NEREIS JAPONICA. 



2 97 



Prof. Ijima), &c 



All the material taken from the above mentioned localities, except 

 those from Kojima Gulf, were sexually immature. It is a remarkable fact 

 that in the Sumida River I have not succeeded as yet to obtain a single 

 adult worm of the species, although I had often searched for them. In 

 that river the worm seems to swarm generally once a year in an immature 

 state of development. But in Kojima Gulf it is very easy to get the 

 mature worm in great quantity during the month of December, and also 

 to observe their breeding swarms, and consequently to follow their early 

 development from the egg by artificial fertilization. 



With regard to N. divcrsicolor, a species which occurs commonly in 

 European waters M'Intosh* says that the breeding season extends from 

 November till May and that he observed no indication of swarming 

 during those months. The same author further states that he obtained 

 numerous postlarval stages of the species in or near the mouths of the 

 burrows in which the annelid lives. 



With these remarks I proceed to record my observations in the 

 following order : — 1) on the immature and mature worms, 2) on the fer- 

 tilization of the eggs and their development, and 3) on the breeding 

 swarms. 



1. Observations on the immature and 

 the mature worms. 



Immature worms. — The immature worms occur in great abundance 

 between the tide marks along the head of Kojima Gulf and also in the 

 mouths of rivers emptying into it. 



The greater part of the gulf is exposed at spring tides, the water then 

 becoming confined to narrow channels leading from the rivers, as shown 

 in the accompanying map of the locality (text-figure 4). 



* M'Intosh :— On the Reproduction of Nereis dherùcolor. (Ann. and Mag. Nat. Hist, 9 

 series, Vol. 20. No. 117. p. 176, Sept. 1907). 



