But in all experiments of this kind, the great principle is, to have a constant current of fresh and aerated water running over the eggs. The uniform supply of air to the fœtus in the egg is essential for its life and growth, and such eggs as are not supplied with water saturated with air are unproductive. The experimenter must be guided exactly by the instinct of the parent fishes, who take care to deposit the impregnated eggs, that are to produce their offspring, only in sources continually abounding in fresh and aerated water. PHYS.—But as every species of fish has a particular and usually different time for spawning, I do not see how it could be contrived to cross their breeds, or how the ova of a trout, which spawns in December, could be impregnated by the seminal fluid of the grayling, which spawns in May; for I conclude it would be impossible to preserve the eggs of a fish out of the body in a state in which they could retain or recover their vitality. HAL.—I believe I mentioned before, that I had found instances, in which the ova of fish were developed at a different period from their natural one; and I have no doubt, that a little inquiry respecting the habits of fishes would enable us to acquire a knowledge of the circumstances, which either hasten or retard their maturity. Plenty of food and a genial season hasten the period of their reproduction, which is delayed by want of proper nourishment, and by unfavourable weather. Males and females likewise, confined from each other, have their generative powers impeded; and trout, grayling, and salmon, will not deposit their ova except in running water; so that by keeping them in tanks, the period of their maturity might be considerably altered. I have seen char even, which had been kept in confined water from September till July; and so slow had been the progress of the ova, that they appeared to be about this time fit for exclusion, though, in the natural course of things, they would have been ripe in the end of October of the year before. By attending to and controlling all these circumstances, I have no doubt many interesting experiments might be made, as to the possibility of modifying the varieties of the salmo, by impregnating the ova of one species with the seminal fluid of another. With fishes of other genera the task would be still more easy. Carp, perch, and pike, deposit their ova in still water in spring and summer, when it is supplied with air by the growth of vegetables: and it is to the leaves of plants, which afford a continual supply of oxygen to the water, that the impregnated eggs usually adhere; so that researches of this kind might be conducted within doors in close vessels, filled with plants, exposed to the sun. I have myself kept minnows and sticklebacks alive for many months in the same confined quantity of water, containing a few confervæ; and their ova and melt increased in the same manner, as if they had been in their natural situation.