148 FLY FISHING FOE TROUT. 



The Grannom comes up in April and lasts about 

 a fortnight : the dates of its appearance and 

 disappearance are clearly marked. The Trans- 

 lation of St. Thomas of Canterbury is 7th July, 

 and I consider the Treatise particularly 

 accurate in dates, and I never saw a Grannom, 

 or heard of one being seen, so late as that. So 

 reluctantly I rejected it. But my scepticism 

 was considerably shaken by finding that Ronalds 

 both uses Shell Fly as a synonym for Grannom 

 and also found the fly, or one like it, in trouts' 

 stomachs in August ; and in his fifth edition says 

 that the Grannom if dressed buzz is a good fly 

 all the summer months into September. Cotton 

 gives the Shell Fly for July, but considers that 

 it was taken by the trout for the palm that drops 

 off the willow into the water, and other writers, 

 who cribbed from the Treatise or Cotton, also 

 give it. But it is one of the flies specifically 

 knocked out by Bowlker, and I do not think it 

 reappears till Ronalds resuscitates it. Ronalds, 

 extremely accurate, says definitely that he 

 found the fly in trout in August : and possibly 

 there is a fly which appears then with a green 

 egg-bunch like the Grannom. This difficulty 

 illustrates the intolerable burden under which 

 we fishermen labour in not having a good modern 

 entomology. Halford's book is not satisfactory. 

 It is the work of one who was a great fisherman 

 but not a naturalist, and I do not think that 

 anyone who is not could possibly succeed. And, 

 apart from this, it suffers from two defects : it 



