1903.] On iht Reduction Phenomena of Animals and Plants. 105 



chromosomes becomes lost at the close of the cycle preceding the 

 synaptic rest ; and they are replaced by half the somatic numbers of 

 new ones which first become visible in the prophase of the heterotype 

 division. During their formation these reduced chromosomes become 

 longitudinally divided twice, in planes at right angles to each other, and 

 in this way the frequently observed rapid succession of hetero- and 

 homotype divisions is accounted for, the fission for both mitoses having 

 taken place almost simultaneously. Many instances have been described 

 of the synaptic chromosomes presenting the appearance of doubly split 

 threads, and this second view of the reduction process consequently 

 appears to rest upon quite as sound a basis of observation as the first. 

 As they stand, the results of existing investigations upon these matters 

 would appear to lead to a belief that the reduction processes are 

 diverse in character and cannot possess the theoretical importance 

 which it has been suggested may attach to them. 



A re-investigation of the whole matter has, however, convinced us 

 that neither of the above interpretations is correct. We find that the 

 appearances, hitherto supposed by nearly all observers* to indicate the 

 mode of formation of the heterotype chromosomes, do not seem to have 

 been correctly apprehended, this being due to certain important stages 

 in the process having escaped adequate appreciation. The entire 

 metamorphosis is admittedly excessively difficult to observe, but 

 the whole may be traced in a manner which seems to leave no room 

 for doubt that both the appearance of a double longitudinal fission, 

 and the origin of the heterotype daughter-chromosomes from split 

 threads, are quite illusory. Moreover, in the numerous plants and 

 animals in which we have followed out the process, we have 

 obtained remarkably concordant results, and we think that the ex- 

 planation we are now able to give of the origin of the reduced chromo- 

 somes is one which incidentally accounts for the many peculiar 

 figures and vagaries that have so frequently been observed and 

 remarked upon. 



At the end of the synaptic rest, that is, in the prophase of the 

 heterotype division, the spireme thread certainly undergoes a longi- 

 tudinal fission (Figs. 1 and 2), often before it segments into the reduced 

 number of chromosomes (Osmunda, Lily, Aneura, among plants; 

 Salamander, Axolotl, Blatta, among animals). Following, however, 

 upon this fission, which may result in a wide divarication of the 

 longitudinal halves of the split thread work, comes a stage when the 



* Schaffner, in 1897 (' Bot. Gazette ') suggested that in Lilium Philadelpliicum 

 the process of reduction is essentially identical with that described in this paper. 

 The value of his correct observations was obscured by errors in other respects, and 

 they have lain practically unnoticed, perhaps on this account. 



Korsohelt (' Zeit. Wiss. Zool.,' vol. 60, 1895) describes a peculiar process for 

 Ophryotrochathat is easily brought into line with our observations, though hitherto 

 it has been regarded as an isolated a:id peculiar case. 



