Cucurbit Genetics Cooperative Report 3:20-21 (article 11) 1980
Heterosis Estimates for Several Characteristics in a Cross
Between a Gynoecious Inbred of Cucumis Sativus
L. and C. hardwickii R.
James Nienhuis and R. L. Lower
University of Wisconsin,
Madison, WI 53706; R. R. Horton, North Carolina State University,
Raleigh, NC 27650
Characteristics such as multiple laterals and prolific
fruiting make Cucumis hardwickii a potentially
useful source of germplasm in a breeding program to improve
yield of pickling cucumbers (1). The use of C. hardwickii
also represents an opportunity to broaden the genetic
base of common cucumber cultivars (3). Information about
the nature of differences in fruit setting and other morphological
characteristics between Cucumis sativus and C.
hardwickii would be very beneficial to plant improvement
programs. The object of this study was to estimate heterosis
for yield components and several vegetative characteristics
in a cross between a gynoecious pickling cucumber inbred
line, Gy 14, and a C. hardwickii line, 'LJ 90430'.
The two parental lines, Gy 14 and 'LJ 90430', (designated
as P1 and P2 respectively), and their F1 were utilized in
this study. The parents had been maintained by selfing for
several generations and were assumed to be homozygous. The
three generations were grown in a randomized complete block
design with ten blocks at the Horticulture Crops Research
Station in Clinton, NC, in 1978. A replication consisted
of five plant single row plots for each generation; the
plants were spaced on 1.5 m centers.
Analysis of variance was used to test mean differences
among generations. Generation by block interaction was used
as an estimate of experimental error. Heterosis above the
midparent was defined as the superiority of the F1 over
the respective high parent for the trait under consideration.
A t-test was used to determine the significance of differences
between generations. Because of the short day nature of
the C. hardwickii line, flowering was reduced and
data on fruiting parameters were estimated from an unweighted
least squares analysis of generation means (40). The estimate
of the fruiting parameters obtained were gathered on individual
plants at maturity as follows: number, weight and length/diameter
ratios of mature fruit, lateral number and main stem vine
length. Lateral number was counted as the number of lateral
branches (primary, secondary, etc.) between the cotyledonary
node and the first 1m of main stem.
Heterosis above both mid and high parent was observed for
fruit weight per plant and main stem vine length (Table
1). Heterosis below the midpoint was observed for lateral
number (Table 1). The length/diameter ration and number
of fruit showed no deviation of the F1 from the midparent.
When estimates of heterosis fall above the high parent,
it is tempting to conclude that this is overdominant type
gene action; however, Moll and Stuber (2) point out that
other non-additive types of gene action and linkage disequilibrium
might result in effects which mimic those of overdominance.
From a plant breeder's point of view, heterosis estimates
indicate whether homozygotes or heterozygotes represent
the more ideal genotype. With no heterozygote advantage,
breeding efforts should be directed towards homozygous inbred
populations; alternatively, if there is a heterozygote advantage,
then breeding methods such as reciprocal recurrent selection,
which capitalize on heterosis, should be explored.
Table 1. Generation means and heterosis estimates for five characteristics in a C. sativus x C. hardwickii cross.z
| Generations |
Fruit no. per plant |
Fruit weight per plant(kg) |
Lateral no. |
Main stem vine length |
Length/diameter ratio
of fruit |
P1 Gy 14C. sativus |
4.03 |
2.21 |
4.10 |
121.31 |
2.24 |
P2 'LJ 90430', C. hardwickii |
93.91 x |
2.20 x |
39.81 |
137.15 |
1.47 x |
F1 |
53.12 |
6.70 |
16.50 |
290.32 |
1.99 |
Heterosis above midparentz |
4.15 NS |
4.49 ** |
-5.46* |
161.09** |
0.14 NS |
Heterosis above high parenty |
- |
4.49** |
- |
153.17** |
- |
z F1 - (P1+P2)/2.
y F1 - high parent.
x Least square estimate from generation means, which included F2, BC1, and BC2.
**Significant at .01 level ; *significant at .05 level.
Literature Cited
- Horst, E. K. and R. L. Lower. 1977. Cucumis hardwickii:
A Source of Germplasm for the Cucumber Breeder. Cucurbit
Genetics Coop. Rpt. 1:5.
- Moll, R. H. and C.W. Stuber. 1974. Adv. Agr.
26:277-315.
- National Academy of Sciences. 1972. Genetic Vulnerability
of Major Crop Plants.
- Nienhuis, J. and R. L. Lower. Unpublished data.
