FISHES sea water, has already been mentioned. In 1888 Raffaele, an Italian zoologist working at Naples, described buoyant ova from the sea with certain peculiarities, and suggested that they belonged to the sardine. About the same time the present writer working at Plymouth obtained similar ova from the sea, and from comparison of these with ripe but unfertilized eggs taken from the pilchard concluded that they belonged to this species. In the two or three years following Professor Pouchet investigated the question on behalf of the French Government at Con- carneau in Brittany, with the object of throwing some light on the cause of the failure of the French sardine fishery at that time. He refused to believe that the eggs of the sardine were buoyant, and maintained that there was no evidence that the fish spawned near the coast at all. At the same time Professor Marion at Marseilles from his own observations sup- ported the conclusions published by Raffaele and the present writer. The question was finally settled in 1893, when the present writer obtained living spawn actually taken from pilchards and artificially fertilized, and found that the ova agreed in every respect with those previously assigned to the species. The spawn thus obtained was actually hatched. The development of the young pilchard was also traced at Plymouth, and the young fish only 2 to 3 inches long were obtained from the stomachs of mackerel in November 1891, and from the Hamoaze in September 1893. In former times, until about the year 1883, pilchards annually visited the north coast of Cornwall, especially St. Ives Bay, in autumn in large shoals, but at present they are seldom found in abundance on this part of the coast, where herrings have taken their place. It is impossible to decide from any evidence at present available what is the cause of this change in the migrations of the pilchard. It would appear that there is on the north coast of Cornwall an alternation of pil- chard periods and herring periods, and it is natural to surmise that this is due to some alternating change in the physical conditions of the sea. As the Cornish coasts form the northern limit of the range of the pilchard, it seems possible that in certain periods the drift of warm water from the south extends further to the north, and that the pilchard then extends its wanderings to the north coast of Cornwall, while in other periods the drift of warm water is weaker or takes another direction, and that for this reason the north coast is deserted by the pilchard and visited by the her- ring. The present writer attempted to get evidence upon this point by comparing observations of the surface temperature of the sea taken by fishermen at various parts of the coast, but constant differences supporting the above suggestion were not discovered. It might be suggested that the change is due to the increase of drift-net fishing, but this does not seem probable, for it is difficult to see how this method of fishing could at the same time cause an increased abundance of herrings. The estuaries of Cornwall are broad and long, but the fresh waters above the reach of the tide are for the most part narrow and of limited 293
