consists mainly of only such species from neighbouring provinces as can endure high sudden variations; and the district is practically impassable. For example, nineteen species of Echinoids are known from the Cape district. Of these twelve are peculiar to the Indo-Pacific province, which stretches from East Africa to the Sandwich Islands and from Japan to Australia; two species are Southern Ocean forms, all but confined to south of 40° S.; four species are peculiar to the Atlantic Ocean: of these eighteen not one gets past the Cape into the next province; the nineteenth is practically a cosmopolitan (A. Agassiz, “Challenger” Reports: “Echinoidea”; compare also C. Chun, Aus den Tiefen des Weltmeeres, pp. 157, 158).
Among the barriers to the horizontal extension of epibenthos must be mentioned a wide deep ocean. The Indo-Pacific fauna ranges from East Africa to about 108° W., stepping from island to island over the Pacific; but this continuity is then broken by 37 degrees of longitude and more than 2000 fathoms of water, and such sessile species as are most Mollusca (cf. fig. 3) are unable to reach the American coast. This is presumably due to the fact that the planktonic larvae of epibenthic adults must settle on a suitable bottom within a certain period or die. In spite of the direct set of the currents from Florida to the British Isles, the epibenthos of the two is absolutely dissimilar; the similarity of the two Boreal provinces (2 and 2′, fig. 3) is to be assigned to a former continuity by way of Greenland, Iceland and Faeroe; a similar continuity, still unbroken, is exhibited by the Aleutian province on both sides of the Pacific. Though larvae cannot cross wide oceans, adults may no doubt traverse great stretches occasionally on floating timber, &c.
This barrier by distance may be instanced in another way. In the Arctic regions land masses are continuous or contiguous, and there are many circumpolar species, as, for example, Rhynchonella psittacea; towards the South Pole the southern continent is almost ice-bound, and the available land consists only of the tips of the continents and of the few oceanic islands. Hence few if any littoral species are circumpolar. For example, not a single littoral Ophiurid surrounds the South Pole, but five or six species are circumpolar in the northern hemisphere.
Taking next the mesobenthos and hypobenthos, living at depths where temperature is constant and current practically negligible, Mesobenthos; Hypobenthos. there appears theoretically to be no reason why an organism which can thrive at 500 fathoms should not have a world-wide range over the bottom of all oceans. Yet this is not often, although occasionally, known to be the case; and although perhaps, speaking generally, hypobenthic species have wider ranges than epibenthic, still they also seem to be limited. It must, however, be remembered that the ocean is large, deep hauls of trawl or dredge few, and individuals at great depths scattered, so that too much stress must not be laid on this point. The “Challenger” results seem to allow of at least one generalization—the deeper the fauna, the wider its range. This is shown by the following table of the “Challenger” benthos: the first column gives the number of benthos species captured at depths indicated in fathoms by the second column, the percentage of these species which is known to have been captured between the tropics, as well as south and north of the tropics, is shown in the third column:—
| Number of Specimens. |
Horizon. | S. T. N. |
| 4248 | 0-100 | 0.6 |
| 1887 | 100-500 | 2 |
| 616 | 500-1000 | 4 |
| 493 | 1000-1500 | 7 |
| 394 | 1500-2000 | 7 |
| 247 | 2000-2500 | 9 |
| 153 | over 2500 | 9 |
We can only guess at the causes of the apparently limited range of many deep-sea types. (a) One of these is probably the limited food supply: presumably, as with a land fauna, there are as many mouths in a given area as it will support, and an equilibrium of species is maintained which will at least hinder the extension of any one. For food the bulk of the deep-water fauna is dependent upon the rain of dead organisms falling from higher levels, these, slowly disintegrating (probably under chemical, not bacterial, action), seem to form with the bottom deposit a kind of nitrogenous ooze, through which many deep-sea organisms slowly swallow their way, as an earthworm goes through earth extracting nutriment. (b) Another hindrance to the extension of many deep-sea species is that they are holobenthic, that is, do not pass through a free-swimming larval stage; the means of dispersal is therefore regulated by the animal's own power of locomotion. Generally speaking, as might be expected, the freely-moving hypobenthos, fish and crustacean, have the widest ranges, and even these are not helped by currents, as are epibenthic or planktonic forms. The larval history of deep-water forms is, however, unfortunately obscure. (c) Lastly, extension of area of a species being at best difficult in deep water for non-swimmers, the place and date of their first migration must be taken into account; forms which have comparatively recently adopted deep-water life cannot be expected to have spread far from their original centre. As regards this point, in the first place, it is with migration, not with local evolution, that we have to deal: no classes and orders, only a few families and genera, rarely sub-orders, are peculiar to the hypobenthos; the deep members of each group consist for the most part of widely separated genera, the species do not grade into each other, as is so often the case in the epibenthos; and evolution could hardly have produced these species and genera under the uniformity of their present environment. This migration downwards from the mud-line has no doubt occurred all over the world, notably in the Southern Ocean, if we may judge by the richness of the deep-water fauna there to-day; probably also largely in Arctic and sub-Arctic regions, less so in tropical and temperate zones. As to the date of migration, the following fact seems to show that it is of comparatively recent origin, and is indeed still in progress: taking the “Challenger” species from the epibenthos, from the mesobenthos, and then from zones of 500 fathoms down to 2500, each zone shares a larger percentage of species with the zone above it than with that below it (except in one case where they are nearly equal). But it is not to be supposed that all our present-day deep-water forms began their migration simultaneously, and we can say with fair certainty that migration to deep water did not begin before the close of the Mesozoic epoch. Had it begun earlier, we should find typical Mesozoic and even older forms, or their conveners, at great depths: so far is this from being the case that the most venerable animals of to-day—Lingula, Amphioxus, Limulus, 75% of Crinoids, 90% of Brachiopoda, &c.—are epibenthic or mesobenthic. On the other hand, it is extremely likely that the Cretaceous epoch marked the commencement of migration. The hexactinellidan sponges are known to have lived in quite shallow water at the date of deposition of the Inferior Oolite; to-day none occur at a less depth than 95 fathoms; and as only two genera are known from the shallow Tertiary deposits, it would seem that the migration began about Cretaceous times (“Challenger” Reports: “Hexactinellida, ” F. E. Schulze). In 1881 (A. Agassiz, “Challenger” Reports: “Echinoidea”) 105 living genera of Echinoidea were admitted; of these 23% were known from Cretaceous but not from Tertiary deposits, 35% from Tertiary but not Cretaceous, and 40% as Recent only. The species of Cretaceous genera constituted only 29% of the epibenthic Echinoids, 44% of the mesobenthic, and no less than 55% of the hypobenthic. These species of Cretaceous genera were distributed fairly evenly over all three zones, but 72% of the species of Tertiary genera and 55% of the Recent forms were confined to the epibenthos. As out of the twenty-five living genera known from the Cretaceous only seven are known also from Jurassic deposits, it is obvious that the close relationship is between Cretaceous and hypobenthos, rather than between any other geological and bathymetric horizons. Other instances, such as that of the Eryonidae, seem to point to similar conclusions.
Excepting the essential air-breathers, practically every phylum and class and most orders are represented in the benthos. The
