EXPERIMENTS IN THE KEEPING OF PLANKTON ANIMALS. 569 



of infection by Bacteria of a different nature, such, for example, as the 

 two forms referred to (pp. 565, 5C6), must of course be considered. But 

 such forms as these were rarely observed, nor did the very irregular rate 

 of mortality among individuals in an experiment, or in different experi- 

 ments, suggest that bacterial action was primarily accountable for the 

 death of the specimens. 



Food-growth again, was in many cases poor and uncertain in experi- 

 ments in Position A, but as the food-supply was with few exceptions 

 renewed at frequent intervals, it seems unlikely that the animals were 

 nmch affected by any such deficiency. It was often observed, too, that 

 when food-growth was vigorous the animals died off independently of 

 this, or even earlier than in experiments in which the growth was poor 

 or stationary. 



A noticeable feature occurs in connection Avitli the special experi- 

 ment in Position A (p. 562), in which, apaii: from temperature, apparently 

 ideal conditions were provided, and a healthy growing food culture was 

 •carried in the changing water-supply through the vessel in which the 

 animals were contained. This vessel was a 1 -S-litre flask, and it is remark- 

 able that the average life of the 5 Calanus it contained was very nearly the 

 same, 21 days, as that for the 19 specimens of the 5 experiments with this 

 species in 1-litre flasks in the same room, 22 days. In the 5 experi- 

 ments with 34 specimens in finger-bowls, in which the volume of water 

 was no more than 300-350 c.c, the low average of 12 days is presum- 

 ably attributable to the changes communicated by the air temperature 

 being more rapid than in the case of the larger vessels. If an instan- 

 taneous change of temperature, probably of 4° or 5° (p. 568), can so 

 affect nauplii as to stun them for 5 or 10 minutes, or if one, apparently 

 of salinity only, can be fatal, as in the case of the 2 Calanus referred to 

 ^(p. 562), it is reasonable to suppose that the daily fluctuations of air 

 temperature to which uncovered vessels are exposed are liable to produce 

 conditions that must sooner or later prove fatal to animals so sensitive 

 to such changes. 



The removal of the experiments to Position C was especially prompted, 

 as it was observed, by the survival of a Calanus for 51 days in a 2-litre 

 jar submerged in one of the tanks. This specimen was then found to have 

 the hairs of the antennae and caudal rami much encrusted with fixed 

 Diatoms and other accretion, and probably succumbed owing to these 

 •causes about a week later. This jar contained a sample of water only, 

 with the plankton naturally present in it, and therefore very few indi- 

 viduals at the outset, nor was any addition made to the food. Yet on 



XEW .SEIIIES. — ViJU X. .NO. 4. MAY, 1915. 2 O 



