338 THE RIVER-SIDE NATURALIST. 



The female, when she throws off her sub-imago state, 

 flies away to find a husband, the two combined falling 

 gradually to the ground, or to the surface of the water ; 

 but just before reaching it the male detaches itself and 

 mounts again in the air, while the female proceeds to 

 deposit her eggs on the surface of the water, which sink 

 at once to the bottom of the river. She then has to go 

 through many dangers in her short career ; she may be 

 snapped up as she is laying her eggs either by fish or bird, 

 or she may get her wings wet and be unable to rise, and 

 so is hurried away by the stream, utterly helpless, and dies 

 exhausted. 



The eggs extrude from the seventh ventral segment in 

 the female, and we observed that she sheds or lays a 

 certain number every time she dips on the surface of the 

 water. We caught them just after dipping, and then 

 found none ready for immediate expulsion ; and we caught 

 them after having risen and remained in the air for a 

 few moments, and then found a cluster of eggs half-exuded 

 and ready for expulsion. 



The May-fly season of 1887 was characterised by the 

 fly first appearing some days later than is usual, and also 

 in some localities by enormous numbers, which pervaded 

 the streams and adjacent bushes and meadows. Another 

 peculiarity was, that there was no morning hatch, although 

 there was for the most part a bright sun, but the wind was 

 either north-east or east. The fly generally did not hatch 

 till after midday, and then in small quantities, the great 

 hatch taking place from 3.30 or thereabouts, and continued 

 till 8 or 9 P.M. On some afternoons the air appeared full 

 of a dense mist, which was nothing but swarms of May- 

 flies, and the accumulation of the dead was so great at 

 some of the hatchways and elsewhere as to produce a most 

 nauseous odour. 



Mr. Eaton states "that many of the characters upon 

 which the classification of the Ephemeridce was formerly 

 based have proved to be unsuitable for the purpose. 

 Originally the number of the caudal setce was deemed a 

 matter of primary importance ; and when forms were dis- 



