GERM CELLS IN PEDICELLINA AMERICANA 35 



rods very probably corresponding to the apices of the V. These 

 new structures divide, preserve their rod-Hke form in the ana- 

 phases (my evidence is complete for the sperm only, on this 

 point), and finally, in the telophases of both oogonia and sper- 

 matogonia give rise to the reduced number of bivalents. What 

 can be the significance of this sudden change in the character 

 of the chromosomes of the last division ? In the light of the evi- 

 dence here adduced, and of that from the other sources already 

 referred to, it is almost impossible to avoid the conclusion that 

 the change is a direct preparation for the process of reduction. 

 Were the V's present in this last division also, twenty-two 

 V-shaped chromosomes would be found at the poles in the telo- 

 phase, instead of the eleven actually found. Under such con- 

 ditions, the appearance of the V's in the post-synaptic phases 

 of the early oocytes and spermatocytes, even if reduced in num- 

 ber, would throw no light on the character of the reduction 

 process. For, in such a case, the reduction might have been 

 attained in any one of a number of ways ; for example, by the 

 superposition and fusion of the V's two by two, as some authors 

 actually conjecture to be the case. But in Pcdicelliua, the true 

 conditions are distinctly otherwise. The V's are not present in 

 the latter half of the last division process and appear as such 

 only later in the telophases as new formations and clearly re- 

 duced in number. 



Without undertaking a review of the already enormous liter- 

 ature on this period in the development of the germ cells, it is 

 here necessary only to point out that the ** synapsis-stage " has 

 been observed in nearly every form in which the early proc- 

 esses of development have been studied. Thus, in such diverse 

 types of spermatogenesis and oogenesis as have been described 

 for the Amphibia^ on the one hand, and the Copepods on the 

 other, the close aggregation at one pole of the nucleus regularly 

 occurs. First designated by the appropriate term "Synapsis " by 

 Moore, '95, this author laid the basis for all later progress in this 

 subject by associating this stage with the process of reduction. 

 To Montgomery, however, we owe the interpretation of synap- 

 sis in the form here adopted. Commenting on the occasional 



