PLANT COMMUNITY STRUCTURE, AND ITS RELATION TO THE VERTICAL COMPLEXITY OF COMMUNITIES - DOMINANCE DIVERSITY AND SPATIAL RANK CONSISTENCY

Plant assembly rules were sought in terms of abundance. Twelve plant communities were sampled, including a range of vertical complexity: crustose lichens, cushion, turf and tall grassland. Community structure was analysed by examining fits to various dominance/diversity models, and by calculating spatial rank consistency. There were significant differences between the communities, as to which dominance/diversity model fitted best. The General Lognormal model fitted best for five communities, the Zipf-Mandelbrot for five others, and the Geometric for the remaining two. However, there was no apparent correlation between the model fitting best and previously published suggestions as to where each model should fit. There was no correlation with the level of vertical complexity. No other patterns could be seen, even a posteriori. Preston's parameter 'gamma' varied between 0.94 and 8.26, but no ecological correlate was apparent for the variation. Spatial rank consistency (i.e. whether community replicates are consistent in the identity of the species occupying each rank) is suggested as a new tool for investigating community structure at any single point in time. It was investigated with an index, C(r) Most communities showed moderate to high C(r) values, but one community showed no rank consistency at all. There was variation in C(r), but again no ecological correlate was obvious, The plant communities examined differ in various indices of community structure. However, the results do not confirm previous, untested suggestions on plant community structure. No new overall pattern emerges. This calls into question whether the models and parameters are measuring the appropriate aspects of plant community structure. If they are, then the question becomes whether the abundance structure really is limited by assembly rules.