236 B. H. GRAVE. 



days or two weeks. This more active spawning period includes 

 two or three days preceding full moon and continues until after the 

 third quarter, or with slight abatement until two or three days 

 after new moon. In 1921 the diminution of egg-laying came 

 promptly at new moon. The periods of excessive spawning ac- 

 tivity are followed by periods of low activity during which spawn- 

 ing is practically suppressed. These periods of low spawning 

 activity extend roughly from new moon until the approach of full 

 moon. 



More specifically stated, there were ten-day periods in each of 

 the months of June, July, and August, during which spawning 

 activity was remarkable for the abundance of eggs extruded, and 

 there were other periods between each of these high points when 

 spawning activity was diminished. 



The low spawning periods were somewhat more distinct in 1921 

 than in 1920. In both years approximately sixty per cent, of the 

 eggs obtained were spawned between full moon and the third quar- 

 ter, and perhaps seventy-five per cent, during the periods beginning 

 three days before full moon and extending to the next new moon. 



The data obtained during the three years agree except that the 

 spawning season of the last two years extended over longer periods, 

 and that the lunar periodicity was less marked than in 1919, when 

 spawning was confined to a few days in each month. 



A further comparison of the results obtained during the three 

 years is interesting in that the total number of eggs secured during 

 1920 was almost double that of 1921 and very far in excess of 

 that of 1919. There is no apparent reason for this variation in 

 egg production from year to year, but similar observations on ether 

 species of animals are common. 



It may not be amiss in this connection to point out that Lillie 

 and Just ('13) have found spawning activities in Nereis limb at a 

 which are quite similar to those of Chcetopleura, but the periodicity 

 of the two species may or may not be identical. 



In what is to follow reasons are given for believing that the 

 spawning periods of Chcctoplcura, under natural conditions, are 

 more restricted than the results of the experiments carried out 

 under laboratory conditions indicate. 



