544 CELL DIVISION IN EGGS OF CREPIDULA. 



ters, and finally polyasters in cases where the division of the cell body is inhibited, 

 while that of the centrosome proceeds. / have been unable to find any evidence 

 whatever that centrosomes arise de novo in the eggs of Crepidula, or that cytasters 

 become the centers of mitotic nuclear division. 



It is a striking fact, and one not easily harmonized with the hypothesis 

 that centrosomes may arise de novo, or from the fragmentation of a single centro- 

 some into many, that the number of nuclear centrosomes in eggs in which cleavage 

 was suppressed by hypertonic sea water is approximately proportional to the 

 length of time during which the eggs have remained in normal sea water following 

 exposure to the hypertonic solutions. The maximum number of cytasters is 

 found while eggs are still in the hypertonic solution (figs. 183, 189, 190) whereas 

 in such eggs the nuclear asters are present in normal numbers. Only after such 

 eggs have been returned to normal sea water for several hours during which 

 time centrosomal division has been progressing while cell division remains sup- 

 pressed do nuclear asters become numerous (figs. 203, 207, 219, et seq.). The 

 only apparent exceptions to this rule are found in those cases in which the 

 hypertonic solution (or other injurious condition) was not strong enough to stop 

 centrosomal division, although strong enough to suppress cell division, and these 

 exceptions are only apparent and really support the proposition that nuclear 

 asters and centrosomes arise by the division of preexisting asters and centrosomes. 

 Further evidence that the polyasters which occur in the eggs of Crepidula are 

 not derived from cytasters may be found in the fact that polyasters may be found 

 at any stage of the cleavage, whereas according to Konopacki and myself, 

 cytasters are rarely found later than the 2-cell stage. 



6. Origin of the Cleavage Centrosomes. 

 (Plate LV.) 



All of the eggs represented in plate LV were placed in 1 per cent. NaCl 

 in sea water for 4 hrs. and were then fixed at once. In all of these figures both 

 polar bodies had been formed and the sperm had entered the egg before the 

 experiment began. The effect of the hypertonic sea water has been to delay 

 the movement of the sperm nucleus toward the egg nucleus and to stimulate the 

 formation of an amphiaster in connection with the egg nucleus. The view was 

 first expressed by Boveri (1892), and has been widely accepted by other investi- 

 gators, that the egg centrosome is a decadent structure, which ultimately under- 

 goes degeneration, while the two cleavage centrosomes arise in connection with 

 the spermatozoon. In cases of normal or artificial parthenogenesis it is supposed 

 that the cleavage centrosomes are derived from the egg centrosome, which in 

 such cases becomes active, or they are new formations. My work on the normal 

 processes of maturation and fertilization in Crepidula (Conklin, 1902) led me 

 to the conclusion that one of the cleavage centrosomes is formed in connection with 

 each of the germ nuclei, though it was impossible to say that they came from the 



