GERM CELLS IX PEDICELLINA AMERICANA 39 



miissen demnach als einfache Chromosomen ncben den anderen 

 lo Doppel Chromosomen vorhanden sein, was auch thatsachlich 

 gefunden worden ist. Ich finde also in dieser Erscheinung eine 

 Bestatigung der Ansicht, dass einerseits bei der Reducktion eine 

 Vereinigung von Chromosomen zu zwei und zwei stattfindet, 

 anderseits, dass hierbei Chromosomen sich paarweise von jedem 

 der Elternindividuen vereinigen. 



In Hke manner, Cannon, '03, in his study of the spermato- 

 genesis of the hybrid peas pointed out that as in Pcripatus, etc., 

 in the anaphases of the last sporogonous division of both hy- 

 brids and the pure form " Fillbasket," the chromosomes unite in 

 pairs, thus reducing the number to one half. No such associa- 

 tion into pairs is ever seen in the telophases of the normal 

 somatic mitoses. 



Still further, Farmer and Moore, '03, Lotsy, '04, Gregory, 

 '04, Gregoire, '04, and Berghs, '04, and finally Strasburger, 

 '04, himself, have in a large number of forms with but slight 

 variations, found the same character of reducing process and 

 ascribe to the synaptic phase the same significance pointed out 

 above. 



It is quite clear then, that in order to understand the reduction 

 process we must turn to the early germ-cells (the oogonia and 

 spermatogonial), and not to the oocytes and spermatocytes ; for, 

 in the latter, the number has already been reduced. There is, 

 therefore, at the very beginning of the growth period, a condi- 

 tion of the chromatin which is known, making possible the 

 study of the later processes. It is in this regard that the classic 

 work of Hacker, Riickert and the other pioneers in the minute 

 study of the chromatin of the germ-cells was weakest ; for cor- 

 rect as their interpretation of the bivalents is now seen to be, 

 the origin of these structures was based on a very broad as- 

 sumption which later research has entirely failed to confirm. 

 In the presence of the characteristic synaptic phases synchro- 

 nous with, or immediately following the last oogonial telophases, 

 in these several Copcpods and the other Arthropods generally, 

 and from the subsequent appearance of the reduced number of 

 bivalents in the early growth period, it is quite evident that the 



