Evidence For Twisted Plane Undulations In Golden Hamster

Download Evidence For Twisted Plane Undulations In Golden Hamster

Preview text


Downloaded from http://rupress.org/jcb/article-pdf/75/3/851/1072788/851.pdf by guest on 08 July 2022

From the Department of Anatomy, Royal (Dick) Schoolof Veterinary Studies, Universityof Edinburgh, Edinburgh EH9 1QH, Scotland. Dr. Woolley'spresent address is the Department of Physiology, The Medical School, Universityof Bristol, Bristol BS8 1TD, England.

Motile spermatozoa from the golden hamster have been arrested by rapid freezing and then fixed with glutaraldehyde at low temperature after substitution with ethylene glycol. As far as can be judged, the flagellar waveforms thus stabilized are similar to those seen in living sperm; in contrast, fixation in glutaraldehyde, without prior freezing, induces agonal changes in flagellar conformation. The characteristic waveform after freeze substitution contains three bends. Approx. half of these flagella are entirely planar. The rest are threedimensional, with the third bend displaced in a regular way from the plane containing the proximal two bends. From the geometry of these flagella, it is concluded that the plane of action of a given bending cycle undergoes a clockwise twist (from a forward viewpoint) as the cycle is succeeded by new bending cycles. This "twisted plane" undulation is quite different from helical movement. The twisting seems to occur abruptly, between cycles, as if each bending cycle has a preferred plane of action. The mechanism underlying the twisting is uncertain. However, on the basis of the angular displacements between the preferred planes, and the findings from electron microscopy, the following idea is presented as a working hypothesis: that, if the most proximal plane of bending is topographically determined by peripheral doublet 1, then successive distal planes of action are influenced predominantly by doublets 2, 3, etc., in clockwise sequence. The merits and weaknesses of this hypothesis are discussed.

KEY WORDS spermatozoa motility flagella 9 axoneme 9 freeze-substitution
Although there have been several detailed studies of the movement patterns of mammalian spermatozoa, in which increasingly fast cine cameras have been used, no agreed description of the flageilar waveform has evolved. The bull sperm flagellum was reported to perform planar bending proximally and a more complex figure-of-eight movement at its tip (13). Alternatively, this fla-

gellum was said to execute waves of helical form but with an elliptical section (18), and planar beating was dismissed as pathological or artefactual (18, 21). Elliptically helical movement has also been suggested for rotating ram sperm, though, because of the flatness of the ellipse, it was conceded that the proximal bending might be planar (6). The idea that a planar bend might become helical as it is propagated was expressed in an earlier review (2). Thus, while there is agreement that rotating spermatozoa have a

THE JOURNAL OF CELL BIOLOGY 9 VOLUME 75, 1977 9 p a g e s 8 5 1 - 8 6 5


Downloaded from http://rupress.org/jcb/article-pdf/75/3/851/1072788/851.pdf by guest on 08 July 2022

waveform which is, over its entire length, threedimensional, the detailed geometry has been in dispute.
Taking the view that a more exact account is unlikely to come from cinephotography (principally because three-dimensional waveforms must always be partially out-of-focus) the author has adopted the approach currently used in studying ciliary bending, namely to attempt the "instantaneous" fixation of the axonemes. If this can be achieved, the flagellar geometry can be studied at leisure, and it should be possible, by arranging the individual waveforms thus preserved, to put together an account of how the configuration of the typical flagellum is changing with time. However, rapid chemical fixation with osmium tetroxide, the method generally used for cilia, only sometimes arrests the metachronal wave satisfactorily (1) and so, for a suspension of spermatozoa, where there is no equivalent pattern of co-ordination which can be checked, a primary chemical fixation was excluded as being likely to produce a significant number of agonal flagellar configurations. Instead, "instantaneous" fixation has been attempted by rapid freezing. The configurations thus physically preserved have been chemically stabilized at low temperatures by substitution fixation of the cells in a glutaraldehyde containing ethylene glycol solvent. The development of this freeze-substitution procedure has already been
reported in detail (22), and only minor technical
changes have been made, except for the decision to omit a cryoprotectant from the sperm diluent. The first conclusion drawn from the work, for the rat spermatozoon, was that the most proximal flagellar bend is planar, occurring in a plane perpendicular to that containing the two central microtubules of the axoneme. In a second report (23) it was provisionally concluded that distal regions of the rat sperm flagellum also had a planar beat but that the plane of action had become angled, apparently without an accompanying twist in the flagellum. Since electron microscopy was used, these were high resolution studies but of very few spermatozoa.
In the present study mainly made by light microscopy, resolution has been sacrificed for the sake of studying a larger sample of cells, and golden hamster sperm have been preferred to those of the rat on account of their greater viability. Initially, a further check has been made on the validity of the method by comparing the proximal waveform of living hamster spermato-

zoa with that obtained after freeze substitution or after glutaraldehyde fixation. Then, from a more
detailed study of the frozen-substituted cells, a description has emerged of a twisted plane pattern
of bend propagation and a hypothesis is proposed which relates the geometry of the twisting to the
internal structure and chirality of the axoneme. The concept of a twisted plane waveform, though
absent from reviews of flagellar function (19, 4, 20) has in fact appeared once previously, in a
description of bull spermatozoa (3).
Mature golden hamsters (Mesocricetus auratus) were
killed by cervical dislocation and left at room temperature for 30 min. Then, from each animal, a single suspension of spermatozoa was prepared by adding some of the contents of a cauda epididymidis to 2 ml of Hanks' solution at room temperature. 5 rain later, when the spermatozoa had become dispersed, successive aliquots were withdrawn and subjected either to (a) flash photography, (b) conventional glutaraldehyde fixation, or (c) rapid freezing (followed later by substitution fixation). These operations could be completed within about 10 min.
For flash photography, several drops of suspension were placed between a siliconized slide and cover slip, the latter supported by two hairs. Without delay, exposures were made of several fields using a 10 x objective, dark ground condenser, Leitz microflash illumination (flash duration approx. 0.001 s) and Ilford FP4 35-ram film. The analysis of the data was done from large prints and consisted of making a decision, for every sperm where a decision was possible, whether the most proximal flagellar bend was in the principal or the reverse direction. (The terminology, taken from Gibbons and Gibbons (12), is discussed later.) The distinction between principal and reverse bends was easily made by relating the curvature of the bend to the disposition of the asymmetrically shaped sperm nucleus (Figs. 1-3).
To compare the normal flagellar geometry with that obtained after chemical fixation, samples of the sperm suspensions were fixed in relatively large volumes of 2.5% glutaraldehyde in 0.1 M s-coUidine buffer (pH 7.4). Small volumes of the sperm suspension were taken as films on aluminum foil and quickly submerged in the fixative. After about 2 h the sedimented spermatozoa were carefully withdrawn and mounted on slides as deep, sealed, wet preparations. The same simple datum was recorded, for 300 spermatozoa, namely, the identification of the most proximal bend as being either in the principal or the reverse direction.
The third technique used in the triple comparison was fixation by freeze substitution. The method employed differed in only a few details from that previously described (22). A rectangular piece of silver foil (6 •


Downloaded from http://rupress.org/jcb/article-pdf/75/3/851/1072788/851.pdf by guest on 08 July 2022

FmURE 1 Phase-contrast micrograph of a golden hamster spermatozoon showing its asymmetrical head and the proximal part of its flagellum. This configuration of the flagellum is defined as a principal bend. x 1,000.
FIGURE 2 Similar to Fig. 1 but showing the alternating flagellar configuration, which is defined as a reverse bend. x 1,000.
FmuaE 3 Principal and reverse bends visible in motile spermatozoa. Flash illumination. Dark ground optics, x 500.

20 mm; Goodfellow Metals Ltd., Esher, England) supported in a cleft applicator stick, was dipped into the sperm suspension and then quickly dropped into Freon22 maintained at its freezing point (-146 to -147~ by liquid N2. After remaining under the Freon for 5 s, the metal foil was quickly transferred to a liquid N2 storage bath before being put into the substituting fluid. This fluid, based on a formula of Pease (14), was prepared in 100-ml lots in screw-capped Pyrex tubes. It is essentially a eutectic mixture of ethylene glycol containing a small amount of glutaraldehyde (ethylene glycol, 70 ml; 0.133 M phosphate buffer, pH = 7.3, 25 ml; 24.6% glutaraidehyde (TAAB Laboratories, Emmet Green, Reading, England), 5 ml). This solution is less acidic than that used earlier (22), owing to its lower glutaraidehyde concentration and greater buffering capacity. The tubes of substituting fluid were immersed in an acetone bath and cooled with dry ice to -50~ For freeze substitution the sperm films were quickly transferred from liquid N2 to the substituting fluid (2 films to each tube), and the tubes were immediately sealed and maintained at - 5 0 ~ (+2 ~) until the opacity of the frozen films had cleared (about 24 h). Thereafter, the preparations were left undisturbed until they had warmed to +10~ (about 24 h). From experiments with gelatin models (22), it was known that glutaraldehyde behaves

as a fixative in ethylene glycol and that it acts at temperatures at least as low as -20~
Further processing involved transferring the spermatozoa to a deep slide in which they could sediment freely before examination by phase-contrast microscopy. The supernate in the tubes above the level of the silver foils was removed by suction. Then the foils were gently agitated and removed. 3 days later, the sedimented spermatozoa were withdrawn by pipette and placed within a wax ring on a slide. A cover slip was sealed in place with wax (Vaseline, 2 vol; and beeswax, 1 vol). The chamber contained a depth of 2 mm of fluid. It became apparent that the geometry of a sperm flagellum influences its orientation in the fluid and ultimately the position in which the cell comes to rest on the slide. Therefore, to allow all the spermatozoa to reach an equilibrium orientation during sedimentation, the preparation was first inverted so that all the ceils fell to the cover slip; then it was abruptly turned back over and placed on the microscope stage. 9 h were allowed for each sedimentation. Spermatozoa in these preparations were initially examined for the direction of curvature of the most proximal flagellar bend. Then a more complete study was undertaken in which each sperm's geometry was recorded using a notation which indicated (a) whether the sperm was lying with its left (L) or right

WOOLLEY Evidence for "Twisted Plane" Undulations in Golden Hamster Sperm Tails 8S3

Downloaded from http://rupress.org/jcb/article-pdf/75/3/851/1072788/851.pdf by guest on 08 July 2022

(R) side uppermost, (b) whether the most proximal bend was in the principal or reverse direction (Pl or rl), (c) whether further bends existed in the flagellum (P2, rz, Pa, etc.), and (d) whether any of the bends were displaced upwards, out of the focal plane (denoted by _P2,etc.). Photographs were taken using a 20• objective and phase-contrast illumination. Many of the spermatozoa were photographed several times, focusing first on the plane of the first bending cycle (plrt and r~p~) and then focusing on the crests of bends displaced from this plane. The vertical displacements of the microscope stage between exposures were recorded from the 1 ~m calibrations of the fine focus knob; these values were adjusted by a small correction factor obtained by measuring the stage's movement (with a micrometer) for the entire travel of the fine focus control. Further measurements on the spermatozoa (as shown in Fig. 8) were made from photographic prints. The curvilinear length (l) was measured with a map measure read to 1/16 inches.
Spermatozoa from one of the hamsters were examined by electron microscopy. Pieces of cauda epididymidis were fixed in 2.5% glutaraldehyde in 0.1 M scollidine buffer (pH 7.4) for 2 h, rinsed in buffer for 30 rain, and postfixed in 1% OsO4 in the same buffer for 1 h. After dehydration in a graded ethanol series, the tissue was embedded in Epon 812. Serial silver sections were mounted on slotted grids (supported by a Formvarcarbon film), stained successively in 3% aqueous uranyl acetate and 0.2% lead citrate, and examined in an AEI 6B electron microscope. To ensure a true reconstruction from serial sections, the order in which the sections left the block was always noted, and care was taken not to produce mirror-inversion during the microscopy and printing. For comparative and more extended observations, epididymal spermatozoa from a rat were processed in the same way and examined in serial sections.
Since the golden hamster spermatozoon has not
previously been used in quantitative studies, it is worth reporting first of all the dimensions of its flagellum. From a series of photographs of 15 moribund sperm from a single male, the length of
the m i d d l e piece is 50.9 +-- 0.3 p.m ( m e a n +-- S E ) , and the total length of the flagellum is 179.9 0.8 p.m. This last measurements includes the end-
piece, approx. 3 /zm long. As in the other mammalian species (15) the
bending cycle of the proximal middle piece in spermatozoa from the cauda epididymidis is asymmetrical. And in the hamster, by reference to the nuclear asymmetry, the principal and reverse bends, as defined ostensively in Figs. 1 and 2, are readily distinguishable. The name "principal" has been given to the bend which has the longer

duration (12). Therefore at any moment more sperm are to be found executing principal first bends (Pl) than reverse ones (rl), and this proportion, expressed as Pd(P~ + rl), is a parameter of the population which can be obtained by photography using flash illumination (Fig. 3). Any fixation method which fails to preserve this ratio cannot be affecting all the cells in the same way and must be of questionable value.
Data from the comparison of living, frozensubstituted, and glutaraldehyde-fixed spermatozoa are given in Table I. • analysis confirmed that in each experiment the ratio p~:r~ in the living sample had been significantly altered by glutaraldehyde fixation but not significantly changed by the freeze-substitution procedure.
One clear effect of the rapid freezing, however, is that some of the sperm tails are either partially or completely fractured. In the three experiments, 27%, 28%, and 23% of the sperm were damaged. These cells were not excluded from the data in Table I provided the shape of the proximal bend could be seen.
Undamaged sperm from the frozen substituted samples were studied further in a search for patterns in the distribution of flagellar configurations. It was noticed, firstly, that most of the flagella contained 1.5 cycles of bending; secondly, that some of these wave forms were three-dimensional and some planar; and thirdly, that the sperm did not sediment such that equal numbers came to lie right-side or left-side uppermost. This last irregularity was accentuated if the planar wave forms were not considered. These initial impressions were established quantitatively in a classification of spermatozoa from two of the experiments (Fig. 4). The classification used the descriptive notation explained in the preceding section. The side of the sperm head defined as the "right" side is that which is seen to be upper-
TABLE I The Values Obtained for the Proportion of Principal First Bends (Pl/[Pt + rj) in the Sperm Population afier Different Fixation Procedures
Glutarald9 Expno. Hashphotography Freezesubstitution fixation
$32 0.75 (338) 0.78 (300) 0.53 (300) $33 0.78 (277) 0.75 (300) 0.41 (300) $40 0.66 (122) 0.62 (300) 0.40 (300)
The numbers of spermatozoa examined are given in parentheses.


Downloaded from http://rupress.org/jcb/article-pdf/75/3/851/1072788/851.pdf by guest on 08 July 2022

most in Figs. 5-7. From Fig. 4 it will be seen that sperm with the common planar wave form plr~p~ sediment about equally as R p~r~p2 and L ptrlP2 whereas those having an elevated third bend (p~rkp~) sediment such that R PlrLP2 greatly outnumber L p~r~_~. In the third experiment in which only these last two categories were analyzed, the same pattern was seen: 48 sperm of the type R plr~_2 were counted, and only two of the type L p~r~_2. The likely explanation of the sedimentation behavior of the three-dimensional flagella is discussed later.
While recognizing that there is a diversity of wave forms in the frozen-substituted preparations, attention has been concentrated on the geometry of the largest class, R par~_2. Two such sperm, having different bend amplitudes are shown in Figs. 5 and 6. In examining the vertically displaced bends,_p2, it was found that their maximum vertical elevations coincided with their maximum projected amplitudes, suggesting that they were still planar bends but now had a plane of action twisted with respect to the plane of action of the

more proximal bends, p~ and rl, i.e., it appeared that a principal bend, from beginning in a plane parallel to the microscope slide (as bend Px) must then, as it propagates (and becomes P2), move through a certain angle, 02, away from this initial plane. Viewed from the sperm head, this angular movement is clockwise. The angle has been calculated by relating the projected amplitude a~, of bend_p2 (defined in Fig. 8) to the upward displacement, d2, of the center of the bend (Fig. 9). These parameters are closely correlated (r = 0.59). The mean angle subtended by the centers of bends _P2 at the axis of progression was 46.6~ with a 1% confidence interval of -+6.0~ A more appropriate analysis, based on normalized data, gave the same value for the mean angle and will not therefore be presented in detail.
Figure 9 also includes a few data for the rarely seen waveform which contains three principal bends. One such sperm is shown in Fig. 7. The estimation of the angle through which the center of bend_ps must have moved is complicated by the fact that bend r2, displaced in the opposite

















6~ -i




6 -I

' I'o 2'o ; o / o // 2.0 . rO. . .

IO 2'~ 3'0 4'0/ / ' 30 2'~ I'O '

Fioo~ 4 Waveformclassificationsfrom two experiments ($32 and $40). The histograms depicting the number of spermatozoa which landed right (R) and left (L) side uppermost are presented side by side for comparison. The descriptive notation, at the left of the figure, is explained in the text. Sperm numbers may be read from the horizontal scale. Detailed analysisof flagellargeometry has been confined to the largest single class (R plr~_2).

WOOLLEY Evidence for "'Twisted Plane" Undulations in Golden Hamster Sperm Tails 855

Downloaded from http://rupress.org/jcb/article-pdf/75/3/851/1072788/851.pdf by guest on 08 July 2022

FXCURE 5 (a) Frozen-substituted spermatozoon having a planar first bending cycle and a vertically displaced second principal bend, lost from the plane of focus. (b) The same spermatozoon, with the second principal bend brought into focus. The plane of this micrograph is 6/.Lm above that of Fig. 5 a. x 1,176.

Downloaded from http://rupress.org/jcb/article-pdf/75/3/851/1072788/851.pdf by guest on 08 July 2022

FmuaE 6 (a and b) Frozen-substituted spermatozoon of waveform R p~rkpz but having bends of greater amplitude than the flagellum in Fig. 5a and b. The vertical displacement of the microscope stage between the two micrographs was 13 /zm. x 1,176.

direction from bends Pz, P3, adds to the total upward displacement of these principal bends from the plane of the microscope slide. Assuming

that bends_pa and r2 have the same amplitude and supposing, as an approximation, that the displacements of p~ and P3 are equally augmented

WOOLLEV Evidence for "Twisted Plane" Undulations in Golden Hamster Sperm Tails 857

Downloaded from http://rupress.org/jcb/article-pdf/75/3/851/1072788/851.pdf by guest on 08 July 2022

FmURE 7 (a-c) A frozen-substituted spermatozoon having the rare waveform R plr~zr~3. Fig. 7a is focused in the plane of the slide; Fig. 7b is displaced upwards by 5 /zm, almost to the center of bend p2; Fig. 7c is displaced a further 5 /~m and brings bend_p3 into focus. • 1,176.







\ :\.


Downloaded from http://rupress.org/jcb/article-pdf/75/3/851/1072788/851.pdf by guest on 08 July 2022


-/ !


Projected amplitude(/~m)

FmuRE 8 A diagram of the characteristic waveform (R plr~.z) seen after freeze substitution. The maximum upward displacement (d~) of bend p2 was recorded from the fine focus control. The projected amplitude ap was obtained from a photograph after judging the center of the bend and locating the progression axis at the midpoint between bends ri and P2. An estimate of the distance (l) traveled by bend C2 was obtained as shown in Fig. 8b, without making allowance for the upward displacement of bend p2.
by the height of rz, then a value for the displacement of 23 from the progression axis can be obtained by subtracting from its total displacement one half of the total upward displacement of bend P2. The estimate of the mean angle subtended by the center of bend P3 at the axis of progression, from five spermatozoa only, was 79.3 -+ 6.6 ~ (1% confidence interval).
Two attempts were made to account for the variation in 02. Firstly, each value of 02 was plotted against the true amplitude of its bend, obtained from x/d22 + t~pz. The relationship between these data can be appreciated indirectly from Fig. 9. The regression coefficient, b = 0.015 /.tin of amplitude per degree, was not significantly different from zero ( P > 0.5). The angle 02 was also plotted against the distance (l) travelled by the bend P2 (see Figs. 8 and 10) but was found

F]oums 9 The relationship between the projected amplitude and the upward displacement of principal bends. The graph is drawn such that the origin represents the progression axis and the points represent the position in space of the centers of principal bends as if viewed from the sperm head. The mean angle subtended by each group of data at the origin is indicated by a line. The data have been pooled from three experiments after ruling out heterogeneity. (O) Second principal bends (p~); and (O) third principal bends (Pa).




~J.5 9


' t35' 9
~ 9







t6o rio ~.o t~o t~o

Distancebetweencenterofbend~ andbaseofflagellum(/zrn)

FIGURE 10 The relationship, within individual spermatozoa, between the angle (02) of the twisted plane at the center of the second principal bend (_P2) and the distance travelled by this bend. The independence of the data indicate that the angle is not determined by whichever region of the flagellum happens to contain the bend.

WOOLLEY Evidence for "Twisted Plane" Undulations in Golden Hamster Sperm Tails 859

Downloaded from http://rupress.org/jcb/article-pdf/75/3/851/1072788/851.pdf by guest on 08 July 2022

not to be dependent on this distance (b = 0.22; P > 0.5).
In order to relate the apparent clockwise twist of the plane of bending to the structure of the axoneme, it was necessary to confirm that the hamster sperm flagellum conforms with others studied previously in having a "clockwise" axoneme. (This means that, when viewed from the base, the axonemal doublets numbered 1-9 appear in clockwise array.) The morphological transition at the distal end of the middle piece, where the axoneme complex is enclosed successively in mitochondria, the annulus and then the fibrous sheath (see Fig. 11), was identified in serial transverse sections and used as the means of knowing the true orientation of the cell. Thus, in the serially sectioned sperm in Figs. 12-15, the axoneme appears anti-clockwise but the series progresses proximally; therefore, the flagellum, by convention, is clockwise. This result was obtained in eight spermatozoa from one hamster.
Although the flagellar configuration R plrlp2
was the dominant three-dimensional waveform,
there were a few flagella (L plr~_p2) which were
twisted anti-clockwise (Fig. 4). Did these few have anti-clockwise axonemes? The question could not be answered for hamster spermatozoa on account of the scarcity of suitable sections. However, in the rat a larger number of spermatozoa could be examined by taking advantage of a more easily found morphological marker, the posterior lobe of the spermatozoan nucleus (Fig. 16). 50 sperm from one epididymis were studied in serial sections (Figs. 17-18), and all had clockwise axonemes.
The significance of the flagellar waveforms observed after freeze substitution depends critically on the assumption that in freezing the cells, the time taken to arrest the flagellum is short in relation to the time scale of bend generation and propagation. There is at present considerable, though indirect support for this assumption. Rapatz et al. (16) have reported the cooling rates of blood samples rapidly frozen in capillary tubes. From their Fig. 3 the fastest rate of cooling, estimated over the freezing range (say to -10~ was about 220~ If such a figure applied, the sperm sample would be cooled from +21~ to -10~ in 0.14 s. Now, under the conditions specified, at room temperature, a few hamster spermatozoa swim extremely rapidly, but nearly

all of them have beat frequencies within the range 1.1-1.7 cycles/s (personal observation). The fastest of this group, therefore, would be arrested within 0.24 cycle. However, where Rapatz et al. used isopentane as their coolant, in these experiments Freon-22 has been used; published estimates of cooling rates indicate that Freon-22 is 1.6 • more efficient than isopentane in cooling a naked thermocouple (17). Also, greater efficiency must have been gained from the higher surfaceto-volume ratio of the sperm specimens compared with the blood samples and from the direct contact between the sperm suspension and the coolant. These considerations indicate that the actual time taken to stop the flagella was considerably shorter than the above estimate.
The comparison which has been made between frozen-substituted and living spermatozoa lends some support perhaps to the view that the sperm are stabilized nearly instantaneously, but the work is interpreted with caution, since, strictly, it reveals nothing about the speed of the quenching procedure; and, furthermore, the flash micrographs can reveal nothing of the important distal part of the flagellum. The results do indicate, however, that there is no time available for at least the proximal part of the flagellum to undergo gross agonal responses during stabilization. Furthermore, it has also been possible to show that a gross agonal response, in this case to immersion in glutaraldehyde, can be detected by this simple technique. Before discussing the waveforms in detail, some explanation is required of the way in which the frozen-substituted sperm sedimented: the relevant observation was that, in a flagellum in which a distal segment deviated from the plane of the first bending cycle (plane P1), the sperm nearly always came to lie with this part of the flagellum projecting upwards from the microscope slide. If we consider a spermatozoon in a fluid, its equilibrium orientation is given by a line connecting the centers of gravity and buoyancy; and, if we assume the sperm head to have a higher specific gravity than the tail, the orientation of a planar tail would be vertical, as shown in Fig. 19a. Any deviation of the tail from plane PI would cause a greater lateral displacement of the
center of buoyancy than of the center of gravity. Such a sperm would therefore adopt a different orientation (Fig. 19b and c), and this orientation would determine the most likely direction of fall after the sperm first touches the slide (Fig. 19d). In fact, gliding might well have occurred also, but



Preparing to load PDF file. please wait...

0 of 0
Evidence For Twisted Plane Undulations In Golden Hamster