REPRODUCTION 



123 



occur, the type difference and certain other characters, according to 

 Sonneborn and Kimball, appear to be under control of genie consti- 

 tuents of the macronucleus. Since the number of chromatin granules 

 appear approximately the same in the macronuclei of different gen- 

 erations of a given species, the reduced number of chromatin gran- 



FiG. 50. Nuclear and cytosomic division of Paramecium caudatum as 

 seen in stained smears, X260 (Kudo). 



ules must be restored sometime before the next division takes place. 

 Calkins (1926) is of the opinion that "each granule elongates and 

 divides into two parts, thus doubling the number of chromomeres." 

 Reichenow (1928) found that in Chilodonella cucullulus the lightly 

 Feulgen positive endosome appeared to form chromatin granules 

 and Kudo (1936) maintained that the large chromatin spherules of 



