284 J. \v. sroTT mac fie. 



of the basal diameter to tlie lenijth of the siphon i> in the ca.^e of C. irtvidios'ns 1 : 8, 

 in C.fatigans 1 : 4, and in Slegoni//i(ifasciata 1 : "J ; for vSen has expressed the opinion 

 that " Stegoimjia can withstand want of oxygen better than Cvlex,^' an observation 

 which he says " appears to have a natural connection with the respective lengths of 

 their respiratory siphons." 



The Age of the Larvae. The age of the larvae is also an important factor, young 

 larvae being able to survive longer than fully developed ones. For example, in an 

 experiment carried out in a fluid rich in organic matter five fully grown and live very 

 young larvae of S.fasciata were used. The fully grown laivae were found dead after 

 3, 4, 5, 5, and (> hours respectively, the average being i'C^ hours, and the young 

 larvae after G, G, 7, 7, and 7 hours, the average being (r() hours. 



It was necessary therefore to use larvae of known ages in subsequent experiments ; 

 that is. larvae in a known phase of their development, for of course a knowledge of the 

 number of days that have elapsed since the larvae emerged from the eggs is no sure 

 guide as to the stage of development to which they have advanced. As it is not 

 always easy to determine rapidly larvae in the second and third phases it was decided 

 to use only larvae in the first and last phases, that is, larvae that had not yet cast 

 their first pelts, and larvae that had already cast their third pelts. 



Individual variations are of course very considerable, but by employing only larvae 

 at known stages of development it was found that much more constant and uniform 

 results were recorded. 



The Temperaliire. Many of the experiments on the survival of submerged larvae 

 lasted a long time, so that although started early in the morning they were not finished 

 by nightfall. In some of them it seemed as if the larvae still alive at sunset survived 

 an abnormally long time, and it was suspected that this might be due to the coolness 

 of the evening and the night. Some experiments were therefore carried out to deter- 

 mine what effect different temperatures might have. All the experiments were con- 

 ducted by the standard method in distilled water which had been kept at the desired 

 temperature for some hours before the larvae were introduced into it. The larvae 

 employed were those of S. fasciaki' in the fourth phase. 



In order to test the action of a lowering of the temperature, five larvae were placed 

 in an incubator at 22° C and five in one at 37*^ C, one evening at 6 p.m. Next morning 

 at 8 a.m., that is after 14 hours, all the larvae exposed to tlu> higher temperature 

 wxre dead, but all those at 22° C. were alive. The experiment was therefore continued. 

 At 6 p.m., all the larvae in the cold incubator were still alive, but at 8 a.m., on the 

 third morning, that is after 38 hours, three were found dead. Of the two remaining 

 larvae one was found dead after 39 hours, and the other after 40 hours. 



As all the larvae in the incubator at 37° C. had been found dead at the first examina- 

 tion in this experiment, five more larvae were exposed to this temperature during the 

 day-time when they could be watched carefully. Two were foiuid dead after 4 hours, 

 and the remaining three after o hours. 



The usual temperatui'e of the laboratorv at Accra where these experiments were 

 carried out was about 30° C. during the day-time. At the natural laboratory tem- 

 peratures S. fasciata larvae in their fourth phase lived about 12 hours in distilled 

 water under the standard conditions, but there was of couise a consideral)le lowering 



