Serviços Personalizados
Journal
Artigo
Indicadores
- Citado por SciELO
- Acessos
Links relacionados
- Similares em SciELO
Compartilhar
Revista Chapingo. Serie horticultura
versão On-line ISSN 2007-4034versão impressa ISSN 1027-152X
Resumo
LOPEZ-SANTIAGO, J. et al. Morphological variable selection for hawthorn characterization (Crataegus spp.). Rev. Chapingo Ser.Hortic [online]. 2008, vol.14, n.2, pp.97-111. ISSN 2007-4034.
Considering plant characterization as the set of characteristics for taxonomical differentiation, in the present research the selection of reliable and discriminating morphological variables was established as objective, using the analysis of variance, coefficient of variation, lineal correlation, stepwise selection analysis and mean component analysis. Hawthorn (Crataegus spp.) leaf, flower and fruit set characteristics from the Chapingo Autonomous University gene bank collection were selected. The statistical methods that contributed more in the selection of these morphologic characteristics and descriptors definition were: the coefficient of variation, the mean components analysis and their analysis of correlation, in addition to the researcher criterion, obtaining 35 characters that contributed more in the morphologic characterization, of a total of 76 variables. The discriminating variables were: from reproducrive shoots leaves: perimeter, major and minor axis lenght, elongation index, feret diameter, compaction index, basal angle and veins number; from short vegetative shoots leaves: perimeter, major and minor axis lenght, elongation index, feret diameter, compaction index, apical and basal angle, veins number; from long vegetative shoots leaves: perimeter, major and minor axis lenght, elongation index, feret diameter, compaction index, apical and basal angle; from reproductive leaves stomata: stomata frequency and stomata lenght; and from flowers: pedicel length, floral receptacle lenght, perimeter and feret petals diameter, style lenght, perimeter, major axis lenght, and feret anther filaments diameter.
Palavras-chave : Rosaceae; plant genetic resources; mean component analysis; multivariate analysis.