Underexploited Germplasm Resources of the Cucurbitaceae

Deena Decker-Walters

The Cucurbit Network, P.O. Box 560483, Miami, FL 33256

Additonal index words. Cucurbitaceae, minor crops, Cucumis, Cucurbita, Citrullus, Lagenaria, Luffa, Momordica, Benincasa, Sechium

Abstract. In addition to the major crops of the Cucurbitaceae (cucumber, melon, watermelon, squashes/pumpkins), more than 20 minor crops are grown for food in localized areas of the globe. Even more species are being cultivated and studied for the pharmacological compounds they contain. Underexploited genera with greater economic potential include Lagenaria (for food, crafting); Luffa (food, sponges); Momordica and Trichosanthes (various bioactive compounds); Sechium and Benincasa (food industry uses); and Telfairia, Hodgsonia, Cucumeropsis, and Fevillea (edible and industrial seed oils).

Major crops of the Cucurbitaceae are cucumber (Cucumis sativus L.), melon (C. melo L.), watermelon (Citrullus lanatus (Thunb.) Masum. & Nakai.), and pumpkins, squashes, and ornamental gourds (Cucurbita pepo L., C. maxima Duch. ex Lam., C. moschata Duch. ex Poir., and to a lesser degree C. argyrosperma Huber). In addition to these taxa, 35 species in 22 genera, 7 tribes, and both subfamilies have been domesticated or cultivated on a smaller scale for use as food or medicine (Table 1). This paper discusses the minor cucurbit crops and their potential for improvement. The arrangement of genera is by relative commercial importance of the genus.


In addition to cucumber and melon, 30 species comprise the predominantly African genus Cucumis. Minor crops include bur gherkin (C. anguria var. anguria) and African horned cucumber (C. metuliferus), the fruit of which are edible, and an ornamental called hedgehog gourd (C. dispaceus). Other species (e.g., C. africanus L., C. kalahariensis Meeuse, C. prophetarum L.) are collected locally for use as food or medicine.

Bur gherkin plants are monoecious annuals bearing ovoid, heavily warted fruit on long pedicels. The mature fruit of var. anguria typically measures 5 cm long and 3.5 cm in diameter. Rind

color is green or yellow and sometimes striped. The spongy white flesh has a cucumber-like taste "with lemony nuances" (Baird and Thieret, 1988). Fruit are best eaten when immature, usually pickled or raw, but also as a cooked vegetable.

Wild populations (C. anguria var. longaculeatus Kirkbride) are native to southern Africa and possibly western Africa (Kirkbride, 1993). The species was apparently introduced to the Americas as a wild-collected food by African slaves during the 17th century. After domestication in the New World, bur gherkin was then carried back to West Africa where it is cultivated today (Burkill, 1985). In the U.S., Europe, and Australia, some naturalized populations of var. anguria stem from the commercial trade of the 19th and early 20th centuries. Only a handful of seed companies offer bur gherkin today.

According to Kirkbride (1988), germplasm for cultivar improvement should be sought in Angola; wild plants of that country most resemble those of the New World cultivars. Gene transfer from C. myriocarpus Naud. and possibly other closely related species may be avenues for improvement as well (Nijs and Visser, 1985).

Meeuse (1958) suggested that some strains of bur gherkin could compete economically with gherkin types of C. sativus because the former produce more fruit in a shorter growing period and possess greater resistance to some pests and

Cucurbitaceae '98

diseases. In the U.S., bur gherkin has been grown as far north as Idaho, with the first harvest of appropriately sized immature fruit taking place 2 months after seeding. In Kentucky, a single planted mound produced 100 mature fruit before the first killing frost (Baird and Thieret, 1988).

The annual monoecious vines of African horned cucumber produce a red to yellow fruit measuring 6 to 15 cm long and 3 to 6 cm in diameter and covered sparsely with stout spines. The flesh is light yellow, but the abundant placental material is dark green. Although sold as an edible fruit, the colorful, bizarre-looking fruit is probably purchased more often as a decorative. The flesh ranges from toxically bitter in wild plants to slightly acidic or insipid in the cultigen.

Cultivation and fruit consumption of this African native is sporadic on that continent. Large-scale commercial culture began in New Zealand under the trade name "kiwano" in 1982 (Morton, 1987), with fruit mostly exported to Europe, Japan, and the U.S.

Although African horned cucumbers have a

long storage life and ship well, research is ongoing to extend shelf life under humid conditions, improve flavor and aroma, enhance seed germination and fruit yield, and control rind color development (Benzioni et al., 1993). Current yields in Israel, with a field growing season of 3.5 months and a planting density of 10,000 plants per hectare, average 46 t·ha­1 of fruit, 60% of which are of export quality (Benzioni, 1997). Breeding investigations should examine wild populations as well as cultigen lines since both exhibit variability in several economically important characters.

Researchers are also trying to transfer some of the disease and nematode resistances in C. metuliferus to melon and cucumber (e.g., Norton and Granberry, 1980). African horned cucumber has been used as a resistant rootstock for these crops, but scientists have had a difficult time rescuing embryos produced from crosses with C. metuliferus (Nijs and Custers, 1990). Crossing studies, isozyme analyses, and DNA comparisons have shown that all three of these species are relatively isolated within the genus, with C.

Cucurbitaceae '98

metuliferus exhibiting a closer relationship to C. melo than to C. sativus.


In addition to the four squash crop species, Malabar gourd (Cucurbita ficifolia) is the only other domesticate in this New World genus of 15 species. Malabar gourd is relatively unique in the genus for its adaptation to high-elevation (up to 2800 m) cultivation in the neotropics. This cold-tolerant species has been cultivated in Mesoamerica and northern South America for several millenia; the earliest archaeological remains are dated 3000 B.C. from Peru. Malabar gourd was carried to the Old World during the 16th century, in part because the fruit could be kept without refrigeration for over a year, providing sustenance during lengthy voyages.

Despite hundreds of years of cultivation on various continents, there is limited genetic diversity in characters of the edible fruit. Even in the New World, the little variation that exists appears to occur throughout the species range (Andres,

1990). The oblong fruit measure 15 to 100 cm long; the rind is white, dark green, or a reticulated pattern of green on white; the bland-tasting flesh is white, coarse, somewhat dry, and fibrous; the seeds are typically black, but sometimes tan. In Latin America, immature fruit are prepared like summer squash, mature fruit are candied, and seeds are consumed raw, roasted, or ground. The flesh contains a protease with potential food industry value (Illanes et al., 1985). In Europe and Asia, C. ficifolia has been grown primarily as an ornamental or as fodder, respectively. More recently, C. ficifolia has been used successfully in The Netherlands, Japan, Korea, and China as a cold-hardy rootstock resistant to fusarium wilt for the winter production of greenhouse cucumber (Lee, 1994).

Because of the lack of variation, there are no distinct cultivars of C. ficifolia, although some lines have been bred to overcome the species' natural dependence on short days for flowering. Consequently, improvement should probably focus on gene exchange with other species. Partially fertile hybrids have been obtained through crosses with

Table 1, continued.

Cucurbitaceae '98

C. maxima, C. pepo, and C. lundelliana L.H. Bailey (Robinson and Decker-Walters, 1997). Andres (1990) suggests that breeding efforts should focus on transferring the disease resistances, long-keeping quality of the fruit, and cold-hardiness of C. ficifolia to the other domesticates, as well as transferring genes for higher carotene and soluble solids content in fruit from these species to Malabar gourd.

A potential domesticate is C. foetidissima, a native of semiarid regions in western North America. Although the herbaceous vines are frost sensitive, the enlarged perennial roots are cold- and drought-tolerant. Asexual reproduction occurs via adventitious rooting at the nodes. Sexual reproduction is prolific as well, with plants being monoecious or gynoecious.

Studies on the feasibility of domesticating wild C. foetidissima have been ongoing for 50 years, mostly in southwestern U.S., but also in Mexico, Australia, and Lebanon. Potential economic prod

ucts include the root starch, which has a digestibility of 95.6% when cooked; the seed oil, which contains 50% to 69% linoleic acid; the seed protein; and the prolific vines, which can by used as fodder, fuel, or as a cellulose source to make ethanol (Gathman and Bemis, 1990). More recently, buffalo gourd fruit and roots, which are high in cucurbitacin content, have been used to create baits for rootworm beetles of the tribe Luperini; the powdered root is laced with carbaryl insecticide and then sprayed onto corn fields (Comis, 1997). Another recent area of research is the use of the root as fuel in deforested underdeveloped countries.

Substantial genetic variation among buffalo gourd populations has already provided the basis for selection for seed and root yields, as well as for fatty acid composition in the seed oil (e.g., Bemis et al., 1978). Seed and fresh root yields have reached 800 to 1000 and 34,500 kg·ha­1, respectively (Gathman and Bemis, 1990). Although the root yield is considered commercially acceptable, the

Table 1, continued.

Cucurbitaceae '98

seed yield is not. Other current impediments to successful commercialization are discussed by Thompson (1990).

Citrullus and Praecitrullus

In addition to watermelon, there are three species in the African genus Citrullus, including C. colocynthis, a wide-ranging, xerophytic perennial. Colocynth has been an important medicine for millenia in Africa, the Middle East, and India. The 5- to 12-cm-wide fruit turn from green and white striped to more uniformly yellow upon maturity. The light-colored flesh is extremely bitter and high in a variety of bioactive and toxic compounds (e.g., colocynthin, elaterin, citrullol; Duke, 1985). The dried fruit pulp is powdered to prepare a strong cathartic and purgative and is quite flammable. Pharmacological studies are ongoing (e.g., Wasfi, 1994).

An isozyme study of Middle East populations

revealed two genetically distinct landracesone in Israel and the other in the Sinai desert (Zamir et al., 1984). In India and the Mediterranean, variation has been found for adaptation to differing soil conditions and for resistance to various diseases (Duke, 1983). A seed yield of 6700 kg·ha­1 has been reported, which translates to a possible seed oil yield of 3000 kg·ha­1 (Duke, 1983).

There is some confusion over the identity of plants called "egusi" that are not Cucumeropsis mannii; this cultigen of tropical West Africa, which is grown mainly for its edible seeds, either belongs to C. colocynthis (cf. Burkill, 1985) or represents bitter-fruited types of watermelon (e.g., C. lanatus var. citroides (Bailey) Mansf.). In either case, cultivars grown in Ghana (e.g., 'Akatewa', 'Nerri') and Nigeria ('Bara', 'Serewe', 'N', 'E') exhibit variation in seed color, size, and protein content (Norman, 1992), as well as other characteristics which may be of use in watermelon breeding.

Although species of Citrullus (n = 11) can be hybridized with varying degrees of difficulty,

Table 1, continued.

Cucurbitaceae '98

they cannot be crossed with one of their closest relatives, Praecitrullus fistulosus (n = 12; Navot and Zamir, 1987). Tinda is mostly cultivated in and apparently indigenous to northwestern India and Pakistan, even though wild populations have yet to be reported. Limited isozyme and morphological surveys indicate little genetic variation within this species, which may be of relatively recent domestication (Seshadri, 1986). The subspherical green fruit of the monoecious annual vine is typically picked when two-thirds grown (5 to 10 cm in diameter) and cooked as a vegetable, with culinary qualities similar to those of summer squash (Stocks, 1851). 'Arka Tinda' is an Indian selection adapted to early summer cultivation; fruit reach the edible stage five days after pollination (Seshadri, 1986).


Lagenaria siceraria, the only monoecious annual in a genus with five wild perennial species, has an ancient pantropical distribution. Archaeological remains suggest the independent use and possible cultivation of bottle gourd by 9,000 to 15,000 years ago in the Americas, 4000 to 5000 B.P. in Africa, and 6000 to 10,000 B.P. in eastern Asia.

Ocean currents apparently carried the buoyant bottle gourd fruit from the species' native home in Africa to parts of Southeast Asia and the New World.

This night-flowering species provides a useful hard-shelled fruit which grows in a variety of shapes and sizes, depending on the cultivar. Utilitarian uses of the dried gourds include fashioning them into food, drink, and medicine containers; culinary utensils; musical instruments such as drums, cordophones, rattles, and horns; birdhouses; penis sheaths; pipes; and more. Cultivars and landraces number in the hundreds, many with origins in tropical West Africa or India. For centuries, cultivars have been disseminated by European seed companies for home gardening. Today, thousands of gourd enthusiasts and crafters belong to the American Gourd Society, Inc., or one of its international counterparts.

Comparative studies of vegetative and reproductive plant parts (Heiser, 1973) indicate that landraces from Africa and the Americas (ssp. siceraria) are relatively distinct from those in Asia (ssp. asiatica (Kob.) Heiser). Genetic diversity exists not only in morphological characters, but in pest and disease resistance (Provvidenti, 1995) and earliness of flowering and fruit set (Martin,

Table 1, continued.

Cucurbitaceae '98

1979) as well. The fruit of some cultivars are not bitter (e.g., 'Cucuzzi', 'Longissima') and are prepared as a vegetable when young, particularly in eastern Asia and the Mediterranean. Selection for growing bottle gourd in specific environments is ongoing in India (Singh and Singh, 1998). Improved cultivars from the Philippines include the open-pollinated 'Tambuli' and the hybrid 'Dalisay' (East-West Seed Co., personal communication).

Hybridization with L. sphaerica (Sond.) Naud., a native of eastern Africa and Madagascar, has been accomplished (Jeffrey, 1967), making gene transfer for improvement of bottle gourd cultivars possible.


Two domesticated species occur in a genus of seven total species, four native to the paleotropics and three indigenous to the neotropics (Heiser and Schilling, 1990). Smooth loofah (Luffa cylindrica, syn. L. aegyptiaca Mill.) and angled loofah (L. acutangula) are monoecious, long-lived plants. In the domesticated varieties, the cylindrical fruit measure 15 to 60 cm long, are typically nonbitter, have a green rind with dry white flesh at the edible immature stage, and have a netted fibrous interior when mature. Fruit of angled loofah have ten longitudinal ribs. This species is mostly grown for its edible young fruit, eaten like a vegetable, whereas smooth loofah is cultivated primarily for the spongy interior of the mature fruit. Loofah sponges have been important commercial items for over a century, having been used as oil filters in steamships, padding in saddles and mattresses, insulation in helmets, and cleaning sponges (Porterfield, 1955).

Under growing conditions in Puerto Rico, loofah germplasm sources expressed variation in earliness, adaptation to the rainy season, longevity, and fruit production (Martin, 1979). Although recent trials have taken place in Israel (Huyskens et al., 1993), few improved cultivars (e.g., 'Hybrid Hercules' from Thailand) are available for angled loofah, with much of the breeding being done in India and focusing on yield, habitat adaptation, and disease and pest resistances (Singh and Singh, 1998). Fresh fruit yields in India reach 15 t·ha­1 under low-input conditions. The cultivar 'Satputia'

is unique in having hermaphroditic flowers. A cross between this cultivar and a normal monoecious cultivar of India called 'Tori' is said to produce five times the yield of either (Chakravarty, 1990).

Much more research is being conducted on smooth loofah. Although Japan was once the leading center for this research, today horticulturists in eastern Asia, including China and India, the U.S., and tropical West Africa are also working on crop improvement, while pharmacologists are studying the bioactive chemistry of the seeds (e.g., Kimura et al., 1997). Cultivars in China exhibit a diversity of characteristics open to selection. These cultivars fall into two general classes: a long-fruited, relatively smooth-skinned type (up to 200 cm long) and the common short-fruited type (20 to 60 cm long) which is more distinctly furrowed (Yang and Walters, 1992). Crop trials and breeding have focused on sponge quality as well as yield (e.g., Davis, 1994). Yields in Japan of 50 t·ha­1 have been reported (Martin, 1979). In the Philippines, improvement of the immature fruit as a vegetable is ongoing (East-West Seed Co., personal communication). Although the two domesticates have been successfully crossed (Dutt and Roy, 1990), interspecific gene transfer has not been used for cultivar improvement.


Although most of the 40 or so species of Momordica are restricted to Africa, a handful are strictly Asian (e.g., M. cochinchinensis and M. dioica), and a few are wild on both continents (e.g., M. balsamina). Most widespread is M. charantia, wild and domesticated forms of which can be found throughout the tropics worldwide. Momordica cochinchinensis (n = 14), M. dioica (n = 14), and M. balsamina (n = 11) are minor cultigens compared to M. charantia (n = 11). The first two species are dioecious tuberous perennials, whereas the latter are monoecious species with fibrous roots. Cochinchin gourd yields a red, 10 to 20 cm long, ovoid fruit covered with conical points, which is usually consumed immature as a vegetable. The small (3 to 8 cm long), softly spiny, yellow fruit of M. dioica is somewhat sweet-tasting. Momordica balsamina produces an orange fruit similar to that

Cucurbitaceae '98

of M. charantia (described below), but usually beaked and less warted.

Within the genus, aggressive crop improvement has only targeted M. charantia. This slender climber has aromatic yellow flowers; a fleshy, bumpy, spindle- or pear-shaped fruit that turns from green or white to orange, retains its bitterness, and dehisces at maturity; and tan seeds covered by sweet-tasting red arils. Fruit of wild and naturalized plants (var. abbreviata Ser.) are 5 cm long; those of domesticated plants can reach 25 cm long and are less prone to dehiscence. The immature fruit of the domesticate is cooked while immature, usually after steeping in salt water to lessen the bitter taste. All parts of wild and cultivated plants have medicinal uses. Because of their numerous bioactive principles, such as alkaloids (e.g., momordicine), saponins, the hypoglycaemic compounds p-insulin and charantin, abortifacient glycoproteins called momorcharins, and cucurbitacin-type triterpenes known as momordicosides (Neuwinger, 1996), bitter melon and related species are being vigorously investigated as treatments for diabetes, viruses, and other ailments (e.g., Raman and Lau, 1996). Researchers are also looking into the use of a seed enzyme, conjugated fatty acid synthetase, as a catalyst to improve the drying action of soybean oil in paints and other coatings (Lee, 1996).

Cultivars offered by U.S. seed companies reveal the Asian origins of the germplasm with names like 'Chinese Bitter Melon,' 'Thailand Bitter Melon,' and 'Bitter Gourd Taiwan Large,' 'Coimbatore Long,' and 'Hong Kong Bitter Melon.' At least three morphologically distinct cultivar groups are known from China (Yang and Walters, 1992), two from India (Chakravarty, 1990), and possibly several from Malaysia (Sands, 1928). The small-fruited group is sometimes referred to as var. muricata (Willd.) Chak. or var. pavel Crantz. Diversity in India has provided the basis for selection of fruit characters (size, shape, color, surface texture, maturation time), growth habit, seasonal adaptation, earliness, yield, and disease and pest resistances. Heterosis for some of these characteristics has led to the production and study of several F1 hybrids (e.g., Mishra et al., 1998). Other techniques being investigated for cultivar improve

ment include tetraploid production, mutation breeding via seed irradiation (Peter et al., 1998), and interspecific hybridization with M. dioica (Vahab and Peter, 1993) and M. balsamina (Singh, 1990).


Plants of domesticated wax gourd (Benincasa hispida var. hispida) are robust, monoecious annuals with large leaves, flowers, and fruit. Bee pollination of the yellow flowers produces a developing green fruit that starts out hairy and typically becomes waxy with age. The mild-tasting, easily digestible, juicy, white flesh is frequently cooked as a vegetable, curried, or candied (Morton, 1971). Fruit range from round to cylindrical and reach sizes of over 1 m long in some of the cylindrical cultivars. Wild populations (var. pruriens (Parkinson) Whistler) produce small, bitter, round fruit which are hard-shelled when mature (Whistler, 1990). Fruit of the cultigen also have a tough rind, which along with the waxy coating, allows for storage of 6 to 12 months. Because of its disease resistance, wax gourd is sometimes used as rootstock for grafted melon.

A variety of landraces have evolved locally in India, China, and Southeast Asia. These exhibit diversity with respect to fruit size, shape, hairiness, and waxiness; seed ridging; and earliness of flowering and fruit maturation. Two to four morphologically distinct groups of cultivars exist (e.g., Herklots, 1972). A preliminary isozyme study suggests that the group of cultivars called "fuzzy gourd" (sometimes classified as var. chieh-qua How) may be genetically diverse (Walters and Decker-Walters, 1989). This group is characterized by having a small, narrowly cylindrical fruit, which is harvested when young. Chinese cultivars are 'Beijing Yi-chuan-lin,' 'Ji-Ling Xiao-Dong-gua,' 'Chendu Wu-ye-zi,' and 'Nan-jing Yi-wuo-fong' (Yang and Walters, 1992). There are at least six large-fruited cultivars in China as well. A large-fruited type called 'Chinese Preserving Melon' was introduced to the U.S. in 1892 (Bailey, 1894). Since then, only a handful of cultivars have been offered by U.S. seed companies.

The limited research on this species has taken place mostly in China, India, and Taiwan (e.g., Wu

Cucurbitaceae '98

et al., 1987). However, western scientists are beginning to show more interest in wax gourd (e.g., Gubler and Bernhardt, 1992). Improved lines from India include Co. 1, which produces a medium-sized fruit with relatively few seeds, and Co. 2, an early maturing (120 days), small-fruited, high-yielding selection which also produces few seeds (Seshadri, 1986). Researchers in Thailand produced 'Hybrid Sapai Thong' and 'Hybrid 4121' using intraspecific hybridization (East-West Seed Co., personal communication). No cucurbit species is cross-compatible with the monotypic B. hispida. However, wild populations are easily hybridized with cultivars and may provide an additional source of disease and pest resistances.


Recent detailed studies of this New World genus have revealed eight (Newstrom, 1990) or more species (Lira et al., 1997), including two cultigens, one grown worldwide (Sechium edule) and the other only in Costa Rica (S. tacaco). Although both domesticates are cultivated primarily for their edible fruit, other parts of chayote are edible or otherwise useful.

Although wild plants of S. edule have bitter fruit, the white flesh of cultivated fruit tastes bland, sweet, or starchy, depending on the cultivar. A tremendous amount of genetic variation exists in other fruit characters as well. Fruit are typically pear-shaped, but may be round, ovoid, or elongated; fruit length (7 to 20 cm) and weight (100 g to 1 kg) are variable; rind color ranges from dark green to white; and the rind is relatively smooth or variously wrinkled and/or spiny. The single fleshy seed sprouts from within the fruit, sometimes while the fruit is still on the vine. Consequently, slightly immature fruit are picked for consumption. Chayote is a monoecious perennial, developing large succulent roots which are a good source of an easily digestible starch (20% to 25%) relatively free of toxins.

Chayote is a vigorous producer easily propagated from sprouted fruit or stem cuttings. It may flower year-round in the humid tropics and subtropics, but requires short days for flowering, thus limiting its cultivation at extreme latitudes. Although cultivated mostly between 500 and 1500

m, ecotypes have evolved that grow well at sea level or above 2000 m and in a variety of soils (Lira-Saade, 1995). Variation also exists in the numerous landraces of Mexico and Central America with regard to phenology, fruit yield, and number of crops per year (1 to 4). In contrast to landrace diversity, commercial growers in Latin America propagate a limited number of inbred cultivars. Only two commercial cultivars are exported from Mexico, Costa Rica, and Puerto Ricoa medium-sized, obovoid, light green, smooth fruit meant for food industry use and a small, round, white, smooth fruit, both with bland-tasting flesh (Newstrom, 1990). Crop improvement would benefit from sampling more of the indigenous genetic diversity, including gene exchange with wild populations of chayote and its close relatives S. compositum (J.D. Sm.) C. Jeffrey and S. chinantlense Lira & Chiang. Enhancement of fruit flavor and disease resistance could improve the crop's success. Interestingly, haploid chromosome number ranges from 12 to 14 in cultivated chayote, 12 to 13 in wild populations, and 14 in S. compositum (Lira-Saade, 1995).

The 6-cm-long ovoid fruit of S. tacaco tastes similar to that of chayote but is more fibrous. Limited studies indicate landrace variation in phenological and productivity characters. Lira-Saade (1995) suggests that improvement could also come from closely related species, particularly C. talamancense (Wunderlin) C. Jeffrey.


About 40 white-flowered, night-blooming species make up the principally Indomalesian genus Trichosanthes. Although many species are used in Chinese and Indian medicine (Yang and Walters, 1992), only a few, principally T. cucumerina (n = 11) and T. dioica (n = 11), are domesticated. Various species, but particularly T. kirilowii, are being studied (e.g., Savary and Flores) for the antitumor and antiviral compounds (e.g., trichosanthin, trichobitacin) they contain.

The mostly widely grown member of the genus is T. cucumerina var. anguina (L.) Haines, wild populations of which are classified as var. cucumerina. The slender immature fruit of this monoecious annual is typically eaten boiled. The young green- and white-striped fruit measures 30

Cucurbitaceae '98

to 180 cm long at the edible stage. The mature orange fruit of some cultivars are bitter. Average yield of edible fruit is 10 to 25 t·ha­1, but mutation breeding in India produced a selection (PKM-1) yielding almost 28 t·ha­1 (Peter et al., 1998). Mutation breeding and intervarietal F1 hybridization have been successful in selecting for modified fruit size and color, rind thickness, flesh texture, yield, earliness in flowering and fruit maturation, and content of punicic acid in the seed oil (Peter et al., 1998).

The dioecious perennial T. dioica produces oblong fruit to 12 cm long. Propagation is usually by stem or root cuttings to maintain favored female plants, and parthenocarpy can be induced in some cases with growth regulators. In India, a single planting of 90% females is maintained for two years, producing two crops, before being dug up (Chakravarty, 1990). Three selections from local landraces were chosen for commercial production in the 1980s: FP-1 (fruit round, green), FP-3 (spindle-shaped, green-striped), and FP-4 (spindle-shaped, light green; Singh, 1989). Little else has been reported on cultivar improvement, and rarely is this species cultivated outside of India, although it has been recently introduced to the U.S. (Lamberts, 1990). In summer trellis trials in Georgia, fruit production lasted from July to mid-October, with a mean seasonal fresh fruit yield of 0.5 kg/plant (Whitehead and Singh, 1998).


Ivy gourd (Coccinia grandis, syns. C. cordifolia Cogn., C. indica Wight & Arn., Cephalandra indica Naud.) is a wide-ranging, bird-dispersed taxon in an otherwise African genus of 30 species. Stem sections of this white-flowered, dioecious, tuberous perennial easily reproduce vegetatively. In some clones, the unfertilized ovaries of female plants can develop parthenocarpically, producing 5-cm-long, red, fleshy fruit. Fruit from wild populations are bitter and oblong­ellipsoid, whereas those of the domesticate are slightly acidic to sweet and more obtusely rounded (Jeffrey, 1980). Young berries are prepared in curries; ripe fruit are eaten raw. Various plant parts have been used medicinally for centuries in southern Asia; pharmacological investigations are ongoing (e.g.,

Platel and Srinivasan, 1997).

India is the primary site of commercial cultivation. Although several landraces have evolved in different parts of that country, little cultivar improvement has been attempted. A recent investigation (Josekutty et al., 1993) showed that tissue culture could be easily performed, providing a potentially valuable tool for crop improvement. Existing landraces and clones exhibit variation in fruit size, shape, color, and yield (Peter et al., 1998). Immature fruit yields average 10 to 13 t·ha­1 (Nath, 1975).


Two of the three species of this tropical African genus have been domesticated for food. Although the foliage and seeds of each are edible, T. occidentalis is grown in western Africa primarily for the nutritious stem tips and leaves (dry portion is 11% protein, 25% carbohydrates, 700 ppm iron; Akoroda, 1990), whereas T. pedata is cultivated, sometimes commercially, in eastern Africa mainly for its nutty-tasting seeds (30% to 60% oil, 30% protein). Both are hardy dioecious perennials producing large ribbed fruit up to 50 to 100 cm long and 20 to 25 cm in diameter. The flattened seeds measure 3.5 to 5 cm across. Plants can be propagated from whole seed, seed section, or vegetative cutting. These wet-season species prefer fertile, well-drained soils and possess a certain degree of drought tolerance, but, during the dry season, grow best when irrigated.

Most of the limited research on these species is being conducted on T. occidentalis in Nigeria, where local landraces and selections exhibit variation in leaf yield, vegetative maturation after planting, fruit size and color, seed size and color, disease resistance, and other characteristics (Norman, 1992). Okoli and Mgbeogu (1983) suggest that polyploidization and interspecific hybridization might offer opportunities for selection of desired traits (e.g., higher vegetative yield in T. occidentalis, the transfer of a thin seed coat from T. occidentalis to T. pedata). Selection for or environmental induction of monoecy may be possible in T. occidentalis (Akoroda et al., 1990). Research on the biochemistry of T. occidentalis seeds has revealed a T-cell mitogenic lectin, which may have pharmacological use (Togun et al., 1994).

Cucurbitaceae '98


The New World genus Cylanthera contains 35 to 40 species. Although a few wild species (e.g., C. brachybotrys (Poepp. & Endl.) Cogn., C. explodens) are sometimes locally cultivated for their edible fruit, the major domesticate is C. pedata, a relatively cold-tolerant monoecious annual. Outside of the New World, this species has been grown in eastern Asia, including mountainous Nepal, in northern New Zealand, and in English greenhouses (Herklots, 1972; National Research Council, 1989). Stuffing cucumber is so-called because the hollow, fleshy-rinded mature fruit is often deseeded, stuffed with other foods, baked, and then eaten. The teardrop-shaped fruit of the cultigen reaches 20 to 25 cm in length (with a rind thickness of 4 cm) 3 months after planting (Yamaguchi, 1979; Lira-Saade, 1995). It tastes somewhat like cucumber or artichoke. There are no reports of cultivar improvement, but the biochemistry of the seeds has been recently studied (De Tommasi et al., 1996).


Cucumeropsis is a monotypic genus native to tropical West Africa. Cucumeropsis mannii (syn. C. edulis (Hook. f.) Cogn.), locally called "egusi," is a monoecious annual adapted to humid forests. Flowers are small and yellow. The cream-colored fruit is ovate to cylindrical (up to 30 cm long ¥ 10 cm diameter), with white flesh and numerous whitish seeds up to 2 cm long. The seeds, tasting like groundnuts and containing 34% protein and 45% oil, are processed for oil extraction or are dried and ground into an edible paste added to soups (Burkill, 1985). The seed oil is semidrying, making it suitable for soap-making and illumination. Most of the very limited research on this species has targeted the seed's chemistry and its value for food or other uses (e.g., Gbenle and Onyekachi, 1995). In Ghana, Chakravarty (1968) documented three landraces differing in fruit shape. Okoli (1984) commented that the plants grown in Nigeria are very susceptible to root-knot nematodes, but fairly resistant to drought and waterlogging.


Hodgsonia is a monotypic genus which is distantly related to its closest neighbor, Trichosanthes. The single species, H. macrocarpa, is a stout, semi-woody, dioecious perennial bearing attractively fringed, yellow flowers and a large, roundish (15 cm long x 25 cm diameter), brown fruit with six or more oily seeds. Seeds from cultivated plants are baked and eaten as a delicacy in China and Malaysia (Burkill, 1935). The seed oil, representing up to 80% of seed weight, is used for cooking, medicine, and illumination (Hu, 1964). The oil tastes like lard and is high in vitamins A and D and the saturated palmitic and stearic acids (Hopkins, 1990). Cultivation trials have been conducted mainly in China, with experiments focusing on improving asexual propagation techniques and enhancing seed oil quality and yield (Chien, 1963). A single plant, with a life span of 70 or more years, can bear 40 to 100 fruit (and hence up to 10 kg of oil) annually (Hu, 1964). However, plants may not flower until the second year after seeding, which is a long time to wait to see if a plant is male or female.


This New World genus possesses two wild species, which appear indigenous to the Caribbean, and one domesticate (Sicana odorifera), which is cultivated sporadically throughout Latin America but unknown in the wild (Lira-Saade, 1995). Casabanana is believed to have been domesticated originally in eastern South America, possibly Brazil. It can be grown at altitudes of up to 1400 m, but requires high temperature for good fruit ripening (Morton, 1987). Although the fruit is edible (usually cooked when young or made into preserves when mature), just as often the plant is grown for its aromatic fruit and flowers. This monoecious perennial is propagated from seeds or stem cuttings. The yellow, red, green, or purple fruit are cylindrical, measuring 20 to 60 cm long and 8 to 12 cm wide. The sweet-smelling, slightly acidic, yellow flesh is tough but juicy. The brown seeds contain 39% oil and 25% protein (Lira-Saade, 1995). Landraces producing purple fruit have purple-tinted vegetation and seeds. No breeding work has been reported.

Cucurbitaceae '98

Other medicinals

In addition to the taxa discussed above, several more cucurbits have long histories of cultivation as medicinals. Those currently under commercial cultivation or being studied for greater exploitation are listed in Table 1. China and Japan are the major commercial producers and exporters of medicinal cucurbit products (Yang and Walters, 1992). One such medicinal is Siraitia grosvenorii, the dried fruit of which are made into a sweet-tasting tonic. Researchers (e.g., Kennelly et al., 1996) are studying the fruit's various glycosides, some of which are over 500 times sweeter than sucrose and might be useful as noncalorific sweeteners. In addition to isolating bioactive compounds from Gynostemma pentaphyllum and related species, a few Chinese publications describe agronomic trials on these important medicinals (e.g., Xiao et al., 1994). Research is also escalating on Chinese species of Hemsleya (e.g., Dinan et al., 1997). Meanwhile, European researchers have focused on the pharmacological compounds in Ecballium elaterium (e.g., Le-Nguyen et al., 1989) and species of Bryonia (Panossian et al., 1997). In the New World, species of Fevillea are being examined, not only for the bioactive compounds in the large seeds (e.g., Valente et al., 1993), but also for commercial production of the seed oil, which is suitable for food preparation and fuel (Gentry and Wettach, 1986).

Literature cited

Akoroda, M.O. 1990. Ethnobotany of Telfairia occidentalis (Cucurbitaceae) among Igbos of Nigeria. Econ. Bot. 44:29­39.

Akoroda, M.O., N.I. Ogbechie-Odiaka, M.L. Adebayo, O.E. Ugwo, and B. Fuwa. 1990. Flowering, pollination and fruiting in fluted pumpkin (Telfairia occidentalis). Scientia Hort. 43:197­206.

Andres, T.C. 1990. Biosystematics, theories on the origin, and breeding potential of Cucurbita ficifolia, p. 102­119. In: D.M. Bates, R.W. Robinson, and C. Jeffrey (eds.). Biology and utilization of the Cucurbitaceae. Cornell Univ. Press, Ithaca, N.Y.

Bailey, L.H. 1894. Some recent Chinese vegetables. Bul. 67, Cornell Univ. Agr. Expt. Sta., Ithaca, N.Y.

Baird, J.R. and J.W. Thieret. 1988. The bur gherkin (Cucumis anguria var. anguria, Cucurbitaceae). Econ. Bot. 42:447­451.

Bemis, W.P., J.W. Berry, C.W. Webber, and T.W. Whitaker. 1978. The buffalo gourd: a new potential horticultural crop. HortScience 13:235­240.

Benzioni, A. 1997. Kiwano. New crop factsheet of Purdue University. unpublished.

Benzioni, A., S. Mendlinger, M. Ventura, and S. Huyskens. 1993. Germination, fruit development, yield, and postharvest characteristics of Cucumis metuliferus, p. 553­557. In: J. Janick and J.E. Simon (eds.). New crops. Wiley, New York.

Burkill, H.M. 1985. The useful plants of west tropical Africa. Royal Botanic Gardens, Kew, U.K.

Chakravarty, H.L. 1968. Cucurbitaceae of Ghana. Bulletin de l'Institut Fondamental d'Afrique Noire 30A(2):400­468.

Chakravarty, H.L. 1990. Cucurbits of India and their role in the development of vegetable crops, p. 325­334. In: D.M. Bates, R.W. Robinson, and C. Jeffrey (eds.). Biology and utilization of the Cucurbitaceae. Cornell Univ. Press, Ithaca, N.Y.

Chien, H. 1963. "Lard fruit" domesticated in China. Euphytica 12:261­262.

Comis, D. 1997. Corn belt growers give areawide IPM a try. Agr. Res. 45(10):4­7.

Davis, J.M. 1994. Luffa sponge gourd production practices for temperate climates. HortScience 29:263­266.

De Tommasi, N., F. De Simone, and G. Speranza. 1996. Studies on the constituents of Cyclanthera pedata (caigua) seeds: Isolation and characterization of six new cucurbitacin glycosides. J. Agr. Food Chem. 44:2020­2025.

Dinan, L., P. Whiting, S.D. Sarker, R. Kasai, and K. Yamasaki. 1997. Cucurbitane-type compounds from Hemsleya carnosiflora antagonize ecdysteroid action in the Drosophila melanogaster BII cell line. Cell. Mol. Life Sci. 53:271­274.

Duke, J.A. 1983. Handbook of energy crops. unpublished.

Duke, J.A. 1985. CRC handbook of medicinal herbs. CRC Press, Inc., Boca Raton, Fla.

Dutt, B. and R.P. Roy. 1990. Cytogenetics of the Old World species of Luffa, p. 134­140. In: D.M. Bates, R.W. Robinson, and C. Jeffrey (eds.). Biology and utilization of the Cucurbitaceae. Cornell Univ. Press, Ithaca, N.Y.

Gathman, A.C. and W.P. Bemis. 1990. Domestication of buffalo gourd, Cucurbita foetidissima, p. 335­348. In: D.M. Bates, R.W. Robinson, and C. Jeffrey (eds.). Biology and utilization of the Cucurbitaceae. Cornell Univ. Press, Ithaca, N.Y.

Gbenle, G.O. and C.N. Onyekachi. 1995. Comparative studies on the functional properties of the proteins of some Nigerian oilseeds: Groundnut, soyabean and two varieties of melon seeds. Trop. Sci. 35:150­155.

Gentry, A.H. and R.H. Wettach. 1986. FevilleaA new oil seed from Amazonian Peru. Econ. Bot. 40:177­185.

Gubler, W.D. and E.A. Bernhardt. 1992. Cavity rot of winter melon caused by Verticillium dahliae. Plant Dis. 76:416­417.

Cucurbitaceae '98

Heiser, C.B. 1973. Variation in the bottle gourd, p. 121­128. In: B.J. Meggars, E.S. Ayensu, and W.D. Duckworth (eds.). Tropical forest ecosystems in Africa and South America: A comparative review. Smithsonian Institution Press, Washington, D.C.

Heiser, Jr., C.B. and E.E. Schilling. 1990. The genus Luffa: a problem in phytogeography, p. 120­133. In: D.M. Bates, R.W. Robinson, and C. Jeffrey (eds.). Biology and utilization of the Cucurbitaceae. Cornell Univ. Press, Ithaca, N.Y.

Herklots, G.A.C. 1972. Vegetables in Southeast Asia. George Allen & Unwin, Ltd., London, U.K.

Hopkins, C.Y. 1990. Fatty acids of Cucurbitaceae seed oils in relation to taxonomy, p. 38­50. In: D.M. Bates, R.W. Robinson, and C. Jeffrey (eds.). Biology and utilization of the Cucurbitaceae. Cornell Univ. Press, Ithaca, N.Y.

Hu, S.-Y. 1964. The economic botany of Hodgsonia. Econ. Bot. 18:167­169.

Huyskens, S., S. Mendlinger, A. Benzioni, and M. Ventura. 1993. Optimization of agrotechniques in the cultivation of Luffa acutangula. J. Hort. Sci. 68:989­994.

Illanes, A., G. Schaffeld, C. Schiappacasse, M. Zuñiga, G. González, E. Curotto, G. Tapia, and S. O'Reilly. 1985. Some studies on the protease from a novel source: The plant Cucurbita ficifolia. Biotech. Lett. 7:669­672.

Jeffrey, C. 1967. Cucurbitaceae, p. 1­157. In: E. Milne-Redhead and R. M. Polhill (eds.). Flora of Tropical East Africa. vol. 4. Whitefriabs Press Ltd., London.

Jeffrey, C. 1980. Further notes on Cucurbitaceae: V. The Cucurbitaceae of the Indian subcontinent. Kew Bul. 34:789­809.

Josekutty, P.C., S. Shah, and G. Prathapasenan. 1993. Direct and indirect organogenesis in Coccinia indica. J. Hort. Sci. 68:31­35.

Kennelly, E.J., R. Suttisri, and A.D. Kinghorn. 1996. Novel sweet-tasting saponins of the cycloartane, oleanane, secodammarane, and steroidal types, p. 1­12. In: G.R. Waller and K. Yamasaki (eds.). Saponins used in traditional and modern medicine. Plenum Press, New York.

Kimura, M., S.-S. Park, and R. Sakai. 1997. Primary structure of 6.5k-arginine/glutamate-rich polypeptide from the seeds of sponge gourd (Luffa cylindrica). Biosci. Biotech Biochem. 61:984­988.

Kirkbride, Jr., J.H. 1988. Botanical origin of the West Indian gherkin (Cucumis anguria L. var. anguria). Bot. Soc. Amer. (Abs.) 75(6):186.

Kirkbride, Jr., J.H. 1993. Biosystematic monograph of the genus Cucumis (Cucurbitaceae). Parkway Publishers, Boone, N.C.

Lamberts, M. 1990. Latin American vegetables, p. 378­387. In: J. Janick and J.E. Simon (eds.). Advances in new crops. Timber Press, Portland, Ore.

Lee, J. 1994. Cultivation of grafted vegetables. I. Current status, grafting methods, and benefits. HortScience 29:235­239.

Lee, J. 1996. Melon could speed soyoil drying. Agr. Res. 44(4):22.

Le-Nguyen, D., D. Nalis, and B. Castro. 1989. Solid phase synthesis of a trypsin inhibitor isolated from the Cucurbitaceae Ecballium elaterium. Inter. J. Pept. Protein Res. 34:492­497.

Lira, R., J. Caballero, and P. Dávila. 1997. A contribution to the generic delimitation of Sechium (Cucurbitaceae, Sicyinae). Taxon 46:269­282.

Lira-Saade, R. 1995. Estudios taxonómicos y ecogeográficos de las Cucurbitaceae latinoamericanas de importancia económica. Systematic and ecogeographic studies on crop genepools. vol. 9. Intl. Plant Genet. Resources Inst., Rome, Italy.

Martin, F.W. 1979. Vegetables for the hot, humid tropics. Part 4. Sponge and bottle gourds, Luffa and Lagenaria. Sci. Educ. Admin., USDA, New Orleans, La.

Meeuse, A.D.J. 1958. The possible origin of Cucumis anguria L. Blumea, Suppl. 4:196­205.

Mishra, H.N., R.S. Mishra, G. Parhi, and S.N. Mishra. 1998. Diallel analysis for variability in bitter gourd (Momordica charantia). Indian J. Agr. Sci. 68:18­20.

Morton, J.F. 1971. The wax gourd, a year-round Florida vegetable with unusual keeping quality. Proc. Fla. State Hort. Soc. 84:104­109.

Morton, J.F. 1987. Fruits of warm climates. J.F. Morton, Miami, Florida.

Nath, P. 1975. Coccinia for commercial cultivation. Financing Agr. 7(2):23­24.

National Research Council. 1989. Lost crops of the Incas: Little-known plants of the Andes with promise for worldwide cultivation. Natl. Acad. Press, Washington, D.C.

Navot, N. and D. Zamir. 1987. Isozyme and seed protein phylogeny of the genus Citrullus (Cucurbitaceae). Plant Syst. Evol. 156:61­67.

Neuwinger, H.D. 1996. African ethnobotany: Poisons and drugs. Chapman & Hall GmbH, Weinheim. Translation by A. Porter of a work first published as Afrikanische arzneipflanzen und jagdgifte, by Wissenschaftliche Verlagsgesellschaft, Stuttgart.

Newstrom, L.E. 1990. Origin and evolution of chayote, Sechium edule, p. 141­149. In: D.M. Bates, R.W. Robinson, and C. Jeffrey (eds.). Biology and utilization of the Cucurbitaceae. Cornell Univ. Press, Ithaca, N.Y.

Nijs, A.P.M. den and J.B.M Custers. 1990. Introducing resistances into cucumbers by interspecific hybridization, p. 382­396. In: D.M. Bates, R.W. Robinson, and C. Jeffrey (eds.). Biology and utilization of the Cucurbitaceae. Cornell Univ. Press, Ithaca, N.Y.

Nijs, A.P.M. den and D.L. Visser. 1985. Relationships between African species of the genus Cucumis L. estimated by the production, vigour and fertility of F1 hybrids. Euphytica 34:279­290.

Norman, J.C. 1992. Tropical vegetable crops. Arthur Stockwell, Ltd., Elms Court, U.K.

Norton, J.D. and D.M. Granberry. 1980. Characteristics of progeny from an interspecific cross of Cucumis melo with C. metuliferus. J. Amer. Soc. Hort. Sci. 105:174­180.

Cucurbitaceae '98

Okoli, B.E. 1984. Wild and cultivated cucurbits in Nigeria. Econ. Bot. 38:350­357.

Okoli, B.E. and C.M. Mgbeogu. 1983. Fluted pumpkin, Telfairia occidentalis: West African vegetable crop. Econ. Bot. 37:145­149.

Panossian, A., E. Gabrielian, and H. Wagner. 1997. Plant adaptogens. II. Bryonia as an adaptogen. Phytomedicine 4:85­99.

Peter, K.V., M.K. Sadhu, M. Raj, and K.P. Prasanna. 1998. Improvement and cultivation: bitter gourd, smake gourd, pointed gourd, and ivy gourd, p. 187­198. In: N.M. Nayar and T.A. More (eds.). Cucurbits. Sci. Publ., Inc.

Platel, K. and K. Srinivasan. 1997. Plant foods in the management of Diabetes mellitus: vegetables as potential hypoglycemic agents. Nahrung Food 41:68­74.

Porterfield, Jr., W.M. 1955. LoofahThe sponge gourd. Econ. Bot. 9:211­223.

Provvidenti, R. 1995. A multi-viral resistant cultivar of bottle gourd (Lagenaria siceraria) from Taiwan. Cuc. Genet. Coop. Rpt. 18:65­67.

Raman, A. and C. Lau. 1996. Anti-diabetic properties and phytochemistry of Momordica charantia L. (Cucurbitaceae). Phytomedicine 2:349­362.

Robinson, R.W. and D.S. Decker-Walters. 1997. Cucurbits. CAB Intl., Wallingford, U.K.

Sands, W.N. 1928. The bitter-cucumber or peria. Malayan Agr. J. 16(2):32­39.

Savary, B.J. and H.E. Flores. 1994. Biosynthesis of defense-related proteins in transformed root cultures of Trichosanthes kirilowii Maxim. var. japonicum (Kitam.). Plant Physiol. 106:1195­1204.

Seshadri, V.S. 1986. Cucurbits, p. 91­164. In: T.K. Bose and M.G. Som (eds.). Vegetable crops in India. Naya Prokash, Calcutta.

Singh, A.K. 1990. Cytogenetics and evolution in the Cucurbitaceae, p. 10­28. In: D.M. Bates, R.W. Robinson, and C. Jeffrey (eds.). Biology and utilization of the Cucurbitaceae. Cornell Univ. Press, Ithaca, N.Y.

Singh, K. 1989. Pointed gourd (Trichosanthes dioica Roxb.). Indian Hort. 33/34:33­37.

Singh, R.D. and J.P. Singh. 1998. Improvement and cultivation: Lagenaria and Luffa, p. 199­204. In: N.M. Nayar and T.A. More (eds.). Cucurbits. Sci. Publ., Inc.

Stocks, J.E. 1851. An account of the dilpasand, a kind of vegetable marrow. Hooker's J. Bot. Kew Gard. Misc. 3:74­77.

Thompson, A.E. 1990. Arid-land industrial crops, p. 232­241. In: J. Janick and J.E. Simon (eds.). Advances in new crops. Timber Press, Portland, Ore.

Togun, R.A., T. Animashaun, J.E. Kay, and A.A. Aboderin. 1994. A galactose-binding T-cell mitogenic lectin from the seeds of Telfairia occidentalis. Phytochemistry 35:1125­1130.

Vahab, M.A. and K.V. Peter. 1993. Crossability between Momordica charantia and Momordica dioica. Cuc. Genet. Coop. Rpt. 16:84­85.

Valente, L.M.M., A.A.L. Gunatilaka, T.E. Glass, and D.G.I. Kingston. 1993. New norcucurbitacin and heptanorcucurbitacin glucosides from Fevillea trilobata. J. Nat. Products 56:1772­1778.

Walters, T.W. and D.S. Decker-Walters. 1989. Systematic re-evaluation of Benincasa hispida (Cucurbitaceae). Econ. Bot. 43:274­278.

Wasfi, I.A. 1994. some phamacological studies on Citrullus colocynthis. J. Herbs Spices Med. Plants 2:65­79.

Whistler, W.A. 1990. The other Polynesian gourd. Pacific Sci. 44:115­122.

Whitehead, W.F. and B.P. Singh. 1998. Yield and fruiting span of tropical vegetable crop parwal in a temperate climate. HortScience 33:470 (abstr.).

Wu, C.-M., S.-E. Liou, Y.-H. Chang, and W. Chiang. 1987. Volatile compounds of the wax gourd (Benincasa hispida, Cogn) and a wax gourd beverage. J. Food Sci. 52:132­134.

Xiao, X.H., S.L. Chen, G.P. Yin, Z.L. Zheng, J.H. Wang, S.P. Le, and L. Le. 1994. Agronomic standardization of Gynostemma pentaphyllum (Thunb.) Makino. Chung Kuo Chung Yao Tsa Chih 19:270­272. In Chinese with English abstract.

Yamaguchi, M. 1979. World vegetables: Principles, production and nutritive values. AVI Publ. Co., Inc., Westport, Conn.

Yang, S.-L. and T.W. Walters. 1992. Ethnobotany and the economic role of the Cucurbitaceae of China. Econ. Bot. 46:349­367.

Zamir, D., N. Navot, and J. Rudich. 1984. Enzyme polymorphism in Citrullus lanatus and C. colocynthis in Israel and Sinai. Plant Syst. Evol. 146:163­170.

Cucurbitaceae '98