SPAWNING HABITS OF CUMINGIA TKI.U NOIDES. 427 



no lunar periodicity in spawning is likely to be observed unless 

 attention is directed to it. Cumingia has been used as a source 

 of embryological material for class work for many years and only 

 rarely has difficulty been experienced in obtaining eggs in 

 abundance at any time that they have been needed. It appeared 

 that eggs could be obtained at any time during the summer. 

 It was only after receiving a suggestion from Heilbrunn that any 

 convincing evidence of periodic spawning was obtained. He said 

 that Cumingia would spawn a second time after its season had 

 apparently ended in August. Morgan gives the spawning season 

 of Cumingia as June, July, and August, so that he evidently 

 overlooked the September spawning after the break in August. 



It was at Heilbrunn's suggestion that I undertook experiments 

 late in the summer of 1922, when, as he said, a break could be 

 expected. 



During the week of August 23 to September I, 1922, no eggs 

 could be obtained from Cumingia, although the characteristic 

 spawning reactions were carried out as usual. The siphons were 

 extended to great length and the whip-like lashing of these 

 organs was carried out. All that was expelled, however, was a 

 considerable amount of mucus containing at best a few immature 

 or defective eggs. There were no mature eggs in the ovaries. 



Bean, working in the same laboratory, was likewise unable to 

 obtain eggs during this period for experimental purposes. Bean 

 worked constantly with Cumingia during the summer of 1922 

 and he thought that he detected a periodicity during the height 

 of the breeding season. The maximum spawning periods, ac- 

 cording to his statement, occur at new and full moon, being 

 therefore bimonthly. This is interesting, if correct, as indicating 

 that the spawning stimulus might be associated with the tides 

 rather than with moonlight. Further study during the summers 

 of 1923, 1924, 1925 and 1926 shows conclusively that the 

 maximum spawning comes at full moon and is not bimonthly. 



The table for 1922 gives the exact data as obtained by the 

 writer. It will be seen that spawning activity is revived some- 

 what before full moon after a period of complete cessation. 

 Furthermore a careful study of the tables for the years 1922, 

 1924 and 1926 shows that the heaviest spawning occurs at the 



28 



