56
M. El. D. P O O R E
histories ; but continuous gradations or ecotones also occur. Similarly,
in the field of abstract units, discontinuities may be detectable due t o
the absence of intermediate habitats; or communities may fall in continuous series. Perhaps we might use two models. In both of them
individual stands of vegetation are represented by points in multidimensional space, each dimension representing a line of variation in
vegetation. If we consider vegetation as essentially continuous the
points will be distributed at random throughout the matrix, and 110
grouping but an arbitrary grouping will be possible. On the other hand,
if the variation of vegetation is not continuous the points will be found
t o be grouped in nebulae.
If the second alternative is true it should be possible t o extract and
define these nebulae and t o use them as points or areas of reference in
terms of which “intermediate” communities may be defined. I have
applied the term “nodum” t o each of these nebulae (Poore, 1955b).
If the first alternative, that there is no grouping of points, is a
situation which is frequently encountered, the problem of classification
must be tackled somewhat differently. There are three possibilities.
(i) Certain arbitrarily chosen stands should be used as types; (ii) stands
should be defined with reference t o their position by a series of coordinates; (iii) the matrix should be divided arbitrarily into a series of
blocks defined by the sum of the characters of the stands within the
blocks.
Any results may then be referred t o a named vegetation unit or
defined area of study -they are not left floating in a multidimensional
void.
Nevertheless a frame of reference must not be allowed t o become a
prison cell. Concepts that have led to great advances in science have
sometimes by their very success, become sacrosanct. They have in their
later stages discouraged a wider vision and have actually impeded progress. Certain definitions of the association and the climax, for example, though leading t o immediate advances in understanding have
later left their exponents in mental blinkers and have restricted rather
than assisted their work. The notions of “nodum” and of the potential
continuous variation of vegetation are flexible concepts which should
not operate restrictively.
In previous papers (Poore, 1955a,b,c, 1956) I have distinguished
noda, because it seemed clear that, in the areas examined, the communities could be grouped into abstract units without distortion of the
facts. But there would be value in the concepts of “type” or “nodum”
even if it were t o be established that the variation of vegetation was
continuous ; although the noda would then lose all their absolute validity
and become arbitrary groupings.
M. El. D. P O O R E
histories ; but continuous gradations or ecotones also occur. Similarly,
in the field of abstract units, discontinuities may be detectable due t o
the absence of intermediate habitats; or communities may fall in continuous series. Perhaps we might use two models. In both of them
individual stands of vegetation are represented by points in multidimensional space, each dimension representing a line of variation in
vegetation. If we consider vegetation as essentially continuous the
points will be distributed at random throughout the matrix, and 110
grouping but an arbitrary grouping will be possible. On the other hand,
if the variation of vegetation is not continuous the points will be found
t o be grouped in nebulae.
If the second alternative is true it should be possible t o extract and
define these nebulae and t o use them as points or areas of reference in
terms of which “intermediate” communities may be defined. I have
applied the term “nodum” t o each of these nebulae (Poore, 1955b).
If the first alternative, that there is no grouping of points, is a
situation which is frequently encountered, the problem of classification
must be tackled somewhat differently. There are three possibilities.
(i) Certain arbitrarily chosen stands should be used as types; (ii) stands
should be defined with reference t o their position by a series of coordinates; (iii) the matrix should be divided arbitrarily into a series of
blocks defined by the sum of the characters of the stands within the
blocks.
Any results may then be referred t o a named vegetation unit or
defined area of study -they are not left floating in a multidimensional
void.
Nevertheless a frame of reference must not be allowed t o become a
prison cell. Concepts that have led to great advances in science have
sometimes by their very success, become sacrosanct. They have in their
later stages discouraged a wider vision and have actually impeded progress. Certain definitions of the association and the climax, for example, though leading t o immediate advances in understanding have
later left their exponents in mental blinkers and have restricted rather
than assisted their work. The notions of “nodum” and of the potential
continuous variation of vegetation are flexible concepts which should
not operate restrictively.
In previous papers (Poore, 1955a,b,c, 1956) I have distinguished
noda, because it seemed clear that, in the areas examined, the communities could be grouped into abstract units without distortion of the
facts. But there would be value in the concepts of “type” or “nodum”
even if it were t o be established that the variation of vegetation was
continuous ; although the noda would then lose all their absolute validity
and become arbitrary groupings.