SUMMARY

Pelagic communities may be described as black boxes where unknown processes transform matter and energy. Forecast and/or control of any community requires the knowledge of its structure and functions.

For the study of pelagic communities, the localities of finds of 713 more- or-less-well-recorded species have been used and these were interpreted in different ways. Areal distributional maps of better-known separate species were prepared (e. g., Figs. 9, 11, 14) which, ecologically, represent the distribution of single-species populations. Biogeographically, they do not represent ranges since they comprise both reproduction areas and sterile expatriation areas. Here I agree with Ekman (1953) that sterile expatriation areas do not belong to the range in the proper sense. To distinguish between reproduction and sterile areas is by no means easy, though. Maps of the average distribution of species based on all locality records are fully comparable only with the distribution of species in bottom sediments.

Areal distributional maps of individual species were subjectively grouped into joint maps according to their shape and position on the globe (Figs. 34—40, 46—54, 60—64). The resulting joint distributions may biogeographically be regarded as types of ranges, and ecologically as recurrent groups. Large-scale pelagic communities are made up of recurrent groups rather than immediately of species (Fager and McGowan, 1963; Kanaya and Koizumi, 1966; Krylov, 1968). The patterns of the groups within the commu nities overlap only partly; there are species with very restricted distribu tions, and faunal areas may be both rich and impoverished.

If one disregards the neritic zones the Antarctic region has no further subdivisions. The Arcto-Boreal region divides into two subregions, the Atlantic and the Pacific, the latter comprising a southern and a northern zone. The Tropical region divides into the Pacific, Atlantic and Indian sub- regions. Each of them has two provinces, the Central and the Equatorial. The Pacific and the Atlantic Central provinces are anti-tropical, the Equatorial province dividing them into halves. The halves may in turn be divided into «zones» and «parts» with minor faunal differences behind them. In the open ocean, «zones» are more or less latitudinal, but nearer to the coasts, in ihe distant-neritic areas (see below), latitudinal boundaries split and diverge, crossing each other (Figs. 44, 59, 66). Consequently, a mosaic of «parts» arises all around the continents (incidentally, this is a clear case of cireum- continental zonation, sensu P. L. Bezrukov, 1964). The boundaries between «parts» are more or less meridional or cut meridians at various angles. «Parts» can also be found in the Equatorial province.

«Zones» of the Central provinces differ in the southern and the northern hemispheres, the northern ones being telescoped (Table 7). In the Central province of the Indian Ocean there is one «zone» less than in two other oceans since Indian equatorial species have less diverse types of ranges.

Biogeographical boundaries are asymmetrical about the meridional plane because of (1) the Transition areas in the Peru, California, Benguela, and Canary Currents and (2) the fact that in the western halves of the oceans there are more species in the boundary zones between regions and provinces than in these areas themselves and less in the boundaries of the eastern halves of the oceans (Figs.

Transition Zones and their «parts», as well as Transition «areas» in the western boundary currents, are just mixed areas and have no biogeographical rank.

Pelagic communities live in a continuously moving medium and in this they differ materially from the communities living on firm substrata. Land and bottom communities can live everywhere provided the environment is favourable. This is also true of pelagic communities but, in addition, they require more or less closed gyres where independent populations of the species can live. Those gyres serve asbasesofbiotopesfor pelagic communities and as bases of ranges for individual species (Fig. 25). The part of the reproduction area outside the basis of a range is a n о n- sterile expatriation area (Beklemishev, 1961).

Movability of the biotopes of pelagic communities creates the problem of Immobilis in Mobile. The essence of this problem is whether gyres are permanent or not. The majority of the oceanic vortices are not permanent; there is only a degree of probability that they will exist at any given moment of time. Recently, R. V. Ozmidov (1965, see Fig. 8) has shown that, in the open ocean, only the vortices of three orders of magnitude can be more stable than the others. Among the more stable vortices, only those of the order of magnitude of 1,000 km can be permanent and thus only they can serve as bases of biotopes (and ranges) in the open ocean. This is why oceanic pelagic communities have the size of the large-scale oceanic circulations. Large-scale oceanic circulation can be described in two-dimensional geographical space as the deformation field which comprises the main vortices and neutral cols between them (Fig. 4). The use of this model allows one to compare pelagic biotopes and their parts to each other. Here the concept of homology of biotopes can be used (see below).

Large-scale gyres extending from coast to coast encompass primary water masses supporting primary oceanic communities.

Homologous pelagic biotopes (and habitats) are similar to each other in their position among other biotopes, in the physical processes which form them as well as in the current pattern (which must be taken into consideration in the study of the areal—rather than vertical—distribution of biotopes) . A general, serial, homology can be found between all the primary biotopes as well as between all the secondary oceanic and all the distant neritic ones. A more strict homology exists between biotopes forming pairs at both sides of the equator and between members of the same pairs in different oceans (e. g., all Central water masses are strictly homologous with each other, as are all the secondary Transition water masses).

In all the three oceans pelagic communities are built after the same pattern. The common pattern of the structure of a complex pelagic community («architectonic complex») of a whole ideal ocean can be abstracted from the structure of complex communities of individual oceans. Typically, they comprise five primary communities (plus their respective secondary ones): two temperate, two central and one equatorial, the latter living in two contacting gyres; the Indian Ocean differs from the type in having no northern half.

Entire complex communities of the whole Pacific and Atlantic Oceans look rather like huge figures-of-eight (Figs. 21, 32, 67). Their empty places are halostases of Central waters with oligotrophic communities and the peripheries of halostases are made up of more poorly balanced and/or more productive equatorial, distant neritic and transition communities. Temperate communities are even productive.

In the case of disjunct ranges there always is a problem whether or not the isolated populations of a given species live in homologous habitats. In species with little morphological variation their isolated populations proved to live in homologous habitats, while different varieties of a species with a disjunct distribution may live in non-homologous habitats (e. g., in the northern Atlantic Conchoecia obtusata is Arcto-Boreal, while in the southern Atlantic C. obtusata var. antarctica is sub-Antarctic, i. e. Transitional). The distribution of Euphausia similis is discontinuous in the Pacific Ocean in the open sea and continuous in the western distant-neritic area. In the Southern Hemisphere E. similis populates all the sub-Antarctic Transition Zone, while in the Northern Hemisphere it was reported only from the distant- neritic mixing zone between the Kuroshio and the Oyashio Currents, but never from the North Pacific Current which is homologous with the sub- Antarctic. It is not quite clear why one of these two homologous zones is populated and the other is not. But the case of E. similis forces one to consider the possibility that the biological contact between a pair of homologous habitats in both hemispheres is not necessarily followed by the colonizalion of one of them and by the appearence of a bipolar range.

Vertically, pelagic communities are not homogeneous but are made up of several depth strata.

A thin intermediate cold layer is typical of subpolar waters with a thick warm intermediate layer below; the latter overlies the deep uniform waters. In the tropics, below the upper thermocline a thick layer with gradually decreasing temperature extends to the uniform deep waters (Fig. 69).

Tropical organisms which undertake extensive diel vertical migrations can benefit from McLaren’s effect (McLaren, 1963) due to a considerable difference in the temperature of their day and night depths. This is why extensive diel migrations of DSL-animals and other macroplankters are common in the tropical pelagic communities. The most efficient migrators seem to sink as deep as the upper limit of the deep waters. The layer with the thermal gradient (the «permanent thermocline») is the habitat of these migrators. In subpolar waters which are much more isothermal than the tropical ones, diel migrations do not lead to any noticeable McLaren’s effect and, consequently, diel migrations are not typical of subpolar macroplankton.

iNet-caught plankton behaves differently. In subpolar waters it performs noticeable, if only seasonal,vertical migrations (which can reach, in the case of Calanus cristatus, as deep as the lower limit of the warm intermediate layer, ca. 3,000 m). In the tropics, only the metridiid copepods seem to migrate deep, but they do so in subpolar waters as well. Diel migrations of net- caught plankton through the 500 m depth are conspicuous in subpolar waters and negligeable in the tropics.

Net-caught plankton and macroplankton may be regarded as two major living forms of zooplankton; they are differently stratified in subpolar and tropical communities. Species of a given community having similar living forms can be grouped into ergocoens and consequently net-caught plankton and macroplankton are two ergocoens of pelagic communities. Only one of them is stratified at a time. Different ergocoens are stratified in communities of each main water structure. In tropical waters with Bruun’s thermosphere, the net-plankton ergocoen is well stratified while the macroplankton ergocoen is not. In subpolar waters without a thermosphere, the net-plankton ergocoen is less stratified than the macroplankton ergocoen.

in the tropics, the deeper strata of the communities are supplied with food from the upper layers every day due to the diel migrations of macroplankton, while in subpolar waters this occurs only once a year, i. e., during the seasonal sinking of net plankton. These differences may be relevant to the problems of utilization of production and of steady state and/or unbalanced trophic cycles.

In anomalous years (see Bjerknes, 1966) the vertical distribution of the above main pelagic ergocoens may greatly differ frow that in normal years. For instance, in 1957—1958 non-migrating species of the upper strata were over-developed at the expense of the equistratial migrating ones. In this case the composition of tropical pelagic communities underwent a change so great that it would be impossible in the cell composition of the body even of a lower Metazoan. Thus the inner information contributes less to the stability of communities than to the stability of Linnean organisms. As cempared with the latter, communities are more open to regulation.