346 Transactions of the Society. 



a pair of m-chroinosomes : on the other hand figs. 2, 5, 6, 3 indicate 

 stages in the disintegration of the nucleolar remnant ; and other 

 sections showed many more granular bodies. In fig. 7 it is not 

 possible to recognize more than 26 elements with certainty. 

 Having made clear the possible sources of error, I should quote 

 this number tentatively for the diploid complex ; this is slightly 

 greater than the number characteristic of Amphibia, uniformly 

 found in the Urodela, and, according to Loyez, in Anguis and 

 Laccrta. As the last-named author gives no figures, and proffers 

 the observation quite incidentally to her main purpose, serious 

 importance is not due to it in view of the difficulty in estimating 

 heteromorphic complices of high number. One suggestion may 

 however be submitted — namely, that the three pairs of chromosomes 

 which show subterminal attachment to the spindle are compound. 

 It is not my purpose to discuss the possible value of studying the 

 heterotype division from this standpoint for the elucidation of 

 numerical chromosome relations in phylogeny. If the above specu- 

 lation is sustained by subsequent research, the correct diploid 

 number for reptiles would be primitively 26 -f- 6, or 32, for which 

 the haploid equivalent would be 16, that cited by Jordan for the 

 Chelonian Cistudo. 



Close scrutiny of the mitotic figures given will reveal a clear 

 correlation in size and shape of the component elements of the 

 different complices. To appreciate this constancy it is of course 

 necessary to take into account the fact that the shape of any 

 element is conditioned by the plane in which it is viewed; and 

 when, as in the present instance, the number of chromosomes is 

 considerable, one rarely finds showing every member of the group 

 lying in the same plane without prolonged and patient searching. 

 Thus the estra pair of apparently J-shaped elements in fig. 2 are 

 seen on changing focus to be bent rods. Furthermore, comparison 

 must always be made at precisely the same stage of condensation 

 upon material, showing as far as possible identical condition as 

 regards preservation. Fig, 4, it should be noted, is an early 

 anaphase. Thus purely negative criticism of the constancy of 

 heteromorphic groups such as that of Foot and Strobell is of little 

 weight in comparison with the positive testimony of a large l)ody 

 of workers in the same field. 1 have considered the possibility 

 that the granular bodies referred to are supei'uumerary chromo- 

 somes such as were originally described by Stevens in insects and 

 in one vertebrate type, Necturus, by King. This is excluded by 

 their inconstancy. There does not appear to be an unpaired 

 element nor a chromatin nucleolus corresponding to such a 

 structure in the spermatocytes ; neither has the study of the 

 reducing divisions indicated the presence of X or Y chromosomes. 

 That the inconstant (? nucleolar) bodies are seen to pass to the 

 poles of the sjjindle is interesting in the light of Carleton's 



