For decades the generally accepted view has been that Texas has no native tree sized palms north of the lower Rio Grande Valley. Sabal mexicana (syn. S. texana), a tall pahn found in Mexico and Central America, was believed to reach the northern limit of its native range along the lower Rio Grande, at the southernmost tip of the state. Sabal minor, the normally trunkless dwarf palmetto of East and South Central Texas, was assumed to be the only native palm found north of there. There were, however, two populations of tree-sized palms far north of the lower Rio Grande that had never been satisfactorily explained. For many years residents of Victoria, in South Central Texas, had cultivated S. mexicana, although no one seemed to know the origin of the population. Since 1941, when Houston botanist Robert A. Vines discovered them, botanists had wondered about some other palms, taller by far than any S. minor, hidden in a thick forest in Brazoria County, south of Houston.

Orator F. Cook (1913) examined the Victoria palms, which he suspected to have been brought from Mexico, and declared them to be a new species, Inodes exul. But at this time, before the assumption that Texas' native tall palms were confined to the lower Rio Grande Valley had become fixed, Cook (1908, 1913) also believed that S. mexicana, which he called Inodes texana, had once been found far north of the Rio Grande. In 1908 (p. 5, n. a) he had noted that "Tall palmettos were seen

in Jackson County as late as 1876 by Mr. J. D. Mitchell, of Victoria." Jackson County is east of Victoria, and 200 miles north of the lower Rio Grande.

After discovering the Brazoria County palms Vines sent specimens to Miriam Bombard, who (1943) identifed them as Sabal touisiana, a small, trunked palm known from Louisiana and Southeast Texas. But shortly thereafter Liberty Hyde Bailey (1944) classified S. louisiana as simply a caulescent form of S. minor. That Bailey was aware of the Brazoria. County palms is evident from his description of S. minor, in which he notes (1944, p. 387) that the species has a "short erect part" that "sometimes emerges into a trunk 23 or reported to 6 m. tall." Six meters is the trunk height Vines (1960) had reported for the tallest Brazoria County palm. Since we know of no S. minor even approaching this height, the reference apparently was to Vines' measurement. Neither Bombard nor Bailey ever visited the Brazoria County site, but on the basis of Bailey's work Texas botanists (except those, such as Vines, who still insisted they were S. louisiana) came to consider the Brazoria County palms to be trunked S. minor. Thus Correll and Johnston (1970, pp. 340-341) write: "According to Bailey, the conspicuously caulescent plants,. such as those found in Brazoria County, represent the optimum emergence of the species. Other than size, there seems to be no botanical difference between the dwarf acaulescent plants and those that develop a prominent trunk. The arborescent plants have been given the name S. louisiana." Consistent with this determination, in 1979 the Texas Forest Service measured the tallest Brazoria County palm at 27 feet, with a 16-foot crown spread and a 43-inch trunk circumference, and pronounced it the national champion Sabal minor. The palm is so listed in the big-tree registries of both the Texas Forest Service (1989, p. 8) and The American Forestry Association (1988, p. 16).

As will be shown, it was a mistake to lump the Brazoria palms in with caulescent S. minor. The botanical differences between them and S. minor, whether caulescent or acaulescent, are many. But there the matter rested, and might still rest, had not the the owner of the 43-acre tract on which most of the Brazoria County palms stand offered his land for sale. In March 1989, when I learned of this development, I contacted the Texas Nature Conservancy and the Texas Parks and Wildlife Department, urging acquisition of the tract. Since neither organization showed much interest, I set out to determine exactly what the Brazoria palms were. Could palms so large really be Sabal minor? If so, why had these particular trees become the giants of their species, towering over all others? I called Natalie Uhl of the Liberty Hyde Bailey Hortorium at Cornell. Uhl referred me to Dennis Johnson, of the International Union for the Conservation of Nature, who in turn referred me to palm taxonomist Robert W. Read, then retiring from the Smithsonian Institution. Read's immediate reaction to specimens and photographs was that, in view of their size and morphology, these palms could hardly be S. minor. But he refused to come to any conclusion without visiting the site, which he insisted on doing only in the fall, when the fruit would be ripe.

Suspecting, however, that the Brazoria palms might be an aberrant population of S. mexicana, Read began to look for evi

dence that S. mexicana was native to the Texas Central Coast. He soon found Cook's 1908 article, with its footnote reference to tall palmettos in Jackson County, as well as notes from Cook's files (now lodged at the Smithsonian) expressing Cook's belief as to the original range of S. mexicana. Hearing of our work, IPS member Dennis O'Connor, of Victoria, sent me a copy of Cook's 1913 article, in which Cook reiterated his belief regarding the former range of S. mexicana, and described "Inodes exul." Cook's "type individual of the new species" was a palm on the lawn of Dennis O'Connor's aunt, Mrs. Martin O'Connor. The palm, now over 40 feet tall, still stands. Inodes exul, however, was later shown by Davis (1942) to be S. texana, which Moore (1971) reduced to synonymy of S. mexicana.

Armed with this information, I started looking for historic references to tall palms north of the lower Rio Grande-and a living population. Within weeks I found both.

In 1685 La Salle established the first European colony in Texas, near Lavaca Bay on the Central Coast. In a careful and impressively accurate account appearing in Margry (1876-86, vol. 3, p. 212) La Salle's historian Henri Joutel describes the local flora and fauna. The account includes a two-sentence description of "trees" with leaves similar to those of the lataniers (fan palms), and bearing an edible fruit. My search for a living population aroused the interest of a reporter who interviewed me for an article in the Victoria Advocate. The day after the article appeared four fishermen called to tell of palms, up to 20 feet tall, along Garcitas Creek, the border between Victoria and Jackson Counties. According to Kathleen Gilmore (1984) archeological evidence proves that La Salle's colony, which was annihilated by Indians in 1688, was located on Garcitas Creek. Further evidence provided by Brownson Malsch (1988, and pers. comm.) indicates that during the last century the original Central Coast population of S. mexicana was almost entirely removed to meet a demand for pilings for wharves, and by transplant to Victoria. The Garcitas Creek population thus appears to be a younger generation, seeded by a few survivors. (A detailed account of this discovery, and a description of the population, appear in Lockett and Read 1990. Also see Fig. 1.)

What was the extent of the native range of S. mexicana? A deposition by survivors of the La Salle colony (in Weddle, 1987, pp. 228, 249) indicates that it probably extended northeastward from Garcitas Creek at least to the Lavaca River, and southwestward at least to the Guadalupe River. But there is also some interesting evidence from San Antonio, 120 miles northwest of Garcitas Creek. In 1716 Fray Isidro Felix de Espinosa, a priest accompanying a Spanish expedition led by Don Domingo Ram6n, reported seeing "palmitos legftimos" at San Antonio Springs, source of the San Antonio River, in what is today the city of San Antonio. (Espi

nosa's diary is on microfilm at the Latin American Collection of the University of Texas Library, Austin.) According to Ignacio Pifia Lujan (1972, p. 85), palmito is the common name for S. mexicana in the Huasteca region of northeastern Mexico. Espinosa was born in Querkaro, capital of the Mexican state of Quer6taro, which apparently has no S. mexicana, but borders on the Huasteca region.

Although one could argue that Espinosa saw S. minor, my own inspection of the remnants of forest in Brackenridge Park and on the Incarnate Word College campus, adjacent to what is left of San Antonio Springs and the present headwaters of the San Antonio River, revealed no S. minor, but many young S. mexicana, and one 30-foot specimen almost hidden in the trees. Since S. mexicana is commonly cultivated in the area there appears to be no way of knowing, without intensive historical research, whether the wild specimens are the progeny of imported transplants, or are descendants of a native population, which itself may have been relocated for landscaping. What we do know is that S. mexicana thrives in the area, while S. minor is lacking.

With it now established that S. mexicana is native to the Central Coast, in November 1989 Read visited the Brazoria County palm site, 60 miles up the coast from Garcitas Creek. After careful study of the palms there he believes, however, that they are not S. mexicana, but hybrids of S. mexicana and S. minor.

According to palm-hybrid specialist Michael J. Balick (1988, p. 29), "Sometimes, in a population comprising several related species (or even genera) of palms, variation that is beyond what is usually expected can be found. When the variation reveals a series of characters intermediate in nature between other members of the population, then hybridization can be suspected." Because Read's examination of the morphology of the palms at the Brazoria site revealed more specimen-to-specimen variation than he would expect in a single (nonhybrid) species at a single site, and because this variation is intermediate between S. minor and S. mexicana, he concluded they were probably hybrids of these two species. As such they would be, according to Balick (pers. comm.), the first known hybrids of Sabal species, and the only known naturally occurring palm hybrids in the United States. (Nauman 1990, reports the recent discovery of apparent intergeneric hybrids in Florida, but, unlike the Brazoria palms, these are sterile.) In his recent monograph of the genus Scott Zona (1990, p. 619) puts S. minor in a class apart from all other Sabal species. Likewise Bailey (1994, p. 383) considered S. minor to be the sole representative of a subgenus, "Eusabar'(correctly Saba~, which was distinct from Inodes, the subgenus to which he assigned all other Sabal palms. Thus the Brazoria palms would be hybrids of two very different species-one a temperate-zone, flatleaved dwarf and the other a tall palm, with highly costapalmate leaves, found

mainly in Mexico. This raises the interesting question of why the only hybrids of the genus would be between two species that, apparently, are not closely related.

Although the hybrid nature of the Brazoria palms is yet to be proven, the morphological and phytogeographic evidence is strong. Two large leaves from separate palms at the site exemplify the variation characteristic of hybrids. The leaf in Figure 2, whose blade measures 225 cm wide by 141 cm long, and has 65 segments, is relatively flat, with an asymmetric hastula. measuring 6 cm on the long side, and a 63-cm costa. In contrast, the leaf in Figure 3 is highly costapalmate. Although the blade size (211 cm wide by 142 cm long, with 68 segments) is similar to that of the first leaf, it has a 13-cm, symmetric hastula and an 86-cm costa. (See Fig. 4 for hastula contrast.)

The variation shown by the Brazoria palms is intermediate between the morphological characteristics of S. minor and S. mexicana. Zona (1990) gives hastula lengths for S. minor (p. 644) and S. mexicana (p. 639) as 0.8-4.7 cm and 9.515.5 cm, respectively. Read found hastulas of the Brazoria palms to be 4.5-13 cm long. Both Bailey (1944, p. 383) and Bombard (1935, p. 44) give 40 cm as the maximum costa length for S. minor (S. louisiana, in the case of Bornhard). Read found Brazoria costas to be 39-86 cm. long, or about 1/i blade length, while indicating that for S. mexicana the costa exceeds 1/i blade length. The Brazoria. costa in Figure 2 is less than 'A blade length, while that of Figure 3 exceeds 1/1 blade length. Bailey (1944, p. 387) indicates 16-40 segments per leaf for S. minor, while Zona (1990, p. 644) gives 15-65. Read found 46-68 for Brazoria leaves, and Zona (1990, p. 639) gives 80-115 for S. mexicana. And we now know, from the discovery of the Garcitas Creek S. mexicana population, that the ranges of S. minor and S. mexicana overlap in the Central Coast, since the range of S. minor extends well to the southwest of Garcitas Creek. Since the Brazoria palms would be hybrids of Texas' two native species of palm, and are endemic to Texas, in a paper Read and I are preparing we propose calling them Sabal xtexensis.

The Brazoria County site consists of approximately 30 trunked palms, plus an uncounted number of young (trunkless) specimens, scattered through a heavily forested area of about 60 acres. Although S. minor abounds at the site, forming the understory of the forest, as far as we know the nearest wild S. mexicana is on Garcitas Creek, 60 miles southwest of the hybrid site. The hybrids are apparently beginning to slowly disperse beyond the main concentration. We found one two miles away.

Although we know the hybrids are reproducing, we do not know whether they are mainly backcrossing with S. minor, or crossing with each other. To the extent they are backcrossing the S. minor population will be enriched by a genetic infusion from a much larger and very different species. This will help S. minor have the genetic variability it needs to adapt to changing conditions. To the extent the hybrids are crossing with each other they could eventually stabilize into a new species, perhaps a tall palm for the middle Gulf Coast, a region which has no such native palm. Either way the site constitutes a natural genetic laboratory where researchers can study an ongoing evolutionary process, in its ecological context. For this reason both Read and Balick (pers. comm.) stress the importance of protecting the Brazoria palms in the forest where they stand. Balick (1988, p. 30) writes: "Since hybrid progeny can develop into distinct species over time, it must be recognizied that these are distinct taxa worthy of conservation efforts." And also: "In the case of palms, much greater emphasis needs to be put on field studies of naturally occurring as well as disturbed populations in the wild in order to fully recognize the importance of hybridization in this family."

This dense bottomland hardwood forest, with an understory of S. minor, seems to have protected the palms from human intrusion, since few people, even in Brazoria County, are aware of them. As a remnant of a continuous hardwood forest that, according to Del Weniger (1984, p. 33), once covered most of two counties, it deserves protection in its own right. None of what remains of this unique coastal forest is now in a preserve.

For decades botanists have puzzled over the rare trunked form of S. minor, with Bombard (1943) noting that they appear to reach their greatest development in Louisiana and Texas. Testing S. minor from the Carolinas to East Texas, Paul Ramp (1989) found an east-west genetic chne. By studying leaf production in the crown and counting leaf scars down the trunk, Read estimates the tallest hybrid (27 feet) to be 150 years old. He does not, however, believe this pahn is necessarily the original hybrid. Since the original hybridization could have preceded this palm by many years, with backcrossing with S. minor already in progress, perhaps seed dispersal up the coast, toward the Mississippi delta, and inland, could explain Texas' and Louisiana's scattered populations of trunked S. minor. Although the morphological differences between trunked S. minor and the putative hybrids are clear, if backcrossing with S. minor has been long in progress genetic influence from the hybrid source could still he the cause of the trunked forms.

In his monograph Zona (1990, p. 645) writes: "Throughout its range, [S. minor] is a palm of the rich soils of floodplains, levees, riverbanks, and swamps. . . . "But, in the Hill Country of Texas, a limestone region with an average annual rainfall of 30 inches, and frequent drouths, I have found a population of S. minor growing amid rocks and cactus on a ridge top 2,000 feet above sea level. (See Fig. 5.) Since S. mexicana is a palm of relatively drier environments, (ses Zona, 1990, p. 640), Read believes the apparent ability of Hill Country S. minor to grow in a xeric environment could also be due to the influence of hybridization.

Conservation Efforts

The 43-acre tract on which most of the Brazoria palms stand is for sale. The availability of the tract presents, Read and I believe, a unique conservation opportunity. The Brazoria palms not only would be the only known hybrids of Sabal species, but the only naturally occurring and reproducing palm hybrids in the United States. Only 50 miles south of Houston, and over 200 miles north of the lower Rio Grande, they are, as far as we know, the northermnost natural population of tall palms west of Florida (Sabal palmetto) and east of Arizona (Washingtoniafilifera), and one of only two such natural populations north of the lower Rio Grande. Further, if the tract can be protected, prospects for successful conservation of the palms, and thus of the evolutionary event taking place, are excellent. They are healthy and reproducing, and slowly dispersing outward from the site. On the other hand, IPS member Mike Rayburn, a Brazoria County rancher, believes that if the tract is sold it will probably be cleared, since there would be no economic use for the land in its present densely wooded state. Most of the remaining palms, outside the 43-acre tract, stand on two adjoining tracts of 12 and 3 acres each. Ideally these tracts would eventually be added to the preserve.

A local conservation organization is now receiving donations toward purchase of the tract. Those wishing to help may send checks, payable to the "BNCAP Palm Fund," to the Brazos port Nature Center and Planetarium, P.O. Box 1464, Lake Jackson, TX 77566. Also please call or write the following persons, urging them to protect the Brazoria County palm site:

DAVID BRAUN, Director, Texas Nature Conservancy, P.O, Box 1440, San Anto

nio, TX 78295-1440. Tel. 512/2248774.

JOHN C. SAWHILL, President, The Nature Conservancy, 1815 North Lynn Street, Arlington, VA 22209. Tel. 703/ 841-5300.

ANDREW SANSOM, Director, Texas Parks and Wildlife Department, 4200 Smith School Road, Austin, TX 78744. Tel. 512/389-4800.

Acknowledgments