Multitrait selection system using populations with a small number of interploid (4x-2x) hybrid seedlings in potato: Degree of high-parent heterosis for yield and frequency of clones combining quantitative agronomic traits
Ja. Buso et al., Multitrait selection system using populations with a small number of interploid (4x-2x) hybrid seedlings in potato: Degree of high-parent heterosis for yield and frequency of clones combining quantitative agronomic traits, THEOR A GEN, 99(1-2), 1999, pp. 81-91
One of the most attractive features of the 4x-2x breeding scheme via unilat
eral sexual polyploidization (USP) is the possibility of obtaining heteroti
c clones with a combination of desirable traits based upon selection in a s
mall number of hybrid seedlings. A set of experiments was carried out to ve
rify this characteristic of the USP scheme using 42 families (with 20 plant
s each) derived from 2x Phureja-haploid Tuberosum or haploid Tuberosum-Sola
num chacoense hybrids as male parents. These clones were 2n-pollan-producer
s by either first-division restitution with crossing over (FDR-CO) or witho
ut crossing over (FDR-NCO). The 4x parents were eight cultivars of USA and
European origin. A total of 168 out 840 clones was initially selected (four
clones/family) based on general tuber appearance (GTA). An additional roun
d of selection (not taking into account the parentage of the clone) for tot
al tuber yield (TTY) reduced the sample to 96 clones. These selected clones
and the original 4x parents were evaluated in two locations at Wisconsin (
Hancock - E # 1, and Rhinelander - E # 2). The average high-parent heterosi
s values for TTY ranged from 27.8% (E # 2) to 48% (E # 1). The population o
f experimental clones also had a TTY range wider than that of the 4x parent
s. The best clone gave a yield of 101.6% and 63.2% over the best 4x parent
at E # 1 and E # 2, respectively. For TTY, a significant genotype x environ
ment (G x E) interaction for the experimental clones was found when a combi
ned analysis of variance for both locations was carried out. However, the G
x E interaction was not significant when only the 4x-parent group was cons
idered. Clones derived from 4x European cultivars had higher GTA scores tha
n clones derived from 4x USA cultivars, With two culling levels being set o
n the TTY 2 4x parent group mean and the GTA 2 three, about 56% (E # 1) and
48% (E # 2) of the clones would be retained for further evaluation. These
percentages of selected clones are much higher than those reported using co
nventional 4x-4x crosses. Our results indicate that a USP strategy with 4x-
2x (FDR) crosses would be more effective than intra-Tuberosum crosses in ge
nerating heterotic clones with a combination of desirable quantitative trai
ts using populations with a small number of hybrid seedlings.