Vavilov, a phenomenon by no means confined to cultivated plants, but often
exhibited by them to a marked degree.”This consideration complicates classifi-
cation of crop complexes, because it means that critical aspects of the genome may
be arrayed in complex ways within a group, and taxonomic recognition of this
partitioning may be a debatable issue.
In biological taxonomy,“natural classifications”(sometimes termed general
classifications) are based on overall genetic similarities and/or phylogeny, while
so-called “artificial” or “special-purpose” classifications are based on selected
similarities of particular (practical) interest to people. Artificial classification is
unrelated to the concept of artificial selection, and is a phrase, sometimes used
pejoratively, to indicate that the merit of such classifications is limited. It is often
claimed that restricting the character base to only certain economic considerations
means that the resulting classification is not based on evolution, and so not an
acceptable basis for biological taxonomy. However, characteristics of domesticated
organismsare the result of evolution, and when they are produced by strong
selective pressures they may merit special taxonomic consideration. This is
important for classifying domesticated plants, particularly forCannabis, because
biological taxonomy is, above all, intended to convey information, and for useful
plants like crops the most useful information often resides in a particular aspect of
the genome, not necessarily the entire genome. Characters or character complexes
that are selected by humans are adaptive for domesticated plants, at least in the
context of cultivation, and using such characters in recognizing taxa does constitute
evolutionary classification. The following classification of Cannabis, slightly
modified from Small and Cronquist ( 1976 ), is based on the recurrent selective
pressures (and associated gene selection) for stemfiber or THC content (between
groups of domesticated plants) and for achene retention or shattering (between wild
and cultivated plants). These principal selective evolutionary pressures onCannabis
are responsible for the generation of the most obvious and important variation
within the genus, and are accordingly appropriate bases for taxonomic delimitation.
1.14 A Practical and Natural Taxonomy forCannabis
The following four-group taxonomic subdivision ofC. sativafirst divides it into
two groups on the basis of THC and CBD content. The genetic determination of
these compounds is probably under the partial genetic control of codominant
alleles, and this may provoke the criticism that the division on the basis of pre-
dominant cannabinoid is essentially a“one-character taxonomy”(a rather pejorative
phrase in classification science). However, there are several morphological and
physiological trends that tend to distinguish plants of the hemp class and those of
the marijuana class.
As shown in Fig.1.10, divergent selection for high THC content in intoxicant
plants versus high stemfiber (and much lower THC) represents a principal
dimension of disruptive evolutionary forces that are responsible for differences in
52 E. Small