DESCRIPTIONS OF AUSTRALIAN ^SCHNIN^. 65 



is eaten ; Agrionid larvae are greedily consumed. When all else fails, 

 cannibalism becomes rampant, and the final result attained is that only two 

 or three nymphs survive, becoming rapidly full-fed at the expense of all the 

 rest. Each survivor jealously guards its own corner of the aquarium, and 

 woe to the one that is taken unawares. 



I have not been able to determine the total number of ecdyses in this 

 or any other dragonfly larva. Larvse captured at about 10-12 mm. length 

 undergo either four or five ecdyses before becoming full-fed. The total 

 number would appear therefore to be ten or more, for at 12 mm. the larva 

 has reached a stage in which the wing-cases can be seen as small rudiments. 

 The full-fed nymph becomes more or less dormant for a period varying from 

 two days to a week. The colder the weather the longer does this " dormant " 

 period last. The nymph of ^. hrevistyla nearly always makes " trial " 

 excursions out of the water for a few days : at first resting for an hour or 

 two in the morning with only its head projecting ; then, with half its body, 

 or more, out of the water. Finally, it climbs up a stick or reed-stem and 

 emerges in the usual fairly rapid manner, the abdomen attaining a fair length 

 before the wings expand, but afterwards reaching its full length and slender 

 shape rather slowly. The emergence nearly always takes place in the early 

 morning ; but, during thundery weather, large numbers may be seen to 

 emerge together, even late in the afternoon, before or e^-en during the 

 storm. 



Habitat. This species is very common throughout the temperate portion 

 of Australia, ranging from Western Australia across to South Queensland, 

 and being also quite common in Tasmania. In Northern Queensland it 

 is exceedingly rare, and I doubt whether it extends to Cape York. It is 

 common in New Zealand and the Kermadecs. This distribution is of con- 

 siderable interest, showing as it does a distinct southern origin for the species. 

 As the New Zealand form, on the average, appears to be somewhat more 

 reduced than the Australian, I am inclined to regard this species as a single 

 remnant in Australia of a once more plentiful Antarctic u^lEseJ ma -group of 

 species, of which ^. dijinis, Rambur, may possibly be another representative 

 in Patagonia, Chili, and Argentina. As nearly half the known species of 

 the genus yEschna occur in the Neotropical Region, it seems possible that our 

 one Australian species is truly an offshoot from that great group, having 

 travelled, with others since extinct, southwards to a temperate Antarctica, 

 from which it was at last driven out to take refuge in Tasmania and New 

 Zealand. Also it must have spread from Tasmania to the mainland before 

 the separation of the two ; for, in spite of its size and powers of flight, it has 

 no migratory instinct, and rarely travels more than a few hundred yards from 

 its breeding-places. We have also the remarkable fact to consider that 

 Anax papuensis, a larger and stronger flier than JE. hrevistyla, and belonging 

 to a group with more migratory tendencies, which has worked into Australia 



LINN. JOURN. — ZOOLOGY, VOL. XXXIII. 5 



