934. Observations and experiments on conjugation of Balan-

tidhnii from the chimpanzee. Am. J. Hyg., 20, 106-134.
Neresheimer, E. 1907. Die Fortpflanzung der Opalinen. Arch. Protistenk.,

Suppl. 1, 1-42.
Noble, E. R. 1938. The life cycle of Zygosovia globosum sp. nov., a gregarine

parasite of Urechis caupo. Univ. Calif. {Berkeley) Pubs. Zoo/., 43, 41-66.
. 1941. Nuclear cycles in the life history of the protozoan genus Cera-

to?nyxa. J. MorphoL, 69, 455-479.

1944. Life cycles in the Myxosporidia. Quart. Rev. Biol., 19, 213-235.

Noland, L. E. 1927. Conjugation in the ciliate, Metopus sigmoides C. and L.
/. Morphol. and Physiol., 44, 341-362.

Nowlin, N. 1922. Correlation of the life-cycle of a parasite with the meta-
morphosis of its host. /. ParasitoL, 8, 153-160.

Padnos, M., and R. F. Nigrelli. 1942. Trichodiiia spheroidesi and Trichodina
halli spp. nov. parasitic on the gills and skin of marine fishes, with special
reference to the life-history of T. spheroidesi. Zoologica, 27, 65-72.

Paehler, F. 1904. Uber die Morphologic, Fortpflanzung und Entwicklung von
Gregarina ovata. Arch. Protistenk., 4, 64-87.

Patefi^, P. 1926. Fortpflanzungserscheinigungen bei Difjiugia mannnilaris Penard
und Clypeolina ?narginata Penard. Arch. Protistenk., 55, 516-544.

Patton, W. S. 1922. Some reflections on the kala-azar and oriental sore prob-
lems. Ind. J. Med. Research, 9, 496-532.

Pearl, R. 1907. A biometrical study of conjugation in Paramecium. Biotnetrika,
5, 213-297.

Penard, E. 1902. Faune rhizopodique du bassi?i du Leman. Kiindig, Geneva.

. 1904a. Etude sur la Chlamydoi}iyxa montana. Arch. Protistenk., 4,

296-334.

. 1904b. hes heliozoaires d''eau douce. Kiindig, Geneva.

-. 1938. Les Infinime?7t petits dans leur manifestatio7is vitales. Georg &

Cie., Geneva.
Penn, A. B. K. 1937. Reinvestigation into the cytology of conjugation in

Paramecium caudatum. Arch. Protistenk., 89, 45-54.
Phillips, N. E., and D. L. Mackinnon. 1946. Observations on a moncysted

gregarine, Apolocystis elongata n. sp. in the seminal vesicles of Eisenia

foetida. Parasitology, 37, 65-74.
Plate, L. 1886. Untersuchungen einiges an den Kiemenblattern des Ganimarus

pulex lebenden Ektoparasiten. Z. iviss. ZooL, 43, 175-241.
Popoff, M. 1911. Uber der Entwicklungscyclus von Amoeba mi7iuta. Arch.

Protistenk., 22, 197-223.
Pratje, A. 1921. Noctiluca miliaris Suriray, etc. Arch. Protistenk., 42, 1-98,

422-438.

SEX IN PROTOZOA 261

Picnant, iM. 1935. Lcgons dc Zoologic, Protozoaircs, I/ifiisoircs cilics. Hermann

& Cie, Paris.
Prowazek, S. von. 1903. Flagcliatcn Studicn. Arch. Protistenk., 2, 195-212.
. 1904a. Die Entwicklung von Herpetoino7ias, einen mit der Trypano-

sonicn verwandrcn Flagcliatcn. Arh. kaiserl. Gcsuvdh., 20, 440-452.
. 1904b. Untcrsuchungcn iiber einige parasitische Flagcllaten. Arh.

kaiserl. Gesundh., 21, 1-41.

1905. Studicn iibcr Saugcsticrtr\'panos()nicn. Arh. kaiserl. Gesundh.,

22, 351-394.

Prythcrch, H. F. 1940. The life cycle and morpliology of Nematopsis ostre-
arwn sp. nov., a gregarinc parasite of the mud crab and oyster. /. Mor-
phol., 66, 39-65.

Rav, H., and M. Chakravartv. 1933. Studies on Sporozoa from Indian milli-
pedes. II. Arch. Protistenk., 81, 352-360.

Reichenow, E. 1921. Die Hamococcidien der Eidechsen, Vorbemerkungen I.
Teil: Die Entwicklungsgeschichte von Karyolysus. Arch. Protistenk., 42,
179-291.

Reukauf, E. 1912. Zur Encystierung von Euglypha alveolata. Zool. Anz., 39,
372-375.

Rhumbler, L. 1895. Beitrage zur Kenntnis der Rhizopoden. III-V. Z. wiss.
Zool., 61, 38-110.

. 1898. Zelleib-, Schalen- und Kern-verschnielzungen bei den Rhizopo-
den, etc. Biol. Centralhl., 18, 21-26, 33-38, 69-86, 113-130.

Rosenberg, L. E. 1940. Conjugation in Opisthonecta he?meguyi, a free-swim-
ming vorticellid. Proc. Am. Phil. Soc, 82, 437-447.

Roskin, G., and S. Schischliaiewa. 1928. Die Kernteilung bei Trypanosomen.
Arch. Protistenk., 60, 460-481.

Russo, A. 1926. I due cicli sessuati di "Cryptochihnn echini Maupas" si svol-
gono con due categoric d' individui, che si dividono e si rinnovano con
modalita diverse. Atti accad. naz. Lincei Classe sci. fis. 7nat. e nat., 3,
720-723.

Schaudinn, F. 1896. Uber die Kopulation von Actifiophrys sol Ehrbg. Sitzher.
Akad. Wiss. Berlin, 1896 [I], 83-89.

. 1903. Untersuchungen iiber die Fortpflanzung einiger Rhizopoden.

Arh. kaiserl. Gesundh., 19, 547-576.

. 1904. Generations- und Wirtswechsel bei Trypanoso7/m und Spiro-

chaete. Arh. kaiserl. Gesundh., 20, 387-439.
Schellack, C. 1907. Uber die Entwicklung und Fortpflanzung von Echinomera

hispida. Arch. Protistenk., 9, 297-345.
. 1912. Die Gregarlnen. In: Prowazek-Noller. Hdhk. path. Protozoen,

I, 487-514.
Schepatief, A. 1910. Amobenstudien. Zool. Jahrh., Aht. Anat. Ontog., 29,

485-526.
Schewiakoff, W. 1926. Die Acantharia des Golfes von Neapel. Fauna e Flora

del Golfo di Napoli, Monogr., 37, 1-755.

262 SEX IN MICROORGANISMS

Schiller, J. 1925. Die planktontisschen Vegetationen der Adriatischen Meeres.

Arch. Protiste?ik., 51, 1-130.
. 1926. ijber Fortpflanzung, geissellos Gattungen und die Nomenklator

der Coccolithophoracean nebst Mitteilung iiber Copulation bei D'mo-

bryon. Arch. Protisteiik., 53, 326-342.
Schilling, C. 1910. Das Vorkommen von Autoganiie bei Trypavosoiiia leivisi.

Arch. Protjstenk., 19, 119-126.
Schirch, P. 1914. Beitrage zur Kenntnis der Lebenscyclus von Arcella vul-
garis Ehrbg. und Pelojnyxa paliistris Greef. Arch. Protistenk., 33, 247-

271.
Schnitzler, H. 1905. Uber die Fortpflanzung von Clepsidrma ovata. Arch.

Protiste?ik., 6, 309-333.
Schwarz, E. 1932. Der Forniwechsel von Ochrosphaera veapoUtava. Arch.

Protistenk., 77, 434-462.
Scott, M. J. 1927. Studies on the Balantidiiim from the Guinea-pig. /. MorphoL,

44, 417-465.
Seshachar, B. R. 1946. Nuclear reorganization in Epistylis. Current Sci., 15,

198.
Shortt, H. E., and C. S. Swaminath. 1927. Monocystis niackiei n. sp., parasitic

in Phlebotonms argentipes, Ann. & Brun. Ind. J. Med. Research, 15, 439-

552.
Skuja, H. 1950. Korperbau und Reproduktion bei Dinobryo7i borgei. Sverisk

Botan. Tid., 44, 96-107.
Smith, G. M. 1950. The Freshwater Algae of the United States, 2d. ed.

McGraw-Hill Book Co., New^ York.

. 1951a. Manual of Phycology. Chronica Botanica Co., Waltham, Mass.

. 1951b. Sexuality of Algae. In: Manual of Phycology. Chronica Botanica

Co., Waltham, Mass.
Sonneborn, T. M. 1936. Factors determining conjugation in Paravieciuvi

aurelia. I. The cyclical factor: The recency of nuclear reorganization.

Genetics, 21, 503-514.

. 1937. Sex, sex inheritance and sex determination in Paramecium

aurelia. Proc. Natl. Acad. Sci. U. S., 23, 378-385.
. 1947. Recent advances in the genetics of Paraniecium and Euplotes.

Advances in Genetics, 1, 263-358.

1951. Some current problems of genetics in the light of investigations

on Chlamydoinonas and Paramecimii. Cold Spring Harbor Symposia
Quant. Biol, 16, 483-503.

Sprague, V. 1941. Studies on Gregarina blattarum with particular reference
to the chromosome cycle. Univ. Illinois Biol. Monogr., 18, 1-57.

Stein, F. 1883. Der Orgauisnms der Infusionsthiere. Ill Abt. W. Engelmann,
Leipzig.

Stranghoner, E. 1932. Teilungsrate und Kernreorganisationsprozess bei Para-
mecium multimicromicleatum. Arch. Protistenk., 78, 302-360.

Studitsky, A. N. 1932. tjber die Morphologic, Cytologic und Systematik von
Ptychostonmm chattoni Rossolimo. Arch. Protistenk., 76, 188-216.

SI X IN FRoro/oA 263

Summers, I". Al., and (1. W. Kiiklcr. l'>.>6. Taxononiic and cytological studies

on ciliatcs associated with the ampliipod family Orchestiidae from the

Woods Hole district. Arch. Protistciik., 86, 377-403.
Swarc/.ewskv, B. 1908. ijber die Fortpflanzungserscheinigungen, bei Arcella

vulgaris F.hrbg. Arch. Frotistciik., 12, 173-212.
. 1914. iJber den Lebcnscyclus einiger Haplosporidien. Arch. Frotistciik.,

33, 49-108.
-. 1928. Beobachtungcn (iber Spirochoim elcgans. Arch. Protistenk., 61,

185-222.
Tittler, I. A. 1935. Division, encystment and endomixis in Urostyla grmidis

with an account of an amicronucleate race. La Cellule, 44, 189-218.
Tregouboff, G. 1914. Sur revolution sexuelle de Ste7iophora juli A. Schneider

et la position s\stematique de la famille des Stenophorides Leger et

Duboscq. Arch. zool. exptl. et gen., 54 (Notes et Rev.), 19-29.
Turner, J. P. 1930. Division and conjugation in Euplotes patella with special

reference to the nuclear phenomena. Univ. Calif. (Berkeley) Pubs. Zool.,

33, 193-258.
. 1941. Fertilization. In: Calkins and Summers, Protozoa in Biological

Research. Columbia University Press, New York.
\^alkanov, A. 1934. Die Kernverhaltnisse der Opaliniden. Arch. Protistenk., 83,

356-366.
. 1935. Untersuchungen iiber den Entwicklungskreis eines Turbellarien

parasiten Monocy Stella artidti. Z. Parasitenk. 7, 517-538.
Vanderplank, F. L. 1944. Identification of trypanosomes by chromosomes.

Nature, 154, 19-20.
. 1947. The chromosomes and part of the sexual cycle of Trypanosonia

rhodesiense. Trans. Roy. Soc. Trop. Med. Hyg., 41, 15-16.
X'^elev, L. 1905. A further contribution to the studv of Peloiiiyxa palustris

Greef. /. Linnean Soc, London, 29, 374-395.
Verworn, M. 1888. Biologische Protisten-studien. I. Z. iviss. Zool. 46, 455-470.

. 1890. Biologische Protisten-Studien. II. Z. wiss. Zool., 50, 443-468.

Visscher, J. P. 1927. Conjugation in the ciliated protozoon, Dileptus gigas,

with special reference to the nuclear phenomena. /. Morphol., 44, 383-414.
W'edekind, G. 1927. Zvtologische Untersuchungen an Barrouxia schneideri.

Z. Zellforsch. iv. jnikroskop., Anat., 5, 505-595.
Weiser, J. 1943. Zur Kenntnis der Mikrosporidien aus Chironomiden-Larven.

Zool. Anz., 141, 255-264.
W'eisz, Paul B. 1950. Multiconjugation in Blepharisvia. Biol. Bull. 98 [3],

242-246.
Wenrich, D. H. 1926. The structure and division of Parameciwu trichimn. J.

Morphol. and Physiol, 43, 81-103.
\\'enrich, D. H., and C. C. Wang. 1928. The occurrence of conjugation in

Paranieciiiin calkinsi. Scie?ice, 67, 270-271.
Wenyon, C. M. 1911. Oriental sore in Bagdad, together with observations on a

gregarine in Stego?Jiyia fasciata, the haemogregarine of dogs and the

flagellates of house flies. Parasitology, 4, 273-344.

264 SEX IN MICROORGANISAIS

, 1920. Observations on the intestinal Protozoa of three Egyptian Uzards,

with a note on a cell-invading fungus. Parasitology, 12, 350-365.
1926. Protozoology. Wm. Wood & Co., New York.

Weschenfelder, R. 1938. Die Entwicklung von Actinocephalus parvus Well-
mer. Arch. Protistenk., 91, 1-60.

Wichterman, R. 1937. Division and conjugation in Nyctotherus cordiformis
(Ehrbg.) Stein (Protozoa, CiUata) with special reference to the nuclear
phenomena. /. Morph., 60, 563-611.

. 1939. Cytogamy, a new sexual process in joined pairs of Para?necmni

caudatmn. Nature, 144, 123-124.

. 1940. Cytogamy, a sexual process occurring in living joined pairs of

Parameciufn caudatmn and its relation to other sexual phenomena. /.
MorphoL, 66, 423-451.

. 1946. Further evidence of polyploidy in conjugation of green and color-
less Paramecimn bursaria. Biol. Bull., 91, 234.

. 1948a. The time schedule of mating and nuclear events in the con-
jugation of Para?neciu7n bursaria. Turtox News, 26, 2-10.

. 1948b. Mating types and conjugation of four different races of Para-

mecium calkinsi and the effect of X-rays on the mating reaction. Biol.
Bull., 95, 271-272.

1953. The Biology of Paramecium. The Blakiston Co., New York.

Wilber, C. G. 1947. Type of reproductive cycle in multinucleate amoebae.

BiodyiiaiJiica, 6, 111-123.
Willis, A. G. 1948. Studies on Lagenophrys tatter salli (Ciliata, Peritricha,

Vorticellidae). II. Observations on bionomics, conjugation and apparent

endomixis. Quart. J. Microscop. Sci., 89, 385-400.
Winter, F. W. 1907. Zur Kenntnis der Thalamorphen. Arch. Protistenk., 10,

1-113.
Wolcott, G. B. 1952. Mitosis in Trypanosoma lewisi. J. MorphoL, 90, 189-199.
Woodcock, H. M. 1916. Observations on coprozoic flagellates, together with

a suggestion as to the significance of the kinetonucleus in the Binucleata.

Phil. Trans. Roy. Soc. B207, 379-412.
. 1921. Notes on coprozoic flagellates. I, II. Arch. zool. exptl. et gen. 60

(Notes et Rev.), 11-18.
Woodcock, H. M., and G. Lapage. 1915. Observations on the life cycle of a

new flagellate, Helkesiiiiastix foecicola n. g., n. sp. Together with remarks

on the question of syngamy in the trypanosomes. Proc. Roy. Soc. (Lon-
don), B88, 253-370.
Woodruff, L. L. 1921. The structure, life history and intrageneric relationships

of Para7iieciwn calkinsi sp. nov. Biol. Bull., 41, 171-180.
. 1941. Endomixis. In: Calkins and Summers, Protozoa in Biological

Research. Columbia University Press, New York.
Woodruff, L. L., and R. Erdmann. 1914. A normal periodic reorganization

process without cell fusion in Paravieciuni. J. Exptl. Zool., 17, 425-518.
Woodruff, L. L., and H. Spencer. 1923. ParaTuecium polycaryum, sp. nov.

Proc. Soc. Exptl. Biol. Med., 20, 338-339.

Sl'X IN PROIOZOA 265

Yagiu, R. 1940. The division, conjugation and nuclear reorganization of

Entorhipidhivi echini Lvnch (Infusoria). /. Sci. Hiroshima Univ. ser. B.

Div. 1 (Zoo!.), 7, 125-256.
Yarwood, E. A. 1937. The life cvcle of Acielina cryptocerci sp. nov., a coc-

cidian parasite of the roach, Cryptocercus punctulatus. Parasitology, 29,

370-390.
Zederbauer, E. 1904. Gcschlcchtlichc und ungeschlechtliche I<\)rtpflanzung von

Ccratiuvi hirwidiiicllci. Vcrhavdl. dent, bat an. Ges., 22, 1-8.
Zellcr, E. 1877. Untcrsuchungen iiber die Fortpflanzung und Entwicklung der

unseren batrachian schmarotzenden Opalinen. Z. iviss. Zool., 29, 352-380.
Zuelzer, M. 1904. Bcitriige zur Kenntnis von Difflugia itrceolata Carter. Arch.

Protistenk., 4, 240-295.
. 1909. Bau und Entwicklung von Wagiierella burealis Mereschk. Arch.

Protistenk., 17, 135-212.

Mating Type Determination
in Paramecium aurelia
A Study in Cellular Heredity *

DAVID L. NANNEYf Indiana University, Bloomington,
and University of Michigan, Ann Arbor

The genetics of Parajneciimi, in its modern sense, had its beginning
in the discovery of mating types in P. aurelia (Sonneborn, 1937),
and the determination and inheritance of mating type was the first
problem attacked with the newly acquired possibility of cross-
breeding analysis. Although many fundamental discoveries essential
for the understanding of this problem were made by Sonneborn
(1947), the determination and inheritance of mating types in this
species have remained less well understood than the genetics of other
characteristics, such as the killer and the antigenic traits (Sonneborn,
1950b).

Two major difficulties have been encountered in trying to un-
derstand the mating type system. One of these has been to account
for differences among nuclei arising by mitotic divisions from a
common source. Another has been to account for what seemed to be
two very different systems of determination and inheritance of mat-
ing type in different races of this species. Recently (Nanney, 1953)

* A contribution (No. 509) from the Zoology Department, Indiana Uni-
versity and from the University of Michigan.

t Part of the work reported was accomplished with the aid of a Pre-
doctoral Research Fellowship from the National Cancer Institute, Public
Health Service, Federal Security Agency. The work was done in part under
grants to Professor T. M. Sonneborn from the American Cancer Society
through the Committee on Growth of the National Research Council, from
Indiana University, and from the Rockefeller Foundation. The work at the
University of Michigan has been aided by a Faculty Research Grant from
the Rackham School of Graduate Studies and by a grant from the National
Science Foundation.

The author wishes to express his gratitude to Professor T. M. Sonneborn
and the entire group of Paraviecimn investigators at Indiana University for
their contributions in developing the ideas set forth in this paper.

266

MA riXC; ]\P\ 1)1.1 I'RMINA 1 ION 267

an arrcnipr w as iiiadc to give a complete, if formal, explanation for
these and other problems. Although this explanation appeared prom-
isino- and early attempts at verification were encouraging (Sonnebom,
1951 ), more recent studies have shown that the interpretation is cer-
tainl\- not correct in detail. On the other hand, strong support has
been received for the more general features of the model proposed.
Before considering any explanation in detail, it is first of all nec-
essary to revie\\' the available information concerning mating type
determination in this species. Many of the observations are not avail-
able in the literature, and the author is deeply indebted to Prof. T. M.
Sonneborn for the use of his unpublished data. It is to be understood
that the present treatment of Sonneborn's unpublished observations
is subject to his later modification or extension. After reviewing the
observations we will return to a consideration of detailed explana-
tions.

PATTERNS OF MATING TYPE DETERMINATION AND
INHERITANCE

General Background, the Varietal System and
Cytogenetic Considerations

Paraviec'min aiirelia is a "taxonomic" species composed of at least
eight so-called varieties which are themselves effective genetic spe-
cies (Sonneborn and Dippell, 1946; Sonneborn, 1947, 1950a). (See
also Metz, 1954.) Each variety contains no more than two mating
types and in relatively few combinations can these conjugate with
the mating types of other varieties. Such intervarietal conjugations as
are completed usually result in the death of the Fl or F2 generations.
However, enough intervarietal combinations give some degree of
reaction to provide evidence of mating type homologies for most
of the varieties. The two mating types within a single variety are
designated by Roman numerals, an odd and an even number, and it
is clear that the odd mating types in most varieties are similar though
not identical. The even mating types in most varieties are also similar
but not identical.

The eight varieties fall into two readily distinguished classes,
those \\ hich have been characterized as showing little cytoplasmic
inheritance: Group A, composed of varieties 1, 3, 5, and 7, and those

268 SEX IN MICROORGANISMS

which show considerable cytoplasmic inheritance; group B, com-
posed of varieties 2, 4, 6, and 8. Most of the study of mating types
has been concentrated on variety 1 in group A and variety 4 in
group B. The mating types in variety 1 are designated as I and II;
those in variety 4 as VII and VIII.

The chief reason for Sonneborn's ( 1 945 ) conclusion that group
A and group B varieties have fundamentally similar systems of deter-
mination and inheritance of traits, including mating types, is their
possession of identical major cytogenetic processes. These consist
chiefly of conjugation, cytogamy, autogamy, and macronuclear re-
generation (Sonneborn, 1947). The details of these processes are dis-
cussed elsewhere in this symposium (Wenrich, 1954) and will be pre-
sented here only briefly.

Conjugation occurs when cells of complementary mating types
are mixed under appropriate conditions. Initially clumps and even-
tually pairs are formed. Within each member of a conjugating pair
the micronuclei undergo meiosis, and all except one of the haploid
nuclei disintegrate. This remaining nucleus divides mitotically to pro-
duce the migratory and stationary nuclei of identical genetic consti-
tution. The migratory nuclei of the pair members are exchanged and
fuse with the stationary nuclei to form the syncarya. From the fact
that the nuclei fusing in the two pair-members are alike, it will be
clear that the syncarya must be of the same genotype.

Cytogamy is essentially similar to conjugation except that an
exchange of gamete nuclei fails to take place and mutual self-fertiliza-
tion results. In this case, therefore, the syncaryon is formed by the
fusion of two sister haploid nuclei of identical genie constitution. The
syncarya must, therefore, be homozygous for all their genes.

Autogamy occurs periodically in all the stocks studied. This
process is identical with cytogamy except that it occurs in single un-
paired cells.

In all three of these reorganization processes the syncaryon nor-
mally gives rise to four mitotic division products, two of which
become micronuclei and two of which become macronuclei. The
anlagen of the two macronuclei are regularly separated into different
cells at the first postzygotic cell division and develop into mature
macronuclei which divide, apparently amitotically, at subsequent cell
divisions. The animals whose macronuclei are derived from a single
macronuclear anlage are called a caryonide.

MATING TYPE DETERMINATION 269

The fate of the fragments of the old niacronucleus is the same
in all three fertilization processes. They are passively distributed to
tlie products of successive iissions. Since the old niacronucleus breaks
down ordinarily into about 30 to 40 fragments, the cells contain on
the average 15 to 20 after one fission, 7 to 10 after two fissions, and
so on. Meanwhile the fragments slowly distintegrate, but their disin-
tegration usually does not begin until after the first or second cell
division.

In 1942 Sonneborn reported a variation in the behavior of these
fragments and of the new macronuclear anlagen which occurs in both
groups of varieties and which provides a useful tool in genetic analy-
sis. The new macronuclear anlagen may either fail to arise from
products of the syncaryon or their development may be temporarily
inhibited so that, although segregated at the first cell division, they
may fail to divide at the second or third cell division. In either case,
cells arise that lack macronuclear anlagen but contain fragments of
the old macronucleus. In these the fragments fail to disintegrate.
Instead they grow and continue to be segregated until only one is
present in each cell. By this time each fragment has reached the
size of the normal macronucleus and thereafter divides at each cell
division. The whole process of development of a new macronucleus
from a single fragment of the old macronucleus is known as macro-
nuclear regeneration.

In so far as present information goes, regenerated fragments con-
trol the same hereditary traits as the macronucleus from which they
were derived. Presumably, therefore, each fragment contains at least
one complete set of nuclear genes. These results would be expected
on either of two very different hypotheses. Sonneborn (1942) origi-
nally interpreted macronuclear regeneration as indicating a compound
macronuclear structure, i.e., a macronucleus composed of discrete
genetically balanced subnuclei. Kimball (1943) proposed as an alter-
native explanation that the macronucleus contains a great many sets
of chromosomes randomly arranged and assorting more or less at
random when the macronucleus divides or breaks down into frag-
ments. It is apparent that either of the postulated nuclear structures
could result in macronuclear regeneration as found, but that the nec-
essary number of chromosome sets would be much greater in a mac-
ronucleus constructed according to Kimball's hypothesis.

Macronuclear regeneration occurs under ordinary conditions

270 SEX IN MICROORGANISMS

with a very low frequency. However, Sonneborn (1942) increased
its frequency greatly by exposing cells to high temperatures during
the period when the syncaryon undergoes its divisions and the new
macronuclear anlagen are developing. If the exposure is begun earlier
in the fertilization processes, the micronuclei may be lost. Macro-
nuclear regeneration may also be induced by exposure to very low
temperatures at the same period (Sonneborn, unpublished).

Group A Pattern

In some varieties of group A a few stocks have been found
which are pure for one mating type (Sonneborn, 1938; Sonneborn
and Dippell, 1946). These few stocks are of the odd mating type.
In all other stocks any isolated animal can, according to a definite
pattern, produce both mating types. To these "two-type" stocks we
will turn first. The main features of inheritance of mating types in
such stocks were first reported by Sonneborn (1937) and Kimball
(1937).

Animals of these two type stocks usually undergo no change of
mating type during vegetative reproduction. However, changes of
mating type may occur at nuclear reorganization. The kind of nu-
clear reorganization (conjugation or autogamy) has little or no influ-
ence on the frequencies of the mating types produced. A single
reorganized animal may give rise to either or both mating types. If
both are produced, segregation of the mating type determiner nearly
always takes place at the first postzygotic cell division, and the two
products of this division are, during subsequent vegetative reproduc-
tion, pure for mating type. It is at this same division that the inde-
pendently developing macronuclear anlagen separate. Hence, the
unit of mating type inheritance is the caryonide, and this sort of
inheritance is said to be caryonidal. This fact at once suggests that
the macronuclei may be differentiated in such a way as to determine
different mating types. In further support of this view, Sonneborn
(1937, 1938, 1939, etc.) marshalled a number of lines of evidence
which will be briefly reviewed.

When macronuclear regeneration occurs the mating type of the
parent cell is perpetuated and no change is observed (Sonneborn,
1942). This same result is obtained whether macronuclear res^enera-
tion occurs as the result of treatment with high temperatures or as

MAIlNCi lYPl 1)1 n RAIINAIION 271

rhc result of rcorgani/ation in aiiiicronuclcatc lines. These observa-
tions sliow tliat new mating tvpcs arise whenever new macronuclei
arc fornicd from micronuclei and that new mating types do not
arise at nuclear reorganization when new macronuclei fail to form.
.Moreover, it is seen that mating type is perpetuated through any
piece of the old macronucleus.

In some stocks mating types fail to segregate at the first post-
zvgotic division, but segregate at the second or third division.
In these stocks multiple macronuclear anlagen have been found
and mating type distribution in these, as well as in "normal" stocks,
is correlated with the distribution of independently developing mac-
ronuclear anlagen (Sonneborn, 1938, 1939). In rare cases in some
stocks mating type segregation continues and is clearly not due to
the segregation of diverse whole macronuclei (Kimball, 1939). Car-
yonides showing this kind of mating type segregation contain both
mating types and conjugation occurs regularly within the caryonides.
These are called, therefore, "selfing caryonides." A more detailed
consideration of these selfers will be presented later. Except for the
selfing caryonides, the facts of mating type inheritance in group A
clearly require the conclusion that macronuclear differences deter-
mine the two mating types.

The relative frequencies with which the different mating types,
and hence the diverse macronuclei, arise after fertilization varies in
dependence upon the temperature prevailing at this time (Sonneborn,
1939, 1942). The relative frequency of the even mating type in-
creases with the temperature over the range of 12° to 32° C. Al-
though this implies some effect of temperature on the developing
macronuclei, the effect could be either direct or indirect, through
the mediation of some cytoplasmic constituent.

Regardless of the temperature within this range, and conse-
quently regardless of the relative frequencies of the two mating
types, the two independently developing macronuclei in a single
reorganizing cell are independently determined as to which mating
type they \\'ill control. This is shown by the fact that the relative
frequencies of the various possible combinations for the tw^o macro-
nuclei agree with the calculations based on simple probability con-
siderations.

An anomalous result was obtained when animals of variety 3
underwent both conjugation and subsequent reorganization below

272 SEX IN MICROORGANISMS

10° C. (Sonneborn, 1939). Under these conditions very little change
of mating type occurred at conjugation, each mate producing a clone
predominantly of its own mating type. These observations were
made before the discovery of macronuclear regeneration, and since
mating types do not change at macronuclear regeneration, it seems
possible that some, perhaps all, of this failure to change mating types
at low temperature is due to the occurrence of macronuclear regen-
eration.

In sum, the evidence available on mating type determination in
group A agrees in indicating macronuclear control of the differences
in mating types. The problem is then raised as to how different
macronuclei arise from genetically identical micronuclei. Pair mem-
bers are identical in their nuclear genes after fertilization, as required
by the cytogenetic details and as shown by genetic analysis (Sonne-
born, 1939, 1947), and yet pair members may differ in their mating
types. Even more remarkable, sister macronuclei developed from
presumably identical mitotic products of a single fertilization nucleus
may determine different mating types. The fact that mating types
may change at any autogamy, despite the fact that complete homozy-
gosis is established by a single autogamy, argues that no micronuclear
differences or genie recombination can account for the differences in
mating types. This conclusion is reinforced by an additional fact set
forth in the next paragraph. Several tentative hypotheses have been
advanced to account for macronuclear differences, but none is satis-
factory. This then remains the chief unexplained phenomenon of
mating type determination in group A, the manner in which unlike
macronuclei arise from identical micronuclei.

The additional fact referred to in the previous paragraph is of
special importance, although the reason for its importance will not
become evident until later. The mating type of an animal before
reorganization is not correlated in any way with the mating type of
its progeny. An animal of mating type I gives rise to the same kinds
of progeny and in the same proportions as animals of mating type II
(Sonneborn, 1937; Kimball, 1937). In other words, animals in the
two-type stocks of group A which differ in mating type do not
differ from each other with regard to either the types of progeny
which they produce or the relative frequencies of these progeny.

The preceding account shows that the micronuclei in animals of
different mating types in group A are alike, while the macronuclei

iMATING TYPE DETERMINATION 273

nrc in sonic unspecified way different. No mention has yet been made
of particular gene differences affecting mating type inheritance. One
effective gene difference has, however, been reported (Sonneborn,
1939). The difference in the mating type phenomena in the two-type
stocks fullv discussed above and in the one-type stocks briefly men-
tioned at the be^innintr of this section is due to a difference in a
single pair of allelic genes. The recessive allele, for which the one-
type stocks are homozygous, restricts mating type to type I: the
dominant allele (homozygous in two-type stocks) permits develop-
ment of either mating type according to the pattern set forth above.
The mode of action of these alleles remains unknown.

The results of previous investigations on mating type deter-
mination in stocks of group A may be summarized as follows:

1 . The differences between the complementary mating types are
due to some as yet unknown differences in the macronuclei.

2. These macronuclear differences arise at the time the macro-
nuclei develop from products of the syncaryon.

3. Temperature increases at this time increase the probabiUty of
origin of a macronucleus that will control the even mating type.

4. The two macronuclei which develop synchronously in the
same fertilized cell are independently determined as to which mating
type they will control.

5. A single gene difference determines whether both mating
types can be produced or only one, the odd mating type.

6. In stocks in which both mating types can be produced, there
is no effect of the mating type of an individual on the mating type
of its sexually produced offspring.

Group B Pattern

Many of the features of mating type determination and inherit-
ance in group A are also found in group B (Sonneborn, unpub-
lished). As a rule mating types are strictly inherited during vegetative
reproduction. Stocks of varieties 4 and 8 show a small percentage of
selfing caryonides, just as some stocks of variety 1. Varieties 2 and
6, like variety 5 in group A, show a much higher frequency of these
selfers. With the exception of the selfers, mating types change only
at the time of nuclear reorganization. After nuclear reorganization
the sister caryonides from a single reorganized cell may show differ-

274 SEX IN MICROORGANISMS

ent mating types. Marked temperature effects on mating type deter-
mination have also been shown for the B varieties under certain con-
ditions and again, as in the A varieties, the higher temperatures favor
the even mating types (Nanney, unpublished).

In only one essential feature have differences between the A
and B groups been noted (Sonneborn, 1947). In group A there is no
correlation between the mating types of parents and their progeny,
nor between the mating types of sister caryonides. In group B,
changes occur rarely at reorganization, and when they do occur they
usually involve both sister caryonides. A strong correlation is thus
found between the mating type of the cytoplasmic parent and the
mating type of the progeny. Similarly, the mating types of sister
caryonides are strongly correlated. At conjugation this parent-prog-
eny correlation usually results in one of the cytoplasmic parents giv-
ing rise to two caryonides of one mating type and the other giving
rise to two caryonides of the other mating type. Since exconjugants
of a single pair are known to be alike in regard to the kinds of genes
which they possess, Sonneborn concluded that the differences which
characterize the mating types, though clearly due to differences in
the macronuclei, are inherited through some cytoplasmic mechanism.
This interpretation was supported by the observation that pair mem-
bers which are allowed to exchange massive amounts of cytoplasm
usually give rise to the same kinds of progeny.

The presence of a cytoplasmic component in the group B system
of mating type determination introduces special problems, and diver-
sities of interpretation have existed in regard to its significance. Be-
cause of the special features of the cytoplasmic involvment, we will
return later to a more extended consideration of nucleocytoplasmic
interactions in the B system.

The general features of the B system may be summarized as
follows:

1. As in group A, mating types are determined by alternative
macronuclear conditions.

2. Higher temperatures, as in group A, favor the formation of
macronuclei controlling the even mating types.

3. In contrast to the group A pattern, mating types in group B
show a strong parent-progeny correlation. Mating types are not
redetermined at random at nuclear reorganization, but tend strongly
to be maintained through reorganization.

MATING TYPE DKTFRAIINATION 275

4. This pnrcnr-progcny correlation is due to cytoplasmic con-
ditions ciiaractcristic of the two mating types which determine the
manner in which the new nuclei develop.

NATURE OF MACRONUCLEAR DIFFERENCES

1 he facts presented in the previous section show a consistent
pattern of mating type determination and inheritance in both groups
of varieties, hut leave unanswered a central question: In what ways
do the macronuclei in cells of different mating types differ? A pos-
sible clue to these differences comes from a study by Chao (1953) on
the killer character in variety 4. Sonneborn (1947) showed that the
killer character is determined by the presence of self-duplicating par-
ticles, called kappa, in the cytoplasm. He further showed that the
presence of kappa is conditioned by the presence of a gene, K, in
the macronucleus. Freer (1950) developed technics for visualizing the
particles cytologically, and these technics have been used by Chao to
study directly the factors influencing kappa concentration. Two of
his observations are pertinent here. ( 1 ) Cells of a given mating type
w^ill support twice as many kappa particles when they have the
genotype KK as when they have the genotype Kk. This observation
shows a direct numerical relationship between the number of K genes
in the nucleus and the number of kappa particles in the cytoplasm.
(2) Cells of a given genotype will support twice as much kappa
when they show the odd mating type as when they show the even
mating type. These observations suggested (Nanney, 1953) that
macronuclei in cells of the odd mating type contain twice as many
K genes as those of the even mating type. Alternatively, it is possible
that certain genes in one mating type are precisely twice as active
as those in the other mating type, but this explanation seems less
reasonable.

Chao's observations form the basis for a gene-dosage hypothesis
of mating type determination. This hypothesis can take a number of
different forms. Differences in the number of K genes could be
achieved though a doubling or halving of the chromosomes bearing
the K gene, thus yielding an aneuploid constitution for one of the
mating types. Similarly, differences in the number of K genes could
be brought about by a doubling or halving of the entire chromosome
complement. This interpretation would ascribe significance to dif-

276 SEX IN MICROORGANISMS

ferent ploidy levels, either in subnuclei or in the macronucleus as a
whole.

An evaluation of these alternative formulations is complicated
by the fact that the macronucleus undoubtedly contains many sets of
nuclear genes (Sonneborn, 1947), and the manner in which these are
organized in the macronucleus is not well understood. Nevertheless,
certain conclusions seem warranted.

1. The polyploid interpretation in any simple form is untenable.
This conclusion is derived from several different lines of evidence.
(a) The experimental production of cells with different ploidy levels
did not give the results expected on this interpretation (Sonneborn,
1953). (b) The mating type system in Tetrahyviena pyrijonms,
which shows a remarkable series of parallels to that in P. aiirelia,
involves at least seven different mating types (Nannev and Caughey,
1953). Although two ploidy levels, characteristic for different mat-
ing types, are conceivable, the necessity for postulating seven or more
different ploidy levels becomes too great a burden for the hypothesis
to sustain. Certainly ploidy alone is not sufficient to account for
caryonidal inheritance of mating types in Tetrahymena. {c) It has
not been possible to demonstrate any significant differences in the
deoxyribose nucleic acid content of cells of different mating types
in variety 4 of P. aiirelia (Guthe, Tefankjian, and Nanney, unpub-
lished).

2. An aneuploid interpretation that postulates the loss of a chro-
mosome while the macronuclear anlagen are still diploid cannot be
supported. This is shown by the fact that all heterozygotes studied,
including specifically the Kk heterozygotes (Sonneborn, 1947), have
shown the dominant phenotype. If the eliminated chromosome con-
tained the dominant allele (as it should at least in some cases), the
recessive allele would have been manifested.

3. An aneuploid interpretation that holds that chromosome dou-
bling occurs either in the diploid stage or later could account for the
observations. Similarly, the loss of certain chromosomes after the
macronucleus has developed could account for the observations. On
the other hand, neither of these interpretations can be tested with
present technics and the mechanism whereby some one type of
chromosome could be regularly, quantitatively and specifically in-
creased or eliminated is obscure. Hence, the gene-dosage hypothesis
must be considered to have lost its chief utility.

MATING TYPE DI TERAIINATION 277

4. Chiio's rcniiirknhlc ohscrvntions concerning mating types and
kapi");! concentration indicate some precise quantitative distinctions
betw een the mating types, hut the nature of the distinctions remains
a puzzle.

NUCLEOCYTOPLASMIC INTERACTIONS IN THE B SYSTEM

There can be no doubt that a cytoplasmic component occurs in
the group B system of mating type determination, nor that this cyto-
plasmic component acts on the developing macronuclei so as to deter-
mine their manner of development. Differences of interpretation are
possible only in regard to the manner in which the cytoplasmic spe-
cificity is maintained. Either the cytoplasmic conditions are self-
perpetuating, or they are to some extent under the control of the
nuclei.

Interpretations regarding the means of perpetuation of the cyto-
plasmic conditions center around the subsequent behavior of the
caryonides which have changed mating type at nuclear reorganiza-
tion. Sonneborn (unpublished) observed two kinds of changed cary-
onides in regard to their stability at subsequent reorganizations. The
first group were those that maintained the changed type with as
great stability as normal clones and included nearly all caryonides
that had changed from mating type VIII to mating type VII and the
majority of those which had changed from type VII to type VIII.
Many caryonides that had changed from type VII to type VIII
showed a considerable amount of reversion from VIII to VII at the
next autogamy. More rarely, a caryonide of type VII showed a con-
siderable change to type VIII at autogamy. Since in these clones the
cytoplasmic conditions usually associated with one mating type were
at least partially maintained in the presence of a macronucleus con-
trolling the other mating type, Sonneborn concluded that the cyto-
plasmic conditions necessary for the inheritance of mating type are
self-maintaining and at least to some extent independent of the nu-
clear condition. Although failing to manifest its activity at one nuclear
reorganization, the cytoplasmic determiner maintains itself and is
manifested at a subsequent reorganization.

This interpretation is called into question first of all by recent
investigations regarding the nature of the caryonides which show con-
siderable reversion at reorganization. In a study of the inheritance of

278 SEX IN MICROORGANISAIS

mating types at conjugation in selfing caryonides (Nanney, unpub-
lished), it was observed that the mating type VIII individuals in such
caryonides change to type VII with high frequency. All selfing
caryonides studied, regardless of the amount of selfing observed,
showed this instability. In some caryonides one of the mating types
was so infrequent that routine observation for selfing was not suffi-
cient to demonstrate that these were in fact selfers and observations
on several subcultures for a period of days were required to detect
selfing. Since most such selfing caryonides appear superficially to be
pure type VIII cultures, but revert largely to type VII at reorganiza-
tion, it appears possible that many of the unstable VIII's studied by
Sonneborn were of this kind. If so, the reversion is in these cases in-
timately connected with the question of the nature of selfing caryo-
nides and is not due simply to the fact that these clones have recently
changed mating type.

Sonneborn's original interpretation is also called into question
in so far as it focuses attention upon the occasional aberrant clones
and ignores the behavior of the majority of clones which have
changed mating types. Usually when changes occur from one pure
mating type to another pure mating type, not only are the mating
types changed but the changed types are perpetuated normally
through nuclear reorganization. Even when sister caryonides, sharing
a common source of cytoplasm, show different mating types, the
type VII caryonide produces almost exclusively type VII progeny
and the type VIII caryonide produces almost exclusively type VIII
progeny (Nanney, unpublished). One finds the usual high correlation
between the mating type of the parent and that of its sexual progeny,

Sonneborn (1953) has recently provided a similar but more
elegant test for the hypothesis of nuclear control, and the results
seem conclusive. This test involves inducing at conjugation (a) cyto-
plasmic exchange between the mates and (b) macronuclear regenera-
tion. The cytoplasmic exchange results in the change of mating type
in one member of the pair, so that new macronuclei controlling one
mating type share the same cytoplasm with fragments of a macronu-
cleus controlling a different mating type. The conditions under which
macronuclear regeneration was induced were such as to suppress
division of the new macronuclei, but not to abort them. This sup-
pression of division of the new macronuclei results in the formation
of diverse lines of descent from a single cell, some containing a new

AlATING lYPI. Hi URMINATION 279

iiiacronuclcus controlling one ninrint:;- type and sonic containing re-
generated macronuclei controlling the other mating type. Since it
seems reasonable to assume that the cytoplasmic elements, if these
arc indeed independent of nuclear control, will he equally assorted in
the various lines of descent, all lines of descent should at a subsequent
reorganization give rise to the same kinds of progeny. This result
was not obtained. Instead, each line at a subsequent reorganization
regularly produced progeny of the same mating type as the parental
line. Therefore, the cytoplasmic conditions must have been modified
to bring them into agreement with the nuclear constitution.

Since the mating type of the parent is clearly controlled by the
macronucleus, and since the cytoplasmic condition necessary for the
inheritance of mating type through nuclear reorganization is strongly
correlated with the parental mating type, it follows that the cyto-
plasmic condition is ultimately controlled by the old macronucleus.
Thus, though the cytoplasm determines the nature of the new macro-
nucleus, the cytoplasm is in turn controlled by the old macronucleus.
This results in a cyclical interdetermination of the nucleus by the
cytoplasm and of the cytoplasm by the nucleus. This system tran-
scends "maternal inheritance" in that the cytoplasmic influence alters
the new nucleus in such a way as to perpetuate a trait indefinitely
through both vegetative and sexual reproduction. Regardless of the
explanation eventually given for those rare clones which revert to an
original mating type, the situation observed in the majority of
changed clones argues strongly for the macronuclear control of the
cytoplasm.

The precise nature of this macronuclear control is problematical.
The nucleus could produce directly some substance or substances nec-
essary for mating type inheritance, but it is also possible that the
macronucleus merely controls the rate of production of a self-repro-
ducing factor such as has been demonstrated for the killer character
in this same organism (Sonneborn, 1947; Chao, 1953). In the absence
of evidence for a self-reproducing factor, it appears advisable to adopt
the simpler interpretation.

NATURE OF SELFING CARYONIDES

One of the perplexing problems concerning mating types, both
in P. aurelia and in other ciliates, involves the nature of the selfing

280 SEX IN MICROORGANISMS

caryonides. These caryonides are clones deriving their macronuclei
from single macaronuclear anlage; yet within such clones are found
cells of diverse mating types (Kimball, 1939; Jennings, 1941; Nanney
and Caughey, 1953), Since the macronuclei control the mating types,
it is presumably some macronuclear instability that permits cells
within a single clone to manifest different mating types. The fact that
diverse pure types may be derived from the selfers indicates that the
unstable macronuclei may "stabilize," though it is not clear what the
stabilization involves.

A recent formal explanation for the selfers in P. aurelia (Nanney,
1953) was based on the idea of structural inhomogeneity. Since the
most widely held interpretation of macronuclear structure considers
that the macromicleus is compound, i.e., contains many sets of nuclear
genes associated into subnuclei with a certain degree of integrity, it
appears plausible to postulate differences between mating types as
due to some characteristic of the subnuclei. If this were correct, some
macronuclei might contain diverse types of subnuclei, and such
"mixed" macronuclei could provide a basis for the observed vegetative
segregation of mating types.

In an attempt to test this interpretation (Nanney, unpublished),
macronuclei which would ordinarily control different mating types
were allowed to fuse. This fusion was induced through starving
exconjugants (Sonneborn, 1947). Stock 90 in variety 1 was used in
this study because in this variety mating types are determined at ran-
dom at reorganization, giving many cells with diverse sister nuclei,
and because selfing is rare in this stock. After macronuclear fusion
many clones were found which behaved like the spontaneous selfers
in other stocks of the same variety. This production of selfers by the
fusion of diverse macronuclei seemed strongly to support the hypoth-
esis of structural inhomogeneity for the spontaneous selfers, but did
not exclude other interpretations.

Further information from another source has cast considerable
doubt upon this interpretation. The hypothesis offered hope that an
analysis of the pattern of segregation of mating types in an unstable
clone would be profitable. It seemed reasonable to assume that the
probability of origin of a "stable" macronucleus would depend upon
the original number of subnuclei, the relative numbers of the differ-
ent kinds and the number of divisions the macronuclei had undergone.
The analysis in P. aurelia was, however, complicated by a relatively

MATING TYPE DIT ERiMINATION 281

short period of vegetative growth before autogamy intervened and
replaced the macronuclei being studied, \\1ien it was discovered
that selfing caryonides occurred in Tetrahymefia pyriformis and that
autojTaniy did not (Nanney and Caughey, 1953), it seemed that
an analvsis would be profitable here. Hence, studies were undertaken
w'nh this end in view. In the course of these studies (Nanney and
Caughey, unpublished) it was found that the mating types in selfing
clones can be stabilized readily simply by starving the cells. This was
not anticipated on the structural hypothesis, since starvation of mac-
ronuclei with stable but diverse subnuclei could not result in the
establishment of macronuclei with only one type of subnucleus.
Hence, the hypothesis is of no value for the selfers in T. pyriforinis,
and since the systems of determination in P. aiirelia and T. pyriformis
are so much alike, the hypothesis for P. mtrelia is severely discredited.
These observations indicate that, while the systems leading to the
manifestation of the different mating types are mutually antagonistic,
the\' may under certain circumstances be maintained simultaneously
for long periods of growth. The failure of the structural hypothesis
suggests that the differences characterizing macronuclei controlling
different mating types are to be sought in physiological rather than
structural features.

SUMMARY AND CONCLUSIONS

A survey of the patterns of mating type determination and in-
heritance in Paramecium aiirelia is presented with an examination
of the significance of these patterns in an understanding of cellular
heredity. The major conclusions may be summarized as follows:

1. Nuclei containing the same genetic materials may be differ-
entiated in regard to the phenotypes (mating types) that they control.

2. The differentiated nuclei normally breed true in vegetative
growth, i.e., the nuclear characteristics are hereditary.

3. An important factor in determining the manner in which the
nuclei develop (in the group B varieties) is the kind of cytoplasm
in w hich they develop.

4. The significant cytoplasmic conditions are in their turn
determined by the kind of nucleus that previously occupied the cell.

Thus, mating type perpetuation through both vegetative and
sexual reproduction is shown to be due to a series of nucleocytoplas-

282 SEX IN iMICROORGz\NISAIS

mic interactions, in which the cytoplasm determines the nature of
the macronucleus and is in turn redetermined by that macronucleus.
This pattern of nucleocytoplasmic cooperation emphasizes the con-
clusion that the cytoplasm is not a passive, but an active partner in
cellular heredity.

REFERENCES

Chao, P. K. 1953. Kappa concentration per cell in relation to the life cycle,

genotype and mating type in Farameciu7n aurelia. Proc. Natl. Acad. Sci.,

U. S. 39, 103-113.
Jennings, H. S. 1941. Genetics of Faravieciinii bursaria. II. Self-differentiation

and self-fertilization of clones. Proc. Avi. Phil. Soc, 85, 25-48.
Kimball, R. F. 1937. The inheritance of sex at endcmixis in Paraviechim

aurelia. Proc. Natl. Acad. Sci. U. S., 24, 112-120.
. 1939. Change of mating type during vegetative reproduction in

Paramecijivi aurelia. J. Exptl. Zoo/., 81, 165-179.

1943. Mating types in ciliate Protozoa. Quart. Rev. Biol., 18, 30-45.

Metz, C. B. 1954. Mating substances and the physiology of fertilization in

ciliates, in Sex in Microorganis7//s, pp. 284-334. A A AS, Washington, D. C.
Nanney, D. L. 1953. Mating type determination in Parameciuvi aurelia, a

model of nucleo-cytoplasmic interaction. Proc. Natl. Acad. Sci. U. S., 39,

113-119.
Nanney, D. L., and P. A. Caughey. 1953. Mating type determination in Tetra-

bytfiena pyrijorDiis. Proc. Natl. Acad. Sci. U. S., 39, 1057-1063.
Preer, J. R. 1950. Microscopically visible bodies in the cytoplasm of the

"killer" strains of Parameciinn aurelia. Genetics, 35, 344-362.
Sonneborn, T. M. 1937. Sex, sex inheritance and sex determination in Para-

7/ieciu7/i aurelia. Proc. Natl. Acad. Sci. U. S., 23, 378-385.
. 1938. Mating types in Paraiueciuvi aurelia: diverse conditions for

mating in different stocks; occurrence, number end interrelations of

types. Proc. Ain. Phil. Soc, 79, 411-434.
. 1939. Paranieciinn aurelia: mating types and groups; lethal interactions:

determination and inheritance. Am. Naturalist, 73, 390-413.

. 1942. Inheritance in ciliate Protozoa. Aui. Naturalist, 76, 46-62.

. 1945. The dependence of the physiological action of a gene on a

primer and relation of the primer to the gene. Am. Naturalist, 79, 318-339.
. 1947. Recent advances in the genetics of Paramecium and Eu plates.

Advances in Genetics, 1, 263-358.
. 1948. Symposium on plasmagenes, genes and characters in Para?;?eciu:/i

aurelia. Introduction. Am. Naturalist, 82, 26-34.
. 1950a. Methods in the general biology and genetics of Paramecium

aurelia. J. Exptl. Zool., 113, 87-148.
. 1950b. The cytoplasm in heredity. Heredity, 4, 11-36.

MATING TYPF DF.TF.RMINATION 283

— . 1951. Some current problems of genetics in the light of investigations
on Chlainydonionas and Farameciuvi. Cold Spring Harbor Symposia
Quant. Bio]., 16, 483-503.
-. 1953. Patterns of nucleo-cytoplasmic integration in Parainecitim. Proc.

IX hnernatioiial Congress of Getietics. (In press.)
Sonneborn, T. AI., and R. V. Dippell. 1946. Mating reactions and conjugation

between varieties of Paravieciian aiirelia in relation to conceptions of

mating type and variety. Phys. ZooL, 19, 1-18.
W'enrich, D. H. 1954. Sex in Protozoa in Sex in Microorganisms, pp. 134-265.

AAAS, Washington, D. C.

Mating Substances

and the Physiology of Fertilization
in Ciliates*

CHARLES B. METZ,t Department of Zoology,
University of North Carolina, Chapel Hill

In metazoan fertilization the egg and sperm meet, these two cells
adhere, the egg is activated, and the gametes fuse. Similarly, protozoan
fertilization involves an initial adhesion, a series of physiological and
morphological changes, and finally partial or complete cellular fusion.
It appears then that no profound physiological difference should
exist between the process of fertilization in metazoa and ciliate pro-
tozoa and that information regarding fertilization in these two groups
of organisms might profitably be considered together. Such an ap-
proach seems timely since much of the available information on ferti-
lization in ciliates (Kimball, 1943; Sonneborn, 1947, 1949; A4etz,
1948) and in metazoa (Tyler, 1941, 1948, 1949; Runnstrom, 1949;
Rostand, 1950; Rothschild, 1951a,b; Chang and Pincus, 1951; Bielig
and von Medem, 1949) has been reviewed extensively, but independ-
ently, in recent years. In keeping with this view certain problems in
the physiology of fertilization are outlined here, the methods em-
ployed to solve these are presented, and the degree to which these
efforts have succeeded is considered and compared. Since the review-
er's main thesis has developed from studies on Farimieciuni, the
physiology of fertilization in this form is discussed in some detail,
pertinent studies on other ciliates are then considered, and finally an
attempt is made to relate these investigations to metazoan fertilization.

* The writer's studies were aided in part by grants from the National
Institutes of Health, U. S. PubHc Health Service, and from the American
Cancer Society.

t Present address. Department of Zoology, Florida State University, Tal-
lahassee.

284

1 HE PHYSIOLOGY OF FI.R TILIZA TION IN CILIATES 285

PROBLEMS AND METHODS

The prinv.irv problem in the physiology of fertilization is the
problem of the activation initiating meclianism. Associated with this
primary event are certain other phenomena which constitute special
problems in themselves. Thus: (1 ) the metazoan gamete or the ciliate
must be sexually ripe or reactive before fertilization can occur; (2)
some degree of attachment or union of the reacting cells would ap-
pear to be essential for their interaction; (3) fertilization is univer-
sally characterized by a high order of specificity; (4) an astonishing
series of biochemical and morphological changes follows immediately
upon activation in most forms; and finally (5) passage of nuclear
material from one cell to another presupposes some degree of func-
tional fusion of cells. In passing it may be noted that some of these
problems also apply to several other major biological phenomena,
notably embryonic induction, host-virus relationships, and specific
tissue affinities. Although major interest has centered about the prob-
lem of the activation-initiating mechanism, a comprehensive account
of fertilization must provide not only for the initiating reaction itself
but for these associated phenomena as well. The second, third, and
fourth of these phenomena, as well as the activation initiating mecha-
nism, \v411 be considered in this review.

The various morphological, physiological, and biochemical
changes that occur at fertilization all follow in a very precise and
orderly sequence under favorable circumstances. This has led to the
view that most, if not all, of these many changes are interrelated, that
they all proceed from a few or even a single event — a chemical reac-
tion between the interacting cells. Considered in this fashion, it would
not seem excessively difficult to identify the activation-initiating
reaction or reactions. However, this and the associated problems have
been studied for over fifty years and many competent investigators
have directed their attention toward them, but no comprehensive
theory based on substantial experimental data is yet available to de-
scribe them. Four general methods have been employed by various
investigators in their attempts to reveal the mechanism of fertilization.
These are outlined below.

The first of these may be called the biochemical approach. By
studying the end effects of activation, particularly the physical and
metabolic effects, and by tracing these back to their initial causes.

286 SEX IN MICROORGANISMS

one might eventually achieve a biochemical description of the activa-
tion-initiating mechanism. This biochemical approach began with
Warburg's (1908) demonstration of the increase in respiration that
follows activation of the sea urchin egg. Although a great deal has
been learned about the biochemistry of the unfertilized as compared
with the activated egg, and although the approach has even been ex-
tended to fertilization in ciHates (Boell and Woodruff, 1941), it is
still largely in the descriptive stage. Even in the most thoroughly
studied form, the sea urchin, general agreement on metabolic path-
ways has yet to be reached (Rothschild, 1951a; Cleland and Roths-
child, 1952). Evidently, then, it will be some time before this ap-
proach to the problem may be expected to yield the desired answers
to the fertilization problem.

The second approach to the problem of the activation-initiating
mechanism is through the medium of artificial parthenogenesis. Arti-
ficial parthenogenesis was first achieved with the silkworm egg
(Tichomiroff) in 1886. However, it was not until the turn of the
century, with Loeb's (1899) experiments on the sea urchin tgg, that
this field of investigation showed real promise. During the early part
of the century it was anticipated that parthenogenesis would provide
the key to the mechanism of action of the sperm in fertilization. How-
ever, a very large number and variety of effective physical and chem-
ical agents were soon discovered, and as yet no common factor has
been found among them. Parthenogenesis has made the important
contribution of demonstrating that the egg contains within itself all
essentials for development and has given substance to the stimulus or
trigger concept of sperm action, but as yet no comprehensive scheme
to explain parthenogenesis or to relate it to sperm activation has been
forthcoming.

A third approach to the activation problem involves the investi-
gation of cell extractives and substances liberated spontaneously by
gametes or protozoa. This field is an attractive one because the effects
obtained are frequently striking and, like the biochemical studies, the
results are interesting in their own right whether or not they contri-
bute to an understanding of fertilization. The usual rationale here is
to attempt to ascribe a role in fertilization to such substances once
they arc obtained. The most thoroughgoing application of this
approach is to be found in the studies on the sperm and egg isoagglu-
tinins, fertilizin and antifertilizin. But, even after forty years of pain-

THE PHYSIOLOGY OF FFRTIMZATION IN CILIATES 287

stakiniT research, no invcsrigaror has presented convincing evidence
that these substances are essential for fertih/ation.

The fourth avenue of attack follows logically from the preceding
one and, in fact, represents a more systematic development of it. This
approach involves the partial or preferably complete isolation of the
activation-initiating mechanism and a study and characterization of
its parts. So far this method has not been employed with success on
metazoa, in spite of many attempts to activate eggs with sperm
extracts or dead sperm. However, this method has met with rather
striking success in the ciliate ParaiJiechnu. Since this approach and
the associated preceding one are concerned with various sex sub-
stances, a further discussion of these agents will be presented before a
comprehensive account of fertilization in Paraniec'mm and other
ciliates is given.

SEX SUBSTANCES AND MATING SUBSTANCES

Beginning with Frank Lillie's (1913, 1914, 1919) now classic
studies on fertilizin in the sea urchin and the annelid worm, Nereis,
the most fruitful approach to many of the problems of fertilization
has been through an analysis of specific interacting substances of sex
cells. Indeed, these have produced the only comprehensive, though
now outmoded, theory of fertilization, namely, Lillie's fertilizin
theory. Such sex substances have been demonstrated or their presence
inferred in many organisms, both plant and animal, unicellular and
multicellular. In ciliates these substances are produced only when the
organisms are in the sexual or mating condition and are capable of
fertilization. These agents, the mating substances, have specific action
upon ciliates of complementary sex or mating type and presumably
perform some function in fertilization. Correspondingly in metazoa,
agents may be obtained from eggs and sperm which specifically affect
gametes of the species. Again these agents are produced (presumably)
by the gametes themselves and are believed to function in fertilization.

In both groups of organisms the more spectacular and more
readily studied sex agents are those which appear in the fluid contain-
ing the gametes or sexually reactive protozoa. Such water-soluble
substances are actively produced and secreted or dissolve passively
from the cell or its surface into the fluid. Among ciliates the only
well-established example of such diffusible mating substances is in

288 SEX IN MICROORGANISMS

the hypotrich Eiiplotes patella. In this form Kimball (1942) has
shown that the fluid from certain clones will induce animals of certain
other clones to conjugate. The most intensively studied agent of this
sort from metazoan gametes is the fertilizin obtained from eggs. In
its most spectacular form such fertilizin specifically agglutinates the
sperm of the species.

Although freely diffusible, water-soluble sex substances of this
sort occur in the most diverse animal groups (Protozoa and Verta-
brata), they have not been demonstrated universally in either pro-
tozoa or metazoa (see Tyler, 1948, concerning metazoa). In the
demonstrable absence of such agents, specific sexual reactions, either
between protozoa or metazoan gametes, may be attributed again to
specific sex substances. However, in such cases the agents must be in-
soluble in the fluid medium, firmly bound to the cell, or both. It will
be seen in the account to follow that sex substances in Faramecium
are exclusively of this type, and on proper analysis this condition may
be expected to be found widely among metazoa and protozoa alike.

FERTILIZATION IN PARAMECIUM

Normal Sexual Phenomena

Two sexual processes occur normally in Farameciwn: conjuga-
tion and autogamy. The essentials of these are presented here as back-
ground for the analysis to follow. For a more detailed account of
normal conjugation and autogamy and a thorough treatment of the
literature, the reader should consult Sonneborn's (1947) excellent
review.

Conjugation. Conjugation in Paramecium involves several types
of union between mates and a variety of internal changes in these
animals. The first step in conjugation is the initial adhesion. Under
suitable conditions this takes the form of the striking mating reaction
described first by Sonneborn (1937). Dozens, even hundreds, of
animals stick together to form large masses or agglutinates. After
some time these mating reaction clumps break down, releasing mating
pairs and single animals. Animals in mating pairs are united at first
only at the holdfast region (Jennings, 1911; Wichterman, 1940; Hert-
wig, 1889; Metz, 1947) near their anterior ends (Fig. la). This hold-
fast union is firmer and more intimate than the mating reaction union.

THE PHYSIOLOGY OF FKR IIUZATION IN CILIATKS 289

bur the marcs can srill be scpararcd wirhour injury. Subscqucnrly
each mare produces a paroral cone (Diller, 1936) more posreriorly
(Fig. lb). The cones of rhe mares overlap and finally fuse. Afrer
paroral cone fusion rhe mares cannor be separared.

Afrer breakdown of rhe agglurinarcs rhe released animals will
nor give maring reacrions. Clearly rhey have undergone a physiologi-
cal change, a loss of maring reactiviry, Ar rhis rime (P. anrelia) rhe
firsr signs of nuclear acriviry appear. These involve a migrarion of
rhe micronuclei from rheir usual posirion in rhe viciniry of rhe macro-
nucleus and an enlargemenr of rhe micronuclei in prepararion for rhe
rirsr meioric prophase. The meioric divisions, rhe posr-meioric division

Fig. 1. Types of union in conjugating and pseudo-selfing paramecia.

(a) Holdfast union in early conjugants and pseudo-selfing animals, (b) Holdfast
and paroral cone union in more advanced conjugants.

forming rhe rwo generically idenrical (Sonneborn, 1939a) pronu-
clei, macronuclear breakdo\\'n, and rhe passage of rhe migrarory pro-
nucleus ro rhe paroral region rhen follow.

The holdfasr and paroral unions, loss of maring reacriviry, and
meioric and macronuclear evenrs all follow in a very precise and
orderly sequence in conjugarion. The obvious facrs rhar rhese changes
are induced in rhe conjugarion process and rhar rhey occur in a pre-
dicrable sequence suggesrs rhar rhese various changes are inrerrelared
and rhar rhey may be iniriared by a few or even a single reacrion in
conjugarion. In orher words, conjugaring paramecia are acrivared in
rhe same sense rhar rhe merazoan egg is acrivared by rhe ferrilizing
sperm. Indeed, some manifesrarions of acrivarion in rhe merazoan tgg
and in ParLWJeciinn are srrikingly analogous.

Autogamy . Essenrially rhe same series of physiological and

290 SEX IN MICROORGANISMS

morphological changes occurs in conjugation and autogamy. The
autogamous animal undergoes loss of mating activity, meiosis, macro-
nuclear breakdown, pronuclei formation, and paroral cone formation
(Diller, 1936). It appears, then, that the autogamous animal, like the
conjugant, is activated and that both undergo the same physiological
cycle. However, autogamy differs from conjugation in three impor-
tant essentials. (1) Autogamy occurs spontaneously, but only under
certain physiological conditions, in single isolated animals. No contact
or union with a mate is involved. (2) Autogamy results in self-ferti-
lization. The two pronuclei fuse to produce an homozygous syn-
caryon. (3) Autogamy and conjugation are initiated through different
mechanisms (Metz, 1948; Metz and Foley, 1949).

It should be noted that exchange of pronuclei fails to occur in
a number of abnormal conjugation processes. The usual genetic result
in such instances is identical with natural autogamy, that is, self-ferti-
lization. All such phenomena are sexually induced and are considered
to be natural or experimental variants of normal conjugation. They
are conveniently termed cytogamy (Wichterman, 1940). They are
to be distinguished from natural autogamy, which occurs spontane-
ously and without sexual contact between animals of opposite mating
type.

Cellular Adhesion and Mating-Type Substances in
Par ante ciu7n

The initial phase of the sexual reaction in Farainecmm is adhe-
sion of potential conjugants through the mating reaction union. There
is no evidence for action at a distance, no chemotaxis leading to this
reaction. Adhesion occurs only upon random contact. Furthermore,
it is not mediated by any agent from the medium. Thus culture fil-
trates or supernates have no effect upon the mating behavior of Para-
meciimi (Sonneborn, 1937, 1939b; Metz, 1947; Kimball, 1943), and
repeated washing does not alter their mating reactivity (Metz, 1947
and unpublished). Therefore the initial adhesion, the mating reaction,
must result from direct interaction of surfaces. Indeed it may be
attributed to the interaction of surface substances. Evidence for such
surface substances is derived from several sources, Sonneborn (1937,
1942b, c) has observed that an animal of one mating type can adhere
briefly to an animal of the same mating type if it has first clumped

THE PnVSK)1.0(;V OF 1 1:111 ILIZAIION IN CILIATF.S 291

witli an animal of opposite type. This suggests the transfer of surface
substances from one animal to another at the contact. Studies of
the effect of various agents on tlie mating behavior of Varainechnn
lend further support to this view . Paramecia that have been killed
by appropriate treatment with a wide variety of agents will give
specific mating reactions with living animals of opposite mating type.
Furthermore this reactivity can be blocked by treatment with certain
mild agents such as antiserum and protein group reagents. Although
no extract with mating substance activity (inhibition of the mating
reaction, action on animals of opposite type) has yet been prepared,
the evidence presented above and other characteristics of the mating
reaction to be outlined below leave no reason to question the view
that the initial adhesion depends upon reactions at the molecular level.
Therefore the initial adhesion, the mating reaction, can be attributed
to an interaction of substances or at least molecular configurations
attached to or built into the surface structure of the paramecium. For
convenience these are called the viating-type substances. The follow-
ing sections will deal mainly with the role of these mating-type sub-
stances in other aspects of fertilization.

Specificity of the Mating-Type Substances
Breeding Systems in Paramecium

The initial adhesion, the mating reaction, is the first step in con-
jugation. If the mating reaction does not occur, the succeeding events
are not observed. Conversely, completion of conjugation rarely fails
to follow the mating reaction under optimal conditions. In view
of this relationship the specificity of the mating reaction is a limiting
factor in conjugantion, and, since the mating reaction results from
interaction of the mating-type substances, the primary specificity in
conjugation will be determined by these mating-type substances. It
will be evident from the account to follow that this is a very high
order of specificity.

Most species of Faraiiieciwn have been examined for mating
reactions, and breeding systems have been worked out in nearly all
of these. Almost from the time of the discovery of mating types in
Faraineciumy two distinct breeding systems were recognized. In one of
these the morphological species consists of a number of sexually iso-
lated varieties each of which contains two interbreeding mating types.

292 SEX IN MICROORGANISMS

Ordinarily the mating reaction and conjugation occur only between
the two mating types in a variety. P. aiirelia (Sonneborn, 1937, 1938a,
1939a, 1942a, 1947), P. caudatimj (Gilman, 1939, 1941, 1950; Y. T.
Chen, 1944; Hiwatashi, 1949a), P. ivoodmffl (Woodruff, cited by
Sonneborn, 1939b), and P. calkijisi (Sonneborn, 1939b; Wichterman,
1951) follow this type of behavior.

In the second group of animals, sexually isolated varieties are
again found, but in this group several mating types are present in
certain varieties and any one of these will mate with all others in the
variety. P. bursaria (Jennings, 1938, 1939a; Jennings and Opitz, 1944;
Chen, 1946a), P. trichhmi (Sonneborn, 1938a, 1939b), and P. imil-
timicromicleattim (Giese, 1941) show this type of behavior.

Ajirelia-Type Systems. Among the aurelia-type species (two
mating types per variety) Paramecium aurelia has been examined
most thoroughly. The fifteen known mating types of this species fall
into eight sexually isolated varieties. With the exception of variety
7, which contains one (type XIII), all varieties contain two mating
types (Table I). On the basis of experiments on the breeding systems
and certain genetic studies, Sonneborn and Dippell (1946a) have
separated the several varieties into two distinct groups, the group A
and the group B varieties (Table I). In some respects the group A
varieties (Sonneborn's odd-numbered varieties 1, 3, 5, 7) are the more
interesting, for in this group intervarietal mating can occur in certain
combinations of types. Indeed every one of the group A mating types
gives at least one such reaction. These intervarietal mating reactions
are exceptional. They are never as intense as their intravarietal count-
erparts, they occur only under unusually favorable conditions, and
only four of them (types I x X; II x V; II x IX; II x XIII) ever lead
to complete conjugation (Sonneborn and Dippell, 1946a). Sonneborn
and Dippell have noted the following interesting relationships in these
group A reactions: (1) all seven group A mating types give a unique
set of reactions and are therefore different (types I and IX react with
different intensities with types II and X); (2) types I, V, IX, and
XIII are similar since they all react with type II; likewise types II, VI,
and X all react with type XIII; (3) mating reactions and conjugation
only occur between odd- and even-numbered types, never between
two odd- or two even-numbered types. From these observations
Sonneborn and Dippell (1946a) conclude that the several group A
mating types consist of two general types, or two series of homologous

C/0

^ c V

6S

M

<u ■ —

>

>

I— I
J— I

>

y.

I— I

>
>

4- I + I +

+

a

3

o

O

>

+ I +

+ I I I +

I + 1 I I

I 1 + I

1 + I
+ I

+

I + I +

I I I I I 11 +

+ I
I +

+ I

+

+

T^ rO lO

00 CN -rfi O

eg
(U

s

l-l
(U

293

294 SEX IN MICROORGANISMS

types, which they designate pins (even-numbered) and minus (odd-
numbered).

In contrast to the group A varieties the three group B varieties
(Sonneborn's even-numbered varieties 2, 4, and 6) show no cross
reactions among themselves. However, Sonneborn and Dippell
(1946b) have described one race (race 31), which conjugates readily
with variety 4 animals of group B and surprisingly also gives a weak
mating reaction, but never complete conjugation, with variety 3 of
group A. On the basis of this cross reaction and certain genetic
properties, Sonneborn and Dippell (1946b) regard race 31 as con-
stituting a distinct variety, variety 8, which links the group A and B
varieties. Finally, this series of cross reactions admits the mating types
of the intermediate variety 8 and the group B variety 4 to positions
in the two series of general mating types of group A (Sonneborn,
1950). Therefore, eleven of the fifteen mating types in P. aurelia may
be assigned general mating-type designations. Types II, VI, X, XVI,
and VIII are plus types; types I, V, IX, XIII, XV, and VII are minus
types (Table I).

From the foregoing account it is evident that a very high order
of specificity obtains in the mating reaction in P. aurelia; an order
of specificity that compares very favorably with that of antigen-anti-
body reactions, fertilizin-antifertilizin reactions, and fertilization in
other organisms. Since this specificity is determined by surface sub-
stances, the mating-type substances, these must be highly complex
substances or molecular configurations.

According to the simplest scheme (Metz, 1948) the mating reac-
tion within a variety in P. aurelia should result from interaction of a
pair of such substances, A and a, which are complementary in struc-
ture and combine in antigen-antibody-like fashion (Fig. 3a). To in-
clude the two general mating-type systems (plus and minus) in this
concept it is only necessary to assume that the mating substances in a
homologous series are structurally similar but not identical. Thus the
plus series should consist of five structurally similar but not identical
substances A\ A\ A'', A\ A^ (superscripts designate the variety).
Correspondingly the minus series of general types should contain six
similar but distinct substances ^\ a\ a'', a\ a'', a^, which are all comple-
mentary in structure to the members of the plus series. In the "lock
and key" terminology of immunology, near perfect complementari-
ness obtains only between the two mating types (A^ and a^) in a

THE PHYSIOLOGY OF FERTILIZATION IN CILIATES 295

variety. Less perfect "fits" resulting in weaker linkages and less strong
mating reactions should, and in most cases do, occur in intervarietal
combinations of plus and minus mating types.

In P. cctudatinn an even larger number of sexually isolated varie-
ties has been reported. Gilman (1950) has described eleven varieties
and has tested these against the four Japanese varieties found by
Hiwatashi (1949a). Two of the latter failed to conjugate with any
of Gilman's cultures (Gilman, 1950) and are therefore distinct varie-
ties. This gives a total of thirteen varieties with twenty-five mating
types (only one mating type has so far been found in variety 10).
Aside from these Y. T. Chen (1944) has described four Chinese varie-
ties. These have not been tested against the American or Japanese
varieties and may include additional varieties.

Although mating generally occurs only between the two types in
a variety, Gilman (1949) does report seven intervarietal reactions.
One of these links varieties 3 and 6; the remaining six reactions link
varieties 2, 8, 9, and 10 in such fashion that two general types appear
(the reaction between types XVIII and XX is so irregular that the
writer suspects an error). Gilman's (1949) data are summarized in
Table II, and the similarity to the situation in P. aiirelia is apparent.
As in the latter species the mating reactions between types within a

TABLE II

Intervarietal Mating Reactions in Paramecium caudatum

Varieties which give no intervarietal reactions are not included in this table.
Type XIX, variety 10, has not yet been discovered.

Variety

3

6

2

8

9

10

Mating
Type

\'

VI

XI

XII

III IV

XV XVI

XVII XVIII

XIX

XX

General
Mating
Types

3

V
VI

-

+

+

-

- -

- -

- -

—

6

XI
XII

-

+

- -

-

-

-

2

UI
IV

+

+

+

+

-

+

8

XV
XVI

+

+

+

+

9

XVII
XVIII

+

+
+

+

10

XIX
XX

-

+

296 SEX IN MICROORGANISMS

variety are most readily explained by assuming interaction of pairs of
complementary surface substances and at least in the varieties that are
linked by intervarietal reactions, two homologous series of substances
may be assumed. Finally, as Sonneborn (1947) points out, De Garis'
(1935) report that P. aiirelia will mate with F. caudatum suggests the
interesting possibility that the two homologous series of mating types
may extend across species lines.

The breeding pattern in F. calk'msi does not differ from the
systems in F. aiirelia and P. caudatwu. Wichterman (1951) finds four
isolated varieties in this species. Mating reactions and conjugation
occur only between the two complementary types within the varie-
ties. Breeding systems in F. ivoodniffl have not been described in
detail.

Bursaria-Type Systems. Immediately after Sonneborn's (1937)
discovery of mating types in F. aurelia, Jennings (1938) examined
P. bursaria for similar breeding behavior. This study and those that
extended it (Jennings and Opitz, 1944; Chen, 1946a) revealed a very
interesting and different system of mating specificity in this form.
As mentioned above, sexually isolated varieties occur in P. bursaria
as they do in other species. However, the number of mating types
within a variety is not limited to two. When more than two mating
types occur in a variety they interbreed freely. Thus four mating
types occur in varieties I, III, and VI. Any one of the mating types
within one of these varieties will mate with the other three members
of the variety. Similarly, each of the eight mating types in variety
II will conjugate with the remaining seven types.

The specificity relations in P. bursaria are therefore of an unusual
sort. There is a high order of specificity between the varieties, but an
apparent lower order of specificity within four of the six varieties.
However, if the multiple types are examined in terms of specific inter-
acting substances, a formal explanation for this apparent low-order
specificity is evident. The key to the problem lies in the observed
number of interacting mating types in a variety. This number con-
forms to the geometric progression 2", which in turn suggests that n
pairs of substances are involved in mating-type specificity. According
to the proposed scheme two independent pairs of specific, comple-
mentary, interacting substances would be required for a four-type
variety. Each mating type would possess two mating-type substances,
one substance of each independent pair. Assuming a random relation

THF FHYSIOI.OGV Ol' 1 TR Ill.l/A HON IN CII.IAII S 297

between the two independent systems, four interacting mating types
may be obtained. This h\pothesis is presented for variety 1, P. bur-
Siiriij (a four-type variety) in Tal)le III. ^ designates tbc reaction be-
tween the two complementary substances, A and J, of one pair (the a
pair); simihirly i^ represents reaction between the members, B and b,
of the second pair C^ pair) of substances.

TABLE III

Proposed Explanation of Multiple Mating Types in
P. bursaria, Var. I

Two independent pairs of specifically interacting substances, a and j8, are so
distributed that each mating type possesses one a substance (either A or a) and one
jS substance (either B or b). Mating occurs when A reacts with a (a reaction) and/or
B reacts with b {(3 reaction), a substances (.4, a) cannot react with jS substances {B, b).

Mating"
Type A B C D

AB Ah aB ah

Substances

Present

A

AB

B

Ah

C

aB

D

ah

a

a,/3

a,^

a

—

0

Note, a, substance A reacts with substance a; jS, substance B reacts with substance b.
° Jennings' alphabetical system for designating mating types in P. bursaria is used here.

This concept may be extended to include an eight-type variety
(variety II) by postulating three independent pairs of specifically
interacting substances. Each mating type should possess three mating-
type substances, one from each independent pair. By proper random
arrangement eight mating types should result. Any one of these types
should react with all the others, but not with itself, through specific
interaction between one or more pairs of substances.

Support for this hypothesis is found in the mating behavior of
"adolescent" clones (Jennings, 1939a) and the intervarietal reactions
reported in F. bursaria (Jennings and Opitz, 1944).

Exconjugant clones in P. bursaria pass through an "immature"
period of weeks or months during which they will not conjugate or
give mating reactions. This may be followed by a period of "adoles-
cence" during which the clones mate poorly and in some cases fail

298 SEX IN MICROORGANISMS

to give the complete or "adult" spectrum of mating reactions. Such
clones are reported in variety I, a four-type variety. Here the adoles-
cent clones in question mate at first only with two of the four mating
types. Upon reaching maturity they mate with a third type, as well,
giving the full spectrum of reactions. As Jennings points out, it is of
considerable interest that such clones mate either with types A and B
or types C and D. The adolescent clones are therefore type CD or
AB, respectively. CD clones mature to C or D, and AB clones mature
to A or B. No BC or AD adolescent clones are reported. These obser-
vations fit the a-(3 concept on the assumption that the two systems of
substances do not appear simultaneously as the clones reach adoles-
cence. Thus the synthetic mechanism which produces the a system of
mating substances may appear at adolescence, w^hereas the mechanism
for i^i substance formation does not function until later (maturity).
During adolescence, then, the clones have one or the other of the a
substances (A or a) and will mate with the two mature types possess-
ing the complementary substance (a or A) indicated in Table III.
Upon reaching maturity, one or the other of the ^ substances (B or
b) is also produced and the final mating type is established.

Only one series of intervarietal reactions has been reported in
P. bursaria (Jennings and Opitz, 1944). One of the two mating types
in variety IV (type R) cross reacts with four of the eight types
(types E, K, L, M) in variety II. Therefore four of the eight mating
types in variety II are similar. They may have one mating substance
in common which in each cross reaction combines with the same
complementary substance of variety IV, type R, animals.

Thus in the irregular "adolescent" and intervarietal reactions
the results support the hypothesis to the extent that half of the mating
types in a multiple-type variety should have one mating substance in
common. Unfortunately, no other data pertaining to this problem
are available.

Activation-Initiating A4echanism in Paramechnn.

Until recently little was known regarding the activation-initiat-
ing mechanism of fertilization in Par ante cium. However, since the
systematic development of methods for partially isolating this system
(Metz, 1946, 1947) our understanding has increased to the point
where more, is now known regarding the activation-initiating mecha-

THE PHYSIOLOGY OF- FERTILIZATION IN CILIATES 299

nisni in raravicciuiii than in any other organism. Partial isolation of
the activating system is achieved by killing and fixing paramecia
without destroying their ability to activate living animals. Such dead
paramecia may be regarded as a collection of highly specific sub-
stances adsorbed to or built into an inert carrier, the bulk of the dead
animal. By utilizing this partially isolated, static system, the activation-
initiating mechanism has been examined most thoroughly in P. mirelia
(Aletz, 1946, 1947, 1948; Metz and Foley, 1949) and somewhat less
extensively in P. calk'msi (Metz, 1948; Aletz and Butterfield, 1951)
and F. cmidatmn (Hiwatashi, 1949b, 1950).

A\Mien properly prepared dead paramecia are mixed with reactive
li\ing animals of complementary mating type, the living and dead
animals promptly adhere and under favorable conditions form large
mating reaction agglutinates as in normal conjugation. These mating
reaction agglutinates break down after 1 to 2 hours {P. mirelia)^ re-
leasing the living animals. These may be freed as single individuals or
as "pseudo selfing" pairs (Fig. la) joined only at the holdfast region*
{P. aurelia and P. calkinsi; Metz, 1947, 1948). The released animals
(singles as well as pairs) then proceed to undergo meiosis and macro-
nuclear breakdown in normal fashion and according to the time
schedule of normal conjugants (Metz, 1947). None of these events
is observed in mixtures of living and dead paramecia of the same mat-
ing type. This activation of living by dead animals of opposite type
has been reported in types 7 and 8, variety IV of P. aurelia (Metz,
1947); type II of P. calkinsi (Metz, 1948); and types 2, 4, 5, and 8 of
P. cazidatum (Hiwatashi, 1949b). With this brief outline of the newer
methods we may now attempt to analyze the activation-initiating
mechanism in Parameciiivi.

Activation through sexual processes in Paramecium requires con-
tact of potential mates. This follows from the facts that culture fluids,
filtrates, and the like have no specific efl^ect upon paramecia of op-
posite mating type (Sonneborn, 1937, 1939b; Kimball 1943; Metz,
1947); that animals which have been killed, fixed, and repeatedly
washed retain their ability to activate living animals; and finally that
the ability of dead (or living animals) to activate is directly related to
their ability to give mating reactions. Since the process of activation
in conjugation requires contact or union of potential conjugants, the

* Hiwatashi (1951a) reports "pseudo selfing" pairs in P. caudatmn which
are united at both the holdfast and the paroral regions.

300 SEX IN MICROORGANISMS

various manifestations of activation must be initiated by one of the
following mechanisms: (a) direct interaction of fixed surface sub-
stances, (b) transfer or diffusion of substance (s), (c) a combination
of surface interactions and transfer of substances.

As Sonneborn (1949) points out, none of these possibilities has
been excluded by direct experiment. Nevertheless, the constitution of
the dead animals which are capable of activating living animals and
the nature of the reaction between them render the first of these
possibilities highly probable. The reaction between living and dead
animals is a superficial one to the extent that mates can be separated
mechanically at any time during their union. As seen in Table V
certain of the effective killing agents are strong fixatives. After treat-
ment with these agents the dead animals must be washed repeatedly
to remove the killing agent. In this process of killing and fixation, any
substance that is appreciably soluble in water would be removed from
the dead animals. Thus, any essential, diffusible agent would nec-
essarily be a rather special, relatively water-insoluble substance.
Unfortunately, dead animals that have been thoroughly extracted
with lipid solvents have never been tested for their ability to activate
living animals. However, it is not unlikely that the activating proper-
ties will withstand such treatment, since lyophilized animals (P. aiire-
lia) give good mating reactions after extraction with absolute acetone,
ether, chloroform, or benzene (Metz and Fusco, 1949). In view of
these considerations it appears highly probable that the activation-ini-
tiating substances are surface substances and that the essential activat-
ing reaction (s) is a reaction between these surface substances. The
question then arises as to what surface substances interact to initiate
activation. Experimentally, the problem is most readily approached
from a consideration of the three types of union that occur in con-
jugation. These are: (1) the mating reaction union, (2) the holdfast
union (Fig. la), and (3) the paroral union (Fig. lb).

Of these several types of union the last is clearly unnecessary
for activation, as shown by the following three observations: (1)
Paroral union occurs after the holdfast union is formed, mating reac-
tivity is lost, and meiosis has begun. (2) Dead animals which are
capable of activating living animals do not possess paroral cones. (3)
The third animal in conjugating "threes" (Chen, 1940a, 1946b; Metz,
1947) is usually joined to the primary pair only by its holdfast region.

THE PHYSIOI.OCIV OK IKRTILIZA 1 ION IN CILIATFS 30I

Nevertheless, this third animal undergoes a complete nuclear cycle
(cvtogamv).

Holdfast union, the second union in conjugation, is also not an
essential preliminary to activation in Paraviechmi. This is evident
from the fact that the reaction between dead and living animals,
which leads to activation of the latter, does not involve a holdfast
attachment. Although this evidence clearly eliminates typical hold-
fast union as a possibility, it cannot be considered a critical elimina-
tion of holdfast substances from participation in the activating reac-
tion. Since the dead animals are derived from normal cultures whose
members can form holdfast unions, it is possible that they possess
special preformed holdfast substances. These might interact with
similar substances on the living animals even though this interaction
does not lead to lasting union. The same argument may be applied to
paroral substances.

Both holdfast and paroral substances were clearly eliminated as
essential factors in the initiation of activation by a study of a mutant
stock of F. aurelia (Metz and Foley, 1949). Animals of this CM
("can't mate") stock give good mating reaction under appropriate
physiological conditions. However, these animals are incapable of
proceeding further in the conjugation process. Thus CM animals
never form holdfast or paroral unions with CM or normal animals.
Furthermore the CM animals do not undergo loss of mating activity,
meiosis, or macronuclear breakdown through sexual association with
CM or normal animals. In short, CM animals cannot be activated by
CM or normal animals of opposite mating type. Nevertheless, CM ani-
mals can induce loss of mating activity, holdfast union ("pseudo

TABLE IV

Activating Properties of CM and Normal Animals of Opposite Type

(P. aurelia, Var. 4)

Living

Normal

Normal

CM

CM

Formalin-killed

Normal

CM

Normal

CM

Mating reaction

+

+

+

-h

Loss of mating reactivity

+

+

—

—

Holdfast union

4-

+

—

—

Meiosis

+

+

—

—

Macronuclear breakdown

+

+

—

—

302 SEX IN MICROORGANISMS

selfing" pair formation), meiosis, and macronuclear breakdown in
normal animals. These relations were discovered in mixtures of living
and dead animals and are so summarized in Table IV, but they apply
equally well to living-living mixtures of animals of opposite type.

Since the CM animals cannot form holdfast or paroral unions, it
is reasonable to conclude that the CM animals lack any special hold-
fast or paroral cone substances. Nevertheless, the CM animals can
activate normal animals, and therefore they possess the activation-
initiating mechanism. Consequently, interaction of holdfast or paroral
substances is not essential for activation. The only other known sub-
stances which could interact to initiate the various changes in con-
jugation, then, are the mating-type substances. Therefore, activation
in Parmnecium must result from interaction of some as yet unknown
substances, or from interaction of the mating-type substances. The
latter alternative is accepted as the simpler hypothesis.

Interrelation of the Activation Phenomena

The interrelationship of the events which follow the activating
reaction have not been investigated as thoroughly as their importance
warrants. However, some information is available and more may be
expected in the future.

Since the subsequent events in conjugation follow in an orderly
sequence from the initial reaction, it is reasonable to suppose that

HOLDFAST SUBSTANCE FORMATIONl?)

INTERACTION CM NATURAL AUTOGAMY /

OF SURFACE • I X INITIATED HERE / ^

• / / » LOSS OF MATING ACTIVITY

(MATtNG TYPE?)
SUBSTANCES

> > > > — * -> -» PARORAL CONE FORMATION

\^-

\ ^ MEIOSIS

\

MACRONUCLEAR BREAKDOWN

a b C d e

Fig. 2. Scheme for activation in Farainecium.

(a) Initiating reaction (mating-type-substance interaction?) in sexually induced acti-
vation, (b) CM block, here assumed to lie "internal" to the initiating reaction, (a),
(c) Position where chain is activated in natural autogamy, (d) Breakup of main
activation chain into side reactions leading to (e) the various end effects of acti-
vation.

THE PHYSIOLOGY OF FERTILIZATION IN CILIATES

303

rhcv arc related to this initial event throug-Ji a predetermined chain of
reactions. One form of such a sclieme has been presented graphically
by Metz (1948), and his figure is reproduced here (Fig. 2). Support
for this type of scheme is derived from the study of abnormal or
mutant stocks. Of these the CM stock in P. aurelia has been examined
most thoroughly (Aletz, 1948; Metz and Foley, 1949). it. will be
recalled (Table I\ ) that the CM animals can activate normal animals
but that thev cannot themselves be activated by sexual means. Appar-
ently some block, the CM block, prevents activation from proceeding
much beyond the initial stages in the CM animals.

A< — >a

—>-»-»

I I
I I

I< — >R-

Rk—^V

I I

I b I

Fig. 3. Two possible activation-initiating mechanisms.

Each series of arrows represents the main activation chain in one conjugant. (a)
Simultaneous activation of conjugants by interaction of a single pair of surface
substances, (b) Simultaneous activation of conjugants by interaction of two pairs
of surface substances.

The CM block is of particular interest because the CM animals
regularly undergo natural autogamy. Two important facts may be
deduced from this: (1) Activation in conjugation (and its experi-
mental variants) and natural autogamy are initiated through different
mechanisms. This will be discussed under parthenogenesis. (2) The
activating system "internal" to the CM block is intact. This suggests
that the block may be a relatively simple deficiency such as the lack
of an enzyme or essential substrate.

At least two possibilities exist for the position of the block. These
depend upon the nature of the activation-initiating mechanism. If
activation is initiated simultaneously in both mates by interaction
of a single pair of substances (Fig. 3a), the CM block must lie internal

304 SEX IN MICROORGANISAIS

to the initiating mechanism. This follows from the fact that CM ani-
mals can activate normal animals. A second possibility requires that
each conjugant activate its mate by an independent system. Thus, an
inducing substance / in one animal combining with a reacting sub-
stance R in its mate might activate the latter. A reciprocal arrange-
ment of similar substances (R\ V) would be required to activate the
second animal. These relations are represented in Fig. 3 b, and it is ap-
parent that the CM block could reside in the activating mechanism if
the CM animals lacked R. Since there is no evidence that two such
independent systems operate in activation, the first alternative view
is accepted. Thus the CM block is placed internal to the initiating
mechanism in Fig, 2.

Sonneborn (1942b,c) has described another "can't mate" stock
in F. aurelia. Animals of this type I (variety 1 ) stock gave good mat-
ing reactions with normal type II animals, but they failed to form
lasting pairs. Sonneborn describes pairs of type II animals from the
mixtures, but his account is not sufficiently detailed to determine if
these were pseudo selfing pairs. Sonneborn's (1942b) statement that
the clumps "continually break up and reform" suggests that loss of
mating reactivity was not induced in the type II animals, but this is
difficult to establish with certainty except by a controlled experiment
(Metz and Foley, 1949). Unfortunately neither the mutant type I
or the normal type II animals from such mixtures were examined
cytologically for nuclear evidence of activation.

As Sonneborn (1949) points out, little information is available
regarding the relationship between holdfast substance formation, loss
of mating reactivity, paroral cone formation, meiosis, and macronu-
clear breakdown. These may arise independently from a main chain
as indicated in Fig. 2, or one or more of them may be sequential to
another. The available information regarding each of these will be
given in the order listed.

Holdfast Substa?]ces. Holdfast union may result from interac-
tion of special holdfast substances which are produced as an early
manifestation of activation and which are not mating-type specific
(Metz and Foley, 1949). Such holdfast substance formation may
branch from the main activation chain between a and c (Fig. 2) since
autogamous animals never form holdfast or other unions. However,
the intimate association of animals obtained only in mating reaction
agglutinates may be necessary for holdfast union. The action of cer-

THF. PHYSIOLOGY OF FERTILIZATION IN CILIA IIS 305

tain "killer" fluids may bear upon this question (Chen, 1945; Jacob-
son, 1948; Preer, 1948). The killer fluids in question induce activation
with and without pairing in certain "sensitive" stocks (see further
under parthenogenesis). The description (Chen, 1945) of the result-
ing pairs suggests that pairing involves a holdfast union. Therefore
it is possible that, in these exceptional pairs, holdfast substances are
produced in response to the activating action of the killer fluids and
that the animals unite directly through holdfast substance interaction.
The fact that pairs form only after fluid and animals have been mixed
for some time favors this view. If this explanation should prove cor-
rect, the mating reaction could not be a prerequisite for holdfast
unions, and failure to find holdfast unions in autogamy could best be
explained by assuming that holdfact substances are not formed in this
process. Since CM animals do not form holdfast unions, holdfast
substance formation must arise beyond b (Fig. 2). Thus the argument
presented here suggests that holdfast substance formation arises be-
tween b and c.

Loss of Mating Reactivity. This reactivity occurs in conjuga-
tion, its natural and experimental variants, and in natural autogamy.
No statement is available regarding loss of mating reactivity in re-
sponse to "killer fluids" and no case of activation without loss of
mating reactivity has been reported. However, mating reactivity can
be regained in a remarkably short time after conjugation, as Diller
( 1 942 ) and others have shown. Loss of mating reactivity does occur
in certain abnormal material (P. biirsaria) which fails to undergo
the normal nuclear cycle and in which conjugants separate prema-
turely (Chen, 1946d). Furthermore, Tartar and Chen (1941) found
that Uving enucleate fragments (P. biirsaria) lost mating reactivity
after clumping for a short time with animals of opposite type. From
these observations it appears that loss of mating reactivity can follow
directly from the mating reaction and is not dependent upon other
activation phenomena. It may be sequential to but not dependent
upon holdfast substance formation, since it occurs in natural au-
togamy. Thus it might arise between c and d in Fig. 2, Paroral cone
formation, macronuclear breakdown, and meiosis could be sequential
to loss of mating activity.

Paroral Cone Formation. Observations on paroral cone forma-
tion are limited. Hertwdg (1889) clearly described these structures in
conjugants. They are formed also in pseudo selfing pairs (Metz,

306 SEX IN MICROORGANISMS

1947), animals activated by "killer" fluids (Chen, 1945), and natu-
rally autogamous animals (Diller, 1936). Paroral cone formation
should be distinguished from the process of cone fusion, since the
presence of cones does not automatically insure fusion. Actual fusion
occurs only in conjugation and possibly certain cases of "pseudo self-
ing" (Hiwatashi, 1951a). The presence or absence of cones has not
been reported specifically in abnormally conjugating material. How-
ever, cones presumably form in conjugating amicronucleate animals
since pronuclei pass to these from normal mates (Chen, 1940c). A
similar situation obtains in "abbreviated" conjugation (Diller, 1949),
where micronuclei are exchanged at precocious stages of meiosis. It
is clear from these cases that paroral cone formation does not depend
upon the presence of a micronucleus or particular meiotic stage.

Macrojjuclear Breakdown. The macronucleus breaks down in
a characteristic way (see Hertwig, 1889; Maupas, 1889; Metz, 1947;
Sonneborn, 1947, for figures) in conjugation, "pseudo selfing" (Metz,
1947), natural autogamy (Hertwig, 1914; Diller, 1936), and in
"killer" fluid activation (Freer, 1948). Presumably macronuclear
breakdown follows the normal pattern in conjugating amicronucleate
animals (Sonneborn, 1938b; Chen, 1940c) and therefore is not
dependent upon a micronucleus. Macronuclear breakdown followed
by some degree of macronuclear regeneration apparently can occur
without other obvious manifestations of activation (Diller, 1936).
However, such hemixis does not follow the normal breakdown pat-
tern, according to Diller's account, and therefore its relation, if any,
to activation is uncertain.

Meiosis. None of the normal non-meiotic events of activation
depends upon the micronucleus, since all these apparently occur in
conjugating amicronucleate animals. The initiation of meiosis may
be sequential to holdfast substance formation and loss of mating reac-
tivity for the reasons given above, but its serial relation to other events
has not been determined. In any event, the meiotic process itself is
subject to many independent variations. It may consist of nothing
more than a swelling of the micronucleus in abnormal stocks (Chen,
1946d), or it may progress in normal or abnormal fashion through
one or both divisions in intervarietal crosses between normal stocks
of P. bursaria. Chen (1946c) suggests that this results from mixture
of incompatible cytoplasms. In "abbreviated" conjugation (P. caiida-
tiim) the meiotic process may be out of phase with other events

THE PUVSlOIXKiV Ol" II ,R 1 Il.l/A HON IN CILIA lES 307

(Dillcr, 1949), and, in P. Ciilkinsi, type 11, activated by formalin-killed
animals, mciosis is frequently arrested in the prophase of the first
division in spite of the fact that the macronucleus subsequently
breaks down in normal fashion (Met/, and Foley, unpublished). Far
more spectacular deviations from the normal process have been de-
scribed by Nanney (1952) in centrifuged Tetrahyviena conjugants.
In summary form, the evidence presented in this section shows
that holdfast substance formation could arise independently from a
main activation chain at a point between b and c in Fig. 2. Loss of
mating activity certainly arises beyond c. It may arise before d.
Therefore paroral cone formation, macronuclear breakdown, and
meiosis could be sequential to holdfast substance formation and loss
of mating reactivity, but there is no evidence for sequential order or
dependence among these three phenomena. Clearly none of the other
activation phenomena is dependent upon meiosis or a micronucleus
since all occur in amicronucleate animals.

Physical Basis for the Mating Reaction

According to the views presented here the mating-type sub-
stances perform three major functions in fertilization: (1) They
effect the initial adhesion of potential conjugants. (2) They supply
the primary specificity in conjugation. (3) Their interaction triggers
the entire series of activation changes.

In view of their primary role in fertilization it is now essential
to characterize the mating-type substances in more clear-cut physico-
chemical terms. Specifically, three items of information are needed,
namely, the location of the mating-type substances, their chemical
nature, and their manner of interaction. These are considered below.

Location of the Matmg-Type S^ibstances. All workers agree
that the initial adhesion involves the cilia of animals of complementary
mating type. Jennings (1939a) and Tartar and Chen (1941) noted
that the surfaces of F. bursaria, when united in the mating reaction,
were separated by a space that approximated the length of one cilium.
These observations are readily confirmed by casual observation, but
unfortunately there are no detailed descriptions of the mating reac-
tion union in the literature. Therefore, Metz and Pitelka (unpub-
lished) undertook to examine this reaction in F. calk'msi with phase
contrast optics. In this form the cilia appeared to agglutinate tip to

308 SEX IN MICROORGANISMS

tip. Not infrequently the tips of several cilia of two mating animals
all adhered together to form a tight knot at the point of union. Even
single isolated cilia prepared by sonic treatment of reactive formalin-
killed animals adhered to living animals of opposite type. Again the
point of attachment involved one end of the isolated organelle. AVhen
tension was placed upon two joined cilia, they drew apart but re-
mained attached by a fine, apparently elastic, thread which finally
broke. This is believed to represent stretching and final breaking of a
cilium sheath. If such a break did not occur at the oriq-inal point of
union of the cilia, a piece of the cilium surface membrane would
necessarily be transferred from one animal to another. This would
readily account for Sonneborn's (1937, 1942b,c) observation that a
Paramecium of one mating type can clump with another of the same
mating type after it has first clumped with one of the opposite type.*
Unfortunately, these observations, though quite suggestive, do not
exclude other regions of the cilium or even the pellicle from participa-
tion in the mating reaction. The animals were necessarily under con-
siderable compression when observed. Furthermore, the behavior of
individual mating cilia could be studied only on very loosely united
animals or at regions where the pellicles of the mates were relatively
far apart. These observations (Metz and Pitelka, unpublished) leave
no doubt that the cilium surface possesses mating-type substances,
and that further study should be directed toward the structure of
these organelles, particularly their tips.

Several workers have examined paramecium cilia with the elec-
tron microscope, but none of these studies has revealed any mating-
type-specific organization. Jakus and Hall (1946) found that cilia
contain a number of fine fibrils, but these workers were unable to
detect a membrane or limiting sheath about the cilium. Recent stud-
ies, however, leave no doubt that such a structure forms the limiting
boundary of the cilium. Such a membrane has been observed in sec-
tioned cilia by Lansing, Hillier, and Sonneborn (unpublished), and
its presence has been confirmed (Heilbrunn, 1952; Wichterman,
1953) in material prepared by the COo-critical-point method (Ander-

*Sonneborn's observation has been confirmed repeatedlv (Metz, unpub-
lished) in mixtures of living and lyophilized (Metz and Fusco, 1949) P. cal-
kinsi of opposite type. The living and lyophilized animals clump on mixing,
and shortl)' thereafter numbers of living animals mav be seen adhering to one
another.

Plate I. Electron micrographs of P. calk'msi, type 11, prepared to show the

ciHum membrane.

Reactive F. calkinsi, type II, were killed in 5% formalin, frozen in liquid air, and
lyophilized to the collodion membrane. The "blisters" in (1) and the expanded
membrane in (2) probably resulted from partial melting of the preparation during
lyophilization.

309

310 SEX IN MICROORGANISMS

son, 1950) and by Metz and Pitelka (unpublished) in lyophilized
preparations of P. calkinsi (Plate I). The last study is the only
serious visual attempt that has so far been made to detect mating-type
differences in paramecium cilia. The material was prepared for elec-
tron microscopy in several ways including drying living and formahn-
killed paramecia in air and lyophilizing living and formalin-killed ani-
mals directly to the collodion membrane. Comparisons were then
made among (1) cilia of animals from reactive and unreactive cul-
tures, (2) cilia from reactive animals of the two complementary
mating types, and (3) ciha of control animals with others that had
been removed from mating reaction agglutinates. These comparisons
revealed no consistent differences within the three categories. These
observations thus confirm the original supposition, namely, that the
mating-type substances constitute part of the molecular pattern of the
cilium surface membrane. No evidence was obtained for an extra-
cellular coat or capsule about the cilium. The tips of the cilia like-
wise showed no specialized organization. However, a sharp reduction
in diameter was observed approximately 1 micron from the cilium
tip. This narrow terminal part of the cilium is evidently more deli-
cate than the remainder of the organelle, since it was distorted in
many of the preparations.

Chemistry of the Mating-Type Substances. A proper chemical
study of the mating-type substances should reveal the general chem-
ical nature of these substances. Furthermore it might be expected to
demonstrate specific differences between complementary substances.
Such differences might account for the specificity of the mating
reaction and provide some information about the manner of inter-
action of these agents.

Ordinarily, characterization of substances of biological origin is
achieved by appropriate extraction and purification followed by an
identification and description based on physical and chemical proper-
ties. A number of different physical and chemical agents have been
employed in attempts to extract mating substances from paramecia.
They include breaking up the reactive animals by mechanical means
(Metz, 1946; Metz and Butterfield, 1950; Hiwatashi, 1950), by
freeze-thawing (Metz, 1946), irradiation with x-rays (Wichterman,
1948), heating (Metz, 1946), acid and alkaUne extraction (Metz,
1946), digestion with enzymes (Metz and Butterfield, 1950), extrac-
tion with salt solutions (Metz, 1946), detergents (Metz, unpub-

THF, PMVSIOI.OGV OF 1 1 ,R IILIZATION IN CILIATES 311

lishcd), and a variety of organic solvents (Mctz, 1946). In spite of
these efforts no extract has yet been obtained which inhibits the
mating reactivity or otherwise specifically affects animals of com-
plementarv mating type.

Since all attempts at extraction have uniformly met with failure,
other methods for characterizing the mating-type substances are re-
quired. The most fruitful of these has been a study of the effect of
various agents on the mating reactivity of intact paramecia. Of the
agents employed (see Table \^), enzymes and "specific group rea-
gents" have provided the most positive information regarding the
nature of the mating substances.

Evidently the mating-type substances (P. calk'uisi) are either
proteins or substances closely associated with protein, since mating
substance activity is destroyed by proteolytic enzymes (Metz and
Butterfield, 1951).

If mating substances are proteins, amino acid residues might
well be essential for their activity. To test this possibility certain
"protein group reagents" were examined for their effect on mating
substance activity. These agents, the conditions under which they
were employed, and their effects are given in Table V and discussed
below. The various limitations in this type of analysis may be found
in several recent works (Glick, 1949; Danielli, 1949, 1953; Barron,
1951; Herriott, 1947; Olcott and Fraenkel-Conrat, 1947) and nu-
merous short papers.

Oxidizing and reducing agents do not affect mating reactivity.
Likewise, rather drastic treatment with mercuric ion and iodoacetate
treatment sufficient to render paramecia nitroprusside negative failed
to destroy mating reactivity. Therefore sulfhydryl and disulfide
groups are not essential constituents of the mating-type substances
(F. calk'msi) . All the amino and phenolic group reagents tested inac-
tivated Paramecium when used under the mild conditions generally
employed in protein group studies. Therefore it is probable that both
of these groups are "essential groups" in the mating-type substances
(P. caJkmsi). Thus nitrous acid, dinitrofluorobenzene, and diazonium
compounds react with both amino and phenolic groups. Iodine reacts
by substitution in phenolic groups (the inactivating action of iodine
cannot result from oxidation of sulfhydryl groups since these are not
essential), and benzoyl chloride and formalin combine with amino
groups. It should be noted that formalin-killed, reactive animals (P.

o

m

H

W

o

<

^

.S E
;r3 <u

H

W)

<

On „- On

3

fclH

3

■ — ^ t/1 On

3 == M

N N

^ ON

3

t3
in

3

a
c

3

S) N

.5 oj •-

^g SSK SSffi S^ ^

ID

^ "S

O tfi

-a _g

03 QJ (J

aj o

.H CN

u

S o

fe; a:;

S =^ S o

fe

l_q tM [iH tin J

nJ

cS

rt

g

>

g

>

O

>

S

•^ -2

<S

W

*i*

W

S «

•^

fi^

W

•^

S

•«s»

3 "^

^

•^

'Tri

V>

^

S s

?!

Sv

s>

:<

3 e

S

t-(

t^

^

a

^

J ^j

O

vo

<o

t3

-CS

O

CI, ci,cx,tx,tt,CL,a,a,

X

1-1
>

00

ID

v^

J3

ds

O^

O

.■ti

W

ffi

5^

>.

a

a

o"

3

-o

o

o

o

_o

3

3

-o

00

■th"

"S

■o

's

ro

'^^

2 >-

'3

<u

11

ffi

o "^

'c5

to

U

o

O

S
o

U4

cTcT
i7) "^

-T3

O
3

o o

o

o o

E

c3

J

"O

^ .

tsT

^ '^

^j

^ ^

«

Ei ^

s.

Ei Ei

Q

Q

■SJ

'-2 ^

•^A

•»S> *r^

■^

1^

s^

<U

j^

tj ."3

■o

<o Ij

"o

§ £

§

"^

c?5

"S ,

^

a:^ Ch

f^

Q> Q^

fe;

fe;-

^ o

10

»<

CU 2 Ph Ph

■^ -^ \M

3 <u

a-o rt <"

>,•- (U 4J 2

C -S O E W ^ 3
.3 O >- >, 3 3 •-

t/^

0)

ryps
lym
yalu
ysoz
ibon
tyali
ecith

E

r

H U ffi H-1 P< Ph ^-5

W

312

^

o

'»-•

U

-o

i><

c

w

rt

,i_)

o

(U

td

W

k^

C

ci

OJ

c

e

J

rt

'3

PLI

<

3

a

J£ .^ T3

O CN

R O R t/}

f^

=^

bC ^

J5 ^

o

.£P 8

o

to

-C I-

^

<ij

u^

-f fo

rS^

^ -^

^~*

W oo"

od"

^ ffi

ffi

^-T^ a

D-

"^ cu ^

^ t^ ^

i»J~

^ c

-S C -s^ *-■ ~

O O <»j O 'o

§ S 5 c g

c< oc; c<

fe

:=^ X

o

(n

OJ

c

OJ

C

_rt

^

"o

a

c3 IJ

o

•-^

>■

^>

^ ■?

"bi

O

5

u

^ t;^

_o

C/J^

o

s

15

;^

<

^ a

3 3

3 3

m-o

m -c) J-

t ^2

cj o

ni o

N V?

N "^ N~

j_i o^

*-! Cv -tJ

4) .^

OJ T-l OJ

^

>

o

-a

>-" '-j^ ID

3 ^ i:

03 O

o o

CS $j fvj fN

M'

cs

00 i^

■J^ .3

J

c ^

>

c^

313

^

J3

3

ID Oh

fcG

o

W

U

>

W

k^l

<

be

H

.S S

e 15

PM

fe5

1—1 t,

^-^

0< fe;

« .2

e

c«

t/}

'13

OJ

OJ

D.

a

S

^^

>,

>-

H

H

1) u

a

^ S

QJ "J

fo

U o

o c

X

[/I

OJ O

>

o

"S ^ "S

>

o

■ cjl-uu ^.

lO .B i*^ ^ .g

^ S; ^ •■§ ^ ^

-^ ^ -^ § -^ ^
fe; a< fe;

u u u

^^

P-i ~

Ph

'^^ KH

^

l3

o

3

OJ

3.

o

o

3
o

73.

N

S

N

'cj

C

.£

j^

a

rt

M

314

o
U

>

<

Pi

o .5
u

j3

a
c
3

<

^ s

r •) °^

^ J"

U £ %

■■s '5 IH

•o »S O

e uo CO

U

3

U

o<

-•2

u

'3b
o

^ ^

-*:

S s^ !3
« • ?

— 5 o --

.ii c o

o

g E 6 .

>- Cu a. <u
J^ >< X S

OJ

C

XI

a.

0)

c

X

.§

<u

w

43

Dh

•y

4-*

X

3

_c

<u

(U

^

-o

X

rt

(L>

E

3

C

;| S

•- -;:<! x;

-^ (L) .3

^ ■" -5 — ;3

1 1 i 1 1

ci

cS

s

(Sj

<u

4>

yj

<u

u^

k.

g",' '^

II

II

bC

II

PiH

P-,

ffi o

J

315

316 SEX IN MICROORGANISMS

calkinsi I) stain deeply (Metz, unpublished) with azo-carmine G
(Monne and Slautterback, 1951), and up to 10 per cent of the Kjel-
dahl nitrogen may appear as free amino nitrogen (standard formol
titration) in such animals, a value which compares favorably with
that for alcohol-killed control animals (Metz, unpublished). Evi-
dently, then, the amino groups of formalin-killed, reactive animals
are not bound to any appreciable extent.

Possible action of periodate on P-hydroxy amino acid residues
(Edsall, 1942) has not been investigated, but apparently hydroxyl
groups of carbohydrates are not essential (P. calkinsi I), for animals
retain reactivity at periodate concentrations which render them
strongly positive to Schiff' s reagent.

The failure of several organic solvents to remove or destroy
mating substance activity (Table V; Metz and Fusco, 1949) would
seem to eliminate loosely bound fat-soluble substances as essential
constituents. However, Chargaff's (1944) report that alcohol splits
lipoprotein, whereas various other solvents do not, is of some inter-
est, for this agent destroys mating reactivity even at low concentra-
tions (Metz, 1946; Metz and Fusco, 1949; Hiwatashi, 1950).

Although a large number of agents have been examined for
inactivating action on the mating-type substances of Paramecium,
only two of these have shown clear-cut differential action on com-
plementary mating types. These agents are formalin and nitrous acid.
Formalin-killed type II P. calkinsi never give strong mating reactions
regardless of the formalin concentration employed for killing. How-
ever, type I formalin-killed animals give intense reactions. On the
other hand, type II animals killed with HgCli, picric acid or by
lyophilization mate as intensively as similarly treated type I animals
(Table V; Metz and Butterfield, 1950). The differential action of
formalin is even more striking in P. caudatnm, for Hiwatashi (1949b)
reports that one of the two mating types in each of four different
varieties is completely unreactive after formalin killing, whereas the
complementary type in each variety gives strong reactions after for-
malin treatment. This action of formalin indicates two series of ho-
mologous mating types in P. caiidatuvi and suggests the interesting
possibility that these two series correspond to the two general mating
types established by intervarietal matings (Table II). Unfortunately
the data of Gilman and Hiwatashi are insufficient to establish this
relationship. Finally, it should be recalled that formaHn has no dif-

THE PHYSIOLOGY OF FERTILIZATION IN CILIATES 3 1 7

ferential action on the two mating types of variety 4, P. aurelia
(Metz, 1947).

A differential action of nitrous acid was found in P. calkinsi
(Table \^). Type 1 animals mate after treatment with nitrous acid
at 0*^0, whereas type II animals fail to do so. The mating reactivity
of both is destroyed by nitrous acid at 25 °C. It is tempting to ascribe
the differential action of formalin and nitrous acid to inactivation of
amino groups. However, the action of both these agents is complex
(see French and Edsall, 1945; Olcott and Fraenkel-Conrat, 1947),
and no such conclusion is warranted until this very interesting rela-
tionship has been subjected to a more thorough biochemical analysis.

From the chemical studies reviewed in this section it is evident
that mating-type-substance activity is dependent upon protein integ-
rity. Among amino acid residues sulfhydryl and disulfide groups are
not essential, but free alpha amino groups or terminal groups of basic
amino acids, phenolic groups, and perhaps other amino acid residues
appear to be essential.

Ivnmmological Studies on the Mating Reaction. Since the
mating-type substances of Farainecium cannot be extracted, an immu-
nological analysis of these substances in sitn might be expected to
yield valuable information regarding their nature. In some experi-
ments antisera affect complementary mating types differentially
(Bernheimer and Harrison, 1941; Hiwatashi, 1951b); in others the
two types behave alike (Bernheimer and Harrison, 1941; Sonneborn,
personal communication). Such variable results are not surprising
since the experiments were based on the immobilization of living
animals, and the presence of a given immobilizing antigen is known
to depend upon both the genotype and the environment (Sonneborn,
1950; Beale, 1951), at least in P. aurelia. In view of the absence of
mating-type specificity in these studies and the ambiguity of the
immobilization test in this regard, Metz and Fusco (1948) designed
a direct test for anti-mating substance antibodies in immune sera.
P. aurelia was used in the original study, and the results have been
confirmed on a less extensive scale with P. calkinsi. Reactive formalin-
killed animals were treated with antiserum for a substantial period
(30 minutes to 2 hours) and then washed free of the excess antiserum
and tested for mating reactivity with living reactive animals of the
same and complementary types. In some tests the antiserum-treated
animals failed to mate. This clearly indicates that the mating-type

318 SEX IN MICROORGANISMS

substances can be blocked by antiserum. However, this action of anti-
serum was not mating-type specific. In fact, several lines of evidence
suggest that the reactive groups of the mating-type substances do
not combine directly with the inhibiting antibody. Thus, inhibiting
antisera can be prepared from both reactive and unreactive para-
mecia; both can absorb out the inhibiting antibody; antisera which
inhibit the reactivity of one race and mating type fail to inhibit ani-
mals of this same mating type but of another race; finally, animals
of a single race and mating type may gain or lose the ability to give
the blocking effect. Granting a chemical constancy of mating sub-
stances within a type, these results must mean that the inhibiting
antibodies react with antigens other than the mating-type substances.
Therefore it is concluded (Metz and Fusco, 1948) that the inhibiting
antibody combines with some "neighboring" antigen and prevents
intimate contact in the mating reaction by a mechanical masking or
steric hindrance of the mating-type substances.

No direct attempt has yet been made to relate the mating-
reaction-inhibiting antigens to the immobilizing antigens of Sonne-
born, but in some cases, at least, the two appear to differ. Thus
cultures were obtained by selection (P. anrelia) in which the mating
reaction was inhibited by a given antiserum but was not immobilized
by the same serum.

It is evident from this study that the mating-type substances are
associated with antigenic material, but until specific anti-mating sub-
stance antibodies are obtained, this approach is not likely to be re-
warding. Perhaps such antibodies can be prepared by immunizing
animals other than the rabbit.

Parthenogenesis

The eggs of many metazoa can develop without the mediation
of a spermatozoon. The developmental machinery may be set in
motion spontaneously (natural parthenogenesis) or through the ex-
perimental application of various physical or chemical agents (arti-
ficial parthenogenesis). In view of the many other parallels that exist
between fertilization in Faramecimn and metazoa, it is not surprising
that phenomena comparable to natural and artificial parthenogenesis
also occur in Paramecium.

The protozoan counterpart of natural parthenogenesis is natural

THE PHYSIOLOGY OF FERTILIZATION IN CILIATES 319

autogamy.* This process occurs frequently and regularly only in
F. poly Ciiry 117/1 and P. aiirelia, and it has been studied intensively only
in tlic latter. Tlie conditions which induce autogamy in P. aurelia
arc identical with those which bring animals into mating condition,
namely, mild starvation. However, autogamy can be induced only
after a certain minimum number of fissions following the previous
autogamy or conjugation. This suggests that autogamy functions in
a "life cycle" in P. aurelht, and this is borne out by Sonneborn's
(1938c) demonstration that, when autogamy is prevented, the clone
eventually dies.

Conjugating and autogamous animals undergo nearly identical
physiological and morphological changes. In both processes the ani-
mals lose mating reactivity, form paroral cones, and undergo macro-
nuclear breakdown and meiosis in identical order. This indicates that
the same chain of reactions is operative in conjugation and autogamy
(Fig. 2). However, the chain of reactions is set in motion by diflFerent
mechanisms in the two types of behavior. In conjugation, initiation
of activation requires interaction of surface substances on mating ani-
mals; in autogamy, no mate is present. Of greater significance is the
fact that different routes or receptors are clearly operative in conju-
gation and autogamy. This is evident from the behavior of the mu-
tant CM stock of P. aurelia (Metz, 1948; Metz and Foley, 1949).
These CM animals possess the activation-initiating mechanism, at
least so far as their ability to activate normal mates is concerned, yet
they cannot themselves be activated by sexual means (Table IV).
Evidently some block, the CM block (Fig, 2b), prevents sexually
induced activation from proceeding beyond initial stages in CM
animals. Nevertheless, these CM animals regularly undergo natural

* Some may object to this use of the term parthenogenesis since self-
fertilization occurs in autogamy. This distinction between parthenogenesis
and autogamy has considerable genetic significance but no physiological im-
portance. The behavior of pronuclei at the critical fertilization stages is prob-
ably controlled by the cytoplasm (see Nanney, 1952) and follows a definite
pattern. Whether or not this pattern includes pronuclear fusion depends upon
the number of pronuclei present. Thus, when amicronucleate and normal ani-
mals conjugate, two haploid exconjugants are produced. These develop by
parthenogenesis in the genetic sense, for no zygote nucleus has been formed.
Yet the animals have been activated and have passed through all the other
fertilization changes (Chen, 1940b,c). For present purposes this is considered
a variant of the normal sexual process. As used here, parthogenesis means non-
sexual activation followed by a normal nuclear cycle with or without pro-
nuclear fusion.

320 SEX IN MICROORGANISMS

autogamy. Therefore, the main reaction chain is activated beyond
the CM block in natural autogamy through a mechanism and recep-
tor other than the normal sexual one. These relations are presented
graphically in Fig. 2.

No systematic effort has been made to induce activation in para-
mecia by physical or chemical agents under conditions that would
exclude natural autogamy. However, one group of agents is known
to activate under such conditions. These agents may therefore be
regarded as artificial parthenogenetic agents for Faramecmm. These
are culture fluids from certain "killer" stocks of Faramecium. The
effect has been reported in P. bursaria (Chen, 1945) and in P. aurelia
(Freer, 1948; Jacobson, 1948). Induction of paroral cone formation
(Chen, 1945), normal type macronuclear breakdown (Freer, 1948;
Jacobson, 1948), and probably meiosis (Chen, 1945) with or without
pairing are reported. When pairs are formed, they appear several
hours after addition of the "killer" fluid. These are not normally
conjugating animals (see under subsequent activation phenomena).
They may be united by the holdfast region. An interesting relation-
ship exists between the animals producing the fluids and those which
give the parthenogenetic response. In the effective combinations the
fluid and animals are of different varieties. In P. bursaria, fluid from
certain variety 5 animals activated animals of two mating types in
varieties 2 and 3 and of one mating type each in varieties 4 and 6.
Chen (1945) states that certain stocks in these varieties failed to
respond, but he gives no details. In P. aurelia the relationship is
highly stock specific. Killer stock G, variety 2 animals activate stock
F variety 1, type I sensitive animals. Fluids from stock G animals of
both mating types (III, IV) are effective. It is clear from these obser-
vations that the parthenogenetic effect is unrelated to mating type.
The parthenogenetic agent in such fluids is probably paramecin, since
the effect is produced by killer lines but not sensitive fines of stock G.
Why this agent activates certain sensitive stocks and not others is
unknown, and its mode of action has been investigated to a limited
extent only. Faramecin-treated animals undergo macronuclear break-
down several hours later than conjugating controls mated simulta-
neously. Evidently the action of paramecin is delayed in some way.
Jacobson (1948) suggests that the site of paramecin action is internal
and that this delay represents time required for the paramecin to
penetrate to this site. If this interpretation is correct, it means that

THE PHYSIOLOGY OF FERTILIZATION IN CILIATES 3 2 1

paramecin initiates activation tlirougli some other receptor than the
one operative in sexually induced activation. A thorough study of
the parthenogenetic action of paramecins is greatly to be desired, and
a systematic search for other agents with similar action should be
made. Since paramecin is a desoxyribonucleoprotein (van Wagten-
donk, 1948), other substances of this nature might well be investi-
gated. A study of the effect of parthenogenetic agents on blocked
stocks such as the CM stock should be particularly rewarding.

As Metz and Foley (1949) suggest, these studies on partheno-
genesis in Faramecium have certain interesting implications for par-
thenogenesis in metazoa. The CM study shows that natural parthe-
nogenesis (natural autogamy) in Faramechivt and sexually induced
activation are initiated through different routes. This situation may
obtain in other forms, such as hymenopterous insects where the tgg
can develop either parthenogenetically or by sperm activation. Like-
wise artificial parthenogenetic agents need not necessarily act through
the same receptor as the sperm (Metz and Foley, 1949; Runnstrom,
1949). In fact, different parthenogenetic agents may act specifically
at different points in an activation chain,

FERTILIZATION IN OTHER CILIATES

Breeding systems containing clear-cut mating types have been
described in detail in only one form other than Paramecium, namely,
Euplotes patella. Mating types also occur in Leucophrys patula, Ony-
chodrovnis grandis, Stylonichia pustidata, and Loxophyllum fasciola
according to Jennings' (1939b) interpretation of Maupas' (1889)
observations. Recently mating types have been reported briefly in
Tetrahymeva (Elliott and Nanney, 1952; Elliott and Gruchy, 1952),
Stylojiicbia putrina (Downs, 1952) and Euplotes bar pa (Katashima,
1952).

Abating types were first reported in Euplotes patella by Kimball
(1939). Subsequently Kimball (1941, 1942) and Powers (1943)
described the breeding system and inheritance of mating type in
greater detail. Kimball (1943) and Sonneborn (1947) have both
reviewed these studies extensively. Therefore, this discussion will be
confined primarily to a comparison of the physiology of fertilization
in Paramecimn and Euplotes.

Kimball (1943) reports two (possibly three) non-interbreeding

322 SEX IN MICROORGANISMS

varieties in Euplotes patella and two more sexually isolated varieties
in a morphologically distinct but similar form. Mating types occur
within all these varieties, but only one of. the E. patella varieties has
been studied in detail. In this variety six "mating types" are reported.
When animals (with some culture fluid) of one type are mixed with
animals (and fluid) of any one of the other five types, conjugation
occurs in the mixture. The course of events in such mixtures differs
in three ways from what has been described for Faraviecium: (1)
no sex reaction of any kind occurs for at least 90 minutes (Kimball,
1939); (2) mass mating reactions never occur; and (3) animal-free
fluids induce conjugation (selfing) in at least some of the other types.
The later events in conjugating Euplotes approximate the usual ciliate
pattern (Turner, 1930; Kimball, 1941; Katashima, 1952) and need
not be considered further, but the three points mentioned above de-
part radically from the scheme devised for Faraviechnu and warrant
serious treatment.

Since the action of culture fluids is the striking feature in Eu-
plotes, this subject will receive primary consideration. Three conju-
gation-inducing agents are found among the six mating types. Ani-
mals may produce one or two of these, depending upon their genetic
constitution (Kimball, 1942). The production of the substances de-
pends upon three allelic genes, one aflele for each substance. There
are, then, six kinds of animals (the six mating types), three which
produce one kind of substance (homozygous animals) and three

TABLE \T

Conjugation-Inducing Action of Animal-Free Culture Fluids
IN Euplotes patella

Conjugation ("selfing") is indicated by +; failure to observe conjugation by — .

Animal-Free

Fluids

Animals

Mating

Fluid Agent

Type

IV

VI

III

I

II

V

1

IV

_

+

+

_

—

+

2

VI

+

—

+

—

+

—

3

III

+

+

—

+

—

—

1,2

I

+

+

+

—

4-

+

1,3

II

4-

+

4-

+

—

+

2,3

\'

+

+

+

+

+

—

mr, piiYsioLocv or i-i r iili/a i ion in cilia it.s 323

which produce two of the three substances (hetero/ygous animals).
A particular Huid agent induces conjugation only in those animals
w hich do not produce the same agent. These relations are summarized
in Table \ \. In view of this action of fluids the constitution of con-
jugating pairs in mixtures of mating types is subject to uncertainty.
Pairs may be composed of animals of the same mating type (induced
by fluid of the other type), or of different type. As Sonneborn
( 1 947 ) points out, this is an important distinction which Kimball
(1942) and Powers (1943) have not made in all combinations of
mating types.

In a searching analysis of Kimball's and Powers' data Sonneborn
(1947) concludes that true interbreeding between animals of different
mating type occurs only in combinations of mating types in which
"each one induces the other to self, that is, when each produces a
conjugation-inducing substance that the other one does not produce"
(Sonneborn, 1947).* Sonneborn marshals considerable support for
this view and finds only one exception which may not be serious in
view of other exceptional conjugations reported by Kimball and
Powers.

Kimball (1943) discusses the mechanism of action of the con-
jugation-inducing fluids in Euplotes and concludes that these may
function as agglutinins or that they may induce the formation of
other adhesive agents. The former possibility seems unlikely since
the action of fluids is not immediate and since it does not readily
account for conjugation between animals of different mating types.
Therefore Kimball's second proposal is developed here in accord
with the ne\\'er knowledge of fertilization in Pmwnechnn.

If one overlooks exceptional cases of "unexplained" conjugation
and intraclonal selfing in the data of Kimball and Powers, it develops
that conjugation occurs only in situations were fluids could induce
the conjugation. This is implicit in Sonneborn's (1947) scheme. It
seems likely, therefore, that cill conjugation in Euplotes is mediated
by a fluid agent whether the mates are of the same or different types.
In other words, two animals cannot conjugate unless each has been

* According to Sonneborn's scheme animals of any one "mating type" will
actuaUv conjugate (h\bridize) with animals of only three of the other five
"types." All conjugation in mixtures with the remaining two types is selfing.
Thus the term mating type has a rather special meaning in Euplotes. Two
clones are of different tvpe if conjugation occurs in the mixture, even if the
conjugation is onh' selfing among animals of the same clone.

324 SEX IN MICROORGANISMS

exposed independently to the action of a fluid agent which it does
not itself produce. Possibly a particular conjugation-inducing agent
activates animals which do not produce this agent. The substances
involved in the initial adhesion might well be formed as a result of
such activation. If these adhesive substances acted in non-specific
fashion, any two activated animals, regardless of mating type, should
adhere on contact and might then proceed to conjugate. The effect
of the fluid agents on Euplotes would then be roughly analogous to
the activating action of mating substance interaction in Paramecium.
Initial union (by ventral cirri. Turner, 1941) of Euplotes would then
correspond to holdfast union in Faraviec'mvi. Viewed in another way,
the conjugation-indulging agents in Euplotes are analogous in action
to the killer fluids which induce pairing and nuclear reorganizations
in Faramecium (see under parthenogenesis in Paramecium) .

This view accounts for the failure to observe an immediate reac-
tion when fluids and animals or complementary types of Euplotes
are mixed. No effect is observed for at least 90 minutes. According
to the scheme this is the time required for synthesis of the "non-
specific adhesive substances." No mass mating reactions occur in mix-
tures of Euplotes mating types, and this also might be expected since
large numbers of animals would not necessarily be in the adhesive
stage simultaneously. In any event, mass agglutination of Euplotes
presents mechanical difficulties for the adhesive organelles (ventral
cirri) are present only on one side of the animal. The importance of
this factor may be seen in Kimball's (1941) observation that double
animals do form small groups or chains.

It might be expected from this view that fluids would induce
other activation phenomena in responsive Euplotes in the absence of
conjugation. Such action has not been reported. In fact Kimball
(1941) describes exconjugants (probably animals from pairs split in
early conjugation stages) which failed to show any nuclear changes.
Similarly Katashima (1952) separated conjugating E. harpa by killing
one pair member and then studied the nuclear behavior of the surviv-
ing member. When the operation was performed after the "prelim-
inary micronuclear division" (2 to 5 hours after beginning of conju-
gation) but before the first meiotic division, the nuclear composition
of the surviving isolate promptly returned to the vegetative condition.
When the operation was performed at later stages, the micronuclei
sometimes completed the nuclear cycle (cytogamy). Clearly, then,

THE PHYSIOLOGY OF FERTILIZATION IN CILIATES 325

intimate union of conjugants is required for micronuclear changes
in E. J.wpih Ne\ertheless, some physiological change does occur in
E. harpa even before permanent union is established, for pairs sepa-
rated at this early stage did not reunite or conjugate with other ani-
mals. AMiether this is a delayed effect of a fluid agent or a result of
initial union remains to be determined. Indeed, fluid agents have yet
to be demonstrated in E. harpa.

The possible relationship between conjugation-inducing agents
from fluids and the responsive animals presented above can be for-
maUzed in terms of specific interacting substances. As Kimball (1943)
is aware, this relationship forms a system resembling the blood groups
in man. However, further speculation on Euplotes is pointless because
Kimball's stocks appear to have been lost and there is little prospect
of further investigation in the immediate future.

Downs (1952) has described five mating types in another hypo-
trich, Stylouichia piitrina. These five types constitute a variety since
conjugation occurs when any two of them are mixed. Apparently
Downs has not yet tested animal-free fluids for conjugation-inducing
action. Therefore any comparison with the related Euplotes would
be premature.

CONCLUSIONS REGARDING METAZOAN FERTILIZATION

Thirty years ago the major problems in metazoan fertilization
seemed on the threshold of solution, for Glaser (1921) confidently
wrote "whatever transformations our views on the initiation of devel-
opment may undergo within the next few years, the zone within
which we seek for understanding is now marked off by the reaction
capacities of perfectly definite physiological compounds." Unfor-
tunately the present outlook is not so optimistic, for the factors
responsible for initial adhesion, specificity, activation, and most other
aspects of sperm-egg interaction remain obscure.

Of the various "sex substances" that have been obtained from
metazoan gametes only the tg^ membrane lysins from sperm have
well-understood action in fertilization (for recent discussions see
Krauss, 1950; Berg, 1950; Swyer, 1951.) These ^o^g membrane lysins
remove mechanical barriers that would otherwise block the approach
of the fertilizing sperm. In certain cases at least, these agents supply
a specificity factor in fertilization.

326 SEX IN MICROORGANISMS

Among other agents the sperm and egg isoagglutinins, fertihzin
and antifertiHzin (obtained from eggs and sperm respectively), have
been studied most extensively. The highly specific nature of their
interaction and the dramatic character of the agglutinating action
suggest that these substances are essential factors in metazoan fertili-
zation. This may prove to be the case, but as yet substantial exper-
imental evidence to support this view is wanting. Initial adhesion
of gametes may involve a fertilizin-antifertilizin reaction (Tyler,
1948), and a modest case may be made for a role in specificity
determination for these agents, but again experimental support for
these possibilities is lacking. A clear demonstration of the function
or functions of these extraordinary substances is much needed and
may well prove to be the turning point in the analysis of fertilization
in metazoa. Unfortunately, we have no more positive information
concerning the role of specific substances in the attachment of sperm
to egg, specificity determination, and activation initiation in metazoan
fertilization. For detailed accounts of the present status of these
problems, the reader should consult the reviews listed on page 284.

For Parmnecium our understanding of the mechanism of fertili-
zation has progressed to a somewhat more satisfactory position. The
analysis of fertilization in Paraiiieckim clearly demonstrates the pri-
mary role of the cell surface in fertilization. It appears that initial
adhesion, fertilization specificity, and initiation of activation depend
upon the interaction of complementary surface substances, the
mating-type substances. No diffusible agents of any kind appear to
operate. Faraviechnu then exemplifies the classical concept of a
"trigger reaction" in fertilization. As many workers have pointed out,
a similar interaction of surface substances may function in metazoan
fertilization. It is certainly clear from the many observations on
normal fertilization in the sea urchin (see Dan, 1950, for a recent
account). Nereis (Lillie, 1911, 1912) and the starfish (Fol, 1879;
Chambers, 1930; Horstadius, 1939) that activation of the egg results
from a rather superficial sperm-egg contact. Apparently this reaction
does not even require a firm union between sperm and egg, for
Goodrich (1920) was able to remove the activating sperm intact (?)
from the surface of the Nereis egg by microdissection. Similarly,
Reverberi (cited by Runnstrom, 1949) found, in certain interspecific
tunicate crosses, that the activating sperm not only failed to penetrate
but actually fell from the egg surface in some cases. Perhaps clarifica-

Tin: PUVSIOLOCn of I'l-RlIMZAriON IN CIMATFS 327

rion of some of the problems of fertilization may he obtained by
anah'/ing tiic "partialh' isolated" system as was done for Paramechivt.
Many scattered attempts to activate eggs with dead sperm and sperm
extracts are recorded in the literature. Althouirh these have been un-
successful, a more systematic effort \\'ould now seem in order. Indeed,
a beginning in this direction has been made by Hutner and Provasoli
(1951) in the alga Clamydomonas. These investigators found that
gametes killed by formalin, heat, ultraviolet light, or Streptomycin
clumped strongly with gametes of complementary mating type. Like-
wise Metz (1945) and Metz and Donovan (1951) showed that sea
urchin sperm killed w'lxh. heat {Strongyloceiitrotiis pnrpuratns) or
with Bouin's fluid {Arbacia) agglutinated specifically with fertilizin.
Evidently the sperm retained its specific organization when killed by
these agents, at least so far as one surface group is concerned. As yet
attempts to activate eggs with such dead sperm have failed, but fur-
ther efforts may yet demonstrate the utility of the method in such
material.

Regardless of the outcome of such attempts with eggs and sperm,
the ciliates offer a \\'ealth of material for the investigation of sexual
phenomena that should not be neglected. In this material a number
of different sexual mechanisms appear to exist, any one of which may
serve as a useful model for studies on other forms. In Paramecium
distinct mating types occur, and sexual reactions depend upon inter-
action of surface substances. In Eiiplotes "mating types" of a different
sort are found, and diffusible substances appear to mediate conjuga-
tion. Vorticella produces macro- and microconjugants by a single
division, and these may then copulate (Finley, 1939, 1952). The pos-
sible relation of sexual cycles to hormonal activity of the host in
parasitic forms such as Nyctotherus (Wichterman, 1937) presents
new opportunities for the investigation of the role of defined sub-
stances in sexual control. As a final attribute mutant stocks deficient
in their sexual mechanisms occasionally become available in this mate-
rial. As Metz and Foley (1949) have shown, analysis of such defec-
tive stocks can provide valuable information concerning fertilization.
xApart from such examples as those cited here, great numbers of cili-
ates remain to be studied by modern experimental methods. A thor-
ough examination of this reservoir of material certainly should extend
our general understanding of the physiology of fertilization,

328 SEX IN MICROORGANISMS

REFERENCES

Anderson. T. F. 1950. A new technique for the study of biological structures

in the electron microscope. Biol. Bull., 99, 315.
Barron, E. S. G. 1951. Thiol groups of biological importance. Advances in

EnzymoL, 11, 201.
Beale, G. H. 1951. Antigen variation in Faratnecium aurelia, variety 1. Gene-
tics, 37, 62.
Berg, W. E. 1950. Lytic effects of sperm extracts on the eggs of Mytilus edulis.

Biol. Bull, 98, 128.
Bernheimer, A. W., and J. A. Harrison, 1941. Antigenic differentiation among

strains of Parameciu?n aurelia. J. hnmunoL, 41, 201.
Bielig, H. J. and F. von Medem. 1949. Wirkstoffe der tierischen Befruchtung.

Experientia, 5, 11.
Boell, E. J., and L. L. Woodruff. 1941. Respiratory metabolism of mating

types in Faraineciufn calkinsi. J. Exptl. Zoo/., 87, 385.
Chambers, R. 1930. The manner of sperm entry in the starfish egg. Biol. Bull.,

58, 344.
Chang, M. C, and G. Pincus. 1951. Physiology of fertilization in mammals.

Physiol. Revs., 31, 1.
Chargaff, E. 1944. Lipoproteins. Advances in Protein Chem., 1, 1.
Chen, T. 1940a. Conjugation of three animals in Paramecium bursaria. Proc.

Natl. Acad. Sci. U. S., 26, 231.
. 1940b. Conjugation in Pararnecium bursaria between animals with

diverse nuclear constitutions. /. Heredity, 31, 185.
. 1940c. Conjugation in Para?7jecium bursaria between animals with very

different chromosome numbers and between animals with and without

micronuclei. Proc. Natl. Acad. Sci. U. S., 26, 243.
. 1945. Induction of conjugation in Parameciwn bursaria among animals

of one mating type by fluid from another mating type. Proc. Natl. Acad.

Sci. U. S., 31, 404.
. 1946a. Varieties and mating types in Paramecium bursaria. L New

variety and types, from England, Ireland, and Czechoslovakia. Proc. Natl.

Acad. Sci. U. S., 32, 173.
. 1946b. Conjugation in Para?neciu?n bursaria. I. Conjugation of three

animals. /. Morphol, 78, 353.
. 1946c. Conjugation in Paramecium bursaria. II. Nuclear phenomena

in lethal conjugation between varieties. /. Morphol., 79, 125.

1946d. Temporary pair formation in Paramecium bursaria. Biol. Bull.,

91, 112.

Chen, Y. T. 1944. Mating types in Parajnecium caudatum. Am. Naturalist, 78,
334.

Cleland, K. W., and L. Rothschild. 1952. The metabolism of the sea-urchin
&g^. J. Exptl. Biol, 29, 285.

Dan, J. C. 1950. Sperm entrance in echinoderms, observed with the phase con-
trast microscope. Biol. Bull, 99, 399.

THE PHYSIOLOGY OF FERTILIZATION IN CILIATES 329

Danielli, J. F. 1949. Studies on the cytochemistry of proteins. Cold Spring

Harbor Sy7nposia Quant. Biol., 14, 32.

. 1953. Cytocheniistry. John Wiley & Sons, New York.

De Garis, C. F. 1935. Lethal effects of conjugation between Paramecium.

aurelia and double monsters of Faraincciuni caiidatwn. Am. Naturalist, 69,

87.
Diller, W. F. 1936. Nuclear reorganization processes in Faramecium aurelia,

with descriptions of autogamy and 'hemixis.' /. Morphol., 59, 11.

. 1942. Reconjugation in Faraniecium caudatuni. J. Morphol., 70, 229.

. 1949. An abbreviated conjugation process in Faramecium trichium.

Biol. Bull., 97, 331.
Downs, L. W. 1952. Mating types in Stylonichia putrina. Froc. Soc. Exptl.

Biol. Med., 81, 605.
Edsall, J. T. 1942. The chemistry of the proteins and amino acids. Ami. Rev.

Biochevi., 11, 151.
Elliott, A. M., and D. F. Gruchv. 1952. The occurrence of mating types in

Tetrahyme7ia. Biol. Bull, 103, 301.
Elliott, A. M., and D. L. Nanney. 1952. Conjugation in Tetrahymena. Science,

116, 33.
Finley, H. E. 1939. Sexual differentiation in Vorticella microst07)ia. J. Exptl.

Zool., 81, 209.
. 1952. Sexual differentiation in peritrichous ciliates. /. Morphol., 91,

569.
Fol, H. 1879. Recherches sur la fecondation et la commencement de I'heno-

genie chez divers animaux. Mhn. soc. de phys. hist. nat. Gejieve, 26,

89.
French, D., and J. T. Edsall. 1945. The reactions of formaldehyde with amino

acids and proteins. Advances in Protein Chem., 2, 278.
Giese, A. C. 1941, Mating types in Parameciwm ?nultimicronucleatu7n. Anat.

Record (Suppl.), 81, 131.
Gilman, L. C. 1939. Mating types in Paramecium caudatum. Am. Naturalist,

73, 445.
. 1941. Mating types in diverse races of Parainecium caudatum. Biol.

Bull, 80, 384.
. 1949. Intervarietal mating reactions in Parainecium caudatin/i. Biol.

Bull, 97, 239.
-. 1950. The position of Japanese varieties of Faramecimn caudatum with

respect to American varieties. Biol. Bull, 99, 348.

Glaser, O. 1921. Fertilization and egg-secretions. Ajiat. Record, 20, 227.

Glick, D. 1949. Techniques of Histo- and Cytochemistry. Interscience Pub-
lishers, New York.

Goodrich, H. B. 1920. Rapidity of activation in the fertilization of Nereis.
Biol Bull, 38, 196.

Heilbrunn, H. V. 1952. General Physiology. W. B. Saunders Co., Philadelphia.

Flerriott, R. M. 1947. Reactions of native proteins with chemical reagents.
Advances in Protein Chein., 3, 170.

330 SEX IN MICROORGANISMS

Hertwig, R. 1889. tjber die Conjugation der Infusorien. Abh. bayer. Akad.
Wiss. Math.-Naturw. Abt, 17 (II CI., Abt. I), 150.

. 1914. tJber Parthenogenesis der Infusorien und die Depressionszustande

der Protozoen. Biol. 2,entr., 34, 557.

Hiwatashi, K. 1949a. Studies on the conjugation of Faraviechnii caiidamvi. I.
Mating types and groups in the races obtained in Japan. Sci. Repts.
Tohoku Univ., Biol., 18, 137.

. 1949b. Studies on the conjugation of Parmiieciinii caiidatinn. II. Induc-
tion of pseudoselfing pairs bv formahn killed animals. Sci. Repts. Tohoku
Univ., Biol, 18, 141.

. 1950. Studies on the conjugation of Paranieciinn caiidatuiii. III. Some

properties of the mating type substances. Sci. Repts. Tohoku Univ., Biol.,
18, 270.

. 1951a. Studies on the conjugation of FaraTiiecium caudatinn. IV. Con-
jugating behavior of individuals of two mating types marked by a vital
staining method. Sci. Repts. Tohoku Univ., Biol., 19, 95.

1951b. A new method for separating animals of one mating type from

their partners in a conjugating mixture of Faranieciuni. Annot. Xool.

Japan., 24, 83.
Horstadius, S. 1939. Uber die Entwicklung von Astropecten aranciacus L.

Fubbl. staz. zool. Napoli, 17, 222.
Hutner, S. H., and L. Provasoli. 1951. The phvtoflagellates. Biocheniistry and

Fhysiology of Frotozoa, 1, 27. Academic Press, New York.
Jacobson, W. E. 1948. Non-mating-type-specific "activation" in Faranieciinn

aurelia: induction of macronuclear breakdown in sensitive stock P of

variety 1 by killer stock G of variety 2. Anat. Record, 101, 708.
Jakus, M. A., and C. E. Hall. 1946. Electron microscope observations of the

trichocysts and cilia of Faraineciiini. Biol. Bull., 91, 141.
Jennings, H. S. 1911. Assortive mating, variability and inheritance of size in

the conjugation of Faraineciwn. J. Exptl. Z,ool., 11, 1.
. 1938. Sex reaction types and their interrelations in Faraniecium bur-

saria. I. Froc. Natl. Acad. Sci., U. S., 24, 112.
. 1939a. Fara7ueciti7/i bursaria: mating types and groups, mating behavior,

self-sterility; their development and inheritance. Am. Naturalist, 73,

414.
-. 1939b. Mating types and their interaction in the ciliate infusoria. Am.

Naturalist, 73, 385.
Jennings, H. S., and P. Opitz. 1944. Genetics of Faramecijim bursaria. IV. A

fourth variety from Russia. Lethal crosses with an American variety.

Genetics, 29, 5 76.
Katashima, R. 1952. Studies on Euplotes. I. Conjugation and cytogamv induced

by split pair method in Euplotes harpa. J. Sci. Hiroshima Univ., ser. B,

13, 111.
Kimball, R. F. 1939. Mating types in Euplotes. Am. Naturalist, 73, 451.
. 1941. Double animals and amicronucleate animals in Euplotes patella

with particular reference to their conjugation. /. Exptl. "Zool., 86, 1.

THR PHYSlOLOCiV OK FI.KTILIZATION IN CILIA IKS 331

— . 1942. The nature and inheritance of mating types in Eiiplotcs jhitclla.
Genetics, 27, 269.
-. 1943. Mating types in the ciHate protozoa. Quart. Rev. Biol., 18, 30.

Krauss, iM. 1950. Lytic agents of the sperm of some marine animals. /. Kxptl.

Zoo!., 114, 239.
Lillie, F. R. 1911. Studies of fertilization in Nereis. L The cortical changes in

the egg. U. Partial fertilization. /. Morphol., 22, 361.
. 1912. Studies of fertilization in Nereis. IIL The morphology of the nor-
mal fertilization of Nereis. IV. The fertilizing power of portions of the

spermatozoon. /. Exptl. Zool., 12, 413.

. 1913. The mechanism of fertilization. Science, 38, 524.

. 1914. Studies of fertilization. VL The mechanism of fertilization in

Arbacia. /. Exptl. Zoo/., 16, 523.
. 1919. Problems of fertilization. University of Chicago Press, Chicago,

111.
Loeb, J. 1899. On the nature of the process of fertilization, and the production

of normal larvae (plutei) from unfertihzed eggs of the sea-urchin. A///.

J. Physiol., 3, 135.
Alaupas, E. 1889. Le rajeunissement karyogamique chez les cilies. Arch. zool.

exptl. gen., ser. 2, 7, 149.
Metz, C. B. 1945. The agglutination of starfish sperm by fertilizin. Biol. Bull.,

89, 84.
. 1946. Effect of various agents on the mating type substance of Para-
mecium aurelia variety 4. Anat. Record, 94, 347.
. 1947. Induction of "pseudo selfing" and meiosis in Parai)iecium aurelia

by formalin killed animals of opposite mating type. /. Exptl. Zool., 105.

115.
-. 1948. The nature and mode of action of the mating type substances.

Am. Naturalist, 82, 85.
Metz, C. B. and W. Butterfield. 1950. Extraction of a mating reaction inhibiting

agent from Paramecium calkinsi. Proc. Natl. Acad. Sci. U. S., 36, 268.
. 1951. Action of various enzymes on the mating type substances of

Parameciuni calkinsi. Biol. Bull, 101, 99.
Metz, C. B., and J. Donovan. 1951. Specific fertilizin agglutination of dead

sperm. Biol. Bull., 101, 202.
Metz, C. B., and M. T. Foley. 1949. Fertilization studies on Parameciuy/i aurelia:

an experimental analysis of a non-conjugating stock. /. Exptl. Zool., 112,

505.
Metz, C. B., and E. M. Fusco. 1948. Inhibition of the mating reaction in Para-
mecium aurelia with antiserum. Anat. Record, 101, 654.
Metz, C. B., and E. M. Fusco. 1949. Mating reactions between living and

lyophilized paramecia of opposite mating type. Biol. Bull., 97, 245.
Monne, L., and D. B. Slautterback. 1951. The disappearance of protoplasmic

acidophilia upon deamination. Arkiv. Zool, ser. 2, 1, 455.
Nanney, D. L. 1952. Cytoplasmic control of nuclear behavior at conjugation

in Tetrahymena. Genetics, 37, 609.

332 SEX IN MICROORGANISMS

Olcott, H. S., and H. Fraenkel-Conrat. 1947. Specific group reagents for pro-
teins. Chem. Revs., 41, 151.
Powers, E. L. 1943. The mating types of double animals in Eiiplotes patella.

Am. Midland Naturalist, 30, 175.
Freer, J. R. 1948. A study of some properties of the cytoplasmic factor,

"kappa," in Fara7neciimi aurelia, variety 2. Genetics, 33, 349.
Rostand, J. 1950. La Parthenoge?iise Aniiiiale. Presses Universitaires de France,

Paris.
Rothschild, L. 1951a. Sperm-egg interacting substances and metabolic changes

associated with fertilization. Biochem. Soc. Symposia {Cainbridge, Eng.)

No. 7, 40.

. 1951b. Sea urchin spermatozoa. Biol. Rev., 26, 1.

Runnstrom, J. 1949. The mechanism of fertilization in metazoa. Advances in

Enzymol., 9, 241. Interscience Publishers, New York.
Sonneborn, T. M. 1937. Sex, sex inheritance and sex determination in Vara-

7jieciu7n aurelia. Proc. Natl. Acad. Sci. U. S., 23, 378.
. 1938a. Mating types in Paramecium aurelia; diverse conditions for

mating in different stocks; occurrence, number and interrelationship of

the types, Proc. Am. Phil. Soc, 79, 411.
. 1938b. Sex behavior, sex determination, and the inheritance of sex in

fission and conjugation in Paraineciimi aurelia. Genetics, 23, 168.
. 1938c. The delayed occurrence and total omission of endomixis in

selected lines of ParaiJieciiim aurelia. Biol. Bull., 74, 76.
. 1939a. Paramecium aurelia; mating types and groups; lethal interac-
tions; determination and inheritance. Am. Naturalist, 73, 390.
. 1939b. Sexuality and related problems in Parameciwii. Collecting Net,

14,1.
. 1942a. More mating types and varieties in Parameciwn aurelia. Anat.

Record, 84, 92.
. 1942b. Sex hormones in unicellular organisms. Cold Spring Harbor

Symposia Quant. Biol., 10, HI.
. 1942c. Evidence for two distinct mechanisms in the mating reaction

in Parameciu'ni aurelia. Anat. Record, 84, 92.
. 1947. Recent advances in the genetics of Parameciw}i and Eiiplotes.

Advances in Ge?ietics, 1, 263. Academic Press, New York.
— . 1949. Ciliated protozoa: cytogenetics, genetics, and evolution. A?7n. Rev.
Microbiol., 3, 55.
-. 1950. Cellular transformations. Harvey Lectures, ser. 44, 145.

Sonneborn, T. M., and R. V. Dippell. 1946a. Mating reactions and conjugation
between varieties of Paraniecium aurelia in relation to conceptions of
mating type and variety. Physiol. Zool., 19, 1.

. 1946b. The significance of race 31 as a link between group A and B

varieties of Paraiuecimn aurelia. Anat. Record, 96, 19.

Swyer, G. I. M. 1951. Hvaluronidase and fertilization. Biochem. Soc. Sym-
posia (Cambridge, Eng.), No. 7, 34.

THE PHYSIOLOGY OF FERTILIZATION IN CILIATI.S 333

1 ;irtar, V., and T. Chen. Mating reactions of enucleate fragments in Para-

7)ieciwn btirsaria. Biol. RiiL, 80, L^O.
Tichomiroff, A. A. 1886. Die kiinstlichc Parthcnogcnesc bci Insekten. Arch.

Anat. Physiol, Lpz., Physiol. Aht., Suppl. p. 35.
Turner, J. P. 1930. Division and conjugation in Euplotes patella Khrenbcrg

with special reference to the nuclear phenomena. U7iiv. Calif. Puhl. Zoo/.,

33, 193.
. 1941. Fertilization in protozoa. Protozoa in Biological Research,

Columbia University Press, New York, p. 583.
Tyler, A. 1941. Artificial parthenogenesis. Biol. Revs. Cambridge Phil. Soc, 16,

291.

. 1948. Fertilization and immunity. Physiol. Revs., 28, 180.

. 1949. Properties of fertilizin and related substances of eggs and sperm

of marine animals. Aw. Naturalist, 83, 195.
van Wagtendonk, W. J. 1948. The action of enzymes on paramecin. /. Biol.

Chevi., 173, 691.
Warburg, O. 1908. Beobachtungen iiber die Oxidationsprozesse im Seeigelei.

Z. physik. Chem., 57, 1.
Wichterman, R. 1937. Division and conjugation in Nyctotherus cordifortnis

(Ehr.) Stein (Protozoa, Ciliata) with special reference to the nuclear

phenomena. /. Morphol., 60, 563.
. 1940. Cytogamy. A sexual process occurring in living joined pairs of

Paraineciian caudatwii and its relation to other sexual phenomena. /.

Morphol., 66, 423.
. 1948. The biological effects of x-rays on mating types and conjugation

of Parameciuyn btirsaria. Biol. Bull., 94, 113.
. 1951. The ecology, cultivation, structural characteristics and mating

types of Paramecium calkinsi. Proc. Penna. Acad. Set., 25, 51.
— . 1953. The Biology of Paramecium. Blakiston Co., New York.

Addendum

Material that appeared too late for the manuscript includes new informa-
tion on sexuality in Tetrahymeva from the Nanney-Elliott group, a detailed
study of the role of irradiations in the mating rhythmicity of Paravieciuni
bursaria (Ehret, 1953), and a comprehensive review of the cell surface in rela-
tion to metazoan fertilization (Runnstrom, 1952).

Elliott and Hayes (1953) found three interbreeding mating types in Tetra-
hymeria pyriforniis material collected at Woods Hole. From the Fi, Fo, F^, and
back cross progeny obtained by mating two of these (Type I, stock WH-6 x
type II, stock WH-14) Nanney and Caughey (1953) isolated a total of seven
pure mating types. These seven types form an interbreeding system; any one
of the types will conjugate with the remaining six types. These workers empha-
size that the first sign of a sexual reaction (pairing) does not occur for at least
90 minutes after mixing complementary types. Immediate agglutinative mating

334 SEX IN A4ICROORGANISA/IS

reactions are never observed when the types are mixed. These investigators
suggest that conjugation may be mediated by substances released into the me-
dium by the organisms. These papers give an account of the cytology of
conjugation in the Woods Hole stocks and in a third paper Nanney (1953)
describes the conjugation cytology, including the conjugation of three animals,
and the effects of centrifugation on nuclear development of conjugants in a
selfing stock (AA 1-8).

Unlike certain P. mirelia stocks which ordinarily mate only at night and
are inhibited by light (Sonneborn, 1938a), P. hursaria generally mates onh'
during the day. However, Ehret (1953) finds that animals maintained in con-
tinuous darkness exhibit the normal rhythm but with reduced intensity. The
reactive period may be shifted and its intensity enhanced or diminished by
appropriate exposure to light. The action spectrum for these effects suggests
a flavin type photosensitive substance. Ehret presents quantitative ultraviolet
and x-ray inactivation data for mating reactivity and describes a phenomenon,
"autoagglutination" (sticking together of animals of the same mating type),
which results from ultraviolet, x-irradiation, or high-intensity visible irradia-
tion. Since this reaction may be induced in unreactive and immature clones and
is associated with various pathological changes (vesiculation, bleb formation,
immobilization), its relation, if any, to the specific mating reaction is problem-
atical. It would not appear to be related to the x-ray induced change of
mating type reported by Lee (1949). The observation that formalin killed
P. biirsaria give mating reactions (Metz and Foley, 1949) takes some force
from Ehret's hypothesis that mating substance is continuously lost and synthe-
sized anew at the cilium surface during the reactive phase.

REFERENCES

Ehret, C. F. 1953. An analysis of the role of electromagnetic radiations in the
mating reaction of Farmnechmi bursaria. Phys. Zool., 26, 274.

Elliott, A. M., and R. E. Hayes. 1953. Mating types in Tetrahyineiia. Biol.
Bull, 105, 269.

Lee, H. 1949. Change of mating type in Faraiiieciwn bursaria following ex-
posure to x-rays. /. Exptl. Zoo/., 112, 125.

Nanney, D. L. 1953. Nucleo-cytoplasmic interaction during conjugation in
Tetrahyiiiena. Biol. Bull., 105, 133.

Nanney, D. L., and P. A. Caughey. 1953. Mating type determination in Tctra-
loyviena pyriformis. Proc. Natl. Acad. Sci. U. S., 39, 1057.

Runnstrom, J. 1952. The cell surface in relation to fertiHzation. Syvip. Soc.
Exptl. Biol, 6, 39.

Comments on the Origin
and Evolution of ''Sex'

D. H. WENRICH, Zoological Laboratory,
University of Pennsylvania, Philadelphia

In the preceding discussions in this volume the reader found that
many variations occur in the way that "sex" expresses itself; as a
result, many different interpretations of its meaning and significance
can be offered. Here some ideas are presented which indicate that
there are many as yet unsolved problems in connection with "sex."

It is difficult to give a definition of sex or of sexuality because of
the variations in the process and the differences of opinion about the
subject. In the foreword of his book on sexuality, Hartmann (Die
Sexualitcit, 1943) states that a controversy between certain botanists
and himself involved the definition of sexuality. He indicates that
one botanist defined fertilization and sex as the change from a hap-
loid to a diploid phase. Hartmann's definition was the union of two
sexually different haploid cells to form a diploid one. It would be
difficult to decide just where sexuality begins in the series of condi-
tions found by Cleveland for the flagellates in Cryptocerciis because
the change from haploidy to diploidy by one endomitotic division
{Barhulauyvipha) would fit into the definition of sex by the botanist
cited by Hartmann. Consequently, no definition of "sex" will be
attempted here.

Naturally there has been much speculation about the origin and
evolution of "sex," most of which has been based upon conditions in
higher organisms. Geddes and Thomson (1889) in their book on the
evolution of sex made the following statement (p. 117): "The num-
ber of speculations as to the nature of sex has well-nigh doubled
since Drelincourt, in the last century, brought together two hundred
and sixty-two 'groundless hypotheses,' and since Blumenbach quaintly
remarked that nothing was more certain than that Drelincourt's own
theory formed the two hundred and sixty-third. Subsequent investi-
gators have, of course, long ago added Blumenbach's ''Bildungstrieb^

335

336 SEX IN MICROORGANISMS

to the list; nor is it claimed that the generalization we have in our
turn offered has yet received 'final form,' if that phrase indeed be
ever permissible in an evolving science, except when applied to what
is altogether extinct." Like many other authors, Geddes and Thom-
son held that sex differences were primarily metabolic in character.

Just how many additional "theories" about sex have been offered
since 1889 is not known, but only three will be outlined here and
they are based largely on conditions in the Protozoa, the group with
which I am most familiar. The first may be designated the "hunger"
theory; the second may be referred to as the "chance fusion" theory;
and the third consists of the ideas suggested by Cleveland and based
on his studies of sex phenomena in the animal flagellates living in
the gut of the wood-feeding roach, Cryptocercus piinctulatus.

THE "HUNGER" THEORY

Dangeard employed a metabolic concept in discussing the pos-
sible evolution of "sex." He repeatedly expressed himself on this
subject but made a somewhat more extended statement in 1913-14.
He based his theory to a considerable extent on the life history of the
chlamydomonad flagellates; in this group rapidly ensuing fissions pro-
duce "spores" (zoospores) which, as gametes, fuse together to
produce zygotes.

Dangeard stated that, with simple bipartition, the intermediate
period of nutrition suffices to maintain nutritive equilibrium. How-
ever, the nutritive condition finds itself unbalanced at the end of a
series of sporulating divisions. The "spores" are enfeebled, famished,
incapable of development. This diminution of vital energy of ordi-
nary "spores," under the influence of insufficient nutrition, is, accord-
ing to him, the cause which has provoked, in the course of evolution,
the appearance of sexuality.

In his general conclusions, Dangeard expressed himself essentially
as follows: (1) Sexual reproduction has had for its cause a nutritive
origin, the results of successive bipartitions without intermediate
periods of nutrition. (2) The gametes are like ordinary "spores"
(zoospores) but are weak and incapable of continuing their own
development. (3) Sexual reproduction has been derived from asexual
reproduction. (4) The attraction which unites the gametes is of the
same nature as that which carries an organism toward its prey or

COMMENTS ON IHl (^RICIN AND KVOI.UTION OF SEX 337

leads to the presence of food. (5) l^'crrili/ation in its primitive sig-
nificance is then a phenomenon of "luirophagy." (6) This autophagy
has introduced into the cycle of dcxclopmcnt a new stage which
constitutes sexual reproduction. (7) Development of a gamete with-
out union with another gamete constitutes parthenogenesis. Dangeard
considered the haploid state the basic one.

THE "ACCIDENTAL FUSION" THEORY

This theory is fairly well expressed by Rhumbler (1898), who
made a study of shelled rhizopods (Testacea) . Referring to the close
similarity of the process of syngamy as described by Schaudinn
(1896) for Actiiiophrys sol with the sexual reproduction character-
istic of the Metazoa, he stated his belief that the ancestral type of
sexuality ^^'as to be found in the testate rhizopods.

Rhumbler takes as a starting point the tendency of certain rhizo-
pods to cluster together in "nests" of individuals. He says that he has
seen as many as fifty individuals of Amoeba verrucosa clustered to-
gether but without fusions. This tendency to form clusters he calls
"cytotropism."

The next step is fusion of cytoplasm, or plasmogamy, which is
very common among the Testacea and has been reported by a num-
ber of observers; and it has also been reported for other members of
the Sarcodina, as, for example, some heliozoans. Rhumbler states that
he has seen hundreds of cases of plasmogamy in Difflugia lobostoma;
also that during several years many pairs of different species of Dij-
fJiigia had been fixed and stained, but no changres in their nuclei or
other structures could be recognized; that is, there were no more
variations than occurred among single individuals. All possible com-
binations were seen — large nuclei, small nuclei, much reserve food,
little reserve food, and so on.

Rhumbler recognized that a complete series of stages of cell
fusion and reduction divisions had not been established, but believed
that they did take place. The evolutionary steps might be: (1) cyto-
tropism, or the clustering of cells without fusion, caused by the
production of chemotactic substances; (2) "accidental" fusion of
cytoplasms, or plasmogamy (eventually such fusions became neces-
sary); (3) mixing of cytoplasms might induce nuclear divisions with
one daughter nucleus degenerating, thus providing for "reduction";

338 SEX IN MICROORGANISMS

(4) finally, nuclear fusion or karyogamy (such nuclear fusion would
eventuaUy become a necessity) .

Apparently Rhumbler believed that ordinary individuals are dip-
loid as they are in Actinophrys sol. Possibly they are haploid. It
might be pointed out that cytoplasmic fusion could be followed by
nuclear fusion producing diploidy. Haploidy could be restored by a
single segregating nuclear division of the type described by Cleveland.

Since Rhumbler's paper was published, several observers have
reported complete fusion in various shelled rhizopods. In 1910 Dan-
geard denied the occurrence of karyogamy in Arc ell a, as claimed by
Hertwig and his followers (see Swarczewsky, 1908), but later (1937)
he stated that "conjugation" including karyogamy took place in
Diffltigia globnJosa. Ivanic (1935) described complete fusion includ-
ing karyogamy in Cochliopodimn digit atimi, and Penard (1938) re-
ported complete fusion for Eiiglypha sciitigera. Other reports are
mentioned in the section on Testacea in the paper "Sex in Protozoa."

CLEVELAND'S IDEAS ON THE ORIGIN OF SEX

In his discussion of the sexual cycle of the hypermastigote flagel-
late, Urinympha, Cleveland (1951) indicates some of his ideas about
the evolution of sex. In Uriiiyvipba, which is diploid, the sexual cycle
consists of meiosis in one division producing haploid nuclei and the
fusion of these nuclei in a process of autogamy. No cytoplasmic
divisions or fusions occur, although the "gametocyte" complement
of extranuclear organelles is lost as is also one centriole. New extra-
nuclear organelles are produced by the centriole which remains.

Referring to the situation in Barhulanyinpha, which is haploid,
variations in its sexual cycle include an endomitotic duplication of
chromosomes producing diploidy. This condition persists for about
10 days, then meiosis restores haploidy. Barbiilanyinpba also has two
other cycles. In one, two gametic nuclei are produced by the mitotic
division of the haploid nucleus. These nuclei fuse autogamously with-
out cytoplasmic division, then meiosis follows. The other cycle is
similar, but cytoplasmic division accompanies the gametogenic mito-
sis, producing separate gametes which can fuse in a random manner.
Nuclear fusion and meiosis follow.

As indicated l)y Cleveland, this group of sexual cycles suggests
an evolutionary series with the following steps: (I) meiosis to relieve

COMMENTS ON THE ORIGIN AND EVOLUTION Ol SI X 339

cndoniirotic diploidv; (2) iiiciosis followed by autogamy, but with-
out cell division, thus without gamete formation; (3) gametogencsis
in haploid cells followed by fertilization, then meiosis, restoring
haploidv; (4) establishment of diploidy, followed by simultaneous
gametogenesis and meiosis, then fertilization, restoring diploidy; (5)
differentiation of the "sexes" as to gonads and secondary sexual char-
acters.

Cleveland states that, before the appearance of meiosis, unicellu-
lar organisms were perhaps haploid. Diploidy could arise by endo-
mitosis in a haploid nucleus without compensating meiosis. When
asexual cells are haploid, meiosis and gametogenesis cannot occur
simultaneously; meiosis is zygotic. When asexual cells are diploid,
gametogenesis and meiosis must occur concomitantly.

In sexual cycles, as pointed out by Cleveland, even when no
gametes are produced, a centriole is lost, as occurs in higher types of
cycle, since one gamete loses its centriole somewhere during the
process of gamete formation and fusion. Hence the loss of a cen-
triole, like meiosis, antedated gametogenesis in sexual evolution.
Therefore a study of the origin of sex should begin with a considera-
tion of environmental agents responsible for meiosis, that is, those
that induce suppression of duplication of centromeres and chromo-
somes. Of secondary importance are the agencies responsible for dif-
ferentiation into male and female gametes, their union, and the for-
mation of male and female characters.

EVOLUTION OF "SEX"

W^ith these theories of the origin of syngamy before us, we may
pursue the subject of the evolution of this process further.

An examination of the status of sexuality in the Protozoa reveals
that the groups in which syngamy is most commonly found are those
which are more highly evolved. Among the Sarcodina these are the
Foraminifera, Heliozoa, and possibly the Radiolaria, with only scat-
tered examples, often unconfirmed, in the Proteomyxa, Amoebaea,
and Testacea. Among the Sporozoa syngamy seems to be general in
the Gregarinida, Coccidia, and Haemosporidia; also common or gen-
eral in the Myxosporidia, Actinomyxidia, some Aiicrosporidia and
Haplosporidia. In the Ciliophora, syngamy of a relatively primitive
type is described for the Protociliata, w^hereas in the Euciliata and

340 SEX IN MICROORGANISMS

the Suctoria it usually takes the modified form of conjugation.
Among the Mastigophora, in the Phytomonadina, which includes the
rather highly evolved Volvocidae, syngamy is extensively repre-
sented, whereas examples are very few and often questionable in the
more primitive Chrysomonadina; syngamy is apparently absent in
the Cryptomonadina, is seldom found or not well authenticated in
the Euglenoidina, and not reported for the Chloromonadina. In the
animal flagellates syngamy was not well-established until Cleveland
described the sexual cycles of the flagellates living in the woodroach,
Cryptocerciis.

Since the higher groups of Protozoa are supposed to have evolved
from animal flagellates, and the latter from plant-like flagellates, we
may well ask whether or not sexuality (syngamy) has been devel-
oped independently in the highly evolved groups mentioned above,
or was passed on from the plant-like flagellates to animal flagellates,
and from the latter to the Sarcodina, Sporozoa, and Ciliophora. Are
the different orders of plant-like flagellates sufficiently closely related
to have been derived from a common ancestor, or did they evolve
somewhat independently? In either case, what were their ancestors?
Some of the primitive fungi, at least, with primitive sexuality (see
discussion by Raper) are thought to have evolved from a "flagellate
complex" (Cantino, 1950), and now we find that there is evidence
for sexuality in the bacteria and possibly in the viruses. What were
the ancestors of the bacteria; of the viruses?

Is it profitable to suppose that "sex" arose only once in some
ancestral group and has been handed down to all descendants, or is
it more probable that it has arisen sporadically in many different
groups? In either case, can we consider any of the three previously
mentioned theories of the origin of "sex" as having exclusive validity,
or partial validity, or any validity at all?

According to Cleveland, the first step in the evolution of "sex"
could have been the production of diploidy by endomitosis, in which
chromosomes are duplicated but not the ccntrioles. This could be
followed by meiosis, which would restore haploidy. As this could
happen in a single individual, no gametes would be involved. In case
diploidy had become established by endomitosis, meiosis could be
followed by autogamy, \\'ithout the production of separate gametes.
The other two theories mentioned would have diploidy result from
cell fusions. If nuclear fusion followed, meiosis could restore hap-

COMMENTS ON THE ORIGIN AND EVOLUTION OF SEX 34 1

loidy. Apparently Cleveland is the only author to provide a detailed
description of one-division meiosis in the Protozoa. This may well be
the priniitiAe tvpc, whether diploidv arose by endomitosis or by
fusion of the nuclei of previously independent cells.

All three of the theories mentioned are based upon the study of
organisms with definite nuclei which divide by mitosis. Is this the
most primitive condition? There is genetic evidence for recombina-
tion of genes ^^■hen different races of bacteria are mixed, suggesting
syngamy and meiosis in that group (see discussion by Lederberg and
Tatum). Genetic evidence suggests syngamy in viruses (see discus-
sion by Visconti). Are we to suppose that bacteria and viruses have
nuclei with chromosomes and that not only mitosis but also syngamy
and meiosis take place in these organisms? Hutchinson and Stempen,
in this volume, have shown that cell fusion may take place in bac-
teria; and there is some evidence for mitosis in bacteria (see discus-
sion by Lederberg and Tatum). Can viruses have nuclei, chromo-
somes, mitosis, and syngamy? Recently Fraser and Williams (1953)
claimed to have identified strands of nucleic acid, or chromosomal
material from viruses when host bacteria M^ere broken up on an elec-
tron microscope screen and photographed.

If "sex" exists in bacteria and viruses, is the phenomenon a spo-
radic and isolated one, or is it more general? If these organisms rep-
resent the most primitive types of living things, and "sex" is found
to be general among them, did "sex" appear with the origin of life?
Is "sex" therefore the universal characteristic of living things that has
often been postulated? If this is so, can there be "sex" without nuclei
and chromosomes, or did the first living beings have nuclei and chro-
mosomes? If the first forms of life did not have nuclei and chromo-
somes, could they have had "sex"? In the blue-green algae (Cyano-
phyta), in which organized chromosomes apparently are absent, no
sexual phenomena have been found. Are the blue-greens to be re-
garded as degenerate, having lost chromosomes and "sex," or are they
primitive, having been derived from still simpler ancestors in which
chromosomes and "sex" did not exist? Dodson (1952) has suggested
that the blue-greens represent the ancestral type from which all the
other algae have been derived. If this is so, when and where did well-
defined nuclei, chromosomes, and "sex" come into existence? Why
is syngamy so uncommon in the Chrysomonadina, Cryptomonadina,
Euglenoidina, Amoebina, and so on? Have these groups evolved in

342 SEX IN MICROORGANISMS

such a way that "sex" has been largely lost? If so, why was it lost?
How can these organisms flourish without "sex," which Nature seems
to have found so advantageous in the evolution of higher forms of
life? Possibly "sex" is more prevalent than present evidence indicates
and more research will bring more examples to light. However, there
is ample evidence that many unicellular organisms can multiply asex-
ually for indefinite periods of time without syngamy. Frankly, we
do not have satisfactory answers to many of the questions asked
above. Future research may possibly provide the answers.

SEX DETERMINATION AND SEX DIFFERENTIATION

Since the formation of gametes and their fusion to produce new
individuals is widespread among microorganisms, what light can be
shed on how gamete determination and differentiation have arisen?
The very expression "sex determination" implies a difference between
the sexes, and much of modern study of lower and higher organisms
indicates that sex determination is genetic and therefore presumably
due to different genie composition of the chromosomes; and indeed,
in many animals, there are visible differences in the chromosomes of
the two sexes, as pointed out by McClung in 1902.

The terms male and female indicate a morphological difference
between the gametes which unite to produce a zygote and, in higher
organisms, differences in the individuals which produce the gametes
and in the organs in which they are developed. In cases of hermaphro-
ditism the sex organs of both sexes occur in the same individual. Ap-
parently most of the numerous theories of the nature of sex suggested
by the quotation from Geddes and Thomson (1889) dealt with the
development of maleness and femaleness.

In cases of "isogamy" morphological differences between uniting
gametes are not apparent, although physiological differences are
presumed to exist. For example, the attraction that brings gametes
together may be centered primarily in one gamete or may involve
both. According to Hartmann's theory of relative sexuality, both
male and female potencies are present in each gamete, and the way
in which they unite depends upon the relative amounts of these
potencies possessed by each gamete. However, with most organisms,
male and female characteristics are fairly stable and definite (see re-
view by Smith, 1951).

COMMENTS ON THK ORIGIN AND FA^OLUTION OF SEX 343

One of rhc most interesting results of Cleveland's studies was
the finding of a differentiating nuclear division in the gametocytes of
haploid flagellates such as Trichonympha and Leptospironympba dur-
ing wiiich male and female vsets of chromosomes are segregated from
each other. Each parental chromosome gives rise to a chromatid of
each sex.

The existence of sex-differentiating mitoses was postulated by
Prokofieva-Belgovskaya (1946). Starting with binucleate cells in the
bark parenchyma of potatoes, in which, after aging, two sister nuclei
come to have different appearances and behavior, then citing exam-
ples from the literature, this author presents some interesting interpre-
tations. The two sister nuclei in the potato cells, by the end of starch
formation, become considerably shrunken. The basophily of the
chromonemata and chromomeres is greatly diminished. In many cases
these sister nuclei differ from each other considerably, for example
in number of nucleoli and in stainability. The heterocyclicity of these
nuclei is also shown when they divide. The differences between the
two nuclei in their resting and divisional conditions are interpreted
to indicate that the mitosis which produced them was a differentiating
process. In anaphase, one set of chromosomes is supposed to be
"younger" and is segregated from the other, which is "older"; that
is, chromatids derived from the same parental chromosomes are not
identical. In line with some of the more recent ideas about gene and
chromosome duplications (for example, Delbriick, 1941), it is postu-
lated that a new unit forms as a "copy'' of the original instead of the
two units being formed by "splitting." Thus we have the concept
of "mother" chromosomes producing "daughter" chromosomes by a
sort of budding, and the latter may differ somewhat from the
"mother." These supposed differences in molecular makeup are
thought to account for the differences in appearance and behavior
of the sister nuclei described above.

As a matter of fact, Cleveland's description of the separation of
male from female chromatids in the gametocyte division of Tricho-
nympha and other species indicates a mechanism that results in the
separation of one group of sister chromatids from the other. A sys-
tem of joining into two groups operates to prevent random orienta-
tion of pairs of chromatids and to insure the separation of all the
"male" from the "female" chromatids. Apparently this process is not

344 SEX IN MICROORGANISMS

always perfect however, since in Trichonympha Cleveland sometimes
found intersexes or "gynandromorphs" among the "gametes."

Just how far the application of this process can be extended is
problematical. Prokofieva-Belgovskaya applied the principle to the
segregation of "male" from "female" sets of chromosomes in various
Protozoa such as Actinophrys sol (Belaf, 1923, 1926), A?noeba dip-
loidea (Hartmann-Schilling, 1917), Adelea ovata (Jollos, 1929),
Stylorhynchus lojigicollis (Leger, 1904), and Bursaria truncatella
(third division of the micronucleus, Poljansky, 1934), as well as to
various other animals and plants. Prokofieva-Belgovskaya also sug-
gested that the segregation of "mother" from "daughter" chromo-
somes was the initial step in the evolution of "sex." Is the "mother-
daughter" system in the potato cells a sex-segregating system? If not,
why does it occur? Is that system of duplication a general one? In
Cleveland's flagellates, sex-segregating mitoses take place only under
the influence of the molting hormone of the host; at other times the
ordinary type of mitosis takes place. Cleveland's cases apparently
show that there are, then, two types of mitosis, a sex-segregating
type and a non-segregating type. What the stimuli to segregation
would be in a potato cell, or in the other cases cited by Prokofieva-
Belgovskaya, is not apparent.

Although heterocyclicity may indicate sex difl"erences, and many
other instances could be cited, it does not necessarily indicate that
"mother-daughter" origin and separation of complete sets of chro-
mosomes are always required to differentiate the sexes. A difference
of one chromosome, or a part of a chromosome, or even one or a
few genes may be enough to differentiate male and female nuclei
from each other. The heterocyclicity, then, could well be the re-
sult of sex-determining influences other than the "mother-daughter"
system.

In a note appended to the paper by Prokofieva-Belgovskaya, H. J.
Muller remarks that, if such mother and daughter chromosomes are
distinguishable, it would be most surprising if all the mother chromo-
somes should go to one pole and all the daughter chromosomes to the
opposite pole of the spindle. Although such a segregation is possible,
it would involve a new principle which might be as important as the
apparent differences between the two sets of sister chromosomes. It
remained for Cleveland to describe a mechanism whereby such a
segregation actually takes place.

COMMENTS ON THE ORIGIN AND EVOLUTION OF SEX 345

Division of a haploid nucleus giving rise to one migratory "male"
and one stationary "female" pronucleus has long been known for
the third division in the gametogenesis of ciliates, but apparently the
mechanism described by Cleveland has not been recognized in such
cases^ although it might apply. However, Sonneborn (1951) has sug-
gested that, since these pronuclei are presumably genetically equiva-
lent, local cytoplasmic differences would constitute the sex-determin-
ing factor. Maupas (1889) and Diller (1936) had earher suggested
cytoplasmic determination of the differential behavior among the
nuclei derived from the original micronuclei of conjugants and ani-
mals undergoing autogamy.

In many higher animals the sexes differ in the chromosomal con-
tent of their cells, the so-called X and Y chromosomes apparently
being the differentiating feature. These differentiating chromosomes
and the evidence for random segregation of chromosomes on the
mitotic spindle, especially in meiotic divisions, would militate against
the general occurrence of the "mother-daughter" type of segregation.
Thus the evidence indicates that in most cases of chromosome dupli-
cation the products are identical, while under certain conditions the
two groups of daughter chromatids may be different. Here, surely,
is an intriguing field for further research.

As presented in this volume, "sex" is seen to have a very wide
distribution amonff microorganisms. On the other hand, in certain
groups it is very uncommon or apparently absent altogether. Its
evolution is as unclear as that of many of the groups of organisms
themselves. Continued research, both extensive and intensive, may
be expected, in time, to find answers to many of the problems that
remain to be solved.

REFERENCES

Belaf, K. 1923. Untersuchungen an Acfinophrys sol Ehrenberg. I. Die Morpho-
logic des Formwechsels. Arch. Protistenk., 46, 1-96.

. 1926. Der For-ynzvechsel der Frotiste7ikerne. G. Fischer, Jena.

Cantino, E. C. 1950. Nutrition and phylogeny in the water molds. Qvart. Rev.
Biol, 25, 269-277.

Cleveland, L. R. 1951. Hormone-induced sexual cycles of flagellates. VII. One-
division meiosis and autogamy without cell division in Urinympha. J.
MorphoL, 88, 385-440.

Dangeard, P. A. 1910. Etudes sur le developpement des organisms inferieurs.
Le Botaniste, ser. 11, 1-311.

346 SEX IN MICROORGANISMS

. 1913-14. La reproduction sexuelle envisagee dans sa nature dans son

origine et dans ses consequences. Le Botaniste, ser. 13, 285-325.
-. 1937. Memoire sur le Difflugia globiilosa Dujardin. Le Botaniste, ser.

28, 229-274.
Delbriick, M. 1941. A theory of autocatalytic synthesis of polypeptides and

its application to the problem of chromosome reproduction. Cold Spring

Harbor Symposia Quant. Biol., 9, 122-126.
Diller, W. F. 1936. Nuclear reorganization processes in Paraineciinn aurelia,

with a description of autogamy and "hemixis." /. Morphol., 59, 11-67.
Dodson, E. O. 1952. A Textbook of Evolution. W. B. Saunders Co., Phila-
delphia.
Eraser, D., and R. C. Williams. 1953. Electron microscopy of the nucleic acid

released from individual bacteriophage particles. Proc. Natl. Acad. Sci.

U. S., 39, 750-753.
Geddes, P., and J. A. Thomson. 1889. The Evolution of Sex. Walter Scott,

Ltd., London.
Hartmann, M. 1943. Die SexiialitHt. G. Fischer, Jena.
Hartmann, M., and C. Schilling. 1917. Die pathogenen Protozoen. Springer,

Berlin.
Ivanic, M. 1935. Uber die "vollkommene Kopulation" bei Cochliopodiwu

digitatwn Penard, nebst Bemerkungen iiber die Kopulation bei Susswasser-

thalamorphoren im allgemeinen. Biol. ^Lentr., 55, 225-245.
Jollos, V. 1929. Handlnich path. Mikroorg., 8, 279-310.
Leger, L. 1904. La reproduction sexuee chez les Stylorhynchns. Arch. Protis-

tenk., 3, 303-357.
McClung, C. E. 1902. The accessory chromosome: sex-determinant? Biol.

Bull, 3, 43-84.
Maupas, E. 1889. Le rajeunissement karyogamique chez les cilies. Arch. zool.

expt. et gen., ser. 2, 7, 149-517.
Penard, E. 1938. Ees infiniment Petit s dans leiir Manifestations vitales. Georg

& Cie, Geneve.
Poljansky, G. 1934. Geschlechtsprozesse bei Bursaria trnncatella O. F. Miill.

Arch. Protistenk., 81, 420-456.
Prokofieva-Belgovskaya, A. A. 1946. "Mother" and "daughter" chromosomes.

/. Heredity, 37, 239-246.
Rhumbler, L. 1898. Zelleib-, Schalen- und Kern-verschmelzungen bei den

Rhizopoden, etc. Biol. Zentr., 18, 21-26, 33-38, 69-86, 113-130.
Schaudinn, F. 1896. Uber die Kopulation von Actinophrys sol Ehrbg. Sitzber.

deut. Akad. Wiss. Berlin, I, 83-89.
Smith, G. M. 1951. "Sexuality of algae," in Manual of Phycology. Chronica

Botanica Co., Waltham, Mass.
Sonneborn, T. M. 1951. Some current problems of genetics in the light of

investigations on Chlamydoinonas and Paraineciinn. Cold Spring Harbor

Symposia Quant. Biol, 16, 483-503.
Swarczewsky, B. 1908. Uber die Fortpflanzungserscheinigungen, bei Arcella

vulgaris Ehrbg. Arch. Protistenk., 12, 173-212.

Author Index

Adclbcrg, E. A., 13
Adie, H. A., 147
AlexeiefF, A., 148, 149, 191
Ames, L. M., 59
Anderson, T. F., 309-310
Andes, J. O., 62
Aratari, A., 169
Aristotle, 104
Austrian, R., 15, 24
Awerinzew, S., 175, 231

Backus, M. P., 58, 68, 69

Bailev, W. T., 3

Balbiani, G., 187

Baldrey, F. S. H., 144

Banerjee, B., 125

Barron, E. S., 311

Barv, A. de, 68

Bastin, A., 190

Bauch, R., 67

Beadle, G. W., 12, 126

Beale, G. H., 317

Becker, E. R., 204

Behlau, J., Ill, 119-121, 123, 126

Belaf, K., 134, 143, 147, 174, 179,

344
Benjamin, R. K., 56
Berg, W. E., 325
Bergon, P., 83
Berliner, E., 137
Bernheimer, A. W., 317
Berthold, G., 117
Bessev, E. A., 46
Biecheler, B., 137, 140, 186
Bielig, H. J., 284
Bishop, H., 57, 69
Bisset, K. A., 29, 38
Blakeslee, A. F., 49, 50, 68
Blochmann, F., 175
Blumer, S., 48
Boell, E. J., 286
Bogdanowicz, A., 235
Bold, H. C, 107, 108, 111, 119, 121
Borgert, A., 185, 186

180,

125

Bott, K., 172

Bowling, R., 190, 191, 243, 244

Brandt, K., 182, 183, 186

Braun, A. C, 25, 37

Breinl, A., 144

Brodie, H. J., 47

Browning, C. H., 14

Brunswik, H., 61

Bruvn, H. L. G. de, 57

Bulier, A. H. R., 58, 59, 61, 67, 68

Bunting, M., 149-151

Burgeff, H., 68

Biitschli, O., 134, 172, 173, 191, 213

Buttcrfield, W., 310, 312, 313

Cadoret, R., 108

Calkins, G. N., 82, 134, 190, 191. 210.

227, 228, 242-244
Cantino, E. C, 56, 340
Cash, J., 174, 175
Cassel, W. A., 30
Catcheside, D. G., 60
Caughev, P. A., 276, 280, 281
Cavalli,'L. L., 19, 21, 23
Cavara, F., 119
Chagas, C, 144
Chakravarty, M., 190
Chambers, R., 326
Chang, M. C, 284
Chao, P. K., 275, 279
Chargaff, E., 316
Chase, M., 5, 6
Chatton, E., 140, 143, 186
Chejfec, M., 223
Chen, T., 209, 214, 215, 217, 222, 292,

296, 300, 305-307, 319, 320
Chen, Y. T., 292, 295
Chodat, F., 105, 116
Christensen, J. J., 48
Cienkowski, L., 139, 183
Clark, J. B., 17
Claussen, P., 56, 68
Cleland, K. W., 286

347

348

AUTHOR INDEX

Cleveland, L. R., 151, 153, 155, 157,
159, 161, 163, 165, 167, 169, 191, 245,

335, 336, 338-340, 342, 344
Coker, W. C, 53, 68

Collin, B., 143, 214, 233, 234, 241, 242

Colwin, L. H., 231

Cook, A. H., 116

Corliss, J. O., 229

Couch, J. N., 48, 49, 56-58, 67

Craigie, J. N., 58, 68

Cull, S.W., 210, 242

Culwick, A. T., 145, 146

Culwick, R. E., 146

Cunningham, J. T., 188, 193

Czurda, V., 104, 107

DaCunha, A. M., 227

Dallinger, W. H., 147, 149

Dan, J. C, 326

Dangeard, P. A., 149, 169, 173, 174,

336, 338
Danielli, J. F., 311
Darlington, C. D., 100
Davis, B. D., 15, 16
Debaiseaux, P., 201
DeGaris, C. F., 222, 296
Dehorne, A., 210

DeLamater, E. D., 19, 21, 25, 29, 37

Dclbriick, M., 1-3, 8, 343

Dennis, E. W., 206

Deschiens, R., 171

Devide, Z., 240

Dienes, L., 25, 35, 36

Diller, W. F., 208, 210, 213, 214, 217,

218-220, 223-225, 228, 240, 242, 244,

289, 290, 305-307, 345
Dippell, R. v., 267, 270, 292, 294
Distaso, A., 180
Diwald, K., 141, 142
Dobell, C. C, 137, 138, 195, 196, 204
Dodge, B. O., 42, 58, 59, 68
Dodge, C. W., 42
Dodson, E. O., 341
Doermann, A. H., 5, 7, 8, 10
Doflein, F., 134, 144, 213, 219
Dogiel, v., 188, 193, 232, 233, 242
Donovan, J., 112, 327
Dowding, E. S., 59
Downs, L. W., 321, 325
Drayton, F. D., 56, 58

Drvsdale, J., 147, 149
Duboscq, O., 143, 188, 193
Dulbecco, R., 2
Dunn, L. C, 100

Edgerton, C. W., 62
Edmondson, C. H., 174
Edsall, J. T., 316, 317
Ehrenberg, C. G., 229
Ehret, C. F., 334
Elkeles, G., 144
Elliott, A. M., 217, 321
Ellis, J. M., 235
Elmassian, M., 172
Elpatiewskv, W., 173
Elrod, R. P., 25, 37
Elvidge, J. A., 116
Emerson, R., 49, 53, 56, 68
Enderlein, G., 30
Enriques, P., 183, 213, 235
Entz, G., 142
Ephrussi-Taylor, H., 24
Erdmann, R., 222, 223

Fairbairn, H., 145, 146

Faure-Fremiet, E., 228

Fiennes, R. N., 145

Finley, H. E., 208, 229, 327

Flu, P. C, 147

Fol, H., 326

Foley, M. T., 290, 301, 303, 304, 307,

319, 321, 327
Foyn, B., 177

Fraenkel-Conrat, H., 311, 317
Fraser, D., 341
French, D., 317
Fried, P. J., 19
Friedman, S., 29
Fries, N., 61

Fritsch, F. E., 83, 91, 101, 103, 135
Fusco, E. M., 300, 308, 312, 315-318

Gabriel, B., 173, 174

Gaumann, E., 42

Ganapati, P. N., 207

Geddes, P., 335, 336, 342

Gee, F. L., 145

Geitler, L., 82-87, 89, 90-93, 97, 98, 102,

103, 114, 117, 119, 135, 240
Gelei, J. von, 222, 223

AUTHOR INDEX

349

Gcorgcvitch, J., 200, 206

Gerlolf, |., 106, 117, 119, 120-122, 124,

125
Gibbs, A. J., 196, 204
Giese, A., '235, 292
Gilman, L. C, 292, 295
Glaser, H., 171
Glaser, O., 325
Gohre, E., 203
Glick, D., 311
Goette, A., 174
Goldschmidt, R., 143
Goodrich, H. B., 326
Goodrich, H. P., 188
Goroschankin, J., 119
Granata, L., 199, 202
Grasse, P.-P., 134
Grav, C. H., 13
Gre'ef, R., 172
Gregory, L. H., 238
Greiner, J., 195
Grell, K. G., 190, 191
Gross, F., 139
Groves, J. W., 56
Gruber, A., 210
Gruchy, D. F., 321
Giinther, F., 137
Guillermond, A., 48, 49, 67

Haase, G., 137

Haeckel, E., 183

Haecker, V., 186

Haldane, J. B. S., 126

Hall, C. E., 308

Hall, R. P., 134, 142

Hamburger, C, 214, 215

Hanna, W. F., 61

Hansen, H. N., 59

Harder, R., 49, 56, 61

Harrison, J. A., 317

Harte, C, 127

Hartmann, M., 42, 104, 107, 116, 126,

171, 335
Hauschka, T. S., 193, 195, 203
Hausniann, L. A., 172
Hayes, W., 17, 38
Heidenreich, E., 227
Heilbrunn, H. V., 308
Henrici, A. T., 30, 36
Hentschel, C. C, 207

Herriott, R. M., 311
Hershcy, A. D., 3-7
Hcrtwig, R., 173, 180, 181, 183, 219,

223, 288, 305, 306
Hcslot, H, 21
Hiil, M. B., 7
Hindlc, E., 144
Hinreiner, E., 115
Hirsch, H. E., 59
Hiwatashi, K., 292, 295, 306, 310, 312,

314-317
Hoare, C. A., 146
Horstadius, S., 326
Hofker, I., 139
Hopkins, D. L., 172
Hopkinson, J., 174, 175
Horvath, J., 227
Hulpieu, H. R., 172
Humphrey, J. E., 53
Hunter, M. E., 29
Hutchinson, W. G., 25, 29, 32, 35, 36,

341
Huth, W., 182, 185, 186
Hutner, S. H., 106, 109, 115, 117, 327
Hyatt, M. T., 56

Ilowaisky, S. A., 227

Ishikawa, C, 139

Ivanic, M., 143, 227, 228, 235, 338

Iyengar, O. P., 84

Jacobson, W. E., 305, 320

Jakus, M. A., 308

Jameson, A. P., 190, 191

Janasson, L., 61

Jennings, H. S., 119, 126, 134, 217, 235,

280, 288, 292, 296, 297, 307, 321
Jickeli, F., 174
Johnson, P. L., 172
Jollos, v., 115,344
Jones, P. M., 172
Joseph, G., 139
Joyet-Lavergne, P., 203

Karling, J. S., 68
Karsten, G., 83, 91
Katashima, R., 322, 324
Kater, J. M., 124
Kay, M. W., 227, 238
Kent, S., 134

350

AUTHOR INDEX

Kidder, G. W., 228, 235, 242
Kimball, R. F., 269-211, 280, 284, 288,

290, 321-325
Klebahn, H., 90
Klebs, G., 68, 101, 107, 110-112, 119,

120, 122-124
Klee, E. E., 227, 228
Klieneberger-Nobel, E., 16, 25, 32
Klitzke, M., 213
Klyver, F. D., 120, 122, 123
Knaysi, G., 29

Kniep, H., 42, 48, 49, 61, 64, 67, 68
Knight, B. C. J. G., 12
Konsuloff, S., 208, 209
Korotneff, A., 172
Korschikoff, A. A., 101-103, 119, 120,

122, 123
Koster, W., 221
Krafczyk, H., 68
Krauss, M., 325
Krichenbauer, H., 137, 138
Krieger, W., 135
Kropp, B., 171
Kudo, R., 134, 199, 201
Kuhn, A., 140, 171, 179, 182, 185
Kurssanow, L. J., 123

Labbe, A., 187

Lackey, J. B., 135, 136

Lamanna, C, 29

Landis, E. M., 221

Lang, A., 116

Lapage, G., 147, 148

Laustsen, O., 60, 67

Lebedeff, W., 144

Le Calvez, J., 179

Lederberg, E. M., 13, 23, 24, 38

Lederberg, J., 12-21, 23-25, 29, 38, 46,

341
Lee, H., 334
Leger, L., 188, 193, 344
Leidy, J., 173, 174
Leonian, L. H., 57
Lerche, W., 101, 104, 108, 111, 115,

119, 124-126
Levinthal, C, 5
Lewin, R. A., 100, 103, 106, 109-112,

117, 119, 121-123, 125-127
Liebisch, W., 85
Lillie, F. R., 287, 326

Lindegren, C. C, 31, 58, 60
Lindegren, G., 60
Link, G. K. K., 42
Lister, J. J., 177
Loeb, j., 112, 286
Lohnis, F., 30, 36, 37
Lucas, C. L. T., 172
Luria, S. E., 1, 2
Luvet, B. J., 101, 134
Lwoff, A., 12

MacDougall, M. S., 235, 242

Mackinnon, D. L., 190, 193, 195

Mack, B., 98, 135, 136

Maher, M. L., 108, 111, 112, 121, 125

Mainx, F., 104, 137

Martin, C. H., 147, 174

Mather, K., 62

Maupas, E., 210, 213, 214, 217, 219, 229,

231, 235, 242, 306, 321, 345
Maurizio, A., 53
McClung, C. E., 342
McCranie, J., 49
McElroy, W. D., 29
McGahen, J. W., 62, 63
Meglitsch, P. A., 172
Mellon, R. R., 31
Mercier, L., 172
Metcalf, M. A4., 208
Metz, C. B., 112, 116, 267, 284, 288,

290, 294, 300, 301, 303-308, 310, 312-

319, 321, 327
Minchin, E. A., 134, 191
Mitra, A. K., 119-122
Mivashita, Y., 231, 235
Moewus, F., 56, 101, 102, 104-117, 119-

126
Molliard, M., 68
Monne, L., 316
Moore, E. L., 227, 228
Moore, J. E. S., 144
Morris, S., 172
A1 ounce, L, 61
MUhl, D., 203

iMiiller, W., 210, 213, 221, 222
MuUer, H. J., 344
Mulsow, K., 190
Muniz, J., 144, 227
Murneek, A. E., 116

AUTHOR INDEX

351

Mvcrs, K. H., 176, 177
Alycrs, J. W., 13

Nabih, A., 195

Njiglcr, K., 171

Nanncv, 1). L., 217, 266, 274-276, 278,

280, 281, 307, 319, 321
Naville, A., 187, 190, 191, 193, 195,200,

206
Nayal, A. A., 101, 119, 124
Nelson, E. C, 235
Nelson, T. C, 16
Nereshcimer, E., 208
Newcombe, H. B., 19
Nicholas, P. A., 229
Nigrelli, R. F., 228, 231
Nipkow, P., 98

Noble, E. R., 190, 197-200, 206
Nobles, M. K., 47
Noland, L. E., 238
Nowlin, N., 207
Nvholm, M. H., 19

Olcott, H. S., 311, 317

Ondratschek, K., 101, 104, 111, 112,

115, 117, 119, 124
Opitz, P., 292, 296, 297

Padnos, M., 228, 231

Patau, I., 127

Paehler, F., 190

Papazian, H. P., 62

Pascher, A., 101-103, 105-107, 116, 119,

122, 126
Pateff, P., 174
Patrick, Ruth, 82, 97
Patton, W. S., 147
Pearl, R., 235

Penard, E., 169, 171, 174, 175, 180, 338
Penn, A. B. K., 210
Perkins, D. D., 12, 14
Philip, U., 126
Phillips, N. E., 190
Pincus, G., 284
Plate, L., 232, 233
Poljansky, G., 344
Pontecorvo, G., 46, 53
Popoff, M., 171
Pothoff, H., 125
Powers, E. L., 321, 323

Pratjc, A., 139, 140

Prcer, J. R., 275, 305, 306, 320

Prcnaiit, M., 208, 209

Pringshcini, E. G., 100, 101, 104, 111,

115, 117, 119, 124
Frokoricva-Bclgovskava, A. A., 343,

344
Provasoli, L., 106, 109, 113, 115, 117,

327
Prowazek, S. von, 144, 147, 149
Prvtherch, H. F., 189, 205
Puck, T. T., 112
Pulvertaft, R. J. V., 32

Ramanathan, K. R., 103, 120, 122
Raper, J. R., 42, 53, 57, 62, 68-70, 116,

340
Ray, H., 190, 193, 195
Regnery, D. C, 126
Reichenow, E., 107, 119, 134, 195, 219
Renner, O., 60
Reukauf, E., 175
Rhunibler, L., 174, 175, 337, 338
Robinow, C. F., 32, 35
Rodenhiser, H. A., 48
Roepke, R. R., 13
Roper, J. A., 19, 46, 53
Rosenberg, L. E., 231
Roskin, G., 145
Rostand, J., 284
Rothfels, K. H., 17, 19
Rothschild, L., 284, 286
Rotman, R., 4, 7
Runnstrom, J., 284, 326
Russo, A., 235

Salvin, S. B., 53

Sass, J. E., 59

Satina, S., 49

Schaudinn, F., 144, 149, 172, 174, 175,

177, 180, 337
Schellack, C, 190
Schemakhanowa, N. M., 123
Schepatief, A., 172
Schewiakoff, W., 182, 187
Schiller, J., 135
Schilling, C, 144, 344
Schirch, P., 172 ']

Schischliaiewa, S., 145
Schnitzler, H., 190

352

AUTHOR INDEX

Schopfer, W. H., 48

Schreiber, E., 105, 111, 126

Schulze, B., 101, 107, 111, 116, 126

Schiitt, F., 83

Schwarz, E., 135, 136

Scott, M. J., 235

Seshachar, B. R., 228

Shanor, L., 56

Shear, C. L., 58, 68

Shortt, H. E., 207

Siebenthal, R. de, 116

Sinnott, E. W., 100

Skuja, H., 103, 119, 122, 135

Slautterback, D. B., 316

Smith, G. M., 100, 101, 103, 105-108,

110, 113, 115, 116, 121, 123, 125, 126,

135, 342
Smith, N. R., 30
Smith, W.E., 31, 35, 36
Snyder, W. C, 59
Sorgel, G., 49, 56, 67
Sonneborn, T. M., 116, 126, 213, 219,

225, 226, 244, 266, 267-280, 284, 288-

290, 292, 294, 296, 300, 304, 306, 308,

317, 319, 321, 322, 345
Sparrow, F. K., 56
Spencer, H., 223
Sprague, V., 190
Stapp, C, 37
Starr, R. C, 123, 125
Stebbins, G. J., Jr., 96, 97
Stein, F., 139

Stempen, H., 25, 29, 32, 33, 35, 36, 341
Stoughton, R. H., 3 1
Stranghoner, E., 223
Strehlow, K., 106, 111, 120, 122
Studitsky, A. N., 235
Stiiben, H., 56

Subrahmanyan, R., 84, 85, 90, 92, 121
Summers, F. M., 82, 134, 242
Swaminath, C. S., 207
Swarczewsky, B., 173, 202, 233, 338
Swyer, G. I. M., 325

Tartar, V., 305, 307

Tate, P., 207

Tatum, E. L., 12-15, 18, 23, 46, 341

Teodoresco, E. C, 120

Thaxter, R., 56

Thimann, K. V., 116

Thomson, J. A., 335, 336, 342

Tichomiroff, A. A., 286

Tittler, I. A., 228

Tregouboff, G., 190

Tulasne, R., 33, 36

Turner, J. P., 208, 236-238, 322, 324

Tyler, A., 284, 288, 326

Valkanov, A., 190, 208
Vandendries, R., 61
Vanderplank, F. L., 144, 145
Vasseur, E., 112
Veley, L., 172
Verworn, M., 174
Visconti, N., 1, 5, 8, 341
Visscher, J. P., 235
von Medem, F., 284
von Stosch, H. A., 84

Waddington, C. H., 100

Wager, H., 68

van Wagtendonk, W. J., 321

Wailes, G. H., 175

Wang, C. C, 221

Warburg, O., 286

Wedekind, G., 195

Weiser, J., 201

Weisz, Paul B., 217

Wenrich, D. H., 134, 219, 221, 335

Wenyon, C. M., 134, 144, 147-149, 207

Weschenfelder, R., 190

Wheeler, H. E., 62, 63

Whitehouse, H. L. K., 50, 57, 60, 61

Wichterman, R., 210, 214, 215, 221,

235, 236, 238-240, 288, 290, 292, 296,

308, 310, 312, 327
Wilber, C. G, 172
Wilcox, M. S., 58
Williams, R. C, 341
Willis, A. G., 228
Wilson, C. M., 49, 67
Winge, O., 49, 60, 67
Winter, F. W., 177
Wislouch, S. M., 120, 123
Witkin, E. M., 21
Wolcott, G. B., 145
Woodcock, H. M., 137, 147, 148, 149
Woodruff, L, L., 221-223, 226, 286,

292

AUTHOR INDEX 353

Yauiu, R., 235 Zcller, E., 208, 209

Varwood, E. A., 195 Zeulzer, M., 174, 181

Zickler, H., 69
Zederbaucr, E., 140, 141 Zimmermann, W., 124

Zelle, M. R., 20 Zinder, N. D., 25

Subject Index

"Accidental fusion" theory of sex, 337
Acanthostaunis pnrpurascens, 182
Acephalina, 190
Acetobacter i/ielaTiogemi?//, 13
Achlyci, 53, 68-71, 74

ainbisex'ualis, 57, 65
Achnanthes brevipes, 94

lanceolata, 85, 93, 94

lojjgiceps, 94

longipes, 85

i/mmtissi7/ia, 93

subsessilis, 91, 95
Acnidosporidia, 202
Actbiocephalus parvus, 190
Actinomyxidia, 200
Actmophrys sol, 179-181, 246, 337,

338, 344
ActhwspaeTiui)! eichhoriii, 180, 181,

246
Activation in Paraiiieciinti, 298, 302
Ad e lea ovata, 195, 344
Adeleidae, 195
Adelina cryptocerci, 195

derofiis, 193, 195, 203, 204
Aggregata eberthi, 195-197, 204
Aggregatidae, 195
Agrobacteriu?)! tinuefaciens, 37
Algae, chemotaxis in, 116

clumping in, 117

cytogamy in, 121

diploid zygospore in, 122

effect of light on, 107

effect of medium on. 111

effects of period of illumination on,
110

effects of quality of light on, 108

"genetyllin" in, 112

heterothallism in, 103

homothallism in, 103

hybridization of, 106

induction of sexual activity in, 107

isolation of sexual strains of, 105

mating in, 103, 116

new haplophase in, 125

outbreeding mechanisms in, 102

pairing in, 1 19

planozvgotes in, 120

relative sexuality in, 105

sex substances in, 114

sexual activity in, 107

sexual differentiation in, 101

sexual dimorphism in, 102

vegetative cells and gametes in, 101

vis-a-vis pairs in, 120

zygospore germination in, 124
Alloviyces, 48, 49, 53, 68, 71

arbiiscitla, 65

cystogeinis, 49
Alphavwvas, 148, 149
Amoeba alba, 171

diploidea, 171, 344

f rose hi, 171

iiibmta, 171

'inira, 171

proteus, 171, 172

verriicosa, 337
Amoebina, 171
Aiiiphidinmm laciistre, 139
Aviphipleura pelhicida, 86, 87, 91, 93

rutilans, 93
A'lnphiprora alata, 93
ATiiphora coffaeforinis, 93

cyinbelloides, 93

?iorwa?iii, 90, 95

ovalis, 93

oval is var. pe die id us, 93

pusio, 93

veneta, 93
Anisogamous fusion in diatoms, 90
AnoDioeoiieis exilis, 85, 92

sc'ulpta, 93

seriavs, 93
Aiwplophrya branehiarwti, 241, 242

orehestiae, 242
Ambophysa vegetans, 149
Apolocystis elongata, 190
Arcella, 174

vulgaris, 173

354

SUBJECT INDEX

355

Arciiicola cccudiata, 207
Ascoboliis, 6S, 71
Asexual cycle in fungi, 45
Asfyciiiillus, 75
Assiiliiia, 175

Audouinia tentaculata, 207
Aidacantha scolyvhwtha, 185, 186
Aurelia type mating systems in ¥ara-

mcciinii, 292
Auricula hyaliiia, 93
Autogann- in raraincchnii, 223-225,

289
Automixis in diatoms, 90
Auxospore in diatoms, 82
formation in diatoms, 82, 91

t\'pes of, 93
Auxotrophic mutants in bacteria, 13,

14
Azotohacter, 30

Babesia, 197, 206
Bacillaria paradoxa, 95
Bad lilts ine gather iuui, 37, 38
Bacteria, auxotrophic mutants in, 13,
14

conjugation in, 36, 37

genetic studies in, 12

"large body" among, 33, 35

L-bodies among, 32

sex in, 12, 29

evidence from morphology, 29
Bacterial viruses, 1

genetic recombination in, 1
Bacteriophage, life cycle of, 4
Bacteriujii Tjialvacearzmi, 31
Bacteroides jundulijonnis, 31, 32, 36
Balantidimn caviae, 235

simile, 111

sp., 227
Barhulaiiyvipba, 335, 338
Barrouxia sp., 195
Biddulpbia iiwbilievsis, 83
Blastoclodiella, 56, 67, 71
Blepharisvia undiilatis, 111, 235
Bodo lacertae, 147, 149
Botrydiinn gramilatinn, 104
Brachiomovas, 101, 111, 122, 126
Brebissoiiia boeckii, 93
Breeding pattern in Para-meciwii cal-
kivsi, 296

Breeding sxstems in I'araiiicciuui aure-

lia,^29\, 293
Bursaria truncatella, 344

C.alyptrospbacra spbaeroidea, 135
Carteria, 119, 120

Iyengar a, 103, 120, 122

ovata, 120
Car\()gam\' in fungi, 44
Castai/idiu/// variable, 186
Cell fusion in Proteus vulgaris OX- 19,

32
Cellular adhesion in Varanieciuiii, 290
Centrales (diatoms), 83
Cevtropyxis aculeata, 175
Cepedea diniidiata, 240
Ceratiuin biruiidinella, 140, 141
CeratoTiiyxa blemiius, 197, 198, 200
Cercoinovas lovgicauda, 147, 148
Cepbalina, 190
Cbaetoceras boreale, 95

cochlea, 83

defismn, 95
Cbagasella, 196, 204, 205
Chaos chaos, 172

diffltievs, 171
Chemistry of mating t\'pe substances

of Parauieciu'iii, 310
Chemotaxis in algae, 116
Chilodonella ciicidhts, 235

dentata, 240

labiata, 242

uncinatus, 214, 228, 235, 240
Chlaviydobotrys, 111, 122

gracilis, 120

stellata, 120
Cblaviydovwvas, 56, 101, 103, 105, 106,
108, 109, 111, 113, 115, 116, 119,
121-123, 125, 126, 143, 327

botryoides, 106

brawiii, 119, 122

cblaviydogavia, 108, 119, 122, 123,
125

dysosiiios, 1 17

elongata, 102, 122

eugametos, 102, 104, 105-108, 110,
112, 113, 115, 117, 119-124, 126

gyiiinogyne, 101, 102, 122

media, 107, 1 10, 122-124

inimitissniia, 110

356

SUBJECT INDEX

nwewusii, 103, 106, 109-113, 115,
117-127

monoica, 123

paradoxa, 106, 120

paupera, 102, 106, 116, 122

praecox, 102, 119

proboscigera, 119

pseudoparadoxa, 111

reinhardi, 108, 110, 126, 127

sphagna phila, 105

St ell at a, 119

upsaliensis, 119, 122

variabilis, 119-121, 126
ChlaDiydoiuyxa viontana, 169
Cblamydophrys, 173

'//!inor, 174

schaudinni, 174

stercorea, 174
Chlorobotrys gracilliuia, 120
Chlorochytrium le?nma, 123
Chlorococcwn, 119, 123, 125
Chlorogoniuvi, 101, HI, 116, 126

ooganmm, 103, 122
Chloromonadina, 139
Chloro7/ionas paradoxa, 120
Chonotrica, 233
Chromosome aggregates, 240
Chrysococcus heviisphaerica, 135

spiralis, 135, 136
Chrysolykos planktoniciis, 135, 136
CbrysoT)ionadina, 135
Ciliates, conjugation in, 234

differentiation of gametes in, 242

fertilization in, 284, 321

mating substances in, 284

nuclear reorganization in, 222, 225-
227
Ciliophora, 208
Ciuiex lecticjilaris, 147
Cis- and fr^77i--diniethvl crocetin, 117
Clostridijn/i perfrivgens, 30
Clumping in algae, 117
CM = "can't mate" stocks of P. au-

relia, 301
Cnidosporidia, 197, 206
Coccidia, 193

meiosis in, 195
Coccidivimn i/iesnili, 140, 143, 186
Coccofieis pedicidus, 94, 95

placentula, 89, 91, 94, 95, 98

Cochliopodiwn digitatwn, 174, 338
Collosphaera, 183
Collozoiwi, 183

iidvinn, 182, 183

inervie, 182, 183

pelagicum, 183

pidvidimi, 183
Colpidiitvi compylwii, 240
Covcbophthirius uiytili, 235
Conjugants, differentiation of, 234
Conjugate division in fungi, 44
Conjugation, comments on, 234

evolution of, 243

in bacteria, 36

in Farainechnn, 210, 211, 288
Conjugation-inducing action of ani-
mal-free culture fluids in Ejiplotes
patella, 322
Conjugation tubes in bacteria, 37
Coprhms, ll, 1^
Coprouw7ias major, 137

rinuinant'nnn, 137

sidnilis, 137, 138, 151
Copulation in fungi, 60, 65
Crocetin, cis- and /rnf//.f-dimethvl,

117
Crvptobiidae, 147
Cryptocercus pwnctidatiis, 134, 151,

168, 169, 206, 245, 335, 336, 340
Cryptochilmn echini, 235
Cryptodifflngia ovifor/nis, 174
Crvptomonadina, 137
Cryptoperidiniin//, 143
Cycloposthhnn bipalniatimi, 232,

242
Cy clot el la, 84, 89, 94

meneghiniana, 84, 95
Cylindrocystis, 111, 124
Cyi/mtopleura elliptica, 95

solea, 95
Cymbella affinis, 93

caespitosa var. pedicidus, 93

cisfula, 93, 95

cyvibijoniiis, 93

gastroides, 93

helvetica, 93

lacnstris, 93

lanceolata, 93

parva, 93

prostrata, 93

SUBJECT INDl'X

357

siiv/inrciisis, 93, 95

ventricosa, 87,91, 93, 95
Cypboderia, 175
Cystohia chiridotae, 188, 193
Cytoganiv, in algae, 121

in PariVJicciuni catidatwii, 214
C\'tosfcnctic considerations in /'. ait-

relia, 267
C)'toIogy of Escherichia coli, 2 1

Dallasia, 244

Dallasia, fromata, 243, 244, 247

Dasycladus, 115

Denticula tenuis, 97

vaiiheiirckii, 93

anisogamous fusion in, 90

automixis in, 90
Diatoms, auxospore formation in, 82

fusion of gametes in, 89, 93

gametogenesis in, 86

isoganious fusion in, 90

sexual reproduction in, 82

types in, 93-95

zygote development in, 91
Dicarvon in fungi, 44
Dicaryotic cycle in fungi, 48
Dictyuchus, 71

vwnosporus, SI
Diffliigia globiilosa, 338

lobostomci, 173, 174, 337

jirceolatii, 174
Dileptus gigas, 235
Dinobryon borgei, 135

cylindriciivi, 135

diver ge7is, 135

sertidaria, 135
Dinoflagellata, 139
Diploid cycle in fungi, 49
Diploid zygospore in algae, 122
Discorbijia vilardebomia, 111, 179
Discorbis mediterranensis, 179

orbicidaris, 179

patellijormis, 176, 177
Diiboscqtiella miwmicola, 143
Dimaliella, 101, 108, 111, 115, 120, 125,
126

EchinoDiera hispida, 190
Ectocarpus, 117
Eimeria, 204

Einicriidac, 195
Endaiiioeha bhittac, 172
Evdoliiiiax blattae, \11
"Endomixis," 222
Entamoeba coli, 172

thoinsoni, 172
Fjitorrhipidiinn echini, 235
Entosoleniis inarginata, 1 79
Epistylis anastatic a, 228
Epistylis plicatilis, 228
Epithemia argns, 93

sorex, 93

tiirgida, 93

zebra, 93
Erentascus, 71

Escherichia coli, 13, 14, 16, 17, 19-25,
32, 38

coliiinitabile, 32
Euciliata, 209

Eticovw7iyi}ipha, 159, 168, 169
Endorina, 126
Euglena sanguinea, 137
Euglenoidina, 137
Euglypha alveolata, 173, 175

scutigera, 171, 175, 338
Eugregarines, meiosis in, 190
Eunotia arcits, 85-87, 90, 95

flexuosa, 86, 87, 90, 95

forjuica, 95

pectinalis, 95
Euplotes, 244, 321-325, 327

charoji. 111, 240

harpa, 321

longipes, 111, 228

pa^f/Za, 227, 236, 237, 321, 322

Fabrea saliiia, 235

Fertilization, in ciliates other than
FarauieciuHi, 321

in Faraviecimn, 288

1/ietazoafi, 325
Foraminifera, 175
Fragilaria crotonensis, 98
Frustulia rhoiiiboides var. saxonica, 85,

86, 90, 92, 93
Fucus vesiculosus, 103
Fungi, asexual cycle in, 45

caryogamy in, 44

conjugate division in, 44

dicaryon in, 44

358

SUBJECT INDEX

dicaryotic cycle in, 48
diploid cycle in, 49
gametangial copulation in, 65
haploid cycle in, 46-48
hermaphroditism, 58
heterothallism in, 50, 54, 62
homothallism in, 51, 52, 59
incompatibility factors, 60, 61
life cycles in, 42, 43, 45
multiple alternate factors in, 57
plasmogamy in, 44
sexuality in, 42, 49, 55-57, 59-61, 64
somatic copulation in, 60, 65

Gametes, differentiation of in ciliates,
242

fusion in diatoms, 89
Gametogenesis in diatoms, 86
Gasteromycetes, rusts, 71
"Genetvlhn" in algae, 112
Giardia Im/iblia, 149
Gle7wdmmiii lubwievsijorvie, 141, 142

putr is cuius, 139
Gloinerella cingulata, 62, 63, 64
Glossiua morsitans, 146
Go-mphoneiim coiistrictui/i, 89, 90, 93,
95

geiiimatwn, 93

riitricatum, 93

longiceps, 93

olivacewu, 93

parvuhnu var. 7//icrop7is, 85, 87, 89,
90, 91, 94

tenelhn)!, 94
Goniwn, 126
Gojwspora arevicolae, 207

varia, 101
Grannnatophora marina, 95
Gregarina blattaru?n, 190

cimeata, 203

ovata, 190

polyviorpba, 203
Gregarinida, 188
Guyciwtia sphaerulosa, 200
Gyiiiiiodinimii, 185

Haeuiatococcus, 101, 104, 107, 119
Haemogregarinidae, 195
Hacmosporidia, 197, 205

Haploid cycle in fungi, 46-48
Haplosporidia, 202
Haplosporidiuvi livnwdrili, 199, 202
Helicosporidia, 202
Helicosporidi'inii parasiticirni, 202
HeHozoa, 179

Helkesiinastix faecicola, 147, 148
Hermaphroditism in fungi, 58
Heterocapsa triqueta, 139
Heterouiita globosa, 148
Heterothallism, 50, 54, 62, 103
Hexamitus "i?Jtesti?iaIis,'" 149
Holdfast substances in Farameciuvi,

304
Holdfast union in Paranieciwii, 301
Homothallism in algae, 52, 103

in fungi, 51, 52, 59
Hormone-induced sexuality, 151
"Hunger" theory of sex, 336
Hyalosporiva canibalopsisae, 190
"Hybrids," Paravieciwii, 111
Hybridization of algae, 106
Hymenomycetes, smuts, 71
Hypermastigotes, 157
Hypomyces, 71

sola/ii f. cucurbitae, 59
H y psoble7mrus gilbert i, 197, 198

I chthy OS poridiimi giga7/teu7/?, 202

hertivigi, 202
Incompatibility factors in fungi, 60, 61

Kablia si7iiplex, 111
Kalpidorhy7ich7is are7iicolae, 188, 193
Karyolysis sp., 195
Klossia belicis, 195
loosei, 195

Lad a tanishi, 231, 235
Lagcjiophrys, 1 3 1

tattersalli, 228
La7nblia intesti77aUs, 149
Lankesteria cidicis, 101
L-bodies among bacteria, 32
Lecythiiriii thalasse7iiac, 193
LeisJyiiiaiiia do7wva7ii, 147
Leptospiro7iy7npha, 159, 168, 169, 343

e7ipora, 161

ivachula, 161, 163
Leucophrys panda, 321

SUBJlcr INDEX

359

Lihelliis constrictus, 94, 95
Life cN'clc of bacteriophage, 4
Life c\clcs, effect of host on, 206

in fungi, 42
Linkage s\steni in viruses, 7
Loxodcs striata, 235
LoxophyUuiii fasciola, 217, 321

ohms 11///, 217

Mastigella vitrea, 143
Mastigb/a by lac, 143

sctosa, 143
iMastigophora, 135
Mating reactions in Para/z/cci/i/// caii-

iiatii///, 295
Mating reactivity of Fara//icchi///, 305,

312
Mating substances in ciliates, 284
Mating s\stems, Parainechnii bursaria,

296, 297
Mating type in algae, 103
Mating t\'pes in Fara///echn/i aiirelia,

266, 294
Mating type substances in Para7//e-

ciuvi, 290, 307
Meiosis in Coccidia, 195
Mejosis in eugregarines, 190

in Para///eciw//, 306
Alelophagus ovimis, 147
Melosira miv/i/zuloides, 83

varians, 84, 94
Meridio?! circulare, 95
Metazoan fertilization, 325
Met opus sigmoides, 238, 240
Microsporidia, 201
Monas tervio, 147

rivipara, 147
Movoblepharella, 120
Aionoblepharis, 71
Monocy Stella arndti, 190
Monocystis agilis, 190

liinatiis, 190

7//ackici, 207

i//razeki, 190

rostrata, 190
"Mother" and "daughter" chromo-
somes, 343, 344
Mucor, 68, 71

rnucedo, 69
Myxosporidia, 197, 206

Naviciila constricta, 95

nucigera, 94

crytocephala var. vc//eta, 86, 87, 91,
95

ciispidata var. a//ibigna, 94

didyvia, 85, 94

dirccta, 94

firv/a, 94

fo7/ticola, 94

gr evil I a, 95

balopbila, 85, 90, 92, 94, 103, 121

byhrida, 94

pygi//aea, 94

radiosa, 85, 87, 90, 91, 93

ra///ossissr//ia, 94

scopiilorm/i, 94

sev/ivuluv/, 85, 86, 91, 95

s/ibtilis, 94

viridula, 94
Nebela collar is, 175
Nei//atopsis ostreariai/, 189, 205, 207
Nenrospora, 12, 13, 47, 58, 68, 71
Niedim/i affine var. ai//pbirby7/cbus,

94
Nitzscbia fo//ticola, 92

bybrida, 94

loi?gissJ7//a, 94

p^/t'«, 96

sigvwidea, 94

subtilis, 86, 94
Noctihtca, 139

7/iiliaris, 140
Nosema aedis, 201
N'ofi/^, 153, 157, 168, 169

proteiis, 155
Nuclearia siu/plex, 169, 173
Nuclear phenomena, preganietic, 236
Nucleocytoplasmic interactions in the

B system of P. aiirelia, 111
Nyctoterus cordiforv/is, 228, 235, 238-
240, 327

Ocbrospbaera veapolitai/a, 135, 136,

245
07/ycbodro7/ius gra7idis, 321
Opali7ia caiidata, 208

d It I lid lata, 208

i7itestinalis, 208

ra7/arn7n, 208, 209, 240
Opist bone eta heTjnegiiyi, 231

360

SUBJECT INDEX

Opisthotrichiwn jatms, 232, 235
Ovivora thalesseiuae, 195
Oxyvwnas, 153, 155, 168, 169
Oxyrrhis uiariva, 143
Oxytricba bifaria, 111, 238

fall ax, 238
Oxytricba sp., 240

Pairing in algae, 1 19

Faraclevelavdia siiiiplex, 228
Faradhiimii poncheti, 143
Faraviecmvi, 119, 134, 316, 320, 321,
326, 327

activation in, 298, 302

aurelia, 213, 214, 219, 222-225, 244,
267, 280, 281, 292, 293, 318-320
breeding systems in, 293
CM = "can't mate" stocks of, 301
cytogenetic considerations in, 267
macronuclear differences in, 275
mating, 266, 267
patterns in, 270, 273, 277
selfing carvonides in, 279
varietal system in, 267

hursaria, 214, 217, 222, 223, 292, 296-
298, 306, 307, 320

calkjmi, 214, 221, 222, 292, 296, 307,
309, 311, 316, 317

caudatiwi, 210, 211, 217, 222, 223,
236, 242, 292, 295, 306, 316

cellular adhesion and mating type
substances in, 290

cellular heredity in, 266

chemistry of the mating type sub-
stances of, 310

fertilization in, 288

"hybrids," 222

holdfast substances in, 304

macronuclear breakdown in, 306

macronuclear regeneration in, 226

mating reactivity in, 305, 307, 312,
317

meiosis in, 306

imilthiiicromiclcatinn, 214, 221, 223,
292

nepbridiatuni, 222, 223

parthenogenesis in, 318

paroral cone formation in, 305

polycarywn, 214, 221, 222, 223, 319

putrhmm, 213, 219

sex substances and mating sub-
stances in, 287, 291, 307

sexual phenomena in, 288

tricbimn, 214, 215, 218-220, 240, 242,
292

ivoodruffi, 214, 221, 222,292
Paroral cone, 210, 305
Parthenogenesis in Farainechnii, 318
Patelli/ia corriigata, 111, 179
Pelo7nyxa carolifiensis, 111

pahistris, 111
fcvicUlhnti votatwn, 46
Pennales (diatoms), 85
Pericystis, 56
Peridinhwi styghnn, 139
Peritricha, 229
Phacus c and at a, 137

pyriim, 137, 138
Phage crosses, 5

genetic recombination in, 7
Pblebotovms argentipes, 101
Pbyco'iiiyces blakesleeamis, 65
Pbyllocardhnn, 101, 119, 120, 123
PbyUomonas striata, 103, 120
Pbyllostaunis mspidatus, 182
Physiology of fertilization in ciliates,

284
Phytomonadina, 137
Pimiularia gibba, 94

hemiptera, 94

stauroptera, 94

viridis, 94
Planozygotes in algae, 120
Plasmogamy in fungi, 44
Pleistopbora periplaiietae, 202
Pleuraspis costatiis, 182
Pleiirosigiiia mtbecida, 84, 85
Plistopbora cbirouoiiii, 201
Polymastigina, 149
Polvmiastigotes, 153
Polystoinella crispa, 176, 177
Poiytoma, 101, 104-106, 111, 112, 119,
" 123, 124, 126, 147

fusiforiiiis, 123
Portinnis de pur at or, 204
Pregametic nuclear phenomena in cili-
ates, 236
Proteomyxa, 169
Proteroino?ias lacertae, 148, 149

SUBJECT INDEX

361

Proteus sp., 32

vulgaris OX-19, 32, 35, 36, 38
Protociliata, 208
Protonionadina, 144
Vrotosipbon, 101, 104, 105, 107, 108,
111, 112, 115, 119, 120-122, 124-126
Protozoa, sex in, 134
Fronwzekella lacertae, 148, 149
Ftcroccphaliis nob His, 193
Ftvchostoz/iuiii {—Lada) chattoni,

235
Fyronevia, 68, 71
Fyxidicula cymbalui)}, 1 74

Raciborskiella, 120, 123
Radiolaria, 181

Reconjugatioii in ciliates, 213
Rhiibdoiievhi adriaticum, 96

arcuatinii, 91, 95
Rhabdostyla venialis, 229
Rhizomastigina, 143
Rhizosoleiua, 83
RI?izopus, 68

■nigricans, 49
Rhoicosphenia curvata, 85, 94, 95
Rhopalodia gibba, 93
Rhyiichonyi/ipha, 167-169

tarda, 167

Saccbaromyces sp., 71
SaccharoDiy codes, 71
Saccinobacuhis, 153, 168, 169

avibloaxostyliis, 153, 155
Salmonella schottniiilleri, 32

typhinmritnn, 25
Sarcodina, 169
Sappinea diploidea, 171, 172
Schizojievia lacustre, 94
Schizophylluni covmnine, 6S
Schlauchkernserie, 182, 185, 186
Schneideria sp., 207
Sciara coprophila, 207
Sep/t7 officinalis, 204
"Sex," origin and evolution of, 335,
339

in algae, 100, 102, 105, 114

in £. co/i, 25

in fungi, 42, 49, SS, 56, 60

in Protozoa, 134

in Sporozoa, 202

Sex determination, 342

in Sporozoa, 203
Sex differentiation, 342

in algae, 101

in Sporozoa, 203
Sexuality, hormone-induced, 151
Sipbonaria, 68, 71
Smuts, 71

Spathidiinn spatbula, 227, 228
Sphaeroniyxa sabrazesi, 200
Sphaerozouni, 1 8 3

piinctatuvi, 182, 185
Spindelkernserie, 182, 185, 186
Spirillina vivipara, 176, 177
Spirochona, 232, 233
Spiroi)io7ias angustata, 148, 149
Sporozoa, 187

Staiironeis phoeiiicenteron, 94
Stein pell a magna, 201
Steinina ovalis, 203
Stenophora jidi, 190
Stephanos pbaera, 102, 115
Streptobacillus nioiiilijoriiiis, 32, 35
Stroniatinia narcissi, 56
Stylocephalus longicoUis, 190
Stylonychia, 240

?ny tills. 111

pustulata, 321

putrina, 321
Stylorhy7ichus longicoUis, 344

oblongatus, 188, 193
Suctoria, 233
Surirella capronii, 95

geiimia, 96

splendid a, 95

striatida, 95
Synchytriuin, 71
Synconjugant, 233
Syndiniujn, 186
Sy?iedra affinis, 91, 95

runipens var. fragilarioides, 85, 90.
92,94

Z///M, 85, 90, 92, 94

Tabellaria, 95
Telosporidia, 203
Tenebrio inolitor, 203
Testacea, 173
Testudo graeca, 149
Tetradonta variabilis, 120

362

SUBJECT INDEX

Tetrabymena pyrijonjiis, 281

sp., 217, 32*1

rostrata, 229
Tetrmnitiis rostratus, 147, 149, 150
Tetraspora, 102, 1 14, 117, 119, 120, 122,
123

lubrica, 114
Thalassicolla, 185

7mcleata, 182, 183, 185

spmnida, 185
Tbelohania legeri, 199, 201

o pack a, 201
Tokophrya cycloptnii, 233, 234
Trichodhm spheroidesi, 228, 231
Trichomastix lacertae, 149
Trichomonas hominis, 149

intestmalis, 149
Trkhonyvipha, 157, 159, 161, 168, 169,

245,'343, 344
Trineiihi enchelys, 175

linear e, 175
''Troglodytes;' 173
Trypaiwsmna bnicei, 146

congolevse, 144-146

cruzi, 144

equiperdw/j, 145

ganihievse, 146

lewisi, 145

rhodesieiise, 144, 145, 146

rotatoriinn, 144
siviiae, 145, 146
Trvpanosomatidae, 144, 147

Urceolaria synaptae, 231
Urinyynpha, 163, 165, 338

fn'/e^, 165
Uropleptns vwhilis, 221, 228, 242
Urospora lagidis, 190, 191, 193
Urostyla grand is, 228

Vainpyrella, 169

Vegetative cells and gametes in algae,

101
Viruses, linkage system in, 7

mutual exclusion effect in, 2
Vis-a-vis pairs in algae, 120
Vorticella microstonia, 229, 231

Dionilata, 231

nebulifera, 228

Wagner ella bore alls, 181

T^oothamiuvi alternans, 228
7vygosaccharoviyces, 71
Xygosonia globoswii, 190
Zygospore germination in algae, 124
Zygote development in diatoms, 91