BV K. J. TILLYAUD. 671 



all the Agriiniid larvfe, and the two Hemicordidia tail, were dead 

 — stretched, flat and dried up, on the sand*. In the "haven of 

 refuge " three larvae of S. eustalacta had settled down into tlie 

 (lamp sand. The other five were wandering round the tank 

 searching for damp corners, under tlie rocks. I noticed that they 

 appeared to be considerably annoyed by a tinj^ fly which settled 

 often on their backs. It will be interesting to find out whether 

 the laiva of this fly is a parasite on the living dragonfly larva. 



On February 16th, there was only one S. eiistalacta larva in the 

 "haven of refuge." This one was caked all over with dry sand, 

 and I picked him up and examined him. He was quite vigorous, 

 and ran acrcss my hand. I put him back again, and noticed that 

 the other two had crawled to the edge of the t^nk and had got 

 partly under the shelter of a large flat rock near the edge. The 

 sand was now absolutely dry, and caked hard. The tank was left 

 without observation from^ this date until Good Friday (March 

 25th). On that day I turned up the slanting rock in the centre, 

 and found the same larva of S. eustalacta sitting under it in 

 e.\;actly the same position as when I left it. It was quite lively, 

 and ran vigorously across my hand. It appeared very hard and 

 dry, caked with sand, but not at all shrivelled or reduced in size. 

 I then turned up the other rocks, and discovered three more 

 larvae of the same species, firmly set into the dry sand. These 

 came out like drj' hard lumps in a cake of sand. They were 

 absolutely inert, and to all appearances dead, but not .shrivelled, 

 and only a little reduced in size. I dropped these three into a 

 jar of water. Two of them revived immediately, and swam to the 

 bottom of the jar, where they buried themselves at once, with 

 just their heads and anal ends appearing above the sand. The 

 third larva floated on the water for haif-an-hour, when he sud- 

 denly revived and descended to the bottom of the jar, where he 

 sat on the sand without attempting to bury himself. 



* It is interesting to note that before the very dry winter of 1907, huge 

 swarms of Hemic ordulia tau appeared during March and April all over New 

 .South Wales. Many followed the watering-carts in Sydney for weeks, and 

 the occurrence was noted in the daily press. 



