88
J. M. LAMBERT A N D M. B. DALE
latter course to be both more realistic and more generally useful; we
shall therefore concern ourselves with the application of phytosociological principles in general to situations where the vegetation is comparatively unknown, but shall not assume a necessary relationship
with pre-existing schemes of classification.
In any scientific investigation, certain assumptions have to be made
at one point or another before the work can proceed; the initial assumption$ may be very general ones -for instance, the assumption that there
is some pattern to be found in the material under study - but without
such assumptions the work would not even begin. Moreover, unless the
investigation is to be completely aimless, the purpose of the study has
to be clearly defined and kept in mind throughout; this constantly
calls for decisions by the investigator as to what information is useful
for this purpose and what can be regarded as irrelevant. The actual
course of the investigation therefore depends on both assumptions and
decisions, both of which are essentially subjective ; however, decisions
are largely based upon assumptions and, unIess the assumptions are
recognized and constantly examined for their validity, the answer
expected by the investigator may be inadvertently built into the results.
It is probably true to say that, in any investigation, the fewer the
assumptions the wider the applicability of the results, irrespective of
whether the results themselves have been obtained by subjective
assessment or by statistical means. We shall therefore attempt to
examine some of the basic assumptions underlying much phytosociological work in an effort to eliminate those which are unnecessary;
and this will involve us in a comparison of concepts as to the nature of
vegetation as material for study. Our key references for this section are
the works of Poore (1955a,b,c; 1956; 1962) and of Whittaker (1962),
whose extensive researches into phytosociological literature provide an
admirable basis for such a comparison.
The first concept to be examined is that of vegetation itself. This
seems to be a fluctuating concept among workers in that some would
restrict it solely to plant material, while others would allow the concept
to include an element of the natural habitat. We have ourselves defined
the term as a threefold plantfsitelhabitat system, which allows us to
include both plant/site and site/habitat relationships and yet dispense
with any assumption that plants and environment are causally related
and should be assessed together. We may later wish to mamine plant/
habitat relationships for ecological purposes ; but this is a secondary
step and does not affect the strictly phytosociological study of plant/
site relationships. Although the plantfsite and sitefhabitat distinction
may seem unnecessarily precise to ecologists concerned essentially
with vegetation in its widest sense, we have nevertheless tried to show
J. M. LAMBERT A N D M. B. DALE
latter course to be both more realistic and more generally useful; we
shall therefore concern ourselves with the application of phytosociological principles in general to situations where the vegetation is comparatively unknown, but shall not assume a necessary relationship
with pre-existing schemes of classification.
In any scientific investigation, certain assumptions have to be made
at one point or another before the work can proceed; the initial assumption$ may be very general ones -for instance, the assumption that there
is some pattern to be found in the material under study - but without
such assumptions the work would not even begin. Moreover, unless the
investigation is to be completely aimless, the purpose of the study has
to be clearly defined and kept in mind throughout; this constantly
calls for decisions by the investigator as to what information is useful
for this purpose and what can be regarded as irrelevant. The actual
course of the investigation therefore depends on both assumptions and
decisions, both of which are essentially subjective ; however, decisions
are largely based upon assumptions and, unIess the assumptions are
recognized and constantly examined for their validity, the answer
expected by the investigator may be inadvertently built into the results.
It is probably true to say that, in any investigation, the fewer the
assumptions the wider the applicability of the results, irrespective of
whether the results themselves have been obtained by subjective
assessment or by statistical means. We shall therefore attempt to
examine some of the basic assumptions underlying much phytosociological work in an effort to eliminate those which are unnecessary;
and this will involve us in a comparison of concepts as to the nature of
vegetation as material for study. Our key references for this section are
the works of Poore (1955a,b,c; 1956; 1962) and of Whittaker (1962),
whose extensive researches into phytosociological literature provide an
admirable basis for such a comparison.
The first concept to be examined is that of vegetation itself. This
seems to be a fluctuating concept among workers in that some would
restrict it solely to plant material, while others would allow the concept
to include an element of the natural habitat. We have ourselves defined
the term as a threefold plantfsitelhabitat system, which allows us to
include both plant/site and site/habitat relationships and yet dispense
with any assumption that plants and environment are causally related
and should be assessed together. We may later wish to mamine plant/
habitat relationships for ecological purposes ; but this is a secondary
step and does not affect the strictly phytosociological study of plant/
site relationships. Although the plantfsite and sitefhabitat distinction
may seem unnecessarily precise to ecologists concerned essentially
with vegetation in its widest sense, we have nevertheless tried to show