Premium
Relative Effectiveness of Hill and Row Plots for Evaluating Alfalfa Yield
Author(s) -
Trimble M. W.,
Barnes D. K.,
Lamb J. F. S.
Publication year - 1994
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/cropsci1994.0011183x003400060020x
Subject(s) - forage , biology , yield (engineering) , agronomy , randomized block design , mathematics , germplasm , metallurgy , materials science
Plant breeders usually use multiple‐row or broadcast plots grown from 2 to 4 yr to evaluate alfalfa ( Medicago sativa L.) populations for yield and agronomic performance. The requirements of land and seed and the cost of evaluation often limit the number of populations that can be evaluated. The objective of this study was to compare the yield of alfalfa populations grown in hill plots planted on either 0.3‐ or 1.0‐m centers and 1‐row (0.5 by 6.1 m) and 6‐row (1.1 by 6.1 m) plots. The study was conducted for 2 yr in Minnesota with 16 alfalfa populations ranging in dormancy, yielding ability, and biological N fixation. The experimental design was a randomized complete block with four replicates at three locations. The strongest association for forage yield between any two plot types was between the 0.3‐m hill plots and the 1‐row plots. Phenotypic and Spearman rank correlations for forage yield between 0.3‐m hill plots and the 1‐row plots in 1984 were 0.73 and 0.67, respectively. No significant correlation was observed in yield between hill plots and 6‐row plots. Yield in 6‐row plots was positively correlated to that in 1‐row plots. A few populations were consistent in rank for forage yield in both years and across all plot types. Forage yield in some populations ranked low in hill plots and high in 6‐row plots, whereas other populations ranked high in hill plots and low in 6‐row plots, illustrating that a population's performance in an evaluation trial could be influenced by the plot type used. Visual estimates of forage yield and actual harvested yield were highly correlated in the hill plots for all three harvests during 1983. These correlations decreased as stands aged, suggesting that use of visual estimates of forage yield would be useful only during the establishment year. Hill plots may be useful in early phases of a breeding program but should not be considered as a replacement for multiplerow yield evaluation plots during the later stages of a breeding program.