670 ON SOME EXPERIMENTS WITH DRAGONFLY LARV.^E, 



the sand to dry up completely, so as to cake into a hard mass, 

 we cannot say from this that it could have survived a moie 

 vigorous drought. But in its natural haunts, supposing at any 

 time a long-sustained drought had occurred, there would have 

 been at times a fall of dew which would have served the same 

 purpose as the sprinkling of water which I gave once a week. 



ii. The larvaj of Si/nlhemis eustalacta Burm. 



About twenty of these were obtained from a swamp at Leura, 

 Blue Mountains, on October 1st and 2nd, 1909, together with 

 about a dozen larvaj of >S'. laacrostigma Selys, all from one-half to 

 three-quarters grown in size. As these larvje are incapable of 

 climbing oat of a deep glass aquarium, I placed them in a 

 specially designed circular tank, one yard in diameter, wath sides 

 only six inches high. From two to three inches of sand covered 

 the bottom all over, rocks were placed here and there, clumps of 

 water-weed planted, and the tank filled with water to within 

 half-an-inch of the brim. A large quantity of duck weed was 

 placed on the water, and clumps of reed and sedge placed so that 

 the larvaj could climb up and emerge. 



In the previous year, nearly all my larvye of these two species 

 died through inability to emerge from a deep aquarium when full- 

 fed. During November and December, 1909, however, all the 

 larvae of S. macrostigma emerged, and also many of the S. eustal- 

 acta larvae. There were, as a matter of fact, eight of the latter 

 larvae in the tank on December 25th, 1909, together with two 

 Hemicordulia tau and a few small Agrionid \a.wis{^Pseudagrioii 

 cyane). T determined to subject these to a rigorous drought and 

 starvation process. To do this naturally, no more food was put 

 in, and the water was allowed to evaporate of its own accord. 

 Also, in the centre of the tank, two flat rocks were arranged, so 

 that one leaned slantwise on the other, forming a small " haven 

 of refuge," which I could at any time examine, by lifting the 

 slanting rock, without disturbing the larvae. 



By February 2nd the water had all evaporated, but the sand 

 was still very damp, especially around the rocks. A week later 



