WATERMELON (Citrullus lanatus) BREEDING HANDBOOK

Gabriele Gusmini

Raleigh NC, 2003

Abstract

The watermelon (Citrullus lanatus) is one of the most important horticultural crop for fruit

production in the U.S. The watermelon is a monoecious species and therefore a

cross–pollinated crop of the family Cucurbitaceae and does not manifest important levels of

inbreeding depression or heterosis.

2

Public and private breeders are developing leading varieties since the 1,800's (Wehner et al.,

2001; Whitaker and Jagger, 1937) and lot of progress has been done particularly for

qualitative traits. However, there are not very detailed informations available about the

genetics of this plant even though several genes have been identified. Priorities for the future

should be the understanding of the mechanisms of inheritance and the variance components

for several important traits, along with the development of new breeding strategies that can

hold the gain obtained in selection for qualitative traits broadening at the same time genetic

variability in the cultivated watermelons to allow the breeders to improve the quantitative

traits of this crop.

Biotechnology has not played a central role yet for the improvement or the genetical

understanding of the watermelon but should be better used as a tool to short-cut the tipically

long timing of every breeding program (gene sequencing, gene overexpression and/or

silencing, transformation techniques, molecular markers assisted selection, etc.).

This report summarizes some important informations on the genetics and the breeding of the

watermelons and presents some new possibilities for its improvement.

3

Introduction

The watermelon (Citrullus lanatus) is a major vegetable crop in the U.S. with an average

production in the period 1999-2001 of about 39 million cwt/year. The major center of

production was Florida (in 1999-2001 about 8.9 million cwt/year), even though the state with

the major acreage in the same triennium was Texas (about 37,000 acres/year) (U.S.D.A.,

2002).

The watermelon has high lycopene content in the red-fleshed varieties: 60% more than

tomatoes. Lycopene has been classified as a useful component of human diet for prevention

of heart-attacks and certain types of cancer (Perkins-Veazie et al., 2001).

In the U.S. the market is more oriented now toward smaller watermelons, either round or

elongate, preferentially seedless. The watermelon can have flesh of different colors, even

though red is the most common. The market recently has welcome yellow and orange

varieties, particularly in processed watermelons. In other countries , i.e. China, small and

yellow watermelons are preferred to the red ones; in Europe, however, the comsumers still

want red and big fruits.

The american market is nowadays divided equally between seeded and seedless types but the

request for seedless on the market either for fresh and processed products is increasing.

4

Taxonomy (G.R.I.N.)

Kingdom: Plantae

Division: Magnoliophyta

Class: Magnoliopsida

Order: Cucurbitales

Family: Cucurbitaceae

Subfamily: Cucurbitoideae

Tribe: Benincaseae

Subtribe: Benincasinae

Genus: Citrullus

Species: lanatus

Varieties: lanatus

citroides

Table 1. Related species of agricultural interest and their capability of intercrossing withwatermelon Species Chromosome Level of Capability of

Number (2n) Relationship Intercrossing Citrullus lanatus 22 - -

Cucurbita pepo 40 Family NoCucurbita maxima 40 Family NoCucurbita mixta 40 Family NoCucurbita moschata 40 Family NoCucumis sativus 14 Family NoCucumis melo 24 Family NoLuffa spp. 26 Family No

5

Table 2. Intercrossablewild related species of watermelon (Mohr, 1986) Species Chromosome Life Sex

Number (2n) Cycle Type Citrullus lanatus 22 Annual Monoecius

Citrullus colocynthis 22 Perennial MonoeciusCitrullus naudinianus 22 Perennial DioeciousCitrullus ecirrhosus 22 Perennial MonoeciusCitrullus rehmii 22 Perennial Monoecius

The species listed in table 2 are all intercrossable and seeds have been recovered and

successfully germinated in all possible crosses (Mohr, 1986). So far only Citrullus

colocynthis has received attention as a source of valuable germplasm and has been included

genetic studies (Levi et al., 2001a).

6

Biogeography

The watermelon is thought to be native from Africa (Maynard et al., 2001; Mohr, 1986;

Whitaker and Jagger, 1937) but the distribution of the PI accessions present in the U.S.D.A.

collection after the recent updates of the past 10 years shows that it is possible to identify at

list other three important centers of germplasm diversity: South-Asia, Turkie, and Middle-

East.

Table 3. Distribution of hypothetic centers of diversity for watermelon based on the origin ofthe PI Accessions present in the U.S.D.A. germplasm collection at Griffin, GA

Until now the germplasm has been collected mainly in Africa, China, Middle-East, South-

Europe, and slightly in America. The major problem of the collection of watermelon

germplasm is the political instability of the area where it is widely present in the wild:

political reasons, for instance, limited exploration of most countries in Africa and Asia

7

(Vietnam, Laos, and Cambogia); protectionistic politics adopted from the middle-eastern

countries limited also the collection of germplasm in that area. For sure future collection

should concentrate in the asian countries since now they are politically stable and much more

open toward international cohoperation and exchange of germplasm. Africa is becoming

even more dangerous to explore and the most stable countries are generally not willing to

allow export of their germplasm.

Breeders can find germplasm directly in the U.S. by request at the U.S.D.A. repository in

Griffin, GA.

Table 4. Watermelon germplasm held at the U.S.D.A. repository in Griffin, GA Species Variety PI Accessions Citrullus lanatus lanatus 1,435

citroides 134Citrullus colocynthis 23Citrullus naudinianus 0Citrullus ecirrhosus 1Citrullus rehmii 1

In addition to the U.S.D.A. collection of PI Accessions, several cultivars have been released

by public and private breeders and are accessible on the market (Maynard, 2000; Wehner,

2002).

8

Table 5. Watermelon cultivars present in North America (Wehner, 2002)

Africa 8 - Breeder: Rev. Rush Wagner, Old Umtali Mission, South Africa. Parentage:selection from accession sent by Wagner to Layton. Characteristics: round fruit, flaccidsweet red flesh, early maturity. Resistance: anthracnose. 1937.

Angeleno - Parentage: possibly from Chilean watermelons or Chilean Black Seededcultivar.

Black Kleckley - Breeder: Iowa. Parentage: Jugoslavia 7, Iowa Bell, Kleckley Sweet.Characteristics: darker rind than original Kleckley Sweet. 1945.

Black Seeded Klondike No.3 - Breeder: Dept. Vegetable Crops, University of California,Davis. 1933.

Brownlee - Breeder: Mr. Walker, Florida. Parentage: Leesburg x Hawkesbury.Characteristics: thin rind, good internal quality, sweet crisp red flesh with hollow heart orwhite heart. Similar: Kleckley Sweet. 1941.

Buist Little Gem - Breeder and vendor: Robert Buist Co. Characteristics: 75 days toharvest; earliest cultivar; fruit nearly round, small in size, 12 to 17 lb.; skin dark greenstripes on gray green background; flesh pinkish red, very firm, sweet, delicious eating.

Buttercup (4505) - Breeder: Dr. Warren Barham. Vendor: D. Palmer Seed Co.Characteristics: tiger stripe, bright yellow flesh. Adaptation: most watermelon growingareas around the world. 1999.

Calhoun Gray - Breeder: North Louisiana Experiment Station, Calhoun, LA. Parentage:Calhoun Sweet x Charleston Gray. Resistance: anthracnose, fusarium wilt races 0 and 1.1965.

California Klondike - Parentage: selection from Klondike. Similar: Klondike. 1933.

Calsweet - Breeder: Mr. Layton, California. Parentage: (Miles x Peacock) x CharlestonGray, the same parents as Allsweet, Crimson Sweet. Characteristics: 21 x 18 inch size,20-30 lb., wide indistinct stripes, bright red flesh. Resistance: fusarium wilt.

Carolina Bradford - Vendor: Robert Buist Co. Characteristics: 85 days to harvest; a fineshipping cultivar having a very tough, elastic rind; fruit large, 25 lb., oblong, deep greenirregularly striped with a darker shading; flesh dark red, fine grained and sweet; seedscreamy white with some slightly mottled.

Carolina Cross #183 - Breeder and vendor: Burpee Seed. Characteristics: gigantic fruit, 200lb. size, oblong fruit shape, light green rind with narrow dark green stripes, 100 days toharvest.

9

Charleston Gray No.133 - Breeder: Mr. Stevenson, India. Parentage: Charleston Grayselected for higher wilt resistance. Characteristics: thinner rind; otherwise similar toCharleston Gray. Resistance: fusarium wilt. 1961.

Colebrook - Breeder: E.M. Meader. Parentage: open pollinated selection from Koreancultivar Shingyamato. Characteristics: round, 10 lb. fruit, thick rind, striped pattern,bright red flesh, small brown seeds. Similar: Merrimack Sweetheart, Yankee Queen,Sweet Sugar. 1948.

Cole's Early (Harris' Earliest) - Breeder: Coles Seed Store, Pella Iowa. Vendor: RobertBuist Co. Characteristics: 80 days to harvest; extra early; a favorite in the North; forhome market; fruit small size, 15 lb, short, nearly round, dark green stripes irregular on alight green background: rind tender; flesh light red, sweet and delicious; seeds black.1892.

Conqueror - Breeder: Orton. Parentage: Eden x citron. Characteristics: insufficient fruitquality for commercial production; used to develop wilt resistant cultivars withacceptable fruit quality. Resistance: fusarium wilt. 1908.

Cooperstown (XP 4590339) - Breeder: David Dean. Vendor: Seminis-Asgrow. Parentage:F1 hybrid. Characteristics: seedless with mottle stripe on green background, oval fruitshape, medium to large size. Resistance: anthracnose, fusarium wilt. Similar: Tri-X-313.Adaptation: North America. 2000.

Corporal (PS 56595) - Breeder: Manual Rosa. Vendor: Seminis Vegetable Seeds-Petoseed.Parentage: F1 hybrid. Characteristics: seeded, all sweet type, late, wide strips, large fruit.Similar: Sangria. Adaptation: U.S., Mexico. 1999.

Cuban Queen - Breeder and vendor: W. Atlee Burpee. 1881.

Dark Icing - Vendor: D.M. Ferry. 1888.

Delta (PS 36594) - Breeder: Jose Gomez. Vendor: Seminis Vegetable Seeds-Petoseed.Parentage: F1 hybrid. Characteristics: Royal Sweet type, crimson Sweet stripes, oblong,18-25 lb. strong plant, vigorous, wide adaptability. Similar: Royal Sweet. Adaptation:United States. 1998.

Dixie Belle (Stone Mountain) - Breeder: H.G. Hastings, Atlanta, Georgia. Characteristics:thick, tough, dark green rind, 17 x 14 inches, 35-40 lb., red, sweet flesh. 1924.

Dixie Queen (White-Seeded Cuban Queen) - Breeder: Johnson. Vendor: Stokes Seed Co.Parentage: Cuban Queen x Kolb Gem. Characteristics: round fruit 30-40 lb., pink flesh,white seeds, narrow dark green stripes over light green rind; 90 days to harvest; popularwith home gardeners, roadside markets, truckers and shippers; fruit oblong or nearlyround; rind thin, but tough; flesh crisp and extremely sweet with very few seeds; seedswhite and small. 1890.

10

Dulce Verde (4506) - Breeder: Dr. Warren Barham. Vendor: D. Palmer Seed Co.Parentage: F1 hybrid. Characteristics: solid dark green rind. Similar: Peacock.Adaptation: most watermelon growing areas around the world. 1999.

Early Kansas - Vendor: Robert Buist Co. Characteristics: 80 days to harvest; a fine shipper;fruit nearly round, 40 lb., light green with wavy stripes; flesh bright red sweet and tender;seeds reddish brown.

Early Resistant Queen - Breeder: Iowa. Parentage: Early Market Queen x Hawkesbury.1945.

Excel - Breeder: D.H. Gilbert of Monticello, Florida. Characteristics: poorly fixed hybrid.1906.

Falcon (RS 56395) - Breeder: Manual Rosa. Vendor: Seminis Vegetable Seeds-Petoseed.Parentage: F1 hybrid. Characteristics: seeded, Allsweet shaped watermelon, earlyplanting, wide stripes, large fruit, high yields, no hollow heart or bottlenecks.Adaptation: United States and Mexico. 1999.

Fandango (SSC 46072) - Breeder: New Avenue Seeds. Vendor: Shamrock Seed Co.Parentage: F1 hybrid. Similar: Summer Sweet 5244. Adaptation: all watermelongrowing regions. 1999.

Far North - Breeder and vendor: Fischer Seed Co. Parentage: Sugar Bush x Peacock.Characteristics: dark green fruit, 6-8 lb., bright red flesh, black seeds; many off-types.

Fenway (XP 6279) - Breeder and vendor: Seminis Vegetable Seeds. Parentage: F1 triploidhybrid. Characteristics: seedless Sugar Baby type, high yield, round to slightly oblongfruit shape, dark green rind with very dark green faint mottled stripe, rind has shownexcellent resistance to sunburn. Similar: Sugar Baby (except seedless). Adaptation:southeastern U.S. 2001.

Florida Favorite - Vendor: Robert Buist Co. Parentage: Rattlesnake and Pearson.Characteristics: 80 days to harvest; excellent for the home garden or home market trade,not for long distance shipping; fruit large, 25 lb., long with round ends; light green withmottled stripes of dark green; rind fairly tough; flesh dark red, crisp and sweet; seedswhite.

Georgia (152185) - Breeder and vendor: Seminis-Petoseed. Parentage: F1 hybrid.Characteristics: small Crimson Sweet type, strong vine, early to medium early maturity;fruit about 10-13 lb. with small broken dark green stripes. Similar: Crimson Sweet.Adaptation: Guatemala, South America. 2000.

11

Georgia Rattlesnake - Characteristics: 90 days to harvest; an excellent shipping melonpopular for its size, productiveness and eating qualities; fruit very long, large, 25 to 30lb., light green, irregularly mottled with dark green stripes; striking appearance; rind verytough; flesh bright scarlet, crisp and sweet; seeds all white with black tips. Adaptation:southern U.S. 1870.

Golden Anniversary - Parentage: Wikara x Kleckley. Characteristics: sweet.

Gray Monarch (Long Light Icing) - Vendor: Robert Buist Co. Characteristics: 85 days toharvest; a large melon mostly adapted for home gardens; fruit 25 lb. cylindrical withblunt ends, a beautiful light gray green faintly mottled darker; rind thin and tender; fleshred, sweet and delicious; seeds white.

Green Seeded Citron - Vendor: Robert Buist Co. Characteristics: 95 days to harvest; bredexclusively for preserving; fruit round, 10 lb., striped alternately with dark and lightgreen. Flesh clear white and very solid. Seeds glossy olive green.

Halbert Honey - Vendor: Robert Buist Co. Characteristics: 85 days to harvest; a fine, largeprolific melon; sweet flavor; for home use and nearby markets; fruit 30 to 35 lb.,cylindrical with blunt ends, dark glossy green with fine veins; rind tender, flesh rich red,extending clear to the rind; seeds white with black tips.

Harris' Earliest (Cole's Early) - Vendor: Robert Buist Co. Characteristics: 80 days toharvest; extra early; a favorite in the North; for home market; fruit small size, 15 lb,short, nearly round, dark green stripes irregular on a light green background: rind tender;flesh light red, sweet and delicious; seeds black.

Hawk (131784) - Breeder: Manuel Rosa. Vendor: Seminis Vegetable Seeds-Petoseed.Parentage: F1 hybrid. Characteristics: Allsweet type, seeded, wide stripes, for earlyplanting. Similar: Sangria. Adaptation: U.S. & Mexico. 1999.

Hawkesbury (synonyms Hawkesbury Wilt Resistant, Dark Seeded Gray Monarch) -Breeder: H.S. Shirlow, N.S.W. Australia. Parentage: selection from a field of GrayMonarch with dark rather than light colored seed. Resistance: fusarium wilt. Similar:Gray Monarch. 1935.

Hungarian Honey - Parentage: introduced from Hungary. 1885.

Improved Stone Mountain No.5 - Breeder: Mr. Younkin, Iowa. Parentage: Iowa Bell xStone Mountain. Resistance: some fusarium wilt tolerance. Similar: Stone Mountain.

Iowa 112 - Breeder: Mr. Younkin, Iowa. Parentage: Iowa Bell x Jugoslavia 7. 1938.

Iowa Belle - Breeder: Porter, Iowa. Parentage: Conqueror x unknown male parent (probablyKleckley Sweet). 1932.

12

Iowa Dixie - Breeder: Mr. Wilson, Iowa. Parentage: Iowa Belle x Japan 7. 1934.

Iowa King - Breeder: Porter, Iowa. Parentage: Conqueror x unknown male parent (probablyKleckley Sweet). 1932.

Irish Gray - Vendor: Robert Buist Co. Characteristics: 90 days to harvest; one of the finestwatermelons; productive, very sweet and tender; earlier than Tom Watson, but similar inshipping quality; fruit large, 25 to 30 lb. oblong, smooth, a distinct mottled greenish gray:rind tough and hard; flesh bright red, firm and sweet; seeds white.

Japan 7 - Parentage: received from Prof. Akewo Hemmi, Japan. Characteristics: whiteflesh, yellow rind, small oval fruit, early maturity. Resistance: fusarium wilt.

Jubilee II - Breeder: Crall, Leesburg, Florida. Parentage: Jubilee x Jubilee 5.Characteristics: fruit and plant resembles Jubilee but selected for higher resistance to wilt.Resistance: fusarium wilt. Similar: Jubilee. 1990.

Kafir - Breeder: selected from African stock. Resistance: fusarium wilt, anthracnose. 1932.

Kengarden - Breeder: H. Mohr, University of Kentucky. Parentage: Dwarf Asahi Yamato x(dwarf Bush Desert King x breeding lines) x New Hampshire Midget. 1975.

Kleckley Sweet (Monte Cristo) - Breeder: W.A. Kleckley, Alabama. Vendor: Robert BuistCo. Parentage: possibly Boss x Arkansas Traveler. Characteristics: thin, glossy, darkgreen rind, large white seeds, medium-red flesh, oblong fruit, 25-40 lb., 85 days toharvest; always popular and a favorite with all; excellent for home use and nearbymarkets; rind thin and brittle; flesh juicy and very sweet. 1887.

Klondike - Parentage: a mutation selected in California. Characteristics: blocky fruit, rindsolid green with stripes, 20-25 lb., fine textured flesh. 1908.

Klondike R7 - Breeder: Mr. Porter, California. Parentage: Iowa Belle x Klondike.Characteristics: fruit 15 x 10 inch size, 20-25 lb.; thin, dark-green rind; sweet red flesh.Resistance: fusarium wilt. 1937.

Klondike Striped - Parentage: selection by grower in California from field of Klondike.Characteristics: oblong fruit with thin, tough, light green rind with narrow dark greenstripes, 20-25 lb., fine textured flesh. 1923.

Klondike Striped Blue Ribbon - Breeder: Mr. Porter, California. Parentage: Klondike R7 xKlondike Striped. Characteristics: thin tough rind, light green with narrow dark stripes,15 x 10 inch size, 20-30 lb., sweet dark-red flesh. Resistance: fusarium wilt, anthracnose,sunburn. 1939.

13

Klondike WR - Breeder: Mr. Layton, California. Parentage: California Klondike 3 xKlondike R7. Characteristics: like Klondike, but sweeter. Resistance: fusarium wilt(better than Klondike). 1957.

Klondike WR 60 - Breeder: Mr. Layton, California. Parentage: California Klondike 3 xKlondike R7. 1958.

Kolb Gem - Breeder: R.F. Kolb, Alabama. Vendor: D.M. Ferry. 1885.

Long Light Icing (Gray Monarch) - Vendor: Robert Buist Co. Characteristics: 85 days toharvest; a large melon mostly adapted for home gardens; fruit 25 lb. cylindrical withblunt ends, a beautiful light gray green faintly mottled darker; rind thin and tender; fleshred, sweet and delicious; seeds white.

Long Mountain - Parentage: selection from Stone Mountain. 1936.

Matador (13786) - Breeder: Manuel Rosa. Vendor: Seminis Vegetable Seeds-Petoseed.Parentage: F1 hybrid. Characteristics: seeded, late season, large fruit, wide stripes,vigorous vines, excellent yields. Similar: Sangria. Adaptation: U.S. and Mexico. 1999.

Mercedes (SSC 9490) - Breeder: Hollar Seed Co. Vendor: Shamrock Seed Co. Parentage:F1 hybrid. Resistance: some races of anthracnose and fusarium wilt. Similar: Sangria.Adaptation: All watermelon production regions. 2000.

Millennium (HMS7928) - Breeder: Bob Schroeder. Vendor: Harris Moran Seed Co.Parentage: F1 hybrid. Characteristics: very dark rind pattern, dark red flesh. Resistance:fusarium wilt race 1, anthracnose. Similar: Tri-X-Shadow. Adaptation: California,Southeast US, Midwest, Atlantic Shore. 1999.

Monte Cristo (Kleckley Sweet) - Breeder: W.A. Kleckley, Alabama. Vendor: Robert BuistCo. Parentage: possibly Boss x Arkansas Traveler. Characteristics: thin, glossy, darkgreen rind, large white seeds, medium-red flesh, oblong fruit, 25-40 lb., 85 days toharvest; always popular and a favorite with all; excellent for home use and nearbymarkets; rind thin and brittle; flesh juicy and very sweet. 1887.

Montreal (SXW 5023) - Breeder: Gary Elmstrom. Vendor: Sunseeds. Parentage: F1hybrid. Characteristics: seeded diploid watermelon; oblong fruit with wide dark greenstripes on a light green background; high sugar content, good flesh color. Resistance:fusarium wilt race 1. Similar: Allsweet. Adaptation: California, Texas, Florida. 2000.

Mountain Sweet - Vendor: Robert Buist Co. Characteristics: 90 days to harvest; for homeand local markets; fruit large, 25 lb.; oblong, dark green; with sweet juicy, light crimsonflesh; seeds brown.

14

Northern Sweet - Breeder: Minnesota Agr. Exp. Sta. Parentage: selection from Russiancultivar. Characteristics: early, round fruit, 10-12 lb. size, medium green rind with widestripes, red-orange flesh. 1932.

Omega (PS 3981428) - Breeder and vendor: Seminis Vegetable Seeds. Parentage: F1triploid hybrid. Characteristics: seedless hybrid; typical Allsweet stripe (dark green stripeon light green background), 18-20 lb. size fruit, very bright red flesh, mid-seasonmaturity. Adaptation: North America. 2001.

Orton - Breeder: Florida. Parentage: Iowa Belle x unknown male parent. 1950.

Peacock - Breeder: G. Stratis, Brawley, CA. Parentage: selection from Klondike.Characteristics: oval fruit, 20-25 lb., dark green rind, sweet red flesh. 1939.

Peacock Special Shipper - Breeder: R. Peacock. Parentage: chance cross of Klondike.Characteristics: smooth, dark green rind, crisp, red flesh, 20-25 lb. 1935.

Peacock WR 124 - Breeder: Mr. Layton, California. Parentage: California Klondike 3 xPeacock x Klondike R7. 1959.

Peacock WR 50 - Breeder: Mr. Layton, California. Parentage: Klondike R7 x Peacock.Characteristics: tough thin rind, dark green with faint dark stripes, 15 x 10 inch size, 20-25 lb., sweet bright red flesh. Resistance: tolerant to fusarium wilt. 1955.

Premiere (SSC460057) - Breeder: Colorado Seeds. Vendor: Shamrock Seed Co. Parentage:F1 hybrid. Characteristics: 82-85 day maturity; round to slightly oblong fruit; 12-24 lb.average weight; sweet, red, firm flesh; medium-dark green stripes and light greenbackground; produces hardy vines, high yields. Similar: Tri-X-313. 1999.

Riviera (015 208) - Breeder and vendor: Seminis Vegetable Seeds-Asgrow. Parentage: F1hybrid. Characteristics: Crimson Sweet type, more round shape, quite early, fruit weightapproximately 5-7 kg. Resistance: fusarium wilt race 1, anthracnose. Similar: CrimsonSweet. Adaptation: Caribbean. 1997.

Samba (SSC 31782) - Breeder and vendor: Shamrock Seed Co., Inc. Parentage: F1 triploidhybrid. Characteristics: a very productive seedless cultivar with impressive fruituniformity; develops very strong vines; high yields of oval-round, large sized fruit with adark striping pattern; appealing firm, red flesh with high sugars; thrives under a widerange of growing conditions; 75-85 days to maturity. Similar: Tri-X-313. Adaptation: allwatermelon growing regions. 2001.

Santa Amalia (35422) - Breeder: Manuel Rosas. Vendor: Seminis Vegetable Seeds-Petoseed. Parentage: F1 hybrid. Characteristics: Royal Sweet type with Crimson Sweetstripes, oblong, 20-27 lb., strong plant vigor. Similar: no cultivar like it. Adaptation:South America. 1999.

15

Schochler - Vendor: Robert Buist Co. Characteristics: 90 days to harvest; a very largemelon used extensively in the South, having outstanding shipping requisites for size,tough rind and quality; fruit very long, 40 to 50 lb., rich dark green with faint stripe; fleshbright crimson, fine grained and very sugary.

Sentinel (PS 36694) - Breeder: Jose Gomez. Vendor: Seminis Vegetable Seeds-Petoseed.Parentage: F1 hybrid. Characteristics: Royal Sweet type, Crimson Sweet stripe, oblong,20-27 lb., strong plant vigor, excellent red flesh color. Similar: Royal Sweet.Adaptation: U.S. and Australia. 1998.

Small Seeded Dixielee - Breeder: Crall et al., Leesburg, Florida. Parentage: Fairfax,Summit, Charleston Gray, Texas W5, Crimson Sweet, W.R. Graybelle. Characteristics:round-oval fruit, light green rind with narrow dark stripes, 10-15 lb., fine textured flesh,dark, red flesh, few small black seeds, high sugar content. 1995.

Smile - Breeder and vendor: American Takii. Parentage: hybrid. Characteristics: iceboxtype, with resistance to fruit cracking and tough rind for shipping; fruit have crisptextured, sweet, dark red flesh, black seeds; fruit are round with narrow dark stripe onlight background; 6 to 8 lb fruit; 38 to 40 day maturity; fruit keep 12 days after harvest.Adaptation: home garden and commercial fresh market. 2000.

Stone Mountain (Dixie Belle) - Breeder: H.G. Hastings, Atlanta, Georgia. Characteristics:thick, tough, dark green rind, 17 x 14 inches, 35-40 lb., red, sweet flesh. 1924.

Summit - Breeder: Mr. Taylor, Louisiana. Parentage: Calhoun Sweet x Black Diamond.Characteristics: large round fruit with tough, dark green-black rind, flesh red, blackmottled seeds. Resistance: fusarium wilt (high level of resistance). 1957.

Sunsplash (25453) - Breeder: Manuel Rosas. Vendor: Seminis-Petoseed. Parentage: F1hybrid. Characteristics: seedless yellow-flesh round with thick rind, exterior color ,narrow dark-green stripes on light green background, medium-large sizes. Similar:Honeyheart. Adaptation: U.S. Southeast Asia. 2000.

Sweetheart - Breeder: Mr. Wittenmeyer, southern Indiana. Vendor: D.M. Ferry. 1890.

The Dixie - Vendor: Robert Buist Co. Characteristics: 85 days to harvest; a splendidshipper; prolific, with fruit large, 25 lb., oblong, very dark green with lighter greenstripes; rind thin but strong; flesh scarlet and sweet; seeds black.

Thurmond Gray - Breeder: Mr. Thurmond, Perry, Georgia. Vendor: Robert Buist Co.Characteristics: large elongated fruit with light green mottled rind, large seeds, pink flesh;90 days to harvest; shipping cultivar similar to Irish Gray, but having fruit much larger;fruit 30 to 40 lb., cylindrical, grayish-green with faint veining; rock hard rind; flesh crispand sweet. 1923.

16

Tom Watson - Breeder: Alexander Seed Co. Vendor: Robert Buist Co. Characteristics:elongate shaped fruit, 25-40 lb., thin tough rind, dark green with dark veining; fleshsweet, firm, crisp, coarse, red; large brown seeds spotted with white; shipping melon.1906.

Tribute (PX 5696) - Breeder: Manuel Rosas. Vendor: Seminis Vegetable Seeds-Petoseed.Parentage: F1 hybrid. Characteristics: seedless, round fruit, striped with medium to largefruit. Similar Tri-X-313. Adaptation: U.S. and Mexico. 1999.

Trident (PX 58596) - Breeder: Manuel Rosas. Vendor: Seminis Vegetable Seeds-Petoseed.Parentage: F1 hybrid. Characteristics: seedless round fruit, striped, medium to large fruit,deep red flesh and high sugars, early. Similar: Tri-X-313. Adaptation: U.S. and Mexico.1999.

Triton (PS 37794) - Breeder: Jose Gomez. Vendor: Seminis Vegetable Seeds-Petoseed.Parentage: F1 hybrid. Characteristics: seedless, round, striped, medium-large, yellowflesh. Similar: Honey Heart. Adaptation: U.S. and Asia. 1998.

White Seeded - Breeder: Mr. Porter, Iowa. Parentage: from African stock. Resistance:fusarium wilt, anthracnose. 1932.

White-Seeded Cuban Queen (Dixie Queen) - Breeder: Johnson. Vendor: Stokes Seed Co.Parentage: Cuban Queen x Kolb Gem. Characteristics: round fruit 30-40 lb., pink flesh,white seeds, narrow dark green stripes over light green rind; 90 days to harvest; popularwith home gardeners, roadside markets, truckers and shippers; fruit oblong or nearlyround; rind thin, but tough; flesh crisp and extremely sweet with very few seeds; seedswhite and small. 1890.

Wills Sugar - Vendor: O.H. Will Seed Co., Bismarck, North Dakota. 1889.

Wrigley (XP 4590249) - Breeder and vendor: Seminis-Asgrow. Parentage:F1 hybrid.Characteristics: seedless, oval blocky shape with excellent flesh color, maturity 80-90days, average fruit size 12-16 lb. Resistance: anthracnose, fusarium wilt. Similar: A andC 5244. Adaptation: U.S. wholesale market. 2000.

Yellow Dragon (PX 33994) - Breeder: Jose Gomez. Vendor: Seminis Vegetable Seeds-Petoseed. Parentage: F1 hybrid. Characteristics: yellow seeded watermelon, mediumblocky fruit, good taste, crisp flesh. Similar: Yellow Doll. Adaptation: wholesale,Pacific rim. 1997.

17

Botany

The watermelon is a cross-pollinated crop even though self-pollination and sibbing are

common. The watermelon is generally monoecious and rarely andromonoecious (sex-type

very common in C. lanatus var. citroides) (Mohr, 1986); wild relatives can be dioecious

(table 2). Andromonoecy is recessive to monoecy and the ratio of staminate to pistillate or

perfect flowers in the commercial varieties usually is 4:1 or 7:1 (Maynard et al., 2001).

Among the adapted varieties Sugarbaby, Mickylee, and Crimson Sweet have a high

female/male flowers ratio; Allsweet and Jubilee, instead, have a low female/male flowers

ratio.

Male flower Female flower

Male-sterility is useful in hybrid production. Two different male-sterile mutants are widely

known, 'glabrous male sterile' (gms) and chinese male sterile (cms) and the latter is preferred

because gms is less vigorous and have poor seeds set. Parthenocarpy is inducible with

18

growth regulator but does not occur naturally and no mutant carrying any gene for

parthenocarpy has been found yet (Maynard et al., 2001).

Apomixis has not been reported so far and asexual propagation has been suggested

particularly for the production of seedles watermelon but it is too expensive compared to

seed production.

Pollination in nature is done by bees even though the flowers of watermelon are not the most

attractive: weed control and wise choices of other cultivated species flowering at the same

time as watermelon in the surrounding areas are very important aspects to be considered.

Controlled pollinations can be done in the early morning in days with good levels of solar

radiation (Mohr, 1986) (no cloudy days and no Winter pollinations in area North of the

Carolina's) by hand. Female flowers must be isolated the evening before placing on them

foam cups or nets to avoid bees visiting them at random. It is a good practice to protect also

the male flowers to avoid contaminations of the pollen with other pollen.

In the greenhouses, as long as they are well insulated and insects cannot access them, there is

no need for flower protection

Pollinations can be easily done by hand as follow:

• Harvest the male flowers and keep them in a cup, without mixing flowers from plants

genetically different.

• Revert the petals of the male toward the petiol and then use the flower as a brush to

impollinate the female: if the female is not completely open, it can be opened but the

19

petals have not to be broken. If the female plant has perfect flowers the anthers

should be removed with a forcip to gain more successful pollinations.

o Pollen is mature for pollination only if bright yellow.

o The stigma is ready to receive the pollen only if yellow-white ; if darker, it is

too old and the pollination will fail.

o The anthers, the stigma, and the ovary have not to be touched with the naked

hand because it seems to compromise the pollination.

Hand-pollination

20

Pollinated flowers must be marked with a label reporting the following data:

• Year and location.

• Cross- or self-pollination ID (plot numbers can be used and then a nursery plan with

detailed ID of every plot should be kept); for self's a x surrounded by a circle might

be used as code.

• Date of pollination.

The label could be of different shape but always plastic, waterproof, and white (other colors

might attract insects). The label should be written with a permanent marker or, in certain

cases, with a pencil: it is important that the codes last long enough to accompany the seeds all

the way from pollination to seed-packaging. The label has not to be attached to the petiol

because it endangers fruit-set: it is also important to establish a convention on the position of

the label before or after the knot with marked flower (before seems wiser because the newer

females will develop after the older ones).

21

Greenhouse and Field Pollination Tags

22

Genetics

The genetics of the watermelon has been widely studied and several genes have been

reported and described (Rhodes, 1999).

Table 6. Gene list for watermelon z Gene symbol Gene name and descriptionIst IInd Citations a - Andromonoecious. Recessive to monoecious

(Poole and Grimball, 1945; Porter, 1937; Rosa, 1928)Aco-1 - Aconitase-1.

(Navot et al., 1990)Aco-2 - Aconitase-2.

(Navot et al., 1990)Adh-1+ - Alcohol dehydrogenase-1. One of five codominant alleles,

each regulating one band.(Navot and Zamir, 1986; Navot and Zamir, 1987; Zamir et al.,

1984)Adh-11 - Alcohol dehydrogenase-11. One of five codominant alleles,

each regulating one band. Found in C. lanatus var. citroidesand C. colocynthis.(Navot and Zamir, 1986; Navot and Zamir, 1987; Zamir et al.,

1984)Adh-12 - Alcohol dehydrogenase-12. One of five codominant alleles,

each regulating one band. Found in C. lanatus and C .lanatus var. citroides.

` (Navot and Zamir, 1986; Navot and Zamir, 1987; Zamir et al.,1984)Adh-13 - Alcohol dehydrogenase-13. One of five codominant alleles,

each regulating one band. Found in Praecitrullus fistulosus.(Navot and Zamir, 1986; Navot and Zamir, 1987; Zamir et al.,

1984)Adh-14 - Alcohol dehydrogenase-14. One of five codominant alleles,

each regulating one band. Found in Acanthosicyosnaudinianus.(Navot and Zamir, 1986; Navot and Zamir, 1987; Zamir et al.,

1984)A f - Aulacophora faveicollis resistance. Resistance to the red

pumpkin beetle. Dominant to susceptibility.(Vashistha and Choudhury, 1972)

23

Aps-1 Acph-A Acid phosphatase-1.(Navot and Zamir, 1986; Navot and Zamir, 1987; Navot et al.,1990; Zamir et al., 1984)

Aps-21 - Acid phosphatase-21. One of two codominant alleles, eachregulating one band. Found in C. lanatus and C. colocynthis.(Navot and Zamir, 1986; Navot and Zamir, 1987; Navot et al.,1990)

Aps-22 - Acid phosphatase-22. One of two codominant alleles, eachregulating one band. Found in Acanthosicyos naudinianus.(Navot and Zamir, 1986; Navot and Zamir, 1987; Navot et al.,1990)

Ar-1 B,Gc Anthracnose resistance to race 1 of Glomerella cingulata var.orbiculare.(Hall et al., 1960; Layton, 1937; Weetman, 1937)

Ar-21 - Anthracnose resistance to race 2 of Colletotr ichumlagenarium derived from PI 299379 and PI 189225.Resistance in Citrullus colocynthis is due to other dominantfactors.(Love and Rhodes, 1988; Love and Rhodes, 1991; Sowell et al.,1980; Suvanprakorn and Norton, 1980; Wimmer et al., 1997)

b l t l Branch less. Meristems for tendrils and branches areultimately replaced by floral meristems.(Lin et al., 1992; Rhodes et al., 1999; Zhang et al., 1996b)

C - Canary yellow flesh. Dominant to pink. ii inhibitory to CC,resulting in red flesh. In the absence of ii, CC is epistatic toYY.(Henderson et al., 1998; Poole, 1944)

d - Dotted seed coat. Black dotted seeds when dominant for r, t,and w.(Kanda, 1951; Poole et al., 1941; Porter, 1937)

d b - Resistance to gummy stem blight caused by Didymellabryoniae from PI 189225. Recessive to susceptibility.(Norton, 1979)

dg - Delayed green. Cotyledons and young leaves are initiallypale green but later develop chlorophyll. First reported tobe hypostatic to I-dg. More recent evidence (submitted forpublication) indicate simple recessiveness.(Rhodes, 1986)

Dia-1 - Diaphorase-1.(Navot and Zamir, 1986)

dw-1 - Dwarf-1. Short internodes, due to fewer, shorter cells thannormal. Allelic to dw-1s.(Liu and Loy, 1972; Mohr and Sandhu, 1975)

dw-1s - Short vine. Allelic to dw-1. Vine length intermediatebetween normal and dwarf. Hypocotyl somewhat longer

24

than normal vine and considerably longer than dwarf. dw-1srecessive to normal.(Henderson, 1991)

dw-2 - Dwarf-2. Short internodes, due to fewer cells.(Liu and Loy, 1972; Mohr, 1956; Mohr and Sandhu, 1975)

e t Explosive rind. Thin, tender rind, bursting when cut.(Poole, 1944; Porter, 1937)

Est-1+ - Esterase-1+. One of six codominant alleles, each regulatingone band. Found in C. lanatus.(Navot and Zamir, 1986; Navot and Zamir, 1987; Navot et al.,1990)

Est-11 - Esterase-11. One of six codominant alleles, each regulatingone band. Found in C. lanatus var. citroides and C.colocynthis.(Chambliss et al., 1968; Navot and Zamir, 1986; Navot et al.,1990)

Est-12 - Esterase-12. One of six codominant alleles, each regulatingone band. Found in C. colocynthis.(Navot and Zamir, 1986; Navot and Zamir, 1987; Navot et al.,1990)

Est-13 - Esterase-13. One of six codominant alleles, each regulatingone band. Found in Praecitrullus fistulosus.(Navot and Zamir, 1986; Navot and Zamir, 1987; Navot et al.,1990)

Est-14 - Esterase-14. One of six codominant alleles, each regulatingone band. Found in C. ecirrhosus.(Navot and Zamir, 1986; Navot and Zamir, 1987; Navot et al.,1990)

Est-15 - Esterase-15. One of six codominant alleles, each regulatingone band. Found in Acanthosicyos naudinianus.(Navot and Zamir, 1986; Navot and Zamir, 1987; Navot et al.,1990)

Est-2+ - Esterase-2+. One of five codominant alleles, each regulatingone band. Found in C. lanatus.(Navot and Zamir, 1986; Navot and Zamir, 1987; Navot et al.,1990)

Est-21 - Esterase-21. One of five codominant alleles, each regulatingone band. Found in C. colocynthis.(Navot and Zamir, 1986; Navot and Zamir, 1987; Navot et al.,1990)

Est-22 - Esterase-22. One of five codominant alleles, each regulatingone band. Found in C. colocynthis.(Navot and Zamir, 1986; Navot and Zamir, 1987; Navot et al.,1990)

25

Est-23 - Esterase-23. One of five codominant alleles, each regulatingone band. Found in Praecitrullus fistulosus.(Navot and Zamir, 1986; Navot and Zamir, 1987; Navot et al.,1990)

Est-24 - Esterase-24. One of five codominant alleles, each regulatingone band. Found in Acanthosicyos naudinianus.(Navot and Zamir, 1986; Navot and Zamir, 1987; Navot et al.,1990)

f - Furrowed fruit surface. Recessive to smooth.(Poole, 1944)

Fdp-1 - Fructose 1,6 diphosphatase-1.(Navot and Zamir, 1986; Navot and Zamir, 1987)

Fo-1 - Dominant gene for resistance to race 1 of Fusariumoxysporum f. sp. niveum.(Henderson et al., 1970; Netzer and Weintall, 1980)

For-1 - Fructose 1,6 diphosphatase-1.(Navot et al., 1990)

F w r - Fruit fly resistance in watermelon. Dominant tosusceptibility to Dacus cucurbitae.(Khandelwal and Nath, 1978)

g d Light green skin. Light green fruit recessive to dark green(D) and striped green (ds).(Netzer and Weintall, 1980; Poole and Grimball, 1945;Vashistha and Choudhury, 1972)

gs ds Striped green skin. Recessive to dark green but dominant tolight green skin.(Love and Rhodes, 1991; Weetman, 1937)

Gdh-1 - Glutamate dehydrogenase-1. Isozyme located in cytosol.(Navot and Zamir, 1986)

Gdh-2 - Glutamate dehydrogenase-2. Isozyme located in plastids.(Navot and Zamir, 1986; Navot et al., 1990)

gf - Green flower color.(Kwon et al., 1999)

g m s m s g Glabrous male sterile. Foliage lacking trichomes; malesterile - caused by chromosome desynapsis.(Ray and Sherman, 1988; Watts, 1962; Watts, 1967)

go c golden. Yellow color of older leaves and mature fruit.(Barham, 1956)

Got-1+ - Glutamate oxaloacetate transaminase-1+. One of fourcodominant alleles, each regulating one band. Found in C.lanatus.(Navot and Zamir, 1986; Navot and Zamir, 1987; Navot et al.,1990; Zamir et al., 1984)

26

Got-11 - Glutamate oxaloacetate transaminase-11. One of fourcodominant alleles, each regulating one band. Found in C.colocynthis and Praecitrullus fistulosus.(Navot and Zamir, 1986; Navot and Zamir, 1987; Navot et al.,1990; Zamir et al., 1984)

Got-12 - Glutamate oxaloacetate transaminase-12. One of fourcodominant alleles, each regulating one band. Found in C.lanatus var. citroides.(Navot and Zamir, 1986; Navot and Zamir, 1987; Navot et al.,1990; Zamir et al., 1984)

Got-13 - Glutamate oxaloacetate transaminase-13. One of fourcodominant alleles, each regulating one band. Found inAcanthosicyos naudinianus.(Navot and Zamir, 1986; Navot and Zamir, 1987; Navot et al.,1990; Zamir et al., 1984)

Got-2+ - Glutamate oxaloacetate transaminase-2+. One of fivecodominant alleles, each regulating one band. Found in C.lanatus.(Navot and Zamir, 1986; Navot and Zamir, 1987; Navot et al.,1990; Zamir et al., 1984)

Got-21 - Glutamate oxaloacetate transaminase-21. One of fivecodominant alleles, each regulating one band. Found in C.colocynthis.(Navot and Zamir, 1986; Navot and Zamir, 1987; Navot et al.,1990; Zamir et al., 1984)

Got-22 - Glutamate oxaloacetate transaminase-22. One of fivecodominant alleles, each regulating one band. Found in C.ecirrhosus.(Navot and Zamir, 1986; Navot and Zamir, 1987; Navot et al.,1990; Zamir et al., 1984)

Got-23 - Glutamate oxaloacetate transaminase-23. One of fivecodominant alleles, each regulating one band. Found inPraecitrullus fistulosus.(Navot and Zamir, 1986; Navot and Zamir, 1987; Navot et al.,1990; Zamir et al., 1984)

Got-24 - Glutamate oxaloacetate transaminase-24. One of fivecodominant alleles, each regulating oneband. Found inAcanthosicyos naudinianus.(Navot and Zamir, 1986; Navot and Zamir, 1987; Navot et al.,1990; Zamir et al., 1984)

Got-3 - Glutamate oxaloacetate transaminase-3.(Zamir et al., 1984)

Got-4 - Glutamate oxaloacetate transaminase-4.(Navot et al., 1990; Zamir et al., 1984)

27

hsp70 - Heat shock protein 70. One gene presequence 72-kDa hsp70is modulated differently in glyoxomes and plastids.(Wimmer et al., 1997)

I-dg - Inhibitor of delayed green. Epistatic to dg: dg dg I-dg I-dgand dg dg I-dg i-dg plants are pale green; and dg dg i-dg i-dgplants are normal. This gene was not present in moreadvanced germplasm.(Rhodes, 1986)

Idh-1 - Isocitrate dehydrogenase-1.(Zamir et al., 1984)

i-C - Inhibitor of canary yellow, resulting in red flesh.(Henderson et al., 1998)

j a - Juvenile albino. Chlorophyll reduced by short days inseedlings, leaf margins, rind.(Zhang et al., 1996b)

l - Long seed. Long recessive to medium length of seed;interacts with s.(Poole et al., 1941)

Lap-1 - Leucine aminopeptidase-1.(Navot and Zamir, 1986; Navot et al., 1990)

m - Mottled skin. Greenish white mottling of fruit skin.(Poole and Grimball, 1945; Weetman, 1937)

Mdh-1+ - Malic dehydrogenase-1+. One of two codominant alleles,each regulating one band. Found in C. lanatus.(Navot and Zamir, 1987; Zamir et al., 1984)

Mdh-11 - Malic dehydrogenase-11. One of two codominant alleles,each regulating one band. Found in Praecitrullus fistulosus.(Navot and Zamir, 1987; Zamir et al., 1984)

Mdh-2+ - Malic dehydrogenase-2+. One of three codominant alleles,each regulating one band. Found in C. lanatus.(Navot and Zamir, 1987)

Mdh-21 - Malic dehydrogenase-21. One of three codominant alleles,each regulating one band. Found in C. colocynthis.(Navot and Zamir, 1987)

Mdh-22 - Malic dehydrogenase-22. One of three codominant alleles,each regulating one band. Found in Praecitrullus fistulosus.(Navot and Zamir, 1987)

Me-1+ - Malic enzyme-1+. One of three codominant alleles, eachregulating one band. Found in C. lanatus.(Navot and Zamir, 1986; Navot and Zamir, 1987; Navot et al.,1990; Zamir et al., 1984)

Me-11 - Malic enzyme-11. One of three codominant alleles, eachregulating one band. Found in Praecitrullus fistulosus.(Navot and Zamir, 1986; Navot and Zamir, 1987; Navot et al.,1990; Zamir et al., 1984)

28

Me-12 - Malic enzyme-12. One of three codominant alleles, eachregulating one band. Found in C. colocynthis.(Navot and Zamir, 1986; Navot and Zamir, 1987; Navot et al.,1990; Zamir et al., 1984)

Me-2 - Malic enzyme-2.(Zamir et al., 1984)

ms - Male sterile.(Zamir et al., 1984; Zhang and Wang, 1990; Zhang et al., 1994)

msdw - male sterile, dwarf.(Huang et al., 1998)

nl - Nonlobed leaves. Leaves lack lobing; dominance incomplete.(Mohr, 1953)

O - Elongate fruit. Incompletely dominant to spherical.(Poole, 1944; Weetman, 1937)

p - Pencilled lines on skin. Inconspicuous; recessive to nettedfruit.(Poole, 1944; Weetman, 1937)

P g d - 1 + 6-Pgdh-1+ 6-Phosphogluconate dehydrogenase-1+. One of threecodominant alleles, each regulating one plastid band. Foundin C. lanatus.(Navot and Zamir, 1986; Navot and Zamir, 1987; Navot et al.,1990; Zamir et al., 1984)

Pgd-11 6-Pgdh-11 6-Phosphogluconate dehydrogenase-11. One of threecodominant alleles, each regulating one plastid band. Foundin Praecitrullus fistulosus.(Navot and Zamir, 1986; Navot and Zamir, 1987; Navot et al.,1990; Zamir et al., 1984)

Pgd-12 6-Pgdh-12 6-Phosphogluconate dehydrogenase-12. One of threecodominant alleles, each regulating one plastid band. Foundin Acanthosicyos naudinianus.(Navot and Zamir, 1986; Navot and Zamir, 1987; Navot et al.,1990; Zamir et al., 1984)

P g d - 2 + 6-Pgdh-2+ 6-Phosphogluconate dehydrogenase-2+. One of fivecodominant alleles, each regulating one cytosolic band.Found in C. lanatus.(Navot and Zamir, 1986; Zamir et al., 1984)

Pgd-21 6-Pgdh-21 6-Phosphogluconate dehydrogenase-21. One of fivecodominant alleles, each regulating one cytosolic band.Found in C. ecirrhosus.(Navot and Zamir, 1987; Zamir et al., 1984)

Pgd-22 6-Pgdh-22 6-Phosphogluconate dehydrogenase-22. One of fivecodominant alleles, each regulating one cytosolic band.Found in Praecitrullus fistulosus.(Navot and Zamir, 1987; Zamir et al., 1984)

29

Pgd-23 6-Pgdh-23 6-Phosphogluconate dehydrogenase-23. One of fivecodominant alleles, each regulating one cytosolic band.Found in C. colocynthis.(Navot and Zamir, 1987; Zamir et al., 1984)

Pgd-24 6-Pgdh-24 6-Phosphogluconate dehydrogenase-24. One of fivecodominant alleles, each regulating one cytosolic band.Found in Acanthosicyos naudinianus.(Navot and Zamir, 1987; Zamir et al., 1984)

Pgi-1+ - Phosphoglucoisomerase-1+. One of three codominant alleles,each regulating one plastid band. Found in C. lanatus.(Navot and Zamir, 1986; Navot and Zamir, 1987; Navot et al.,1990)

Pgi-11 - Phosphoglucoisomerase-11. One of three codominant alleles,each regulating one plastid band. Found in C. colocynthis.(Navot and Zamir, 1986; Navot and Zamir, 1987; Navot et al.,1990)

Pgi-12 - Phosphoglucoisomerase-12. One of three codominant alleles,each regulating one plastid band. Found in Acanthosicyosnaudinianus.(Navot and Zamir, 1986; Navot and Zamir, 1987; Navot et al.,1990)

Pgi-2+ - Phosphoglucoisomerase-2+. One of six codominant alleles,each regulating one cytosolic band. Found in C. lanatus.(Navot and Zamir, 1986; Navot and Zamir, 1987; Navot et al.,1990; Zamir et al., 1984)

Pgi-21 - Phosphoglucoisomerase-21. One of six codominantalleles,each regulating one cytosolic band. Found in C.lanatus and C. colocynthis.(Navot and Zamir, 1986; Navot and Zamir, 1987; Navot et al.,1990; Zamir et al., 1984)

Pgi-22 - Phosphoglucoisomerase-22. One of six codominant alleles,each regulating one cytosolic band. Found in C. ecirrhosus.(Navot and Zamir, 1986; Navot and Zamir, 1987; Navot et al.,1990; Zamir et al., 1984)

Pgi-23 - Phosphoglucoisomerase-23. One of six codominant alleles,each regulating one cytosolic band. Found in Praecitrullusfistulosus.(Navot and Zamir, 1986; Navot and Zamir, 1987; Navot et al.,1990; Zamir et al., 1984)

Pgi-24 - Phosphoglucoisomerase-24. One of six codominant alleles,each regulating one cytosolic band. Found in C. lanatus var.citroides.(Navot and Zamir, 1986; Navot and Zamir, 1987; Navot et al.,1990; Zamir et al., 1984)

30

Pgi-25 - Phosphoglucoisomerase-25. One of six codominant alleles,each regulating one cytosolic band. Found in Acanthosicyosnaudinianus.(Navot and Zamir, 1986; Navot and Zamir, 1987; Navot et al.,1990; Zamir et al., 1984)

Pgm-1+ - Phosphoglucomutase-1+. One of four codominant alleles,each regulating one plastid band. Found in C. lanatus.(Navot and Zamir, 1986; Navot and Zamir, 1987; Navot et al.,1990; Zamir et al., 1984)

Pgm-11 - Phosphoglucomutase-11. One of four codominant alleles,each regulating one plastid band. Found in C. colocynthis.(Navot and Zamir, 1986; Navot and Zamir, 1987; Navot et al.,1990; Zamir et al., 1984)

Pgm-12 - Phosphoglucomutase-12. One of four codominant alleles,each regulating one plastid band. Found in Acanthosicyosnaudinianus.(Navot and Zamir, 1986; Navot and Zamir, 1987; Navot et al.,1990; Zamir et al., 1984)

Pgm-13 - Phosphoglucomutase-13. One of four codominant alleles,each regulating one plastid band. Found in Praecitrullusfistulosus.(Navot and Zamir, 1986; Navot and Zamir, 1987; Navot et al.,1990; Zamir et al., 1984)

Pgm-2+ - Phosphoglucomutase-2+. One of four codominant alleles,each regulating one cytosolic band. Found in C. lanatus.(Navot and Zamir, 1987; Zamir et al., 1984)

Pgm-21 - Phosphoglucomutase-21. One of four codominant alleles,each regulating one cytosolic band. Found in Acanthosicyosnaudinianus.(Navot and Zamir, 1987; Zamir et al., 1984)

Pgm-22 - Phosphoglucomutase-22. One of four codominant alleles,each regulating one cytosolic band. Found in C. lanatus.(Navot and Zamir, 1987; Zamir et al., 1984)

Pgm-23 - Phosphoglucomutase-23. One of four codominant alleles,each regulating one cytosolic band. Found in Praecitrullusfistulosus.(Navot and Zamir, 1987; Zamir et al., 1984)

p m - Powdery mildew susceptibility. Susceptibility toSphaerotheca fuliginea.(Robinson et al., 1975)

Prx-1+ - Peroxidase-1+. One of seven codominant alleles, eachregulating one band. Found in C. lanatus.(Navot and Zamir, 1986; Navot and Zamir, 1987; Navot et al.,1990)

31

Prx-11 - Peroxidase-11. One of seven codominant alleles, eachregulating one band. Found in C. colocynthis.(Navot and Zamir, 1986; Navot and Zamir, 1987; Navot et al.,1990)

Prx-12 - Peroxidase-12. One of seven codominant alleles, eachregulating one band. Found in Praecitrullus fistulosus.(Navot and Zamir, 1986; Navot and Zamir, 1987; Navot et al.,1990)

Prx-13 - Peroxidase-13. One of seven codominant alleles, eachregulating one band. Found in C. lanatus.(Navot and Zamir, 1986; Navot and Zamir, 1987; Navot et al.,1990)

Prx-14 - Peroxidase-14. One of seven codominant alleles, eachregulating one band. Found in C. ecirrhosus.(Navot and Zamir, 1986; Navot and Zamir, 1987; Navot et al.,1990)

Prx-15 - Peroxidase-15. One of seven codominant alleles, eachregulating one band. Found in C. lanatus and C. colocynthis.(Navot and Zamir, 1986; Navot and Zamir, 1987; Navot et al.,1990)

Prx-16 - Peroxidase-16. One of seven codominant alleles, eachregulating one band. Found in Acanthosicyos naudinianus.(Navot and Zamir, 1986; Navot and Zamir, 1987; Navot et al.,1990)

Prx-2 - Peroxidase-2.(Navot and Zamir, 1987)

Prx-3 - Peroxidase-3.(Navot and Zamir, 1987)

r - Red seed coat. Interacts with w and t.(Poole et al., 1941)

s - Short seeds. Epistatic to l.(Poole et al., 1941)

Sat - Serine acetyltransferase. Catalyzes the formation of O-acetylserine from serine and acetyl-CoA.

Skdh-1 - Shikimic acid dehydrogenase-1.(Zamir et al., 1984)

Skdh-2+ - Shikimic acid dehydrogenase-2+. One of six codominantalleles, each regulating one band.(Navot and Zamir, 1986; Navot and Zamir, 1987; Navot et al.,1990)

Skdh-21 - Shikimic acid dehydrogenase-21. One of six codominantalleles, each regulating one band. Found in C. colocynthis.(Navot and Zamir, 1986; Navot and Zamir, 1987; Navot et al.,1990)

32

Skdh-22 - Shikimic acid dehydrogenase-22. One of six codominantalleles, each regulating one band. Found in C. colocynthis.(Navot and Zamir, 1986; Navot and Zamir, 1987; Navot et al.,1990)

Skdh-23 - Shikimic acid dehydrogenase-23. One of six codominantalleles, each regulating one band. Found in Acanthosicyosnaudinianus.(Navot and Zamir, 1986; Navot and Zamir, 1987; Navot et al.,1990)

Skdh-24 - Shikimic acid dehydrogenase-24. One of six codominantalleles, each regulating one band. Found in C. ecirrhosus.(Navot and Zamir, 1986; Navot and Zamir, 1987; Navot et al.,1990)

Skdh-25 - Shikimic acid dehydrogenase-25. One of six codominantalleles, each regulating one band. Found in Praecitrullusfistulosus.(Navot and Zamir, 1986; Navot and Zamir, 1987; Navot et al.,1990)

slv - Seedling leaf variegation. Conferred by a single recessivegene. Dominant allele at same locus in PI 482261.(Provvidenti, 1994)

Sod-1+ - Superoxide dismutase-1+. One of three codominant alleles,each regulating one band. Found in C. lanatus.(Navot and Zamir, 1986; Navot and Zamir, 1987; Navot et al.,1990; Zamir et al., 1984)

Sod-11 - Superoxide dismutase-11. One of three codominant alleles,each regulating one band. Found in C. colocynthis.(Navot and Zamir, 1986; Navot and Zamir, 1987; Navot et al.,1990; Zamir et al., 1984)

Sod-12 - Superoxide dismutase-12. One of three codominant alleles,each regulating one band. Found in Acanthosicyosnaudinianus.(Navot and Zamir, 1986; Navot and Zamir, 1987; Navot et al.,1990; Zamir et al., 1984)

Sod-2+ - Superoxide dismutase-2+. One of two codominant alleles,each regulating one band. Found in C. lanatus.(Navot and Zamir, 1987)

Sod-21 - Superoxide dismutase-21. One of two codominant alleles,each regulating one band. Found in Acanthosicyosnaudinianus.(Navot and Zamir, 1987)

Sod-3 - Superoxide dismutase-3. One of two codominant alleles,each regulating one band. Found in C. lanatus.(Navot and Zamir, 1987)

33

Sod-31 - Superoxide dismutase-31. One of two codominant alleles,each regulating one band. Found in Praecitrullus fistulosus.(Navot and Zamir, 1987)

Sp - Spotted cotyledons, leaves and fruit.(Rhodes, 1986)

Spr-1 - Seed protein-1.(Navot and Zamir, 1986)

Spr-2 - Seed protein-2.(Navot and Zamir, 1986)

Spr-3 - Seed protein-3.(Navot and Zamir, 1986)

Spr-4 Sp-4 Seed protein-4.(Navot and Zamir, 1986; Navot et al., 1990)

Spr-5 Sp-5 Seed protein-5.(Navot and Zamir, 1986; Navot et al., 1990)

su Bi, suBi Suppressor of bitterness. Non-bitter fruit. Bitterness in C.colocynthis is due to Su Su genotype.(Chambliss et al., 1968; Navot et al., 1990)

t bt Tan seed coat. Interacts with r and w.(Mc Kay, 1936; Poole et al., 1941)

tl - Tendrilless. After 4th or 5th node, vegetative axillary budsare transformed into flower buds and leaf shape is altered.(Rhodes et al., 1999; Zhang et al., 1996a)

Tpi-1+ - Triosephosphatase isomerase-1+. One of four codominantalleles, each regulating one band. Found in C. lanatus.(Navot and Zamir, 1986; Navot and Zamir, 1987; Navot et al.,1990)

Tpi-11 - Triosephosphatase isomerase-11. One of four codominantalleles, each regulating one band. Found in C. colocynthis.(Navot and Zamir, 1986; Navot and Zamir, 1987; Navot et al.,1990)

Tpi-12 - Triosephosphatase isomerase-12. One of four codominantalleles, each regulating one band. Found in Praecitrullusfistulosus.(Navot and Zamir, 1986; Navot and Zamir, 1987; Navot et al.,1990)

Tpi-13 - Triosephosphatase isomerase-13. One of four codominantalleles, each regulating one band. Found in Acanthosicyosnaudinianus.(Navot and Zamir, 1986; Navot and Zamir, 1987; Navot et al.,1990)

Tpi-2+ - Triosephosphatase isomerase-2+. One of three codominantalleles, each regulating one band. Found in C. lanatus.(Navot and Zamir, 1987)

34

Tpi-21 - Triosephosphatase isomerase-21. One of three codominantalleles, each regulating one band. Found in Acanthosicyosnaudinianus.(Navot and Zamir, 1987)

Tpi-22 - Triosephosphatase isomerase-22. One of three codominantalleles, each regulating one band. Found in Praecitrullusfistulosus.(Navot and Zamir, 1987)

Ure-1 - Urease-1.(Navot and Zamir, 1987)

w - White seed coat. Interacts with r and t.(Poole et al., 1941)

Wf W White flesh. Wf is epistatic to the second gene b (or C?)which conditions yellow (Canary yellow?) and red flesh.Wf_B_ and Wf_bb are white fleshed, wf wf B_ is yellowfleshed, and wf wf b b is red fleshed.(Shimotsuma, 1963)

y rd Yellow flesh (`Golden Honey' type). Recessive to Y (redflesh).(Henderson, 1989; Henderson et al., 1998; Poole, 1944; Porter,1937)

yo - Orange flesh (from `Tendersweet Orange Flesh'). Allelic toy. Y (red flesh) is dominant to yo (orange flesh) and y (yellowflesh); yo (orange flesh) is dominant to y (yellow flesh).(Henderson, 1989; Henderson et al., 1998)

Yl - Yellow leaf (from `Yellow Skin'). Incompletely dominant togreen leaf.(Warid and Abd el Hafez, 1976)

z Currently being reviewed by Guner, N. and Wehner, T. C.

Various traits determine the overall phenotypic value of the watermelon and can be in

general grouped as traits determining the plant-habit and traits determining the fruit-

phenotype. Both the categories are very important since they concur together to generate

varieties with certain standard in terms of type, adaptation, yield, quality, and resistance to

different stresses.

35

Inbreeding depression is not a concern for watermelon breeders. Heterosis is not as

important as in other cross-pollinated crops and hybrid varieties are useful only as a way to

combine valuable dominant traits together: however, inbred lines are far enough for breeding

for yield and quality (Wehner et al., 2001).

For a complete description and discussion of these traits refer to (Wehner et al., 2001) and

(Mohr, 1986).

Fruit Traits

• Size

o Polygenic trait

12 lb 18 lb 24 lb 32 lb

Ice box Small (syn. pee-wee) Medium Large Giant

36

• Shape

Round Oval

Blocky Elongate

o Round vs. Elongate => single gene

o Blocky => heterozygote (F1)

37

• Rind Color (syn. Pattern)

Solid (light, dark) Narrow Striped Medium Striped

Wide Striped Gray

38

• Rind Thickness (at blossom end) and Toughness

Good Rind Thickness

Too Thin Rind Too Thick Rind

o Rind Toughness can be measured in different ways

ß Using a penetrometer at stem-end

ß Pressing with a thumb at stem-end

ß Dropping the fruit from knee-heigth to see if it split-opens

39

• Flesh Color

White Dark Red Light red

Canary Yellow Salmon Yellow Orange

o Light Red (YY) > Orange (y0y0) > Salmon Yellow (yy)

o Canary Yellow (CC) > Non-Canary Yellow (cc)

o White > Red

o Dark Red => inheritance is still unknown (hypothesis: intensiphire gene)

40

• Flesh Texture

o Soft (syn. Watery) vs. Firm (syn. Crispy)

o Fibrous vs. Non-Fibrous

• Soluble Solids Content and Flavor

o "Soluble Solids = Sugars" is a generally accepted approximation

o Measured with a refractometer (Brix-Meter)

ß Min. Marketable 10%

ß Average Today 14%

o Polygenic trait

o Fructose tastes sweeter than sucrose but this difference is not measurable with

the refractometer: tasting the fruit is also very important to determine presence

of bitternes or other desirable/undesirable characteristics

• Seeds and seedlessness

Seeded (Diploid) Seedless (Triploid)

41

Too big Seed-Pockets Vestigial Seedsare a difect in seeded fruits are often present in seedless fruits

o In seeded watermelons too many or too big seeds are difects

o In seedless watermelons a few seeds (3-4) are accepted

42

o Seeds can be of different colors

White Seeds Black Seeds

Red Seeds Green Seeds

Specled Seeds Tan Seeds (pink is fungicide)

43

o Seeds can vary in size from very small (tomato seeds-size) to very large

(pumpkin seeds-size)

Small Seeds Large Seeds

o Seeds can vary in type (Non-Egusi > Egusi; Egusi seeds are present only in

wild relatives of the cultivated watermelon)

Non-Egusi (Normal) Seeds Egusi Seeds (only in C. colocynthis)

44

Plant Traits

• Leaf Type

Small vs. Large Leaf Grades of Lobation (also Non-lobed)

• Stem Shape and Habit

Straight Stem Zig-Zag Stem

45

• Plant Habit

Normal Type Bushy Type (Highly Branching)

o Recently bushy-types have been considered by breeders for the production of

small and highly androecious pollinizer for seedless production

• Plant Size

Small (6 Plants/Plot) Large (6 Plants/Plot)

o Watermelons can vary in size from dwarf to large

o Usually small-fruited varieties have shorter vines

46

Production Traits

• Yield

o Polygenic trait

o # of genes involved still unknown

o Growers' goal => 1 load/acre = 45,000 lb/acre

o Breeding for yield is in progress but there has been no valuable advance yet

ß Genetic variability among watermelon cultivars has been demonstrated

to be very low (Levi et al., 2001a)

ß Yield is phenotipically given by[( # of fruits – culls) x average weight]

Table 8. Heritability (h2B) estimates for yield

(Gopal et al., 1996) z

Character s2

P s2G h2

B # of fruit/vine 28.45 27.87 85.93m fruit weight 16.43 16.33 98.84Yield/vine 35.43 35.02 77.07 z Data from analysis of F1 hybrids of 6 diverse varieties

• Earliness

o Polygenic trait

o Growers' goal => "The earlier, the better!"

o Measured as # of nodes to the 1st female flower or # of days to the first mature

fruit from planting

47

Table 9. Heritability (h2B) estimates for earliness

(Sidhu et al., 1977) z

Character h2

N h2B

# of nodes to the 1st female flower 24.70 35.85# of days to the 1st mature fruit 15.78 22.89 z Data from analysis of F1 hybrids of 7 diverse varieties; thetest should be repeated with an NC Design to confirm thesedata

Stress Tolerance Traits

For a complete pathological description of diseases and insects affecting the watermelon

refer to (Zitter et al., 1996) and for fisiological disorders to (Maynard et al., 2001).

• Fusarium Wilt

Table 10. Resistance to Fusarium Wilt(Wehner et al., 2001) Cultigen Resistance to Race

0 1 2 Black Diamond, Sugar Baby S S SQuetzali, Mickylee R S SCharleston Gray, Crimson Sweet R M SCalhoun Gray R R SPI 296341, PI 271769 R R R

• Anthracnose (Colletotrichum lagenarium)

o 7 races: 1, 2, and 3 the most dangerous

o Resistance to race 1 and 3 => single dominant gene (Ar-1)

o Resistance to race 2 => polygenic

48

Table 11. Resistance to Anthracnose(Wehner et al., 2001) Cultigen Resistance to Race

1 2 3 Congo, Fairfax, Charleston Gray,Crimson Sweet, others R S RPI 189225, PI 271775, PI 299379,PI 271778, PI 203551, PI 270550,PI 326515, PI 271775, PI 271779,PI 203551 R ? RR 143, PI 512385 ? R ?

• Gummy Stem Blight (Didymella bryoniae)

o Resistance => single recessive gene (db)

Table 12. Resistance to Gummy Stem Blight(Song et al., 2002) Cultigen Level of Resistance PI 164248 HighPI 244019 HighPI 254744 HighPI 271771 HighPI 279461 HighPI 296332 HighPI 379243 HighPI 482276 HighPI 482284 HighPI 482379 HighPI 490383 HighPI 526233 HighPI 189225 ModeratePI 271778 Moderate

49

GSB – Resistance (front plots) vs. Susceptibility (border and back plots)

• Powdery Mildew (Sphaerotheca fuliginea) (Wehner et al., 2001)

o Susceptibility => single recessive gene (pm)

o Resistant germplasm => PI 189225, PI 271778

• Bacterial Fruit Blotch (Acidovorax avenae subsp. citrulli) (Wehner et al., 2001)

o Resistant germplasm => not yet found

o Germplasm screening is in progress

o Surface sterilization of fruits before seed-extraction seems to control the

spresding of the disease among the breeding-material

50

• Bacterial Rind Necrosis (Erwinia spp.)

Table 13. Resistance to Bacterial Rind Necrosis(Wehner et al., 2001) Cultigen Level of Resistance Sweet princess RJubilee RKlondike Blue Ribbon SLouisiana Queen S

• Yellow Vine (Wehner et al., 2001)

o New important disease

o Studies are in progress

• Verticillium Wilt (Wehner et al., 2001)

o New important disease

o Studies are in progress

• Alternaria Leaf Spot (Wehner et al., 2001)

o Resistant germplasm => 'Sugar Baby', 'Faifax', 'Calhoun Gray'

• Root Knot Nematodes (Meloidogyne spp.) (Wehner et al., 2001)

o Resistant germplasm => not yet found

o Germplasm screening is in progress

• Viruses (Strange et al., 2002; Wehner et al., 2001)

51

o Resistant germplasm

ß WMV-2 => PI 244018, PI 244019

ß ZYMV => PI 482299, PI 482261, PI 595203, PI 255137

ß PRSV (WMV-1) => PI 278005, PI 277972, PI278009, PI244017

o Germplasm screening is in progress

• Aphiis gossypii (Wehner et al., 2001)

o Resistant germplasm => PI 299563

• Spider Mites (Wehner et al., 2001)

o Resistant germplasm => 'Congo', 'Giza 1'

• Dacus cucurbitae (Wehner et al., 2001)

o Resistance => single dominat gene (Fwr)

o Resistant germplasm => PI 299563

• Aulacophora foveicollis (Pumpkin Beetle) (Wehner et al., 2001)

o Resistant germplasm => 'Afghan'

• Spotted, Striped, Banded Cucumber Beetles (Wehner et al., 2001)

o Germplasm screening is needed

• Pickleworm (Wehner et al., 2001)

52

o Resistant germplasm => 'Blue Ribbon', 'Crimson Sweet'

• Hollowheart (Wehner et al., 2001)

o The most important physiological disorder

o Mechanism still not clear but high environmental variation is common

o Some degrees of resistance among cultivars

o Resistant germplasm => not yet found

Severe Hollowheart

• Rind Necrosis, Blossom-End Rot, Cross Stitch (Wehner et al., 2001)

o Less important physiological disorders

o More informations are needed

o Breeding for resistance might be a plus for leading varieties

53

• Chilling (Provvidenti, 1992; Xu et al., 2000)

o The most important environmental stress

o Breeding for resistance is very important for certain areas of early production

o Resistance => single dominat gene

o Resistant germplasm => PI 482322, Zimbabwe collection

54

Classical Breeding Techniques

Breeding techniques appliable to the watermelon are widely and accurately described and

discussed in (Wehner et al., 2001); the following schemes summarize the main points of the

cited chapter. Of course breeders might develop programs based on mixtures of these basic

techniques presented. How to generate the original base population to start a breeding

programs still is a matter of choices and art of the breeder. Limiting factors for watermelon

breeding might be the size of the plants and consequently the amount of land needed; the

natural cycle also is quite long but a good choice of the locations and maybe cohoperation

with farms and institutions in the Southern Emisphere might allow 3 to 4 generations per

year. Other limitations are arising now from the necessity to keep the seeds clean from

pathogens: if fruits have to be sterilized before cutting and cut away from watermelon plants

(one of the practice suggested for prevention of fruit blotch contaminations), harvest and

handling of the fruits might become challenging for recurrent selection designs.

Recurrent selection

• Two main designs for population improvement

o Recurrent selection for phenotypic traits

o Reciprocal recurrent selection – half-sib families testing

• Advantages

o Possible gain for quantitative traits (i.e. yield)

o Improvement of the base population and contemporary production of elitè

inbreds

o Open pollination (not in RRS-HS)

55

• Disadvantages

o Lots of resources required

ß Labor

ß Field space for testing

Pedigree breeding

• Advantages

o Possible to combine genes from two-three different elitè parents

o High control of crosses

o Continue selection

o Possible to produce elitè inbreds and hybrids

• Disadvantages

o Time consuming

o Reduces variability in the germplasm

o Limited to the programmed crosses

o Possible gain only for qualitative traits

Backcross breeding

• Same advantages and disadvantages of pedigree breeding

• Very useful to add a single gene (usually for resistance to pathogens) to an adapted

breeding line or cultivar

56

Inbreds development

• Developed out of the breeding designs above

• Lines become inbreds at the S6-S8 generation of self-pollination

o Usually open-pollination by bees in isolation blocks (or cages)

• Usually performing as well as hybrids

• Final selection required at S6-S8 before release or use as parental for hybrids

development

Hybrids development

• Developed usually by crossing two (rarely more) inbreds

• Very useful to protect property of the inbred parents

• Require a lot of resourches for pollination (hand-pollination) or for use of male-

sterility systems

Seedless development

• See below for description of the technique

• Useful to use tetraploids with grey rind for crossing with dyploids with striped rind

o Intercrossing rate in open–pollination is about 84%

The watermelon has been bred mainly using pedigree breeding and using a narrow base sets

of parentals until now so that genetic variation among modern cultivars is very low (much

lower than among old cultivars from the late 1,800's and early 1,900's). Selection for

57

qualitative traits has been done gaining very good results in terms of quality, type, and

adaptability; selection for quantitative traits has enhanced the average levels of sugar content

and yield but the variability has been decreased.

It does look like that there is plenty of possibilities to experiment other breeding designs to

try to keep the high quality gained over about 150 years of selection increasing the value of

certain quantitative traits.

58

Special Breeding Techniques

Biotechnologies are applied at a certain extent nowadays also to watermelons:

• Molecular markers have been found for tagging genes for pathogen resistance (Levi,

2001)

• Linkage maps have been constructed using the RAPD technology (Levi et al., 2001b)

• Genetic diversity among cultivars has been tested using the RAPD technology (Levi

and Thomas, 2001; Levi et al., 2001a)

Tissue culture has been proposed as a technique for transformation and for propagation of

valuable phenotypes but the high costs cannot compete with seed propagation: so far tissue

culture is still used for propagation of valuable tetraploid watermelons for the productions of

seedless triploids.

Interspecific hybridization could be used to introgreed valuable traits from wild relatives and

in particular fom Citrullus colocynthis which is a valuable source of pathogen-resistant

germplasm.

Transformation could be used to increase genetic variability in the watermelon; seeds or

plantlets in vitro could be trasformed and then screened and characterized for several

different traits generating to a new collection of transformed germplasm. Mutation breeding

has already been applied in China for male-sterility and triploid production and has given

encouraging results even if not definite or marketable (Zhang and Rhodes, 1992).

59

Polyploidization (the technique was initiated by Kihara and his group (Eigsti, 1979)) has

been used for the production of triploid seedless watermelons. The technique implies several

passages:

• Valuable diploid cultigens are treated with chromosomal doubling agents (colchicine

or dinitroaniline herbicides) to obtain somatic tetraploids (somatic ploidy level can be

detected by screening of the number of chloroplast in the guard-cells: 6 in diploids,

12 in tetraploids) that are self-pollinated to obtain tetraploid seeds

• Tetraploids undergo to several cycles of self-pollination and selection particularly to

increase their seed-production and reduce defects

• Once the tetraploid line is selected and tested, it can be used as female parent for

crossing with a selected diploid male for the production of triploid hybrids (the

reciprocal does not produce seeds)

• Triploidy causes in watermelon male-sterility: though, triploid females need a diploid

pollinizer to producefruits in the grower- fields

It has been reported that russian scientists found spontaneous watermelon polyploids, but

direct informations are not available: further investigations should be required (Eigsti, 1979).

Other biotechnologies might be employed on the improvement of the watermelon but

limiting factors are deriving more from social and political issues rather than scientific

limitations. Gene sequencing, biopathway studies, transformation techniques and gene

tagging techniques (molecular markers, AFLP, RAPD, Microarrays, etc.) should be used

routinarily at least from public geneticists to gain better understanding of the genetic and

biomolecular characteristics of this species. In the future techniques such as gene silencing

60

and/or overexpression might be very useful in solving problems related particularly to 1)

pathogenic or environmental stress resistance, and 2) sex determination and parthenocarpy (it

would allow easier and more economic ways to produce hybrid seeds and seedless varieties).

61

Selection Techniques and Trialing Methods

Methods for selection can be fairly different among breeders but there are some common

guidelines that can be followed.

Type

Selection for type of fruit or plant (dwarf types, busch types, etc.) is very subjective to the

breeder judgement and to the goals of the breeding programs. Attention has to be payed to

do not narrow the presence in the base populations of different types because they could

become necessary to adapt the breeding strategy to the everchanging market. In the past

selection for fruit-type and plant-habit was done at the seedling stage by looking at size

and/or shape of cotyledons and stem (Mohr, 1986) but it is not possible anymore: the actual

market requires a very accurate selection for these traits so that approximation is not

possible; seedling selection can be useful to eventually reduce high numbers of lines to a

pool that can be easily tested in replicated trials.

Breeders have different personal preferences in grading methods for shape and size of the

fruits and for habit of the plants: the techniques range between simple visual evaluation with

the aid of grading scales and the precise measure of the characters (length/diameter ratio for

the fruits, foliage densities, leaf area indexes, vine length, etc. for the plant-habit).

Yield

Yield is measured as number of fruits per plant (average of multiple plants per plot and count

of all the marketable fruits in the plot) and average weight of the fruits (can be averaged over

the most representative 10-20 fruits per plot or, if the plot is fairly uniform, as the weight of

62

the single most representative fruit). Usually breeders measure yield at later stage in cultivar

development because variability for this trait is believed to be quite low and highly

environmental (however, specific research should be done). Fruits must be weight at

picking–maturity.

Quality

Multiple traits have to be measured to estimate the quality of a fruit:

• Sugar content : measured with refractometer

• Flavor: usually panel-taste at later stage of development (in earlier stages breeding

lines are evaluated in terms of good vs. bad-taste for a gross selection)

• Texture, color, and seedness can be measured by simple observations at every

generations

• Size, shape, color: determined by visual evaluation or as explained above for type

Adaptability

Selection for adaptability usually should be done at the beginning of the breeding program by

choosing as parents of the base population adapted or adaptable lines for the area of

marketing of the future varieties. In later stages adaptability can be tested in grower-trials

directly (see below).

Resistance

Resistance to diseases and insects is so important in modern varieties that several specific

tests (ranging between the extreme of biological assay and DNA analysis) have been recently

63

developed: the techniques are becoming day by day more accurate and difficult to perform so

that the cohoperation with professional pathologists, enthomologists and biotechnologists is

vital for the modern breeder. Usually every breeder includes in his team a professional

pathologist specialized on cucurbits; enthomologists and biotechnologists, however, can be

hired as consultant and in the seed-companies usually they assist all the different breeding

programs. The recent trend is to include in the breeding-team directly also biotechnologist

for the exploration of marker-assisted selection techniques.

Resistance to physiological stresses in watermelon almost entirely relates to hollowheart

susceptibility: the only available screening method is the observation of the fruits all the way

through the breeding program: the genetic base of this problem is not yet known and the

environmental variance component is thought to be very high but further analysis are

required to better understand how to simulate stressing conditions to test the germplasm or

how to detect directly the genetic susceptibility with molecular or phenotypic markers.

Resistance to adverse environmental conditions are included in selection for adaptability (see

above).

The data can be collected on paper even though the general opinion is that direct recording

on laptop computers is more reliable because it decreases the sources of error during retyping

for statistical analysis or for preparation of the following breeding-plans.

64

Example of electronic data-sheet for cultivars evaluation

65

Trials of watermelons can be devided in the following general categories:

• Developmental trials of early generations: run directly at the breeding farm or in other

farms in different locations but still part of the breeding program

• Breeding trials of advanced generations: run in grower-fields; they can include:

o Breeding lines: tested in small replicated plots

o Lines for release: tested in long plots at grower-field density and plot width

• Public trials: every generation can be sent for evaluation by public breeders in their

own trials. This is a very interesting source of verification of the breeding lines but it

present some problem to the private breeder: 1) trialing methods are not yet

standardized among public breeders, 2) the competitors will be able to see the lines

and read the measurements and evaluations (even though it is possible to pay a higher

price per plot and keep the data reserved).

• Grower trials: usually growers are very cohoperative in including in their fields some

plots for trialing: sales agents for private companies and extension agents for public

breeders are the best cohoperators to organize and run these type of trials; usually the

breeder visits himself the plots only for evaluation at harvest.

66

Obvious recommendation would be to spend as much time as needed in evaluating the plots

and take every possible measurement of interest maybe along with a digital picture of the

fruits (the picture should include at once all the traits that can be of interest for a future visual

evaluation).

Correct layout for digital imaging of evaluated lines

67

Equipments, Organization, and Protocols

Every breeder would be able to organize his own program in a different way. In this chapter

some suggestions are reported and might be used as guide-lines to develop own protocols and

techniques.

Nursery

Besides certain locations where it is possible to gain 3 generations of watermelons in

open–fields, usually the combination field-greenhouse is the most common for the breeding

nursery.

The fields can be used for testing and for crossing watermelons. For testing purposes it has

been demonstrated (Neppl et al., data not published) that small plots with single rows are

sufficient. For intercrossing breeders use different designs but they all follow some common

rules of thumb:

• Open-pollination: maximize mixture of the vines by reducing inter-row distance;

• Manual-pollination: increase distance in the row and between rows to be able to trace

plants and flag correctly the flowers.

Plants in crossing fields should be transplanted on plastic mulch with drip-irrigation; in

testing fields if seeds are available in high quantity the watermelons can be seeded directly.

It is very important to keep the weed population under control and to constantly scout the

fields for detecting eventual disease development or insects damages.

The most of the watermelon acreage in the U.S. is cultivated with plastic mulches and

drip–irrigation so that all the tests should be run in this fashion: however, in developing

cultivars for other countries this might be incorrect and the cultivation techniques should be

68

adjusted to the prevalent in the marketing-areas of the future cultivars (i.e. in Europe most of

the watermelons are cultivated in plastic-houses).

In the greenhouse the plants should be grown in pots or plastic disposable bags (3 gallons

usually is a good size) and plants have to be grown on trellises as single vine and trained and

pruned often. With wild accessions sometimes the fertile flowers are on the laterals so that

pruning should be delayed after fruit-setting. The lateral close to the set-flower should not be

pruned but tipped at the 2nd true-leaf. Tips are very delicate and their excision would

compromise the possibility to get flowers: therefore, training and pruning become very

delicated operations. Plants need a specific fertirrigation program depending on the soil-

mixture that is used.

Greenhouse nursery

69

Germination can be favored in cold seasons by using heating-mats under the flats

(soil–T=80°F) Plants are transplanted from 6x12 square-holes flats into the pots at the 1st

true–leaf stage.

Germination Nursery Transplanting in Pot

During the Winter and with cloudy weather in certain areas artificial illumination might be

useful to favor pollinations otherwise very difficult.

The greenhouse should be kept free from insects and pathogen: once a pathogen as been

established in a greenhouse is very challenging to sanitate the nursery completely and

sometimes the whole destiny of a breeding program depends on it. Chemical control of pests

should be as precise and accurate as possible and insect-damage should be prevented by

adding granular insecticide to the pot at transplanting and keeping yellow chromotropic traps

on every section of the greenhouse (they prevent white-flies, aphids, and thrips).

70

Once the fruits reach the size of a soft-ball they have to be suspended to the trellises to avoid

the breakage of the petiol and the loss of the fruit: a bag made of cheese–cloth might be the

best material to use.

Fruits into Cheese-Cloth Bags

Harvest and Seed-Treatments

The fruits should be harvested on a calendar base (about 30 days after pollination), surface-

sterilized with clorine, rinsed and cut. The flesh and the seeds should be left in single

buckets for 24 hours because fermentation helps to kill pathogens: if the fruit contains not

enough juice or if there is urgence to rinse and dry the seeds sooner than the next day,

fermentation can be helped with artificial enzymes.

71

One-day Harvest Fruit-Sterilization

Single-Fruit Extraction Flesh Fermentation

Fruits from the fields can be cut and seeds extracted as above for single-fruit extraction or

can be cut in pieces and dumped in a bulk-extractor for seed-increase purposes: it is

important to have a lot of fruits and with a lot of seeds per fruit because the bulk-extractor

wastes about a 10% of the seeds. The seeds need to be fermented in trash-cans for 24 hours

72

as above and then rinsed: a sluice (small version of the industrial types) might be very

helpful.

Bulk-Extraction in Process Sluice

After extraction seeds must be dried in an air-flow dryer for no-more than 12 hours

(T=80°F).

The seeds have to be kept free from heat and humidity: paper envelopes and wood sorters or

tissue-bags are the best containers and a cooler with controlled environmental conditions

would be the best long-term storage-facility (T=38°F and RH=25%). A broad-spectrum

fungicide (i.e. Captan) should be added in small quantity to the seeds to prevent formation of

molds and survival of pathogens.

73

Dryer Seed-Storage

Single-Fruit Seeds Bulk-seeds

Record Keeping

Every breeder usually has his own codification to identify his seeds so that it is impossible to

determine what is the best way to keep record of the genealogy of a nursery. In this chapter

is presented the technique used from several years by the Cucurbit Breeding Group at N.C.

State University: this method is in use since 1979 and has always given very good results.

74

Every generation (nursery cycle) is projected in advance on a pollination-plan that contains

all the informations about each experiment, each plot, and each cross- or self-pollination that

is needed.

• The pollination-plan reports also the seed-source from the previous generation (the

cross- or self-pollination number of the previous generation)

• Every year the plot-numbers start from 0 and are associated with the binumeric year

code

• Every plant (usually 1 plant=1plot) is marked with is plot-number which is later

reported also on the pollination tag as explained above

• At harvest seeds are counted and the number is recorded on the pollination-plan

75

Example of Pollination-Plan

76

The pollination-plan should be completed also with the calendar of the cycle (dates of

seeding, transplanting, pollination, and harvest).

With the new technologies now available the whole process could be made more efficient:

pollination-tags could be substituted by bar-codes and every technician could carry a hand-

held computer (i.e. Palm)connected to a wireless network to imput the date of pollination of

every other practice; the calendar of harvest could be calculated by a main database with

regards to different parameters per each cultigen, plot, greenhouse range, fertirrigation

program, etc. and then downloaded to each hand-held computer every day for the harvest.

The same bar-code could be printed on the seed-envelopes and the informations (seed-

number or other ratings) cold be imput directly in the main database with the hand-held

computers during fruit-cutting or seed-packaging.

The general data-base should report data of every introduction used in the breeding program:

it might indicate the general history of the introduction itself, the dates of introduction in the

breeding program, the codes of the experiment in which it has been included, and notes

concerning particular traits of interest or specific needs for cultivation, pollination and seed-

set.

77

Hybrid Seeds Production

Hybrid seeds production usually is not of any concern for the private breeder because

seed–companies have specific units that are devoted to the optimization of this operation.

Since most of the seed–companies produce their hybrid seeds in areas of the world where

labor is very cheap, they mostly rely on manual emasculation of the female parent and

tagging of the female pollinated flowers. Another technique often adopted for the same

reasons above is hand-pollination.

Incorporation of recessive phenotypic markers in the inbreds used as parents is an alternative

but it implies selection of the hybrid seedlings and it is not feasible in a commercial scenario.

Male-sterility would become of greater importance soon because the recent presence of

seed–borne diseases in the usual seed–production areas of the world are forcing the

seed–companies to move their production units in areas with greater labor costs.

Male–sterility in watermelon is genic and there as been no reports yet of cytoplasmic or

cyto–genic male–sterility: further research is needed in this area.

78

Sources of Information

Breeders need to remain up-to-date on every aspect of their crop. There several sources of

information available: the most important for the U.S. are listed below.

Table 14. Suggested sources of information for breeders Source Notes Meetings and ConferencesISHS World Meeting Every 4 yearsASHS National Meeting YearlyASHS Regional Meetings YearlyWatermelon Breeders Meeting YearlyCucurbitaceae Every 4 yearsEucarpia Every 4 years

Journals and PublicationsJ. of American Society for Horticultural ScienceAmerican J. of Plant PathologyCucurbit Genetics Cooperative ReportU.S.D.A. Agricultural Statistics

AssociationsNational Watermelon AssociationNational Watermelon Promotion BoardNational watermelon associations: go to (Maynard, 2001)

79

Conclusions and perspectives

The watermelon is a crop of always increasing interest and different types are now avaliable

on and requested by the american market. Future breeding efforts will be directed towards

both the creation of new types, particularly for nitsch markets, and the improvement of the

current types.

New types will have different flesh colors for the processing market such as yellow and

orange flesh fruits. Size and shape will be variable but more directed towards smaller sizes

for the fresh market and both elongated and round fruits, seeded and seedless. There are now

on the market some new very small watermelons (round, 3 to 5 lbs, striped, seedless) called

generally "Personal Size". Yellow watermelons are already successfully present on the

market but higher levels of sugars are needed and there is not yet an elongated yellow type

which will fit very well in the consumers needs for a fruit easy to carry and store. In general

elongated fruits will be more valuable if their ends will be flat instead of pointed, because

shippers report high damages to fruits due to contact with pointed ends of other fruits.

However, the major priority for breeding will be disease resistance with particular attention

to gummy stem blight, Fusarium wilt race 2, and powdery mildew. Second priority will be

breeding for rootstock production with particular attention to rootstocks resistant to

nematodes and soil–borne pathogens: from 2005 no more fumigations will be allowed in the

U.S. so that these parassites and pathogens will increase fastly to dangerous levels if

resistance will not be present in the cultivated plants. In absence of resistance the

watermelon will be cultivated with rotations of about 6 to 8 years.

From an academic prospective, a better understanding of the genetics of the species will be

necessary and biotechnology will become increasingly more important for a targeted strategy

80

in breeding for disease resistance. Populations used for genome mapping will be necessary

moved from the actual ones that includes 'NH Midget' as adapted parental line to new ones

including 'Allsweet' as a parent, since most of the commercial varieties in the U.S. are

related to the latter.

81

Literature Cited

Barham, W.S. 1956. A study of the 'Royal Golden' watermelon with emphasis on the

inheritance of the chlorotic condition characteristics of this variety. Proc. Amer. Soc.

Hort. Sci. 67:487-489.

Chambliss, O.L., H.T. Erickson, and C.M. Jones. 1968. Genetic control of bitterness in

watermelon fruits. Proc. Amer. Soc. Hort. Sci. 93:539-546.

Eigsti, O.J. 1979. Improvement of watermelons with polyploids. Cuc. Gen. Coop. Rep. 2:25-

26.

Gopal, Y.H., K.B. Reddy, and R. Srinivasulu. 1996. Studies on variability, heritability and

genetic advance in watermelon. New Bot. 23:125-127.

Hall, C.V., S.K. Dutta, H.R. Kalia, and C.T. Rogerson. 1960. Inheritance of resistance to the

fungus Colletotrichum lagenarium (Pass.) Ell. and Halst. in watermelons. Proc.

Amer. Soc. Hort. Sci. 75.

Henderson, W.R. 1989. Inheritance of orange flesh color in watermelon. Cuc. Gen. Coop.

Rep.:59-63.

Henderson, W.R. 1991. Gene List for Watermelon. Cuc. Gen. Coop. Rep. 14:129-138.

82

Henderson, W.R., S.F. Jenkins, and J.O. Rawlings. 1970. The inheritance of Fusarium wilt

resistance in watermelon, Citrullus lanatus (Thunb.) Mansf. Proc. Amer. Soc. Hort.

Sci. 95:276-282.

Henderson, W.R., G.H. Scott, and T.C. Wehner. 1998. Interaction of Flesh Color Genes in

Watermelon. J. Her. 89:50-53.

Huang, H., X. Zhang, Z. Wei, Q. Li, and X. Li. 1998. Inheritance of male-sterility and

dwarfism in watermelon [Citrullus lanatus (Thunb.) Matsum. and Nakai]. Scien.

Hort. 74:175-181.

Kanda, T. 1951. The inheritance of seed-coat colouring in the watermelon. Japan. J. Genet.

7:30-48.

Khandelwal, R.C., and P. Nath. 1978. Inheritance of resistance to fruit fly in watermelon.

Can. J. Genet. and Cytol. 20:31-34.

Kwon, Y., F. Dane, and Y.S. Kwon. 1999. Inheritance of light green flower color (gf) in

watermelon (Citrullus lanatus). Cuc. Gen. Coop. Rep. 22:31-33.

Layton, D.V. 1937. The parasitism of Colletotrichum lagenarium (Pass.) Ells. and Halst.

Iowa Agri. Exp. St. Ann. Bull. 223:37-67.

83

Levi, A. 2001. Biotechnology, p. 74-77, In D. N. Maynard, ed. Watermelons. Characteristics,

production, and marketing. ASHS.

Levi, A., and C.E. Thomas. 2001. Low genetic diversity indicates the need to broaden the

genetic base of cultivated watermelon. Hort Sci. 36:1096-1101.

Levi, A., C.E. Thomas, A.P. Keinath, and T.C. Wehner. 2001a. Genetic diversity among

watermelon (Citrullus lanatus and Citrullus colocynthis) accessions. Genet. Res. and

Cr. Evol. 48:559-566.

Levi, A., C.E. Thomas, X. Zhang, T. Joobeur, R.A. Dean, T.C. Wehner, and B.R. Carle.

2001b. A genetic linkage map for watermelon based on Randomly Amplified

Polymorphic DNA Markers. J. Amer. Soc. Hort. Sci. 126:730-737.

Lin, D., T. Wang, and Y.J. Wang. 1992. The effect of the branch less gene bl on plant

morphology in watermelon. Cuc. Gen. Coop. Rep. 15:74-75.

Liu, P.B.W., and J.B. Loy. 1972. Inheritance and morphology of two dwarf mutants in

watermelon. J. Amer. Soc. Hort. Sci. 97:745-748.

Love, S.L., and B.B. Rhodes. 1988. Single gene control of anthracnose resistance in

Citrullus? Cuc. Gen. Coop. Rep. 11:64-67.

84

Love, S.L., and B. Rhodes. 1991. R309, a selection of Citrullus colocynthis with multigenic

resistance to Colletotrichum lagenarium race 2. Cuc. Gen. Coop. Rep. 14:92-95.

Maynard, D.N. 2000. Watermelon cultivars in the United States in 2000. Cuc. Gen. Coop.

Rep. 23:51-53.

Maynard, D.N. 2001. Watermelon organizations, p. 197-198, In D. N. Maynard, ed.

Watermelons. Characteristics, production, and marketing. ASHS.

Maynard, D.N., K.R. Barker, G.E. Brust, and F.J. BDainello. 2001. Watermelons.

Characteristics, production, and marketing. I ed. ASHS.

Mc Kay, J.W. 1936. Factor interaction in Citrullus. J. Her. 27:110-112.

Mohr, H.C. 1953. A mutant leaf form in watermelon. Proc. Assoc. South. Agri. Work.

50:129-130.

Mohr, H.C. 1956. Mode of inheritance of the bushy growth characteristics in watermelon.

Proc. Assoc. South. Agri. Work. 53:174.

Mohr, H.C. 1986. Watermelon breeding, p. 37-66 Breeding vegetable crops. AVI Publishing

Co.

85

Mohr, H.C., and M.S. Sandhu. 1975. Inheritance and morphological traits of a double

recessive dwarf in watermelon, Citrullus lanatus (Thunb.) Mansf. J. Amer. Soc. Hort.

Sci. 100:135-137.

Navot, N., and D. Zamir. 1986. Linkage relationships of 19 protein coding genes in

watermelon. Theor. App. Genet. 72:274-278.

Navot, N., and D. Zamir. 1987. Isozyme and seed protein phylogeny of the genus Citrullus

(Cucurbitaceae). Plant Syst. & Evol. 156:61-67.

Navot, N., M. Sarfatti, and D. Zamir. 1990. Linkage relationships of genes affecting

bitterness and flesh color in watermelon. J. Her. 81:162-165.

Netzer, D., and C. Weintall. 1980. Inheritance of resistance in watermelon to race 1 of

Fusarium oxysporum f. sp. niveum. Plant Dis. Rep. 64:853-854.

Norton, J.D. 1979. Inheritance of resistance to gummy stem blight caused by Didymella

bryoniae in watermelon. Hort Sci. 14:630-632.

Perkins-Veazie, P., J.K. Collins, S.D. Pair, and W. Roberts. 2001. Lycopene content differs

among red-fleshed watermelon cultivars. Journal of the Science of Food and

Agriculture 81:983-987.

Poole, C.F. 1944. Genetics of cultivated cucurbits. J. Her. 35:122-128.

86

Poole, C.F., and P.C. Grimball. 1945. Interaction of sex, shape, and weight genes in

watermelon. J. Agri. Res. 71:533-552.

Poole, C.F., P.C. Grimball, and D.R. Porter. 1941. Inheritance of seed characters in

watermelon. J. Agri. Res. 63:433-456.

Porter, D.R. 1937. Inheritance of certain fruit and seed characters in watermelons. Hilgardia

10:489-509.

Provvidenti, R. 1992. Cold resistance in accessions of watermelon from Zimbabwe. Cuc.

Gen. Coop. Rep. 15:67-68.

Provvidenti, R. 1994. Inheritance of a partial chlorophyll deficiency in watermelon activated

by low temperatures at the seedling stage. Hort Sci. 29:1062-1063.

Ray, D.T., and J.D. Sherman. 1988. Desynaptic chromosome behavior of th gms mutant in

watermelon. J. Her. 79:397-399.

Rhodes, B. 1986. Gene affecting foliage color in watermelon. J. Her. 77:134-135.

Rhodes, B. 1999. Gene list for watermelon. Cuc. Gen. Coop. Rep. 22:61-77.

Rhodes, B., X. Zhang, V. Baird, H. Knapp, and X.P. Zhang. 1999. A tendrilless mutant in

watermelon: phenotype and inheritance. Cuc. Gen. Coop. Rep. 22:28-30.

87

Robinson, R.W., R. Provvidenti, and J.W. Shail. 1975. Inheritance of susceptibility to

powdery mildew in the watermelon. J. Her. 66:310-311.

Rosa, J.T. 1928. The inheritance of flower types in Cucumis and Citrullus. Hilgardia 3:233-

250.

Shimotsuma, M. 1963. Cytogenetical studies in the genus Citrullus. VII. Inheritance of

several characters in watermelon. Japan. J. Breed. 13:235-240.

Sidhu, A.S., J.S. Brar, and S.P.S. Gill. 1977. Mode of inheritance of earliness and quality

characters in watermelon (Citrullus lanatus (Thumb) mansf). J. Res. Punj. Agr. Un.

14:423-426.

Song, R., T.C. Wehner, and G. Gusmini. 2002. Screening the Watermelon Germplasm

Collection for Resistance to Gummy Stem Blight. Acta Hort. in press.

Sowell, G., B. Rhodes, and J.D. Norton. 1980. New sources of resistance to watermelon

anthrachnose. J. Amer. Soc. Hort. Sci. 105:197-199.

Strange, E.B., N. Guner, Z. Pesic-Van Esbroek, and T.C. Wehner. 2002. Screening the

watermelon germplasm colletion for resistance to Papaya Ringspot Virus type-W.

Crop Sci. 42:1324-1330.

88

Suvanprakorn, K., and J.D. Norton. 1980. Inheritance of resistance of anthracnose race 2 in

watermelon. Hort Sci. 15:277.

U.S.D.A. 2002. Statistics of vegetables and melons. Agricultural Statistics 1:IV.1-IV.38.

Vashistha, R.N., and B. Choudhury. 1972. Inheritance of resistance to red punpkin beetle in

muskmelon, bottle gourd and watermelon. Proc. 3rd Intl. Symp. Sub-Trop. and Trop.

Hort.:75-81.

Warid, A., and A.A. Abd el Hafez. 1976. inheritance of marker genes of leaf color and ovary

shape in watermelon. The Lybian J. of Sci. 6:1-8.

Watts, W.M. 1962. A marked male-sterile mutant in watermelon. Proc. Amer. Soc. Hort. Sci.

81:498-505.

Watts, W.M. 1967. Development of disease resistance and seed production in watermelon

stocks carrying the msg gene. Proc. Amer. Soc. Hort. Sci. 91:579-583.

Weetman, L.M. 1937. Inheritance and correlation of shape, size and color in the watermelon,

Citrullus vulgaris Schrad. Iowa Agri. Exp. St. Ann. Bull. 228:224-256.

Wehner, T.C. 2002. Vegetable cultivar descriptions for North America. List 26-2002. Hort

Science 37:15-78.

89

Wehner, T.C., N.V. Shetty, and G.W. Elmstrom. 2001. Breeding and seed production, p. 27-

73, In D. N. Maynard, ed. Watermelons. Characteristics, production, and marketing.

ASHS.

Whitaker, T.W., and I.C. Jagger. 1937. Breeding and improvements of Cucurbits, p. 207-232

Yearbook of Agriculture.

Wimmer, B., F. Lottspeich, I. van der Klei, M. Veenhuis, and C. Gietl. 1997. The

Glyoxysomal and plastid molecular chaperones (70-kDa heat shock protein) of

watermelon cotyledons are encoded by a single gene. Proc. Nat. Acad Sci. 94:13624-

13629.

Xu, Y., H.Y. Zhang, G.B. Kang, and Y.J. Wang. 2000. Studies on physiological and

biochemical characteristics and inheritance of chilling tolerance of seedlings of

watermelon wild germplasm. Acta Agr. Bor. Sin. 15:67-71.

Zamir, D., N. Navot, and J. Rudich. 1984. Enzyme polymorphism in Citrullus lanatus and C.

colocynthis in Israel and Sinai. Plant Syst. & Evol. 146:163-170.

Zhang, X.P., and M. Wang. 1990. A genetic male-sterile (ms) watermelon from China. Cuc.

Gen. Coop. Rep. 13:45.

90

Zhang, X.P., and B. Rhodes. 1992. Watermelon variety improvement in China. Cuc. Gen.

Coop. Rep. 15:76-79.

Zhang, X.P., H.T. Skorupska, and B. Rhodes. 1994. Cytological expression in the male-

sterile ms mutant in watermelon. J. Her. 85:279-285.

Zhang, X.P., B. Rhodes, V. Baird, and H.T. Skorupska. 1996a. A tendrilless mutant in

watermelon: phenotype and development. Hort Sci. 31:602.

Zhang, X.P., B.B. Rhodes, W.V. Baird, H.T. Skorupska, and W.C. Bridges. 1996b.

Phenotype, inheritance, and regulation of expression of a new virescent mutant in

watermelon: juvenile albino. J. Amer. Soc. Hort. Sci. 121:609-615.

Zitter, T.A., D.L. Hopkins, and C.E. Thomas. 1996. Compendium of cucurbit diseases, p. 27-

28. APS Press, St. Paul.