Premium
Predicting Forage Yield from Morphological Traits in Reed Canarygrass 1
Author(s) -
Casler M. D.,
Hovin A. W.
Publication year - 1985
Publication title -
crop science
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.76
H-Index - 147
eISSN - 1435-0653
pISSN - 0011-183X
DOI - 10.2135/cropsci1985.0011183x0025000500014x
Subject(s) - forage , biology , tiller (botany) , agronomy , phalaris arundinacea , fodder , yield (engineering) , mathematics , ecology , wetland , materials science , metallurgy
Yield components and morphological traits have been investigated as a means of more efficiently selecting for improved economic yield in many crop species. The objectives of this research were to develop equations for predicting forage yield from morphological traits of reed canarygrass ( Phalaris arundinacea L.) and to determine the potential usefulness of these equations as selection indices in forage yield improvement. Ten parental clones and their topcross progenies were evaluated for 2 yr at four Minnesota locations for forage yield and two locations for morphological traits. Forage yield was predicted by multiple regression analysis from among the following traits of spaced plants and simulated solid‐seeded plots (microplots): leaf blade width, leaf blade thickness, specific leaf weight, leaf rigidity, culm diameter, tiller dry weight, tiller number or density, plant height, spaced‐plant area, and spaced‐plant circumference. Predictive equations were generated by screening models for high coefficient of determination (R 2 ), low average relative influence (diagonal elements of the hat matrix), and low residuals. Of the 16 predictive equations developed (based on data for three harvests and means over harvests for spaced plants and microplots at two locations in 1979) and evaluated, only four had R 2 ≥ 0.68. Of these, the equation developed from means over harvests on Rosemount spaced plants (RS‐M) had uniformly the best predictive value across both locations and years, the lowest average relative influence, and the lowest average residuals. Expected genetic gain for seasonal forage yield based on predicted yield from the RS‐M equation, which included leaf rigidity and tiller number, was 4.3% per cycle for 10% selection intensity and phenotypic selection compared with 4.6% per cycle for yield per se. Increasing the selection intensity to 5% would increase the expected response to 5.0% per cycle. We concluded that this equation appears to have application for yield selection in spaced plant populations of this reed canarygrass germplasm.