350 Transactions of the Society. 



definitive chromosomes of the heterotype prophase has provided 

 more fruit for controversy than any other stage in the meiotic 

 phase ; but from the material examined the nature of the process 

 in Sphenodon was readily elucidated. Heterotype prophase nuclei 

 were abundant and well preserved. The point of unique interest 

 here concerned is the reference of the longitudinal cleavage in the 

 diplotene filaments to the axis of fission of the derivative chromo- 

 somes in the maturation division which ensues. In a large number 

 of animals the plane of division in both succeeding mitoses is 

 undoubtedly the same as that in which the long axis of the 

 diplotene segments lies — that is to say, the lieterotype chromo- 

 somes divide by the drawing apart of each half of the longitudinally 

 split diplotene filaments ; and the subsequent division is normal. 

 This fully coincides with the requirements of the parasynaptic 

 view, since the fission of the diplotene loops lies in precisely the 

 same plane as the interspace between the conjugating elements in 

 synapsis. In diakinesis (late heterotype prophase) the chromo- 

 somes assume the form of rings in the case of Sphenodon. The 

 annular type of bivalent according to the earlier telosynaptic view 

 was thought to be formed individually from a single diplotene 

 loop by the junction of its free extremities after the dissolution of 

 the bouquet ; and since the rings segment diametrically it thus 

 seemed that the segregation of homologous chromosomes could 

 only be effected on the assumption that the diplotene filaments 

 consisted of univalent halves united end-to-end. This view was 

 put forward by Meves for Amphibia, Farmer and Moore for 

 Pcriplaneta and Fishes, and by various workers on Orthoptera. Sub- 

 sequent study by Jannsens, Morse and other investigators has 

 however demonstrated that a much more intricate and elaborate 

 process intervenes with, in some cases, a rapidity that readily 

 explains the failure of earlier workers to obtain the entire sequence 

 of events. The diplotene loops are first abbreviated into stout 

 double rods, and then drawn apart along the line of cleavage to 

 form the rings, each half retaining contiguity with the other by 

 its extremities only. Sometimes in the process of division the 

 halves become detached by one extremity first, thus confirm- 

 ing the impression of telosynaptic junction, when previous events 

 have escaped observation. In any case the diameter of the ring 

 represents the actual line of cleavage in the diplotene stage. This 

 is precisely what occurs in Sphenodon, in which the heterotype 

 prophase is reminiscent of Amphibian spermatogenesis. 



The actual course of events in Sphenodon is as follows : — After 

 a period of rest in the bouquet stage the diplotene loops become 

 detached, losing their polar orientation ; they become less curved 

 and progressively abbreviated, and finally assume, in anticipation 

 of the formation of the definitive bivalents of the first spermatocyte 

 mitosis, the form of straight rods or arcs. Many figures clearly 



