13 DESCRIPTION, CLASSIFICATION, AND 



Both are undoubtedly correct. In the Grande De- 

 charge I have had exactly the same experience as 

 Mr. Creighton. But in addition to taking adults 

 there in September with milt and ova well devel- 

 oped and all other characteristics of the spawning 

 salmon present, I have seen thousands of the try in 

 some of the gravelly pools in the early summer. 

 Though there is a well-established case recorded by 

 Mr. Creighton of ouananiche spawning in a small 

 pool half-way up the wild falls in the Peribonca 

 Kiver known as the ckuie au diable, I believe that they 

 usually avoid the larger streams in selecting their 

 spawning-grounds and deposit their ova in the shal- 

 low water of the smaller rivers, where the newly 

 hatched fry incur less danger of being snapped up by 

 the hungry pike {Esox lucius), the water-wolf of Lake 

 St. John and its larger tributaries. In the peculiar 

 case mentioned by Mr. Creighton, parental instinct 

 could scarcely have devised a safer spawning-bed for 

 the immunity from this danger of the future fry. Not 

 many years ago, one of the best-known resorts of the 

 ouananiche for the spawning season was the Salmon 

 Biver, a small tributary of the Ashuapmouchouan. 

 The erection of a mill-dam upon the stream not far 

 from its mouth, and the absence there of any flshway, 

 have driven the fish elsewhere, and now perhaps the 

 favorite tributary of the Ashuapmouchouan as a spawn- 

 ing-ground for the ouananiche is the Eiviere du Cran. 

 Similarly conditioned feeders of the other large tribu- 

 taries of Lake St. John are also resorted to for the 

 same purpose. But there are spawning-beds for the 

 fish in the lake itself, and, it is probable, in most 



