1905.] of Permanent Forms among Chromosomes, etc. 567 



no half-way set of gemini between the forms a and b given in the table on 

 p. 566, or between c and d, e and /, and so on. 



The gemini in Triton, as we believe is the case in every other instance of 

 the first maiotic division, are produced by the conjugation of premaiotic 

 {somatic) chromosomes in pairs during the synaptic rest ; * and since there 

 are 24 premaiotic chromosomes in the particular instance of Triton, it follows 

 that there must be only four individual chromosomes which can unite with 

 each other to form the two gemini belonging to each of the six types. 



The above results, based upon a study of the gemini in the first maiotic 

 divisions in Triton, are interesting in themselves, but they immediately raise 

 the further question as to whether the order here observed is simply a 

 curious instance, or an individual expression of a wider law; on account of 

 this we have studied in a similar manner the first maiotic division in man, 

 rats, and Periplaneta ; that is to say, in two more typical vertebrates and a 

 representative arthropod. 



In the testes of rats the first maiotic division occurs in groups of cells, 

 and it is by no means difficult to bring under observation in a short time 

 thirty or forty instances of the early spindle and late prophase. In this we 

 have material amply sufficient to arrive at a decision upon the matters with 

 which we are concerned. Fig. 3 is a drawing of a portion of a tubule from 

 the testes of a piebald rat. It shows the outer wall of the tubule with some 

 premaiotic nuclei and three dividing cells belonging to the first maiotic 

 (heterotype) mitosis. 



In these it will be seen that the gemini are of very different forms, 

 and that as in Triton the same forms are repeated in different 

 individual cells. Analysis of a large number of similar cells reveals the 

 fact that in this particular example there are again six varieties of gemini 

 {see table, p. 566). In the rat, however, instead of the premaiotic 

 chromosomes being 24 in number as in Triton, there are 32 ; consequently 

 we shall have to ascertain the relative numbers of the different types. 

 Further it will be seen that the six types present in the rat are not all the 

 same as those catalogued for Triton (table, p. 566). It will indeed be obvious 

 from this that only four of the Amphibian varieties are represented in the 

 mammal ; to these four two new forms of gemini are added. 



In regard to the type h in the table, p. 566, it is obvious that this particular 

 form of the gemini might be regarded as an opened out 'J, which in this case 

 appears as a straight rod ; but although this is so, the fact remains that the 



* Farmer and Moore, ' Roy. Soc. Proc.,' 1903, loc. cit. ; Farmer and Moore, ' Quart. 

 Journ. Micr. Sci.,' vol. 48, loc. cit. ; Moore and Embleton, ' Roy. Soc. Proc.,' 1905, loc. cit. ; 

 Moore and Walker, ' Thomson Yates Reports,' loc. cit. 



