The Plant Communities of Table Mountain: II. Life-Form Dominance and Succession

Abstract

THE general features of the plant communities of Table Mountain have been described in a previous paper (1)1, where the outlines of their structure and their relations have been traced. These communities are characterised by large and varied populations and generally by a striking absence of dominant species and of species of decided social habit. The lack of dominants is seen both in seral communities and often in the climax. While there are a few communities here which exhibit true dominance of one or a few species these are quite the exception, the majority having a mixed population. In spite, however, of the absence of dominance these communities have a definite physiognomy and can be readily distinguished from one another in the field, though the ordinary canons of description are not readily applied. Since the floristic composition of a community must ultimately be its most important characteristic (cf. 6, 5, 7) an investigation was made with a view to finding some explanation of the apparent variability of the floras. For special investigation a number of communities were selected on the northand west-facing slopes of Table Mountain at an altitude of 300-360 m. (1000-1200 ft.). Communities from other parts were investigated for comparison and the results have been tested on a large number of examples scattered over a wide area. The selected communities were on a granite slope more or less covered by sandstone talus. The area taken for detailed study was restricted at first in order to limit the number of variable factors and to allow a readier comparison of different stages in succession. A study by purely statistical methods was made but did not give results that seemed to afford any clue to the peculiarities of these communities (cf. 2, 3, 5, 11, 19) and a broader basis of study was therefore employed. The general method used was that of sample quadrats (20). A number of these were taken in a limited area of a definite community. In most cases the samples were scattered at random, but in some examples a definite plan was followed, the quadrats being arranged along straight lines. The results, however, did not show significant differences. In the majority of cases quadrats 1 m. square were used, but for stages near the climax, where large shrubs are present, the size was increased to 2 m.