SALMON— LAND-LOOKED RACES. 105 



S. solar in all points, and are not identical with any of the Scandinavian races 

 described by Dr. Giinther in the Oatalogue of the Fishes in the British Museum. 



There appears to be one remarkable point of divergence between the Scan- 

 dinavian and American land-locked salmon. In the first Malmgren found that 

 the eggs had decreased in number and diminished in size ; whereas Mather says 

 of the latter it " may be noted that though in maturity smaller, in embryonic 

 stages the land-locked salmon are larger ; the eggs perhaps being ten per cent, 

 greater in diameter." Atkins, however, found them smaller, so they may vary in 

 difEerent localities, possibly in accordance with the food supply. 



It is certainly remarkable that among the Scandinavian land-locked races some 

 are found in a number of lakes with broad outlets into the sea. Also respecting 

 the land-locked salmon of America, Mr. Atkins, in 1884, observed that he did 

 not think we have any evidence that the land-locking of the species has occurred 

 during recent geological periods. " There is nothing at present to prevent any of 

 these salmon going out to sea from any of those waters where they are now found. 

 There are obstructions to their coming back if they once went to the sea ; and 

 these same obstructions would hinder the sea salmon having access to the upper 

 waters, where the land-locked salmon now live."* 



Should it be decided to etockf some of our larger and clear lakes with this 

 race of land-locked salmon, J it might be worth considering -whether eggs 

 obtained from Lake' Wenern, which is slightly north of the British Isles, would 

 not be more likely to succeed than those from the United States, where the 

 waters are entirely to the south of Great Britain. While a trial of the two forms 

 would be of greater utility than all the theories which could be advanced on the 

 subject. 



It seems very questionable whether these fishes are suited to our rivers, which 



* The land-locked race of salmon found in the State of Maine and neighbouring portions of 

 Canada, in Lake Sebago, Lake Sabec, and the Schoodic lakes, are also identical in external form 

 with Salmo salar. That similar conditions in America may occasion the same changes is exceed- 

 ingly interesting, but only what might be expected. These salmon Agassiz considered to be 

 identical with the lake sahnon, or SaXmo argenteus (silfverlax) of Sweden. 



Mr. Atkins, who has made a special study of these American forms of land-locked salmon, 

 tells U3 that there are differences which point towards a theory of arrested development, 

 such as the dark par bands which are retained through life, as may be seen by removing the 

 scales. Such appearances were not present in any of the Wenern fish as have come under my 

 observation, perhaps twenty or thirty in all. In the United States some land-looked forms give 

 fewer and smaller eggs than the anadromous race, while ovarian disease has been found to be 

 frequent. They only cease feeding for a few weeks at the breeding time, and undertake two 

 migrations yearly frorn the lakes to the streams. The first is in May, and continues untU August ; 

 it seems to be for the sole purpose of feeding. The second or " fall " migration begins in 

 September, and ceases at the commencement of winter, and is for breeding. 



The land-locked salmon in the breeding season never assumes such bright colours as do male sea 

 or river salmon, while Atkinson observed of the Schoodic lake, " Here, as in many other instances 

 that I know of, the salmon move down from the lake into its outlet at the spawning season, 

 instead of up into the tributaries. It follows that the young fish, instead of dropping down with 

 the current, as many sea-salmon do, are in the habit of ascending their native streams tiU they 

 reach the deep water above." 



These salmon, in the United States, as well as in Canada, are inferior in size to the anadro- 

 mous race for in the Penobscot the average adult is given at about 13 lb., while among the Sebago 

 race the largest average 5 lb., although a solitary instance has been recorded of one of 25 lb. that 

 had been found dead. Although these forms seem to thrive best where the waters are very deep, 

 still to this general rule there are some exceptions. Large numbers of eggs from this race, some 

 of which are much larger than those of our salmon, have been brought into this country, and 

 many hundreds of yearlings are thriving at Howietoun. 



f In some cases it has been found that the introduction of sabnon into pieces of water 

 frequented by trout has not been conducive to the well-being of the latter species. 



{ In case at a future date the following observation from Land and Water, 1886, should raise 

 discussion, I may mention that the fish alluded to below (of which I have a series of specimens) 

 were trout, S. purpuratus, but not S. salar. The extract alluded to was as foUows : — ■ 



"Laud-Looked Sadmon. Three specimens of this Canadian fish, about 14 in. in length, 

 which have been presented to Mr. Thomas Spreckley by the Science and Art Department at 

 Kensington, were taken to Chertsey on Thursday last, and placed in one of the tanks belonging 

 to Mr. James Forbes, of Chertsey Bridge, prior to their being turned into the Thames in that 

 neighbourhood." 



