METAPHORS AND TYPOLOGY IN THE DEVELOPMENT OF BOTANICAL SYSTEMATICS 1690–1960, OR THE ART OF PUTTING NEW WINE IN OLD BOTTLES

<div class="line" id="line-5"> Aspects of the development of systematics, and particularly the systematics of higher plant groups, in the period 1690&ndash;1960 are discussed. Botanists throughout this period have considered higher taxa in natural classifications to be largely polythetic. There are no characters the possession of which are necessary and sufficient for group membership, but of the character common in a group (but neither exclusive to it or constant within it) the possession of several is needed for group membership. The practice of character weighting, developed towards the end of the eighteenth century, permitted some consistency in the use of characters and yet allowed the recognition of these polythetic groups. The largely superficial knowledge of plants led to apparent affinities being recognizable between one group and a variety of other groups, with each character indicating affinity behaving independently. Thus with the demise of the&nbsp; <i> scala naturae&nbsp; </i> in the late 1700's, the overwhelming majority of botanists adopted analogies of relationships to webs, nets, maps, or other basically horizontal, planar, reticulating structures, and use of these analogies was general by 1859. Discussion of relationships emphasized intermediates, and sometimes continuity between groups. Relations were considered to be directly between groups and were not immediately hierarchical. Typology, with types as abstractions of groups, was not developed to any great extent, largely because of the limited information provided by anatomy, development, and paleontology. In zoology, on the other hand, the large amount of data accruing from these disciplines (connected with the different nature of the organisms) permitted the strong development of typological thought. Analogies of natural relationships as vertical, tree&hyphen;like structures were developed&mdash;relationships were hierarchical and non&hyphen;reticulate, and were via the type of the group, rather than directly between the members of the groups themselves.</div><div class="line" id="line-36"> <br/></div><div class="line" id="line-26"> With the publication of &ldquo;The Origin of Species&rdquo; in 1859, systematists did not change the way they constructed classifications, but only the explanations as to what they thought these classifications represented. This set up a fundamental conflict in botany. Classifications were produced by a methodology that was developed when it was expected that relationships would be reticulate (horizontal), yet they explained the classifications in nested hierarchical (vertical) terms. Classifications were to represent phylogeny, in addition to the tasks such as information retrieval and predictivity that they had historically assumed. There was little explicit information as to how groups were to be formed. In practice the factors that influenced group membership were: the nature of the plant body, reliance on external features to provide evidence of relationship, the primacy of a prior belief in relationship, the equation of parallelism with homology, the extent of comparison of characters, and the effect of current group membership on future group membership. These factors were also largely operative in the pre&hyphen;1859 period. Apart from the development of floral formulae and diagrams, typology did not develop strongly after 1859 for the same reasons that it did not develop earlier&mdash;many plant groups were neither morphologically coherent nor unambiguously separable from coordinate groups. Thus the evidence available did not favor the adoption of abstractions, and the plasticity of the plant body allowed authors to justify very different typological schemes. The style of phylogenetic discussion still focused on concerns such as intermediate taxa, taxa as grades or levels of organization arising from other taxa of the same rank, and relationships being directly between taxa. Intermittent attempts to emphasize characters denoting lineages were generally rejected as showing too few characters to be useful.</div><div class="line" id="line-38"> <br/></div><div class="line" id="line-28"> Reactions to phylogenetic botany (and zoology as well) stressed the circularity problems in going from classification to phylogeny and the generally &ldquo;unscientific&rdquo; nature of systematics, but did not focus on the fundamental problem of horizontal methodology and vertical explanation. Attempts to make taxonomy more &ldquo;scientific&rdquo; emphasized the adoption of objective numerical methods and the avoidance of a priori weighting of characters. These attempts led to the development of the phenetic school. Classifications were to be general purpose using all characters, not special purpose (this included phylogenetic classification) using few characters. However, the uses of a general purpose classification were never really discussed in the context of mid&hyphen;twentieth century biology. Additional concerns over stability of classification and the failure to see a uniform result of the various and decidedly nonuniform speciation mechanisms then known seem to have influenced these deliberations. Attempts were also made to produce classifications that more clearly reflected phylogenetic relationships. These attempts were affected by the feeling that taxonomy should become more &ldquo;scientific,&rdquo; but otherwise they did not address the particular concerns that motivated the development of the phenetic school.</div><div class="line" id="line-40"> <br/></div><div class="line" id="line-30"> From Linnaeus onwards, the development of taxonomy can be seen as a complex interaction between groups more or less intuitively formed on the basis of rather restricted evidence and the subsequent theoretical justification of these groups. In the early 1800's, the general adoption of horizontal analogies in the discussion of the relationships of groups influenced the methodology of group formation. However, both the methodology and style of discussion of relationships persisted largely unchanged into the post&hyphen;1859 era in which a different method of group formation and a different style of discussion were required. The tensions arising in systematics from this unperceived conflict, as well as other perceived problems with phylogenetic systematics, led to two reactions: one in some sense strengthened the connection of twentieth century taxonomy with an early nineteenth century pre&hyphen;evolutionary world view, the other led to an attempt to reform taxonomy so that it became consistent with evolutionary ideas.</div><div class="line" id="line-34"> <br/></div><div class="line" id="line-32"> &ldquo;But if he kick away the Ladder, the phylogeneticist cannot fall back on the Net: a species or genus cannot be connected in all directions: the emblem of a phylogeneticist is a Tree&rdquo; (Bather, 1927, p. lxxxiv).</div>