IDENTIFICATION OF IDEOTYPES BY CANONICAL ANALYSIS IN Panicum maximum

Ciência e Agrotecnologia

Janaina Azevedo Martuscello¹  Thiago Gomes Dos Santos Braz¹  Liana Jank¹  Daniel De Noronha Figueiredo Vieira Da Cunha¹  Ana Luiza Silva Carvalho¹ 

关键词: Multivariate analysis;    breeding value;    agronomic performance;    leaf dry matter;    Análise multivariada;    valor genotípico;    desempenho agronômico;    massa seca de folhas;   

DOI  :  10.1590/S1413-70542015000200006

来源: SciELO

【 摘 要 】

Grouping of genotypes by canonical variable analysis is an important tool in breeding. It allows the grouping of individuals with similar characteristics that are associated with superior agronomic performance and may indicate the ideal profile of a plant for the region. The objective of the present study was to define, by canonical analysis, the agronomic profile of Panicum maximum plants adapted to the Agreste region. The experiment was conducted in a completely randomized design with 28 treatments, 22 genotypes of Panicum maximum, and cultivars Mombasa, Tanzania, Massai, Milenio, BRS Zuri, and BRS Tamani in triplicate in 4-m² plots. Plots were harvested five times and the following traits were evaluated: plant height; total, leaf, and stem; dead dry matter yields; leaf:stem ratio; leaf percentage; and volumetric density of forage. The analysis of canonical variables was performed based on the phenotypic means of the evaluated traits and on the residual variance and covariance matrix. Genotype PM34 showed higher mean leaf dry matter yield under the conditions of the Agreste of Alagoas (on average 53% higher than cultivars Mombasa, Tanzania, Milenio and Massai). It was possible to summarize the variation observed in eight agronomic characteristics in only two canonical variables accounting for 81.44 % of the data variation. The ideotype plant adapted to the conditions of the Agreste should be tall and present high leaf yield, leaf percentage, and leaf:stem ratio, and intermediate values ​​of volumetric density of forage.

【 授权许可】

CC BY   
 All the contents of this journal, except where otherwise noted, is licensed under a Creative Commons Attribution License