Deterioration. Regarding the growth of trout in the mountain-lakes of eastern Norway in comparison with the degenerate condition of trout on the western side, Dahl (1919, p. 28) notes as follows:— — West. East. Food Mostly insects and small organisms Mostly large animals, fresh-water shrimps, snails, and Lepidurus. Lakes Often deep, and therefore little productive Often shallow, and therefore more productive. Spawners Small and young, therefore vigorous reproduction Larger and older, therefore slower reproduction. Ova Small, with small growth-capacity Large, with better growth-capacity. Dahl's researches are of great interest and importance; but much further investigation seems to be required before these reasons and results may be accepted in their entirety. Armistead (1920, p. 58) states as follows: “A stock of mountain-trout subjected to a favourable environment may grow and improve for some years. After a time a recoil takes place and the improvement is replaced by a deterioration, apart, as far as I can tell, from the question of food. It is as though the vitality accumulated originally through generations of hardship was exhausted in the process of growth.” Dahl (1919, p. 33) states that “growth depends on the qualities of the mother fish and the size of the ovum (i.e., the size of the yolk-sac of the ovum).” Thus it would seem that, apart from the amount of food-supply available at the stage when feeding commences, the whole future history of the trout depends (1) on the amount of nutriment available for the embryo in the yolk-sac, (2) on what may be termed the inherent constitutional vigour imparted by the parents to their progeny. There can be no doubt that decreasing food-supply has a direct bearing on the question in the thermal lakes, but I am of the opinion that this is not the solution of the whole problem. In regard to Dahl's tabulated observations, it would be interesting to ascertain whether outside fry or young trout were introduced into any of the lakes to augment the parent stock; also whether the trout of the eastern lakes of Norway had greater natural facilities for sexual intermixing than was afforded the trout on the western side. Further, it may be that certain inorganic constituents of the separate waters have been responsible for the predominance of two different types of plankton and benthos.* Phillipps and Grigg (1922) have given considerable data on the relations of organic and inorganic geochemistry to fish life. In the thermal-lakes district I have examined the yolk-sac of the ovum of fishes of different sizes and ages, and have found that in large trout, six to eight years, the yolk-sac is relatively larger than in younger and older trout. The rainbow trout of these lakes reach their maximum weight and condition at about six years. In my opinion it is these large trout (which at the age of six years weigh anything up to 9 lb.) that more than others will be likely to produce a strong and healthy progeny, and thus aid in maintaining the basic standard of the race. Progeny derived from the same parents may not impress upon their progeny a strong constitutional vigour. Milne (1917, p. 37) writes: “It has been noticed that if eggs are collected annually to the full capacity of a minor tributary in a large watershed, and some of the progeny are