UPSTREAM WIND 93 



which I had found in other years the most lucrative length 

 at the beginning of the season. 



The length which I selected to begin upon ran nearly 

 north-west to south-east, down to a short bend, when the 

 stream turned due east. The bend is always a satisfactory 

 place at which to observe fly coming down, as the stream 

 flows sharply and deeply along the western bank, and seems 

 to concentrate the hatch of duns into a yard-wide causeway 

 under the bank. And so, when I reached the spot, quite 

 naturally I ignored the absurdly early hour, and gave up 

 watching the snipe and the strings of wild duck and the 

 two pairs of swans preparing to nest, and the hundred and 

 one charming things with which the water meadows teem, 

 and concentrated on the current in the corner. It was 

 by now ten minutes to eleven summer time (9.50 a.m. by 

 Greenwich), and consequently at least two hours before 

 there was any reasonable hope of the fly beginning to show. 

 But I had my cast soaked and a Rough Olive floater tied 

 on and oiled, and I didn't feel a bit unreasonable. Pres- 

 ently eleven o'clock (summer time) chimed, and almost 

 before the echoes of the last stroke had died there appeared a 

 little dark form on the surface at the lower end of the bend. 

 It drifted, fluttering, for a little space, and then there was 

 a cheerful smack and a widening ring, and it was not. 

 It did not take long to get my distance ; the wind, blowing 

 from south-east by south, though strong, was not un- 

 friendly, and promptly I was covering the trout. Twice 

 he rose at natural flies, and three other fish began above 

 him in the same bend. The next one looked like a bigger 

 trout. However, I stuck to my fish, and at length brought 

 him up to my Dark Olive — short. I then cast across to 

 a fish I had seen rise just by a tussock at the point of the 

 inner corner of the bend. Up he came boldly — and missed. 

 I pondered. I knew my pattern was as good imitation of 



