The southern cold temperate coasts, with special reference to South Africa

Abstract

My own connexion with the southern ocean has been primarily during a ten-year residence in South Africa. I have also spent a year in Australia, but this was concerned with tropical shores. During the South African period a survey was made which revealed the presence of a cold temperate area on the South African coast, lying outside the Subtropical Convergence but affected by the upwelling of a body of cold water which may be antarctic intermediate water, although this attribution has been queried. Since then similar areas have been identified in southern Australia, Tasmania and New Zealand, and it was always clear that South America must possess such an area, although only recently do we know much about the ecology of its seashores. It has been possible to compare the populations of these several areas with one another and with those of some of the southern islands, not only by means of the published accounts, but in the light of conversations and correspondence with those who have visited them, assisted by very large numbers of photographs, both in colour and in monochrome. General studies of seashores are apt to lead to unexpected conclusions, especially if anything is taken for granted beforehand. For example, such work leads one to expect that any well-stocked shore will have representatives of various animal groups on it, such as periwinkles, barnacles, whelks, crabs, anemones and limpets. But in practice one can find shores on which any one of these groups (if not more than one) is missing. This is not necessarily because of special conditions, such as those prevailing in estuaries, where reductions would seem natural, but for no reason which has (as yet) been clearly determined. Similarly, where there is a circumpolar marine climate and a West Wind Drift, one would expect circumpolar marine populations. But this is just what we do not find. While acknowledging immediately that there are circumpolar species, genera and even families, we must realize that these have attracted so much attention as to obscure the other side of the picture, which is equally important and perhaps more surprising. The South African, South Australian and South American shores all have a high proportion of endemic species, particularly marked in Australia; they are very independent. More than this, the combinations of common plants and animals which form the communities characteristic of the South African shores are very distinctive, and they are not found, so far as we know, on any other shores. The same applies to South Australia and South America: nobody familiar with any one of these would confuse them with the others. Probably the same applies to the oceanic islands, or at least to those which are widely separated. This view of the situation does not conflict with the recent work of Knox, though at first sight it may appear to do so. We know from this work that certain dominant species are recurrent round the southern ocean or part of it, and that there is a standard type of zonation. But similarities in zonation exist not only in the southern ocean but all over the world, and they demonstrate not so much a similarity between communities, as a similar reaction to environmental circumstances of any communities which exist at a particular place. Widespread dominant species show more affinity between one area and another, but it may still remain true that the combinations of common species (i. e. the communities) characteristic of any one southern cold temperate area are distinctive of that area and on the whole paralleled but not repeated in the others. The limpets of South Africa may be mentioned in this connexion. South Africa possesses a limpet population which is perhaps unrivalled in the world. There are more than 25 common species, belonging to more than nine genera, and some of these are present in countless thousands and are large in size. Now southern Australia, New Zealand and Tasmania are by no means lacking in limpets, but few if any of them are the same as the South African ones, most if not all of which are endemic to South Africa. South America again has limpets, but a different selection. These three limpet populations are therefore very independent of one another, and although a few circumpolar limpets are known, they occur in regions farther south than South Africa.