Origin of the dessert watermelon, Citrullus lanatus

H.S. Parisa
A.R.O. Newe Ya’ar Research Center, PO Box 1021, Ramat Yishay, Israel.

Abstract
Watermelons, Citrullus species, are native to Africa and have been cultivated
since ancient times. The fruit flesh of wild watermelons is watery, but typically hard-
textured, pale-colored and bland or bitter. The familiar sweet dessert watermelons, C.
lanatus, featuring non-bitter, tender, well-colored flesh, have a narrow genetic base,
suggesting that they originated from a series of selection events in a single ancestral
population. The objective of the present investigation was to determine where dessert
watermelons originated. Archaeological remains of watermelons, mostly seeds, that
date from 5,000 years ago have been found in northeastern Africa. An image of a large,
striped oblong fruit on a tray has been found in an Egyptian tomb that dates to at least
4,000 years ago. The Greek word pepon, Latin pepo and Hebrew avattiah of the first
centuries CE were used for the same large, thick-rinded, wet fruit which, evidently,
was the watermelon. Wild and primitive watermelons have been observed repeatedly
in Sudan and neighboring countries of northeastern Africa. The diverse evidence,
combined, indicates that northeastern Africa is the center of origin of the dessert
watermelon and that watermelons were domesticated for water and food there over
4,000 years ago. Next-generation ancient-DNA sequencing and genomic analysis offer
opportunities to rigorously assess the relationships among ancient and living wild
and primitive watermelons from northeastern Africa, modern sweet dessert
watermelons and other Citrullus taxa.

Keywords: archaeobotany, citron watermelon, colocynth, crop history, egusi watermelon

INTRODUCTION
The dessert watermelon, Citrullus lanatus (Thunb.) Matsum. & Nakai, is one of the
most refreshing and appreciated food items on hot summer days. With over 3,400,000 ha
planted and over 100,000,000 t harvested annually, watermelons are among the most widely
grown vegetable crops in the warmer regions of the world (Wehner, 2008).
The xerophytic genus Citrullus Schrad. is native to Africa (Whitaker and Davis, 1962).
Besides the dessert watermelon, other cultivated members of the genus include the citron
and egusi watermelons and the colocynth. Widely different beliefs have been expressed
concerning the progenitor and the center of origin of the dessert watermelon. I will here in
review these opinions and present evidence indicating a plausible alternative progenitor and
center of origin.

SOURCES OF EVIDENCE
A multidisciplinary approach that encompasses botany, horticulture, cookery,
philology and archaeology is necessary to best assess crop plant history (Dalby, 2003).
Independent lines of evidence have taken the form of artefacts, iconography and literature
(Paris, 2000; Janick et al., 2007). Modern distribution of wild relatives can also provide some
indication of where a crop was first cultivated.

DIAGNOSTIC FEATURES OF CITRULLUS

The Cucurbitaceae exhibit much parallel variation among genera and species.
Nevertheless, Citrullus is readily distinguished from other cucurbit genera by the pinnatifid
shape of its leaf laminae. Also, the flowers are borne singly in leaf axils, with a pistillate or

a
E-mail: hsparis@agri.gov.il

  Acta Hortic. 1151. ISHS 2017. DOI 10.17660/ActaHortic.2017.1151.15 87

Proc. V International Symposium on Cucurbits
  Eds.: M.L. Gómez-Guillamón et al.
hermaphroditic flower appearing at every seventh or eighth leaf axil and staminate flowers
occupying the remainder. The ovaries and primordial fruits are hairy, becoming smooth and
lustrous as they grow. The exocarp almost always is green and exhibits striping and the
stripes have acute, jagged edges. Immediately underneath the green exterior is the thick,
watery, pale green or white rind which is the mesocarp. This encloses the endocarp or fruit
flesh within which are distributed the seeds. The seed coats, depending on accession or
cultivar, can be brown, black, beige, yellow or other colors. Watermelons are often lumped or
confused with melons, Cucumis melo, in literature but the two differ sharply in the shape of
the leaf laminae, distribution of the flowers on the plant, glossiness and striping of the fruit
exterior, distribution of seeds within the fruit, and color of the seed coats.

Table 1. English names and equivalent specific, sub-specific, botanical-variety and cultivar-
group names (non-inclusive) of cultivated Citrullus.
Botanical variety Cultivar-
English name Species1 Subspecies (ssp.)
(var.)2 group(s)
Dessert watermelon C. lanatus (Thunb.) vulgaris (Schrad.) Fursa; lanatus; Dessert;
Matsum. & Nakai cordophanus Ter-Avan. colocynthoides Schweinf. Cordophanus
Citron watermelon C. amarus Schrad. lanatus citroides Bailey Citroides

Egusi watermelon C. mucosospermus mucosospermus Fursa mucosospermus Fursa Mucosospermus

(Fursa) Fursa
Colocynth C. colocynthis (L.) Schrad. -- -- --
1AfterChomicki and Renner (2015).
2There are very many designated botanical varieties, only four equivalent to the four listed subspecies are listed here.

NOMENCLATURE AND TAXONOMY OF CITRULLUS

Citrullus nomenclature and taxonomy have been highly confused. Various names at
differing taxonomical levels have been applied to cultivated Citrullus (Table 1). Moreover,
reported results of crossing among colocynths, dessert, citron and egusi watermelons
cannot always be relied on because of questionable taxonomic identity of one of the parents.
Not all bitter-fleshed Citrullus fruits are colocynths and not all hard-fleshed Citrullus fruits
are citron watermelons (Figure 1). Among the more reliable traits for distinguishing the
colocynths are the small size of their seeds, which are ≤8 mm length. Egusis lack seed coats.

Figure 1. Ripe fruits of Citrullus having hard, pale-colored, bitter flesh. Left, whole fruit and
center, equatorially cut fruit of citron watermelon PI 296341 (South Africa). Right,
whole and equatorially cut fruit of egusi watermelon PI 457916 (Liberia).
Young, developing citron and dessert watermelons can be distinguished by the
hairiness of the former (Figure 2). DNA sequencing and genomics offer reliable methods for
taxonomic identification of watermelon germplasm (Guo et al., 2013; Chomicki and Renner,
2015).

88

Figure 2. Young developing fruit of, left, the citron watermelon ‘Red-Seeded Citron’ and,
right, the dessert watermelon ‘Crimson Sweet’.
Three Citrullus species, C. ecirrhosus Cogn., C. rehmii De Winter and C. naudinianus
(Sond.) Hooker f., are not cultivated. C. colocynthis Schrad. is sparingly cultivated. The citron,
egusi and dessert watermelons have been variously treated as subspecies, botanical
varieties or cultivar-groups of C. lanatus, but Chomicki and Renner (2015), based on
genomic analysis, proposed that each be designated as separate species (Table 1). Indeed,
barriers to crossing among these three, although rather weak, have been documented (Levi
et al., 2003; Gusmini et al., 2004), which would justify treating them as separate species. The
citron watermelon is thus C. amarus Schrad., the egusi C. mucosospermus (Fursa) Fursa and
the dessert watermelon retains the name C. lanatus (Table 1) even though it is in fact less
lanate than the citron watermelon (Figure 2).

CITRULLUS FROM ANCIENT EGYPT AND SUDAN

Artefacts
De Vartavan and Asensi Amoros (1997) listed ten sites in Egypt, dating from the pre-
dynastic (5,000 years ago) and dynastic eras, that contain watermelon remains. A fruit with
seeds, from ca. 1500 BCE, was recovered from the foundation of a temple in Sudan (Van
Zeist, 1983). Schweinfurth (1883) identified foliage placed in a coffin, from ca. 1,000 BCE, as
of a desert ecotype, which he designated var. colocynthoides, of the dessert watermelon.
Germer (1988) listed remains, including seeds >4,500 years old and fruits >4,000 years old,
belonging to C. lanatus var. colocynthoides. Roman-era seeds have also been found in Egypt
(Cox and van der Veen, 2008). Photographs show the seeds as 10-11 mm long, with bulges
adjacent to the hilum, and therefore clearly not of C. colocynthis.

Iconography
Depictions of numerous ancient plants and fruits have been preserved in Egypt. One
depiction shows a round, striped fruit attached to a short section of vine that includes two
highly divided leaves, representing watermelon (Figure 3) (Keimer, 1924). Another shows
nine large, round, striped fruits in a basket (Feliks, 2005) and a third shows a large, oblong
striped from on a tray (Manniche, 1989).

89

Figure 3. Ancient images of watermelons from Egyptian tombs. Top, a spherical, striped fruit
with an adjacent section of foliage (Keimer, 1924). Center, various fruits, including
a basket of small, round, striped watermelons (Feliks, 2005). Bottom, an oblong
striped fruit on a tray (Manniche, 1989).

Literature
Five vegetables are mentioned together in the Hebrew Bible (Numbers 11:5), the
qishu’im (snake melons), avattihim (watermelons), hazir (leeks), bezalim (onions), and
shumim (garlics). The Hebrew word avattihim is probably derived from an ancient Egyptian
language root (Manniche, 1989).
Greek, Latin and Hebrew literature of Roman times, collectively, are rich in references
to watermelons although the descriptions of the fruits are meager. The pepon was described
by Greek-writing physicians as cold and wet. Dioscorides, in On Medical Matters (ca. 70 CE)
wrote that the rind of the pepon is to be applied on top of the head of a child suffering from
heat stroke. Quintus Gargilius Martialis, writing in Latin (ca. 260 CE), wrote that the pepone
were good to eat after removal of the rind and pits and Anthimus (ca. 516 CE) wrote that
they were to be eaten when fully ripened. In Hebrew-language codices of Jewish Law, the
avattihim (watermelons), in most instances, are considered together with other edible
cucurbits, the melafefonot (round melons), qishu’im (snake melons), and delu‘im (bottle
gourds). In one instance, though, they are considered together with figs, table grapes, and
pomegranates, which is the first literary implication of the existence of dessert watermelons
that were actually sweet.

DISCUSSION
Although it is agreed that the genus Citrullus is of African origin (Whitaker and Davis,
1962; Zohary et al., 2012), there has been major controversy concerning where in Africa the
dessert watermelon originated. Much of the disagreement stems from phenotypic variations
that overlap among citron, egusi and dessert watermelons and colocynths. Wild and
primitive Citrullus fruits, regardless of taxon, typically have hard, bitter or bland, weakly
colored flesh (Figure 1) and, as discussed by Wehner (2008), this situation has repeatedly
been a source of questionable taxonomic identifications. Thus, some conclusions drawn
from crossing experiments within Citrullus have been based on incorrect taxonomic
identities of the parents. Confused taxonomy together with the weakness of crossability

90
barriers among species of Citrullus have resulted in several hypotheses for the origin of the
sweet dessert watermelon.
One widely promoted hypothesis states that the dessert watermelon is descended
from the colocynth of northern Africa. When these two are crossed, though, the hybrid
plants exhibit chromosome irregularities and lower fertility (Whitaker and Davis, 1962;
Wehner, 2008). Results of investigations using techniques of molecular genetics and
genomics indicate that colocynths are more distant to dessert watermelons than are citron
watermelons (Levi et al., 2013; Chomicki and Renner, 2015).
A second, widely accepted hypothesis for the origin of the dessert watermelon is that
it is descended from the citron watermelon of southern Africa. However, the findings of
4,000-year-old artifacts of watermelon in Egypt are contradictory, as farming was not yet
underway in southern Africa at that time (Zohary et al., 2012). Massive preferential
segregation (Levi et al., 2003) occurs in progenies of crosses between citron and dessert
watermelons. Moreover, there are major differences in genome organization between the
two (Guo et al., 2013; Chomicki and Renner, 2015).
A third, newer hypothesis is that ‘the recent progenitor of modern cultivated
watermelon’ is the egusi watermelon of western Africa (Guo et al., 2013). Egusi and dessert
watermelons are sister species that diverged only 3.1 million years ago (Chomicki and
Renner, 2015). However, the egusi watermelons have peculiar soft, thick, moist,
mucilaginous seed coats not reported in any other Citrullus (Figure 1). Besides the
unidirectional crossing ability between dessert and egusi watermelons (Gusmini et al.,
2004), egusi watermelons have high nucleotide divergence of reproductive barrier genes
from the dessert watermelons, for which has been offered the tenuous explanation that ‘the
domestication of watermelon could be a possible force responsible for the rapid evolution of
reproductive barriers’ (Guo et al., 2013).
A fourth hypothesis is that the dessert watermelon originated and was first
domesticated in northeastern Africa. Accordingly, the wild, often hard, bitter or insipid
watermelons given the botanical variety name colocynthoides or subspecies name
cordophanus (Table 1) are living representatives of the ancestor of the sweet dessert
watermelon (Figure 4). Wild and “semi-cultivated” populations of watermelons are
widespread in semi-arid and arid Sudan (Andrews, 1950), the largest, apparently, occurring
in the Nile Valley (Mariod et al., 2009). They have been observed in the desert where they
are exploited by local residents for water and sustenance (Abdel-Magid, 1989). Their
distribution extends north into Egypt (Muschler, 1912).

Figure 4. Longitudinally cut and whole fruits of wild or primitive watermelons from
northeastern Africa, as grown in the USA. Left, PI 481871 (Darfur, Sudan); center,
PI 254622 (Kordofan, Sudan); right, PI 525084 (Qena, Egypt). Photographs by A.
Davis, USDA/ARS.
Unfortunately, Citrullus samples derived from Sudan and Egypt have not been
subjected to extensive molecular-genetic scrutiny. Results from DNA-sequence-based,

91
systematic investigations of Citrullus have been presented for only a few accessions from
Egypt and none from Sudan. Two of the Egyptian accessions, listed as United States plant
introductions PI 525081 and PI 525083, which had been previously thought, based on
phenotype, to be a colocynth and a citron, respectively, were observed by Levi et al. (2013)
to cluster with dessert watermelons. ‘Next-generation’ sequencing of ancient DNA (Brown et
al., 2015), if applied to comparing watermelon artefacts with modern Citrullus, can be
expected to identify the artefacts taxonomically. The latest sequencing and genomics
technologies (Levi et al., 2013; Guo et al., 2013; Chomicki and Renner, 2015) also should be
directed toward comparing the wild and primitive watermelons from northeastern Africa
with modern, sweet dessert watermelons as well as other Citrullus taxa. If applied to such
germplasm, the results can be expected to determine whether the dessert watermelon,
Citrullus lanatus, indeed exhibits a clear wild/domesticated dimorphism or merely a
continuum from wild to domesticated gene pools reflective of continual selection for
improved horticultural traits (Abbo et al., 2014).

Literature cited
Abbo, S., Pinhasi van-Oss, R., Gopher, A., Saranga, Y., Ofner, I., and Peleg, Z. (2014). Plant domestication versus
crop evolution: a conceptual framework for cereals and grain legumes. Trends Plant Sci. 19 (6), 351–360. PubMed
http://dx.doi.org/10.1016/j.tplants.2013.12.002
Abdel-Magid, A. (1989). Plant Domestication in the Middle Nile Basin (Oxford, UK: BAR International Series 523).
Andrews, F.W. (1950). The Flowering Plants of Anglo-Egyptian Sudan (Arbroath, UK: Buncle).
Brown, T.A., Cappellini, E., Kistler, L., Lister, D.L., Oliveira, H.R., Wales, N., and Schlumbaum, A. (2015). Recent
advances in ancient DNA research and their implications for archaeobotany. Veg. Hist. Archaeobot. 24 (1), 207–
214 http://dx.doi.org/10.1007/s00334-014-0489-4.
Chomicki, G., and Renner, S.S. (2015). Watermelon origin solved with molecular phylogenetics including
Linnaean material: another example of museomics. New Phytol. 205 (2), 526–532. PubMed
http://dx.doi.org/10.1111/nph.13163
Cox, A., and van der Veen, M. (2008). Changing foodways: watermelon (Citrullus lanatus) consumption in Roman
and Islamic Quseir al-Qadim, Egypt. Veg. Hist. Archaeobot. 17 (S1), 181–189 http://dx.doi.org/10.1007/s00334-
008-0164-8.
Dalby, A. (2003). Food in the Ancient World from A to Z (London: Routledge).
De Vartavan, C., and Asensi Amoros, V. (1997). Codex of Ancient Egyptian Plant Remains (London: Triade
Exploration).
Feliks, Y. (2005). Talmud Yerushalmi: Massekhet Ma‘asrot (Ramat Gan, Israel: Bar-Ilan).
Germer, R. (1988). Katalog der Altä gyptischen Pflanzenreste der Berliner Museeum. Agyptologische
Abhandlunger, Vol. 47 (Wiesbaden, Germany: Otto Harrassowitz).
Guo, S., Zhang, J., Sun, H., Salse, J., Lucas, W.J., Zhang, H., Zheng, Y., Mao, L., Ren, Y., Wang, Z., et al. (2013). The draft
genome of watermelon (Citrullus lanatus) and resequencing of 20 diverse accessions. Nat. Genet. 45 (1), 51–58.
PubMed http://dx.doi.org/10.1038/ng.2470
Gusmini, G., Wehner, T.C., and Jarret, R.L. (2004). Inheritance of egusi seed type in watermelon. J. Hered. 95 (3),
268–270. PubMed http://dx.doi.org/10.1093/jhered/esh031
Janick, J., Paris, H.S., and Parrish, D.C. (2007). The cucurbits of mediterranean antiquity: identification of taxa
from ancient images and descriptions. Ann. Bot. 100 (7), 1441–1457. PubMed http://dx.doi.org/10.1093/
aob/mcm242
Keimer, L. (1924). Die Gartenpflanzen im Alten Agypten (Hamburg, Germany: Hoffmann & Campe).
Levi, A., Thomas, C.E., Newman, M., Zhang, X., Xu, Y., and Wehner, T.C. (2003). Massive preferential segregation
and nonrandom assortment of linkage-groups produce quasi-linkage in an F2 mapping population of
watermelon. HortScience 38, 782.
Levi, A., Thies, J.A., Wechter, W.P., Harrison, H.F., Simmons, A.M., Reddy, U.K., Nimmakayala, P., and Fei, Z. (2013).
High-frequency oligonucleotides: targeting active gene (HFO–TAG) markers revealed wide genetic diversity
among Citrullus spp. accessions used for enhancing disease or pest resistance in watermelon cultivars. Genet.
Resour. Crop Evol. 60 (2), 427–440 http://dx.doi.org/10.1007/s10722-012-9845-3.
Manniche, L. (1989). An Ancient Egyptian Herbal (London: British Museum).

92
Mariod, A.A., Ahmed, Y.M., Matthä us, B., Khaleel, G., Siddig, A., Gabra, A.M., and Abdelwahab, S.I. (2009). A
comparative study of the properties of six Sudanese cucurbit seeds and seed oils. J. Am. Oil Chemists’ Soc. 86 (12),
1181–1188 http://dx.doi.org/10.1007/s11746-009-1459-3.
Muschler, R. (1912). A Manual Flora of Egypt (Berlin: R. Friedlaender.)
Paris, H.S. (2000). History of the cultivar-groups of Cucurbita pepo. Hortic. Rev. (Am. Soc. Hortic. Sci.) 25, 71–170.
Schweinfurth, G. (1883). The flora of ancient Egypt. Nature 28 (709), 109–114 http://dx.doi.org/10.1038/
028109a0.
Van Zeist, W. (1983). Fruits in foundation deposits of two temples. J. Archaeol. Sci. 10 (4), 351–354
http://dx.doi.org/10.1016/0305-4403(83)90072-9.
Wehner, T.C. (2008). Watermelon. In Handbook of Plant Breeding, Vegetables I, J. Prohens, and F. Nuez, eds. (New
York: Springer), p.381–418.
Whitaker, T.W., and Davis, G.N. (1962). Cucurbits (New York: Interscience).
Zohary, D., Hopf, M., and Weiss, E. (2012). Domestication of Plants in the Old World, 4th edn (Oxford, UK:
Clarendon Press).

93


94