NATURAL HISTORY OF THE SALMON. 67 



sively, as the figures of the earlier months. In the montlis 

 of July and August, nearly eight-tenths of the whole 

 grilse of the year are captured ; in the month of August 

 alone, more than four-tenths. But September, as compared 

 with August, shows a diminution of two-thirds, or 66 per 

 cent., the capture of that month being three-twentieths of 

 the capture of the whole season ; October shows a diminu- 

 tion of another third ; and November, were the fishing 

 continued, would probably show pretty nearly a blank. 

 Turning, however, to the salmon, we find that in those 

 two months, July and August, when eight-tenths of the 

 whole grilse ascend, only four-tenths of the whole salmon 

 ascend ; that ■ in September, when the grilse have de- 

 creased 66 per cent., salmon have decreased only 30 per 

 cent. ; and that in October, as compared with Sep- 

 tember, when grilse show a decrease of another third, 

 salmon show an increase of nearly one-third, and have 

 become again almost as numerous as they had been in 

 September. Briefly, salmon ascend in every month of 

 the year, in numbers, comparatively speaking, not very 

 unequal ; grilse, speaking roundly, do not ascend at all 

 in the first half of the year ; all of them, but a fraction, 

 ascend within two consecutive months in the middle of 

 the year, and in the latter months of the year their 

 ascent almost ceases. Mr. Mackenzie would account for 

 all this by saying that these are two different species of 

 fish ; and he finds it necessary to go the length of saying 

 that the one is a spring, and the other a summer fish. 

 It might be possible, were there no facts beyond those 

 we are at present dealing with, to assume that there 

 are two species of migratory fish, one of them, not a 



