Temperature Effects on Anaphase Chromosome Movement in the Spermatocytes of Two Species of Crane Flies {Nephrotoma Suturalis Loew and Nephrotoma Ferruginea Fabricius)

J. Cell Sci. 39, 29-52 (1979) 29 Printed in Great Britain © Company of Biologists Limited

TEMPERATURE EFFECTS ON ANAPHASE CHROMOSOME MOVEMENT IN THE SPERMATOCYTES OF TWO SPECIES OF CRANE FLIES {NEPHROTOMA SUTURALIS LOEW AND NEPHROTOMA FERRUGINEA FABRICIUS)

CATHERINE J. SCHAAP AND ARTHUR FORER Biology Department, York University, Dovmsview, Ontario MJj, 1P3, Canada

SUMMARY Using phase-contrast cinemicrography on living crane fly (Nepkrotoma suturalis Loew and Nephrotoma ferruginea Fabricius) spermatocytes, we have studied the effects of a range of temperatures (6-30 °C) on the anaphase I chromosome-to-pole movements of both autosomes and sex chromosomes. In contrast to previous work we have been able to study chromosome-to- pole velocities of autosomes without concurrent pole-to-pole elongation. In these cells we found that the higher the temperature, the faster was the autosomal chromosome movement. From reviewing the literature we find that the general pattern of the effects of temperature on chromosome movement is similar whether or not pole-to-pole elongation occurs simultaneously with the chromosome-to-pole movement. Changes in cellular viscosities calculated from measurements of particulate Brownian movement do not seem to be able to account for the observed velocity differences due to temperature. Temperature effects on muscle contraction speed, flagellar beat frequency, ciliary beat frequency, granule flow in nerves, and chromosome movement have been compared, as have the activation energies for the rate-limiting steps in these motile systems: no distinction between possible mechanisms of force production is possible using these comparisons. The data show that even the different autosomes within single spermatocytes usually move at different speeds. These velocity differences cannot simply be related to chromosome size as the autosomes are visually indistinguishable. The sex chromosomes start their anaphase poleward movement after that of the autosomes, and move more slowly (by a factor of about 4), but their velocities appear to be affected by temperature in the same fashion as those of the autosomes. The interval between the onset of autosome anaphase and sex chromosome anaphase is also affected by temperature: the higher the temperature, the shorter the interval between the 2 stages. We have observed abnormalities in sex chromosome segregation, which may be due to temperature, but have not determined what the exact temperature shift conditions are that cause these abnormalities.

INTRODUCTION The mechanisms which produce the movement of chromosomes during anaphase are still unknown, and the hypotheses are subject to much debate (for reviews see Schrader, 1953; Gruzdev, 1972; Nicklas, 1975). Anaphase movement can be sub- divided into 2 distinct processes which may or may not occur simultaneously: Send correspondence to C. J. Schaap. 3 CEL39 30 C.J. Schaap and A. Forer chromosome-to-pole movement and spindle pole-to-pole elongation. These 2 processes may have different mechanisms (e.g. Ris, 1949; Oppenheim, Hauschka & Mclntosh, 1973; McDonald, Pickett-Heaps, Mclntosh & Tippit, 1977). To under- stand better the mechanisms of anaphase chromosome movement, workers have tried to alter chromosome-to-pole movement, or pole-to-pole elongation, or both, by means of experimental treatments including altered temperature, increased pressure, addition of drugs, ultraviolet microbeam irradiation, and micromanipulation of chromosomes (e.g. Inoue", 1952; Taylor, 1959; Inoue", 1964; Forer, 1965; Nicklas, 1973; Oppenheim et al. 1973; Inoue & Ritter, 1975; Salmon, 1976). These studies have led to the formulation of several models for anaphase chromosome movement (e.g. Mclntosh, Hepler & Van Wie, 1969; Bajer, 1973; Forer, 1976; Inoue, 1976; Margolis, Wilson & Keifer, 1978). Each model has faults, however, and there is no universally accepted model for anaphase chromosome movement. This paper represents the first part of a study of the effects of temperature on anaphase chromosome movement in the first meiotic division of spermatocytes of crane flies {Nephrotoma suturalis and Nephrotoma ferruginea). There are several factors which make these animals suitable for studies of anaphase: (1) there are only 3 pairs of autosomes in large cells, making it possible to follow the movement of individual chromosomes; and (2) the major part of the chromosome-to-pole movement precedes the pole-to-pole elongation (Forer, 1964, 1965, 1966). Thus we are able to measure the effect of temperature on chromosome-to-pole movement separately from the effect of temperature on pole-to-pole elongation, which was not possible in previous studies (Barber, 1939; Ris, 1949; Fuseler, 1973, 1975). In this paper we report the effects of a range of temperatures on anaphase velocities in crane-fly spermatocytes: anaphase chromosome-to-pole velocities tend to increase with temperature within the range studied (6-30 °C) and at any given temperature chromosomes in N. suturalis spermatocytes tend to move faster than those in N. ferruginea spermatocytes.

MATERIALS AND METHODS Animals and living spermatocyte cell preparations The 2 species of crane flies (Nephrotoma suturalis Loew and Nephrotoma ferruginea Fabricius) used in these experiments are reared in the laboratory, using methods described in detail in Forer (1964). The N. suturalis stock originated from North Carolina, while the N. ferruginea stock originated from Toronto. Under our conditions spermatocytes enter prometaphase of the first meiotic division about 10-12 days after the male larvae enter the fourth instar stage. The testes are dissected out of the appropriate larvae while the larvae are immersed under oil (Halocarbon Oil series 10—25: Halocarbon Products Corporation, Hackensack, New Jersey, U.S.A.). The testes are then transferred to a drop of Voltalef oil (Huile 10 S: Ugine Kuhlmann, Division Plastiques; II, bd. Pershing, 75017-Paris), where the surrounding fat is carefully dissected away. (In this transfer and all subsequent steps, caution is exercised to avoid exposure of the testes to the air.) Testes are then placed under Voltalef Huile 10 S on the bottom of a holder designed to fit the temperature-control slide. The testes are punctured and the cells spread out onto the glass surface of the bottom of the holder. After the temperature-control slide is mounted on the microscope stage, the preparation is scanned and a flat cell in metaphase I is located. In our experiments, all cells were at the experi- Temperature and anaphase movement 31 mental temperature at least 10 min before the start of anaphase. Cells were photographed from metaphase until at least the end of anaphase. Pictures were taken at intervals of 20—60 s depending on the temperature (e.g. 60 s at 10 °C or 20 a at 25 °C).

Temperature-control slide and supporting apparatus The temperature-control set up is similar to that used by Stephens (1973). Modifications to the Stephens system were however necessary because we used an inverted microscope. We used a Zeiss 40 x (N.A. 075) phase-contrast water-immersion lens with a long working distance. To control the temperature of the coolant, 50 % ethanol, we used a Lauda K2/RD (Brinkmann Instruments) constant temperature circulator: this both brings the solution to the proper temperature and circulates it. Thermistor probes (Teflon Probe, Part no. 44104) con- nected to a YSI Tele-Thermometer (Model 43 TZ Yellow Springs Instrument Co., Inc., Yellow Springs, Ohio, U.S.A.) were used to monitor the temperature of coolant going into and coming out from the temperature-control slide. When care was taken to dislodge air bubbles from the thermistors, the incoming and outgoing temperatures differed by only about 0-5 °C or less. The Lauda circulator temperature varied by only ±o-oi to ±0-03 °C during the course of an experiment.

Time-lapse apparatus and photography The experiments were recorded on 16-mm film (Kodak Plus X Negative Film 7231). We used a Bolex (H-16-J) with a time-lapse camera drive controlled by a time-lapse control panel (R. J. Matthias and Associates, Houston, Texas, U.S.A.). The negatives were developed with Diafine in a Cramer automatic processor (Sarasota, Florida, U.S.A.). Positive working prints were made at a commercial laboratory.

Film analysis and statistics The films were analysed frame by frame using an Athena projector (Model 224 by L-W Photo Inc., Van Nuys, California, U.S.A.). The working prints were projected onto a table via a mirror on a retort stand in front of the projector. For the analysis the final magnification of the cells was 2000 times. The magnification of the film was calibrated using photographs of a stage micrometer. To determine the velocities of the separating autosomal chromosome pairs, we measured interkinetochore distances parallel to the pole-to-pole axis in successive frames of the film as anaphase proceeded. These measurements were plotted with respect to time, and a typical result is given in Fig. 1. The chromosome separation velocity was determined by standard least mean squares linear regression line (Sokal & Rohlf (1969), p. 419) through the points from the linear portion of the distance-v.-time graph (e.g. the points between the arrows in Fig. 1). To determine the velocities by which the sex chromosomes separate the distance between the farthest kinetochores was measured in successive frames. Distance-v.-time graphs were plotted and the slopes of the initial steep parts of the graphs were calculated by linear regression. The slopes of the distance-v.-time curves are the velocities at which the autosomes or sex chromosomes separate. These will be referred to as the chromosome velocities in the results. Since there is no pole-to-pole elongation until at least the end of autosomal anaphase, the chromosome velocities calculated for autosomes are directly proportional to the chromosome-to- pole velocities. Since the 2 separating autosome half-bivalents usually move at the same speed, then their velocities to the poles are half the calculated autosomal chromosome velocities. On the other hand, the pole-to-pole distances often increase as the sex chromosomes move poleward, so that the calculated sex chromosome velocities include pole-to-pole elongation as well as chromosome-to-pole movement. By analysis of covariance (Sokal & Rohlf (1969), pp. 448-458) we compared the 3 autosome velocities within individual cells. Thus we determined the frequency with which an autosome pair within an individual cell has a velocity different from the other autosome pairs. Viscosity values were computed from measurements of the movement of cytoplasmic particles. Equations and techniques developed by Fiirth (1917, 1930) and Pekarek (1930) enable one to compute the viscosity of a Newtonian fluid from measurements of Brownian movement. 3-2 C. J. Schaap and A. Forer

14 O

5 8 oo°°c?

12 16 20 24 28 32 Time, min Fig. i. Typical graph of interkinetochore distances of autosomes (O, D, and <]) and sex chromosomes (•) throughout anaphase. (N. ferruginea cell at25 °C). Chromosome- separating velocities were determined by calculating the slopes using the points on the graph between the arrows. The time labelled 'o' on the abscissa indicates the start of autosomal anaphase. The ordinate is the distance between chromosomes in microns.

The technique was later used by Taylor (1965), who extensively discusses the limitations of the technique. The equation we used (Pekarek, 1930) is 1] = (kTt)/(2naLhi) (?/ is viscosity, k is the Boltzmann constant, T is the absolute temperature, t is the time it takes the particle for n first crossings of parallel grid lines spaced L units apart, and a is the particle radius). We controlled T, and we measured a and corresponding n and t values; from these we calculated •>/ at various temperatures. Fig. 2 shows diagrammatically how the measurements were made. As many 'first crossings' as possible were counted for each particle, but these were limited due to particles leaving the focal plane, or single particles being mixed with other particles.

RESULTS General description of first meiotic division The first meiotic division in N. suturalis spermatocytes appears to be similar to that in N. ferruginea spermatocytes (Figs. 3, 4) in the following respects: there are 3 pairs of autosomes and 2 sex chromosomes; all chromosomes can be seen clearly at metaphase in a flat cell; the 3 bivalents (autosome pairs) begin anaphase together, while the sex chromosomes remain at the equator; the sex chromosomes begin to move Temperature and anaphase movement 3f

Fig. 2. Schematic drawing of a hypothetical particle moving randomly in the cytoplasm by Brownian movement.' Crossings' are counted only the first time a particle crosses a particular grid line, until that particle has crossed a second grid line. After that if it crosses the first grid line again, this may be counted as a crossing. The path of the particle is represented by the 'jagged' series of lines. The particle starts at position l which is the first crossing. Subsequent crossings are numbered 2-4, L is the distance between grid lines and a is the particle radius (redrawn from fig. 1 of Pekarek, 1930).

Figs. 3, 4. Series of pictures from the anaphase of N. suturalis spermatocyte (Fig. 3) and N. ferrugi-nea spermatocyte (Fig. 4) at 15 CC. The times, in min from the start of anaphase were: 3A, B, C, D, — 22, +5, + io, +48; and 4A, B, c, D, — 10, +5, +21, + 55, respectively. The bars are 10/Jm apart. 34 C. J. Schaap and A. Forer polewards only when the autosomes have completed or almost completed their poleward movement; there is no pole-to-pole elongation until the autosomes are near the poles; there may be pole-to-pole elongation during the poleward movement of the sex-chromosomes; and the separating autosomes attain their maximum separating velocities almost immediately (the time required depends upon the temperature).

A _ B 3-2 19)

3 1 l I||S 30 ~ u

2-9 cell CN 28 27 2-6 25 2-4 • , 23 2-2 cells 3) 21 un a s 20 1 17) m/m = U "tn a. 1 9 eel gc ~"~ O 1 8 == CN 8 to part 1-7 Si v— 1-6 • o

1-5 CO CN - "53 1-4 • — "53 1 3 - & tn in o 1 2 - -

Avera 3. * 3) 1-1 11 cell T n — 1 0 _ _ o ^__ __ CO 0-9 - o ^ - II eel 08 _ • 7 •- _ ~z - _

07 eel i

i = (01 06 - 8 8 - II 05 - T * 0-4 - i 1 - 1/ 8

0 3 (/) ell 1 J II 0-2 - 0 1 - — ~X 5 J 1 i i 1 1 • I i 10 15 20 25 30 5 10 15 20 25 Temperature,°C Fig. 5. Effect of temperature on the average velocities of separation for N. suturalis (A) and N.ferruginea (B). The vertical bars represent the standard deviations from the means. (All of the standard deviations were calculated assuming a normal distribution about the mean using n— 1 degrees of freedom as described in Sokal & Rohlf (1969), pp. 53-55.) The number of cells and the total number of autosome separating velocities (n) which are part of the mean are also indicated. We thank A. Barkas for permission to analyse his films in order to gain additional data for N. suturalis, which we have incor- porated with our own. (Note that for various technical reasons it was not always possible to get data from all 3 autosome pairs in each cell. A typical reason for this would be that not all of the chromosome pairs were in the same focal plane.) Temperature and anaphase movement 35 More detailed descriptions of meiosis in crane fly spermatocytes can be found in Bauer (1931), Wolf (1941), Dietz (1956, 1959), Bauer, Dtetz & Robbelen (1961), Dietz (1963), Forer (1964), Ullerich, Bauer & Dietz (1964), Forer (1965, 1966), Behnke & Forer (1966) and LaFountain (1972).

Effect of temperature on autosomal chromosome velocities As measured from our phase-contrast micrographs, the pole-to-pole distances do not appear to increase until the autosomes are near the poles. Phase-contrast microscopy does not allow one to locate the positions of the poles accurately, to within more than a few microns, but data from polarization microscopy confirm that pole-to-pole distances do not increase until the autosomes are near the poles (see illustrations and graphs in Forer, 1965, 1966, 1976 or Dietz, 1969). The pole-to-pole distances did not appear to increase during autosomal anaphase at any of the temperatures studied. Thus the autosomal chromosome velocities which we calculated really are proportional to velocities of chromosome-to-pole movements: increase of the pole-to- pole distance does not contribute to autosome velocity at this time. Our results on autosomal velocities at different temperatures are illustrated in Fig. 5. These figures demonstrate clearly that as the temperature increases the autosomal velocities also increase. However, at any temperature the range of possible velocities is very large, and these velocities often overlap with those seen at other temperatures. A wide range of autosomal velocities was also observed by Forer (1964, 1965) for N. suturalis studied at room temperatures, though some of that spread may have been due to variations in room temperature. There are some differences between the cells of the 2 species. At any given temperature autosomes of N. suturalis spermatocytes tend to move faster than those of N. ferruginea spermatocytes. Using Student's£test(Sokal&Rohlf(i969),pp. 143-145), these differences are significant (a = 0-05) at 20, 15, and 10 °C but not at 25 °C. The temperature ranges are also different. Autosome movement in N. ferruginea spermatocytes is limited to a range between about 6 and 25 °C: outside this the cells die, spindles collapse and/or the chromosomes clump. The limits for N. suturalis spermatocytes have not been determined, but the upper limit is at least 30 °C. Our N. ferruginea culture derives from flies caught in Toronto while the N. suturalis culture derives from flies caught in North Carolina, which has a much warmer climate than Toronto. This may have an effect on the range of temperatures at which normal divisions can occur. The different autosomes (half-bivalents) within a cell do not necessarily move polewards at the same velocity. The chromosome velocities within cells were compared statistically by an analysis of covariance (Sokal & Rohlf (1969), pp. 448-458). This test allowed statistical comparison of the velocities of autosomes within a cell, while at the same time taking into account variations inherent in the distance-v.-time measurements. In other words, were the differences in velocities calculated from regression lines statistically significant (a = 0-05) or were they merely due to errors in measuring? In the usual case it is clear that, in both species, the autosome pairs in any given cell do not all move polewards at the same speed, as illustrated in Fig. 6. C. J. Schaap and A. Forer

7 6 5 4 3 2 1 A'B'C'D'E1 FLnVB'C'D'E, 1 FLEVB'C'D'Eb1 'A'B'C'D'Ecofb' YB'C'D'En m1 , 'A'B'C'D'Eam' 'A'B'CD'E h B ° 9 i 8 6 5 4 3 2 1 A'B'C'D'E1 Bin'A'B'C'D'E. ' ru.f'A'BCDEi fABCDi EBE - A BCD E ABCDE ABCDE 10° 15° 20° 25° 30° Total Fig. 6. Summarized data of analysis of covariance on the separating velocities of autosomes within the same cell (A, N. suturalis; B, N.ferruginea). As described in the text this type of data analysis has allowed us to determine whether one or more pairs of separating autosorr.es actually have the same separating velocities. Since the autosome pairs cannot be visually differentiated they are arbitrarily named aa', bb'', and cc' as one moves from left to right along the metaphase plate of the cell (bb' is always the centre autosome pair). The separating velocity of aa' is designated as a, of bb' as b and of cc' as c. The shaded boxes mean that there were only 2 measurable autosome pairs in that particular cell. (The data from the films of A. Barkas were not included in the analysis of covariance and thus also not in Fig. 7 and Fig. 8, but we have no reason to believe they would significantly alter the results.) Categories: A, a = b or b = c; B, a = c; C, a — b = c; D, a 4= b 4= c; E, statistically strange, e.g. a = b, b = c, a 4= c.

Usually at least one pair moves poleward at a different speed from the others. One might argue that since the outside autosomes must travel a greater distance, they should travel faster in order to get to the poles at the same time. One can rule this out, however, because if 2 of the 3 autosome pairs move at the same speed, these are more likely to be one middle autosome pair and an outside one, as compared to 2 autosomes positioned on the outside (Fig. 6). The variabilities of autosome velocities appear to be independent of temperature (Fig. 6); therefore these differences are not merely temperature effects. This is corroborated by comparing the percentage differences, higher velocity — lower velocity lower velocity x ioo, between chromosome pairs considered statistically different: the percentage differences do not vary with temperature (Fig. 7), which they would do if the temperature changes caused different autosomes in the same cell to move with different velocities. On the other hand the differences in velocities between autosomal pairs in the same cell are dependent on chromosome velocities (Fig. 8), and the shapes of these curves are very Temperature and anaphase movement 37

120 A B

110 - -

>• 100 - -

I 90 - - 80 f - - CO i70 - - £ 60 - -

| 50 _ - 1

•Si 1 - I - 1

•c° 30 3 1 2 20

10 -

1 1 1 1 1 1 0' 5 10 15 20 25 30 0 5 10 15 20 25 30 Temporature, °C Fig. 7. The mean percentage differences between the velocities found to be statistically different by the analysis of covariance are represented in relation to temperature (A, N. suturalis; B, N.ferruginea). The vertical bars represent the standard deviations from the means. UILU 1-1 ~ A - B "e 10 _ a. > 09 _ ISA

08 - i - O) O7 _ _ e i 06 - - c1 0-5 - 1

04 - •I

1O3 - - 0-2

i ; to 01 $ ; {'. c I $ 3 1 FIII i 0 5 10 15 20 25 30 0 5 10 15 20 25 30 Temperature, °C Fig. 8. The relationship of mean absolute differences in autosome separating velocities within cells to temperature (A, N. suturalis; B, N. ferruginea). The vertical bars represent the standard deviations from the means. similar to the curves of autosome velocity versus temperature (Fig. 5). Thus, the differences in autosome velocities within cells are proportional to the autosome velocities, and are not temperature-dependent effects. In summary, increased temperature results in increased autosome velocities; at any given temperature autosomes in N. suturalis spermatocytes tend to separate faster than those of N'. ferruginea spermatocytes; and for both species temperature does not 38 C. J. Schaap and A. Forer appear to control the observed differences in autosome velocities within individual cells.

Temperature effects on sex-chromosome velocities The sex chromosomes do not start their anaphase poleward movements until the autosomes have reached or almost reached their respective poles. Although pole-to- pole elongation as well as chromosome-to-pole movement may occur simultaneously, the sex chromosomes move much more slowly than the autosomes (Fig. 9), often by a factor of about 3 or 4. The shapes of the autosome and sex chromosome velocity- v.-temperature curves are comparable (cf. Fig. 9 with 5). Thus temperature appears to affect the 2 types of movement similarly, and we would expect to be able to predict the approximate separating velocity of the sex chromosomes if the average autosome- separating velocity for that cell were known. This is roughly true, as illustrated in Fig. 10: a plot of autosome velocity-v.-sex chromosome velocity in the same cell is indeed linear, but there is considerable scatter. (In Fig. 10 there is one point which seems particularly out of place; we have no explanation for this apparent misfit.)

A 8 0-6 A

05 - -

0-4 - -

A A A

0-3 - - A A A

A * "A" 0-2 - - A

A A 0-1 - - I t

1 1 1 1 1 1 1 1 1 1 1 10 15 20 25 30 0 10 15 20 25 Temperature, °C Fig. 9. The effects of temperature on the velocity of sex chromosome separation of N. suturalis (A) and N. ferruginea (B). (Note that we could not get the sex chromosome- separation velocity for each cell studied: for example if the 2 sex chromosomes \s'ere not in the same focal plane.)

Relative timing of sex chromosome movements In the first meiotic division of crane-fly spermatocytes the anaphase of the autosomal bivalents precedes that of the sex chromosomal univalents. How are the 2 anaphases linked? What is it that controls the time lag between them? Temperature affects the mechanisms which control the lag time between the 2 anaphases (Fig. 11): in both species the time between the start of autosomal anaphase and the start of the sex chromosome segregation depends on temperature. This time difference seems to Temperature and anaphase movement 39 depend more on temperature in N.ferruginea than in N. suturalis: in the former it is clear that the lower the temperature the longer it takes for the sex chromosomes to start their anaphase, but the trend is not as pronounced in N. suturalis (Fig. 11). However, more data are needed in order to see whether or not there is a difference between the spermatocytes of the 2 species in this regard.

0-7 -

6 - £ °' • .2 E 0-5 • o =i 0-4 • • 11 0-3 • E 2 •• 2 a 0-2 — A U W g oO'1 © 1 1 1 1 1 1 1 1 1 1 1 t 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 t t t 1 05 10 1-5 20 2-5 30 3-5 Average autosomal velocity of separation, ^m/min Fig. 10. A comparison of sex chromosome velocity of separation with the autosome velocity of separation in the same cell (N. suturalis, # ; N.ferruginea, A): the autosome velocities are averages of up to 3 velocities per cell, or are single velocities (when we had only one autosome velocity for that cell).

A 100 -

- A 50 - o

e A - o o o o • O A o A o o O

10 - o A A O - O cx> 1 1 1 1 1 1 10 15 20 25 30 Temperature, °C Fig. 11. The log of the time difference between the start of autosomal anaphase and the start of the sex chromosome segregation is plotted on the ordinate and temperature is plotted on the abscissa (N. suturalis: individual cells, O; mean for a given temperature, • ; N. ferruginea: individual cells, A; mean for a given temperature, A). C. J. Schaap and A. Forer

A 34 32 30 0 28

26 A 24

22 o 1 20 A 0 * 18 -

6 14 A 0 > 12 - o 10 - o o A 8 0 A 6 - o

4 - o o

2 - I 1 1 1 1 1 10 15 20 25 30 Temperature, °C Fig. 12. Cytoplasmic viscosity, on the ordinate, is plotted versus temperature, on the abscissa (N. suUiralis, O'» N. ferruginea, A)-

38 -

36 -4 34 - 32 - 30 - 28 - 26 i ! 24 - 22 - '• 20 - - i 18 - . 16 - ' 14 - 12 - 10 - 8 - 6 - 4 - I 1 I I I I I I I I I I I I I I I I I I ) 0 0-5 10 15 20 2-5 30 Mean velocity of autosome separation, yni Fig. 13. The measured cytoplasmic viscosity within a cell (ordinate) is compared with the average velocity of autosome separation within that cell (abscissa) (AT. suturalis, # ; N.femiginea, A). Temperature and anaphase movement 41

Effect of temperature on cytoplasmic viscosity In order to see if the observed differences in velocity might be due to temperature- induced viscosity changes, cytoplasmic viscosity was calculated from measurements of Brownian movement velocities of particles in the cytoplasm. The particles measured were about 0-5-1 -o /im in size. The areas studied in these measurements were not chosen at random, but rather had relatively few particles and were away from inter- ferences of the spindle area and the cell membrane. Because it was very difficult to find suitable cytoplasmic particles and because there is extensive literature on similar measurements in other cells relatively few measurements were made. As expected, the viscosity increases as the temperature decreases (Fig. 12). The range of viscosities (about a factor of 6-7) is not as large as the range in velocities (about a factor of 8-10). We have compared viscosities directly with autosome velocities measured in exactly the same cells (Fig. 13), and while viscosity might influence chromosome velocities at lower velocities (less than about o-6 /

DISCUSSION Autosomal velocities We have shown that autosomal velocities in anaphase depend on external temperature: the higher the temperature the faster the movement (Fig. 5). In previous work, Barber (1939), Ris (1949) and Fuseler (1973, 1975) have also shown that chromosome velocities in anaphase, in various cell types, depend on temperature, and we have summarized these data in Fig. 14. While these earlier data are similar to ours in indicating that anaphase velocities increase as temperature increases, the results are not directly comparable to ours, because our measurements are of chromosome-to-pole velocities. The previous measurements included both chromosome-to-pole velocity and velocity of pole-to-pole elongation. In the previous experiments, for example, it was possible that pole-to-pole elongation was greatly affected by temperature while chromosome-to-pole movement was unaffected, or the converse. As far as we know, ours are the first data which directly measure the effect of temperature on chromosome-to-pole motion during anaphase, separately from pole-to-pole elongation. Other previous workers described the effect of temperature on anaphase, but these data deal primarily with the duration of anaphase at different temperatures (Fig. 15). These data might be compared with our data on velocity by assuming that total anaphase distances are constant: in that case velocities would be inversely proportional to anaphase durations. We have made such calculations from published data, and we summarize these data in Fig. 15. These recalculated data agree with ours in showing that anaphase chromosome velocities increase as temperature increases, but here too, these data are of possible mixtures of both pole-to-pole elongation and chromosome- to-pole motion. C. J. Schaap and A. Forer

50 - a

- A A 3 • , A , A A • -

A,'

1 0 -

• c 0 •

A — A '0 B 0-5 — A ' •

— A f o 6 cf o/ i

0-1 - •

0-05 • /

n-m 1 i i i i 10 15 20 25 30 35 Temperature, °C Fig. 14. Mean anaphase velocities of separation from each of several systems are plotted on the log scale of the ordinate to show their relationship to temperature (abscissa) and to compare these data directly with ours (AT. suturalis: autosomes, A; sex chromosomes, V; N. ferruginea: autosomes, •; sex chromosomes, |£; Barber (1939) Tradescantia virginiana, A; Ris(i949) Chorthophaga viridifasciata, •; Fuseler (1975) (estimated from fig. 7), Asterias forbesi, O, Tilia americana, •). The broken lines show the trends in velocity changes due to temperature, of both autosomes (A) and sex chromosomes (B). Temperature and anaphase movement 43

10 r

0-1

0-01

0001 10 15 20 25 30 35 40 45 Temperature, °C Fig. 15. Review of data from the literature on the effects of temperature on the length of anaphase in several systems. Since we assume that the duration of anaphase is inversely proportional to anaphase velocity, we have plotted the inverse of the time of anaphase on a log scale on the ordinate and compared it with temperature on the abscissa (Laughlin (1919), Allium cepa, @; Bucciante (1927), chick fibroblasts, •; Makino & Nakahara (1953), Yoshida sarcoma cells, A, MTK sarcoma I cells, •; Agrell (1958), Echinus esculentus, O,Psammeckinus miliaris, <^>; Evans & Savage (1959), Vicia faba, (E; Dettlaff (1963), Acipenser gilldenstadti colckicus V. Marti, #; Lopez- Saez, Gim£nez-Martfn & Gonzalez-Ferndndez (1966), Allium cepa, A; Stephens (1972), Strongylocentrottis droebachiensis, <*). 44 C. J. Schaap and A. Forer Might the temperature dependence of chromosome-to-pole movements give a clue to the mechanism of force production ? Some suggest that the forces for chromosome- to-pole motion are produced by muscle proteins such as actin (e.g. Forer, 1976), while others suggest that the forces arise from the action of dynein (e.g. Mclntosh et al. 1969; Mclntosh, 1974; Margolis et al. 1978); if the temperature dependence of anaphase chromosome movement were like muscle contraction but not ciliary beating, for example, this might suggest that chromosome movement was due to the action of muscle proteins, or the converse. We have looked at the literature describing the effects of changing temperature on various motile processes, such as muscle contraction, particle transport, ciliary beating and flagellar beating, and we summarize these data in Fig. 16. These data show that these processes occur faster at higher than at lower temperatures, but actin-based mechanisms (muscle, movements of cytoplasmic particles) cannot be distinguished from dynein-based mechanisms (cilia, flagella). Thus the data do not suggest that the force for chromosome movement arises from one mechanism and not the other. Another way to look at data on temperature effects is to calculate activation energies (from the Arrhenius equation) for the rate-limiting step in the motile system in ques- tion. To do this one must assume that the same rate-limiting step applies at all the temperatures considered. Activation energies calculated from data on temperature dependence have been compiled in Table r. In our case, we plotted log velocity- v.-1/absolute temperature, and calculated activation energy (Ea) from the slope of the regression line. If the activation energies of Table 1 are compared with activation energies charac- teristic of enzyme-catalysed reactions (Table 2), it appears that usually the rate- limiting step of these motile systems is not enzyme-catalysed. From the literature reviewed by Lineweaver (1939) and Sizer (1943) it appears that the activation energies of enzyme-catalysed reactions are commonly below about 13 kcal per mol (e.g. see Table 2), while those of the motile systems discussed tend to be higher. Thus although enzymes are probably involved in these processes they appear not to be catalysts in the rate-limiting reaction. The rate-limiting step may however be catalysed by in- organic ions (e.g. hydrogen, calcium or platinum). Our data are also relevant to all hypotheses of force production for chromosome movement in which all chromosomes would be expected to move at the same speed in anaphase: as a general rule, they do not move at the same speed (Figs. 6-8). This implies that one cannot invoke a simple temperature-regulated microtubule assembly—disassembly equilibrium as the motive force for chromosome movement (e.g. Inoue, 1976), because that would imply that all chromosomes move at the same speed. At the least one must superpose upon a mechanism such as this, and indeed upon the other proposed mechanisms, some further control, such that some chromosomes can move at a velocity up to 50 % different from that of other chromosomes in the same cell (Figs. 6-8). These differences in velocities are not related to obvious differences in chromosomal size, because the autosomes in crane-fly spermatocytes are not distin- guishable by size. We have calculated viscosity in spermatocytes at different temperatures from Temperature and anaphase movement 45

A B 1 4 - 34 • • • 1-2 • - 30

| 1-0 • - 26 • E. 0-8 - • - 22 • I 0-6 • - 18 • 0-4 • - 14 • • 0-2 • - 10 0 C D 2-8 • • • 2-4 • •

| 2-0 • 5 C o c :•. H4 •• • • • • • 08 2 ir 0-4 - 1

0 E F 28 - 7

, 24 • 6 in ! 20 5 S J 4 CD • .Q 4 ; 16 £

CD • [i2 o j C • S • • 8 2 x • 1 4 - • m ° o i i i i 1 I 1 1 0 5 10 15 20 25 30 35 40 0 5 10 15 20 25 30 35 40 Temperature,1 °C 1 1 1 1 1 1 1 Fig. 16. Compilation of representative data showing the influence of temperature (abscissa) on several types of biological movement. A. Rate of ciliary movement on gills of Mytilus as determined by the rate of particle transport (from the data of Gray, 1924). B. Ciliary beat frequency of Stentor polymorphic (from the data of Sleigh, 1956). C. Rate of retrograde intra-axonal organelle transport in Rana catesbeiana (redrawn from Forman el al. 1977). D. Rate of fast axonal amino acid transport in Anodonata cygnea (redrawn from Heslop & Howes, 1972). E. Flagellar beat frequency of Strigomonas oncopelti (from the data of Holwill & Silvester, 1965). F. Rate of heart beat in rabbit (•) and dog (O) (from data of Frank, 1907).

CEL 39 Table I. Activation energy values of some motile systems

System Process Reference Nephrotma suturalis Autosome anaphase movement Sex chromosome anaphase movement Autosome anaphase movement Sex chromosome anaphase movement Anaphase chromosome movement (Calc. from max. velocity) Tilia americana Anaphase chromosome movement Asterias forbesi Anaphase chromosome movement Chorthophaga viridayasciata Anaphase chromosome movement Strigomonas oncopelti Flagellar beat frequency Stmtor polymorphus Ciliary beat frequency My tilus Rate of particulate transport by cilia on the gills Rana catesbeiana Rapid retrograde axoplasmic organelle Forman et al. (1~7)t transport Rapid anterograde axoplasmic transport in C-fibres Cat Rapid anterograde axoplasmic transport Rabbit Rapid anterograde axoplasmic transport \Cited in Cosens er al. (1976) Rana temporaria Rapid anterograde axoplasmic transport Frog Velocity of contraction of heart muscle Velocity of contraction of heart muscle Dog Cited in Crozier (1926) Cat Velocity of contraction of heart muscle Rabbit Velocity of contraction of heart muscle I Calculated from tabular data by present authors (s.E. of regression line calculated as in Sokal & Rohlf (1969) pp. 417-436). t Estimated from graphed data by present authors. Temperature and anaphase movement 47

Table 2.* Activation energies of selected chemical reactions with and without enzyme catalysts

Reaction Catalyst Ea (kcal/mol) Hydrogen peroxide None (dust) 18-0 decomposition Fe(OH)3 "1 IVInOs ^ US I~ J Colloidal Pt n-7 Liver catalase 5'5 Sucrose inversion H+ion 260 Yeast invertase us Malt invertase 130 Casein hydrolysis HC1 206 Trypsin-kinase 144 Trypsin 1 Crystalline trypsin ,- 12-0 Crystalline chymotrypsin J Ethyl butyrate H+ion 13-2 Pancrease lipase 4-2 • Partial reproduction of table i Lineweaver (1939). Activation energies for other enzyme- catalysed reactions can be found in Sizer (1943). measurements on granule movements: while viscosity does change somewhat with temperature (Fig. 12), the increase in viscosity as temperature is lowered cannot account for the reduced chromosome velocity (Fig. 13), especially at temperatures from 15 to 30 °C. The technique we used to calculate viscosities certainly has limitations (see discussions in Heilbrunn, 1928; Taylor, 1965; Heilbrunn, 1958), but our results agree with those using other methods. We have summarized in Fig. 17 results of other calculations of viscosity at different temperatures in various cells. One can see from this that some other calculations give results similar to ours. Heilbrunn (1958), in reviewing calculations of cytoplasmic viscosity at different temperatures, suggests that there are 2 general groups of cells, (1) those in which ' the protoplasm... shows a progressive decrease in viscosity as the temperature is raised', and (2) those in which 'the viscosity goes through a maximum with rising temperature'. Our results seem to fall into the first class, and we conclude that changes in viscosity are not the main reason for the observed changes in chromosome velocity. Barber (1939) reached a similar conclusion, although he himself made no viscosity measurements.

Movements of sex chromosomes Sex chromosome velocities change with temperature in much the same way as do autosomal chromosome velocities, but the sex chromosome velocities are considerably slower than the autosomal velocities, by a factor of about 4. This difference in speed might be a reflexion of the different orientations of the 2 kinds of chromosomes: the autosomes are syntelically orientated while the sex chromosomes are amphitelically 4-2 C. J. Schaap and A. Forer

20 11U 100 1 5 - 90 £ s 80 S I 1-0 70 > so I 50 tr 40 I I I I I I I I I I 1 1 I I I I 30 280 E 240 £ ? 3 200 S o 160 5

> 2 120| % 80 £ 40 0 I I I I I I I I I I I I I I I I 0

230 20 I 220. i 50' 1-5 2r S § 40 Si g 30 10 >

I 20 05 01 10 0 0 :0 5 10 15 20 25 30 35 10 15 20 25 30 35 40 ^ 20 I 15 Z 10 5 5 II 0 1_ i ' i L _l L 5 10 15 20 25 30 35 40 45 50 55 60 Temperature, °C

Fig. 17. Compilation of data on the effects of temperature on cellular viscosity. (For discussion of methods and data see Heilbrunn, 1958). A. Relative viscosity of immature eggs of Nereis diversicolor as determined by centrifugation (redrawn from Pantin, 1924). B. Relative viscosity of Cwningia eggs as determined by centrifugation (redrawn from Heilbrunn, 1924, as quoted by Heilbrunn, 1928). C. Viscosity of Spirogyra protoplasm as determined by studying particle Brownian movement in the cytoplasm (redrawn from Bass Becking, Sande Bakhuyzen & Hotelling, 1928). D. Viscosity of Arbacia eggs as determined by centrifugation (redrawn from Costello, 1934, as quoted in Heilbrunn, 1958). E. Relative protoplasmic viscosity of Ascaris megalocephala eggs as determined by centrifugation (from data of Faur6-Fremiet, 1913, as quoted by Heilbrunn, 1958). F. Protoplasmic viscosity of Amoeba dubia as determined by centrifugation (redrawn from Murphy, 1940, as quoted by Heilbrunn, 1958). G. Protoplasmic viscosity of Phaseolus multiflorus as determined by the time required for a starch gTain to fall through the protoplasm (from data of Weber & Weber, as quoted by Heilbrunn, 1958). Temperature and anaphase movement 49 orientated (Bauer, Dietz & Robbelen, 1961). That is to say, the autosomes have one chromosomal spindle fibre extending between autosome and pole; this spindle fibre shortens as the autosomal half-bivalent moves poleward in anaphase. On the other hand the sex chromosomes have chromosomal spindle fibres extending to both poles; during anaphase one spindle fibre shortens and the other one elongates as any given sex chromosome moves polewards. The different spindle fibre arrangements might give rise to different poleward velocities. Temperature affected not only chromosome velocities but also the time interval between the onset of autosome anaphase and sex chromosome anaphase (Fig. n): the higher the temperature the shorter the interval between the 2 stages. With regard to a cellular signal giving rise to sex chromosome anaphase, one might guess that sex chromosome anaphase begins after the autosomes reach the poles, or after the autosomes have moved a certain distance, but this seems not to be true: if one plots autosomal velocity (from Fig. 5) versus the time interval between autosome and sex chromosome anaphase at the same temperatures (from Fig. 11), one does not get the straight line which would be expected if sex chromosome anaphase began after the autosomes had moved a certain distance. The signal to start sex chromosome anaphase, and how temperature affects this, remain mysteries. We have observed abnormalities in sex chromosome segregation throughout these experiments: for example, a single sex chromosome sometimes moved polewards before autosomes, or together with the autosomes. Such abnormalities rarely if ever occur in cells studied at room temperature without experimental treatment (e.g. Dietz, 1969; Forer & Koch, 1973), and we presume that the temperature shifts somehow affected the sex chromosome segregation mechanism. However, we have not been able to determine what, exactly, in our procedures might be causing alterations in sex chromosome segregation.

We should like to thank Dr F. Eckhardt for aiding us in our understanding of the German viscosity papers. We are also grateful to our colleagues in the Biology Department at York University for useful discussions, Mrs D. Gunning for excellent secretarial assistance and the National Research Council of Canada for financial support.

REFERENCES AGRELL, I. (1958). The thermal dependence of the mitotic stages during the early embryonic development of the sea urchin embryo. Ark. Zool. 11, 383-393. BAAS BECKING, L. G. M., SANDE BAKHUYZEN, H. V. d. & HOTELLING, H. (1928). The physical state of protoplasm. Verh. K. Akad. Wet. (Tiveede Sectie) 25, 1-53. BAJER, A. S. (1973). Interaction of microtubules and the mechanism of chromosome movement (zipper hypothesis). 1. General principle. Cytobios 8, 139—160. BARBER, H. N. (1939). The rate of movement of chromosomes on the spindle. Chromosoma 1, 33-5°- BAUER, H. (1931). Die Chromosomen von Tipula paludosa Meig in Eibildung und Spermato- genese. Z. Zelljorsch. mikrosk. Anat. 14, 138-193. BAUER, H., DIETZ, R. & ROBBELEN, C. (1961). Die Spermatocytenteilungen der Tipuliden. Ill Das Bewegungsverhalten der Chromosomen in Translokarionsheterozygoten von Tipula oleracea. Chromosoma 12, 116—189. 50 C. J. Schaap and A. Forer BEHNKE, O. & FORER, A. (1966). Some aspects of microtubules in spermatocyte meiosis in a crane fly (Nephrotoma suturalis Loew); intranuclear and intrachromosomal microtubules. C. r. Trav. Lab. Carlsberg 35, 437-455. BUCCIANTE, L. (1927). Ulteriori richerche sulla velocita della mitosi nelle cellule coltivate 'in vitro' in funzione della temperatura. Arch. exp. Zellforsch. 5, 1-24. COSENS, B., THACKER, D. & BRIMIJOIN, S. (1976). Temperature-dependence of rapid axonal transport in sympathetic nerves of the rabbit. J. Neurobiol. 7, 339-354. COSTELLO, D. P. (1934). The effects of temperature on the viscosity of Arbacia egg protoplasm. J. cell. comp. Physiol. 4, 421-433. CROZIER, W. J. (1926). The distribution of temperature characteristics for biological processes; critical increments for heart rates. J. gen. Physiol. 9, 531-546. DETTLAFF, T. A. (1963). Mitotic dynamics of the first cleavage divisions in the eggs of sturgeon (at various temperatures) and of trout. Expl Cell Res. 29, 490-503. DIETZ, R. (1956). Die Spermatocytenteilungen der Tipuliden II. Graphische Analyse der Chromosomenbewegung wahrend der Prometaphase I im Leben. Chromosoma 8, 183-211. DFETZ, R. (1959). Centrosomenfreie Spindelpole in Tipuliden-Spermatocyten. Z. Naturf. 14 b, 749-752- DIETZ, R. (1963). Polarisationsmikropische Befunde zur chromosomeninduzierten Spindel- bildung bei der Tipulide Pales crocata (Nematocera). Zool. Anz. 26, Suppl., 131-138. DIETZ, R. (1969). Bau und Funktion des Spindelapparatus. Natunuissenschaften 56, 237-248. EVANS, H. J. & SAVAGE, J. R. K. (1959). The effect of temperature on mitosis and on the action of colchicine in root meristem cells of Vicia faba. Expl Cell Res. 18, 51-61. FAURE-FREMIET, E. (1913). Le cycle germinatif chez VAscaris megalocephala. Arclu. Anat. microsc. Morph. exp. 15, 435-757. FORER, A. ( I 964). Evidence for Two Spindle Fiber Components: a Study of Chromosome Movement in Living Crane Fly {Nephrotoma suturalis) Spermatocytes, tising Polarization Microscopy and an Ultraviolet Microbeam. Ph.D. Thesis, Dartmouth College, New Hampshire, U.S.A. FORER, A. (1965). Local reduction of spindle fiber birefringence in living Nephrotoma suturalis (Loew) spermatocytes induced by ultraviolet microbeam irradiation. J. CellBiol. 25, 95-117. FORER, A. (1966). Characterization of the mitotic traction system, and evidence that birefringent spindle fibres neither produce nor transmit force for chromosome movement. Chro- mosoma 19, 44-98. FORER, A. (1976). Actin filaments and birefringent spindle fibers during chromosome movements. In Cell Motility, Book C. Microtubules and Related Proteins (ed. R. Goldman, T. Pollard & J. Rosenbaum), pp. 1273-1293, Cold Spring Harbor, N.Y.: Cold Spring Harbor Laboratory. FORER, A. & KOCH, C. (1973). Influence of autosome movements and of sex-chromosome movements on sex-chromosome segregation in crane fly spermatocytes. Chromosoma 40, 417-442. FORMAN, D. S., PADJEN, A. L. & SIGGENS, G. R. (1977). Effect of temperature on the rapid retrograde transport of microscopically visible intra-axonal organelles. Brain Res. 136, 215- 226. FRANK, O. (1907). Einfluss der Herztemperatur auf die Erregbarkeit der beschleunigenden und verlangsamenden Nerven. Z. Biol. 49, 392-420. FCRTH, R. (1917). 1. Einige Untersuchungen iiber BrownscheBewegunganeinemEinzelteilchen. Amxln Phys. 53, 177-213. FORTH, R. (1930). Uber die Messung der ViskositUt sehr kleiner Fliissigkeitsmengen mit Hilfe der Brownschen Bewegung. Z. Phys. 6o, 313-316. FUSELER, J. W. (1973). The Effect of Temperature on Chromosome Movement and the Assembly- Disassembly Process of Birefringent Spindle Fibers in Actively Dividing Plant and Animal Cells. Ph.D. Thesis, University of Pennsylvania, Pennsylvania, U.S.A. FUSELER, J. W. (1975). Temperature dependence of anaphase chromosome velocity and microtubule depolymerization. J. Cell Biol. 67, 789—800. GRAY, J. (1924). The mechanism of ciliary movement. III. The effect of temperature. Proc. R. Soc. B 95, 6-15. GROSS, G. W. (1973). The effect of temperature on the rapid axoplasmic transport in C-fibers. Brain Res. 56, 359-363. Temperature and anaphase movement 51

GRUZDEV, A. D. (1972). Critical review of some hypotheses concerning anaphase chromosome movements. (In Russian.) Tsitologiya 14, 141-149. (English translation as NRC Technical Translation 1758, National Research Council of Canada, Ottawa). HEILBRUNN, L. V. (1924). The colloid chemistry of protoplasm. III. The viscosity of protoplasm at various temperatures. Am. J. Physiol. 68, 645-648. HEILBRUNN, L. V. (1928). The Colloid Chemistry of Protoplasm, pp. 102-124. Berlin: Born- traeger. HEILBRUNN, L. V. (1958). The viscosity of protoplasm. In Protoplasmatologia. Vol. II. Cyto- plasma C. Physik, Physikalische Chemie, Kolloidchemie 1 (ed. L. V. Heilbrunn & F. Weber), pp. 1-109. Wien: Springer-Verlag. HESLOP, J. P. & HOWES, E. A. (1972). Temperature and inhibitor effects on fast axonal transport in a molluscan nerve. J. Neiirochem. 19, 1709-1716. HOLWILL, M. E. J. & SILVESTER, N. R. (1965). The thermal dependence of flagellaractivit y in Strigomonas oncopelti. J. exp. Biol. 42, 537-544- INOUE, S. (1952). The effect of colchicine on the microscopic and submicroscopic structure of the mitotic spindle. Expl Cell Res. Suppl. 2, 305-318. INOUE, S. (1964). Organization and function of the mitotic spindle. In Primitive Motile Systems in Cell Biology (ed. R. D.Allen & N. Kamiya), pp. 549-598. New York and London: Academic Press. INOUE, S. (1976). Chromosome movement by reversible assembly of microtubules. In Cell Motility. Book C. Microtubules and Related Proteins (ed. R. Goldman, T. Pollard & J. Rosenbaum), pp. 1317-1328. Cold Spring Harbor, N.Y.: Cold Spring Harbor Laboratory. INOUE, S. & RITTER, H. (1975). Dynamics of mitotic spindle organization and function. In Molecules and Cell Movement (ed. S. Inou6 & R. E. Stephens), pp. 3-30. New York: Raven Press. LAPOUNTAIN, J. R. Jr. (1972). Spindle shape changes as an indicator of force production in crane-fly spermatocytes. J. Cell Sci. io, 79—93. LAUGHLIN, H. H. (1919). Duration of the several mitotic stages in the dividing root-tip cells of the common onion. Publ. Cam. Instn Wash. 265, 1-48. LINEWEAVER, H. (1939). The energy of activation of enzyme reactions, and their velocity below 0°. J. Am. chem. Soc. 61, 403-408. L<3PEZ-SAEZ, J. F., GIMENEZ-MART/N, G. & GONZALEZ-FERNANDEZ, A. (1966). Duration of the cell division cycle and its dependence on temperature. Z. Zellforsch. mikrosk. Anat. 75, 591-600. MAKINO, S. & NAKAHARA, H. (1953). Cytologische Untersuchungen an Tumoren. VIII. Beobachtungen iiber den Mitoseablauf in lebenden Tumorzellen der Ascitessarkome der Ratten. Z. Krebsforsch. 5, 298-309. MARCOLIS, R. L., WILSON, L. & KIEFER, B. I. (1978). Mitotic mechanism based on intrinsic microtubular behaviour. Nature, Land. 272, 450-452. MCDONALD, K., PICKETT-HEAPS, J. D., MCINTOSH, J. R. & TIPPIT, D. H. (1977). On the mechanism of anaphase spindle elongation in Diatoma vulgare. J. Cell Biol. 74, 377-388. MCINTOSH, J. R. (1974). Bridges between microtubules. .7. Cell Biol. 61, 166-187. MCINTOSH, J. R., HEPLER, P. K. & VAN WIE, D. G. (1969). Model for mitosis. Nature, Lond. 224, 659-663. MURPHY, Q. (1940). The effect of temperature on the viscosity of Amoeba dubia protoplasm. J. cell. comp. Physiol. 16, 401-404. NICKLAS, R. B. (1973). Experimental and cellular control of chromosome movement. In Chromosomes Today, vol. 4 (ed. J. Wahman & K. R. Lewis), pp. 215-223. New York and Toronto: Wiley. Jerusalem: Israel Universities Press. NICKLAS, R. B. (1975). Chromosome movement: current models and experiments on living cells. In Molecules and CellMovement (ed. S. Inou6&R. E. Stephens), pp. 97-117. New York: Raven Press. OPPENHEIM, D. S., HAUSCHKA, B. T. & MCINTOSH, J. R. (1973). Anaphase motions in dilute colchicine. Evidence of two phases in chromosome segregation. Expl Cell Res. 79, 95-105. PANTIN, C. F. A. (1924). Temperature and the viscosity of protoplasm. J. mar. biol. Ass. U.K. 13.331-339- 52 C. J. Schaap and A. Forer PEKABEK, J. (1930). Absolute Viskositatsmessung tnit Hilfe der Brownschen Molekular Bewe- gung. 1. Prinzip der Methode, Voraussetzungen, Fehlerquellen der Messungen. Protoplasma io, 510-532. Ris, H. (1949). The anaphase movement of chromosomes in the spermatocytes of the grass- hopper. Biol. Bull. mar. biol. Lab., Woods Hole 96, 90-106. SALMON, E. D. (1976). Pressure-induced depolymerization of spindle microtubules. IV. Pro- duction and regulation of chromosome movement. In Cell Motility. Book C. Microttibules and Related Proteins (ed. R. Goldman, T. Pollard & J. Rosenbaum), pp. 1329-1341. Cold Spring Harbor, N.Y.: Cold Spring Harbor Laboratory. SCHRADER, F. (1953). Mitosis. New York: Columbia University Press. SIZER, I. W. (1943). Effects of temperature on enzyme kinetics. Adv. Ensymol. 3, 35-62. SLEIGH, M. A. (1956). Metachronism and frequency of beat in the peristomal cilia of Stentor. J. exp. Biol. 33, 15-28. SOKAL, R. R. & ROHLF, F. J. (1969). Biometry: The Principles and Practice of Statistics in Biological Research. San Francisco: Freeman. STEPHENS, R. E. (1972). Studies on the development of the sea urchin Strongylocentrotus droebachiensis. I. Ecology and normal development. Biol. Bull. mar. biol. Lab., Woods Hole 142, 132-144. STEPHENS, R. E. (1973). A thermodynamic analysis of mitotic spindle equilibrium at active metaphase. J. Cell Biol. 57, 133-147. TAYLOR, E. W. (1959). Dynamics of spindle formation and its inhibition by chemicals. J. biophys. biochem. Cytol. 6, 193-196. TAYLOR, E. W. (1965). Brownian and saltatory movements of cytoplasmic granules and the movement of anaphase chromosomes. In Proc. IVth int. Congr. on RJieology, Brotvn University (ed. E. H. Lee), pp. 175-191. New York: Interscience. ULLERICH, F.-H., BAUER, H. & DIETZ, R. (1964). Geschlechtsbestimmung bei Tipuliden (Nematocera, Diptera). Chroviosoma 15, 591-605. WEBER, F. & WEBER, G. (1916). Die Temperaturabhangigkeit der Plasmaviskositat. Ber. dt. bot. Ges. 34, 836-846. WOLF, E. (1941). Die Chromosomen in der Spermatogenese einiger Nematoceren. Chroviosoma 2, 192-246.

(Received 9 January 1979)