The Annual Breeding-Swarm of the Atlantic Palolo. III 
overcast or cloudy weather, so that even diffuse moonlight appears to be 
capable of calling forth the breeding-swarm. 
In the Atlantic palolo the annual breeding-season is only of 1 to 6 days’ 
duration, and the males outnumber the females in the ratio of about 3 to 2, 
whereas in Nereis, where the breeding-season is fully 100 days long, the 
males greatly outnumber the females. It is evident that a shortening of 
the breeding-season would cause a greater concentration of breeding indi- 
viduals, and would therefore permit of a relative decrease in the number 
of males and a corresponding increase in the number of females; for when- 
ever a female swarms it is important for the preservation of the species that 
there should be a male near her to fertilize her eggs. If the breeding-season 
be of long duration, the males must greatly outnumber the females to secure 
this fortuitous proximity, but if all of the females swarm within a few 
days very much fewer males will suffice to accomplish this purpose. 
We have advanced beyond the period in the history of biology when one 
had but to discover an advantage to determine a cause ; but that some such 
cause may have contributed to shorten the breeding-season in such animals 
as the Atlantic, Pacific, and Japanese palolo worms is shown by the fact 
that more eggs are fertilized when males are near the female than when 
they are far away. For example, I took a female Atlantic palolo from the 
midst of the swarm and placed her in sea-water 200 meters away from the 
nearest swarming males. Of the eggs which were laid by this female in 
the water removed from the place of the swarm only an occasional one 
developed, whereas practically every egg developed in the sea-water where 
males were near. 
The polar bodies are given forth as soon as the eggs are cast out from 
the female, and fertilization occurs in the water; but the egg does not 
mature if it be cast out at any time other than that of the normal breeding- 
swarm. When about 10 to 15 hours old the larve are nearly all negatively 
phototactic either in diffuse light or in sunlight. When about 28 hours old, 
however, they mainly become positively phototactic and remain thus, even 
after the eighth day, when they will have ceased to swim through the water 
and have sunken to the bottom. Within 24 hours after sinking to the 
bottom, however, they become indifferent to light in so far as their move- 
ment is concerned. 
The segmentation closely resembles that of Nereis and the larva is 
telotrochal. 
Further accounts of the Atlantic palolo will be found in Bulletin of 
the Museum of Comparative Zoology at Harvard College, 1900, vol. 36, 
Pp. I-14, plates 1-3, and in the Science Bulletin of the Museum of the 
Brooklyn Institute of Arts and Sciences, 1902, vol. 1, No. 3, pp. 93-103, 
I plate. 
The Japanese palolo is treated of in detail by A. Izuka, 1903, Journal 
