POLVVITELLINY IN POND SNAILS. 323 



..." and attributes this to the fact that cleavage is determinate 

 in animals of this phylum. From his explanation and his review 

 of the works of writers who claim that twins, at least in the forms 

 they studied, are produced by a process of budding (Patterson), 

 fission (Stockard) or fusion (Gemmill) it is evident that the 

 process of twinning begins later than the first cleavage of the egg. 

 However, it might be possible for two ova to fuse before either 

 has undergone cleavage ; whether this would produce twins or 

 not is another question. 



We have been unable to find anything to indicate that the 

 supernumary individuals developed from pond snails eggs arose 

 from a single vitellus. Although some four dozen ovotestes of 

 L. s. appressa were sectioned (Crabb, '2/0) not more than ten and 

 seldom as many as four vitelli were found in the ovotestis and 

 hermaphrodite duct in any one snail and in no instance were ova, 

 and rarely sperms, found in the filliform part of the hermaphrodite 

 duct (i.e., craniad to the region " O," Crabb, '27/7, Fig. i) which 

 in life bends around the gizzard. Thus it appears probable that 

 while actively contracting the gizzard inhibits the passage of 

 ova to such an extent that several ova accumulate in the enlarged 

 part of the hermaphrodite duct and during a period of reduced 

 activity of the gizzard all pass into the convoluted uterus at one 

 time. In this, or some other way, a number of vitelli become en- 

 veloped by the albumen and the egg membrane which would 

 normally cover a single vitellus. Often a mass of sperms is en- 

 closed with the vitelli (Figs, i, 5). Thus, probably by sheer 

 accident, a polyvitelline egg is produced, and for this reason the 

 young hatched from such an egg should not be considered true 

 twins any more than should all the other young hatched from that 

 mass of eggs. 



SUMMARY. 



That polyvitelline eggs do not contain true twins is shown by 

 the following facts : 



1. The vitelli are normally enveloped by the albumen and egg 

 membrane before any of them have undergone the first maturation 

 division (Figs. 2, 6). 



2. The vitelli often occur in uneven numbers (Plate I, Tables 

 L, II.). 



