All fish are not oviparous. There is a well-known blenny which is viviparous, the young of which at the time of their birth are so perfect as to be able to swim about with great ease; and this fish is also very productive. Our skate fishes (Raiæ) are all viviparous. “The young are enclosed in a horny capsule of an oblong square shape, with a filament at each corner. It is nourished by means of an umbilical bag till the due period of exclusion arrives, when it enters upon an independent existence.” I could name a few other fish which are viviparous. In the fish-room of the British Museum may be seen one of these. It is known as _Ditrema argentea_, and is plentifully found in the seas of South America. But our information on this portion of the natural history of fish is very obscure at present. There are many facts of fish biography that have yet to be ascertained, and which, if we knew them, would probably conduce to a stricter economy of fish life and the better regulation of the fisheries. Beyond a knowledge of mere generalities, the animal kingdom of the sea is a sealed book. No person can tell, for example, how long a time elapses from the birth of any particular sea fish till the period when it is brought to table. Sea fish grow up unheeded—quite, in a sense, out of the bounds of observation. Naturalists can only guess at what rate a codfish grows. Even the life of a herring, in its most important phase, is still a mystery; and at what age the mackerel or any other fish becomes reproductive, who can say? The salmon is the one particular fish that has as yet been compelled to render up to those inquiring the secret of its birth and the ratio of its growth. (See _Natural and Economic History of the Salmon_.) We have imprisoned this valuable fish in artificial ponds, and by robbing it of its eggs have noted when the young ones were born and how they grew. It would be equally easy to devise a means of observing sea fish. Why should we not erect a great marine observatory, where we could, as in the case of the Stormontfield-bred salmon, watch the young fish burst from its shell, and for a year or two observe and study the progress of the animal, and ascertain its rate of growth, and especially the period at which it becomes reproductive? The government might act upon this suggestion, and vote a few thousand pounds annually for the support of a series of marine fish-ponds; for something more is required than the resources of an amateur naturalist to determine how fish live and grow. What naturalists chiefly and greatly need in respect of our sea fish is, precise information as to their rate of growth. We have a personal knowledge of the fact of the sea fish selecting our shores as a spawning-ground, but we do not precisely know in some instances the exact time of spawning, how long the spawn takes to quicken into life, or at what rate the fish increase in growth.