probably the chief barrier to its vertical extension; the food supply is sufficiently plentiful in, at any rate, the upper parts of the mesobenthic zone to present no obvious barrier. The chemical constitution of the water (except to animals in brackish water near river mouths) and the pressure appear to exert little or no influence; and only those species which attach themselves to clean hard substances would be repelled by the mud-line. The chief barrier to a horizontal extension of the epibenthos is undoubtedly temperature. As an example of its distribution may be taken the Gastropod and Lamellibranch Molluscs, as groups of which the distribution has been studied for many years by specialists. The shallow-water species fall into provinces (compare Cooke, Camb. Nat. Hist. vol. “Molluscs,” ch. xii), and a comparison of figs. 1 and 3 shows at once the profound influence upon them of the great currents. Taking the Atlantic Ocean, we find the Arctic species, tempted southwards by the cold Labrador Current, repelled northwards by the warm North Atlantic Drift. The Boreal or sub-Arctic species, many of which are identical on both sides of the ocean (2 and 2′, fig. 3), lie much farther southwards on the west than on the east side, from the same causes. The warm-water molluscs of West Africa (5) are cut off from those of the east side (7) by the cold water from the great easterly Antarctic Drift, which impinges on the Cape, giving it a special fauna (6). On the South American coasts the tropical and temperate fauna reach respectively to 28° S. and 45° S on the east coast, owing to the warm Brazil Current, but the corresponding groups on the west coast only to 5° S. and 37° S., being kept back by cold upwelling and Humboldt's Current. This influence is visible in individual species as well as in the facies of a fauna: Purpura lapillus, a temperate form, reaches on the east side of the Pacific to 24° N. and on the East Atlantic to 32° N.; but on the West Pacific only to 41° N and the West Atlantic to 42° N., being repelled by the Japan stream (and other warm currents of the south-west monsoon) and Gulf Stream respectively.
| |||||
|
Fig. 3.—Diagram showing the Coastwise (not seaward) Extension of the Provinces of Epibenthic Gastropods and Lamellibranchs Provinces.— | |||||
| 1. Arctic. | 7. Indo-Pacific. | 14. Peruvian | 16. Argentinian. | ||
| 2. Boreal of East Atlantic. | 8. Japanese. |
|
17. Caribbean. | ||
| 2′. Boreal of West Atlantic. | 9. Australian. | 18. Transatlantic. | |||
| 3. Celtic. | 10. New Zealand. | ||||
| 4. Lusitanian. | 11. Aleutian. | ||||
| 5. West African. | 12. Californian. | ||||
| 6. South African. | 13. Panama. | ||||
But while some species may be confined to a bay, others to a province, others to an ocean, there are cosmopolitan species which either vertical or horizontal barriers, or both, fail to restrain. In relation to temperature the wide-ranging species are termed eurythermal, the limited, stenothermal (Moebius); the terms are useful to record a fact, but are not explanatory. It seems to be the case that to every organism is assigned a minimum temperature below which it dies, a maximum temperature above which it dies, and an optimum temperature at which it thrives best; but these have to be studied separately for every species. Similarly, in regard to depth, species have been classed as eurybathic and stenobathic, but, since increased depth practically means diminished temperature, these are probably merely expressions of the same fact in another form. That an Arctic shallow-water species should stretch to considerable depths is not surprising, but it is remarkable to find such forms as, for example, Venus mesodesma on a New Zealand beach at 55° F. and in 1000 fathoms at 37° F. off Tristan d'Acunha. The provinces of zoological distribution, like the geographical divisions of mankind, must be taken merely to indicate the facies of a well-characterized fauna, not to imply the restriction of all its habitants to that area.
In considering the effect of temperature (and this applies to plankton as well as to benthos down to 100 fathoms), attention must be directed not only to the question of general warmth or cold as expressed by the mean annual temperature, but also to the range between the annual extremes: these ranges of variation have been carefully mapped by Sir J. Murray (Geog. Journ. xii. 113; compare ibid. xiv. 34). Still more important to the death-rate than these is the suddenness with which such variations occur: many animals are known to endure great extremes of heat and cold if exposed to them gradually, but to succumb to rapid alterations of temperature. Hence the frontier districts (Mischgebiete) between opposing currents are characterized by a heavy death-rate, and constitute marked barriers. A conspicuous instance of such a barrier in distribution is afforded at the Cape. The warm Mozambique Current, with a southwesterly direction off Natal, meets a north-east branch of the cold Antarctic Drift, and is beaten back eastwards: a result of the constant warring of these hot and cold currents is a high range of sudden temperature variation. Hence the Cape fauna
