A Preliminary Account of the Spermatogenesis of Spltenodon. 34T 



demonstration of the continuity of an intranucleolar element. 

 Carleton's suggestive work makes it imperative that the whole 

 subject of supernumerary, and it may be added, many instances of 

 m-chromosomes, should be handled very carefully, particularly in 

 the absence of specific staining diagnosis of chromatin ; it would 

 be profitable to repeat such observations on material fixed by the 

 newer methods introduced for the study of the extranuclear cell 

 inclusions. 



Tlie nuclear history of the first spermatocyte in Sphenodon so 

 closely resembles earlier meiotic stages of Anguis that Trinci's 

 excellent illustrations of the latter may be referred to as typical 

 of the salient features of corresponding phenomena in the former. 

 As in Vertebrates generally, the seriation of the various stages of 

 maturation roughly corresponds to the radial axis of each testi- 

 cular tubule ; but for practical purposes no assistance is afforded 

 by the structure of the testis in the study of individual stages in 

 the sequence of events which intervene between the last sperma- 

 togonial telophase and the first reducing division. The order of 

 these can only be recognized by the characteristic peculiarities of 

 the synaptic processes and actual continuity based on the study 

 of a large number of cells. 



A few late spermatogonial telophase figures indicate as in 

 fig. 8, which is somewhat diagrammatic, that the telophase chromo- 

 somes spin out directly into the fine attenuated and convoluted 

 filaments of the earliest meiotic stage or leptonema without the 

 intervention of a diffuse condition. Since with the exception of 

 the interphase (resting nucleus) structure the leptotene threads are 

 more difficult to fix than any other stage in the history of chromo- 

 somes in the germ cell cycle, the value of the data obtained is 

 limited by the technique employed, and none of the preparations 

 examined were in such a perfect condition that the number of - 

 loops could be estimated in optical sections, as I have been able 

 to do in the study of Peri'planeta. Clearly the number of leptotene 

 loops is very much greater than in the succeeding stages, when 

 reduction has taken place ; clearly, too, the leptotene stage does 

 not partake of the nature of a continuous spireme such as is 

 described by those who advocate telosynapsis. In good sections 

 the leptone loops show the orientation so characteristic of this 

 stage, and, as I have elsewhere insisted, the reduction in number 

 of loops that retain their polarized disposition throughout the 

 synaptic processes can only be effected by lateral approximation. 

 The filaments at first lie peripherally in contact with the nuclear 

 membrane, contracting slightly in the succeeding phase. In 

 process of contraction they become more closely approximated, 

 and reduction in number of the loops is effected thereby ; in a few 

 cases it could be definitely seen that this does actually take place 

 by theii- lateral association in pairs. The slight degree of con- 



2 B 2 



