Zealand, 108; Hawaii, 144; West Indies, 299, and the Panama region, 256.
Common to Japan and the Mediterranean are 79, all but two being of wide distribution; to Japan and the Red Sea, 111; to Japan and Hawaii, 82; to Japan and Australia, 135; to Japan and the West Indies, 113; to Japan and Panama, 91. To the Mediterranean and the Red Sea, 40 genera are common, all of wide distribution; to the West Indies and the Mediterranean, 70, 59 being of wide distribution; to the West Indies and Panama, 179, only 101 being of wide distribution.
It is evident from an analytical table that the warm-water fauna of Japan, like that of Hawaii, is derived from that of the East Indies and Hindostan; that the fauna of the Eed Sea is derived from the same source; that the Mediterranean fauna bears no special resemblance to that of Japan rather than to that of the other parts of Eastern Asia with like conditions of temperature and no greater than is borne by the West Indies; that the fauna of the two sides of the Isthmus of Suez have relatively little in common, while those of the two sides of the Isthmus of Panama show a remarkable degree of identity.
When the fishes of Panama were first described, it was claimed that their species were almost entirely identical with those of the West Indies; this statement was followed by speculations on the relation of the depression of this Isthmus to the Gulf Stream, and to the glacial epoch. Further investigations by Jordan, and by Evermann and Jenkins showed the fallacy of this claim of identity. Of about 1,400 species now known from the two sides of the Isthmus, only 70 are identical, or five per cent, of the whole, and about 10 of these are almost cosmopolitan in the tropics. Dr. Paul Fischer finds about three per cent, of the mollusks identical on the two coasts.
Dr. E. T. Hill goes on to show that there is neither geological nor biological evidence of the submergence of the Isthmus of Panama since Tertiary times, and that such a barrier existed as far back as Jurassic times. There is, however, evidence of a brief connection in Tertiary time at the end of the Eocene period.
Assuming this to be true, the actual facts of distribution seem to be in accord with it. The period of depression was before the life-time of most of the present species. It was, however, not earlier than the period of most of the present genera. It was relatively shallow, but wide enough to permit the infiltration from the Caribbean Sea to the Pacific of species representing most of the genera of sandy bays, rocky tide pools and brackish estuaries. Since the channel was closed, the species left on either side have undergone modification in varying degrees, mostly retaining generic identity, while losing some of their specific characters.
