STUDIES ON SEX IN CREPIDULA 13 



3, 4, 14, etc.) have a reticular nucleus, the chromatin being dis- 

 posed upon an indistinct network of linin whose meshes are 

 dense enough to give the whole nucleus an opaque appearance. 

 Measurements of type A and type B cells have been made at 

 different stages of the sexual cycle, and the results have been 

 brought together in table 1. The average diameter of the A 

 cells in any specimen is always greater than that of the B cells. 

 The two series are grouped about different modes. 



2. The type A cells, though they may become ever so numer- 

 ous, do not ever come to lie in the lumen of the gonad. Their 

 divisions always leave the daughter cells attached to the periph- 

 ery. The type B elements, on the other hand, when undergoing 

 a period of mitotic division, proliferate cells like themselves 

 into the lumen of the gonad where they lie free, which continue 

 to divide. 



3. The significance of the dimorphic germ cells becomes evi- 

 dent when one observes their times of activity. As male de- 

 velopment takes place great numbers of type B cells are pro- 

 duced and fill up and enlarge .the lumen of the gonad. After 

 their period of multiplication they become transformed into the 

 spermatocytes and the cells which are destined to give rise to 

 the atypical sperm. 



The type A cells undergo their period of greatest activity 

 after the male phase is over (if that occurs at all), and when they 

 have become very numerous, are gradually transformed into the 

 oocytes. Knowing the fate of the two series of germinal ele- 

 ments, we may speak of the type A cells as the primordial egg 

 cells and the type B's as the primordial sperm cells. Crepidula 

 plana produces its sperm and eggs at widely separated times, and 

 correspondingly its primordial male and female sex cells undergo 

 their periods of activity at different times. The presence of 

 large numbers of either type of cell is accompanied by a paucity 

 of cells of the other type. 



4. There is no possibility that we are dealing with different 

 phases in the activity of a single series, for two different series 

 of division figures have been found, the behavior of each of which 

 identifies it with one of the types of cells above described. Just 



