Jr. Caradus et Df. Chapman, SELECTION FOR AND HERITABILITY OF STOLON CHARACTERISTICS IN 2 CULTIVARS OF WHITE CLOVER, Crop science, 36(4), 1996, pp. 900-904
White clover (Trifolium repens L.) herbage production in intensively g
razed pasture is related to clover stolen growing point density. High
stolen density is associated with small leaf size. The objective of th
is study was to examine the probability of breaking the genetically co
ntrolled link between leaf size and stolen density by manipulating int
ernode length and branching propensity. Two white clover cultivars, Gr
asslands Tahora and Grasslands Kopu, were grown in two New Zealand env
ironments, full sunlight or 50% sunlight (shade), during 1991 and 1992
. Selections were made for high or low frequency of stolen branching,
long or short stolen internodes, and large or small leaves. These sele
ctions were polycrossed within groups to give 24 lines. Lines were eva
luated in pots grown outside to determine the effect of selection on t
hese and other shoot characters. Selection of populations with high or
low frequency of branching gave progeny with different frequency of b
ranching for all cultivar-environment combinations except for selectio
n within Tahora grown in shade. Selection of populations with long or
short internode length gave progeny with different internode length in
all cultivar-environment combinations. Selection for large and small
leaflet width gave progeny with different leaflet width in all cultiva
r-environment combinations except for selection within Kopu grown in s
un. Two estimates were made of narrow sense heritabilities. Estimates
from mid-parent versus progeny regression heritabilities were higher f
or internode length and leaflet width than for frequency of branching.
Realized heritabilities were again generally higher for internode len
gth and leaflet width than for frequency of branching although there w
ere some exceptions. Selection for high frequency of branching in Taho
ra grown in full sunlight had a high realized heritability. Stolen cha
racters, such as internode length and frequency of branching, can be e
ffectively manipulated by selection. The degree of selection success c
an be affected by environment and cultivar. Selection for variation in
internode length is likely to be more successful than for frequency o
f branching.