30 THOS. H. MONTGOMERY jr., 



mined. But it is probable that the coalescence is produced by fusion 

 of chromosomes end to end, rather than in any other way, since in 

 the postsynapsis the chromosomes have at least double the length 

 characteristic of them at the commencement of the synapsis, without 

 corresponding increase in diameter. It is unfortunate that all the 

 details of this important process cannot be determined. 



The nucleus in the postsynapsis, owing to its numerous, sharply- 

 marked chromatin loops, has considerable resemblance to a spirem 

 stage (see Figs. 78 — 88). But an important difference is found by 

 the fact that in the postsynapsis these loops are not confined to the 

 periphery of the nucleus, but cross the nuclear cavity in all directions. 

 The contours of the loops are not smooth, but slightly roughened, 

 and the diameter of a loop frequently varies at different points on 

 its surface; their ends may or may not be pointed. Their form is 

 very irregular, but the majority have a resemblance to a U or a V 

 (Figs. 78 — 88). Their diameter is less than in the synapsis. The 

 ends of these chromosomes most frequently terminate against the 

 nuclear membrane; perhaps this is to be explained by assuming that 

 the chromosomes are more or less elastic, tending to straighten out 

 as far as possible; but more probably, owing to the fact that they 

 are not as a rule apposed to the nuclear membrane and so could 

 hardly be regarded as elastic, it is due to a possible connection by 

 linin threads with the nuclear membrane. 



The idiozome during this stage has the same position as in the 

 preceding {Id. 8 Figs. 78, 79, 85, 86). 



There occur globules in the cytoplasm of the spermatocytes, which 

 may be termed yolk globules, since their mode of origin is the same 

 as that of the yolk in ova, and since they stain in the same way; 

 but they do not become blackened by the action of osmic acid. They 

 must be either nutritive masses, or products of nutritive metabolism, 

 since their first appearance is coincident in point of time with the 

 beginning of the growth period of the spermatocytes. These yolk 

 globules (Fand Yk.Gl Figs. 73, 76, 77) arise about the synaptic and 

 postsynaptic stages, though their most marked period of growth 

 coincides with the telophase (Figs. 90—93). At their first origin (in 

 the synapsis) they appear as small granules forming masses of varying 

 dimensions ( F Figs. 73, 77, 92, 93), are homogeneous, and stain 

 like the cytoplasm. Subsequently neighboring ones fuse together to 

 form larger, dense, homogeneous, spherical or oval globules {Yk. Gl 

 from Fig. 90 on). These are the mature yolk globules, and though 



