J. Cell Sci. 17, 471-493 (i975)
Printed in Great Britain
471
CYTOFLUORIMETRIC ANALYSIS OF
NUCLEAR DNA DURING MEIOSIS,
FERTILIZATION AND MACRONUCLEAR
DEVELOPMENT IN THE CILIATE
TETRAHYMENA PYRIFORMIS, SYNGEN 1
F. P. DOERDER* AND L. E. DE BAULT
Departments of Zoology and Psychiatry,
University of Iowa, Iowa City, Iowa 52242, U.S.A.
SUMMARY
Fluorescence cytophotometry was used to study nuclear DNA content and synthesis patterns
during meiosis, fertilization and macronuclear development in the ciliated protozoon, Tetrahymena pyriformis, syngen 1. It was found that cells entered conjugation with a G1 (45C) macronucleus and a G3 (4C) micronucleus. During meiosis the micronucleus was reduced to 4 haploid
nuclei, each with a i C amount of DNA; each meiotic product then replicated to 2C, but only
the nucleus next to the attachment membrane in each conjugant divided to form the two i C
gametic nuclei. The gametic nuclei replicated to 2C prior to fertilization; hence there was no
S-period in the 4C fertilization nucleus (synkaryon). The first postzygotic division products
immediately entered an 5-period to become 4C, and at the second postzygotic division, each of
the two 4C nuclei in each conjugant divided to form one 2C micronucleus and one 2C macronuclear Anlage. The macronuclear Aniagen began DNA synthesis immediately and were about
8C at the completion of conjugation; the micronuclei did not undergo rapid DNA doubling and
measured between 2C and 3C when the conjugants separated. The old macronucleus did not
participate in any >S-period during conjugation and began to decompose after the second postzygotic division; it contained an average of 24C at the end of conjugation. From this sequence
of nuclear divisions a pattern emerges that, unless a general cytoplasmic signal for DNA synthesis is suppressed, DNA synthesis always occurs in micronuclear division products immediately following separation of sister chromatids.
Nuclear development continued in the first two cell cycles after conjugation. In exconjugants
(the first cycle), macronuclear Aniagen underwent two rounds of DNA synthesis to become 32C
and both micronuclei also underwent DNA synthesis. However, prior to the first cell division,
one micronucleus and the old macronucleus completely disintegrated, and at the first cell division the remaining 4C micronucleus divided and one macronuclear Anlage was distributed to
each resulting caryonide. At the end of the second cell cycle, the dividing macronucleus of each
caryonide contained about 128C.
These results relate to the question of ploidy of macronuclear subunits. It is argued that the
G, macronucleus contains 22 or 23 diploid subunits, each subunit being a copy of the diploid
micronuclear genome. It is suggested that unequal macronuclear division relates to the question
of subunit ploidy by playing a role in the phenomenon of macronuclear assortment.
• Present address: Department of Biology, University of Pittsburgh, Pittsburgh, Pennsylvania, U.S.A. 15260.
472
F. P. Doerder and L. E. De Bault
INTRODUCTION
Although the nuclear cytology of conjugation in TetraJiymena pyriformis, syngen i,
has been previously described (Nanney, 1953; Ray, 1956), it has never been examined
with techniques sensitive enough to determine relative amounts of DNA at various
stages of the nuclear reorganization process. In this paper we will report such a study
using the technique of fluorescence cytophotometry.
The main objective was to probe more deeply into the nature of the macronucleus
which, by a variety of techniques, has been shown to be a compound structure consisting of many independent subunits. Various inquiries into the nature of these subunits have been made, but the results are conflicting (see Discussion). Extensive
genetic evidence indicates that the G1 macronucleus contains 45 diploid subunits,
each organized as a complete diploid genome (reviewed by Nanney, 1968; Doerder,
1973). However, microspectrophotometric comparison of macronuclear and micronuclear DNA reveals that in order for there to be 45 subunits they must be haploid
(Gibson & Martin, 1971; Woodard, Kaneshiro & Gorovsky, 1972).
We sought to resolve this paradox in 2 ways. First, since the cytophotometric
studies just cited rely on the assumption that the G2 micronucleus contains the 4C
amount of DNA, we examined more closely the structure and DNA content of this
nucleus, particularly for the possibility that it too might be a compound structure.
Either it might enter meiosis with a Gu rather than G2 amount of DNA (see Wolfe,
1973), or its chromosomes might be differentially polyneme. Second, we measured
the DNA content of macronuclear Anlagen throughout their development during
conjugation and in the first 2 cell cycles after conjugation. Loss of DNA during
development (chromatin diminution; see Ammermann, Steinbruck, von Berger &
Hennig, 1974) or extensive over-replication beyond the normal macronuclear DNA
content would have important implications concerning the nature of macronuclear
structure.
As will be seen, our results do not completely resolve the problem of subunit ploidy,
but they do suggest new approaches to the problem. In addition, several unanticipated
observations concerning the control of micronuclear DNA synthesis were made. They
will be reported here.
MATERIALS AND METHODS
Strains and culture conditions
Inbred strains A and B of Tetrahytnena pyriformis syngen 1, derived from wild isolates
(Nanney, Caughey & Tefankjian, 1955; Nanney, 1959), were used in this study.
Cultures were grown axenically in proteose peptone at 10 g/1. plus yeast extract at 15 g/1.
Samples containing the dividing cells used to establish Gl and G, DNA values were prepared
by harvesting, washing and fixing log phase cultures (<io* cells/ml). For crosses, strains A
and B were grown to stationary phase, harvested, washed free of culture medium, and resuspended in Dryl's physiological salt solution (Dryl, 1959). Twelve hours later the starved cells
were mixed in equal numbers in sterile depression slides. Normally, mating began within 1 h
and at 25 h half of the cells had formed pairs. Samples for cytophotometry were taken for
fixation at intervals beginning at 1 h after mixing and continuing until the second postzygotic
cell division. Samples were also taken from the left-over (non-mating) cultures in Dryl's solu-
Tetrahymena micro- and macronuclei
473
tion to serve as controls for any effect starvation might have on nuclear DNA content. In
order to be certain that normal conjugation had occurred (see Allen, 1967), 60-120 individual
pairs were isolated from the duplicate mating mixtures at 6 h and tested for true conjugation
(viability) as described by Nanney & Caughey (1953). In both replicates the viability was
greater than 99 %, meaning that nearly every conjugating pair underwent normal nuclear
reorganization.
All samples for Feulgen staining were air dried on clean microscope slides, fixed for 15 min
in 3:1 ethanol: acetic acid, rinsed with water, and transferred to absolute ethanol for storage at
4 °C until stained. The slides were then rehydrated and hydrolysed for 20 min in I N HC1 at
40 °C, since preliminary studies had determined that the 20-min hydrolysis time produced
specimens with an extinction coefficient suitable for cytofluorimetry (Bohm & Sprenger, 1968).
The cells were stained for 2 h in SchifT's reagent (Elftman, 1959), bleached, dehydrated and
mounted. In order to lessen fluorescence decay during measurement the slides were allowed to
age in the dark for at least 3 weeks prior to measurement.
Fluorescence cytophotometry
Relative micronuclear and macronuclear DNA contents were determined by fluorescence
cytophotometry (Bohm & Sprenger, 1968). For these measurements the Leitz MPV-fluorometer
(E. Leitz GmbH, Wetzlar, Germany) was used in an optical arrangement that allowed
successive and/or simultaneous illumination with transmitted and incident light and was
similar to that described by Ruch (1970). Since the pararosaniline dye of the Schiff reagent has
a low fluorescence efficiency, high levels of exciting irradiation are required to produce detectable amounts of red fluorescence. This was achieved by illuminating the specimen from the top
with a Ploem vertical illuminator, which uses the objective in reverse to focus the exciting light
on to the specimen at a very high flux density (Ploem, 1967). The specific optical arrangement
in the Leitz-MPV-FIuorometer used for the measurements utilized a high-pressure Xenon
lamp XBO-150 W (Osram, Berlin) in combination with an interference heat-protecting filter,
a red absorbing filter 5 mm BG38, and an HgS46 interference filter to produce a near monochromatic light source at 546 nm. This intense green light was reflected and focused on to the
specimen by a dichroic minor (Leitz) TK580 and a Phaco NPL 100/1.30 objective, respectively.
The induced red fluorescence was then collected by the objective lens, passed through the
dichroic minor and a K580 barrier filter (both of which transmit red light), the areas of the
individual nuclei optically isolated by the MPV diaphragm, and the image of a given nucleus
projected on to the photomultiplier S-20 type 9558AQ with a quartz window. Only a narrow
band of the red spectrum was used for the measurements (Bohm & Sprenger, 1968) and was
selected by a 60-mm interference wedge filter set at 660 nm. Thus, fluorescence intensities at
660 nm were recorded as photomultiplier voltage output and were proportional to the DNA
content.
Nuclear identification
Microscopic identification of the various stages of conjugation was facilitated by the fact that
each micronuclear division is characterized by a particular number and configuration of nuclei.
These stages are shown in Figs. 7-20 and are described in the figure legends. A complete
description of conjugation in Tetrahymena has been given by Ray (1956). In the 2 crosses from
which nuclei were measured, conjugation lasted from 16 to 18 h. Since the nuclei could be
visually identified as to stage, no attempt was made to use synchronous matings or to time
precisely the intervals between successive nuclear divisions. A rough estimate, however,
indicates that for any given pair about 5 h elapsed between pair formation and metaphase of
the first prezygotic division; about 4 h more until the end of the second postzygotic nuclear
division; and 6-8 additional hours until separation of the conjugants. The time between
separation and the first postzygotic cell division was 8-10 h, and between the first and second
cell divisions 4-6 h.
F. P. Doerder and L. E. De Bault
474
RESULTS
Gx and G2 standards and nuclei of sexually reactive cells
In order to determine G1 and G2 macronuclear and micronuclear standards, the
nuclei of log phase cells of strains A and B were measured (Fig. i). The DNA content
of G2 macronuclei (shaded in Fig. i) was determined by averaging the DNA content
16
12 -
8 -
4
-
14 18 22 26 30 34 38 42 46 50 54 100
500
300
Mlcronuclei
900
700
1300
1100
1500
Macronuclei
Fig. i. Relative DNA content of micronuclei and macronuclei in log phase cells
(strains A and B (A, B)) and in sexually reactive cells (c). Shaded areas indicate
micronuclei in various stages of mitosis (probably in S; see text) and macronuclei in
dividing animals (in G,). Ordinate: number of nuclei. Abscissa: relative DNA
content in arbitrary fluorescence units.
of macronuclei of animals beginning cell division (recognized by the prior division of
the micronuclei and by the characteristic 'peanut' cell-shape), since these nuclei have
been shown to be in the replicated, G2 phase (McDonald, 1962; Woodard et al. 1972).
Similarly, the DNA content of G2 micronuclei was determined by averaging the
Tetrahymena micro- and macronuclei
475
micronuclear DNA content of animals not in cell division. This is valid because, as
these studies also show, the micronucleus begins its 5-period in anaphase of its
division and enters G2 shortly after the completion of cytokinesis. (Note that most of
the newly divided micronuclei in dividing animals, which are shaded in Fig. 1,
are evidently in S and therefore have more than half the mean G2 DNA amount.)
Table 1. DNA content of G2 nuclei and of nuclei in sexually reactive cells and
premeiotic conjugants
Macronucleus
Nuclei measured
Fig. 1 data
Gt standard
Sexually reactive cells
Second experiment
Gt standard
Sexually reactive cells
Premeiotic conjugants
Micronucleus
Feulgen DNAf
(AO
9°4 ±73
4io± 19
(5°)
(13)
Feulgen DNAf
354 ±048
349 ±046
43-o± 1-36
36710-72
422 ± 12
407 ± 0-94
• Macronuclear to micronuclear size ratio.
t Mean ± standard error in arbitrary units.
908163
4S6±:7
(9)
(41)
(50)
(N)
MA/MI*
(90)
(5°)
25-5
n-8
(22)
(4i)
(So)
2I-I
124
IO-3
These averages are shown in Table 1, which also summarizes data pertaining to the
nuclei from sexually reactive cells and includes data from a second, identical experiment. Since the mean macronuclear DNA content of the sexually reactive animals is
about half that of known G2 macronuclei, and since there is little difference between
the micronuclear means, it is concluded, as shown in Fig. 1, that Tetrahymena enters
conjugation with a Gx macronucleus and a G2 micronucleus.
Table 1 also shows that the ratios between G2 macronuclei and G2 micronuclei are
close to 22-5, or half of 45, the number of assorting subunits. This observation confirms the cytochemical data of Gibson & Martin (1971) and Woodard et al. (1972)
which show that if the Gx micronucleus possesses the 2N number of (unineme)
chromosomes, then there is sufficient macronuclear DNA for only 45 haploid subunits.
This assumes, of course, that each subunit would be a complete, non-diminished
haploid genome.
It should also be noted here that the micronuclei assumed to be in G2 (frame c of
Fig. 1) show an almost 2-fold variation in DNA content. About 10% can be attributed
to machine error, and a great deal, perhaps all, of the variability may be due to
differing degrees of nuclear compaction (Garcia, 1970). It should be pointed out that
such variability is a consistent feature of micronuclei at other stages of conjugation
and that it has been observed in other studies (Woodard et al. 1972; J. D. Berger,
personal communication). The meaning of this variability, if not artifactual, is thus
not clear, but it does not alter the conclusions presented here.
F. P. Doerder and L. E. De Bault
476
4
8
12 16 20 24 28 32 36 40 44 48 S2
Fig. 2. Relative DNA content of prezygotic micronuclear division products, A, micronuclei of sexually reactive cells; B, micronuclei in prophase of meiosis I (clear area),
division products of meiosis I (shaded area), and division products of meiosis II, with
chromosomes visible (hatched area); C, division products of meiosis II, condensed;
nuclei at attachment membrane are shaded; D, nuclei at attachment membrane
undergoing 3rd prezygotic division; E, gametic nuclei formed by 3rd prezygotic
division with chromosomes visible (hatched area), fully condensed (clear and shaded
area) and at exchange (shaded area).
Nuclear DNA during conjugation
In Tetrahymena, 5 nuclear divisions - 3 prezygotic and 2 postzygotic - occur in
each animal during conjugation (see legends to Figs. 7-20; see also Nanney, 1953;
Ray, 1956). Meiosis I and II occur during the first 2 prezygotic divisions; the third
prezygotic division forms the haploid gametic nuclei by mitosis. The 2 postzygotic
divisions of the fertilization nucleus (synkaryon) are also mitotic and ultimately give
rise to 2 new micronuclei and 2 new macronuclei. The measurements of prezygotic
Tetratiymena micro- and macronuclei
477
micronuclear divisions are shown in Fig. 2 (see also Table 2); the data pertaining to
fertilization and the postzygotic nuclear divisions are presented in Figs. 3 and 4 (see
also Table 3). The relative DNA content of the (old) macronuclei during conjugation
is shown in Fig. 6 (p. 481). Since these results are the pooled data obtained from
several slides (fixed at different times after mixing of the 2 strains, but all stained at
Table 2. DNA content of nuclei resulting from prezygotic
micronuclear divisions
Fig. 2 data
Stage of conjugation
Supplementary (data*
Feulgen DNAf
(N)
37-7 ± 1-08
39-9 ±1-70
(20)
46-014-14
(4)
Pre-crescent pairs
(9)
(12)
1st division products
19-2 ± 1-50
(10)
52-8 12O2
22-2 1 0 6 4
8010-37
(13)
II-81O-25
(24)
21-810-75
(52)
17-410-36
(156)
24I I 0 6 0
I9-91O-6O
Unpaired cells
2nd division products
Chromosomes visible
Condensed products
Attachment membrane
Relic nuclei
Feulgen DNAf
(N)
(12)
(16)
21-9 ± I-IO
(12)
3rd division, metaphase
Gametic nuclei
14-410-87
(12)
Chromosomes visible
25-210-70
(46)
Fully condensed
21-3 lo-8o
—
At exchange
(12)
• Data obtained from same: cross as data in Fig. i, but using a slide hydrolysed for
f Mean ± standard error in arbitrary fluorescence units
(48)
—
—
—
—
30 min
Table 3. DNA content of nuclei inpostzygote nuclear and cell divisions
{data from Figs. 3, 4)*
Micronuclei
Macronuclear Anlagen
A
Stage
Feulgen DNA|
(N)
Feulgen DNAf
(N)
Synkarya, metaphase
1st postzygote division
Early products
Entering second division
45-i ±1-58
(27)
22-5 10-84
40-410-58
(20)
2nd postzygote division
Stage 1
Stage 2
Stage 3
18-110-94
21-2 1O-45
26-3 iO4O
(14)
(52)
(106)
19-610-55
39-610-87
61 0 1 i-6o
(52)
(106)
Exconjugants
With 2 micronuclei
With 1 micronucleu8
3OOiO79
37'4l2-IO
(20)
76-413-96
(20)
189119
(16)
•
—
—
—
—
(84)
(8)
(16)
Caryonides
37-Oil-24 (4C) (33)
G, nuclei
1165152 (126C)
(5)
• The data summarized here are directly comparable to those in Table 2 (Fig. 2! data).
f Mean 1 standard error in arbitrary units.
31
C EL 17
F. P. Doerder and L. E. De Bault
47 8
.—,—.
0
.
.
i
i n
i' n i n
i
i
i
i—i
i 16
i 20
i 24
i 28
I 32 36
i 40
i 44
i 48
I 52
i 56
i 60
i 64
I 68
I 72
4• 8i 12
Fig. 3. Relative DNA content of postzygotic nuclear division products and micronuclear development. A, fertilization nuclei (synkarya) in metaphase; B, newly
formed first postzygotic division products (shaded area), entering second postzygotic
division (clear area); C, new micronuclei (shaded area) and presumptive macronuclear Anlagen (clear area) formed by second postzygotic division; D, E, micronuclear development during conjugation; F, c, exconjugants with 2 and 1 micronuclei, respectively. For stage designations, see text and legends to Figs. 16-19.
the same time), it is important to note that betvveen-slide variation was small, as for
example is indicated by the close agreement between the means of the second experiment in Table 1 and those of the same stages in Table 2. Each nuclear division may be
summarized as follows.
(1) First prezygotic division (Fig. 2, Table 2). Although the data suggest a small
amount of DNA synthesis during prophase I, this may be artifactual and therefore not
true meiotic DNA synthesis (see Discussion). This conclusion is consistent with the
observation that the 2 nuclei formed as a result of this division each have half the
Tetrahymena micro- and macronuclei
12
AC
8C
I
I
479
32C
16C
I
8
4
40
30
20
10
- c
30
20
10
12
8
n
JL
40
80
120
160
200
240
280
320
360
Fig. 4. Relative DNA content of developing macronuclear Anlagen in conjugants
(stages 1, 2, 3 (A, B, c)) and in the first cell cycle after conjugation (0 and E, exconjugants
with 2 and 1 micronuclei, respectively). For stage designations, see text and legends to
Figs. 16-19.
amount of Gz micronuclear DNA rather than half the amount of prophase I DNA.
As expected for meiosis I, the products of this division do not synthesize DNA.
(2) Second prezygotic division (Fig. 2, Table 2). Four nuclei, each with the haploid
complement (N = 5) of chromosomes (Ray, 1956) and each with one-fourth the
amount of G2 micronuclear DNA, are formed by this division. The 2 nuclei (one in
each conjugant) closest to the attachment membrane (the nuclei which divide in the
third prezygotic division), and in most pairs all of the remaining second division
products as well undergo a round of DNA synthesis.
(3) Third prezygotic division (Fig. 2, Table 2). The 2 gametic nuclei formed by
this division each receive an amount of DNA equivalent to one-fourth that of a Ga
31-2
F. P. Doerder and L. E. De Bault
480
micronucleus. These haploid gametic nuclei immediately increase in DNA content,
apparently beginning an S-period before the chromosomes uncoil to form the typical
gametic nucleus. Thus by the time migratory gametic nuclei are reciprocally exchanged, all 4 gametic nuclei of a pair contain an amount of DNA equivalent to half
that of a G2 micronucleus. The relic nuclei do not participate in this, or any subsequent, DNA syntheses. Since the minimum amount of DNA present when the haploid
number of chromosomes is clearly discernible (e.g. after both the second and third
A
10
—
64C
8
6
4
:
-
JTJ
•
8
6
4
2
—
i
B
i
i
i
1
J
•
J
i
2
10
128C
—1 1
r
1
1
1
1
1
T
0
200 400 600 800 1000 1200 1400 1600 1800
Fig. 5. Relative DNA content of macronuclear Anlagen in the second postzygotic cell
cycle. Samples A and B were taken 3 h apart. Shaded area in B indicates macronuclei
(Gj) in dividing animals.
prezygotic divisions, but before the ensuing 5-periods) is always about one-fourth that
of a premeiotic micronucleus, it may be considered the iC value. The Gx micronucleus,
therefore, contains the diploid number of chromosomes and the 2C amount of DNA,
whereas the G2 micronucleus contains twice that amount (4C). This observation
virtually eliminates the possibility of differential chromosomal polynemy and favours
the concept of a simple, non-compound micronucleus (see Discussion).
(4) First postzygotic division (Fig. 3, Table 3). Since the gametic nuclei contain
the 2C amount of DNA, the fertilization nucleus (synkaryon), which immediately
enters division after its formation, apparently divides without a further 5-period. The
increase in the DNA content in the synkarya (which were in anaphase) to more than
the 4C level probably represents the onset of DNA synthesis in the products.
(5) Second postzygotic division (Figs. 4, 5, Table 3). It is at this division that the
micronuclear and macronuclear genetic material for future asexual generations is
separated. Each replicated (4C) product of the first synkaryotic division divides
Tetrahymena micro- and macronuclei
481
equally, with one 2C product going to the anterior end of the conjugant, the other 2C
product to the posterior end.
The 2 nuclei which reach the anterior end of each conjugant develop into new
macronuclei in 3 defined stages (see Figs. 16-19). In the first stage, telophase of the
second postzygotic division, each macronuclear Anlage receives the diploid (2C)
amount of DNA; it is small and condensed. In the second stage, the Anlagen are still
A
20
-
16 -
12
8
4
-
-^n—i_n m
i
32
28
24
20
16
12
i
i
i
i
i
i
i
i
i
i
i
i
i
i
i
B
-
8
4
r
24 " c
20 16 12 8 4
HI
1
24
20
16
12
8
4
1
|
|
- D
-r L
• 1L
J
1
i i 'r i
0
400
800 1200 1600
Fig. 6. Relative DNA content of old macronuclei at various stages of conjugation.
A, unpaired cells; D, end of second prezygotic stage; c, end of first postzygotic;
D, stage 3 Anlagen.
located in the anterior portion of the cell, but they are much larger and more diffuse;
they measure about 4C (Fig. 4, Table 3). At this stage the old macronucleus, which
apparently does not synthesize any DNA during conjugation (Fig. 6) begins to change
physical form (see Fig. 17). In the third stage, the macronuclear Anlagen greatly enlarge and migrate to the centre of the conjugant; the posteriorly located products of
the second postzygotic division (micronuclei) also migrate and come to lie between the
2 Anlagen. The Anlagen average 6C during this stage (range, 4-8C) (Fig. 4, Table 3).
482
F. P. Doerder and L. E. De Bault
In addition the old macronucleus begins to lose its DNA (Fig. 6) and becomes positioned in the posterior end of the cell. The old macronucleus averages about 24C in
this stage. It is with this nuclear configuration that the conjugants separate some
6-8 h later.
The 2 nuclei reaching the posterior end of each conjugant at the second postzygotic
division become new micronuclei (actually, one is autolysed in the next cell cycle).
Unlike the two anterior products of this division and unlike products of previous
divisions, the new micronuclei remain at 2C, apparently not beginning DNA synthesis
until after the macronuclear Anlagen have undergone one round of replication and
migration (stage 3). During this 6-8 h stage the micronuclei average about 3C, range
2-4C (Fig. 3l Table 3).
Nuclear development in the first and second cell cycles after conjugation
Since synchronously growing exconjugants (cells in the first cell cycle) and caryonides (cells in the second cell cycle) were not isolated for measurement, the course of
nuclear development in the first 2 cell cycles after conjugation is based upon inference
from somewhat asynchronous populations of cells. Despite this shortcoming, the
major steps in nuclear development are unambiguous.
The data pertaining to micronuclear development are presented in Fig. 3 (see also
Table 3). They show that there is only a small increase in the DNA content of the new
micronuclei between stage 3 of macronuclear development and exconjugants containing 2 micronuclei. There is not, however, any significant difference between the
DNA content of the 2 micronuclei within a given cell. This latter observation is consistent with autoradiographic analysis (McDonald, 1973) and suggests that both
micronuclei, including the one destined to disintegrate, undergo DNA synthesis. In
exconjugants in which the old macronucleus and one micronucleus have disintegrated
(these events appear to occur almost simultaneously), the remaining micronucleus
measures close to 4C. Thus, it appears that after the second postzygotic nuclear
division 14-18 h are required for the micronuclear DNA content to double; this is
compared to the normal replication time of 0-5-0-75 h (Woodard et al. 1972). The
micronuclear cycle is apparently normal during the second cell cycle; all of the
micronuclei measured, except those in dividing cells, contained the G2 (4C) amount
of DNA.
The data on macronuclear development during the first 2 cell cycles are presented
in Figs. 4 and 5 (see also Table 3). Conjugation ends when the Anlagen have reached
between 4 and 8C, and the first cell cycle apparently ends when the Anlagen have
reached 32C. Assuming there is no diminution of DNA (see Discussion), this means
there are 1-2 DNA replications in conjugating cells and 2-3 DNA doublings in the
first cell cycle, for a maximum of 4 doublings of the diploid complement in each Anlage by the end of the first cell cycle. The data also suggest that in the first cell cycle
the disappearance of the old macronucleus and one micronucleus is not correlated
with any particular amount of DNA present in the Anlagen.
The development of the macronuclei in the second cell cycle is more difficult to trace
because the 2 samples (taken 3 h apart at the time when cells in the second cell cycle
Tetrahymena micro- and macro-nuclei
483
should have been present) may have contained some cells which had not mated. Thus,
we are not absolutely certain that every cell was indeed a caryonide; however, because
the proportion of unmated cells in earlier samples was small and since both of these
samples consisted exclusively of cells with one macronucleus and one micronucleus
with only a few exconjugants, we are reasonably certain that most of the cells measured
were caryonides. This conclusion is perhaps reinforced by the data themselves
(frames A and B of Fig. 5), since they are reasonably homogeneous despite the small
sample sizes. The Anlagen contain, on the average, nearly twice the amount of DNA
as the average G± macronucleus, ranging from 32C in the smallest cells to an average
of about 128C (108-157C) in the 5 nuclei in dividing cells.
The value 128C has important implications concerning the nature of the macronucleus, and will be referred to in the Discussion.
DISCUSSION
Structure of the micronucleus
As noted in the Introduction, the question of micronuclear structure, particularly
chromosomal polynemy, is especially important because it has important implications
concerning the structure of the larger compound macronucleus. Although previous
genetic and cytogenetic inquiries into the nature of the micronucleus have demonstrated that it is diploid (Ray, 1956; for review, see Nanney, 1968; Allen & Gibson,
1973), there have also been occasional reports of micronuclear polyploidy and polyteny
(Alfert&Balamuth, 1957; Ray, 1958). In addition, because micronuclear chromosomes
become visible to resemble those of other eukaryotes only during certain stages of conjugation (Ray, 1956) there was also the possibility that the DNA content of diploid
micronuclei during conjugation would differ from the DNA content of diploid (Gx)
micronuclei of asexually dividing cells.
The evidence presented here, however, clearly shows the quantitative equivalence
of diploid micronuclei in all stages of the life cycle, and also demonstrates the conventionality (except for the crescent stage) of the meiotic process in Tetrahymena. In
addition, the third prezygotic division, which is known to be mitotic, provides the
crucial test for the remaining possibility of differential micronuclear polynemy. The
chromosomes could, for example, be replicated such that the third prezygotic nuclear
division would resolve polynemy by proceeding without intervening DNA synthesis.
However, since DNA synthesis does occur and since each gametic nucleus receives
the iC amount of DNA it is unlikely that the micronuclear chromosomes are polyneme. Since polynemy is also unacceptable for a number of other reasons (Thomas,
1971), the simplest interpretation of the data is that the micronucleus is a 'simple'
nucleus containing the 2N number of unineme chromosomes in the G1 phase.
Control of micronuclear DNA
interactions
synthesis during conjugation and nucleo-cytoplasmic
The results obtained here amply confirm earlier conclusions (Wolfe, 1973; McDonald, 1973) that Tetrahymena enters conjugation and proceeds into meiosis with
484
F. P. Doerder and L. E. De Bault
the G2 (4C) amount of micronuclear DNA. This is in interesting contrast to the
related holotrich Paramecium aurelia, where cells enter conjugation with a Gx micronucleus which doubles in DNA content before meiosis (Woodard, Woodard, Gelber
& Swift, 1966) and to certain hypotrichous ciliates which undergo nuclear divisions
after pair formation as preparation for meiosis (Heckmann, 1963; Nobili & Luporini,
1967).
Discounting the possibility of meiotic DNA synthesis (McDonald, 1973) which
the present technique is unable to discriminate, 4 periods of DNA synthesis during
conjugation can be demonstrated: (1) in all 4 products of the second prezygotic
division (meiosis II); (2) in the gametic nuclei formed as a result of the third prezygotic
division; (3) in the 2 products of the first postzygotic division; (4) in the 2 anteriorly
located products (macronuclear Anlagen) of the second postzygotic division.
From these results the pattern emerges that during conjugation and during the
vegetative cycle as well (McDonald, 1962; Woodard et al. 1972), each nuclear division
in which sister chromatids (centromeres) are separated is immediately followed by
DNA synthesis in the division products. Interestingly, this pattern holds for division
products which need not, a priori, undergo such synthesis, i.e. 3 products of the
second prezygotic division and the 2 gametic nuclei produced by the third prezygotic
division. Although a cause-and-effect relationship between centromere separation and
DNA synthesis might be suggested, this need not be the case. Equally acceptable is a
general cytoplasmic signal for micronuclear DNA synthesis to which only mitotically
dividing (anaphase to telophase) nuclei are sensitive; thus, by failing to divide again,
relic nuclei would not respond to such a signal even though they may lie in cytoplasm
where such synthesis occurs.
Two additional considerations bear on this problem of nucleo-cytoplasmic interactions (for discussion see Nanney, 1953; Sonneborn, 1954).
(1) As also occurs in Paramecium aurelia (Berger, 1969), the failure of the relic
nuclei to divide, even though they have undergone DNA synthesis, suggests that the
signal for DNA replication is separate from the signal for division itself. The nature of
the division signal is not clear, but it is likely that cell membrane is involved. Perhaps
as suggested for P. aurelia by Sonneborn (1954), the close proximity of the nucleus to
the attachment membrane protects it from a transient signal for the initiation of autolysis
in the free nuclei; thus, relic nuclei in autolysis would be unable to respond to a subsequent signal for nuclear division (Nanney, 1953).
(2) In an important exception to the pattern that nuclear division is followed immediately by DNA synthesis in the products, both autoradiographic analysis (McDonald, 1973) and the present results show that after the second postzygotic nuclear
division the 2 nuclei which reach the posterior end (new micronuclei) apparently do not
begin DNA synthesis until after the 2 anteriorly located products (macronuclear
Anlagen) have doubled their DNA content at least once. Since this division appears
to be equational, and since the first postzygotic division products also come to lie in
similar locations but do not undergo differential nuclear development, transient,
regionalized differences which effect nuclear differentiation are probably present in the
cytoplasm. That these differences are indeed cytoplasmic (possibly membrane) and
Tetrahymena micro- and macronuclei
485
not intrinsically nuclear has been clearly demonstrated by Nanney (1953). Unfortunately, there are few clues as to what the cytoplasmic differences might be. One
possibility suggested by recent evidence is that the initiation of RNA synthesis might
play a role. In both P. aurelia and T. pyriformis macronuclear Anlagen begin RNA
synthesis early after the second postzygotic division (Berger, 1973; McDonald, 1973).
In addition, Berger (1973) has found that in P. aurelia developing macronuclear
Anlagen can inhibit DNA synthesis in nuclear fragments that remain in the cytoplasm.
Thus, the first round of DNA synthesis in the anterior products may occur in response
to the general DNA synthesis-inducing signals postulated above, with the necessary
prerequisite for macronuclear differentiation being the initiation of RNA transcription
by substances transiently present in the anterior half of the cell. As a result of this
transcriptional activity, DNA synthesis in the posteriorly located nuclei could be suppressed. Other possibilities of course also exist.
The structure of the macronucleus
As pointed out earlier the question of Gx macronuclear ploidy in Tetrahymena first
arose when it was found that genetic evidence in favour of 45 diploid subunits (reviewed by Nanney, 1968; Allen & Gibson, 1973; Doerder, 1973) is contradicted by
cytochemical evidence which shows that there can be only 23 diploid or 45 haploid
subunits (present results; Gibson & Martin, 1971; Woodard et al. 1972).
The genetic evidence is based largely on the peculiar phenomenon of macronuclear
assortment (also called allelic exclusion). Macronuclear assortment was first observed
in a study of selfers (intraclonal conjugation; Allen & Nanney, 1958), and since then it
has been found that heterozygotes for nearly every locus for which allelic variants exist
give rise during vegetative growth to sublines which express only one allele or the other
(for references, see Sonneborn, 1974). In every case in which it has been measured,
the equilibrium rate of assortment (Rf) of a clone into phenotypically pure subclones
is close to 0-0113 pure lines per fission. Recent experiments suggest that the unit of
assortment is a linkage group consisting of at least one haploid set of chromosomes
(Doerder, 1973), and, assuming randomness of assortment, Schensted (1958) has
derived the equation Rf = i/(2iV— 1) to find that the number of subunits in the Gx
macronucleus is 45.
The only variable associated with macronuclear assortment appears to be the time
of onset of assortment, which for some loci may be during the first or second cell cycle
after conjugation and for others as late as 50fissionsafter conjugation. The occurrence
of the late-assorting loci has been taken as strong support for the concept of diploid
subunits. If, for the sake of argument, there were 45 haploid subunits, it would be
impossible without ad hoc assumptions to explain the apparent locus-specific initiation of assortment: all loci in a haploid subunit would be simultaneously determined
in macronuclear development because alternate alleles would go to another subunit
(see Nanney, 1964).
Several attempts to resolve the problem of subunit ploidy have been made, but none
has so far been successful (D. L. Nanney, unpublished; Allen & Gibson, 1972, 1973).
486
F. P. Doerder and L. E. De Bault
As a complication, both evidence obtained with the electron microscope (Wolfe, 1967;
Nilsson, 1970) and analysis of the kinetic complexity of macronuclear DNA (Flavell &
Jones, 1970; Allen & Gibson, 1972) have been contradictory. And, as has been argued
earlier in this paper, the resolution of the problem is probably not to be found in the
structure of the micronucleus. In addition, our results and cross-hybridization studies
(M.-C. Yao & M. A. Gorovsky, personal communication) virtually eliminate extensive
chromatin diminution of the type seen in Stylonychia (Ammermann, 1971, 1973;
Ammermann et al. 1974) as a possible explanation.
Although the present results also fail to resolve the paradox, they relate to the
problem of subunit ploidy in yet another way. That is, they allow a test of the hypothesis that one allele at each assorting locus is physically lost without replacement by a
similar amount of DNA from each subunit at the beginning of assortment (Nanney &
Doerder, 1972). Assuming, as suggested by the timing of assortment, that both alleles
at each locus are present in a newly elaborated macronucleus, this hypothesis would
predict that the G2 macronucleus in a caryonide should contain an amount of DNA
proportional to the number of loci which would subsequently undergo assortment in
addition to the amount of DNA in a fully assorted cell. Since nearly every locus known
undergoes assortment, we would predict that the amount of DNA in a G2 caryonidal
macronucleus should be nearly twice that of a completely differentiated macronucleus,
or as calculated from Table 1 (using the G2 micronucleus as the 4C standard), about
180-190C. From Table 3 it can be calculated that the macronuclei in dividing caryonides contain neither 90 nor 180C, but about 128C (range 108-157C). Since this
amount of DNA is less than 180C and since it is part of a geometric progression
beginning with 2C, it suggests (1) that non-compensatory gene loss is not the cause
of subunit determination and (2) that macronuclear development probably consists of
full rounds of DNA replication [limited chromatin diminution and amplification of
ribosomal DNA (M.-C. Yao, A. R. Kimmel & M. A. Gorovsky, personal communication) are not eliminated].
Independent experimental results support these 2 conclusions. In T. pyriformis,
strain HSM (possibly phenoset D), Cleffmann (1968) found that the inequality of
macronuclear division was about 8 % of the mean Gx amount of DNA, and recently
for syngen 1, strain D, average inequalities of 6-2 and 10-5% have been found in 2
separate experiments (F. P. Doerder, J. Frankel, L. Jenkins & L. E. DeBault, in
preparation). In addition, Cleffmann observed that DNA-containing bodies were
occasionally extruded from the macronucleus during division. He also found that after
a lower limit was reached, the macronucleus undergoes an additional round(s) of DNA
synthesis without intervening cell division. If in the absence of frequent DNA extrusion during macronuclear division in log phase syngen 1 animals (F. P. Doerder &
L. E. DeBault, unpublished; D. L. Nanney, personal communication), an upper
limit is compensated for by nuclear divisions without an intervening 5-period, then
it is simple to imagine the regulation from a G2 DNA content of 128C to a G2 amount
of 90C. Thus, it would appear that even though there is sufficient DNA for only 45
haploid subunits, the macronucleus is probably diploid throughout the life cycle.
Obviously, a critical test would be to bring the silent allele back into expression, but
Tetrahymena micro- and macronuclei
487
so far, all attempts to do this have failed (Nanney, Reeve, Nagel & DePinto, 1963;
F. P. Doerder, unpublished).
Finally, as an extension of the above discussion, we suggest the possibility that unequal macronuclear division plays a role in the assortment phenomenon. Consider an
impure G1 macronucleus with 23 diploid subunits, some determined (in an unknown
manner) to express one allele, others to express the alternate. If newly replicated subunits retain the same determined state, and if at the next macronuclear division they
assort at random to progeny macronuclei as assumed in the original subunit model
(Allen & Nanney, 1958; Schensted, 1958), that is if the probability of subunit disjunction (going to different division products) is the same as the probability of subunit
non-disjunction (moving to the same division product), then assortment should occur
at Rf = 0-0222, or almost twice the observed rate. Now, if the probability of subunit
disjunction were increased, for example, by a greater tendency for the macronuclear
division furrow to pass through dividing subunits, then macronuclear division would
become more 'mitotic' and Rf would be decreased, perhaps becoming as low as 0-0113.
The unequalness of macronuclear division may thus be due in part to the nondisjunction of some subunits.
However, in the absence of more adequate data and in the absence of computer
simulation it is not clear whether this mechanism would be sufficient to account
simultaneously for the apparent randomness of assortment (E. Orias & M. Flacks,
personal communication) and for assortment at Rf = 0-0113. Appropriate experiments to examine these possibilities are currently under way.
We wish to thank Dr Joseph Frankel for originally suggesting this project and for his advice
and helpful discussion throughout its progress. We also want to thank Nora C. Doerder,
Jonathan Lief, James Berger, David Nanney and Martin Gorovsky for their suggestions concerning the manuscript.
This investigation could not have been completed without the able technical assistance of
Leslie Jenkins.
This project was supported by a Biological Sciences Development award NSF-GU-2591 to
the University of Iowa and by NIH research grant GM-18966 to L.E.D., and by NSF grant
GB-32408 to J.F.
REFERENCES
ALFERT, M. & BALAMUTH, W. (1957). Differential micronuclear polyteny in a population of the
ciliate Tetrahymena pyriformis. Chromosoma 8, 371-379.
ALLEN, S. L. (1967). Cytogenetics of genomic exclusion in Tetrahymena. Genetics, Princeton
55. 797-822.
ALLEN, S. L. & GIBSON, I. (1972). Genome amplification and gene expression in the ciliate
macronucleus. Biochem. Genet. 6, 293-313.
ALLEN, S. L. & GIBSON, I. (1973). Genetics of Tetrahymena. In Biology of Tetrahymena (ed.
A. M. Elliott), pp. 307-373. Stroudsburg, Pa.: Dowden, Hutchinson & Ross.
ALLEN, S. L. & NANNEY, D. L. (1958). An analysis of nuclear differentiation in the selfers of
Tetrahymena. Am. Nat. 92, 139-160.
AMMERMANN, D. (1971). Morphology and development of the macronuclei of the ciliates
Stylonychia mytilus and Euplotes aediculatus. Chromosoma 33, 209-238.
AMMERMANN, D. (1973). Cell development and differentiation in the ciliate Stylonychia mytilus.
In The Cell Cycle in Development and Differentiation (ed. M. Balls & F. S. Billett), pp. 51-60.
Cambridge: University Press.
488
F. P. Doerder and L. E. De Bault
D., STEINBRUCK, G., VON BERGER, L. & HENNIG, W. (1974). The development
of the macronucleus in the ciliated protozoan Stylonycliia mytilus. Chromosoma 45, 401-429.
BERGER, J. D. (1969). Nuclear Differentiation and Nucleic Acid Syntliesis in Paramecium aurelia.
Ph.D. Thesis, Indiana University, Bloomington.
BERGER, J. D. (1973). Nuclear differentiation and nucleic acid synthesis in well-fed exconjugants of Paramecium aurelia. Chromosoma 42, 247-268.
BOHM, H. & SPRENGER, E. (1968). Fluorescence cytophotometry: a valuable method for the
quantitative determination of nuclear Feulgen-DNA. Histocheniie 16, 100-118.
CLEFFMANN, G. (1968). Regulierung der DNS-Menge im Makronucleus von Tetrahymena.
Expl Cell Res. 50, 193-207.
DOERDER, F. P. (1973). Regulatory serotype mutations in Tetrahymena pyriformis, syngen 1.
Genetics, Princeton 74, 81-106.
DRYL, S. (1959). Antigenic transformation in Paramecium aurelia after homologous antiserum
treatment during autogamy and conjugation. J. Protozool. 6, 25 (Abstr.).
ELFTMAN, H. A. (1959). A Schiff reagent of calibrated sensitivity. J. Histocliem. Cytochem. 7,
93-97FLAVELL, R. A. & JONES, I. G. (1970). Kinetic complexity of Tetrahymena pyriformis nuclear
deoxyribonucleic acid. Biochem. J. 116, 155-157.
GARCIA, A. M. (1970). Stoichiometry of dye binding versus degree of chromatin coiling. In
Introduction to Quantitative Cytochemistry, vol. 2 (ed. G. L. Wied & G. F. Bahr), pp. 153171. London: Academic Press.
GIBSON, I. & MARTIN, N. (1971). DNA amounts in the nuclei of Paramecium aurelia and
Tetrahymena pyriformis. Cltromosoma 35, 374-382.
HECKMANN, K. (1963). Paarungssystem und genabhangige Paarungstypdifferenzierung bei
dem hypotrichen Ciliaten Euplotes vannus O. F. Muller. Arch. Protistenk. 106, 393-421.
MCDONALD, B. B. (1962). Synthesis of deoxyribonucleic acid by micro- and macronuclei of
Tetrahymena pyriformis. J. Cell Biol. 13, 193-203.
MCDONALD, B. B. (1973). Nucleic acids in Tetrahymena during vegetative growth and conjugation. In Biology of Tetraliyntena (ed. A. M. Elliott), pp. 287-306. Stroudsburg, Pa.: Dowden,
Hutchinson & Ross.
NANNEY, D. L. (1953). Nucleo-cytoplasmic interaction during conjugation in Tetrahymena.
Biol. Bull. mar. biol. Lab., Woods Hole 105, 133-148.
NANNEY, D. L. (1959). Genetic factors affecting mating type frequencies in variety 1 of
Tetrahymena pyriformis. Genetics, Princeton 44, 1173-1184.
NANNEY, D. L. (1964). Macronuclear differentiation and subnuclear assortment in ciliates. In
The Role of Chromosomes in Development, 23 rd Symp. Soc. Study of Development and Growth
(ed. M. Locke), pp. 253-273. New York: Academic Press.
NANNEY, D. L. (1968). Ciliate genetics: patterns and programs of gene action. A. Rev. Genet.
AMMERMANN,
2, 121-140.
D. L. & CAUGHEY, P. A. (1953). Mating type determination in Tetrahymena pyriformis.
Proc. natn. Acad. Sci. U.S.A. 39, 1057-1063.
NANNEY, D. L., CAUGHEY, P. A. & TEFANKJIAN, A. (1955). The genetic control of mating type
potentialities in Tetrahymena pyriformis. Genetics, Princeton 40, 668—680.
NANNEY, D. L. & DOERDER, F. P. (1972). Transitory heterosis in numbers of basal bodies in
Tetrahymena pyriformis. Genetics, Princeton 72, 227-238.
NANNEY, D. L., REEVE, S. J., NAGEL, J. & DEPINTO, S. (1963). H serotype differentiation in
Tetrahymena. Genetics, Princeton 48, 803-813.
NILSSON, J. R. (1970). Suggestive structural evidence for macronuclear subnuclei in Tetrahymena pyriformis. J. Protozool. 17, 548-551.
NOBILI, R. & LUPORINI, P. (1967). Maintenance of heterozygosity at the mt locus after autogamy in Euplotes minuta (Ciliata, Hypotrichida). Genet. Res. 10, 35-43.
PLOEM, J. S. (1967). The use of a vertical illuminator with interchangeable dichroic mirrors for
fluorescence microscopy with incident light. Z. tuiss. Mikrosk. 68, 129—142.
RAY, C , Jr. (1956). Meiosis and nuclear behavior in Tetrahymena pyriformis. J. Protozool. 3,
88-96.
RAY, C , Jr. (1958). Tetraploidy in TetraJiymena pyriformis. Proc. Xtli Int. Congr. Genet. 2, 229
(Abstr.). University of Toronto Press.
NANNEY,
Tetrahymena micro- and macronudei
489
RUCH, F. (1970). Principles and some applications of cytofluorometry. In Introduction to
Quantitative Cytochemistry, vol. 2 (ed. G. L. Wied & G. F. Bahr), pp. 431-450. New York:
Academic Press.
SCHENSTED, I. V. (1958). Model of subnuclear segregation in the macronucleus of ciliates. Am.
Nat. 92, 161-170.
SONNEDORN, T . M. (1954). Patterns of nucleocytoplasmic integration in Paramecium. Caryologia 6 (Suppl. i), 307-325.
SONNEBORN, T. M. (1974). Genetics of Tetrahymena pyriformis. In Handbook of Genetics,
vol. 2 (ed. R. C. King), in press.
THOMAS, C. A., Jr. (1971). The genetic organization of chromosomes. A. Rev. Genet. 5, 237256.
WOLFE, J. (1967). Structural aspects of amitosis: a light and electron microscope study of the
isolated macronuclei of Paramecium aurelia and Tetrahymena pyriformis. Chromosoma 23,
59-79WOLFE, J. (1973). Conjugation in Tetrahymena: the relationship between the division cycle and
cell pairing. Devi Biol. 35, 221-231.
WOODARD, J. E., KANESHIRO, E. & GOROVSKY, M. A. (1972). Cytochemical studies on the
problem of macronuclear subnuclei in TetraJiymena. Genetics, Princeton 70, 251-260.
WOODARD, J., WOODARD, M., GELBER, B. & SWIFT, H. (1966). Cytochemical studies of conjuga-
tion in Paramecium aurelia. Expl Cell Res. 41, 55-63.
{Received 23 July 1974)
490
F. P. Doerder and L. E. De Bault
Fig. 7. Cell a: normal cell with 1 micronucleus and 1 macronucleus.
Fig. 8. Pair b: initiation of micronuclear meiosis. After the pairing of sexually reactive
cells, the micronucleus moves from its notch in the macronucleus and begins to
elongate to form the peculiar crescent. Pair c: beginning of crescent formation.
Fig. 9. Pair d: crescent (or parachute) stage. The crescent condenses to reveal 5 bivalents, which then proceed into metaphase of meiosis I.
Fig. 10. Pair «: first prezygotic division (meiosis I) products.
Fig. 11. Pair^: uncondensed products of second prezygotic division (meiosis II).
Fig. 12. Pair/: initiation of second prezygotic division. Pairs h and »: condensed
products of second prezygotic division. Note in pairs h and i that one nucleus in each
conjugant is at the attachment membrane between the cells.
Fig. 13. Pair _;': metaphase of third prezygotic division. This and all subsequent
nuclear divisions are mitotic. Relics of the second prezygotic division may be seen in
the anterior part of each cell.
Tetrahymena micro- and macronuclei
492
F. P. Doerder and L. E. De Bault
Fig. 14. Pair k: condensed products (gametic nuclei) formed by third prezygotic
division. The nuclei in the anterior-most part of the pair are the migratory nuclei
which are exchanged prior to fertilization. Two small relics of the second prezygotic
division persist in the posterior part of each cell.
Pair /: fertilization nucleus (synkaryon) at metaphase of first postzygotic division.
Fig. 15. Pair n: products of synkaryotic division. Note that they are larger than
gametic nuclei (pair m) just prior to fertilization.
Fig. 16. Pair o: stage 1 of macronuclear development. The products of the first postzygotic division align in the interior margin of each pair and divide to produce 4 nuclei.
The 2 nuclei which reach the anterior part of the cell become macronuclear Anlagen;
those which move to the posterior are new micronuclei. The morphology of the old
macronucleus has not changed.
Fig. 17. Pair£: stage 2 of macronuclear development. Macronuclear Anlagen have
begun to enlarge; the 2 new micronuclei in the posterior part of each cell are small
and round. One relic persists in each conjugant of this pair; often no relic nuclei are
seen at this stage.
Fig. 18. Pair q: beginning of stage 3 of macronuclear development. The micronuclei
are migrating to a position between the developing Anlagen. The old macronucleus,
which is now condensed, moves to the posterior part of the cell.
Fig. 19. Pair r: stage 3 of macronuclear development.
Fig. 20. Pair r: stage 3 of macronuclear development.
Cell s: an exconjugant in which 1 micronucleus has undergone autolysis. Usually
the old macronucleus has also disappeared. This stage is later than cell t.
Cell t: an exconjugant after completion of conjugation. This cell loses 1 micronucleus and the old macronucleus before division.
Cell u: presumably a caryonide, in the second cell cycle after conjugation.
Tetrahymena micro- and macronuclei
493
CEL
17