landek: anatomy of hemiurus crenatus. 17 



The large sperm-morulce of Hemiurus crenatus (Plate 2, Fig. 18) 

 measure about 18 /* in diameter. The first chauge indicative of the 

 development of their cells into spermatozoa is seen in the nuclei, which 

 first become oval, then wedge shaped, and later comma shaped (Figs. 18- 

 20). Shortly after this change in the form of the nuclei begins, the cell 

 boundaries become indistinct and at length disappear (Fig. 19). The 

 wedge-shaped and comma-shaped nuclei have their more pointed ends 

 directed toward the centre of the sperm-morula. The nuclei continue 

 to elongate until finally they become thread-like and so intertwined as to 

 resemble a dense chromatic reticulum occupying the central portion 

 of the sperm-morula, which thus takes on an appearance very much like 

 that of a large cell with a single large nucleus (Plate 2, Fig. 23), The 

 intertwining nuclei at first form a nearly spherical mass, but later they 

 appear to become arranged in the form of a shallow cup. The tails of 

 the spermatozoa, which are outgrowths of the nucleus, in some manner 

 break through the cytoplasm at a point corresponding with the centre 

 of the concave surface of the cup-shaped mass, and protrude from the 

 sperm-morula, while the heads remain within, or partly within, the 

 cytoplasm, till all the spermatozoa have taken up this position and form a 

 sheaf-like mass (Plate 3, Fig. 26). After the spermatozoa have detached 

 themselves from the sperm-morula, the cytoplasm apparently assumes a 

 nearly spherical form. These cytoplasmic remains of sperm-morulce are 

 quite common in the testes ; their cytoplasm stains but slightly, and is 

 apparently homogeneous, showing, however, occasionally a few small 

 vacuoles (Plate 3, Fig. 27). Wright and Macallum state that in Sphyra- 

 nura the cytoplasm degenerates, becomes fluid, and is probably partly 

 absorbed by the developing spermatozoa. 



The entire spermatozoon appears to be formed from the nucleus, 

 though I have not been able to follow the development of the thread- 

 like nuclei into mature spermatozoa.^ Wright and Macallum find the 

 entire spermatozoon to be developed from the nucleus in the species 

 which they studied, while Schwarze, on the other hand, states that the 

 head is formed from the nucleus and that the cytoplasm becomes trans- 

 formed into the tail. 



The spermatozoa apparently remain together in the sheaf-like bundles 

 until they leave the testis. The mature spermatozoa measure 20 fx in 



^ In view of the results of studies on spermatogenesis in other invertebrates, as 

 well as in vertebrates, this appearance cannot have great weight. Only extended 

 observations, directed to this one point, on favorable material, could give trustworthj 

 evidence on this difficult question. 



VOL. XLV. — NO. 1 2 



