91 DE. G. H. FOWLEE— BISCATAN PLANKTON : 



it consisted almost entirely of Ijlastozooids. Eor example, haul 21 h at 100 fathoms 

 yielded the folloTring determinable and measurable specimens * in half an hour : — 

 Blastozooids : 2 mm. and less in length 369 



3 „ „ 43 



4 „ „ IG 



10 



G „ „ 10 



7" „ „ 4. 



Gonozooids : 2 ,, „ 2 



3 ,1 11 1 



This puzzling appearance of numerous specimens of the same generation termed a 

 " swarm " has been discussed at some length by Dr. Borgert. He refers the swarming to 

 the more or less seasonal occurrence of optimal conditions, such as food and temperature 

 {o2). cit. pp. 61-03) ; but he rejects (o/j. cit. pp. 49, 50) the idea that species of Doliolwu 

 have regularly recurrent periods of reproduction, on the following grounds : — (1) that all 

 the eggs of the gonozooid do not ripen simultaneously, and. therefore the resulting blasto- 

 zooids will be of different ages and sizes; (2) that the blastozooids and phorozooids are 

 productive for a considerable time, and. their successive " progeny " therefore will 

 be even more markedly of different sizes and ages : thus the effect of a swarm produced 

 under optimal conditions will be dissipated and uatraceable in the covirse of a complete 

 cycle of reproduction. 



On the other hand, the presence of such large numbers of minute blastozooids as were 

 taken on July 8 certainly seems to me to demand the pre-existence of a "swarm" of 

 gonozooids, since the ovary of the specimens examined contained generally only one 

 large ovum, although plenty of smaller ova. 



The swarm was apparently confined to a small space; on the next day, July 9, at 

 a distance varying from 8 to 20 miles away in a N.W. direction, it was not certainly 

 recognizable. On July 11 we were again 8 miles from the original position, t])is time 

 a little N. of E. ; the average catch per haul that day was 2-6 instead of 314. On 

 July 12, 9 miles S.E. of the original position, the average catch f per haul was 4'5. 

 Yet the whole area covered by our stations was so small that the conditions of food 

 and temperature were presumably uniform throughout, and, if optimal at one point, 

 ought, on Dr. Borgert's theory, to have affected the whole area, and to have produced a 

 imiform swarming : this was obviously not the case. 



But though the " swarm " was apparently local, yet sporadic sexual reproduction 

 appeared to have taken place simultaneously over, at any rate, a large part of the area 

 under observation. This would appear from the following table % : — 



* Actual numbers, not multiplied by a time-factor. 



t The daily average catches per haul have been plotted on the chart of positions, but yielded no obvious result. 



i Days of 24 hours, midnight to midnight. The specimens were roughly measured as 2 mm. or less ; 3 mm. or 

 less, but more than 2 ; 4 mm. or less, but more than 3, &c. ; and the average length struck on these measurements. 

 They seem to show the comparative daily lengths, but would not be accurate enough to determine (say) rate 

 of growth. 



