ORIGIN OF MYCETOCYTES IN PSEUDOCOCCUS. 285 



Numerically, the same number could be attained by the combina- 

 tion of 3 cleavage cells. The latter possibility is more than doubt- 

 ful if it is considered that giant cells are formed only in a certain 

 and limited area at the periphery, whereas if a fusion of cleavage 

 cells alone is a possibility such cells with 30 chromosomes should 

 then be formed at any part of the periphery where cleavage cells 

 are forming the blastoderm. Weighty although negative evi- 

 dence against this hypothesis is to be adduced from the seeming 

 nonexistence of giant cells with 20 chromosomes. The failure to 

 find such cells is all the more significant if it is considered that the 

 chances of two cleavage cells coming together for fusion are 

 greater than the chances of three combining in that way. 



The 35 chromosome type can originate only* from a combina- 

 tion of a polar nucleus derivative with two cleavage cells. Other 

 chances of combination to bring about this type have been ruled 

 out by the preceding considerations. 



Why a 40 chromosome cell should not be found at this stage it 

 is not possible to say. Possibly there are such cells but their 

 less frequent occurrence has prevented their discovery. More 

 probably the three types already mentioned represent all the 

 combinations possible. 



THE GIANT CELL CHROMOSOMES AT LATER STAGES. 



As already mentioned, the giant cells originate always at the 

 time that the cleavage cells are migrating to the periphery of 

 the egg to establish the blastoderm. With the completion of the 

 latter and the first stages of germ band formation, another period 

 in the history of the giant cells is initiated. In the course of this, 

 giant cells with still greater numbers of chromosomes than those 

 already described, are encountered. Beside them, the three older 

 numerical types may continue to exist. The increase in the 

 number of chromosomes, together with the fact that divisions in 

 the giant cells tend to decrease their size, make certain counts in 

 these later cells a much more difficult matter. Apparently most 

 of these greater chromosome numbers hover in the neighborhood 

 of 60 or 70. Only one certain count could be made, that being of a 

 plate containing 60 chromosomes (Fig. 18). 



Cytological evidence to be considered later, makes it probable 



