30 ART. 11. — A. IZUKA : OBSERVATIONS ON THE 



and further, that in some years it occurs in two or three, instead of 

 four periods and rarely only once during the months mentioned. 



In order to make observations for myself, I spent many 

 evenings on the river, and it was not long before I became con- 

 vinced of the general accuracy of the fishermen's predictions as 

 to when the swarming was to be expected. On October Stli 1896, 

 I had the satisfaction of observing the swarming for the fust 

 time. It was the first swarming of that year. On that day the 

 worms that swarmed out were not numerous, but on the following- 

 day the swarm proved to be one of the largest I have ever known. 



As an illustration I may describe my experience on that 

 particular day (Oct. 9th, 1896). In good time to see the begin- 

 ning of the swarming, I was on the river in a boat manned by 

 two fishermen and provided with a lantern, nets of various kinds 

 and such other utensils as might be required for observing, 

 capturing and preserving the worms. About half past six in the 

 evening, the place was reached which was considered likely to be 

 favorable for the accomplishment of my purpose. The flood was 

 to occur at 6.54 p.m. While we were waiting for the beginning 

 of the swarming, a crowd of other boats, each provided with a 

 light to attract the worms, assembled near mine. They had 

 eome to catch the " Bachi," which, as I have said, are much 

 used as bait in fishing and therefore are a marketable com- 

 modity. About 7 o'clock or a little later, the first swimming 

 worms were observed. It was only individuals of small size, 

 measuring o0-40 mm. in length, that were seen in the first part 

 of the swarming ; about 15 minutes later, larger ones began to 

 join the swarm, in which a few individuals were noticed still 

 trailing the shrivelled posterior portion of the body, such as are 

 shown in fig. 5, PI. I. (" Ya-bachi "). About half an hour 



