T H E ANATOMICAL RECORD 202:339-348 (1982) Preovulatory Changes in Cumulus-Oocyte Complex of the Hamster M. LORRAINE LEIBFRIED A N D NEAL L. FIRST Endocrinology-Reproductive Physiology Progrum, Department of Meat and Animal Science, University of Wisconsin-Madison, Madison, Wisconsin S3706 ABSTRACT An orderly series of associations was found between the cumulus and the stages of nuclear development in the hamster oocyte as the oocytecumulus complex underwent preovulatory maturation. Ten hours prior to ovulation meiotic resumption was first seen in the oocyte. The onset of cumulus dissociation with corresponding morphological changes in cumulus cells was observed eight hours prior to ovulation. Completion of the first meiotic division in the oocyte and full cumulus dissociation occurred 2 hours prior to ovulation. Cumulus dissociation was found to occur only in vesicular follicles destined to ovulate and not in follicles undergoing atresia. Cells with large vacuolar inclusions with distorted, hyperchromatic nuclei were found in the maturing cumulus. Resumption of the first meiotic division in the oocyte and dissociation of the cumulus oophorus from the membrana granulosa are two major events occurring during preovulatory maturation of the follicle in most mammals. The importance of nuclear maturation as a prerequisite for fertilization and its causation by the preovulatory gonadotropin surge has been recognized (Lindner et al., 1974).Dissociation of the cumulus-oocyte complex from the underlying granulosa, although long noted, has been studied very little in relation to ovulation. Dissociation involves loss of cohesiveness both between cumulus cells and between cumulus cells and oocyte, and entrapment of the cells in the vicinity of the oocyte by a viscous, intercellular cement. Loss of cohesiveness between cumulus cells is due to the disappearance of intercellular junctions (Gilula et al., 1978;Dekel et al., 1976b).The process of mucification can be due to either de novo synthesis and accumulation of mucinous material during preovulatory maturation (Dekel and Kraicer, 1978)or enzymatic liquification of an existing intercellular matrix. The result of dissociation of the cumulus is an oocyte surrounded by single cells maintained in loose association with each other and the egg by a viscous matrix. Dissociation has been thought of as a process that allows release of the oocyte from the follicle at ovulation. The greater mass added to the oocyte by expansion of the cumulus is thought to ensure transport of the ovum from the point of rupture to the 0003-276X/82/2023-0339$03.00 0 1982 Alan R. Liss, Inc. ostium abdominale. The stickiness of the mucinous material could facilitate entrapment of the oocyte by the fimbriated infundibulum and transport of the ovum to the site of fertilization. Cumulus cells may also aid in the capacitation of spermatozoa prior to fertilization (Gwatkin et al., 1972; Gwatkin and Andersen, 1973);hence the need for the viscous matrix to retain cumulus cells in the vicinity of the oocyte after ovulation. Changes in the cumulus related to dissociation may be the outward manifestation of a function that is antecedent to ovulation. Cumulus dissociation in the rat occurs after the preovulatory gonadotropin surge (Dekelet al, 1976b) and can be stimulated in vitro in several mammalian species by gonadotropins (Dekel and Kraicer, 1978; Thibault et al, 1975; Eppig, 1979) and adenosine monophosphates (Dekel and Kraicer, 1978; Eppig, 1979). The disappearance of cell-to-cell junctions during cumulus dissociation, occuring coincidentally with preovulatory maturation of the oocyte (Dekel et al., 1976b Gilula et al., 1978), has been implicated as a mechanism whereby resumption of the meiotic process prior to ovulation can be controlled. The purpose of this study is to describe the time relationship between meiotic resumption in the oocyte and cumulus dissociation during the preovulatory period. The hamster was Received M a y 7. 19x1: accepted September 15. 19x1 340 M.L. LEIBFRIED AND NEAL L. FIRST chosen as an animal model owing to the constancy of its estrous cycle. MATERIALS AND METHODS Preparation o f animal materal Female golden hamsters between 4 and 6 weeks of age and maintained on a lighting schedule of 1 4 hours of light and 10 hours of dark were used. Only animals exhibiting three or more consecutive 4-day estrous cylces were included in the study. Day 1 of the cycle was defined as the day of the proestrous vaginal discharge (Ward, 1946). Starting at 15:OO on Day 4 of the cycle and ending at Day 1,four animals were killed every two hours. This time interval corresponds to the period just prior to or at the beginning of the endogenous gonadotropin surge preceding ovulation (Sheela Rani and Moudgal, 1977)and lasting until ovulation is completed in our colony. In addition, four animals were killed at 9:OO on Day 1to obtain the average number of ovulations expected for hamsters killed earlier. At autopsy the ovaries and oviducts were removed and fixed in Bouin’ssolution, embedded in paraffin, serially sectioned at 7 pm, and stained with Mallory’s triple stain (Humason, 1976). Handling of histological material Since we wished to consider oocyte-cumulus relationships only in Graafian follicles, examination of an animal‘s follicular population was restricted to vesicular follicles 2 400 pm in diameter, the smallest expected diameter for normal antral follicles in a cycling hamster (Greenwald, 1973). Size was estimated by counting the number of serial sections in which the follicle appeared and multiplying by the thickness of a section. Using pyknosis, sloughing of cells, and karyorrexis as indicators of atresia, antral follicles were divided on the basis of their granulosa into two categories: normal or atretic. A follicle was classified as atretic if the granulosa showed one or more signs of atresia. The categories used for evaluating stage of meiosis in the oocytes were: (1) Dictyate- An intact germinal vesicle was still present. (2) Prometaphase I-The germinal vesicle had dissappeared and the chromatin was condensing in the cytoplasm. (3) Metaphase I-The bivalents were in association with the spindle apparatus. (4) Anaphase to telophase I-The bivalents were separating. (5) Metaphase 11-The first polar body and second metaphase plate were present. (6) Other-Either no discernible chromatin was present or the configuration did not fit the preceding categories. Cumulus dissociation was evaluated according to the following categories: (1) Compact - Cumulus cells were closely associated in layers around the oocyte. The outline of the discus proligerus was smooth. (2) Loose-The hillock assumed an irregular outline. Cells were losing their epitheliallike arrangement and, beginning centripetally, were showing intercellular gaps. (3) Expanding - Cumulus cells were individually isolated from one another except for those forming the corona radiata. Large deposits of a mucopolysaccharide-staining material were seen as a granular precipitate between the cells. (4) Free-The description noted for the category termed expanding applied to this category but intercellular distances were further accentuated. The original area of the discus proligerus was no longer discernible and the oocyte had been pushed toward the center of the antrum (5) Ouiductal- Cumulus masses and associated oocytes were found in the oviducts. (6) Degenerate - Pyknosis andior karyorrexhis was apparent in the cumulus. Cell sloughing had caused thinning of the cumulus layers (five to eight cellular layers normally surround the oocyte). (7) Nude -Cellular degeneration and sloughing had left the oocyte free in the antrum surrounded by the zona pellucida. Both mitotic figures and formation of large lipid-containing vacuoles were observed in the cumulus cells. For this study two sample sections of the cumulus were rated as having mitotic figures present or absent. The sections chosen were those on either side of the oocyte where both zona pellucida and ooplasm are first encountered. A follicle was scored: 0 if neither of the samples possessed a mitotic figure, 1 if one of the sections had one or more mitotic figures, or 2 if both sections had one or more mitotic figures present. This scoring system was also used to score the presence or absence of the vacuolar inclusions in cumulus cells. Statistical methods Analysis of variance (Barr et al., 1976) was performed on both number and size of follicles. Cramer’s v (Dixon and Brown, 1977)was used 34 1 PREOVULATORY CHANGES IN CUMULUS-OOCYTECOMPLEX t o obtain a measure of association for cumulus and chromatin categories within a time period. Log linear analysis (Dixon and Brown, 1977) was used to analyze the data concerning mitotic figures. After preliminary analysis showed no effect of replicate for the categorical data on formation of vacuoles in the cumulus cells, numbers were collapsed across replicates and a chi square contingency table analysis was performed (Snedecor and Cochran, 1967). RESULTS Means for number and size of preovulatory follicles for each time period are shown in Table 1. After eliminating the values for animals killed at 15:OO. there were no differences between time periods in total number of follicles or the subcategories of normal and atretic follicles. Graafian follicles, therefore, were committed to ovulation or atresia at the time interval chosen for the study. Follicular diameter was different between time periods (P < 0.001), indicating that follicles are still increasing in size during the 12-hour period preceding ovulation (Table 1).Normal follicles were larger than atretic follicles during the preovulatory period (P < 0.001). From the results of the analyses for size and number of follicles, we concluded that classification of follicles based on appearance of the granulosa allowed separation of two distinct types of follicles. All further discussion and analysis will treat normal and atretic follicles separately. The univariate frequency distributions for chromatin and cumulus categories in normal follicles are shown in Figure 1. The cumulus category termed nude was deleted since follicles classified as normal never showed this character. With the use of Cramer’s v (Table l), a high degree of association was found between the appearance of these two characteristics within time. The sequence of associations between chromatin and cumulus will be described by individual time periods. 15:OO The discus proligerus had the smooth, rounded outline characteristic of the oocytic hillock (Fig. 2). With few exceptions the cumulus category found was that termed compact. The cells forming the cumulus were cuboidal and tightly packed in layers around the oocyte in a pseudoepithelial arrangement (Fig. 3). All of the oocytes remained in the dictyate stage (Fig. 3) with an intact nuclear membrane and nucleoli present. The exceptions to the compact cumulus were cumuli classified as degenerate (Fig. 12). Analysis of variance on the number of follicles present across all observation times (Table 1) indicated that a greater number of normal, vesicular follicles (P < 0.05) were present at 15:OO yet the total number of follicles was not different. The presence of folli- TABLE 1. Means and statistics for follicular data’ Time Mean No. of follicles 2 400pm’lanimal Total No. of normal2 follicles observed Mean No. of normal follicles2/animd Mean diameter of normal follicles (pm) Total No. of atretic follicles observed Mean No. of atretic folliclesianimal Mean diameter of atretic follicles (pm) Cramer’s v3 Mean score for mitotic index3 (x2 = 329.0) Mean score for index of vocuolated cells’ l:oo 3:OO 900 19.25 18.50 21.25 18.25 54 63 57 13.50 15.75 14.25 735.00 - - 17:OO 19:OO 19.25 19.50 20.50 21.75 69 53 59 57 17.25 13.25 14.75 14.25 13.50 497.04 631.32 661.56 700.29 707.78 10 25 23 30 23 20 5.00 480.28 1.00 495.78 0.709 528.85 0.317 1.37 1.52 1.34 1.12 1.24 1.96 1.a9 1.90 6.25 414.00 490.68 1.00 475.35 0.523 0.71 0.44 o,O1 0 7.50 54 5.75 5.75 2.50 - 21:OO 23:OO 15:OO 22 16 5.50 4.00 475.36 470.31 - - - - - - ‘Data are from four animals within a time period. ’Ovulated follicles were included in the number of folliclescategorized as normal and total number of follicles. The diameter of ruptured follicles was not measured. ’Statistics derived from data on normal. unruptured follicles to obtain a measure of association hetween stages of nuclear development of the m y t e and cumulus dissociation for an individual time period. 342 M.L. LEIBFRIED AND NEAL L. FIRST = CHROMATIN ESSI CUMULUS 50 CHROMATIN CUMULUS OTHER DEGENERATE 0 100 50 MET P OVIDUCT 0 100 50 WA-TELO I FREE MET I EXPANDING PROMET I LOOSE DICTYATE COMPACT 0 100 50 0 15:oo 1700 19:00 23:OO 300 21:OO 1:00 TIME Fig. 1. Univariate distribution of the chromatin and cumulus oophorus categories of normal follicles. cles classified as normal but containing degenerating cumuli account for this difference at 1500 compared to the other time intervals. These may represent the last follicles to begin atresia in the wave of follicles stimulated for this ovulation. The most accurate identification of a follicle as atretic should combine appearance of both the granulosa and cumulus oophorus. 17:m Again the majority ( > 90%)of cumuli in nor ma1 follicles were compact (Fig. 4). Although about 60% of the oocytes were still in the dicty- PREOVULATORY CHANGES IN CUMULUS-OOCYTE COMPLEX ate stage, the germinal vesicle was characterized by chromatin condensation and decreasing clarity of the nucleoli and nuclear membranes (Fig. 5). The remainder of the oocytes had undergone germinal vesicle breakdown with chromatin condensation continuing in the cytoplasm. The relatively low measure of association obtained at 17:OO hours (Table 1)indicated that resumption of meiosis began before discernible changes in the cumulus occurred. Such has been reported previously for the rat (Hillensjo et al., 1976). 343 3:OO Animals in our colony have completed ovulation by this time. The follicles remaining on the ovary were all atretic. The number of ruptured follicles counted on the pairs of ovaries corresponded to the number of oocytes found in the oviducts. Atretic follicles Cumulus stage and chromatin stage of the oocyte were not associated consistently nor sequentially in atretic follicles. Cumulus dissociation like that in normal follicles was never 19:oo seen in atretic follicles. The oocyte investCumulus dissociation was first observed by ments in atretic follicles were classified exclulight microscopy at this time period. The sively as degenerate (Fig. 12) and nude (Fig. majority of the cumuli were divided between 13). At all experimental time periods, oocytes the categories termed loose (Fig. 6) and ex- could be found in any stage of the first meiotic panding (Fig. 8).In loose cumuli the discus pro- division and also in the chromatin category ligerus had a rough outline and the cellular termed other. This category included oocytes layers had lost their epitheloid arrangement with no stainable chromatin or scattered (Fig. 7). Expanding cumuli were characterized clumps of chromatin and those possessing one by increased intercellular distances and or two female pronuclei. At no time period were increased density of the intercellular material oocytes in atretic follicles found consistently in demonstrated by the staining procedure (Figs. the stage of nuclear development considered 8,9).The majority of the oocytes were found in predominant for oocytes in normal follicles. Of the stage of chromatin development classified 169 oocytes observed in atretic follicles, only as prometaphase I. One-fourth of the oocytes ten remained in the dictyate stage of nuclear had reached metaphase I with a distinct development. spindle structure present. Mitotic figures and vacuole formation 21:OO The formation of vacuolar inclusions in cells Cramer’s v for this time period was unity, in- of the cumulus was not independent of time (P dicating a high degree of association for the < 0.001). Vacuoles were not observed until chromatin and cumulus categories typical for 1900, the time when the onset of cumulus disthis time. Over 85% of the follicles observed sociation is first observed (Table 1). These had oocytes with the chromatin configuration vacuoles (Fig. 14),probably having a high lipid at metaphase I . The predominant cumulus content, fill the cytoplasm in subsequent type was that termed expanding (Fig. 9). stages and seem to distort and displace the nucleus to an eccentric location in the cell. 23:OO Analysis of the categorical data for mitotic Although the majority of cumulus-oocyte figures (Table 1)gave a log linear model of best complexes remained in the same stage as the fit when rating of mitosis is considered as time preceding time period, i.e., expanding and dependent. A quadratic expression best metaphase I, about a third had proceeded to described the linear component of mitotic f r e free and anaphase-telophase I categories. quency with time. As seen in Table 1, frequency of mitotic figures increased until 21:00, 1:OO then declined. Hamsters in our colony begin ovulating at this time. Two out of four animals killed at this DISCUSSION period had one or two ovulations. Twclthirds of An ordered, sequential relationship between the oocytes had completed the first meiotic division with the expulsion of the first polar cumulus dissociation and oocyte maturation body and formation of the second metaphase has been demonstrated for the hamster. A simplate. The other oocytes were still in amphase- ilar relationship has been shown for the rat (Dekel et al., 1979). In both our study with telophase I (Fig. 11). The predominant hamsters and others using rats (Hillensjoet al, cumulus class was free (Fig. 10). 344 M.L. LEIBFRIED AND NEAL L. FIRST Fig. 2. Normal follicle observed a t 15:OO. The cumulus is compact and oocyte contains an intact germinal vesicle. x 63. Fig. 3 . Normal follicle observed at 15:OO. The nucleus of the oocyte and epithelial arrangement of the cumulus cells are apparent. x 160. Fig. 4 . Normal follicle observed a t 17:OO. ' f i e cumulus remains in the immature or compact category. x 63. Fig. 5. Normal follicle observed a t 17:OO. Mitotic figures are seen in the cumulus. Chromatin condensation is occurring in the germinal vesicle. x 160. PREOVULATORY CHANGES IN CUMULUS-OOCYTE COMPLEX Fig. 6. Normal follicle observed a t 19:OO. The cumulus is classified as loose with the uocyte at prometaphase I. X 63. Fig. 7. Normal follicle observed a t 19:OO. The loosening cumulus shows the loss of epithelial arrangement and development of intercellular spaces. X 160. 345 Fig. 8. Normal follicleobserved a t 2190. The cumulus is expanding. Increased intercellular distance and density of staining of the intercellular matrix characterize this cumulus category. x 63. Fig. 9. Normal follicle observed a t 21:OO. The expanding cumulus and metaphase Z are predominant at this time period. x 160. 346 M.L. LEIBFRIED AND NEAL L. FIRST Fig. 10. Normal follicle observed a t 23:OO. The ooeyte shows expulsion of the first polar body. The cumulus is classified as free. x 63. Fig. 11. Normal follicle observed a t 23:OO. The oocyte is in late telophase I. The vacuolated cells of the free cumulus are prominent during this time period. x 160. Fig. 12. Atretic follicleobserved a t 15:OO. Thecumulus is degenerate. x 63. Fig. 13. Atretic follicle observed a t 21:OO. One polar body and a pronucleus are seen in the oocyte. The cumulus category is that termed nude. x 63. PREOVULATORY CHANGES IN CUMULUS-OOCYTECOMPLEX Fig. 14. Normal follicle observed a t 23:OO. Mitotic f i g ures and vacuolated cells are seen in the cumulus. x 160. 1976; Dekel et al., 1979),meiotic resumption in the oocyte preceded observable cumulus disso ciation. Electron microscopic evaluation is needed to confirm this point since early ultrastructural changes would have been missed by the methods used to assess the onset of cumulus dissociation in these studies. When the onset of cumulus expansion was first observed, morphological changes in cumulus cells were also seen. A morphological change in cumulus cells corresponding to the onset of cumulus dissociation was also reported for the rat (Hillensjo et al., 1976; Dekel et al., 1979). The change observed was described as blebbing, or formation of pseudo podia. The major changes found in the cells of the hamster cumulus were the formation of large vacuoles and the distortion and hyperchromicity of the nucleus. Such changes could be signs of cellular degeneration. LH in the rat decreases cumulus cell respiration and destroys their ability to form monolayers in culture (Dekel et al., 1976a).These also could be taken as signs of impending cellular degen- 347 eration. Cumulus cells of oviductal eggs recovered after ovulation from mice show signs of cellular degeneration (Zamboni, 1970). It is possible that the gonadotropin surge prior to ovulation may be a terminal signal for the cumulus cells. Cumulus expression and associated morphological changes in cumulus cells were not observed in follicles undergoing atresia. Cumulus dissociation must be controlled by hormonal events of the preovulatory phase of the estrous cycle and it occurs in follicles “selected to undergo ovulation. In atretic follicles with advanced degeneration of the cumulus, oocytes were commonly found to have resumed meiosis. The presumed loss of intercellular association in cumulus cells may have allowed release of the oocyte from meiotic arrest. Hay and Cran (1978)have already suggested this after studying atretic, vesicular follicles of sheep. Loss of intercellular association or communication as part of cumulus dissociation has been shown to occur in the rat oocyte cumulus complex prior to ovulation (Gilula et al., 1978).This termination of cell-tocell communication is hypothesized to be the mechanism allowing meiosis to resume in the oocyte. It is striking that under two such disparate conditions as follicular atresia and preovulatory follicular development, loss of integrity of the cumulus is thought to allow for release of the oocyte from meiotic arrest. If meiotic arrest during folliculogenesisis originally dependent on formation of intercellular communication with follicle cells, as has been suggested (Ohno and Smith, 1964; Anderson and Albertini, 1976), then meiotic resumption during follicular atresia or development is the reverse process or loss of communication. Control of meiosis would then be analogous to contact inhibition. In summary, we have shown a sequential series of associations between the cumulus as dissociation occurs and the stages of nuclear development in the hamster oocyte prior to ovulation. Meiotic resumption was observable prior to the onset of cumulus dissociation, as shown by light microscopy. Corresponding to the onset of cumulus dissociation, large vacuolar inclusions are found in the cells of the cumulus. Loss of integrity of the cumulus under the conditions of atresia or development takes separate pathways. Cumulus dissociation was found to occur only in preovulatory follicles and not in follicles undergoing atresia. Cumulus dissociation, therefore, must be initiated by hormonal events of the preovulatory phase of the estrous cycle. 348 M.L. LEIBFRIED AND NEAL L. FIRST ACKNOWLEDGMENTS The authors are grateful to Dr. Fields C. Gunsett for statistical consulting and to Julie Busker for assistance in histologic preparation of the tissue. This research was supported by the College of Agricultural and Life Sciences, University of Wisconsin-Madison; U.S.D.A S.E.A. grant No. 12-14-5001-313 Mod. 1; Public Health Service training grant No. 5-T01-00104-10 from the National Institute of Child Health and Human Development; and grant No. 630-0505B from the Ford Foundation. This article is Department of Meat and Animal Science Paper No. 788. REFERENCES Anderson, E., and D.F. Albertini (1976) Gap junctions between the oocyte and companion follicle cells in the mammalian ovary. J. Cell Biol., 71.680-686. Barr, A.J., J.H. Goodnight. J.P. Sall, and J.T. Helwig(1976) A Guide to SAS 76. SAS Institute, P.O. Box 10066, Raleigh, NC. Dekel, N., R. Hultborn, T. Hillensjo, L. Hamberger, and P.F. Kraicer (1976a)Effect of luteinizing hormone on respiration of the preovulatory cumulus-oophorus of the-rat. Endocrinology, 98t498-504. Dekel. N., G. Messer, and S. Dabush (1976b)Effect of LH on the fine structure of rat preowlatory cumulus oophorus. Proceedings of the 6th European Congress of Electron Microscopy, pp. 596-597. Dekel, N., and P.F. Kraicer (1978)Induction in uitro of mucification of rat cumulus oophorus by gonadotropins and adenosine 3'. 5'-monophosphate. Endocrinology. 102:1797-1802. Dekel, N., T. Hillensjo, and P.F. Kraicer (1979)Maturational effects of gonadotropins on the cumulus-oocyte complex of the rat. Biol. Reprod., 20:191-197. Dixon, W.B., and W.B. Brown (1977)Biomedical Computer Programs (P-Series). University of California Press, Berkeley. Eppig, J.J. (1979) Gonadotropin stimulation of the expansion of cumulus oophori isolated from mice: General conditions for expansion in uitro. J. Exp. Zool., 208..111-120 Gilula, N.B.. M.L. Epstein, and W.H. Beers (1978) Cell-tocell communication and ovulation: A study of the cumulus-oocyte complex. J. Cell. Biol., 78t58-75. Greenwald, G.S. (1973) Quantitative aspects of follicular development in the untreated and PMS-treated cyclic hamster. Anat. Rec., 178:139-143. Gwatkin, R.B.L.. O.F. Andersen, and C.R. Hutchinson (1972)Capacitation of the hamster spermatozoa in uitrot The role of cumulus components. J. Reprod. Fert. 3Ot389-394. Gwatkin, R.B.L., and O.F. Andersen (1973)Effect of glycosidase inhibition of the capacitation of hamster sperm by cumulus cells in uitro. 3. Reprod. Fertil., 35:565-567. Hay, M.F., and D.G. Cran (1978) Differential response of components of sheep Graafian follicles to atresia. Ann. Biol. Anim. Biochim. Biophys., 18t453-460. Hillensjo, T., N. Dekel, and P.F. Kraicer (1976) Effects of gonadotropin on the cumulus oophorus of isolated rat graafian follicles. Acta Physiol. Scand., .96:558-569. Humason, G.L. (1967) Animal Tissue Techniques. 2nd ed. W.A. Freeman and Co., San Francisco, p. 160. Lindner, H.R.. A. Tsafriri, M.E. Lieberman, U. Zor, Y. Koch, S. Bauminger. and A. Barnea (1974)Gonadotropin action on cultured Graafian follicles: Induction of maturation division of the mammalian oocytes and differentiation of the luteal cell. Recent Prog. Horm. Res., 3W9-138. Ohno. S., and J.B. Smith (1964) Role of fetal follicular cells in meiosis of mammalian oocytes. Cytogenetics, 3:324-333. Sheela Rani, C.S., and N.R. Moudgal(1977) Role of the proestrous surge of gonadotropins in the initiation of folliculm maturation in the cyclic hamster: A study using antisera to follicle stimulating hormone and luteinizing hormone. Endocrinology, 101:1484-1494. Snedecor, G.W., and W.G. Cochran (1967)Statistical Methods. 6th ed. Iowa State University Press, Ames, Iowa. Thibault, C., M. Gerard, and Y. Menezo (1975)Preovulatory and ovulatory mechanisms in oocyte maturation. J. Reprod. Fertil., 45:605-610. Ward, M.C. (1946) A study of the estrous cycle and the breeding of the golden hamster, Cricetus auratus. Anat. Rec.. 94:139-161. Zamboni, L. (1970) Ultrastructure of mammalian oocytes and ova. Biol. Reprod. Suppl., 2t44-63.