S U C C E S S I V E A P P R O X I M A T I O N I N D E S C R I P T I V E ECOLOGY
53
His next step is t o arrange these communities into classes or sequences
according t o changes in their composition or ecology. This can be done
by tabulation of the data followed by inspection (see for example
Nordhagen, 1928 and 1943 ; Braun-Blanquet, 1951 ; Poore, 1955a,b,c
and 1956; Dahl, 1956 and Ellenberg, 1956);t or by calculating various
quantitative measures such as coefficients of community (Jaccard,
1902 ; Kulczynski, 1928 ; and Srarensen, 1948) or the “climax adaptation
number” and “continuum index” of Curtis and McIntosh (1951).
These methods lead either t o discrete abstract units (sociations,
associations etc.) or t o continuous series of communities which can be
arranged along various axes of variation. Which of these two results is
obtained appears t o depend partly on the type of vegetation examined
and partly on the preconceptions of the worker about the nature of
vegetation.
The question, whether vegetation is a continuum or can be divided
into discrete units, has been debated ever since Gleason formulated his
“individualistic hypothesis” and seems no nearer now than then t o a
clear unequivocal answer. Workers who have long familiarity with
vegetation and who have examined the question critically can come to
opposite conclusions.
For example, Dahl writes: “The repeated occurrence in nature of
vegetational discontinuities is, in the writer’s opinion, the basis which
makes a real classification of vegetation possible. If all vegetational
change were gradual, each species reacting independently of the external factors there would be no basis of classification only of typification ... . The occurrence of more or less well marked vegetational discontinuities is, in the writer’s opinion, an indication that we have t o
deal with different communities which have to be analysed separately.
If the difference is not only in the frequency of one species but can be
characterized by more criteria there is a basis for referring the vegetation on either side of the discontinuity t o different sociological units.” A t
the opposite extreme Curtis (1959) maintains : “Communities . . . are not
precise entities of fixed and unvaiying composition, but rather are loose
aggregations of species, whose make-up changes from place t o place and
from time t o time in a more or less continuous fashion. The communities,
therefore, and the entire vegetation which they compose cannot be described in the exact language of physical science but must be treated in
a statistical manner as a continuous variable.” Both these views gain
support from the data their authors bring forward. What are the
reasons for this divergence?
The results which are obtained in the final tabulation or ordination
t For a detailed description of the construction and manipulation of tables see Ellenberg (1958).
53
His next step is t o arrange these communities into classes or sequences
according t o changes in their composition or ecology. This can be done
by tabulation of the data followed by inspection (see for example
Nordhagen, 1928 and 1943 ; Braun-Blanquet, 1951 ; Poore, 1955a,b,c
and 1956; Dahl, 1956 and Ellenberg, 1956);t or by calculating various
quantitative measures such as coefficients of community (Jaccard,
1902 ; Kulczynski, 1928 ; and Srarensen, 1948) or the “climax adaptation
number” and “continuum index” of Curtis and McIntosh (1951).
These methods lead either t o discrete abstract units (sociations,
associations etc.) or t o continuous series of communities which can be
arranged along various axes of variation. Which of these two results is
obtained appears t o depend partly on the type of vegetation examined
and partly on the preconceptions of the worker about the nature of
vegetation.
The question, whether vegetation is a continuum or can be divided
into discrete units, has been debated ever since Gleason formulated his
“individualistic hypothesis” and seems no nearer now than then t o a
clear unequivocal answer. Workers who have long familiarity with
vegetation and who have examined the question critically can come to
opposite conclusions.
For example, Dahl writes: “The repeated occurrence in nature of
vegetational discontinuities is, in the writer’s opinion, the basis which
makes a real classification of vegetation possible. If all vegetational
change were gradual, each species reacting independently of the external factors there would be no basis of classification only of typification ... . The occurrence of more or less well marked vegetational discontinuities is, in the writer’s opinion, an indication that we have t o
deal with different communities which have to be analysed separately.
If the difference is not only in the frequency of one species but can be
characterized by more criteria there is a basis for referring the vegetation on either side of the discontinuity t o different sociological units.” A t
the opposite extreme Curtis (1959) maintains : “Communities . . . are not
precise entities of fixed and unvaiying composition, but rather are loose
aggregations of species, whose make-up changes from place t o place and
from time t o time in a more or less continuous fashion. The communities,
therefore, and the entire vegetation which they compose cannot be described in the exact language of physical science but must be treated in
a statistical manner as a continuous variable.” Both these views gain
support from the data their authors bring forward. What are the
reasons for this divergence?
The results which are obtained in the final tabulation or ordination
t For a detailed description of the construction and manipulation of tables see Ellenberg (1958).