352 WHITTAKER 



devoid of order. It is neither well ordered as it seemed in Clements' system 

 nor simply disorderly; the fundamental character of natural communities 

 with which ecologists must deal might be described as loosely ordered com- 

 plexity. In this, it is the task of the ecologist not to discover simple order 

 inherent in his material, but to find such means of effective abstraction and 

 generalization as the loosely ordered condition permits. In the author's view, 

 it is the gradual coming to terms with this loosely ordered complexity, as it 

 affected one ecological concept after another, that has characterized the 

 changing climate of interpretation in synecology. The negative aspect of this 

 change is a loss of the kind of uncritical faith in ecological concepts and 

 systems which once prevailed. Ecological concepts, man's means of inter- 

 preting natural communities, were projected back into natural communities 

 and seen as part of nature itself. They were hypostatized or reified and were 

 directly identified with the natural communities from which they were abstrac- 

 tions. Not only were plant associations part of the order of nature (Conard, 

 1939; cf. Clements, 1916, 1928; Tansley, 1920; Du Rietz, 1921, 1929; 

 Alechin, 1925); plant communities were formations and associations, vege- 

 tation was succession and the climax. More recent experiences have led to in- 

 creasing wariness of such identification, and increasing awareness of the com- 

 plexities which bedevil all ecological generalizations, the limitations affecting 

 all ecological concepts. The positive aspect of the change has been the building 

 of inductive knowledge and a more realistic understanding of the function of 

 ecological concepts. Older concepts of continuing usefulness, like the climax 

 and association, are seen in new lights, while newer concepts, like the ecosys- 

 tem and continuum, share with them in the understanding of vegetation. 



The changing view of the role of ecological concepts may be summarized 

 in a few points: 



1. All are necessarily abstractions; they are essentially human creations 

 serving to order, interrelate, and interpret some of the information about 

 natural communities available to us. None can be thought inherent in vegeta- 

 tion in the sense earlier authors assumed; none can be thought to represent 

 the real, whole, ultimate truth about natural communities. 



2. There is nothing in the nature of vegetation which compels us to adopt 

 one or another of these concepts as the primary basis of vegetation interpre- 

 tation in general — as Americans have often adopted succession and the 

 school of Braun-Blanquet associations. Nothing in the nature of vegetation 

 forces us to choose a particular way of defining a given concept. The various 

 approaches and concepts have only different degrees of general usefulness 

 and different appropriateness to particular circumstances and purposes. 



3. There is in all these concepts a dependence on choice and assumption, 

 an element of subjectivity and artistry, which can never be wholly escaped, 

 even though it is an objective of science to minimize its influence. The best 

 means of controlling subjectivity is through a recognition, as frank and clear- 



