27 



on was concerned, very few being shaken off. In the 

 course of a week after arrival, hatching of the larvae com- 

 menced from the more advanced eggs, and continued 

 during the next few weeks, but only in small numbers at 

 a time. With one or two exceptions no larvoe hatched 

 out in the day-time, practically all emerged at night, 

 between 10 p.m. and 2 a.m. 



Each morning the larvae were removed from the collect- 

 ing boxes. At first they were kept in glass aquaria, where 

 a constant circulation of water was maintained. After- 

 wards we tried keeping them in the dark, and finally in 

 the floating boxes of the Dannevig apparatus, all light 

 being carefully excluded. 



During the first few days the larvae fed vigorously, and 

 swam about actively. Towards the end of the first week 

 of their life they ceased feeding, and kept more to the 

 bottom of the apparatus. From seven to ten days after 

 hatching the larvae commenced to moult for the first time. 

 Many died at this stage, some with the skin partly shed. 

 Many of the survivors failed to recover from the strain of 

 moulting a second time, and before the period of the third 

 moult had arrived, all had gradually died off. The longest 

 time that any lived was just over three weeks. 



Various methods were tried to keep the larvae alive. 

 The food, water supply, and light were varied from time 

 to time, but with no effect. 



The food found most suitable was nnnute fragments of 

 the liver of freshly-killed shore crabs. Some larvae took it 

 readily, clinging tenaciously to the pieces, which could be 

 seen gradually passing into the stomach ; while others 

 refused it altogether, although it was held in front of them 

 at the end of a thin piece of wood. The larvae made no 

 attempt to follow this food if it fell to the bottom, but 

 sometimes, when walking over the fine gravel, they would 



