COCOONS OF ALLOLOBOPHORA FCETIDA. 213 



Spermatophores, however, are certainly not received into the 

 spermatheca; in the form of spermatophores, for we have vivisected 

 many copulating worms having spermatophores at each end of the 

 slime-tube (/. c., over the spermathecal openings of both worms) 

 and the contents of the spermathecae, in every case, was quite 

 unlike the spermatophore substance ; the former mixing freely 

 with water, while the spermatophores do not disintegrate even 

 after twenty-four hours in water. 



The granular substance secreted by the spermathecal glands l 

 which forms the center of the spermatophores also fails to disinte- 

 grate in water, and we have not been able to identify this sub- 

 stance in the spermathecse. 



When copulating worms are allowed to withdraw from the 

 double slime-tube, we sometimes find aggregations of spermato- 

 phores in the definite areas of the slime-tube where the two 

 cocoons are formed. The constrictions of the slime-tube that 

 close each end of the cocoons, confine the spermatophores to 

 these areas, and when filled with spermatophores, these areas 

 bear such a striking resemblance in form and color to freshly 

 deposited cocoons, that in our earlier observations we misinter- 

 preted them as partly formed, or nearly completed cocoons. The 

 true "partly formed cocoon " has a white semi-transparent cover- 

 ing, contains more or less albumen and several eggs. Prior to 

 1901 \ve had found these cocoons only in the compost not 

 around the worm. We found only a few, and in every case the 

 cocoon contained eggs, but as we did not preserve the albumen, the 

 absence of spermatozoa was not discovered and we supposed these 

 cocoons to represent a later stage of development than the aggre- 

 gations of spermatophores in the slime-tube. 



The marked difference in structure of the spermatophore sub- 

 stance and the contents of these cocoons was noted, as well as 

 the fact that in freshly deposited cocoons we found only a very 

 small number of spermatozoa, in comparison with the large quan- 

 tity found in a slime tube containing spermatophores. It was 

 these inconsistencies that prompted us to undertake the experi- 

 ments described above. 



1 Ibid. , p. 496. 



