Answers are grounded in this paper and cite their sources.
Since a previous study was published, a number of refinements have been made to a group-forming system for ecological survey data using homogeneity functions (Hall 1967a). The data for this work are the records of some parameter of 'abundance' of the species occurring at a set of chosen sample sites in the area under study. The sites are clustered by a computer program and the resulting structure is shown by the linkages in a dendrogram such as that given in Fig. 2 of this paper. Several types of structure can be resolved with reasonable precision in dendrograms. A set of major groups appear as clusters subtended by relatively long stems which link at comparatively low homogeneity levels in the dendrogram. A major group may be made up of two or more sub-groups, each subtended by a comparatively short stem. Another kind of structure, the 'sibling group' (Hall 1969b), has been largely overlooked in theoretical studies in spite of its common occurrence. This consists of a relatively compact subset, lying among a 'background' of comparatively diverse members. The sibling group should be given a special level of recognition that should not be shared by its diverse allies in a classification. Finally, by inserting some extra linkages in the dendrogram, a few simple patterns of intermediateness can be shown (Hall 1968, 1969a). Interpretation in greater depth than this is needed in ecology. A feature of many situations is an extensive pattern of intergrading zones between communities. The importance of this aspect has been fully treated elsewhere (e.g. Whittaker 1962; McIntosh 1967). Such zones may form narrow belts, or be so broad as to leave few areas that are not significantly transitional in one way or another. To meet these needs, the dendrogram approach may have to be used with an auxiliary procedure to show the patterns of intergrading zones. The procedure given below allows a full interpretation of the zones in the context of geographically mapped data. The system for group-forming to give the dendrogram is described first.
No takes yet. Share an insight, caveat, or question.
A. V. Hall (1970) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: