MIGRATION- AXD DISPERSAL OF INSECTS ! ODON'ATA. 188 



difference between the dates May 80th-8lst, 1889, and June 2Hth, 

 1816, is also noticed. The suggestion is made that the great multi- 

 plication of the insects in 1889, was due to the excessively rainy 

 -weather from May 1st to 21st. Much of the lower ground near the 

 ri\er was inundated. Hence myriads of pup;e which would have 

 remained in deep water under ordinary circumstances, were brought 

 into shallow water, and the hot weather from ]\Iay 21st to May 29th 

 converted these shallow pools into true hot-beds. It is further 

 suggested that the numerous thunderstorms about that time at Weimar 

 greatly encouraged the rapid development of the imagines, and so the 

 insects rose almost at once from the temporary swamps, and were 

 ol)liged to migrate to satisfy their voracious appetites. Weissenborn 

 further notes that on their migrations the dragontlies followed the direc- 

 tion of the river, appearing always to fly with the current, spreading over 

 wide districts on either side of the river in order to subsist. Hope, at 

 the meeting of the iMit. Soc. of Lond., October 7th, 1889, stated that 

 he witnessed a nearly similar migration of the same insects some years 

 previously at Netley, which he attributed to the drying up of some 

 large reservoirs of water. In this instance several distinct species 

 were observed. 



Ghiliani records that, in 1867, there was a most unusual appearance 

 of swarms of A)ia.r nicditcrranrus in Italy. Previous to this the species 

 had actually been erased from the European list, although the original 

 description was made from a single individual taken in Sardinia. 



Campbell notes a migration of .KscJnia iiii.rta that took place on 

 September 28rd, 1884, on the banks of the (rironde, about seven miles 

 from St. Estephe. The insects were first noticed about 0.0 p.m., and 

 the flight lasted from one and a half to one and three-(juarter hours. 

 The dragonflies were from live feet to fifteen feet apart, and were 

 taking a steady up-river course, at a height of from ten to fifteen feet 

 above the ground, Avhilst the width of the flight was about 150 yards. 

 The observer suspected, from the size of the examples, that there were 

 two species, the larger proving to be .Jo. )iii.rta. The weather was fine 

 and warm, but the sky was clouded, and rain had fallen during the 

 day, whilst there was little or no wind. In the lienir ilcs (hlonatcn, 

 p. 181, Selys says that A-'l^clma <iran(li>i sometimes undertakes con- 

 siderable migrations. Another observer describes a migration 

 which he witnessed in the Alps, and states that "all the way 

 from Haspenthal to Andermatt the dragonflies were to be seen 

 coming up against the wind, which v.'as blowing from the west. There 

 was no mistake about it ; countless numbers were steadily passing up 

 tlie valley, but whither they were going it was hopeless to ascertain. 

 They did not seem to turn up the St. Gothard road, for I remarked 

 them the whole way up the valley to the foot of the Furka Pass 

 westwards." 



Walker, writing from the yacht " Aline," Palermo, April 5th, 1873, 

 records {Hiitmn., vi., ]). 457) that from the time tlie yacht left Tunis, 

 all along the ^lalta Channel, and on to It.Uy, the sea was covered 

 with large broAvn butterflies, motlis of all sizes, and dragonfiios, 

 evidently just dead, as they had apparently not been long in the water. 

 For four or five days, he adds, the yacht -;ailed througli many miles of 

 water without linding any diminution in their number. 



