18 
larvEe was g-ot about th-e midflle of September. Some 350 of 
these were placed in the tanks of the private aquarium. Of 
these 120 died on October 14th, and tlie remainder with the 
exception of '4 were lost in the third stage. The remain- 
ing 3 passed througdi the fourth stage and died on 
October 31st. 
IVe are thus led to conclude from our experiences of 1010 
that it is essential to retain the berried lobsters in a large 
pond during the period before and during hatching, and that 
the greatest cleanliness must be observed in dealing with the 
larvae after hatching. Great difficulty was experienced in 
keeping the small tanks in Avhich the larvae Avere kept clean, 
for the Avater pumped from the bay was found, especially 
after a slight sea, to contain a great deal of matter in suspen- 
sion. This settled to the bottom of the tanks, and, in spite of 
the tanks being frequently cleaned, doubtless caused the 
death of many of the larvae Avhen these settled on the bottom, 
esi)ecially when they AA’ere ca.sting. The larvie left in the 
centre pool Avere healthy and moA’ed about Avith great Augour, 
many reaching the third stage, but the number Avas obseiwed 
to diminish rapidly. This for a time could not be accounted 
for, as no dead were found at the bottom of the pool. The 
majority of them disappeared before the Laboratorj^ 
Attendant found late one evening that some shore fish had 
been placed amongst the j’oung lobsters by a visitor. The 
fish were four blennies and a three-bearded rockling, and 
the}' kept in the shelter of the rockAvork during the day. 
It is eAudent, therefore, that the small tanks used for 
rearing the laiwce must have a supply of Avater free from 
sediment and eA'en certain Protozoa. So that filtering 
appears to be necessary. And careful feeding Avill do much 
to keep doAvn the death rate from the cannibalistic propen- 
sities of the larvae. 
