Cucurbit Genetics Cooperative Report 7:73-75 (article 32) 1984

Natural Hybridization of Wild Cucurbita sororia Group and Domesticated C. mixta in Southern Sonora, Mexico

Merrick, Laura C.

Cornell University, Ithaca, NY 14556 and University of California, Davis, CA 95616

Gary P. Nabhan

Native Seeds/SEARCH, and University of Arizona, Tucson, AZ 85721

Despite numerous experimental hybrids that have been made between Cucurbita species to determine taxonomic relationships among taxa in this genus, evidence of uncontrolled hybrids between taxa within their natural ranges has rarely been reported (1). The following observations and ethnobotanical interviews suggest that C. mixta land races long-cultivated by Sonoran Indians may continue to exchange genes with wild or weedy Cucurbita native to northwest Mexico.

In Sinaloa and adjacent Sonora and Chihuahua, botanists have collected numerous herbaria specimens of a wild Cucurbita belonging to the C. sororia group as it is defined by several authors (2,3). This group is supposedly represented by C. palmeri in the north, but their leaves are not as deeply lobed as C. palmeri described by Bailey (4). Sonoran specimens have sometimes been referred to as the more southerly C. sororia. Merrick (work in progress) is finding high seed set in hybrids made between various species in this group. If additional genetic and biogeographic work indicate that these taxa form one highly interfertile but variable, widespread compilospecies, the name C. radicans Naud. (4) has precedence. These taxa are closely clustered with C. mixta in numerical taxonomic studies and, as a group including C. mixta, are distinct from other Cucurbita species (3,5). Supporting evidence from crossing studies (Merrick, unpublished data) indicates fertile hybrids can be produced in controlled reciprocal crosses between C. mixta and C. sororia group taxa collected throughout the geographical range of these species in Mexico. More restricted crossability was found in experimental crosses between C. mixta/C. sororia group taxa and other Cucurbita wild and cultivated species. In native habitats, the flowering phenology, basic floral morphology, and pollinators of neighboring wild and cultivated Cucurbita are often the same, resulting in potentially suitable conditions for gene flow. Have C. sororia group taxa possibly contributed genes to C. mixta during its evolution under domestication?

Ethnobotanical evidence from areas of native agriculture in Northwest Mexico suggests occasional gene exchange. In August 1970, ethnobiologist Campbell Pennington collected seed of "calabasa caliente" in the Mountain Pima Indian locality of Yecora, Sonora. Dr. Thomas Whitaker gave them a preliminary i.d. of C. mixta "Taos type". Upon growout as Native Seeds/SEARCH #EO1-015, its seed produced round, nearly softball size fruit with extremely bitter flesh, and a peduncle less corky or enlarged than most C. mixta. Expression of the dominant Bi gene for cucurbitacin content, intermediate fruit size and relatively thin peduncle suggest that this may represent progeny of a C. mixta X C. sororia group hybrid from Sonora. Drs. Whitaker and Bemis have viewed the fruit and concurred with this hypothesis; progeny will continue to be evaluated.

While collecting plants in southern Sonora, each of us has encountered other possible evidence of such introgression. In September 1982, L.M. visited the garden of Indian Valentin Sasueta in the Mayo-Warihio community of San Bernardo. From his fence, she harvested fruits (#293) which matched the description given above for #EOI-015. Two other accessions (#294A and #294B) with relatively large fruit were obtained at a nearby airport landing strip not far from cultivated fields in the Alamos municipality. Grown out in Davis, #294B appeared to be an "escaped" C. mixta. It produced large fruits with non- bitter flesh, thickened peduncles, large seeds and C. mixta-type flowers. In Davis, plants of #293, #294A, and Sonoran accessions #EO2-001 and #EO2-003 from C.N. produced fruits of intermediate size (compared to over 90 accessions of C. mixta and over 40 accessions of wild C. sororia group taxa), bitter flesh, enlarged peduncles and relatively large C. mixta-like foliage and stem characteristics. The seeds of these accessions resembled those of the C. mixta "green striped cushaw type" (6) land races cultivated in Northwestern Mexico (seed body shiny white, seed margin tan), but were distinctly smaller. The preceding characters match those found in experimentally produced C. mixta/C. sororia group taxa hybrids. The bitter Sonoran accessions exhibited variation between and within accessions for fruit size, shape, and coloration; floral characteristics; seed size and shape; and leaf coloration. These accessions have traits similar to those reported by Whitaker and Gentry (pers. comm.) as wild C. mixta from the Warihio Indian region around Alamos, and are often called "Chi Chi Coyote" or "Calabasa de Coyote" in Spanish, and "Ha'la'we Chipu" (Bitter Squash) in the native Warihio tongue (Eric Rowell, pers. comm.). Fruits of #293, #294A, and #294B were called "Chi Chi Coyote" by San Bernardo residents.

In August 1983, G.N. was accompanied by Mayo Indian José Valenzuela into the sierras between San Bernardo, Sonora and the Chihuahua border. At the edge of a slash-and-burn field, he pointed out a wild Cucurbita which he said was very bitter. He then volunteered in Spanish: "It is called Chi Chi Coyota. If Chi Chi Coyota is on the edge of your fields, and you plant squashes, you are going to lose them (for use) because the bad will enter the good (squashes)." The trouble, he added, is that it is difficult to tell this weed from the domesticated plant until the fruit begins to mature. He said that only the burros will try to eat its bitter gourds.

The folk term "Chi Chi Coyote" or its variants are used as well for the xerophytic wild Cucurbita of Northwest Mexico. In Baja California Norte, in the Sierra Juarez, a Mexican told G.N. that this was because they are sometimes used by mothers to discourage their children from further nursing. To wean a child, the cucurbitacin-rich, wet pulp of maturing fruit is rubbed on the woman's breast. After one "surprise", the child no longer wants to nurse. Perhaps this folk name is derived from "Coyote's Breasts" (breasts=chichis in Mexican vernacular), the Coyote being a supreme trickster in New World folklore. Sonoran and Arizona Indians associate Coyote with other wild relatives of their crops, including species of Nicotiana, Proboscidea, Gossypium, Phaseolus, and even Cucumis (C. angularis) (7). Their recognition of the close affinity of C. sororia group taxa with C. mixta predates that of Western scientists. Further studies will hopefully confirm this genetic introgression hypothesis and possibly establish that C. mixta evolution is non-centric, or diffuse in the geographical area from which it evolved and diverged from wild Cucurbita progenitors. Since southern Sonora borders on the true Sonoran Desert, this gene pool may be a source of genes conferring drought and heat resistance of use in the improvement of stress tolerance in C. mixta and C. pepo.

Literature Cited

1. Bemis, W.P. and T.W. Whitaker. 1965. Natural hybridization between Cucurbita digitata and C. palmata. Madrono 18:39-47.
2. Hurd, P.D., Jr., E.G. Linsley and T.W. Whitaker. 1971. Squash and Gourd Bees Bees (Peponapsis, Xenoglossa) and the origin of the cultivated Cucurbita. Evolution 25:218-234.
3. Rhodes. A.M., W.P. Bemis, T.W. Whitaker and S.G. Carmer. 1968. Numerical taxonomic study of Cucurbita. Brittonia 20:251-266. Bailey, L.H. 1943. Species of Cucurbita. Centes Herb.VI (V):266-322.
4. Bemis, W.P., A.M. Rhodes, T.W. Whitaker and S.G. Carmer. 1970. Numerical taxonomy applied to Cucurbita relationships. Amer Jour. Bot. 57:404-412.
5. Cutler, H.C., and T.W. Whitaker. 1956. Cucurbita mlxta Pang.: Its classification and relationships. Bull. Torr. Bot. Club 83:253-260.
6. Nabhan, G.P. 1982. The Desert Smells Like Rain: A Naturalist in Papago Indian Country. North Point Press, Berkeley CA.