Spermatogonia and primary diplotene and zygotene spermatocytes contain an electron-dense, finely granular substance which is usually closely associated with mitochondria; small patches of this substance also occur close to the nuclear membrane, often in the nuclear pores, and within the nucleus of primary spermatogonia. The fine structure of this substance is very similar to the fine structure of germ plasm in other stages of development, and since an ontogenetic continuity with germ plasm can be demonstrated, it was concluded that this substance is also germ plasm. The substance disappears about pachytene, earlier than in oogenesis where it persists until mid-diplotene, a difference which may be due to the fact that the oocyte stores large quantities of germ plasm in its cortex for the next generation of primordial germ cells. If the presence of the substance in the nuclear pores and within the nucleus is an indicator of synthesis of germ plasm, then synthesis stops in the secondary spermatogonium, which correlates with the subsequent absence of germ plasm from the pachytene spermatocyte stages. It is suggested that the function of the germ plasm in specifying germ line cells is carried out between the gastrula stage and the beginning of meiosis. The three events which take place during this period are (i) the migration of the presumptive primordial germ cells from the endoderm to the genital ridges, (ii) mitosis of the primordial germ cells and subsequently of the oogonia and spermatogonia in the developing gonads and (iii) preparations for meiosis. It is suggested that the mechanism of action of the germ plasm may be in the control of one or more of these processes. Other types of granular cytoplasmic deposits are also described, and their possible relationship to germ plasm discussed.

In all anuran amphibians which have been closely studied, cells of the germ line from which the gametes will develop are distinguished by the presence of a cytoplasmic factor called germ plasm, which is recognizable with the light microscope because of its affinity for certain stains, or with the electron microscope by its electron-dense, finely granular appearance and its close association with mitochondria (see reviews by Blackler, 1958,1965,1970; also seeMahowald & Hennen, 1971 ; Williams & Smith, 1971; Czolowska, 1972; Kalt, 1973). The fact that this substance does specify germ line cells has been confirmed experimentally, most convincingly, by Smith (1966). Blackler (1970) reviews other supporting experiments. However, although the general role of the germ plasm seems established, the mechanism by which it performs its role is not known, and even the time at which it acts is not defined.

Implicit in the idea of a separate germ line is the concept that germ plasm is continuous from generation to generation (Weismann, 1892) and also therefore, within a generation. Recent studies, to be discussed in detail, have now provided good evidence that germ plasm is present in germ line cells at all stages of the life-cycle. For continuity between generations, it is only necessary for the female germ cells to carry the germ plasm, since it is reasonable to assume that germ plasm, as a cytoplasmic factor, cannot be passed on to succeeding generations through the spermatozoa. However, the sex of the early gonad and the primordial germ cells is not determined until relatively late in development. The question therefore arises as to the fate of the germ plasm in the spermatogenic stages which arise from an initially indifferent primordial germ cell, which is capable of developing into either a male or female sex cell.

The purpose of the work reported here was to compare the behaviour of germ plasm during spermatogenesis with that during oogenesis, as recently described by Al-Mukhtar & Webb (1971) and independently confirmed by us; it was expected that any differences or similarities might be significant in determining whether the germ plasm performs any function during gametogenesis and, if so, what that function is.

The observations we have made show, as far as morphological studies permit, that in the male cells synthesis of germ plasm ceases early in the first meiotic prophase, and germ plasm can no longer be distinguished after zygotene; whereas in females, synthesis of germ plasm continues, and aggregations of germ plasm persist, until mid-late diplotene.

A brief report of this work has already appeared in abstract (Kerr & Dixon, 1973).

The results presented in this study were obtained from an examination of the gonads from metamorphosing tadpoles and immature and adult toads of Xenopus laevis. Developmental stages were based on the normal tables of Nieuwkoop & Faber (1967). Gonads were excised from tadpoles which had been previously anaesthetized with 2,2,2-trimethylsulphonate and from toads which had been decapitated. For histological examination, gonads were fixed with Carnoy’s fixative and then routinely processed and embedded in paraffin and cut at 5 – 7 μm. The sections were stained with either haematoxylin and eosin or Cason Mallory trichrome (Cason, 1950).

For electron microscopy, gonads were fixed in either 2 % glutaraldehyde and 2 % acrolein buffered to pH 7·2 in Sorensen’s phosphate buffer, or 5 % glutaraldehyde and 4% formaldehyde buffered to pH 7·2 in Millonig’s phosphate buffer.

After fixation for 1–2 h at 5 °C, the tissue was washed overnight in ice-cold buffer and post-fixed in 1 % osmium tetroxide for 2 h. Samples were then dehydrated through alcohols and embedded in Araldite. Thick (1 μm) sections were cut on an LKB Ultrotome, stained with either 0·1 % azure B (Dodge, 1964) or 1 % methylene blue in 1 % borax (Richardson, Jarett & Finke, 1960), and were examined with the light microscope. Thin sections were cut using glass knives, collected on naked copper grids and stained with lead citrate (Venable & Coggeshall, 1965). Sections were examined with a Philips EM 300 electron microscope operating at 60 kV.

The testis

Our observations on testis formation agree very closely with those of Witschi (1929) on Rana sylvatica and Cheng (1932) on R. cantabrigensis. In X. laevis, the gonads begin to differentiate sexually at about stage 54–55, when the testis can be distinguished from the ovary, based mainly on the criterion that the testis has an enlarged region at its tip composed of a number of densely packed, basophilic cortical cells (Fig. 1) whereas the ovary develops a central lumen. Other exclusive features which distinguish testes from ovaries are the medullary development of the testis and the cortical development of the ovary (see Witschi, 1929).

Fig. 1.

Early testis from a tadpole undergoing metamorphosis (stage 57) illustrating prominent ‘growth tip’t. Primary spermatogonia are prominent and some somatic cells with flattened nuclei (arrows) are presumably beginning to differentiate into follicle cells.

Fig. 1.

Early testis from a tadpole undergoing metamorphosis (stage 57) illustrating prominent ‘growth tip’t. Primary spermatogonia are prominent and some somatic cells with flattened nuclei (arrows) are presumably beginning to differentiate into follicle cells.

The earliest specimens we describe here are from stage 57 tadpoles, the stage at which metamorphosis first becomes evident and in which the sex of the gonads can be determined unequivocally. During its later development, the testis shortens in length, relative to the kidney, and at stage 66, when metamorphosis is just completed, it has the cylindrical appearance characteristic of the adult. During this developmental period, the seminiferous tubules form. The earliest stages (approximately stage 57) are recognizable mainly because of the associated somatic cells which form circular arrays about single spermatogonia. As development proceeds, the seminiferous tubules become more obvious as the germ cells and associated somatic cells form more definite and discrete circular profiles (Fig. 2) within a section, with the germ cells arranged on the basal laminae.

Fig. 2.

Testis from a stage 63 tadpole in which the two types of spermatogonia are seen. Primary spermatogonia (arrows) have a cytoplasm of light density, while secondary spermatogonia (asterisks) contain prominent nucleoli and present an optically much denser cytoplasm. The smaller and darker nuclei of the somatic cells of the testis surround the spermatogonia, and will form outlines of seminiferous tubules in later stages.

Fig. 2.

Testis from a stage 63 tadpole in which the two types of spermatogonia are seen. Primary spermatogonia (arrows) have a cytoplasm of light density, while secondary spermatogonia (asterisks) contain prominent nucleoli and present an optically much denser cytoplasm. The smaller and darker nuclei of the somatic cells of the testis surround the spermatogonia, and will form outlines of seminiferous tubules in later stages.

The cells of the testis

Three different stages in spermatogenesis were examined for their role as potential carriers of germ plasm : spermatogonia, spermatocytes and spermatids. For our own convenience, we define primordial germ cells as germ cells in an indifferent gonad, following Nieuwkoop (1947). Once the sex of the gonad can be distinguished, we call these cells spermatogonia or oogonia, as appropriate. We are not aware of any less arbitrary method of distinguishing between these stages. In the premetamorphic testis, two types of spermatogonia can be recognized : (a) primary spermatogonia which occur singly and which have a lightly stained, highly lobed nucleus and (b) secondary spermatogonia which occur in small clusters and are optically denser with a less polymorphic nucleus (Fig. 3). Spermatogonia are the only germ cells in the premetamorphic testis and persist in the adult testis. Our interpretations agree with previously published descriptions of spermatogonial stages in Anura, e.g. Bufo arenarum (Burgos & Mancini, 1948; Burgos & Fawcett, 1956); X. laevis (Kalt, 1973). In X. laevis, Reed & Stanley (1972) distinguished two cell types in the testis, primordial germ cells and spermatogonia, which appear to be the equivalent of primary and secondary spermatogonia respectively.

Fig. 3.

Spermatogonium from a 2-year-old adult toad, illustrating a lobular cell nucleus and denser cytoplasm than the primary spermatogonium (cf. Fig. 7). Note the lack of electron-dense nuage material within the nucleus and the relatively small number of nuclear pores containing dense material (arrows), (cf. primary spermatogonium). Only a few mitochondria (w) are associated with patches of dense material in the cytoplasm. The dense cytoplasmic substances consist of coarsely granular deposits, gb, sometimes associated with more finely granular material (asterisk), nl: nucleolus; g: single electron-dense cytoplasmic granules.

Fig. 3.

Spermatogonium from a 2-year-old adult toad, illustrating a lobular cell nucleus and denser cytoplasm than the primary spermatogonium (cf. Fig. 7). Note the lack of electron-dense nuage material within the nucleus and the relatively small number of nuclear pores containing dense material (arrows), (cf. primary spermatogonium). Only a few mitochondria (w) are associated with patches of dense material in the cytoplasm. The dense cytoplasmic substances consist of coarsely granular deposits, gb, sometimes associated with more finely granular material (asterisk), nl: nucleolus; g: single electron-dense cytoplasmic granules.

Primary spermatocytes were seen in testes taken from maturing frogs two months past metamorphosis, by which time the testis is well developed, suggesting that primary spermatocytes are probably produced shortly after metamorphosis. Primary spermatocytes in adult testes were also examined. The different stages in the maturation of the primary spermatocytes can be distinguished on both cytoplasmic and nuclear criteria. In leptotene spermatocytes, the mitochondria appear quite normal and the nucleus contains finely clumped chromatin and a fibrillar nucleolus. Zygotene spermatocytes are recognizable because the cytoplasm has a prominent Golgi body and a few, flattened vesicles, the chromatin is more heavily clumped than in leptotene and synaptinemal complexes are present, but as yet not prominent (Fig. 4). In pachytene spermatocytes, the cytoplasm has more flattened vesicles, the mitochondria appear quite changed, and in the nucleus, the synaptinemal complexes are obvious and the chromatin is more clumped. Diplotene spermatocytes were not recognizable and secondary spermatocytes as such were also rare, although spermatocytes in division were sometimes encountered. We were unable to determine whether this was the first or second meiotic division.

Fig. 4.

Portion of the cytoplasm of a zygotene primary spermatocyte from an adult testis. Few mitochondria are associated with deposits of intermitochondrial material (asterisks), and free patches of dense material are not present in the cytoplasm (cf. primary spermatogonium, Fig. 7). Nuclear pores (np) do not contain nuage material. Numerous flattened vesicles (long arrows) are positioned close to the plasma membrane, and Golgi bodies, (G) with associated vesicles are also present in the cytoplasm, sc, synaptinemal complexes.

Fig. 4.

Portion of the cytoplasm of a zygotene primary spermatocyte from an adult testis. Few mitochondria are associated with deposits of intermitochondrial material (asterisks), and free patches of dense material are not present in the cytoplasm (cf. primary spermatogonium, Fig. 7). Nuclear pores (np) do not contain nuage material. Numerous flattened vesicles (long arrows) are positioned close to the plasma membrane, and Golgi bodies, (G) with associated vesicles are also present in the cytoplasm, sc, synaptinemal complexes.

In both tadpole and toad testes, the cytoplasm of secondary spermatogonia has many fibrous deposits, and the dense, granular intermitochondrial substance is not as prominent (Fig. 3). In post-metamorphic testes, the intermitochondrial deposits tend not to be as frequent or as obvious. Spermatids are recognised because the chromatin is evenly dispersed through the nucleus, in later stages becoming denser as it begins to condense. Our descriptions of the characteristics by which the different stages can be recognized are in substantial agreement with previous accounts (Burgos & Fawcett, 1956; Reed & Stanley, 1972; Kalt, 1973).

The germ plasm

In primary spermatogonia, the cytoplasm contains conspicuous aggregations of small numbers of mitochondria closely associated with a dense, finely granular substance, situated usually 1 –2 μm from the nuclear envelope (Figs. 57). The granular substance usually lies between mitochondria; i.e. a number of mitochondria are grouped around the periphery of a granular deposit, and often the mitochondrial membranes are modified, either showing an increased density and distinctness, or contrarily, long discontinuities in the membrane. In some mitochondria, the surface in contact with the granular substance is flattened and the interior is distorted by large vesicles (Figs. 5, 7). Around the aggregations of mitochondria and the dense granular substance, small clear vesicles (0 ·15— 0 ·2 μm diameter) frequently are visible (Fig. 5). Ribosomes are also present, but they appear no more numerous than elsewhere in the cytoplasm.

Fig. 5.

High magnification view of part of the cytoplasm from a spermatogonium in a stage 57 tadpole, showing differences between the three main types of granular cytoplasmic inclusions. Close to the nuclear envelope (ne), two mitochondria are associated with a dense, finely granular deposit similar in appearance to the material associated with the nuclear pores (asterisks). Another patch of more granular, less-dense material is farther from the nucleus and is also associated with mitochondria disposed around three sides, but there is not the same intimate association as with the first substance. The third type of inclusion consists of single large granules (arrows) dispersed randomly through the cytoplasm, n, Nucleus.

Fig. 5.

High magnification view of part of the cytoplasm from a spermatogonium in a stage 57 tadpole, showing differences between the three main types of granular cytoplasmic inclusions. Close to the nuclear envelope (ne), two mitochondria are associated with a dense, finely granular deposit similar in appearance to the material associated with the nuclear pores (asterisks). Another patch of more granular, less-dense material is farther from the nucleus and is also associated with mitochondria disposed around three sides, but there is not the same intimate association as with the first substance. The third type of inclusion consists of single large granules (arrows) dispersed randomly through the cytoplasm, n, Nucleus.

Fig. 6.

(A) High-magnification view of cytoplasm and part of nucleus (n) from a spermatogonium in a stage 57 tadpole, showing details of the fine structure of dense, finely granular intermitochondrial substance and its association with mitochondria. Note isolated piece of membrane (arrowed), smooth vesicles (v) and large single granules (g). (B) Similar substance in an oogonium, for comparison, × 24000.

Fig. 6.

(A) High-magnification view of cytoplasm and part of nucleus (n) from a spermatogonium in a stage 57 tadpole, showing details of the fine structure of dense, finely granular intermitochondrial substance and its association with mitochondria. Note isolated piece of membrane (arrowed), smooth vesicles (v) and large single granules (g). (B) Similar substance in an oogonium, for comparison, × 24000.

FIGURE 7

(A) Spermatogonium from a stage 57 tadpole, resting on basement membrane (bm). Nucleus is very lobed; nucleolus is large and conspicuous and granules (g) appear to be derived from it (see inset (B) for higher ( × 28,000) magnification view of granules). Dense material within the nucleus (thick arrow) resembles substance within nuclear pores, which sometimes protrudes into the cytoplasm (asterisk). Similar bodies occur free in the cytoplasm (thin arrows) and are intimately associated with mitochondria, towards the top of the micrograph. See also inset (C) for higher ( × 28,000) magnification view of mitochondria; note semicylindrical bodies of unknown origin and function associated with the mitochondrial cluster. A more granular substance (s) of lower electron-density lies in the entrance to the deep nuclear invagination.

FIGURE 7

(A) Spermatogonium from a stage 57 tadpole, resting on basement membrane (bm). Nucleus is very lobed; nucleolus is large and conspicuous and granules (g) appear to be derived from it (see inset (B) for higher ( × 28,000) magnification view of granules). Dense material within the nucleus (thick arrow) resembles substance within nuclear pores, which sometimes protrudes into the cytoplasm (asterisk). Similar bodies occur free in the cytoplasm (thin arrows) and are intimately associated with mitochondria, towards the top of the micrograph. See also inset (C) for higher ( × 28,000) magnification view of mitochondria; note semicylindrical bodies of unknown origin and function associated with the mitochondrial cluster. A more granular substance (s) of lower electron-density lies in the entrance to the deep nuclear invagination.

In secondary spermatogonia, particularly those from mature testes, the substance is not as intimately associated with mitochondria and frequently has a fibrous appearance (Fig. 3). In primary spermatocytes, small amounts of the substance are associated with mitochondria in leptotene, but the substance is rarely seen in zygotene and even more rarely in pachytene spermatocytes. It does not occur at all in spermatids.

In primary spermatogonia, particularly from tadpole testes, small patches of the dense, granular material identical in appearance to the intermitochondrial substance are common close to the nuclear envelope, often apparently issuing from the nuclear pores, on the nuclear side of which similar deposits can be seen (Figs. 5, 7). In secondary spermatogonia and in primary spermatocytes, the free cytoplasmic patches are much less frequent and there is rarely any material of this nature associated with the pores or in the nucleus (Figs. 3, 4). André & Rouiller (1956) have called this substance ‘nuage-material’.

In the zygotene spermatocyte, the mitochondria start to take on a different appearance; they become dense and laterally flattened, the number of cristae is commonly reduced and the spaces between the membranes of single cristae become prominent because of their relative lack of density (Fig. 4). In spermatids, the modifications are similar but greater in degree; in some mitochondria, no cristae can be distinguished and instead most of the interior of the mitochondrion is taken up by a clear, more or less circular space. The mitochondria are also dispersed from their predominantly juxta-nuclear position in early primary spermatocytes towards the periphery of the cell ; in spermatids, the mitochondria then again come to lie close to the posterior end of the nucleus.

Other cytoplasmic inclusions

A second type of cytoplasmic inclusion differs from the germ plasm in forming large, roughly circular (i.e. spherical) bodies up to 3 μm in diameter, which are usually located 1 – 2 μm from the nuclear membrane or, more rarely, situated peripherally (Fig. 5). They occur in spermatogonia from both tadpole and frog testes and in primary spermatocytes, but have not been seen in spermatids. The substance of which the bodies are composed is less electron-dense, less compacted and much more obviously granular than the intermitochondrial substance, the individual particles having a diameter of 20 – 40 nm, although some particles are larger than this, perhaps as a result of fusion. Occasionally, in post-metamorphic testes, there is a more finely granular component as well, resembling in texture but not in electron density, the intermitochondrial substance (Fig. 8).

Fig. 8.

Spermatogonium of a testis from a 2-year-old toad showing dense, finely granular patches (arrowed) associated with mitochondria but not as closely as in the early stage 57 testis (see Fig. 7). Note aberrant appearance of some of the mitochondria (asterisks) and coarsely granular body (gb) associated with more finely granular substance.

Fig. 8.

Spermatogonium of a testis from a 2-year-old toad showing dense, finely granular patches (arrowed) associated with mitochondria but not as closely as in the early stage 57 testis (see Fig. 7). Note aberrant appearance of some of the mitochondria (asterisks) and coarsely granular body (gb) associated with more finely granular substance.

Both these types of bodies have been described previously by Al-Mukhtar & Webb (1971) and Reed & Stanley (1972) as ‘nucleolus-like bodies’ and by Kalt (1973) as analogous to the chromatoid body of mammalian sperm, described by Fawcett, Eddy & Phillips (1970).

Although these bodies have some resemblance to the dense intermitochondrial substance, their texture and density are not sufficiently similar to support the thesis of a close relationship, unless they represent a different phase of the same material. Lacking any evidence from more suitable techniques such as autoradiography (see, for example Eddy & Ito (1971) and Weakley (1971) for work on oocytes, discussed later), we regard them as separate bodies. In this conclusion we are in agreement with Kalt (1973), who has discussed in detail the origin and affinities of this body.

The cytoplasm of primary spermatocytes contains a large number of flattened vesicles, approximately 0 · 25 μm in length, which appear very suddenly. They tend to be concentrated near the periphery of the cell and are often aligned close to (ca. 10 nm) and parallel with the plasma membrane. Reed & Stanley (1972) have previously reported the presence of these vesicles, and although they noted that the vesicles were more numerous in the peripheral regions of the cell, they did not comment on any relationships with the plasma membrane.

Other nuclear inclusions

In addition to the patches of dense, finely granular material resembling the intermitochondrial substance, the nucleus of the primary spermatogonium also sometimes contains a particulate, granular substance found close to the nucleolus, giving the impression that the nucleolus is fragmenting (Fig. 7B). Similar observations have been made in the oogonium (J. B. Kerr, unpublished data).

On the periphery of these aggregations, there are other patches of material with a texture and density similar to the intermitochondrial substance. Again, in the absence of any stronger evidence to connect these particulate bodies with the intermitochondrial substance, we regard them as unrelated to the germ plasm.

These studies have shown that in X. laevis, electron-dense finely granular intermitochondrial deposits occur in spermatogonia and, in much smaller amounts, in primary spermatocytes up to pachytene. Our observations largely confirm and amplify those of Reed & Stanley (1972) that the granular intermitochondrial substance is more abundant in cells of the mature testis, that it is not present in stages later than the primary spermatocyte and that it is often associated with the nuclear pores.

Reed & Stanley (1972) have drawn attention to the similarity between the granular deposits in spermatogonia and primary spermatocytes, and the granular substances reported in primordial germ cells, oogonia and primary oocytes of X. laevis by Al-Mukhtar & Webb (1971). We agree that the substances appear morphologically identical, and independently have confirmed the observations of Al-Mukhtar & Webb (1971) as indicated in part in Fig. 6B. We also believe that the substance is identical with that composing the dense, granular, intermitochondrial deposits described in X. laevis primary oocytes by Balinsky & Devis (1963), in two-cell embryos by Czolowska (1972) and in various stages by Kalt (1973). A summary of the descriptions is given in Table 1, from which we conclude that the same substance exists in primordial germ cells, in oogonia and oocytes, and in two-cell embryos. The substance Czolowska described is the so-called germ plasm, and a number of light-microscope studies have shown that germ plasm occurs in germ cells up to the time of entry of the presumptive primordial germ cells into the genital ridges. We conclude therefore on the basis of this ontogenetic history, that the substance we have described here in spermatogonia and early primary spermatocytes is germ plasm. In the light of observations that primordial germ cells are indifferent (i.e. neither male nor female) for some time after arrival in the genital ridges, and that the sex of tadpoles can be reversed (see Witschi, 1951), it is not surprising that the early spermatogenic stages contain germ plasm. These conclusions support those of Kalt (1973) who, in a study of several different stages of development of X. laevis, has demonstrated the continuity of germ plasm through the life cycle and has also concluded that germ plasm is present in the male germ cells.

Table 1.

Descriptions of the ultrastructure of germ plasm in Xenopus laevis

Descriptions of the ultrastructure of germ plasm in Xenopus laevis
Descriptions of the ultrastructure of germ plasm in Xenopus laevis

These findings are strengthened by the observations that in a number of species and in several different developmental stages of Rana, dense, granular intermitochondrial substances, identical to those reported here have been described (see Table 2). Although there are minor differences in these descriptions, it is clear that the substances described are morphologically identical, within the limits of the slight changes noted by Mahowald & Hennen (1971) in Rana embryos at different stages. This substance has been shown to be germ plasm by careful comparative light microscopy and electron microscopy, and by experimental studies (for review, see Blackler, 1970). Taken together, these studies indicate that the germ plasm and the dense, granular intermitochondrial deposits in gonocytes are one and the same substance.

Table 2.

Descriptions of the ultrastructure of germ plasm in Rana spp.

Descriptions of the ultrastructure of germ plasm in Rana spp.
Descriptions of the ultrastructure of germ plasm in Rana spp.

In the primary spermatogonia of X. laevis, the nucleus and the nuclear pores contain a substance -nuage material (André & Rouiller, 1956)-which is morphologically similar to the small cytoplasmic juxta-nuclear deposits, and these in turn are similar to the intermitochondrial substance. This correlation, although the criteria for identification are few, permits us to propose, in agreement with others (see for example, Clérot (1968), Eddy & Ito (1971),

Al-Mukhtar & Webb (1971)) that the germ plasm or a component of it, is synthesized in the nucleus and then transferred to the cytoplasm as nuage material. Some support for this proposal comes from the autoradiographic results of Eddy & Ito (1971) and Weakley (1971) and from the cytochemical investigations of Kalt (1973). The final proof for this interpretation will have to come from experiments using techniques like these.

The purpose of this study was to determine whether germ plasm occurs in male germ cells, and if so, whether there are any differences between germ plasm in spermatogenesis compared to oogenesis, in the expectation that if there were differences, they might provide a clue to some aspects of the biology of germ plasm. The major difference lies in the length of time for which the germ plasm is present in its usual form (i.e. as a distinctive, dense, granular aggregate intimately associated with mitochondria). In the female germ line, germ plasm is present in this form through prophase into mid-diplotene, when the oocyte has a diameter of approximately 350 μm, as reported by Balinsky & Devis (1963) and Kalt (1973) (see also Dumont, 1972; Van Gansen & Schram, 1972). Kalt (1973) estimates that the amount of germ plasm in the oocyte increases through diplotene. In the male germ line, however, the amount of germ plasm begins to decline early in prophase I, probably during leptotene (see also Kalt, 1973).

Another difference between male and female sex cells is in the length of time for which the nuage material is present. In males, the amount of nuage material is greatest in primary spermatogonia, whereas secondary spermatogonia and early primary spermatocytes (up to zygotene) contain very little, if any. In females, nuage material is present in late diplotene oocytes (stage 4 of Dumont (1972); diplotene D of Van Gansen & Schram (1972); Kalt (1973)). If, as we have suggested earlier, the presence of nuage material is an indication of the intra-nuclear synthesis of germ plasm, then synthesis in males occurs for a much shorter time than it does in females -a conclusion which is consistent with our earlier finding that germ plasm persists for a longer time in females than in males. In both sexes, synthesis of germ plasm takes place when the germ cells are dividing rapidly, presumably to satisfy the requirement that cytoplasmic levels are maintained.

Balinsky & Devis (1963) have shown that in late diplotene oocytes, the germ plasm is dispersed to the cortical region. In the spermatocyte, the fate of the germ plasm could not be determined. Its disappearance coincides with the production of large numbers of flattened vesicles which later appear to fuse with the plasma membrane. The disappearance of the germ plasm in the spermatocyte therefore bears some resemblance to the events which take place in oocytes (Balinsky & Devis, 1963), but whereas the germ plasm is stored in the cortex of the oocyte, to be released at fertilization, there has never been any suggestion that the sperm carries germ plasm.

From a study made in our laboratory of the behaviour of germ plasm during early embryogenesis, Whitington & Dixon (1974) suggest that the germ plasm is not active during cleavage. Since, in both males and females, the germ plasm is either absent or dispersed in the cortex after the beginning of meiosis, it seems reasonable to suggest that the part of the life-cycle in which the germ plasm is active is sometime between gastrulation and meiosis. The three events which take place during this period are (i) the migration of the presumptive primordial germ cells from the endoderm to the genital ridges, (ii) mitosis of the primordial germ cells and subsequently of the oogonia and spermatogonia in the developing gonads and (iii) preparations for meiosis. The actual mechanism of action of the germ plasm may be in the control of one or more of these processes. At the moment however, it is only possible to describe the function of the germ plasm in such general terms as ‘to specify the germ cells’.

Dense granular intermitochondrial substances have often been reported from the germ cells of various animals (Table 3). But, whereas in anuran amphibians, in particular Xenopus and Rana, a well-defined ontogenetic continuity has been established which is consistent with the general functions which have been ascribed to the germ plasm as a result of experimental studies, as yet, similar claims cannot be made for substances with similar morphology in other animals. Furthermore, Sauzin (1966) and Morita, Best & Noel (1969) have reported similar dense granular intermitochondrial substances in planarian blastema cells, and Chaigneau (1971) has observed a similar substance, again in association with mitochondria, in cells of a suspected endocrine organ in Isopods. In both of these instances, the cells involved are somatic cells; caution must therefore be exercised in any attempt to relate the conclusions from the study of germ plasm in anuran amphibians to the role of morphologically similar material in other groups of animals.

Table 3.

Descriptions of dense granular cytoplasmic deposits closely associated with mitochondria, in germ cells of animals (excluding Anura)

Descriptions of dense granular cytoplasmic deposits closely associated with mitochondria, in germ cells of animals (excluding Anura)
Descriptions of dense granular cytoplasmic deposits closely associated with mitochondria, in germ cells of animals (excluding Anura)

On the other hand, it would be prudent to consider here the applicability of suggested functions for these substances in other animals to the problem of the role of germ plasm in Anura. Two functions have been proposed: (a) in mitochondriogenesis, particularly when the intermitochondrial substance occurs in oocytes and (b) as a precursor of the chromatoid body, in spermatocytes.

We cannot disagree with the first suggestion, since it is an attempt to explain the intimate morphological relationships between the mitochondria and the intermitochondrial substance, a correlation which we have not attempted. However, although the number of mitochondria probably increases during some stages of spermatogenesis, particularly in the secondary spermatogonium, the increase, if any, appears to be of the same order as that for normal somatic cell division where germ plasm is not required. Furthermore, Chase & Dawid (1972) have shown that the amount of mitochondrial protein and DNA (and therefore presumably the number of mitochondria) increases after hatching, a proliferation which takes place in the absence of intermitochondrial germ plasm, at least in the somatic cells of the embryo. Eddy & Ito (1971) have discussed other aspects of this question more fully. Furthermore, on the basis of our current knowledge of anuran germ plasm, this seems unlikely to be the sole function of the substance in anuran germ cells, since it does not explain the effect of removal or inactivation of the germ plasm in preventing the presumptive primordial germ cells from reaching the genital ridges.

The second suggestion, originally made by Fawcett et al. (1970), seems to us to be more tenuous, even in animals other than Anura. Kalt (1973) has summarized the major part of the evidence against the homology of germ plasm with the chromatoid body, although he has not been able to rule out the possibility that one substance is a precursor of the other. Recent autoradiographic studies by Weakley (1971) suggest that in hamster oocytes, the intermitochondrial substance incorporates both lysine and uridine, whereas the granulo-fibrillar bodies which are the precursors of the chromatoid body do not incorporate lysine and incorporate uridine only occasionally. In a similar study of tadpole oocytes, Eddy & Ito (1971) did not differentiate between the ability of the granulo-fibrillar bodies and the intermitochondrial mass to incorporate tritiated amino acids. However, this evidence suggests that the chromatoid body arises from the granulo-fibrillar masses, the intermitochondrial substance having a separate function, as yet undefined.

This paper was written with the aid of the library facilities of the Marine Biological Laboratory, Woods Hole, Massachusetts. We express our thanks to Miss B. Züst for advice and access to her unpublished results and also to Mr G. Sobels for able technical assistance.

Adams
,
E. C.
&
Hertig
,
A. T.
(
1964
).
Studies on guinea pig oocytes. I. Electron microscopic observations on the development of cytoplasmic organelles in oocytes of primordial and primary follicles
.
J. Cell Biol
.
21
,
397
^427.
Al-Mukhtar
,
K. A. K.
&
Webb
,
A. C.
(
1971
).
An ultrastructural study of primordial germ cells, oogonia and early oocytes in Xenopus laevis
.
J. Embryol exp. Morph
.
26
,
195
217
.
André
,
J.
&
Rouiller
,
C.
(
1956
).
L’ultrastructure de la membrane nucléaire des ovocytes de l’Araignée (Tegeneria domestica Clark
).
In Electron Microscopy
(ed.
F. S.
Sjostrand
&
J.
Rhodin
), pp.
162
164
.
New York
:
Academic Press
.
Balinsky
,
B. I.
(
1966
).
Changes in the ultrastructure of amphibian eggs following fertilization
.
Acta Embryol. Morph, exp
.
9
,
132
154
.
Balinsky
,
B. I.
&
Devis
,
R. J.
(
1963
).
Origin and differentiation of cytoplasmic structures in the oocytes of Xenopus laevis
.
Acta Embryol. Morph, exp
.
6
,
55
108
.
Blackler
,
A. W.
(
1958
).
Contribution to the study of the germ cells in the Anura
.
J. Embryol. exp. Morph
.
6
,
491
503
.
Blackler
,
A. W.
(
1965
).
The continuity of the germ line in amphibians and mammals
.
Année biol
.
4
,
627
635
.
Blackler
,
A. W.
(
1970
).
The integrity of the reproductive cell line in the Amphibia
.
Curr. Top. Devi Biol
.
5
,
71
87
.
Brôkelmann
,
J.
(
1963
).
Fine structure of germ cells and Sertoli cells during the cycle of the seminiferous epithelium in the rat
.
Z. Zellforsch. mikrosk. Anat
.
59
,
820
850
.
Burgos
,
M. H.
&
Fawcett
,
D. W.
(
1956
).
An electron microscopy study of spermatid differentiation in the toad Bufo arenarum Hensel
.
J. biophys. biochem. Cytol
.
2
,
223
240
.
Burgos
,
M. H.
&
Mancini
,
R. E.
(
1948
).
Ciclo espermatogenico anual del Bufo arenarum Hensel
.
Revta Soc. argent. Biol
.
24
,
328
336
.
Cambar
,
R.
,
Delbos
,
M.
&
Gipouloux
,
J.-D.
(
1970
).
Premières observations sur l’infrastructure des cellules germinales à la fin de leur migration dans les crêtes génitales chez les Amphibiens Anoures
.
C. r. Séanc. Soc. Biol
.
164
,
1686
1688
.
Cason
,
J. E.
(
1950
).
A rapid one-step Mallory-Heidenhain stain for connective tissue
.
Stain Technol
.
25
,
225
226
.
Chaigneau
,
J.
(
1971
).
L’organe de Bellonci du Crustacé Isopode Sphaeroma serratum (Fabricius)
.
Z. Zellforsch. mikrosk. Anat
.
112
,
166
187
.
Chase
,
J. W.
&
Dawid
,
I. G.
(
1972
).
Biogenesis of mitochondria during Xenopus laevis development
.
Devi Biol
.
27
,
504
518
.
Cheng
,
T. H.
(
1932
).
The germ cell history of Rana cantabrigensis Baird
.
Z. Zellforsch. mikrosk. Anat
.
32
,
497
541
.
Clérot
,
J. C.
(
1968
).
Mise en évidence par cytochimie ultrastructurale de l’émission de protéines par le noyau d’auxocytes de batraciens
.
J. Microscopie
1
,
973
992
.
Coggins
,
L. W.
(
1973
).
An ultrastructural and autoradiographic study of early oogenesis in the toad Xenopus laevis
.
J. Cell Sci
.
12
,
71
93
.
Czolowska
,
R.
(
1972
).
The fine structure of the ‘germinal cytoplasm’ in the egg of Xenopus laevis
.
Wilhelm Roux Arch. EntwMeeh. Org
.
169
,
335
344
.
De Kretser
,
D. M.
(
1969
).
Studies on the structure and function of the human testis. M.D. Thesis, Monash University, Victoria, Australia
.
Delbos
,
M.
,
Gipouloux
,
J.-D.
&
Cambar
,
R.
(
1971
).
Observations sur l’infrastructure des différents types cellulaires constituant la gonade larvaire chez la Grenouille agile Rana dalmatina Bon. (Amphibien anoure)
.
C. r. hebd. Séanc. Acad. Sci., Paris D
272
,
2372
2374
.
Dhainaut
,
A.
(
1970
).
Etude en microscopie électronique et par auto-radiographie à haute résolution des extrusions nucléaires au cours de l’ovogenèse de Nereis pelágica (Annélide Polychète)
.
J. Microscopie
9
,
99
118
.
Dodge
,
J. D.
(
1964
).
Cytochemical staining of sections from plastic-embedded flagellates
.
Stain Technol
.
39
,
381
386
.
Dumont
,
J. M.
(
1972
).
Oogenesis in Xenopus laevis. 1. Stages of oocyte development in laboratory maintained animals
.
J. Morph
.
136
,
153
180
.
Eddy
,
E. M.
(
1972
).
Fine structural studies on the distribution of nuage material in germ cells of the rat
.
Anat. Rec
.
172
,
304
.
Eddy
,
E. M.
&
Ito
,
S.
(
1971
).
Fine structural and radioautographic observations on dense perinuclear cytoplasmic material in tadpole oocytes
.
J. Cell Biol
.
49
,
90
108
.
Fawcett
,
D. W.
,
Eddy
,
E. M.
&
Phillips
,
D. M.
(
1970
).
Observations on the fine structure and relationships of the chromatoid body in mammalian spermatogenesis
.
Biol. Reprod
.
2
,
129
153
.
Gondos
,
B.
&
Conner
,
L. A.
(
1973
).
Ultrastructure of developing germ cells in the fetal rabbit testis
.
Am. J. Anat
.
136
,
23
42
.
Gondos
,
B.
&
Hobel
,
C. J.
(
1971
).
Ultrastructure of germ cell development in the human fetal testis
.
Z. Zellforsch. mikrosk. Anat
.
199
,
1
20
.
Gondos
,
B.
,
Renston
,
R. H.
&
Conner
,
L. A.
(
1973
).
Ultrastructure of germ cells and Sertoli cells in the postnatal rabbit testis
.
Am. J. Anat
.
136
,
427
440
.
Herbaut
,
C.
(
1972
).
Etude cytochimique et ultrastructurale de l’ovogenèse chez Lithobius forficatus L. (Myriapode, Chilopode); Evolution des constituants cellulaires
.
Wilhelm Roux Arch. EntwMech. Org
.
170
,
115
134
.
Hsu
,
W. S.
(
1962
).
An electron microscope study on the origin of yolk in the oocytes of the Ascidian Boltenia villosa Stimpson
.
Cellule
62
,
145
155
.
Jeon
,
K. W.
(
1972
).
Electron microscopic studies on primordial germ cells in early mouse embryos
.
J. Cell Biol
.
55
,
125
a.
Kalt
,
M. R.
(
1973
).
Ultrastructural observations on the germ line of Xenopus laevis
.
Z. Zellforsch. mikrosk. Anat
.
138
,
41
62
.
Kerr
,
J. B.
&
Dixon
,
K. E.
(
1973
).
Association of cytoplasmic dense particles and meiosis in Xenopus
.
J. Anat
.
114
,
149
.
Kessel
,
R. G.
(
1966
).
An electron microscope study of nuclear-cytoplasmic exchange in oocytes of Ciona intestinalis
.
J. ultrastruct. Res
.
15
,
181
196
.
Kessel
,
R. G.
(
1969
).
Cytodifferentiation in the Rana pipiens oocyte. I. Associations between mitochondria and nucleolus-like bodies in young oocytes
.
J. Ultrastruct. Res
.
28
,
61
77
.
Kessel
,
R. G.
(
1971
).
Cytodifferentiation in the Rana pipiens oocyte. II. Intramitochondrial yolk
.
Z. Zellforsch. mikrosk. Anat
.
112
,
313
332
.
Kress
,
A.
&
Spornitz
,
U. M.
(
1972
).
Ultrastructural studies of oogenesis in some European amphibians. I
.
Rana esculenta and Rana temporaria. Z. Zellforsch. mikrosk. Anat
.
128
,
438
456
.
Lanzavecchia
,
G.
(
1962
).
Organisation of frog oocytes before the yolk synthesis
.
In Proc. Fifth Int. Congr. Electron Microscopy
(ed.
S. S.
Breese
).
New York
:
Academic Press
.
Mahowald
,
A. P.
(
1962
).
Fine structure of pole cells and polar granules in Drosophila melanogaster
.
J. exp. Zool
.
151
,
201
215
.
Mahowald
,
A. P.
&
Hennen
,
S.
(
1971
).
Ultrastructure of the ‘germ plasm’ in eggs and embryos of Rana pipiens
.
Devi Biol
.
24
,
37
53
.
Massover
,
W. H.
(
1968
).
Cytoplasmic cylinders in bull frog oocytes
.
J. Ultrastruct. Res
.
22
,
159
167
.
Miller
,
O. L.
(
1962
).
Studies on the ultrastructure and metabolism of nucleoli in amphibian oocytes. In Proc
.
Fifth. Int. Congr. Electron Microscopy
, (ed.
S. S.
Breese
).
New York
:
Academic Press
.
Morita
,
M.
,
Best
,
J. B.
&
Noel
,
J.
(
1969
).
Electron microscopic studies of planarian regeneration. I. Fine structure of neoblasts in Dugesia dorotocephala
.
J. Ultrastruct. Res
.
27
,
7
23
.
Nieuwkoop
,
P. D.
(
1947
).
Experimental investigations on the origin and determination of the germ cells, and on the development of the lateral plates and germ ridges in urodeles
.
Archs néerl. Zool
.
8
,
1
—205.
Nieuwkoop
,
P. D.
&
Faber
,
J.
(
1967
).
Normal Table of Xenopus laevis (Daudin)
, 2nd ed.
Amsterdam
:
North Holland Publ. Co
.
Odor
,
D. L.
(
1965
).
The ultrastructure of unilaminar follicles of the hamster ovary
.
Am. J. Anat
.
116
,
493
522
.
Ornstein
,
L.
(
1956
).
Mitochondrial and nuclear interaction
.
J. biophys. biochem. Cytol
.
2
(Suppl),
351
352
.
Reed
,
S. C.
&
Stanley
,
H. P.
(
1972
).
Fine structure of spermatogenesis in the South African clawed toad Xenopus laevis Daudin
.
J. Ultrastruct. Res
.
41
,
277
295
.
Richardson
,
K. C.
,
Jarett
,
L.
&
Finke
,
E. H.
(
1960
).
Embedding in epoxy resins for ultrathin sectioning in electron microscopy
.
Stain Technol
.
35
,
313
323
.
Rowley
,
M. J.
,
Berlin
,
J. D.
&
Heller
,
C. G.
(
1971
).
The ultrastructure of four types of human spermatogonia
.
Z. Zellforsch. mikrosk. Anat
.
112
,
139
157
.
Sauzin
,
M. J.
(
1966
).
Etude au microscope électronique du néoblaste de la Planaire Dugesia gonocephala (Turbellaire, Triclade) et de ses changements ultrastructuraux au cours des premiers stades de la régénération
.
C. r. hebd. Seanc. Acad. Sci., Paris D
263
,
605
608
.
Schwalm
,
F. E.
,
Simpson
,
R.
&
Bender
,
H. A.
(
1971
).
Early development of the kelp fly, Coelopa frígida (Díptera). Ultrastructural changes within the polar granules during pole cell formation
.
Wilhelm Roux Arch. EntwMech. Org
.
166
,
205
218
.
Smith
,
L. D.
(
1966
).
The role of a ‘germinal plasm’ in the formation of primordial germ cells in Rana pipiens
.
Devi Biol
.
14
,
330
347
.
Szollosi
,
D.
(
1972
).
Changes of some cell organelles during oogenesis in mammals
.
In Oogenesis
(ed.
J. D.
Biggers
&
A. W.
Schuetz
), pp.
47
64
.
Baltimore
:
University Park Press
.
Van Gansen
,
P.
&
Schram
,
A.
(
1972
).
Evolution of the nucleoli during oogenesis in Xenopus laevis, studied by electron microscopy
.
J. Cell Sci
.
10
,
339
367
.
Venable
,
J. H.
&
Coggeshall
,
R.
(
1965
).
A simplified lead citrate stain for use in electron microscopy
.
J. Cell Biol
.
25
,
407
408
.
Weakley
,
B. S.
(
1969
).
Granular cytoplasmic bodies in oocytes of the golden hamster during the post-natal period
.
Z. Zellforsch. mikrosk. Anat
.
101
,
394
400
.
Weakley
,
B. S.
(
1971
).
Basic protein and ribonucleic acid in the cytoplasm of the ovarian oocyte in the golden hamster
.
Z. Zellforsch. mikrosk. Anat
.
112
,
69
84
.
Weismann
,
A.
(
1892
).
Aufsâtze fiber Vererbung und verwandte biologische Fragen
.
Uber die Vererbung (1883)
, pp.
73
121
.
Jena
:
Gustav Fischer
.
Whitington
,
P.
&
Dixon
,
K. E.
(
1974
).
J. Embryol. exp. Morph. (In the Press
.)
Williams
,
M. A.
&
Smith
,
L. D.
(
1971
).
Ultrastructure of the ‘germinal plasm’ during maturation and early cleavage in Rana pipiens
.
Devi Biol
.
25
,
568
580
.
Witschi
,
E.
(
1929
).
Studies on sex differentiation and sex determination in amphibians. I. Development and sexual differentiation of the gonads of Rana sylvatica
.
J. exp. Zool.
52
,
235
265
.
Witschi
,
E.
(
1951
).
Gonad development and function. Embryogenesis of the adrenal and the reproductive glands
.
Recent Prog. Horm. Res
.
6
,
1
—23.
Yamamoto
,
K.
&
Onozato
,
H.
(
1965
).
Electron microscope study on the growing oocyte of the gold fish during the first growth phase
.
Mem. Fac. Fish. Hokkaido Univ
.
13
,
79
106
.
Zaffagnini
,
F.
&
Lucchi
,
H. L.
(
1970
).
Osservationi ultrastrutturali sul determinante germinale dei Dafindi
.
Atti Accad. Naz. Lincei Rc. Ser
.
849
,
141
146
.
Zhand
,
J. P.
&
Porte
,
A.
(
1966
).
Signes morphologiques de transfert de materiel nucléaire dans le cytoplasme des ovocytes de certaines espèces de Poissons
.
C. r. hebd. Séanc. Acad. Sci., Paris D
262
,
1977
1978
.
Zamboni
,
L.
(
1972
).
Comparative studies on the ultrastructure of mammalian oocytes
.
In Oogenesis
(ed.
J. D.
Biggers
&
A. W.
Schuetz
), pp.
5
45
.
Baltimore
:
University Park Press
.