once in these two years has a shoal been seen on the surface, and then for such a short period of time that it had disappeared before the net could be shot. Reports have been received from time to time of the occurrence of these fish at Cape Saunders and the Nuggets, but there has been no confirmation of these rumours. In the case of fish-stomachs examined at the station, sprats have been found in red cod, blue cod, southern king-fish, barracouta, and spined dog-fish; and pilchards in barracouta. The examination of other records would seem to show that both species of fish occur more or less permanently all the year round in the West Coast and Marlborough Sounds, and perhaps all round the coast, but that they only come to the surface when disturbed by some enemies; that they move along the coast at times in enormous shoals, usually between November to April, but most commonly during January to March; that this movement is most erratic, sometimes no fish being recorded for two or even three seasons; and that the shoals are harassed continually by porpoises, such predatory fish as barracouta, and countless flocks of sea-birds. What causes these migrations is a matter of conjecture. Clupeids are known to feed chiefly on free-swimming Copepoda, and these in turn depend upon diatoms, Peridineae, and other minute forms of vegetable and animal life. Therefore every investigation into the occurrence and movements of these fish must be preceded by an examination of the conditions of the sea, and the occurrence in it of these smaller forms. The routine work of the station at the Portobello Fish Hatchery for the future is to include regular observations on (1) the temperature and (2) the salinity of the sea in the sub-antarctic current which sweeps in a north-easterly direction along the coast to the east of Otago, and (3) the collection of tow-net material at the same time that the hydrographical observations are being taken. When shoals are met with the size and direction will be noted, the size of the fish, the condition of the reproductive organs, and the stomach-contents will be recorded, and specimens preserved. The regular work of the station has always included the examination of the stomach-contents of the fishes taken, and this will be continued in the future. The drift of the current has been observed by means of a drift-bottle experiment which was carried on for a considerable period, but the rate of movement has never been estimated by observation since Captain Stokes made his survey of the coast in 1840. As opportunity offers the scope of these observations will be extended both north and south, and other observers secured to assist in the research. Ultimately it is hoped to obtain an amount of accurate information on which will be based some estimate of the possibility of establishing an industry in the capture and utilization of these fish. At present it is clear that no definite statements can be made. The problem illustrates the importance of systematic and careful scientific investigation of marine conditions, and this is work which should be undertaken by all the States of Australia in conjunction with New Zealand. By co-ordination of scientific effort success in this and similar investigations is most likely to occur.