Cucurbit Genetics Cooperative Report 14:53-54 (article 19) 1991
Progress in Breeding Melons for Watermelon Mosaic Resistance
Henry M. Munger
Department of Plant Breeding, Cornell University, Ithica, N.Y. 14853
When melon breeders and pathologists met in 1986 and 1987, there was general agreement that, although some watermelon mosaic resistance (WMR) in melons had been suggested by the work of Moyer et al. (2), high priority should be given to locating better sources. At that time the causal virus was usually called WMV2 but now is simply WMV. In November 1989 we began screening a few seed lots already on hand, melons for which we had either experience or reports of resistance to some virus disease. We then obtained from the Plant Introduction Station at Ames, Iowa, 50 introductions from India and 50 from Afghanistan for a greenhouse planting in March 1988.
The November test consisted of 30-40 plants per entry, but space limited the March test to 12 per entry. No entry in either test appeared to be uniformly resistant, but the introductions from Afghanistan were judged to be uniformly susceptible to WMV. No other conclusions can be made about the frequency of resistance because of disparity of numbers between the two plantings. Resistance might have been found in some discarded accessions if larger numbers had been grown. The following summary is based on the November and March tests jointly.
Although none of the germplasm screened was immunt to WMV, a small proportion of plants had the ability to grow and produce fruit, while the standard varieties did neither. The best plants were found in 4 distinct types of melons:
- 'Freeman Cucumber', C. melo conomon, described by Enzie as resistant to cucumber mosaic (1) and subsequently used and maintained at Ithica NY.
- PI 371795 from India, PI 414723 derived from it, and PI 414723-4 selected for resistance to ZYMV by Dr. R. Provvidenti. Probably C. melo momordicaI (3).
- PI 182938 from India, a small hard melon with no edible flesh, probably C. melo agrestis (3).
- C. melo dudaim, a monoecious melon collected in Louisiana by Wall (5).
With the exception of 'Freeman Cucumber' (FC) female flowers were scarce in this greenhouse planting and pollinations were necessarily made to some extent on the basis of expediency. They included selfs, crosses to susceptible melons, and intercrosses of resistant plants.
A field planting made in June 1988 included all seeds available from the greenhouse test and repeats of the more promising introductions. It was inoculated with WMV in the seedling stage. A number of entries with supposed WMR were badly stunted or died without producing fruit, probably because of natural CMV infection. This was especially evident with PI 371795 and selections from it.
At the other extreme, the outstanding rows were crosses between FC selected for WMR and either PI 4147823 or 414723-4, likewise selected for WMR. These rows were pollinated with several susceptible varieties or breeding lines. The resulting hybrids were the starting point for most of the backcross progenies on which our subsequent effort has concentrated. We thought that the superiority of these parents might be due to the complementary effect of genes for WMV from the very different parents and that it might be difficult to retain such resistance through several backcrosses. This assumption appears to have been erroneous. The partially dominant cucumber mosaic resistance of the FC parent may have been largely or wholly responsible for the superiority of the hybrids.
The FC x 414723 F1 was crossed with 'UC Topmark FR', 'TAM Uvalde', and CPM339, a Cornell monoecious PMR line with moderate cucumber mosaic resistance (CMR). Three successive backcrosses to each parent were made in two greenhouse generations and one intervening in the field, selecting each time for WMR. The resulting BC3 seed was used for the 1990 field planting. All segregating generations were inoculated at the first true leaf stage, but as Pitrat had reported (4), differences in resistance were not apparent for several weeks. Approximately 40 plants from backcrosses to each recurrent parent were transplanted to the field with only a little elimination of segregates thought to be clearly susceptible. For each recurrent parent four or five plants selected for WMV in BC2 were represented.
At flowering, about half the plants in most BC3 progenies were larger than their recurrent parents and had milder leaf mottling. These were used to make BC4. At that time the CPM329 parent row had distinctly better vines than 'Topmark' or 'TAM Uvalde' , suggesting the CMV had spread naturally in the field. It was the onlyrecurrent parent to set much fruit. Then as fruit approached half size in early September, all thge recurrent parents and part of the plants in the BC3 progenies suddenly wilted and died, the CPM339 parent as much as the others. Most of the BC3 plants selected as resistant matured fruit and seed, but the backcrosses to CPM339 had much better fruit size and quality and vines remained green longer. It was clear that there had been natural CMV infection and that melons with resistance to both it and WMV were far superior to those with resistance to either one singly.
Backcrosses to 'Honeydew' were comparable in behavior to the other two CMV susceptible backcroses but their background was different. They included two BC3 progenies inwhich the original cross was to a WMR selected plant in C. melo dudaim and three BC1 progenies in which resistance came from the same F1 used to start the other WMR backcrosses. All had the same proportion and performance of resistant plants in the field even though progenies with dudaim parentage had seemed less resistant in the greenhouse generation.
The 1990 results indicate that it may be possible to transfer a usable level of WMV rapidly and with relatively small populations and that similar resistance may be derived from different sources. They also suggest that this resistance may be more important than we previously thought. Progenies selected CMR have frequently looked resistant during most of the season but died suddenly as maturity approached. It now seems that this may not have been due to an indeqeuate CMR level as much s to natural spread of WMV.
Much further study of watermelon mosaic resistance is needed in melons. We do not know whether the resistance in our backcross progenies comes from 'Freeman Cucumber', from PI 414723, from the same gene in both, or from complementary genes in the two parents. Studies on the value of resistance from various sources as well as its inheritance are now underway.
Acknowledgments: Prof. Lin Depai from Zinjaing August 1st Agricultural College, China, was an active participant in the initial phases of this work from November 1987 to August 1988. Mr. Yiping Zhang likewise devoted much effort to the project during all of 1990. Their contributions were indispensable and are gratefully acknowledged.
- Enzie, W.D. 1943. A source of muskmelon mosaic resistance found in the Oriental Pickling Melon, Cucumis melo var. conomon. Proc. Amer. Soc. Hort. Sci. 43-195-198.
- Moyer, J.W., G.G. Kennedy, and Louise R. Romanow. 1985. resistance to watermelon mosaic virus 11 multiplication in Cucumis melo. Phytopathology 75:201-205.
- Munger, H.M. and R.W. Robinson. 1991. Nomenclature of Cucumis melo L. Cucurbit Genet. Coop. Rept. 14:43-44.
- Pitrat, M. 1978. Tolerance of melon to watermelon mosaic virus II. Cucurbit. Genet. Coop. Rpt. 1:20.
- Wall, J.R. 1967. Correlated inheritance of sex expression and fruit shape in Cucumis. Euphytica 16:199-208.