Nomenclature in the dock. Overclassification leads to instability: a case study in the horticulturally important genus Cyclamen (Myrsinaceae)

Nine different classifications have been produced in the last 70 years for the horticulturally valuable genus Cyclamen, a small genus with fewer than 30 species. These classifications, generated by intuitive methods and cladistic analyses, incorporated a total of four infrageneric ranks above that of species and were based on data from morphology, cytology and DNA sequencing. Our results, based on cladistic analyses of three independent data sources − nrDNA ITS, cpDNA trnL intron and morphological data − reveal good resolution only in nrDNA sequence data. However, when these three data sources are combined they provide stronger resolution and support for three major clades, only one of which, subgenus Psilanthum, has been consistently supported in previous classifications. The differing infrageneric classifications produced in Cyclamen result from varying taxon sampling, differing interpretation of morphological data, changes in the sources and analysis of data, and inconsistent application of names. Extensive subdivision of small genera in the absence of adequate data that could provide evidence for consistent patterns of relationship is premature and leads to a proliferation of names.© 2004 The Linnean Society of London, Botanical Journal of the Linnean Society, 2004, 146, 339-349.

Artificial keys for botanical identification using a multilayer perceptron neural network (MLP)

In this paper, practical generation of identification keys for biological taxa using a multilayer perceptron neural network is described. Unlike conventional expert systems, this method does not require an expert for key generation, but is merely based on recordings of observed character states. Like a human taxonomist, its judgement is based on experience, and it is therefore capable of generalized identification of taxa. An initial study involving identification of three species of Iris with greater than 90% confidence is presented here. In addition, the horticulturally significant genus Lithops (Aizoaceae/Mesembryanthemaceae), popular with enthusiasts of succulent plants, is used as a more practical example, because of the difficulty of generation of a conventional key to species, and the existence of a relatively recent monograph. It is demonstrated that such an Artificial Neural Network Key (ANNKEY) can identify more than half (52.9%) of the species in this genus, after training with representative data, even though data for one character is completely missing.

Interactive effect of cowpea variety, dose and exposure time on bruchid tolerance to botanical pesticides

Callosobruchus maculatus has for years remained a serious menace in cowpea in Sub-Sahara Africa. The objective of this study was to investigate the effect of genotypic cowpea (Vigna unguiculata (L.) Walp) varieties, time and dose on C. maculatus exposed to powders of Piper guineense and Eugenia aromatica. Irrespective of duration and botanicals, bruchid reared on KDV showed the highest tolerance to both plant materials; while their counterparts from IAR48V were the most susceptible. Median lethal time (LT50) also varied according to the plant materials; with the highest in KDV reared bruchid [P. guineense: KDV (18.31), IAR48V (9.27), IFBV (13.17); E. aromatica: KDV (76.01), IAR48V (5.59), IFBV (6.49)]. There was a significant impact of cowpea variety (V), exposure time (T) and dose (D) on the tolerance of C. maculatus to both plant materials. The effect of all two-way (VxT, VxD, DxT) and three way interactions (V×T×D) on the tolerance of C. maculatus to both plant materials was also significant. Varietal effect was more pronounced in bruchids exposed to E. aromatica; while exposure time was more pronounced in bruchids exposed to P. guineense.