17

11.—12

size of the ova, thoroughly clean, and boiled to kill all vegetable or animal matter in it, should be placed at the bottom of the boxes, and a board covering should be put over them, to keep the light from the ova. Every morning the dead eggs should be picked out by means of the catcher, and as the young ones begin to hatch out, the stream of water should be gradually-increased, as young fish consume more oxygen than eggs do. The eggs, when dead, turn opaque and white : some of the eggs contain no embryo, but yet they do not turn white. They should not be removed till they do turn white. The formation of the embryo is shown by two small black specks, which are the eyes of the fish. The young ones, on emerging from the egg, are provided with a bag or sack (the " umbilical vesicle"), which is absorbed as the fish grow. When this vesicle has nearly disappeared, the young fish should be fed with red worms, or the yolk of egg boiled hard and cut fine, or the food of larger fish very finely powdered, namely, rump steak and biscuit, as below described. During this time the troughs should have a little charcoal, in small lumps, sprinkled in them, to prevent'the putrefaction of the food. The moment the fish begin to die they should be thinned; and my advice is that they should be placed in wooden troughs 5 or 6 feet long, 18 inches wide, and 1 foot deep. Into these troughs they should have certainly double the current of water running that was passed through the hatching boxes. The water should be let to run into the troughs through a series of jets, as shown in the diagram. (Diagram No. 3.) If they still die, thin them again, and immediately make more boxes ; if they still die, they should be turned out. I never have approved, and never will, of semi-stagnant ponds as a nursery for salmon. If I were in New Zealand, I should distribute my fish, when they were five or six months old, up and down in the shallow parts of the tributaries of the stream in which I wished to acclimatize them, choosing a locality where the water was generally running about six or eight inches deep. I would arrange stones (the natural stones of the bottom of the brook), which should be of gravel, in a shape V (Diagram No. 4), covering them with larger stones or slates, leaving free ingress and egress between the stones: these would form hiding-places for the fry. Before doing this, I would carefully drive out of the stream, by walking in it or netting it, all natural enemies, such as eels, &c ; and choosing about half a mile of stream for this purpose, I would place fine gratings at the top A (Diagram No. 5) and at the bottom B, aud cutting down trees or poles, I would make shelters, about 3 feet wide, for the fish right across the stream, about two feet from its surface, by placing on these poles the boughs of the trees growing near, or rough hurdles, which should be placed at intervals of 3or 4 feet: this should be done to keep off' the sun. I would appoint a man to carefully watch this mile or half-mile of water, and see how things went on. When these fish arc in their natural nursery, I think they should be fed for two or three months with rump steak, powdered finely up with biscuit powder or oatmeal, in such proportions as to make a thick paste, so as not to stick to the fingers. This should be thrown in very small particles at the head of the stream, and allowed to float down. The only thing I am afraid of is, that possibly floods may come, and sweep the fish all away before they are strong enough. If there is any danger of this, a hatchway should be made at A, and an artificial channel ABC (Diagram No. 6) should be cut, so as to allow a current of water throughout, but to be out of the influence of floods. I am certain that this will be foitnd a more successful way of breeding salmon than in any ponds, and I would beg to refer to my report on the artificial breeding of salmon in Scotland (p. 134), of which I enclose a copy. A careful watch should be kept on the salmon, and they should be allowed to go to the sea when they please. A great difficulty now occurs as regards their prosperity in the sea. I should first have the mouth of the river and the estuary carefully netted with small meshed nets at once, aud samples taken of all sea fish that will be proved to be indigenous to the estuary. These fish should be submitted to some scientific authority, to see if they are likely to be injurious to the young salmon, —I may make bold to say that I should like to have samples of them myself, simply packed in salt, with a hole cut in the intestines to admit the salt. I do not know at what month the young salmon would migrate to the sea in New Zealand : in England, I may remark that it is in May and June, and the corresponding months of the New Zealand seasons would probably be the time. At any rate, just previous to the time of navigation the small meshed nets should be constantly going at the estuary, and every enemy frightened away as much as possible. Careful observations should be kept as regards the relative temperatures of the sea and river. Not knowing the habits of the New Zealand sea fish, I am unable to advise how they should be frightened from the estuary; local fishermen will probably give information on this point. In order to ascertain whether the experiment has been successful or not, and whether the young fry return as grilse or salmon, I should most decidedly recommend at the end of the second year a fishing weir should be built, right across the river, at some convenient spot, according to enclosed plan (Diagram No. 7), and this will test, at a very slight expense, whether the salmon really come back or not. If this had been done in Australia, the question would have been settled long ago. 3—H. 12.