332 JOHN W. SCOTT. 



new moon and the day following, July 23 and 24 ; in 1907 best 

 results happened on July 26, two days after full moon ; and in 

 1908 at new moon, July 28. The full meaning of these facts, 

 however, is best brought out in Table II. Here I have shown 

 the day on which eggs and sperm were deposited in the labora- 

 tory in reference to spring tide. In all I have recorded 38 

 instances where Amphitrite have shed eggs or sperm in an appar- 

 ently normal manner. Twelve of these were known to be in- 

 jured, and consequently their data are unreliable. But of the 

 twenty-six uninjured worms twenty-three (88 per cent.) shed their 

 products on or within two days of spring tide ; twenty-four worms 

 (92 per cent.) shed products within three days of spring tide ; only 

 one worm deposited eggs five days, another seven days, after the 

 period named. Of the injured worms nearly sixty per cent, de- 

 posited products within two days of this period, while thirty -three 

 per cent, missed the spring tide by more than four days. This 

 suggests the possibility that, of those worms given as uninjured, 

 perhaps the last two were injured internally. Again, the table 

 shows that more worms laid eggs or sperm on the day of spring 

 tide than on any other day ; that the next largest number deposited 

 products on the day following ; that practically all sexual products 

 are deposited within three days of spring tide with a tendency to 

 follow rather than to precede this period. That the distribution 

 of these figures is not due to a like distribution of the number of 

 times worms were collected, is shown clearly by comparing them 

 with the figures in the last line of the table. 



In order to make a further test of the hypothesis that ripe 

 sexual products are deposited most abundantly at the time of 

 spring tide, on July 17, 1909, the date of the new moon, twenty 

 adult Amphitrite were collected. Of this number twenty-five 

 per cent., three females and two males, deposited ripe products, 

 thus giving excellent confirmation to the hypothesis. The ques- 

 tion arises how may we account for the periodicity in the time 

 of egg-laying. Observations that have been made upon some 

 other forms will help in our explanation. 



Mayer ('02) has described the interesting case of the breeding 

 habits of the Atlantic palolo. This worm " swarms at the sur- 

 face before sunrise within three days of the day of the last 



