THE ANATOMICAL RECORD 227:167-174 (1990)

Staging Equine Seminiferous Tubules by Nomarski Optics in Unstained Histologic Sections

and in Tubules Mounted In Toto to Reveal the S pe rmatog en ic Wave

LARRY JOHNSON, VINCE B. HARDY, AND MICHAEL T. MARTIN Departments of Veterinary Anatomy and Veterinary Large Animal Medicine and Surgery, College of Veterinary Medicine, Institute of Equine Science and Technology, Texas A & M

University, College Station, Texas

ABSTRACT Nomarski optics were used to identify stages of the spermato- genic cycle of seminiferous tubules in sectioned tissue or in whole dispersed tu- bules and to characterize the equine spermatogenic wave. Embedded tissues were sectioned a t 20 pm. Whole dispersed tubules were obtained by enzymatic digestion of thin slices of fresh testis. Dispersed tubules were fixed, dehydrated in graded levels of alcohol, infiltrated with Epon, and mounted in toto on glass slides. Stages of the spermatogenic cycle could be identified under Nomarski optics in both his- tologic sections and tubules mounted in toto. Stage dependent nuclear chromatic and cytoplasmic changes in spermatogonia, spermatocytes, and spermatids were evident. Spermatid development included chromatin condensation, nuclear elon- gation, acrosomal development from the Golgi and proacrosomic granules, migra- tion of the annulus and mitochondria1 alignment, and the transient appearance of the chromatoid body and manchette. Both nuclear and cytoplasmic details of Ser- toli cells were revealed. In tubules mounted in toto, the spermatogenic wave along the length of the tubules occurred as a consecutive set of stages occupying small regions along the tubular length. The spermatogenic wave in the horse is more similar to that of humans than that of rats. The combination of enzymatic isolation of seminiferous tubules and identification of spermatogenic stages by Nomarski optics facilitates examination of the spermatogenic wave in species whose tubules are tightly bound and not easily teased apart.

Whole-mounted rat seminiferous tubules (stained with Harris hematoxylin) have been used t o study cell kinetics of differentiating spermatogonia (Clermont and Bustos-Obregon, 1968; Huckins, 1971). Similar studies using tubules mounted in toto have been con- ducted for hamsters and rams (Lok et al., 1982). Observations of whole tubules also have facilitated isolation and partitioning of tubules into specific sper- matogenic stages and have revealed stage-dependent changes in enzymes, endogenous steroids (Parvinen and Vanha-Perttula, 1972; Parvinen and Ruokonen, 1982; Vihko et al., 19871, and Sertoli cell gene expres- sion (Morales e t al., 1987). Confirmation of stages in whole tubules has been based on histologic observation of the same tubules following embedding, sectioning, and staining (Parvinen and Vanha-Perttula, 1972; Parvinen and Ruokonen, 1982).

Nomarski optics reveal nuclear and cytoplasmic de- tail of unstained cells, and the small (shallow) depth of focus facilitates optical sectioning of tissues (Allen et al., 1969). Capitalizing on these features, Nomarski op- tics have been used to measure the maximum diameter of nuclei totally embedded in 20 pm Epon sections of a variety of testicular cells from humans (Johnson et al., 1980, 1981, 1986, 1989; Neaves et al., 1984), rats

0 1990 WILEY-LISS, INC

(Johnson et al., 1980,1984), and horses (Johnson, 1985, 1986; Johnson and Neaves, 1981; Johnson and Tatum, 1989). Nomarski optics were used on humans (Johnson et al., 1987) and horses (Johnson, 1985; Johnson and Tatum, 1989) to measure the nuclear size of specific germ cells found in certain stages of the spermatogenic cycle. However, these studies did not characterize de- velopmental differences in germ cells among spermato- genic stages and did not address the spermatogenic wave along the length of tubules.

The objectives of this study were (1) to develop a method to identify spermatogenic stages under Nomar- ski optics in tubules mounted in toto, and (2) to study the spermatogenic wave in equine seminiferous tu- bules. It was found that Nomarski optics facilitated staging of unstained seminiferous tubules in histologic sections as well as those mounted in toto and that stages appeared in consecutive order in isolated seg- ments of whole mounted equine tubules.

Received June 20, 1989; accepted October 9, 1989. Address reprint requests to Larry Johnson, Ph.D., Dept. of Veteri-

nary Anatomy, College of Veterinary Medicine, Institute of Equine Science and Technology, Texas A&M University, College Station, TX 77843-4458.

168 L. .JOHNSON ET AL.

MATERIALS AND METHODS

Testes from five adult (4-10 yr) stallions were per- fused with 2% glutaraldehyde in cacodylate buffer. Slices of fixed tissue were further fixed in 1% osmium tetroxide, embedded in Epon 812, sectioned a t 20 pm, and observed unstained by Nomarski optics (Johnson and Neaves, 1981; Johnson, 1985).

Whole-mounted tubules were obtained from enzy- matically digested, fresh testicular parenchyma from five adult (4-10 yr) stallions. Thin (2-8 mm) slices of fresh testicular parenchyma (1 g) were placed in 15 ml of phosphate buffered saline containing 0.075 mg hy- aluronidase, 5 mg pronase (Lok et al., 19821, and 0.875 mg collagenase a t room temperature with shak- ing. After an hour of digestion, dispersed and teased tubules were fixed in 2% glutaraldehyde followed by osmium tetroxide. Tubules were dehydrated in graded levels of alcohol, infiltrated with Epon, mounted in toto in Epon on glass slides, and observed unstained by Nomarski optics.

In both thick histologic sections and tubules mounted in toto, stages of the cycle of seminiferous ep- ithelium were based on the eight-stage classification of Swierstra et al. (1974). This classification is based on the presence of specific germ cell types (i.e., presence of B spermatogonia or secondary spermatocytes and num- ber of generations of primary spermatocytes or sper- matids) and on the changes associated with nuclear elongation, location, and release of spermatids. The steps of spermatid development (Sa, Sb,, Sb,, Sc, Sd,, Sd,) are based on those previously described for the human (Clermont, 1963) and horse (Johnson, 1985). Following tedious teasing of dispersed tubules, tubular segments measuring up to 1.5 cm were obtained. These tubules were mounted in toto as described above and used to characterize the equine spermatogenic wave.

RESULTS Observation of germ cell nuclear chromatic and cy-

toplasmic details in histologic sections by Nomarski optics enabled identification of the spermatogenic cycle (Fig. 1). The identity of stages I through VIII was based on the nuclear shape, bundle formation, luminal align- ment, or release of spermatids (Fig. 1). In stage I, sper- matids with spherical nuclei were the most advanced cell type present. These nuclei were partially elongated in stage I1 and appeared in bundles in stage 111. Stage IV had secondary spermatocytes and meiotic meta- phase figures. When secondary spermatocytes divided, a second generation of spermatids was found in stages V-VIII. Type B spermatogonia were found in stages VI and VII. The Sd, spermatids found in stage VI were still deeply embedded in the Sertoli cell apex and had not migrated toward and lined up along the lumen as had Sd2 spermatids of stage VII. B spermatogonia di- vided at the end of stage VII to give preleptotene pri- mary spermatocytes in stage VIII. At the end of stage VII, the entire length of each Sd, spermatid was pro- jected into the lumen. Residual bodies from recently released spermatozoa lined the lumen and could be seen in transit through columnar Sertoli cell cytoplasm on their way to the base of Sertoli cells, where they remained until digested.

Stage-dependent changes were seen in various germ

cell types (Fig. 1). Several types of A spermatogonia were observed. The largest type of A spermatogonia was seen in stages VII, VIII, I, and 11. The chromatin clumps were small, and a large nucleolus was present. In stage I, A spermatogonial nuclei were spherical and had distinct clumps of chromatin throughout each nu- cleus. Stage-dependent changes in chromatin patterns of primary spermatocytes were seen. These included early polar displacement of chromatin material, thread formations of chromosomes, and number and location of nucleoli. Also, there was a transient appearance of a large spherical Golgi apparatus in late primary sper- matocytes.

The cytoplasm of secondary spermatocytes was not remarkable (Fig. 1). Neither the Golgi apparatus, chro- matoid body, nor mitochondria1 cluster was large enough to be discerned by Nomarski optics. However, the cytoplasm of secondary spermatocytes had discern- ible granular structure.

Spermatids in stages V and VI contained the chro- matoid body and a cluster of mitochondria around one pole of their nuclei (Fig. 1). The Golgi apparatus, dis- tinguished by its spherical shape, and developing ac- rosomic vesicle appeared in stage V. The acrosomic ves- icle, which often indents the nuclear envelope, also was seen in stages VI-VIII, but developed into the acroso- ma1 cap in stage VIII.

The developing tail was attached to the nucleus of late Sa spermatids in stage VIII (not shown). The an- nulus had appeared, but became distinct on these cells in stage I1 (Fig. 1). The manchette appeared as the nucleus elongated in stage I1 and persisted through stage VI. The annulus was located near the attachment of the nucleus to the developing tail until late stage VII, a t which time it migrated to its final location at the junction of the developing middle piece and princi- pal piece of the tail. Shortly after annulus migration, mitochondria migrated around the developing middle piece. In stage VIII, Sd, spermatids could be seen pro- jecting into the lumen. The middle piece of the tail was enlarged with mitochondria and the cytoplasmic drop- let.

In addition to germ cells, Nomarski optics revealed nuclear and cytoplasmic details of Sertoli cells, myoid cells, and Leydig cells (Fig. 1). Leydig cells were char- acterized by long, thin mitochondria, large regions of clear cytoplasm (smooth endoplasmic reticulum), and granules (lipidilipofuscin granules). Myoid cells were thin, flattened cells containing irregular flattened nu- clei. The large, pear-shaped Sertoli cell nucleus con- tained mostly euchromatin and a large nucleolus. Due to long, slender mitochondria streaming from the base to the apex of the cell, Sertoli cell cytoplasm could be distinguished in columns within the seminiferous epi- thelium. The orientation of spermatids between lateral borders or in the apex of a Sertoli cell changed with the stage of the cycle. In stage VIII, residual bodies were observed in the apex of Sertoli cells (lining the lumen) and in transit toward the base of Sertoli cells. Diges- tion granules of these bodies were reduced in volume in stages VI and VII; however, granules were seen at the base of Sertoli cells in all stages.

While many details observed in histologic sections (Fig. 1) could not be discerned in tubules mounted in toto, sufficient cellular detail was evident with Nomar-

E Q U I N E SPERMATOGENIC WAVE 169

ski optics (Fig. 2 ) to allow staging of equine tubules. Due to the difficulty of identifying B spermatogonia, stages V and VI could not be distinguished. Stage I was identified by the presence of Sb, spermatids as the most advanced cell type; for stage 11, it was by the presence of Sb, spermatids. In fact, the classification could be based solely on 1) the presence of specific cell types characteristic of that stage; 2) the spermatid de- velopment, bundle formation, and alignment along the tubular lumen; and 3) the presence of residual bodies.

Differences in stages were observed between sides (top and bottom) of the whole-mounted tubules. The wave of equine seminiferous epithelium along the tu- bular length was revealed as consecutive adjacent stages (Fig. 2). The region occupied by a given stage was quite small but was varied. Also, modulations in the wave (reversal of the order of consecutive stages) occurred.

Modulation occurred at any point in the entire waves (Fig. 3). The number of stages in the modulation varied from one to five, but returned by reversing its order starting with the last stage of the modulation. Latter stages (more developed, higher numbers) were ob- served in the same direction on tubular segments con- taining more than one wave. However, the direction of the rete testis, in relation to the direction of the latter stage, was not determined from these isolated tubular segments.

DISCUSSION Nomarski optics revealed details of germ cell devel-

opment during spermatogenesis. Details observed in thick, unstained histologic sections (Fig. 1) supple- mented those reports based on conventional histologic sections (Swierstra et al., 1974; Johnson et al., 1978) or dispersed equine germ cells (Johnson and Thompson, 1983). Stage-dependent changes in spermatid tail de- velopment, nuclear condensation and elongation, or- ganelles (i.e., Golgi apparatus, mitochondria1 cluster- ing), transient organelles (manchette, chromatoid body, acrosomic vesicle and granule), and acrosomal development of spermatids were visible in 20 km Epon sections viewed by Nomarski optics. Observation of these structures made i t possible to follow spermatids in their different steps of development.

Due to optical sectioning with Nomarski optics and progressive changes within a stage, rarely observed structures could be seen by focusing in the third dimen- sion. Migration of residual bodies from the apical re- gion of the seminiferous epithelium to the base of Sertoli cells is rarely seen in conventional histologic sections of non-rodent species. In stage VIII, residual bodies, left behind at spermiation in stage VIII, could be found at high density in the apex of the seminiferous epithelium, in transit (Fig. l ) , and at the base of the seminiferous epithelium. While migration of residual bodies could not be localized to Sertoli cells in these preparations, it is known that Sertoli cells engulf re- sidual bodies and digest them in their basal cytoplasm (Smith and Lacy, 1959).

The present study revealed that Sa spermatids of stage VIII have developing tails attached to their nuclei. This finding is consistent with the scanning electron microscopic observation of tails from two gen-

erations of spermatids (Sa and Sd,) in the lumen in stage VIII (Johnson et al., 1978).

Optical sectioning by Nomarski optics and the abil- ity to observe cellular structural detail in unstained specimens facilitated identification of stages in tubules enzymatically isolated and mounted in toto (Figs. 2, 3). The greatest advantage of this combination is found in species whose seminiferous tubules are not as straight nor as easily isolated as in the rat (Perey et al., 1961). In such species, the spermatogenic wave can be visu- alized without flat embedding, precise sectioning to ob- tain the lumen throughout the length of the segment, or tedious serial sectioning.

The spermatogenic wave is the spatial, sequential order of stages along the length of seminiferous tubules at any given time (Fig. 3). The origin of the wave is unknown, but it results from synchronized but not si- multaneous division of stem cell spermatogonia in adjacent tubular segments along the length of the seminiferous tubule (Johnson, 1990). The spermatogen- ic wave may function as a mechanism to assure a con- stant release of spermatozoa. This function is particularly important as a constant release of sperma- tozoa is essential to maximize the opportunity for fer- tilization as female gametes only are available intermit- tently (cyclically). Other possible functions of the wave are to 1) reduce competition for hormones and metab- olites used in a given stage; 2) reduce tubular conges- tion that could be produced if spermiation occurred simultaneously along the entire length of the tubule; 3) assure a constant flow of seminiferous tubular fluid to maintain the vehicle for spermatozoan transport and hormones needed by the epididymal epithelium; and 4) facilitate maturation of spermatozoa in the epididymis by a constant flow of spermatozoa and fluid from the testis. Synchronization of spermatogonial division along the length of tubules can be modified by treat- ment as noted by the reduction of the spermatogenic wave to three or four stages by retinol injection in vi- tamin A-deficient rats (Griswold et al., 1988).

As in other species (rat, Perey et al., 1961; human, Schulze and Rehder, 1984), features of the equine sper- matogenic wave explain findings reported in histologic sections. Given that more than one stage often is seen in a particular cross section of the equine seminiferous tubule (Johnson et al., 1978), i t follows that the length of a tubule in a given stage could be quite small (Fig. 3 ) .

In the “unit-segment’’ theory of wave evolution (Perey et al., 1961), long tubular segments are made up of smaller unit segments, all of which are in similar stages of development. Hence, the difference in the length of tubules occupied by a given stage depends on the number of similar unit segments in the same stage. The number of unit segments in a tubular segment is dependent upon the number of adjacent synchronized and simultaneous spermatogonial divisions.

In humans, a small number of unit segments (which do not entirely extend around the tubular wall) are in the same stage. The wave can only be seen if one con- siders the spiral course of the unit segments as differ- ent stages along the length of the tubule (Schulze, 1982; Schulze and Rehder, 1984).

The unit-segment theory of wave evolution is sup- ported in the rat by differences in development within

170 L. JOHNSON ET AI,.

c

ci, ii

EQUINE SPERMATOGENIC WAVE 171

a tubular segment occupied by a single stage (i.e., sub- stage), the variation in size of tubular segments, and differences in stages around the tubule in a given seg- ment (Perey et al., 1961). In the horse (Fig. 3) and human (Clermont, 19631, the unit-segment theory is supported by the occurrence of multiple stages per tu- bular cross section and some stages occupying ex- tremely small regions.

Variation in the size of individual tubular segments occupied by a given stage differs in the horse (Fig. 3), mouse, rabbit (Curtis, 19181, rat (Perey et al., 19611, and bull (Hochereau, 1963) and prevents the size of these segments from being proportional to the duration of that stage. However, when averaged over several tubular segments, length occupied by a stage is consid- ered to be proportional to the duration of that stage (Courot e t al., 1970). In species such as the human, and less so in the horse (which has multiple stages per tu-

Fig. 1. A clockwise arrangement of eight stages of the cycle of the equine seminiferous epithelium observed by Nomarski optics in un- stained, 20 pm histologic sections. Both nuclear and cytoplasmic de- tails are revealed in Leydig cells (LC), Sertoli cells (SC), various types of germ cells, and myoid cells (MC). A spermatogonia (A) of different types and Sertoli cells are found in all stages of the spermatogenic cycle. Classification of spermatid development was based on that de- scribed for humans (Clermont, 1963) and used in the horse (Johnson, 1985). Briefly, the Sa spermatid is the earliest form as it contains a spherical nucleus and a large Golgi with either no acrosomic vesicle, a developing acrosomic vesicle, or an acrosomal cap. The Sb, sperma- tid has a spherical nucleus but also has an attached tail and distinct acrosome covering half of the nuclear surface. The Sb, spermatid has begun nuclear elongation with the appearance of the newly formed manchette. The Sc spermatid has a distinct manchette and a more elongated nucleus than the Sb, spermatid. The Sd, spermatid is un- dergoing final maturation with the removal of the manchette and migration of mitochondria around the tail. The Sd, spermatid is the final form, and a large portion of the cell including the head projects into the tubular lumen in Stage VIII. Stage I is characterized by preleptotene or leptotene (Ll and pachytene (PI primary spermato- cytes as well as Sb, spermatids (Sb,). These spermatids are charac- terized by spherical nuclei, attached acrosomal caps (AC), and devel- oping tails (DT). Stage I1 is characterized by leptotene primary spermatocytes (L), pachytene primary spermatocytes (P), and Sb, spermatids (Sb,). The labelled pachytene primary spermatocyte is displaying its large spherical Golgi apparatus (open arrow). The Sb, spermatid has an elongating nucleus, a manchette (Mn), and an an- nulus (An). Stage I11 has zygotene (Z) and pachytene (PI primary spermatocytes, and bundles of Sc spermatids (Scl. The Sc spermatid has an elongating and condensing nucleus, a distinct manchette (Mn), and annulus (An). Stage IV is characterized by the presence of zy- gotene primary spermatocytes (Zl, diplotene primary spermatocytes or secondary spermatocytes (SS), meiotic figures (MF), and Sc sper- matids (Sc). Stage V is composed of pachytene primary spermatocytes (P), newly formed Sa spermatids (Sa), and Sd, spermatids (Sd,). The Sa spermatid has a distinct Golgi apparatus (G) , developing acrosomic vesicle (AV), and chromatoid body (CB); and Sd, spermatids form bundles which are deeply embedded in the seminiferous epithelium. Stage VI has the B spermatogonium (B), pachytene primary sperma- tocyte (PI, Sa spermatid (Sa) with its grouped mitochondria (GMI, and Sd, spermatid (Sd,) with its annulus (An) and less distinct manchette. Stage VII is characterized by the B spermatogonium (B), pachytene the primary spermatocyte (PI, Sa spermatid @a), and Sd, spermatid (Sd,). Sd, spermatids, the most advanced form, are migrating toward the tubular lumen in this stage. Stage VIII has the newly formed preleptotene (P1) and pachytene (PI primary spermatocytes; the late Sa spermatid (Sa) with its acrosomic granule (AG) and newly at- tached developing tail; and the Sd, the spermatid lining the lumen with its migrated annulus, enlarged middle piece (MP), and attached cytoplasmic droplet (CD). Residual bodies (RBI left behind by spermi- ation of spermatids can be seen near the luminal surface and in tran- sit toward the base within Sertoli cell cytoplasm. Bar equals 10 pm.

bular cross section; Johnson et al., 19781, the average length of tubular segments occupied by a given stage would not be proportional to the duration of the stage or the volume of the testis occupied by that given stage.

The spermatogenic wave in the horse was similar t o that of other species in modulations and possibly direc- tion of the rete testis. Reversal of the wave a t modula- tions was seen in the horse seminiferous tubules, but it returned in sequential order to continue the original wave (Fig. 3). This is consistent with observations for the mouse, rabbit, dog (Curtis, 1918), and rat (Perey et al., 1961). While single off-phase stages occasionally were found in the wave, Curtis (1918) noted that the wave continued unbroken in the order even when tu- bules branched. While we were unable to determine the direction of the wave in relation to the rete testis in the horse, multiple waves were found in the same di- rection. The general course of the waves was in de- scending order from the rete testis in the rat (Perey et al., 1961) and in the mouse, but was irregular in the rabbit (Curtis, 1918).

Considering the high degree of convolution in equine seminiferous tubules (Fig. 31, some adjacent tubular profiles in a given histologic section represent rather closely spaced views of the same tubule as in the hu- man (Amann, 1981). However, due to the short tubular segment occupied by a given stage (Fig. 31, these adja- cent tubular profiles usually are at different spermato- genic stages and represent the spermatogenic capabil- ity of the testis. In fact, in the absence of local degenerative tubules, little histologic or quantitative difference can be detected in different regions of the entire testis in humans (Johnson et al., 1980) and var- ious animal species (Amann, 1970) including horses (Gebauer et al., 1974; Johnson et al., 1978).

In addition to the rat, in which the wave has been most vividly illustrated (Perey et al., 19611, the wave has been reported to occur in the mouse, bull, guinea pig, rabbit, ram, boar, dog, cat (Benda, 1887; Curtis, 19181, marsupials (Furst, 18871, and humans (Schulze, 1982; Schulze and Rehder, 1984). The current paper extends these studies by revealing the spermatogenic wave in the horse.

The spermatogenic wave has been exploited to deter- mine stage-specific changes in the seminiferous epithe- lium (Parvinen, 1982). Both follicle-stimulating hor- mone (FSH) and androgens, essential hormones for spermatogenesis, seem to have different preferential stages in which they act. Also, there are stage-depen- dent differences in the production of androgen-binding protein, maximal secretion of plasminogen activator, and of meiosis-inducing substance (Parvinen, 1982). These kinds of studies were made possible by staging live, dispersed rat seminiferous tubules by transillumi- nation (Parvinen and Vanha-Perttula, 1972). Seminif- erous tubules in domestic species are tightly bound in the testicular parenchyma and are not easily isolated for these types of studies. Recently, transillumination has been successfully employed to stage enzymatically dispersed seminiferous tubules in the horse (Johnson and Hardy, 1989). This procedure should prove fruitful to improve our understanding of spermatogenesis in domestic animals. In the present study, the first step in this process was presented as a method to obtain iso- lated seminiferous tubules from a species whose tu-

172 L. JOHNSON ET AL.

Fig. 2. Consecutive stages of the spermatogenic cycle along the length of a seminiferous tubular segment mounted in toto and ob- served by Nomarski optics (I, VIII, VII, VIVI) or brightfield micros- copy (center). Boxes in the center photograph indicate corresponding regions observed by Nomarski optics. Stage I is identified by a single generation of Sb, spermatids (Sb,) with spherical nuclei as the most advanced germ cells present. Adjacent to a relatively large stage I,

stage VIII is distinguished by the presence of Sd, spermatids (Sd,) largely extending into the lumen and the presence of residual bodies (RB). Stage VII, adjacent to VIII, has Sd, spermatids (Sd,) lining the tubular lumen, but residual bodies have not formed. Stage ViVI is distinguished by newly formed Sa spermatids (Sa) and Sd, spermatids (Sd,) deeply embedded into seminiferous epithelium. Bars equal 100 km for the center photograph or for I, VIII, VII, VIVI.

bules are tightly bound to each other and a method to stage whole mounted tubules by Nomarski optics. With the aid of enzymatic isolation and staging by Nomarski optics, it was revealed that the equine spermatogenic wave was more similar to that of humans than to that of rats. NIH Grant HD16773

ACKNOWLEDGMENTS

Thanks are extended to Dr. Janice S. Grumbles for assistance in preparing and Ms. Judy Gloyna for typing this manuscript. This work was supported in part by

LITERATURE CITED Allen, R.D., G.B. David, and G. Nomarski 1969 The Zeiss-Nomarski

differential interference equipment for transmitted-light micros- copy. Z. Wissen. Mikro. Mikro. Tech., 69r193-221.

Amann, R.P. 1970 Sperm production rates. In: The Testis, Vol. 1. A.D. Johnson, W.R. Gomes, and N.L. VanDemark, eds. Academic Press, New York, pp. 433-482.

Amann, R.P. 1981 A critical review of methods for evaluation of sper- matogenesis from seminal characteristics. J. Androl., 2t37-58.

Benda, C. 1887 Untersuchungen uber den Bau des Funktoinierenden Samenkanalchens einiger Saugetiere und Folgerungen fur die Spermatogenese dieser Wirbeltierklasse. Arch. Mikro. Anat., 30: 49-110.

Clermont, Y. 1963 The cycle of the seminiferous epithelium in man. Am. J. Anat., 112t35-51.

Clermont, Y., and E. Bustos-Obregon 1968 Re-examination of sper- matogonial renewal in the rat by means of seminiferous tubules mounted in toto. Am. J. Anat., 122:237-248.

Courot, M., M.T. Hochereau-de-Reviers, and R. Ortavant 1970 Sper- matogenesis. In: The Testis, Vol. 1. A.D. Johnson, W.R. Gomes, and N.L. VanDemark, eds. Academic Press, New York, pp. 339- 432.

Curtis, G.M. 1918 The morphology of the mammalian seminiferous tubule. Am. J . Anat., 24r339-394.

Furst, C. 1887 Uber die Entwicklung der Samenkorperchen bei den Beutethieren. Arch. Mikro. Anat., 3Ot336-365.

Gebauer, M.R., B.W. Pickett, and E.E. Swierstra 1974 Reproductive physiology of the stallion. 11. Daily production and output of sperm. J . Anim. Sci., 39:732-736.

Griswold, M.D., C. Morales, and S.R. Sylvester 1988 Molecular biol- ogy of the Sertoli cell. Oxford Rev. Reprod. Biol., 10t124-161.

Hochereau, M.T. 1963 Etude comparee de la vague spermatogenet- ique chez le taureau et chez le rat. Ann. Biol. Animale Biochim. Biophys., 3:5-20.

Huckins, C. 1971 Cell cycle properties of differentiating spermato- gonia in adult Sprague-Dawley rats. Cell Tissue Kinet., 4:139- 154.

Johnson, L. 1985 Increased daily sperm production in the breeding season of stallions is explained by an elevated population of sper- matogonia. Biol. Reprod., 32:1181-1190.

Johnson, L. 1986 A new approach to quantification of Sertoli cells that avoids problems associated with the irregular surface. Anat. Rec., 214:231-237.

Johnson, L. 1990 Spermatogenesis. In: Reproduction in Domestic An- imals. P.T. Cupps, ed. Academic Press, New York, 4th edition (In press).

Johnson, L., and V.B. Hardy 1989 Isolation and staging of equine seminiferous tubules by transillumination. J . Androl., 10t41.

Johnson, L., and W.B. Neaves 1981 Age-related changes in the Leydig cell population, seminiferous tubules, and sperm production in stallions. Biol. Reprod., 24t703-712.

Johnson, L., and M.E. Tatum 1989 Temporal appearance of seasonal changes in numbers of Sertoli cells, Leydig cells, and germ cells in stallions. Biol. Reprod., in press.

Johnson, L., and D.L. Thompson, J r . 1983 Age-related and seasonal variation in the Sertoli cell population, daily sperm production, and serum concentrations of follicle-stimulating hormone, lutein- izing hormone, and testosterone in stallions. Biol. Reprod., 29: 777-789.

Johnson, L., R.P. Amann, and B.W. Pickett 1978 Scanning electron and light microscopy of the equine seminiferous tubule. Fertil. Steril., 29t208-215.

Johnson, L., C.S. Petty, and W.B. Neaves 1980 A comparative study of daily sperm production and testicular composition in humans and rats. Biol. Reprod., 22r1233-1243.

Johnson, L., C.S. Petty, and W.B. Neaves 1981 A new approach to quantification of spermatogenesis and its application to germinal cell attrition during human spermiogenesis. Biol. Reprod., 25: 217-226.

Johnson, L., R.M. Lebovitz, and W.K. Samson 1984 Germ cell degen- eration in normal and microwave-irradiated rats: Potential soerm oroduction rates a t different develoomental stem in suer-

Fig. 3. Enzymatically isolated equine seminiferous tubule, fixed in glutaraldehyde followed by osmium tetroxide, infiltrated with Epon, mounted in toto in Epon, and observed by brightfield microscopy. The corresponding drawing reveals the consecutive stages along the length of the tubule. While modulations (reversal of order) occur, adjacent stages are in consecutive order. Latter stages (more devel- oped; higher numbers) are observed in the same direction of two sets of all stages I-VIII. Bar equals 200 pm.

matoge‘nesis. Anat. Rec., 209:501-507. Johnson, L., F.W. George, W.B. Neaves, I.M. Rosenthal, R.A. Chris-

tensen, A. Decristoforo, H.-U. Schweikert, M.V. Sauer, M. Leshin, J.E. Griffin, and J.D. Wilson 1986 Characterization of the testic- ular abnormality in 5 alpha-reductase deficiency. J . Clin. Endo- crinol. Metab., 63t1091-1099.

Johnson, L., H.B. Nguyen, C.S. Petty, and W.B. Neaves 1987 Quan- tification of human spermatogenesis: Germ cell degeneratlon

174 L. JOHNSON ET AL.

during spermatocytogenesis and meiosis in testis from younger and older adult men. Biol. Reprod., 37t739-747.

Johnson, L., J.S. Grumbles, A. Bagheri, and C.S. Petty 1989 Germ cell degeneration during meiosis is related to increased serum FSH concentrations and reduced daily sperm production in aged men. Biol. Reprod., (in press).

Lok, D., D. Weenk, and D.G. De Rooij 1982 Morphology, proliferation, and differentiation of undifferentiated spermatogonia in the Chi- nese hamster and the ram. Anat. Rec., 203:83-99.

Morales, C., S. Hugly, and M.D. Griswold 1987 Stage-dependent lev- els of specific mRNA transcripts in Sertoli cells. Biol. Reprod., 36t1035-1046.

Neaves, W.B., L. Johnson, J.C. Porter, C.R. Parker, J r . , and C.S. Petty 1984 Leydig cell numbers, daily sperm production, and serum gonadotropin levels in aging men. J. Clin. Endocrinol. Metab., 59t756-763.

Parvinen, M. 1982 Regulation of the seminiferous epithelium. En- docr. Rev., 3t404-417.

Parvinen, M., and T. Vanha-Perttula 1972 Identification and enzyme quantification of the stages of the seminiferous epithelial wave in the rat. Anat. Rec., 174t435-450.

Parvinen, M., and A. Ruokonen 1982 Endogenous steroids in the rat seminiferous tubules. Comparison of the stages of the epithelial cycle isolated by transillumination-assisted microdissection. J. Androl., 3:2 11-220.

Perey, B., Y. Clermont, and C.P. Leblond 1961 The wave of the sem- iniferous epithelium in the rat. Am. J. Anat., 108t47-77.

Schulze, W. 1982 Evidence of a wave of spermatogenesis in human testis. Andrologia, 14t200-207.

Schulze, W., and U. Rehder 1984 Organization and morphogenesis of the human seminiferous epithelium. Cell Tissue Res., 237:395- 407.

Smith, B.V.K., and D. Lacy 1959 Residual bodies of seminiferous tu- bules of the rat. Nature, 184:249-251.

Swierstra, E.E., M.R. Gebauer, and B.W. Pickett 1974 Reproductive physiology of the stallion. I. Spermatogenesis and testis compo- sition. J . Reprod. Fertil., 40t113-123.

Vihko, K.K., J. Toppari, and M. Parvinen 1987 Stage specific regula- tion of plasminogen activator secretion in the ra t seminiferous epithelium. Endocrinology, 120t142-145.