Segregation of the Determinate (de) Allele in Crosses Between Cucumis sativus L. and C. sativus var. hardwickii (R.) Alef.

Cucurbit Genetics Cooperative Report 8:2-3 (Article 1) 1985

D.E. Delaney and R.L. Lower
University of Wisconsin, Madison, WI 53706

In 1982, we initiated a breeding program for increasing the number of lateral branches on determinate plant types. The objective was to alter the plant architecture of current determinate lines by incorporating the multiple lateral branching character from Cucumis sativus var. hardwickii. The new plant type would capitalize on both the shorter length of determinate genotypes and the larger photosynthetic area and sequential fruit-setting ability of var. hardwickii. Previous unpublished data had suggested that the frequency of determinate individuals in determinate C. sativus x var. hardwickii crosses was lower than expected based on single gene recessive inheritance. Thus, another objective of the study was to further investigate this phenomenon.

A C. sativus var. hardwickii line (LJ 90430) was crossed onto four determinate C. sativus lines (‘Spacemaster’, NCSU M21, NCSU M27, and USDA 1909) to create four populations. Seed of the four populations was planted in 1983 at the Hancock Experimental Farm, Hancock, WI, on 1.5m centers. Individual plants in the F₂ and BC₁ generations to the determinate parents (BC₁ P₁) were classified visually into two classes, determinate (dede) and indeterminate (De-), based on vine and internode length, termination of apical growth, and leaf color (determinate genotypes generally have a darker green color). Termination of growth was determined by tagging the apex of the main stem at 10 day intervals. If the stem had not lengthened in 10 days, it was considered determinate.

Two additional populations were evaluated in 1984. In both populations the determinate C. sativus parent was NCSU M11de, which was crossed to two different indeterminate lines, LJ 90430, and a near-isogenic line, NCSU M11. Plants in the F₂ and BC₁P₁ generations were classified by plant type, as before, except the main stems were not tagged.

In all four populations in the 1983 study, the frequency of determinate plant types in the F₂ was lower than the expected 3:1 ratio (Table 1). The chi square values for goodness of fit were all significant at least at the .10 level. In contrast, segregation ratios in the BC₁P₁ were not significantly different from the expected 1:1 ratio, except in the ‘Spacemaster’ x LJ 90430 population. Some of the discrepancy in the F₂ ratios could be due to misclassification. However, since fruit set was not controlled, it is likely that misclassification could have occurred in either direction (i.e. classifying a determinate individual as indeterminate or an indeterminate plant as determinate). The presence of early fruit set is known to reduce node number and deter growth in C. sativus (1,3). While other investigators have expressed difficulty in classifying determinate and indeterminate segregates (1,2), none have reported segregation ratios deviating from that expected with single gene inheritance. These investigators were not utilizing var. hardwickii germplasm, however.

Table 1. Chi square (X²) tests for plant type segregations in the F₂ and BC₁P₁ generations of crosses between five determinate C. sativus lines and C. sativus var. hardwickii (LJ 90430), and a C. sativus x C. sativus cross (NCSU M11de x NCSU M11).

		Generation
			F2				BC1P1
Population	Genotype	Expected	Observed	X2	P	Expected	Observed	X2	P
1983
Spacemaster x LJ 90430	De-	178.5	200	10.36	.001	57	69	4.28	.02-.05
	dede	59.5	38			57	45
NCSU M21 x x LJ 90430	De-	178.5	191	3.50	.05-.10	57.5	51	1.47	.30-.40
	dede	59.5	47			57.5	64
NCSU M27 x x LJ 90430	De-	180	197	6.10	.01-.02	58.5	51	1.92	.10-.20
	dede	60	43			58.5	66
USDA 1909 x x LJ 90430	De-	177	188	2.73	.10	54.5	63	2.65	.10-.20
	dede	59	48			54.5	46
1984
NCSU M11de x LJ 90430	De-	155	181	17.08	<.001	40	45	1.25	.25-.50
	dede	52	26			40	35
NCSU M11de x NCSU M11	De-	156	148	1.64	.20	42	37	1.19	.25-.50
	dede	52	60			42	47

The comparison of two additional populations in 1984, revealed that the F₂ segregations failed to fit a 3:1 ratio only when var. hardwickii was used as the indeterminate parent. The NCSU M11de x NCSU M11 population gave an acceptable fit to a 3:1 ratio, but the NCSU M11de x LJ 90430 population again had fewer determinate plant types than expected. However, as in 1983, both populations segregated 1:1, indeterminate:determinate, in the BC₁P₁.

In crosses between C. sativus and var. hardwickii most of the determinate segregates in the F₂ and BC₁P₁ resembled the respective determinate parent for lateral number. The determinate plant types that seem to be missing are those with high numbers of lateral branches. It seems likely that the plant growth regulators may be involved. More investigations are continuing.

Literature Cited

1. Denna, D. W. 1971. Expression of determinate habit in cucumbers (Cucumis sativus L.). J. Amer. Soc. Hort. Sci. 96:277-279.
2. Kauffman, C. S. 1973. An investigation of the inheritance of determinate habit and short internode dwarf cucumbers, Cucumis sativus L. M.S. Thesis. North Carolina State University, Raleigh, N.C.
3. McCollum, J. P. 1934. Vegetative and reproductive responses associated with fruit development in cucumber. Cornell Agr. Exp. Sta. Memoir 163.