Fishes. 4991 



under the gravel of fresh -water streams, and that the young springing 

 from these eggs appeared in the rivers as smolts, or silvery-scaled fish, 

 in May, and in May only ; but it was not determined whether the 

 May smolt was the product of the incubation of the preceding winter 

 or of the one or ones before that ; in other words, the age of the May 

 smolt, that is, of the young of the salmon, after it has undergone its 

 last metamorphosis before descending to the ocean, had not been 

 determined. This problem is still unsolved. 



2. At the period I allude to it was also perfectly well known that 

 numerous small fish, with natural history characters seemingly iden- 

 tical w T ith young salmon before they have undergone their last meta- 

 morphosis, were to be found at all times in salmon rivers ; that these 

 questionable and dubious fish were male and female ; that they were 

 of various sizes and strongly resembled young salmon in shape; that 

 the male acquired occasionally a well-developed milt capable of fecun- 

 dating artificially the ova of a full-grown salmon, whilst, on the con- 

 trary, the roe of the female of these fishes was never found developed. 

 Some naturalists fancied these little fishes to be a peculiar kind or 

 species of fish ; practical fishermen believed them to be simply the 

 young of the salmon ; the physiological difficulties in the way of such 

 opinions remain just as they were. I proved, however, by direct ob- 

 servation that in a few days or weeks after the smolts had descended 

 to the ocean, these peculiar salmon-resembling fish, known by various 

 names as parr, brandling, &c, were to be taken larger than the smolts 

 which had just left; that the milt was developed in these same fishes 

 at times when no such thing happened in the grown salmon, and that 

 therefore their whole history was still a mystery. To this was added 

 the fact, which I believe to be correct, that in some salmon-frequented 

 rivers the parr (the fish I speak of) is not found. 



3. The habitat of the salmon whilst in the ocean is not known : its 

 rapid growth in a few weeks from a May smolt to a well-grown salmon 

 has been proved experimentally, so also has the fact that salmon do 

 not feed in fresh waters, but fall off constantly and would perish if 

 not permitted to return to the ocean. 



4. The transplanting of salmon and salmon-trout into fresh waters, 

 as a permanent residence, has uniformly failed. 



5. The metamorphoses of the salmon from the egg to the smolt have 

 not been accurately traced, the changes having been noted only in 

 those subjected to restraint in artificial ponds and reservoirs. By the 

 Drumlanrig experiments the fact was made out, that by confining the 

 young in small fresh-water ponds the metamorphosis of the young 



