4o8 NATURAL SCIENCE. June, 



It is, however, obvious that the probability of this or of any other 

 process representing in reality a true reduction, depends absolutely on 

 the universality of its occurrence. If, therefore, the process now put 

 forward by Hacker, Riickert, and Vom Rath, as the " Reductions- 

 theilung " can be proved to be in no sense universal, then we are not 

 only without assurance that the process described by these authors is 

 a " Reductions-theilung " at all, but there arises the greatest proba- 

 bility that, the final phenomena of maturation being thus conclusively 

 demonstrated to be diverse, the theoretical value of reduction is 

 annihilated altogether. 



In 1894 the work which directly substantiated this new type of 

 reduction was almost if not entirely restricted to observations made 

 upon arthropods by the above-mentioned three authors. In 1894, 

 however, Vom Rath (4) published a paper on the spermatogenesis of 

 Salamandva maculosa, which not only flatly contradicted the previous 

 careful work of Flemming (6), but which, if true, went far to establish 

 the wide-spread and possibly universal occurrence of what we may 

 call " the new reduction " in animal cells. It was the most important 

 paper that the supporters of this type of reduction had produced, 

 since in it Vom Rath unhesitatingly declared that the processes pre- 

 viously described in Arthropoda, also occurred in Amphibia in full. 



It was impossible for me at the time to investigate the subject 

 anew ; but I subsequently (7) became quite convinced that so far as the 

 maturation process in elasmobranchs was concerned, tetrad formation 

 and reduction in Hacker's sense did not exist. It is hardly probable 

 on the face of it that the cardinal points in the spermatogenesis of 

 amphibians and elasmobranchs differ widely each from each ; and the 

 diametrically opposed character of Vom Rath's work and my own, led 

 naturally to the enquiry whether one or other set of observations was 

 not altogether wrong. 



The answer to this question appears in a recent paper by Meves (8), 

 who, with the full concurrence of Flemming, refutes almost the whole 

 of Vom Rath's work. He shows in the first place that Vom Rath has 

 fallen into the grossest error concerning the number of post-synaptic 

 generations, that tetrad formation is in Amphibia, as Flemming had 

 long ago ascertained, a rare abnormality, generally associated with 

 degeneration phenomena. Lastly, and most important of all, he 

 shows that the final division is a true example of longitudinally split, 

 not transversely separating, chromosomes (fig. 8) — being in fact the 

 very type from which Flemming originally made his description of the 

 homotypic form of division. There is moreover, as Meves himself 

 points out, the closest agreement between the processes I describe in 

 elasmobranchs and the spermatogenesis of Amphibia. We have thus 

 two more types which rank with Brauer's observations upon Ascaris (9), 

 as directly contradicting the universality of the "Reductions-theilung " 

 described by Hacker. Such being the case, it is obvious that these 

 authors can no longer maintain the universal occurrence of this process, 



