![Thumbnail: [Volume 87, 1959 page 57]](/journals/rsnz/images/rsnz_87/thumbnails/rsnz_87_01_0076_0057_ac_02.gif)
Observations
The Spermatocyte
In Pl. 6, fig. 3 is part of a spermatocyte showing about eleven mitochondria, mostly chondriomites. In a juxta-nuclear position (n) is a mass of Golgi bodies (g) and small vesicles belonging to the Golgi apparatus. In the lamellated dictyosomes at (g), there are three to five complete lamellae (the so-called collapsed saccules), but these dictyosomes consist largely of aggregated small vesicles of Dalton and Felix (1953). In places, masses of electron opaque vesicles occur, but no lamellae can be seen, as at (g) right. This ill-defined condition persists until spermio-genesis begins, the typical spermatid dictyosome being shown in Pl. 7, fig. 5 (g). In the chondriomites, the cristae run in the length of the mitochondrion, but have sidelines giving a herring-bone effect. This is well shown in Pl. 6, fig. 1 (m), where the matrix is seen to be denser than is usual in the mitochondria of higher animals. Ergastoplasmic strands can be seen in places (in Pl. 6, fig. 3 (ep)), but associated punctate granules are not usually clear in the present material.
In the vicinity of the Golgi apparatus are spherical or ovoid bodies, (x) in Pl. 6, fig. 3, containing fine vesicles, indicating that some form of secretion has already appeared in the spermatocyte. It is considered that these vesicles are proacrosomic granules, and the full grown acrosome (Pl. 7, fig. 5, a) is formed by fusion of such elements.
Second Maturation Division
In Pl. 8, fig. 10, and Pl. 9, fig. 12, are two metaphase stages Other stages are not yet known. In Pl. 9, fig. 12, the metaphase stage is oriented so that the centriole astral region would be at six o'clock, but the centriole has not been cut across. at (g) is part of the Golgi apparatus, the partly vesiculated condition being evident. The object in the lower part of this cell is not a centriole but is a part of the Golgi apparatus. It has been known to light microscopists for many years that, during dictyokinesis, the Golgi bodies in some cases become associated with spindle or astral fibres, and are thus shepherded into the daughter cells in a sub-equal manner, so that each spermatid gets its moiety. In Pl. 8, fig. 10 (h), upper, the centromere attached to the chromosome is clear: it consists of a pyramidal body down the centre of which is a canal (see also Text-figs. 15 and 16, H). The centromere continues on into a hollow spindle fibre, one of which is clearly shown in Pl. 8, fig. 10 (p) lower left. At (h) upper, the spindle canal is shown attached to the centromere, but cut almost transversely. At (p) right, the astral region has been cut across, but the centriole is not in this section. From the study of a number of micrographs, the amphiastral figure has been reconstructed in Text-fig. 15, and the centromere-spindle attachment and chromosome is drawn semi-diagrammatically in Text-fig. 16. In the micrographs, the canal in the centromere does not appear to connect directly with the hollow spindle fibre, for there is a membrane or stopper at (S) In addition, the spindle fibre tube seems to have side branches (ST) in some cases near the astral region which is here reconstructed from such evidence as that in Pl. 8, fig. 10 (p) left, but the extent of the astral fibres (tubes) is not properly known In all cases where cut suitably, the centromere sits in a crater in the chromosome As fixed in Dalton's chrome-osmic mixture, the chromatin is seen to consist of electron opaque granules and micro-vesicles In Lumbricus the ergastoplasm is scanty as in Pl. 6, fig. 3 (ep), and Pl. 9, fig. 12 (ep), and there is no evidence that this cell element is associated with the spindle fibre tubes, but the possibility cannot be completely excluded on the basis of the present micrographs.
The Structure of the Dictyosome
As will be seen from Pl. 6, fig. 3, and Pl. 9, fig. 12, the dictyosome of the pre-spermatid stages is irregular and shows no more than three or four organised lamellae. The main part of the dictyosome is formed of microvesicles of the same
![Thumbnail: [Volume 87, 1959 page 58]](/journals/rsnz/images/rsnz_87/thumbnails/rsnz_87_01_0077_0058_ac_02.gif)
nature and diameter as the lamella in transverse section. As will be seen, this condition persists during the maturation divisions where, as in Pl. 9, fig. 12 at (e), imperfect lamellae and masses of microvesicles together make up the dictyosome body.
In Pl. 7, fig. 5, in the original photograph at 71,000 X, the very regular lamellated condition of the dictyosome is seen. Assuming that the position of the acrosome (a) marks the ventral surface of the dictyosome, this micrograph would be cut transversely From a study of many examples, it appears that the spermatid dictyosome is formed of a pile of somewhat oval or rectangular plates tending to be of slightly different sizes Text-fig. 17 gives a somewhat inadequate impression of the construction of the dictyosome, Pl. 7, fig. 5 being cut through A-B and Pl. 6, fig. 1 through C-D.
The dictyosome in Pl. 7, fig. 5 appears to be oriented in the sense that the upper region (o) is different from the lower (a), in which lies the acrosome (a) and the proacrosomic material. The presence of the microvesicles (o) cannot always be ascertained, and it is quite certain that the Lumbricus dictyosome is never provided with the elaborate double cover seen in gryllid dictyosomes (Beams, Tahmisian, Devine, and Anderson, 1954), and shown at C1 and C2 in Text-fig. 12.
Nevertheless, in the present material the acrosome always appears on that side of the dictyosome turned partly or wholly away from the nucleus. It will appear to the observer that a model of the Lumbricus late spermatid dictyosome could be made by a pile of eight to twelve irregularly oval or rectangular pieces of three-ply wood. The pieces may be flat, but in such a figure as Pl. 7, fig. 9 (upper right) the plates may be conical or warped That the plates are independent is shown in the acroblast remnant, where (Pl. 7, fig. 7, g) individual parts often drift apart as happens when a pile of dinner plates begins to fall over This phenomenon has been shown to be the case in numbers of other micrographs not published in this or the previous paper with Dalton: this is a matter of considerable importance in trying to assess the function of the dictyosome Previous electron microscopists, and the author and his colleagues (Gatenby, Tasmisian, Devine, and Beams, 1958) have tentatively accepted the idea that the lamellae are really collapsed sacs, such as would be formed when a pile of empty grain sacs is thrown on the ground In Pl. 7, fig. 5, the parts (e) appear to be broken, but as if originating, from the same lamella At (d) in this dictyosome the edges of the lamellae are becoming blebbed, and at both (b) and (d) the blebs appear to originate from single electron opaque lamellae We know from such examples as that in Pl. 6, fig. 2 (g), Pl. 7, fig. 9 (right-upper), and Pl. 7, fig. 7, that the lamellae are independent, and the condition existing in Pl. 7, fig. 5, may be due to bad fixation In Lumbricus, as in gryllids studied, the lamella is undoubtedly formed like three-ply wood, the central layer being electron translucent, the upper and lower layers electron opaque Thus in the micrographs, two dark lines represent the upper and lower layers of a single lamella or saccule.
The Centriole
It is now known that the animal cell centriole is a hollow bead or collar as viewed by the electron microscope. In the present micrographs centrioles appear in Pl. 7, figs. 6, 7, 8 and 9 (c). In all cases they are hollow, and show different stages in the 9 + 1 subdivisions now known to occur universally in cilia and flagella. The centriole of the spermatocyte in Pl. 7, fig. 9 (c) is believed to be elongating, preparatory to binary fission. The subsequent stages have been given by Gatenby and Dalton in the previous paper In the lengthening spermatid, there is evidence that the centriole (c, in Pl. 7, fig. 6) passes forward into a tube which either is, or arises from, the centriole adjunct Inspection of Figs. 6 and 7 will show that at this stage, the centriole is a mere shell, from the back of which emerges the 9 +1 (or 9 + 2) flagellar filaments In Pl. 7, fig. 8 the centriole has become
![Thumbnail: [Volume 87, 1959 page 59]](/journals/rsnz/images/rsnz_87/thumbnails/rsnz_87_01_0078_0059_ac_02.gif)
subdivided into its 9 parts, the central filament still being undeveloped. In Pl. 6, fig. 1 the middle filament has appeared. The method of origin of this filament is unknown. The middle filament is clear in the flagella cut across in Pl. 8, fig. 11, left and upper In Pl. 7, figs. 6 and 7, the middle filament appears to originate from the bottom of the centriole and not from its hollow centre. The two stages intermediate between centriole and flagellum are shown at (c) and (f) in Pl. 7, fig. 8. Where the centriole flagellum complex emerges from the cell, as in (c) in Pl. 7, fig. 5, a thickening of the surrounding cell wall takes place. The peculiar position of the head centriole with reference to the middle-piece is mentioned in the Discussion.
The Acrosome Carrier
In Pl. 9, fig. 14 (a), the acrosome in situ has been cut in transverse section. all the other sections in this figure have cut across the spermatozoa somewhat lower down, that is through the nucleus. In Fig. 13, at (a) upper, the acrosome is passing out of its carrier into the upper part of the cell, preparatory to fusing with the nucleus, at (a) lower, the acrosome is still in the carrier, and just above this and at the right upper (ac) are two spent acrosome carriers At Pl. 8, fig. 11 (a) a normal unspent acrosome carrier is shown.
Pre-Secretion in the Spermatocyte Golgi Field
In Pl. 6, fig. 3 spherical or ovoid bodies marked (x) are found associated with the Golgi apparatus field Identical bodies have been found by Gatenby and Roth in micrographs of the spermatocytes of the snail (Helix aspersa). In the Mammalia, light microscopists have previously described in spermatocytes what have been called intra-archoplasmic spherules, which have been traced into the spermatid as the pro-acrosomic material. In Pl. 9, fig. 12, x, one of these spherical bodies can be seen at (x) imbedded in a cloud of small vacuoles, which shows that such bodies are carried through the maturation divisions presumably a number of these fuse, or in some other way contribute to the formation of the acrosome (Pl. 7, fig. 5 (a)). This is a matter of some importance, for the presence of such presumed pro-acrosomic bodies in the spermatocyte implies that aggregation, but not secretion, of the acrosome has taken place in the spermatid stage, as for example in Pl. 7, fig. 5; it follows that the lamellae of the Golgi apparatus must now have some function other than secretion at the spermatid stage.
The Nuclear Fin
It was shown in the previous paper, that as the ripening spermatid nucleus begins to pass from the spherical to the ovoid condition an aggregation of granules on one side appears and eventually stretches from the anterior to the posterior end of the elongated spermatid. The position of this fin is put in Text-figs. 9–11 (F). In transverse section the fin is shown at (n) in Pl. 9, fig. 14. Its function is not known, but similar structures have been described in spermatozoa of other animals by light microscopists.