XB ISSN 0038-3872 SOUP ERN CALIFORNIA ACADEMY OF SCIENCES BOULLETIN Volume 86 Number 1 BCAS-A86(1) 1-56 (1987) APRIL 1987 Southern California Academy of Sciences Founded 6 November 1891, incorporated 17 May 1907 © Southern California Academy of Sciences, 1987 OFFICERS Robert G. Zahary, President Charles E. Oxnard, Vice-President Camm C. Swift, Secretary Takashi Hoshizaki, Treasurer Jon E. Keeley, Technical Editor Gretchen Sibley, Managing Editor BOARD OF DIRECTORS 1985-1987 1986-1988 1987-1989 Michael H. Horn Daniel M. Cohen Larry G. Allen Margaret C. Jefferson Takashi Hoshizaki Hans M. Bozler Susanne Lawrenz-Miller Susan S. Lieberman Peter L. Haaker John D. Soule Camm C. Swift Charles E. Oxnard Gloria J. Takahashi Robert G. Zahary June Lindstedt Siva Membership is open to scholars in the fields of natural and social sciences, and to any person interested in the advancement of science. Dues for membership, changes of address, and requests for missing numbers lost in shipment should be addressed to: Southern California Academy of Sciences, the Natural History Museum of Los Angeles County, Exposition Park, Los Angeles, California 90007. Annual:Meéembers: 20S ee ee I ee a $ 20.00 Student Membersy nc) 3 soe a ee ee 12.50 Life Members Fellows: Elected by the Board of Directors for meritorious services. The Bulletin is published three times each year by the Academy. Manuscripts for publication should be sent to the appropriate editor as explained in “Instructions for Authors” on the inside back cover of each number. All other communications should be addressed to the Southern California Academy of Sciences in care of the Natural History Museum of Los Angeles County, Exposition Park, Los Angeles, California 90007. Date of this issue 9 April 1987 Bull. Southern California Acad. Sci. 86(1), 1987, pp. 1-12 © Southern California Academy of Sciences, 1987 Rediscovery of Radiocentrum avalonense (Hemphill in Pilsbry, 1905) (Gastropoda: Pulmonata) F. G. Hochberg, Jr.,! Barry Roth,? and Walter B. Miller? '.2Department of Invertebrate Zoology, Santa Barbara Museum of Natural History, 2559 Puesta Del Sol Road, Santa Barbara, California 93105 3Department of Ecology and Evolutionary Biology, University of Arizona, Tucson, Arizona 85721. Abstract.—The land snail species originally described as Oreohelix avalonensis Hemphill in Pilsbry, 1905, was rediscovered on Santa Catalina Island, California, after being “‘lost’’ for nearly 80 years since its original collection. It is narrowly distributed on south-facing slopes on the southeastern part of the island, in as- sociation with a coastal sage scrub plant community. Dissection of the reproduc- tive system confirms placement in the oreohelicid genus Radiocentrum. Disjunc- tions in the range of Radiocentrum correspond to the arid environments of the Sonoran and Chihuahuan deserts and probably date from late Pleistocene time. A lectotype is designated for the species. In January or February of 1902 the conchologist Henry Hemphill visited Santa Catalina Island, California, and collected a large series of live oreohelicid snails, presumably in the vicinity of the town of Avalon. From this original collection Hemphill distributed specimens to collectors and museums throughout the coun- try, many under the name “Helix alternatus var. avalonensis.”’ The find was first reported by Stearns (1902) and the snail later described by Pilsbry (1905), quoting a manuscript of Hemphill’s, as Oreohelix avalonensis. Over the years many prom- inent malacologists have visited Catalina Island in search of this snail. Repeated failure to find it led to the suggestion that it might have become extinct (Bequaert and Miller 1973), perhaps through over-collecting (Smith 1970) or expansion of the settlement of Avalon. Because Catalina Island is well outside the range oth- erwise known for the Oreohelicidae, some workers suspected a mixup of locality labels, which has been known to happen with Hamphill’s material (Roth 1982). During a field survey of land mollusks on Catalina Island in December 1978, the senior author (FGH) unexpectedly found recently dead shells of an Oreohelix- like snail on a steep, south-facing slope about 4 km south of Avalon. No living snails were located in a brief survey of the area, nor on a second reconnaissance in 1979. A third trip was made to the same locality in January 1982 by all three authors. On the second day of field work, during a heavy, early-morning rainstorm, the junior authors (BR & WBM) discovered six live snails in an area which had been thoroughly examined the day before. Subsequent dissection and comparisons revealed that these snails were indeed the elusive species originally collected by Hemphill and confirmed that the proper generic assignment is to the genus Ra- diocentrum. Additional field work extended the known range (based on recently dead shells) approximately 0.5 km to the northwest and 1.6 km to the southeast. No Radiocen- 1 2 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES trum have yet been found on other parts of the island in spite of extensive searching by the senior author. Since Hemphill did not specify his original locality, except as “Santa Catalina Island,” we can only speculate that it was near the town of Avalon or in Avalon Canyon as the name implies. The original locality may have been destroyed in the course of Avalon’s growth or through natural vegetational changes. Our findings indicate that the species’ present range is probably restricted to the southeast end of the island on south-facing slopes, in coastal sage scrub habitats dominated by Salvia and Opuntia. Because the species is rare and localized and would be threatened by unre- strained collecting, in this report locality information is given only in general terms. Detailed data are on file at the Santa Barbara Museum of Natural History and available for research and management purposes. The following institutional abbreviations are used: ANSP, Academy of Natural Sciences, Philadelphia; CAS, California Academy of Sciences; CM, Carnegie Mu- seum; SBMNH, Santa Barbara Museum of Natural History; SDMNH, San Diego Museum of Natural History; SUPTC, Stanford University Paleontological Type Collection (now at CAS); UCM, University of Colorado Museum; USNM, Na- tional Museum of Natural History, Smithsonian Institution. Order Sigmurethra Family Oreohelicidae Radiocentrum Pilsbry, 1905 Type-species: Oreohelix chiricahuana Pilsbry, 1905, by original designation. Diagnosis.—Oreohelicids with embryonic shell of about 1% radially ribbed whorls; penis tripartite, walls of anterior part plain or furnished with pilasters, middle part papillose or with oblique ridges internally, posterior part ridged inside, wide and truncate at the end; epiphallus about as long as the penis, slender anteriorly, the penial retractor inserted on it a short distance from its entrance in the penis; albumen gland large; reproduction oviparous. Babrakzai, Miller, and Ward (1975) found a large number of submetacentric chromosomes and a haploid chromosome nnumber of n = 32 to be characteristic of Radiocentrum. The few-whorled, radially costulate protoconch is the only diagnostic con- chological character. The shells are otherwise much like Oreohelix, depressed- helicoid to lenticular in shape, the periphery ranging from rounded, through obtusely subangular, to distinctly carinate. Remarks.— The endemic North American pulmonate family Oreohelicidae con- sists of two genera: Oreohelix Pilsbry, 1905, and Radiocentrum, the latter origi- nally proposed as a subgenus of the former. Conchologically, Radiocentrum has a distinctive, radially ribbed 1.5-whorled embryonic shell. Living Radiocentrum deposit eggs, whereas Recent Oreohelix are OvOviviparous. On the basis of reproductive characters and chromosome data, Babrakzai, Miller, and Ward (1975) elevated Radiocentrum to full generic status. Radiocentrum avalonense (Hemphill in Pilsbry, 1905) (Figs. 1-6) Pyramidula hemphilli Newcomb (in part), Stearns 1902:62. Oreohelix (Radiocentrum) avalonensis Hemphill, in Pilsbry 1905:283-284, pl. 11, figs. 4~7.—Cockerell 1905:71.—Pilsbry 1916:341, 354.—Berry 1931a:73.— REDISCOVERY OF RADIOCENTRUM AVALONENSE 3 Figs. 1-5. Radiocentrum avalonense. 1-2, Lectotype, ANSP 8667 1a; top and basal views; diameter 11.2 mm. 3, 5, SBMNH 33997, live-collected specimen from south-facing slope approximately 4 km south of Avalon, Santa Catalina Island, Calif., B. Roth coll. 5 January 1982; apertural and top views; diameter 9.9 mm. 4, SBMNH 33998, from same locality, F. G. Hochberg coll. 4 December 1978; top view; diameter 12.3 mm. Berry 1931b:115.—Pilsbry 1934:408.— Keep 1935:312, fig. 329.—Pilsbry 1939: 552-553, fig. 367.—Miller 1973:333-334. Helix var. avalonensis Hemphill 1911:104—-106, pl. 4. Helix avalonensis Hemphill, Orcutt 1915:159. Oreohelix avalonensis ‘““Hemphill” Pilsbry, Cockerell 1938:11.—Ingram 1946: 89.—Baily 1953:22.—Baker 1962:4.—Smith 1970:42.—Roth 1972:11.—Be- quaert and Miller 1973:32, 35. Radiocentrum avalonense Hemphill in Pilsbry, Coan, and Roth 1987:325. Type material.—Lectotype (designated herein): ANSP 86671a. Santa Catalina Island, California, H. Hemphill coll. Paralectotypes: ANSP 86671 (2 specimens). CAS 029012 (18) and 032996 (4). CM 6160 (3) (Brooks and Brooks 1931:248). SBMNH 33996 (9) and 34882 (2). SDMNH 1376 (5). SUPTC 6162 (51). UCM 20746 (8) and 21273 (4) (Wu and Brandauer 1982:29-30). USNM 174685 (6). The data for the paralectotypes are the same as for the lectotype, except that later labels with CAS 032996 and SBMNH 33996 add “‘Avalon,’”’ SUPTC 6162 adds the date “1902,” and SBMNH 33996 gives the date as “I-II-1902” (pre- sumably January and February). Referred material.—SBMNH 33997-34001, Santa Catalina Island, California, south-facing slopes approximately 4 km south of Avalon, elevation ca. 300-380 m, under Opuntia patches and in rock rubble areas around Salvia, F. G. Hochberg 4 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES O mm 5 Fig. 6. Reproductive system of Radiocentrum avalonense. ag, albumen gland; ec, epiphallic cae- cum; ep, epiphallus; fo, free oviduct; go, genital orifice; hd, hermaphroditic duct; Ipe, lower part of penis; ot, ovotestis; pr, penial retractor muscle; pt, prostate; sd, spermathecal duct; sp, spermatheca; upe, upper part of penis; ut, uterus; va, vagina; vd, vas deferens. Drawn from stained whole mount, WBM 7251. and others coll., 1978, 1980, 1982, 1983, 1984. Specimens also in private col- lections of the junior authors (WBM 7251, BR 1347). Description.—““Whorls 412, granulated above and below, the last one wide; aperture large. Alt. 6, diam. 11 mm.’ (Hemphill) (Pilsbry 1905:283). “The shell is lens-shaped, sharply carinate; umbilicus contained slightly more than 5 times in the diameter. Spire dull brown, the last whorl cream white with steaks of light brown and gray, and two pale brown bands, the upper one a short distance above, the lower immediately below the periphery. Embryonic shell of 12 strongly convex whorls, the initial half turn smooth, REDISCOVERY OF RADIOCENTRUM AVALONENSE 5 the following whorl with strong radial ribs with wider, smooth intervals. Later whorls with rough, irregular striation, irregularly cut into long granules by spiral impressions; base with irregular radial striae and fine, weak spiral Sstriation, with about 5 stronger spirals at wide intervals. The last whorl is subangular in the middle of the upper surface, at least in front; and it does not descend to the aperture. The peristome is thin, its ends connected by a thin, rather long parietal callus. “Height 6 mm., diameter 11.2 mm; 4'2 whorls. Height 6.3 mm., diameter 11.4 mm; 4% whorls. ‘Height 8 mm., diameter 14 mm; 5 or 5% whorls.’ (Hemphill.)”’ (Pilsbry 1939:553). Pilsbry (1905) quoted a brief diagnosis from an unpublished manuscript by Hemphill and illustrated shells, sufficient to validate the name six years before Hemphill (1911) published a full description. ANSP 8667 1a, which Pilsbry (1939) and Baker (1962) referred to as “‘the type,”’ is hereby designated lectotype of the species (Figs. 1, 2). Paralectotypes CAS 029012 are accompanied by a label in Hemphill’s handwriting in ink, ““H. var. avalonensis/Hemph/Santa Catalina Is./ Cal.’ with a later addition in pencil, ““Type lot.’’ They range from juveniles 4.9 mm in diameter with 3.5 whorls to a single specimen 14.0 mm in diameter, 8.8 mm in height, with 5.5 whorls. This specimen, evidently the one Hemphill (1911) regarded as his “‘single mature shell,” is slightly worn, the carina is obsolete on the last half of the body whorl, and the color ranges from flesh color on the spire to mottled cream white on the body whorl. The ribbed Radiocentrum embryonic shell is present. There seems to be no reason to doubt that this large shell came from the same source as the rest of Hemphill’s material. The next largest specimens in the lot are 12.3 and 12.1 mm in diameter, both with 4.9 whorls. There is a tendency in all material we have seen for shells greater than about 11.3 mm in diameter to have inflated body whorls. The newly collected specimens do not differ significantly in shell characters from the type material. The largest is 12.3 mm in diameter and has 5.0 whorls (Fig. 4). A dissected specimen 9.9 mm in diameter with 4.5 whorls was sexually mature. Anatomy.— The reproductive system of R. avalonense (Fig. 6) has been critically compared with those of R. chiricahuanum (Pilsbry, 1905), R. labrenanum (Pils- bry, 1948), R. clappi (Ferriss, 1904), R. exorbitans (Miller, 1973), and R. discus Christensen and Miller, 1976. It is typical of the genus in the following aspects: the albumen gland is large, as in other oviparous genera (in Oreohelix, which is Ovoviviparous, it is much reduced); the lower part (approximately one-third) of the spermathecal duct is swollen; the upper part of the penis is much enlarged and “hatchet-shaped.”’ It differs, however, from other species of Radiocentrum in having a swollen lower part of the penis, containing three distinct longitudinal pilasters. This characteristic was observed in all adult specimens dissected. Externally the animal is light tan, the mantle collar somewhat grayish. Speci- mens in terraria were active infrequently—far less than Micrarionta and Xera- rionta kept under the same conditions. They were not observed to eat fresh lettuce but apparently fed on dead stems and leaf litter collected under Sa/via clumps at their native locality. Habitat.— All our collections of R. avalonense have been made in a limited area near the southeast end of Catalina Island (Fig. 7), near the top of steep, south- 6 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES Santa Barbara Miguel mae w A - Los Angeles Anacapa sg Rosa Barbara Catalina NS Clemente S Nicolas | i 118°25'N 33°22'w— Avalon Fig. 7. Map of Santa Catalina Island, Calif., showing generalized distribution of Radiocentrum avalonense (stippled). facing slopes sparsely covered with coastal sage scrub vegetation. The dominant plants are black sage (Salvia mellifera) and prickly-pear cactus (Opuntia littoralis). Sclerophyll shrubs such as lemonade berry (Rhus integrifolia) and toyon (Heter- omeles arbutifolia) are also present (Figs. 8, 9). Few herbaceous plants occur on these upper slopes, probably because of persistent grazing by feral goats, whose trails, bare of vegetation. crisscross the slopes. The vegetation is less dense and arborescent than that on north-, east-, or west-facing slopes in the same area. Living specimens were first discovered early on 5 January 1982, after a night of drenching rain that continued into the morning. They were on a slope that, the day before, had yielded only a few empty shells. At 10:00 A.M. the snails were all active, quite turgid, and crawling either on stems or on the ground around the edge of Salvia clumps, or in one case on a stone deep in a patch of Opuntia. The helminthoglyptid snails Xerarionta kelletti (Forbes, 1850) and Micrarionta bea- tula Cockerell, 1929, were relatively common in the same area. One of the living R. avalonense was observed to be rasping the bark of a Salvia twig, and the first feces of specimens in captivity were grayish brown, the same color as the Salvia stems and leaf litter. The breakage pattern on some of the empty shells, body whorl broken from the side and spire decollated, suggests predation by small mammals such as mice. A field survey of the land mollusks of Catalina Island organized by the Santa REDISCOVERY OF RADIOCENTRUM AVALONENSE 7 Figs. 8-11. 8, Radiocentrum avalonense habitat: slope with coastal sage scrub vegetation. Low plant cover predominantly Salvia mellifera and Opuntia littoralis; larger shrubs are mainly Rhus integrifolia. 9, R. avalonense shells among litter of Salvia mellifera. 10, Avalon, Santa Catalina Island, looking northwest, about 1903; photograph courtesy of Catalina Island Museum. Vegetation on distant slope largely coastal sage scrub. 11, Similar view in 1983. Note increase in density of large chaparral shrubs, principally Rhus integrifolia, on distant slope. Barbara Museum of Natural History has not found R. avalonense elsewhere on the island. The eastern part of Catalina that includes Avalon Canyon is more completely dissected by canyons than any other sector of equivalent size. The bedrock is an intrusive dacite porphyry of Miocene age, which, however, also extends a considerable distance to the west (Platt 1976: fig. 3). This eastern sector also contains the largest uninterrupted area of coastal sage scrub on the island (Minnich 1980: fig. 2). Beyond these general observations on exposure, vegetation, and topography, we have been unable to correlate the distribution of R. avalonense with any more specific environmental factors. In order to be able to collect over 100 specimens, most apparently live-taken, of a species that we found rare and difficult to locate, Hemphill must have en- countered more populous colonies, perhaps nearer to the town of Avalon. His- torical photographs show that around Avalon, upon release from grazing pressure, a landscape of sparse grass, prickly-pear cactus, and open chaparral in the 1880s developed rapidly into coastal sage scrub by 1900. Thereafter, chaparral species such as Rhus integrifolia invaded at a slower pace (Minnich 1980) (Figs. 10, 11). The scrub has now been protected from brush fires for many years and in con- sequence is very mature and woody. Young or fire-renewed scrub would yield more soft litter and perhaps support a denser snail population like that which Hemphill may have found. There is also the possibility, hinted at by Smith (1970), that Hemphill over-collected the species and that the effects are still being felt today. Intemperate collecting by early naturalists was an important factor in the 8 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES 125° 120° 115° 110° 105° 100° ' L | t ! LJ L} ! ! 35° ( i} 1 | l 30° 25° Fig. 12. Map of western North America showing Recent and fossil ditribution of genus Radi- ocentrum. Solid circles, living; open circles, Quaternary (Pleistocene or early Holocene); triangles, upper Cretaceous through Oligocene occurrences. Stippled areas, approximate limits of Sonoran and Chihuahuan deserts (after Axelrod 1979). decline of the slow-maturing Achatinella snails on Oahu (Hadfield and Mountain 1980). Without comparable data on reproductive rate in R. avalonense we cannot automatically rule out the effects of early collecting. Biogeography.—The Pleistocene to Holocene distribution of Radiocentrum (Fig. 12) includes the following taxa: REDISCOVERY OF RADIOCENTRUM AVALONENSE 9 Southern New Mexico R. hachitanum hachitanum (Pilsbry, 1905) R. h. cadaver (Pilsbry, 1915) R. ferrissi ferrissi (Pilsbry, 1915) R. f. morticinum (Pilsbry, 1915) Southern Arizona . Clappi clappi (Ferriss, 1904) . c. cataracta (Pilsbry and Ferriss, 1910) . c. emigrans (Pilsbry and Ferriss, 1910) . chiricahuanum chiricahuanum (Pilsbry, 1905) . ch. obsoletum (Pilsbry and Ferriss, 1910) . ch. percarinatum (Pilsbry and Ferriss, 1910) Chihuahua R. caenosum (Pilsbry, 1948) R. labrenanum (Pilsbry, 1948) R. almoloya (Drake, 1949) Northern Coahuila R. orientalis Metcalf, 1980, probably Pleistocene Franklin Mountains, El] Paso County, Texas R. ferrissi ferrissi, late Pleistocene fossils (Metcalf and Johnson 1971) Baja California Sur R. exorbitans (Miller, 1973) R. discus Christensen and Miller, 1976 Santa Catalina Island R. avalonense (Hemphill, in Pilsbry, 1905). In addition to the species assigned a Pleistocene age, R. almoloya, R. hachi- tanum cadaver, and R. ferrissi morticinum are known from their empty shells only and may represent extinct colonies. Tozer (1956) found Oreohelix angulifera (Whiteaves, 1885) from the St. Mary River and Edmonton Formations, upper Cretaceous of western Alberta, and O. thurstoni (Russell, 1926) from the Paskapoo, Porcupine Hills, and Willow Creek Formations, Paleocene of western Alberta, to have regular costae on the embryonic whorls, strongly suggestive of Radiocentrum. Two undescribed species of Radio- centrum occur in the Bozeman Group, Eocene and/or Oligocene, of Broadwater and Gallatin counties, Montana (D. W. Taylor in Robinson 1963:68, table 4; Taylor 1975:209; where cited as Oreohelix n. sp.; Roth, in press). Additional fossil oreohelicids probably assignable to Radiocentrum include R. grangeri (Cockerell and Henderson, 1912) from the Eocene of Park County, Wyoming, and R. hen- dersoni (Russell, 1938) from the Oligocene of Colorado. Cockerell (1914) assigned Helix nacimientensis White, 1886, from the Paleocene of New Mexico, to Ra- diocentrum; Taylor (1975), however, regarded this species as probably helminth- oglyptid. These Cretaceous and early Tertiary species are all north of the present range of the genus, along the eastern Cordillera (Fig. 12). The northernmost occurrences are the oldest: late Cretaceous and Paleocene. Eocene and Oligocene localities are somewhat farther south. The genus has evidently undergone a southward restric- tion or displacement of range since Paleogene time. Evidence from paleobotany, oceanic foraminifera, and fossil vertebrates (Savin 1977; Wolfe 1978; Lillegraven 1979) indicates progressive cooling through the AAAARAD 10 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES Tertiary Period. a steepening latitudinal temperature gradient, and restriction of tropical and subtropical biota to lower latitudes. Over the same interval. precip- itation in the western interior has become largely confined to the winter months (Axelrod 1979). The extinction of Radiocentrum in the northern regions from present-day Alberta to Colorado probably accompanied a climatic change from tropical to temperate and from summer-wet to summer-dry. The Quaternary range of Radiocentrum includes two significant disjunctions, across the Sonoran and Chihuahuan deserts (Fig. 12). The Sonoran Desert is interposed between R. avalonense on Santa Catalina Island and the Baja Cali- fornian and the Arizona-New Mexico-Chihuahuan groups of species. The main mass of the Chihuahuan Desert separates the latter group from the Pleistocene R. orientalis in Coahuila: the occurrence of R. ferrissi ferrissi in the Franklin Mountains of west Texas (where it is apparently extinct) is about 180 km east of the living R. ferrissi of New Mexico and separated by a tongue of Chihuahuan Desert. Many trans-Sonoran disjunctions are recognized among plants, including woody species of the woodland and chaparral vegetation of southern California and Anzona (Axelrod 1979: fig. 4, table 1) and unique taxa of the insular flora of southern California with their closest allies in the Sierra Madre of Mexico (Axelrod 1979: fig. 6). According to Clements (1936), the disjunct woodland and chaparral taxa imply a former continuity over the area at a time when temperatures were much milder and annual rainfall was near 500 mm at a minimum. Studies of late Pleistocene plant remains (summarized by Axelrod 1979:9) indicate that oak- conifer woodland covered much of the Sonoran Desert in Arizona and also the northern Chihuahuan Desert as recently as 10.000-12.000 years ago. Some of the component taxa persist now as relicts in the desert ranges. where they retreated as aridity increased after the last pluvial stage. Similarly. the arid environment of the Chihuahuan Desert separates the montane woodland of the eastern and western cordilleras of north-central Mexico. The emergence of the Sonoran Desert as an environment of regional extent dates from the latest Cenozoic; its current broad area. the maximum ever achieved, 1s a post-Wisconsinan feature (Axelrod 1979). The disjunct type of distribution shown by Radiocentrum is what one would expect of a formerly widespread taxon now persisting as relicts in scattered en- claves of favorable habitat. The fact that several taxa are known only from empty shells may indicate that recent extinction of localized populations is still taking place—consistent with Axelrod’s timetable of modern desert development. Ra- diocentrum avalonense is thus probably a relict outlier of a once more continuously distributed genus that has undergone its most severe fragmentation during the late Cenozoic. Another molluscan taxon that may have had a similar history yet gone all the way to extinction on the Pacific Coast is the urocoptid genus Holospira. The present westernmost range of Holospira is in eastern Arizona and Sonora (Bequaert and Miller 1973: fig. 6), but Gregg (1944) reported fossils apparently referable to Holospira from the late Pleistocene deposits of Rancho La Brea in Los Angeles, California. Acknowledgments We wish to thank especially Douglas Probst (Catalina Island Conservancy) for permission to conduct our studies on Conservancy property. Mark Hoefs (Wrigley REDISCOVERY OF RADIOCENTRUM AVALONENSE 11 Memorial Garden and Foundation) for identification of island plants and for partial funding support, and Patricia Moore (Catalina Island Museum) for use of historical photographs housed at the Museum. Field trip participants include Robert Given (Catalina Marine Science Center); Paul Scott, Eve Shipp, and Jim Shipp (SBMNH); Timothy Pearce (University of California); Betty Sue Miller, Jane Deisler, Edna Naranjo Garcia, Bob Carey, Russel Duncan, Jim Hoffman, Jennifer Titley, and Diana Warr (University of Arizona). Robert Robertson and Mary Garback (ANSP) loaned type specimens; Harald Rehder and the late Joseph Rosewater helped locate material at the USNM. The late S. Stillman Berry pro- vided access to his bibliographic file and specimens in his collection. Carl Chris- tensen and Art Metcalf critically reviewed the manuscript. Literature Cited Axelrod, D.I. 1979. Age and origin of Sonoran Desert vegetation. California Acad. Sci. Occas. Pap. 132, 74 pp. Babrakzai, N., W. B. Miller, and O. G. Ward. 1975. Cytotaxonomy of some Arizona Oreohelicidae (Gastropoda: Pulmonata). Bull. Amer. Malacol. Union, 40:4—-11. Baily, J. L. 1953. Oreohelix in San Diego County, California. Nautilus, 67:20—22. Baker, H. B. 1962. Type land snails in the Academy of Natural Sciences of Philadelphia. I. North America, north of Mexico. Proc. Acad. Nat. Sci. Philadelphia, 114:1-21. Bequaert, J. C., and W. B. Miller. 1973. The mollusks of the arid southwest, with an Arizona check list. Univ. Arizona Press. Berry, S.S. 193la. The genus Oreohelix in California. Nautilus, 44:73-75. . 1931b. New helicoid snails from the Mohave Desert.—V. The genus Oreohelix in southern California and Nevada, Ann. Mag. Nat. Hist. (10), 8:115-120. Brooks, S. T., and B. W. Brooks. 1931. List of types of Amphineura and Gastropoda catalogued in the collection of the Carnegie Museum on January 1, 1931. Ann. Carnegie Mus., 20:179-253. Christensen, C. C., and W. B. Miller. 1976. A new Radiocentrum (Pulmonata: Oreohelicidae) from Baja California, Mexico. Veliger, 18:378—380. Clements, F.E. 1936. The origin of the desert climax and climate. Pp. 87-140 in Essays in geobotany in honor of William Albert Setchell (T. H. Goodspeed, ed.). Univ. California Press. Coan, E. V., & B. Roth. 1987. The malacological taxa of Henry Hemphill. Veliger, 29:322—339. Cockerell, T. D. A. 1905. The snails of New Mexico and Arizona. Nautilus, 19:68-71. . 1914. Tertiary Mollusca from New Mexico and Wyoming. Bull. Amer. Mus. Nat. Hist., 33: 101-107. 1938. Studies of island life. Univ. Colorado Stud., 26:3-20. , and J. Henderson. 1912. Mollusca from the Tertiary strata of the west. Bull. Amer. Mus. Nat. Hist., 31:229-234. Drake, R. J. 1949. A new species of Oreohelix, subgenus Radiocentrum, from southeastern Chi- huahua. Nautilus, 62:109-112. Gregg, W. O. 1944. A Rancho La Brea landsnail. Min. Conchol. Club South. California 41:30. Hadfield, M. G., and B.S. Mountain. 1980. A field study ofa vanishing species, Achatinella mustelina (Gastropoda, Pulmonata), in the Waianae Mountains of Oahu. Pac. Science, 34:345-358. Hemphill, H. 1911. Descriptions of some varieties of shells, with short notes on the geographical range and means of distribution of land shells. Trans. San Diego Soc. Nat. Hist., 1:99-108. Ingram, W. M. 1946. A check list of the helicoid snails of California from Henry A. Pilsbry’s monograph. Bull. South. California Acad. Sci., 45:61—93. Keep, J. (J. L. Baily, reviser). 1935. West coats shells. Stanford Univ. Press. Lillegraven, J. A. 1979. A biogeographical problem involving comparisons of later Eocene terrestrial vertebrate faunas of western North America. Pp. 333-347 in Historical biogeography, plate tectonics, and the changing environment (J. Gray and A. J. Boucot, eds.). Oregon State Univ. Press. Metcalf, A. L. 1980. A new fossil Radiocentrum (Pulmonata: Oreohelicidae) from northern Coahuila, Mexico. Nautilus, 94:16-17. , and W. E. Johnson. 1971. Gastropods of the Franklin Mountains, E] Paso County, Texas. Southwest. Nat., 16:85-109. 12 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES Miller, W. B. 1973. A Recent Oreohelix (Gastropoda: Pulmonata) from Baja California Sur, Mexico. Veliger, 15:332-334. Minnich, R. A. 1980. Vegetation of Santa Cruz and Santa Catalina Islands. Pp. 123-137 in The California Islands: Proceedings of a multidisciplinary symposium (D. M. Power, ed.). Santa Barbara Mus. Nat. Hist. Orcutt, C. R. 1915. Molluscan World, vol. 1. San Diego, privately printed. Pilsbry, H. A. 1905. Mollusca of the southwestern states, I: Urocoptidae; Helicidae of Arizona and New Mexico. Proc. Acad. Nat. Sci. Philadelphia, 57:211-—290. . 1916. Notes on the anatomy of Oreohelix, with a catalogue of the species. Proc. Acad. Nat. Sci. Philadelphia, 68:340-359. 1934. Notes on the anatomy of Oreohelix, —III, with descriptions of new species and sub- species. Proc. Acad. Nat. Sci. Philadelphia, 85:383-410. 1939. Land Mollusca of North America (north of Mexico). Acad. Nat. Sci. Philadelphia Monog. 3, 1 (1):1-573. 1948. Inland mollusks of northern Mexico.—I. The genera Humboldtiana, Sonorella, Or- eohelix and Ashmunella. Proc. Acad. Nat. Sci. Philadelphia, 100:185-203. Platt, J.P. 1976. The petrology, structure, and geologic history of the Catalina Schist terrain, southern California. Univ. California Publ. Geol. Sci., 112:1-111. Robinson, G. D. 1963. Geology of the Three Forks Quadrangle, Montana. U.S. Geol. Surv. Prof. Pap. 370, 143 p. Roth, B. 1972. Rare and endangered land mollusks in California. Sterkiana, 48:4—16. . 1982. The identity of “Helminthoglypta” tularica (Bartsch) (Gastropoda: Pulmonata). Bull. South. California Acad. Sci., 81:101-105. —. In press. Land mollusks (Gastropoda: Pulmonata) from early Tertiary Bozeman Group, Montana. Proc. California Acad. Sci. Russell, L. S. 1938. New species of Gastropoda from the Oligocene of Colorado. J. Paleontol., 12: 505-507. Savin, S. M. 1977. The history of the earth’s surface temperature during the past 100 million years. Ann. Rev. Earth Planet. Sci., 5:319-355. Smith, A.G. 1970. American Malacological Union Symposium: Rare and endangered mollusks. 6. Western land snails. Malacologia, 10:39—46. Stearns, R. E.C. 1902. Helix var. circumcarinata and Pyramidula elrodi. Nautilus, 16:61-62. Taylor, D. W. 1975. Early Tertiary mollusks from the Powder River Basin, Wyoming-Montana, and adjacent regions. U.S. Geol. Surv. Open-file Rept. 75-331, 515 p. Tozer, E. T. 1956. Uppermost Cretaceous and Paleocene non-marine molluscan faunas of western Alberta. Geol. Surv. Canada Mem. 280, 125 p. Wolfe, J. A. 1978. A paleobotanical interpretation of Tertiary climates in the northern hemisphere. Amer. Sci., 66:694—-703. Wu, S.-K., and N. E. Brandauer. 1982. Type specimens of Recent Mollusca in the University of Colorado Museum. Nat. Hist. Inventory of Colorado, no. 7, 47 pp. Accepted for publication 20 September 1985. Bull. Southern California Acad. Sci. 86(1), 1987, pp. 13-26 © Southern California Academy of Sciences, 1987 Distribution and Stability of Grasslands in the Los Angeles Basin David O. Freudenberger, Brian E. Fish and Jon E. Keeley Department of Biology, Occidental College, Los Angeles, California 90041 Abstract. — Present grassland distribution (as of 1980) was mapped from modern aerial photographs for 21 7.5 minute quadrangles in the Los Angeles Basin. These patterns were compared with the distribution of grasslands mapped from aerial photographs from 1928 to 1936. Grasslands increased in all but three quadrangles, the greatest increases being in the northwestern portion of the basin. Vegetational changes, however, were not unindirectional as shrubland, notably coastal sage scrub, replaced grassland in sections of nine quadrangles. In general, areas subject to frequent fires and grazing moved from shrubland to grassland while grassland areas with infrequent disturbance were invaded by shrubs. An intensive study was undertaken in a quadrangle in which vegetation had changed in both direc- tions. Nine islands of coastal sage scrub surrounded by grassland were selected for study. Vegetation pattern was not dictated by topographic position or soil charateristics. Density of seedlings was high within patches of mature shrubs but seedlings were largely absent outside their boundaries. Shrub sizes in transects across the ecotone suggested that in eight of the nine coastal sage scrub patches shrubland was not invading grassland. Recent burning (within 10 years), coupled with intensive grazing, appears to inhibit the invasion of shrubs into adjacent grasslands. We hypothesize that the vegetation of the Los Angeles Basin is a mosaic of community types differing in their tolerances to disturbance. Annual grasslands are a major vegetation type in California. Throughout much of their range they are dominated by non-native species and the vegetation com- position prior to the introduction of these exotic species in the late 18th century is unknown. One hypothesis is that these grasslands were originally dominated by native bunchgrass species, notably Stipa pulchra. An alternative is that these regions were originally brush covered. Clements (1934) noted that certain grassland areas contained isolated patches of Stipa pulchra which he interpreted as “‘relicts” of a previous native grassland. It has been suggested that these native grasslands succumbed to competition with annual exotics during periods of intensive grazing and drought in the mid 19th century (Heady 1977). Alternatively, Cooper (1927) noted what he interpreted as “relict” stands of brush in grassland areas and hypothesized that early settlers cleared brushlands by repeated fires. Since brush species are intolerant of repeated fires they were replaced by the exotic annual species which are good colonizers. Wells (1962) suggested that the present annual grasslands were not derived from grasslands dominated by native perennial bunchgrasses on deep clay soils, but were derived from woody sage and chaparral vegetation on shallow rocky soils. 13 14 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES Today, areas dominated by native grasses are invariably on fine-textured sub- strates. Attempts to regenerate native grass species off such soils have largely failed (Kay et al. 1981). Although the theory of widespread native grasslands is based largely on “presumed relicts” of Stipa pulchra and other native grasses, the al- ternative theory is based on “‘known relicts” of chaparral. Documents show that many grassland areas once were covered by brush. Land managers have long used repeated prescribed fires to clear brush and Zedler et al. (1983) have shown this can result from frequent wildfires. Annual grasslands are widely distributed throughout the Los Angeles Basin. The purpose of this study was to determine the distribution and stability of these grasslands and to identify the factors responsible for their formation. A major focus was to document present grassland distribution and determine if any dis- tributional changes had occurred in the last 50 years by comparisons with early aerial photographs. A second focus was to examine site characteristics in a grass- land area containing patches of coastal sage scrub. Demographic trends of shrubs at the grass: sage interface were examined in order to evaluate the stability of such zonation patterns. Sites and Methods Present and Past Grassland Distribution Twenty-one 7.5-minute USGS quadrangles were selected for study. These in- cluded the most extensive grasslands in the Los Angeles Basin (Fig. 1) excluding the Santa Monica Mtns. where grassland distribution has been studied by Goode (1981) and Hobbs (1983). USGS quadrangle maps were used as a starting point for outlining the present-day grassland distribution. These maps give vegetation types, although in many cases they are out of date. Therefore, 1974-1976 aerial photographs were obtained for all quadrangles and used to update vegetation distribution. Updated maps were further checked by on-site inspection of most areas in 1979-1980. The historical record for these areas was constructed by examination of 1928-1936 aerial photographs from the Fairchild Collection at Whittier College, Whittier, California. Interpretation of these photographs was aided by vegetation maps produced by Wieslander circa 1930 (available at the Rancho Santa Ana Botanic Garden). USGS topographic maps were also a source of historical data since vegetation types were mapped, but often not updated for over 25 years. All historical vegetation data were mapped on acetate sheet overlays of the 7.5-minute topographic maps. From the present-day vegetation maps and historical overlays, the total area of vegetation change was estimated to the nearest hectare. Data on the fire history were gathered from county, state, and federal agencies for areas where such data were available. For various reasons the area under study has very incomplete data on fire history. USDA soil survey maps were consuited for areas for which they were available. Localized Patterns at the Grass : Sage Ecotone— Intensive Site The intensive study site was the Calabasas Quadrangle between Palo Comado and Cheeseboro Canyons, 3-5 km N of Hwy 101. This area contains a mosaic of grasslands with patches of coastal sage scrub. Nine sites were selected; each in- cluded one scrub patch plus adjoining grassland. LOS ANGELES BASIN GRASSLANDS 15 POINT MUGU. LOS ANGELES SANTA MONICA LONG BEACHe ao_— DANA POINT Fig. 1. Los Angeles Basin 7.5 minute quadrangles included in this study. Slope aspect and inclination were recorded for each site. Soil samples from O- 15 cm depth were taken from the shrub stand and adjoining grassland, in most cases on two sides. For each soil sample, pH, organic matter content and soil texture were determined according to procedures in Cox (1980). Vegetation at each site was sampled in spring 1982 by recording height and foliage diameter of all shrub species in 3.5 x 4.0 m quadrats placed contiguously from grassland through the coastal sage scrub patch and into the grassland on the other side. Results Grassland Distribution During the past 50 years, vegetation has changed throughout the Los Angeles Basin (Table 1). Since urbanized areas and other human artifacts such as agri- cultural land have been excluded from this analysis, decreases in one vegetation represent an increase in another vegetation type. Grasslands have increased or decreased in many areas. Quantitatively, the major change has been an increase in grasslands at the expense of coastal sage scrub and chaparral. A brief description of the changes observed in each quadrangle follows. Northwestern Los Angeles Basin In the Thousand Oaks Quadrangle, coastal sage increased extensively at the expense of grassland over the past 40 years. This is evident in a comparison of 1936 aerial photos with the 1952 USGS topographic map. Further increases in coastal sage scrub were seen in the 1976 aerial photo. Except for a small fire around Medea Creek in 1970, this region had not burned since 1910. The area burned in 1970 was grassland in 1976. In general, grasslands predominated on 16 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES Table 1. Grassland distribution changes in the Los Angeles Basin between 1928 and 1980. Area subject Area changed to change i eens Quadrangle (ha) (ha) % change Increase Decrease Northwestern: Thousand Oaks 400 8000 5 xX xX Calabasas 700 12,000 6 X xX Santa Susana 1000 12,000 8 xX Oat Mountain 300 12,000 3 xX San Fernando 1100 8000 14 xX Burbank 50 8000 >1 X Northeastern: Glendora 400 15,000 3 xX San Dimas 500 8000 6 xX >< Baldwin Park 0 1700 0 La Habra 300 4000 8 xX Yorba Linda 200 23,000 2 xX Prado Dam 250 10,000 2 xX Corona North 20 1500 >1 xX Riverside West 100 1500 6 m4 Southern: Steele Peak 600 8000 8 X Lake Mathews 1500 12,000 13 »4 »4 Corona South 100 12,000 >1 »4 Black Star Canyon 100 12,000 3 xX Orange El Toro 150 8000 7 X Canada Gobernadora 500 16,000 D) xX xX San Juan Capistrano 100 16,000 >1 xX Laguna Beach 900 9000 10 xX soil such as Linne silty clay (30-50% slope), Rincon silty clay loam (2—9% slope) and Cropley clay (2-9% slope). Coastal sage tended to predominate on Calleguas shaley loam (30-50% slope). However, field examination detected numerous ex- ceptions. In the Calabasas Quadrangle, grassland has increased substantially, particularly in Las Virgenes, Cheeseboro, and Palo Comado Canyons. Increases were evident from 1928 to 1952, and from 1952 to 1976. Most of this area burned in 1970 and is now grazed by cattle. A small area in the southern portion showed an increase in coastal sage scrub over grassland, however. The Santa Susana Quadrangle was dominated by grasslands in 1930. Since then, these grasslands have expanded at the expense of coastal sage scrub along Chivo, Gillibrand, Windmill, and Meier Canyons. The last known fire in this area was in 1970, 10 years prior to these observations. Today this area is heavily grazed by cattle. An area within a fenced-in water tank remained coastal sage while grassland replaced coastal sage outside. A 1-ha area of grassland, which appeared to be under cultivation in 1928, was coastal sage scrub in 1980. Grasslands increased substantially on the south-facing slopes of the Santa Su- sana Mountains in the Oat Mountain Quadrangle. However, a few patches of LOS ANGELES BASIN GRASSLANDS 17 oot ~ nS if PAS EehS ‘ ‘ sage 1928 + , ‘ grass 1980 , 7 sage 1928 grass 1980 7’ sage 1928 oo Nee au “grass 1980- wisage 1928 = grass 1980 “sage 1928 *., “| grass 1980°, 1928 1980 sage 1928 grass sage 1980 2000060 ° ° ° 00% ° 2 ©, ° ° Co ° o 20000000000° Deer Lake Highlands Fig. 2. Historical and present day vegetation patterns (as of 1980) on a portion of the Oat Mountain Quadrangle. coastal sage scrub appeared in 1980 in areas that were grassland in 1928. These changes are illustrated in Fig. 2. Parts of this area burned in either 1969 and/or 1970. Most of this area is currently grazed by cattle. In the San Fernando Quadrangle, grassland has increased markedly since a fire in 1975. Shrub resprouts were most evident on north-facing slopes and ravines, however, and the entire area is likely to return to the dense chaparral recorded on the 1966 USGS map. 18 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES The vegetation of the Verdugo Mountains within the Burbank Quadrangle has remained unchanged, although the density of chaparral between Elmwood and Stough Canyons has increased since 1928. Northeastern Los Angeles Basin Glendora Quadrangle includes the USDA San Dimas Experimental Forest. A large section of chaparral in this area was converted to grassland as part of a type- conversion experiment. Within the San Dimas Quadrangle there have been diverse changes. Three hectares on the south facing slope of Buzzard Peak were coastal sage or chaparral in the 1928 aerial photo and Wieslander’s 1930 vegetation map. On the 1964 USGS map they were sparse shrub-grassland and in the 1975 aerial photo and 1980 field checks they had progressed to pure grassland at the lower elevations and open grass-shrubland higher on the slope. Several areas in this quadrangle had changed from grassland or sparse shrub cover circa 1930 to dense brush in 1980; e.g., east of Diamond Bar, within the Puddingstone Reservoir State Rec- reation Area and northeast of Tonner Canyon. The Baldwin Park Quadrangle has remained relatively stable. Wieslander mapped this area in 1930 and much of his grassland-coastal sage scrub mosaic was still evident in the 1976 aerial photo. Much of this area burned in 1976. In La Habra Quadrangle grassland increased somewhat at the expense of coastal sage scrub or chaparral, especially in the proximity of oil fields which were not present in 1928. Parts of the Puente and Chino Hills, included in the Yorba Linda Quadrangle, have remained quite stable over the past 50 years despite the considerable mosaic of vegetation. Aerial photos showed that from 1928 to 1975 the density of scrub oak chaparral increased between Soquoel and Telegraph Canyons. An area along Brea Canyon Road burned in 1970, and the 1975 photo showed no change in vegetation since 1928 other than more open chaparral due to incomplete recovery from the fire. Soil maps for the San Bernardino section of this quadrangle showed that both grasslands and shrublands occurred on Fontana clay loam and Gaviota rock outcrop complex soils. The vegetation in the Prado Dam Quadrangle has remained relatively stable over the past 50 years. One obvious change is a decrease in live oak (Quercus agrifolia) density in this region. The soils of the San Bernardino County portion of this quadrangle have been mapped and no obvious correlation exists between soil type and vegetation; scrub, grasses, and oak trees were found on Alo clay and Fontana clay loam as well as on the Gaviota rock outcrop complex. Topography seemed to be important since coastal sage scrub was concentrated on steep slopes, grassland on valley floors, oak woodland on gentle slopes and stream floodplains. The Corona North Quadrangle was dense coastal sage scrub in 1931; however, the 1967 USGS map indicated it was <25% shrub cover and in 1980 it was still sparse coastal sage scrub. This may derive from a fire which covered most of the area in 1959. Three hectares on the north facing slope of Arlington Peak in Riverside West Quadrangle were dense coastal sage scrub in the 1975 photos but burned in 1978. Field observations in 1980 indicated that these areas were predominantly grass covered. LOS ANGELES BASIN GRASSLANDS 19 In 1931, the Pedley Hills were covered by coastal sage which was denser on north-facing than on south-facing slopes. This area burned in 1970 and in 1975 was covered by a mixture of grass and sage. Southern Los Angeles Basin Steele Peak Quadrangle has had several changes in the valleys. An area adjacent to the Galvian Mine was partially cultivated and entirely grassland in 1930; however, the eastern three quarters of the valley was sage in 1980. Many of the other valleys covered by grassland in 1930 are now orchards and not considered in the estimates in Table 1. Several areas of rugged terrain southwest of Steele Peak were covered with sage in 1931 but were dominated by grassland in 1980. In general this replacement of coastal sage scrub by grassland did not occur on north-facing slopes. Between 1960 and 1979 there were at least five fires in various parts of this region. In the Lake Mathews Quadrangle large areas that were grassland in 1930 are now coastal sage scrub. This area was heavily grazed in the 1930’s and before. In the 1950’s, as part of the Lake Mathews Reservoir project, a chain-link fence was constructed around much of the area. This fence bounds the area which has gone from grass to shrubland, possibly due to the elimination of grazing plus the fact that there have been no recorded fires since 1910. A small area outside this fence site, which was dense coastal sage scrub in 1974, burned in 1978. In 1980 it was still dominated by grasses and a few resprouting shrubs. Another site was dense coastal sage in 1930, burned in 1957, and was still sage in 1974. The Corona South Quadrangle showed some expansion of grassland at the expense of coastal sage scrub. In the Black Star Canyon Quadrangle some area has gone from woodland to grass and several hectares of grassland have been invaded by coastal sage scrub. In the northern half of the quadrangle the chaparral vegetation has become notice- ably sparser since 1931. This area burned in 1948 and 1967. The chaparral east of Black Star Canyon has not burned since 1914 and is considerably denser than in 1931. A grassland area north of Fresno Canyon to the Santa Ana River burned in 1928, 1958, and 1967 and was still grassland in 1975. The Orange Quadrangle was largely grassland in 1928 and remains so today. Riparian woodland has declined in the Peralta Hills and sage scrub is less extensive along the lower hills around Chapman Ave. The hills above Rattlesnake and Siphon Reservoirs in the El Toro Quadrangle were sage in the 1931 photo and grassland in the 1974 photo. These areas burned in 1948 and 1967. The coastal sage on the hills between Williams and Silverado Canyons was denser in 1974 than in 1931. Within the Canada Gobernadora Quadrangle pockets of coastal sage surrounded by grassland were evident in 1929. By 1974 the coastal sage had expanded and dominated the area, particularly the region west of Trampas Canyon. The scrub cover was also denser in 1929 than in 1974. A similar pattern was evident along the eastern wall of Bell Canyon. In 1958 wildfire burned from Canada Chaquita east into the Cleveland National Forest. An area near Trampas Canyon that escaped this fire showed an increase in density of scrub. Much of the open area within the San Juan Capistrano Quadrangle has remained grassland over the past 50 years. This area has been cultivated and grazed heavily. 20 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES A 1-ha patch on the western bank of Arroyo Trabuco was coastal sage in 1929 and grassland in 1974. In the Laguna Beach Quadrangle many localities changed from grassland in 1931 to sparse coastal sage scrub in 1964, and to dense coastal sage scrub in 1980: e.g.. Emerald. Laguna, East Moro and Los Trancos Canyons, and the southwest- facing slopes of Mustard Hill. The 1964 USGS indicated that the San Joaquin Hills had <25% coastal sage coverage. whereas the 1974 photos and 1980 field checks showed this area to be dense coastal sage scrub. Today most grasslands are on 9—50% slopes on Alo clay or Balcom clay loam. Much of the coastal sage is on 30—75% slopes on Cieneba rock outcrop complex and Callegaus clay loam. However. there were numerous exceptions. These areas have not burned since 1931. This quadrangle was heavily grazed by sheep in the late 1800’s and early 1900’s (Fusch 1968) but is no longer being grazed. Localized Patterns at the Intensive Study Site The study area was in the Calabasas Quadrangle and had not burned for 10 years as indicated by ring counts of Salvia leucophylla shrubs. This corresponds to the Clampitt Fire of 1970. This area has been intensively grazed in recent years. The nine patches of coastal sage scrub were distributed on a range of slope aspects and inclinations (Table 2). Observations of these and other sage patches confirmed that they are not related to topographic position. Soils data likewise suggest that they are not distributed according to soil differences. Grassland soils tended to be more similar to those of adjacent sage stands than to those of other grassland areas (Table 2). A reasonable hypothesis is that the mosaic of coastal sage patches in grassland is a dynamic system controlled by disturbance. If the boundaries between these communities are changing. then in what direction? Is coastal sage invading grass- land or vice versa? Leak and Graber (1974) suggested a technique for determining the direction in which a shrubland-grassland boundary is moving. Their theory suggests that if sage scrub species are advancing into grassland. one should find an advancing front of young plants. On the other hand if the grasslands are gradually replacing the sage. perhaps due to disturbance such as frequent fires, the sage plants along the margins should be as old as those throughout the stand. A rough test of this prediction was made by examining the distribution of shrub heights, as height and age are related in these shrub species. Figure 3 shows such a height profile for a belt transect 4 m wide through Site 1. In this stand Salvia leucophylla was the dominant shrub (Table 3). It is clear that the distribution of size classes is not greatly different in the border areas as compared to within the stand itself; mature shrubs occurred in the ecotone as well as within the stand. Seedlings (plants <25 cm high) were abundant (Table 3) within the stand but not at the advancing front. Seven of the other eight sites showed a pattern broadly similar to Site 1. Site 2 was co-dominated by Salvia leucophylla and Artemisia californica (Table 3). One difference from the Site 1 pattern was that while seedlings were not found outside the boundary of the stand, there was an area within the stand devoid of mature shrubs (possibly from disturbance) but with abundant seedlings and “sap- lings.” This area extended for three quadrats (out of 22) and had 1.9 seedlings/ m- compared to <0.1/m7? in all other plots. LOS ANGELES BASIN GRASSLANDS 21 Table 2. Slope aspect and inclination for each coastal sage patch studied and soil characteristics within each stand and in the adjacent grassland in the Calabasas Quadrangle. Soil characteristics Organic Inclina- matter 0 Bes (% dry Texture (%) Site Area Aspect (°) PH weight) Clay Silt Sand Gravel #1 Upper grass N 20 6.5 6.7 15 11 74 1.1 Shrubs 7.5 8.9 10 14 76 7.1 Lower grass 7.8 7.9 6 17 77 6.1 #2 Upper grass WNW 30 6.5 7.3 11 11 78 0.2 Shrubs 6.7 6.2 9 12 79 0.5 Lower grass 7.6 8.8 11 12 76 3.4 #3 Upper grass NNW 35 6.8 7.5 11 13 76 2.0 Shrubs 6.9 7.9 10 13 77 2.9 Lower grass 6.7 7.0 10 12 78 3.7 #4 Upper grass NW 25 6.9 8.2 9 14 77 4.2 Salvia 6.7 11.8 12 13 75 9.0 Haplopappus 6.8 8.0 12 12 76 5.5 Haplopappus 15 6.5 7.5 14 10 76 2.6 Lower grass 6.4 7.0 13 13 74 6.5 #5 Upper grass E 25 Uo! 6.0 7 16 77 0.4 Shrubs 7.2 6.2 9 11 80 1.5 #6 Upper grass NW 40 6.7 6.9 11 10 79 0.1 Shrubs 6.7 6.8 9 11 80 0.9 Lower grass 6.9 6.9 8 13 79 3.6 #7 Upper grass W 30 7.6 8.4 9 15 76 6.2 Shrubs ES 9.1 8 14 78 5.0 #8 North grass ESE 15 7.8 5.6 8 12 80 1.7 Shrubs 7.8 6.3 7 15 78 3.2 South grass 7.7 6.4 7 13 80 2.6 #9 Upper grass N 30 6.5 6.8 11 11 78 1.0 Shrubs 6.8 6.6 10 11 79 1.1 Lower grass 6.6 6.4 8 13 79 7.5 Salvia leucophylla and Artemisia californica dominated Site 3. One variation from the pattern seen for Site 1 (Fig. 3) was that half of the site lacked seedlings altogether whereas the other half had a seedling density of 0.6/m7?. Figure 4 shows the profile observed in Site 4 which showed the greatest deviation from the pattern noted at Site 1 (cf. Fig. 3). The area was dominated by Haplo- Dappus venetus. The distribution was the closest example we observed of a shrub population expanding outward into the grassland. Salvia leucophylla dominated Site 5 and all shrub seedlings were within the boundary of the largest shrubs. Co-dominants at Site 6 were Salvia leucophylla and Artemisia californica. This stand had an abundance of seedlings which followed a pattern comparable to Fig. 3 although the greatest concentration of seedlings was in the center of the stand. Site 7 was essentially pure Salvia leucophylla and no shrub <50 cm height occurred within 12 m of the outer boundary of the stand. 22 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES 225 i 200 1 i i 1 1 1 175 1 1 i 1 2 2 i 1- i 1 1 150 1 i 2 1 3 i 1 1 i 1 ~ 1 = i 1 1 2 2 i S 125 ai ip eee : ae i 1 2 2 ‘€ 2 fo)) 1 1 i 1 2 1 1 = 100 i i i 2 2 1 i 1 1 2 i i 1 1 1 1 1 2 75 i 1 i 1 1 1 1 1 i 1 1 4 3 1 1 1 i i 50 1 1 3 2 4 2 i 1 4 2 i 1 1 1 2 6 1 4 1 1 2 3 cS) 1 S 4 i 2 1 i5 8 2 4 2 2 2 1 25 i 22 21 5 2 i 2 3 3 “si (22 & 3 1 5 id 2 13 3 2 & 24 34 5 3 3 2 3 i 6 i 5 20 36 3 3 2 S 6 2 1 1 2 7 Tees 3 1 8 11 1 2 i 2 2 2 4 6 8 10 12 14 Quadrat number Fig. 3. Shrub height profile for 4 m wide 3.5 m deep contiguous quadrats through Site 1 in the Calabasas Quadrangle (number of individuals in 5 cm size class are plotted). Sampling began and ended in pure grassland although for presentation only the first and last quadrats with shrub species present are included. Shrub height distribution at Sites 8 and 9 followed closely the pattern observed at Site 1 (Fig. 3). Discussion We believe that Well’s (1962) model of grassland dynamics applies well to the Los Angeles Basin. He argued that the vegetation patterns in the coastal ranges of California are determined largely by different tolerances to disturbance and by different colonizing abilities. Grasslands represent the rapid-recovery extreme of a continuum of resilience. This community persists under intensive grazing and frequent fires. Coastal sage scrub. due to its subligneous character, falls between grassland and chaparral with respect to its resilience to disturbance. Coastal sage recovers rapidly after fire from resprouts, although seedling establishment is poor in the first post-fire season (Keeley and Keeley 1984). Fires in successive or LOS ANGELES BASIN GRASSLANDS 23 Table 3. Shrub density in nine coastal sage sites in the Calabasas Quadrangle. Relative density (%) Species Site: 1 2 3 4 5 6 7 8 9 Salvia leucophylla 92 60 85 26 85 82 99 79 5 Artemisia californica 1 35 14 3 2 18 1 6 67 Haplopappus venetus 5 4 1 71 15 28 Malacothamnus fasciculatum 1 13 Eriogonum fasciculatum 2 Shrub seedlings/m? 2.1 0.3 53) 133.5) <0.1 5.9 0.2 <0.1 2.9 alternate years are highly destructive to coastal sage scrub species (Zedler et al. 1983). Chaparral regenerates after fire with resprouts and seedlings but requires the longest fire-free period to maintain itself (Keeley 1981). The Los Angeles Basin in prehistorical times was likely dominated to a great extent by ligneous formations such as chaparral and oak or walnut woodlands. Natural lightning fires were a source of wildfires although there is reason to believe the frequency of such disturbance was lower than at present (Keeley 1982). Even then, boundaries between communities were probably dynamic, due to the ran- dom nature of wildfires or disturbances from the large grazing fauna (Stock 1956). Grasslands at that time may have been dominated by Stipa pulchra, particularly on heavier clay soils, although annuals were undoubtedly important components (Wester 1981, Bartolome and Gemmill 1981). Early human occupation of coastal California was accompanied by an increase in fire frequency (Knowles 1953) and a concomitant shift in the dominant vegetation (Aschmann 1959, Heusser 1978). When the first Europeans entered California, grasslands were noted and it ap- pears from diaries such as Fray Juan Crespi’s that they were commonly in close proximity to Indian Villages and maintained through frequent Indian burning (Bolton 1927; Timbrook et al. 1982). With European occupation came most of the annual grasses and forbs which now dominate the grasslands. These species were adapted to frequent disturbances such as fires and grazing through millenia of selection under such conditions in mediterranean Europe (Naveh 1967). Not only were these species resilient to disturbances, they were aggressive colonizers of such sites (Stebbins 1965). As land use intensified (Mooney and Dunn 1972) grasslands expanded. In the Los Angeles Basin during the second half of the 19th century, demand for grazing sites increased resulting in the conversion of brush covered sites to herbaceous vegetation (Burcham 1957). Historically, type conversion of this sort has been achieved by repeatedly burning coastal sage or chaparral (Sampson 1944; Armold et al. 1951). These sites are readily colonized by grassland species and with suc- cessive fires their dominance increases. The process begins as a gradual thinning due to the fact that a portion of the resprouting shrub population, for all resprouting species, is killed in each successive fire and obligate seedling species require a decade or more to reach reproductive age (Keeley and Zedler 1978). In the past 5O years land use practices have changed; in some areas grazing is less extensive and fire protection more effective. These changes have resulted in shrub reestablishment. Initially, coastal sage species are favored over chaparral 24 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES 200 175 150 125 1 e S me 100 1 1 i 1 oa 1p) ene ® 1 1 ae: 2 1 @ 75 1 1 gi ha} 1 1 1 1 ie sl SS NE 2 1 sit 1 A 1 50 aie 5) 3) ace von eee 2 i Ge Wee Te ie anes 1 ee GA etek 19e 7y ole ye re kos 5; 3) ss) Go lee seers 12 13 6) Swets 17 z6 8 14 48 4 6 8 10 Quadrat number Fig. 4. Shrub height profile for Site 4 in the Calabasas Quadrangle. species due to their higher dispersability (Wells 1962) and seeds which germinate readily without fire (Keeley and Keeley 1984). Such patterns of coastal sage in- vasion into grasslands following reduced disturbance were evident in several quadrangles (e.g., Thousand Oaks, Lake Mathews, and Laguna Beach) and similar patterns have been described by others (Haines 1966; McBride 1974; Westman 1976; Oberbauer 1978). The detailed studies of coastal sage demography patterns in the Calabasas Quadrangle (Fig. 3) suggest that a 10 year fire frequency, coupled with intensive grazing, are sufficient to inhibit sage invasion into grassland. The hypothesis that disturbances such as grazing inhibit shrub invasion is at odds with the opinion of some (e.g., Dodge 1975 and others), who have suggested that grazing has enhanced coastal sage scrub invasion into grasslands in Southern California. Oberbauer (1978), however, points out that this opinion is based on extrapolation from other ecosystems such as desert grasslands where it has been demonstrated that grazing favors the spread of unpalatable spiny shrubs. Indeed, Oberbauer (1978) argues that one should not expect the same effect in coastal sage : grassland areas since the shrub species are not as noxious as thorny desert shrubs. Wells (1962) provided good evidence of the inhibitory effect of grazing LOS ANGELES BASIN GRASSLANDS 25 on coastal sage scrub and suggested that factors other than direct predation of seedlings could be involved, e.g., soil compaction and trampling. McBride and Heady (1968) also demonstrated that grazing inhibits the invasion of scrub vege- tation into annual grasslands. Cessation of grazing and fires in their Berkeley Hills sites resulted in a return of scrub vegetation and a concomitant reduction in those grasslands. Similar conclusions were reached for Point Reyes Peninsula in north- ern California by Elliot and Wehausen (1974), for Santa Cruz Island by Brumbaugh (1980), and for the Santa Monica Mountains in southern California by Hobbs (1983). Certain parts of the Los Angeles Basin, possibly due to increased population density, have experienced increased fire frequency. Here, grasslands have ex- panded at the expense of brush (particularly evident in Oat Mtn— Fig. 2, Calabasas, and Santa Susana quadrangles). The patterns and processes involved in the vegetation mosaic in the Los Angeles Basin today are not unlike those in primeval times. The extent of grasslands has undoubtedly increased and the species composition has changed. The concept of a climax vegetation in this instance must be tempered with a recognition of the dynamic nature of the vegetations. Even so, successional patterns will be influ- enced to varying extents by species tolerances to such site characteristics as ele- vation, topography, aspect and substrate. Literature Cited Arnold, K., L. T. Burcham, R. L. Fenner, and R. F. Grah. 1951. Use of fire in land clearing. Calif. Agric., 5(3):9-11, 5(4):145. 13. 15(5):11-12, 5(6):13-15, 5(7):6, 15. Aschmann, H. 1959. The evolution of a wild landscape and its persistence in southern California. Assoc. Amer. Geogr., Ann. 49 (Suppl. No. 3, Pt. 2):34—-56. Bartolome, J. W., and B. Gemmill. 1981. The ecological status of Stipa pulchra (Poaceae) in Cali- fornia. Madrono, 28:172-184. Bolton, H. E. 1927. Fray Crespi-Missionary explorer on the Pacific Coast 1769-1774. Univ. Calif. Press, Berkeley. 402 pp. Brumbaugh, R. W. 1980. Recent geomorphic and vegetal dynamics on Santa Cruz Island, California. Pp. 139-158 in The California islands. Proceedings of a multidisciplinary symposium. (D. M. Power, ed.), Santa Barbara Mus. of Nat. Hist., Santa Barbara, Calif. 787 pp. Burcham, L. T. 1957. California range land; An historic ecological study of the range resources of California. State of Calif., Dept. of Nat. Res., Div. For. 261 pp. Clements, F. E. 1934. The relict method in dynamic ecology. J. Ecol., 22:39-68. Cooper, W. S. 1927. The broad-sclerophyll vegetation of California. Carnegie Inst. of Wash. Publ. No. 319. 124 pp. Cox, G. W. 1980. Laboratory manual of general ecology. William C. Brown Co., Dubuque, Iowa. Dodge, J. M. 1975. Vegetational changes associated with land use and fire history in San Diego County. Ph.D. Dissertation, Univ. Calif. Riverside. 216 pp. Elliot, H. W., III., and J. D. Wehausen. 1974. Vegetational succession on coastal rangeland of Point Reyes Peninsula. Madrono, 22:231-238. Fusch, R. D. 1968. Irvine Ranch: An analysis of changing land use patterns. M.S. Thesis, San Diego State Univ., San Diego, Calif. 120 pp. Goode, S. 1981. The vegetation of La Jolla Valley. M.S. Thesis, Calif. State Univ., Los Angeles. 45 pp. Haines, B. L. 1966. Invasion of grasslands and their inhibition by Artemisia californica Less. M.S. Thesis, Univ. Calif., Santa Barbara. Heady, H. F. 1977. Valley grasslands. Pp. 491-514 in Terrestrial vegetation of California. (M. G. Barbour and J. Major, eds.), Wiley-Interscience, New York. Heusser, L. 1978. Pollen in the Santa Barbara Basin, California: A 12,000-yr record. Geol. Soc. Amer. Bull., 89:673-678. 26 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES Hobbs. E. R. 1983. Factors controlling the form and location of the boundary between coastal sage scrub and grassland in southern California. Ph.D. Dissertation, Univ. Calif., Los Angeles. Kay, B. L., R. M. Love, and R. D. Slayback. 1981. Discussion: Revegetation with native grasses. I. A disappointing history. Fremontia, 9(3):11-15. Keeley. J.E. 1981. Reproductive cycles and fire regimes. Pp. 231-277 in Proceedings of the conference fire regimes and ecosystem properties. (H. A. Mooney, T. M. Bonnicksen, N. L. Christensen, J. E. Lotan and W. A. Reiners. eds.), USDA Forest Service, General Technical Report WO- 26. 594 pp. 1982. Distribution of lightning and man-caused wildfires in California. Pp. 431—437 in Proceedings of the symposium on dynamics and management of mediterranean type ecosystems. (C. E. Conrad and W. C. Oechel, eds.), USDA Forest Service, General Technical Report PSW- 58. Berkeley. 637 pp. —. and S. C. Keeley. 1984. Postfire recovery of California coastal sage scrub. Amer. Midl. Natur., 111:105—-117. ,and P.H.Zedler. 1978. Reproduction of chaparral shrubs after fire: A comparison of sprouting and seedling strategies. Amer. Midl. Natur., 99:142-161. Knowles, C. 1953. Vegetation burning by California Indians as shown by early records. Pamphlet 16, Fire Vol. 28, Forestry Library, Univ. Calif., Berkeley. Leak, W. B.. and R. E. Graber. 1974. A method for deleting migration of forest vegetation. Ecology, 55:1425-1427. McBride. J. R. 1974. Plant succession in the Berkeley Hills, California. Madrono, 22:317—380. Mooney. H. A.. and E. L. Dunn. 1972. Land-use history of California and Chile as related to the structure of the sclerophyll scrub vegetations. Madrono, 21:305-319. Naveh, F. 1967. Mediterranean ecosystems and vegetation types in California and Israel. Ecology, 48:445-459. Oberbauer, A. T. 1978. Distribution dynamics of San Diego County grasslands. M.S. Thesis, San Diego State Univ. Calif. Sampson, A. W. 1944. Plant succession on burned chaparral lands in northern California. Univ. Calif. Agric. Exper. Stn., Berkeley, Bull. No. 685. 144 pp. Stebbins, G. L. 1965. Colonizing species of the native California flora. Pp. 173-195 in The genetics of colonizing species. (H. G. Baker and G. L. Stebbins, eds.), Academic Press, New York. Stock, C. 1956. Rancho LaBrea—A record of Pleistocene life in California. Los Angeles Co. Mus. Nat. Hist.. Sci. Ser. No. 20 Paleon No. 11 (6th ed). Timbrook, J., J. R. Johnson, and D. D. Earle. 1982. Vegetation burning by the Chumash. J. Calif. Great Basin Anthrop., 4:163-186. Wells, P. V. 1962. Vegetation in relation to ecological substratum and fire in the San Luis Obispo Quadrangle, California. Ecol. Monogr., 32:79-103. Wester, L. 1981. Composition of native grasslands in the San Joaquin Valley, California. Madrono, 28:231-241. Westman, W.E. 1976. Vegetation conversion for fire control in Los Angeles. Urban Ecol., 2:119- iS 72 Zedler, P. H., C. R. Gautier, and G. S. McMaster. 1983. Vegetation change in response to extreme events: The effect of a short interval between fires in California chaparral and coastal scrub. Ecology. 64:809-818. Accepted for publication 4 December 1985. Bull. Southern California Acad. Sci. 86(1), 1987, pp. 27-33 © Southern California Academy of Sciences, 1987 A New Species of Hippolytid Shrimp from the West Coast of Mexico Mary K. Wicksten Department of Biology, Texas A&M University, College Station, Texas 77843 Abstract.— Thor algicola n. sp. is described from the Gulf of California and southwestern Mexico. Related to 7. manningi and T. floridanus of the warmer western Atlantic, T. algicola can be distinguished by the spines on its first per- eopods and its larger size. T. algicola usually lives on rocky bottoms, often among algae. Species of Thor are shallow-water hippolytid shrimps living in warm-temperate and tropical marine areas. During studies of the caridean fauna of the Gulf of California, a common species of Thor was found among algae. Chace (1972) reported 7. manningi-from the eastern Pacific, but the species from western Mexico was considerably larger than 7. manningi and had a more robust rostrum. I originally identified the species as T. paschalis (Heller) (Wicksten, 1983). Further study of the eastern Pacific species, T. manningi, and T. paschalis indicated that the eastern Pacific species was in fact undescribed. The new species is described herein. Thor algicola new species Figs. 1-3 Thor manningi.—Chace, 1972:137 (part: the Islas Tres Marias specimens).— ?Rios and Carvacho 1982:459. Thor paschalis (Heller): Wicksten 1983:24—25. (Not Hippolyte paschalis Heller 1862). Diagnosis. —Small hippolytid shrimp lacking supraorbital spine. Anterior mar- gin of carapace with antennal spine only, rounded elsewhere. Spine at base of stylocerite. Ischium of first pereopod bearing small spine on distal flexor margin, two small spinules on flexor margin of merus. Dactyl of third pereopod bearing two stout claws and 3-4 spinules, merus with one stout distal spine. Merus of fourth pereopod with one stout subdistal spine, merus of fifth pereopod with one or no stout subdistal spine. Free-living, not commensal with cnidarians. Description. —Rostrum at most barely reaching anteriorly as far as distal margin of first segment of antennular peduncle, reaching pigmented area of eye, armed with one spine on dorsal midline of carapace and 3-6 spines on dorsal surface of rostrum proper, one ventral spine on rostrum in line with tip, giving rostrum bifid appearance. Rostrum of larger females slightly arched over eye, deeper than rostrum of males or immature females. Spacing of rostral teeth variable. Eyes large, pigmented; with ocellus. Carapace with prominent antennal spine, otherwise rounded on anterior margin. 27 28 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES Fig. 1. Thor algicola n. sp. Female in lateral view. Carapace length 5.1 mm. Specimen from Bahia Bocochibampo, Mexico. Scale in all figures is 1 mm. No supraorbital spine or prominence, but slight supraorbital ridge visible in largest individuals. Pleura of abdominal segments 1-3 rounded, those of fourth and fifth segments with posterolateral points, sixth segment with distolateral point. Sixth segment longer than fifth. Telson with 4-5 pair prominent posterolateral spines, 3 pair terminal spines and long setae. Stylocerite nearly as long as third segment of antennular peduncle, with curved spine at base. First segment of antennular peduncle with spine on ventromesial Margin, setae at distal margin, about 0.66 length of stylocerite. Statocyst ob- solescent. Second segment of antennular peduncle with sharp distolateral spine. Third segment with subtriangular dorsal scale. Dorsolateral flagellum stout and heavily setose, ventromesial flagellum thin and whip-like. Second antenna with sharp distolateral spine on basicerite. Carpocerite less than 0.5 x length of antennal scale. Antennal scale broad, about 3 x as long as wide, blade greatly exceeding tooth. Mandibles asymmetrical. Molar process swollen, armed with numerous teeth; incisor process narrow, armed with 5 teeth. First maxilla with slender lower endite, broad upper endite and bilobed palp. Second maxilla with large scaphognathite, well developed palp, bilobed upper endite and small lower endite. First maxilliped with exopod, two-jointed palp, and bilobed epipod. Second maxilliped with ex- opod, podobranch, and epipod. Third maxilliped with setose terminal segment, ending in sharp dark spines. Penultimate segment about 0.4 x length of ultimate, A NEW SPECIES OF THOR FROM WESTERN MEXICO 29 Fig. 2. Thor algicola n. sp. A, frontal region of female in dorsal view; B, telson and uropods; C, third maxilliped; D, first pereopod; E, second pereopod; F, third pereopod of female; G, third pereopod of functional male; H, second pleopod of male. with tuft of setae on mesial distal margin. Antepenultimate segment with tooth and stiff setae on distal margin, about equal to ultimate segment. Exopod and small epipod present. First pereopod stout, chelate. Dactyl about 0.4 x length of propodus, with tuft of setae at end. Fixed finger with tufts of setae, row of setae on outer proximal margin of propodus. Carpus about 0.5 x length of propodus. Merus shorter than propodus, with 2 small spines on proximal flexor margin. Ischium about 0.5 x merus, with small distal spine on flexor margin. No epipod or exopod. Second pereopod slender, chelate. Fingers of chelae with tufts of setae. Dactyl about 0.3 x length of propodus. Carpus of 6 divisions, the sixth the longest. Merus shorter than carpus, ischium shorter than merus. No epipod. Third pereopod of female with slender dactyl, 3 x as long as wide; tip with two stout darkly pigmented claws, that on flexor margin the broader, and 3-4 spines on flexor margin. Propodus 4 length of dactyl, with 10-14 spinules occurring singly or in pairs on flexor margin, tufts of setae on extensor margin. Carpus about 0.3 x length of propodus. Merus about equal to propodus, with prominent dis- tolateral spine. Ischium about 0.4 length of merus. Third pereopod of male prehensile, subchelate, barely overreaching scaphocerite. Dactyl with bifid tip and 9 closely appressed spines on flexor margin. Propodus 3 x length of dactyl, distal third of flexor margin converging toward extensor margin, armed with spinules; carpus about 0.5 x propodus, merus 1.5 x propodus, with distolateral spine, is- chium without spine. No epipod in either sex. Fourth and fifth pereopods similar to third of female in both sexes. Fourth pereopod with one meral spine, fifth pereopod with one or no meral spine. 30 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES Fig. 3. Thor algicola n. sp. A, second maxilliped; B, first maxilliped; C, second maxilla; D, first maxilla; E, mandible. Female with appendix interna on second pleopod. Male with first pleopod smaller than other pleopods. Male second pleopod with setose appendix mas- culina. Uropods longer than telson, fringed with setae. Outer branch of uropod with small distolateral tooth and movable spine. Holotype. —Female, ovigerous, total length in millimeters 18.4, carapace length 5.1. Bahia Bocochibampo, Guaymas, Sonora (27°57'N, 111°02’W), 5 m, among Sargassum sp., 19 June 1978, Alex Kerstitch, Allan Hancock Foundation (Uni- versity of Southern California) type number 786. Paratypes.—Cholla Bay, Sonora, on sand, 16 Aug. 1966, Tom and Beatrice Burch, 2 specimens.—S. shore, Isla Tiburon, Gulf of California, shore, among shingle, 25 Jan. 1940, Velero III sta. 1045-40, 2 specimens.—Isla San Nicolas, Sonora, 20 m, rubble, 2 July 1978, Alex Kerstitch, 1 specimen.—Isla San Pedro Nolasco, Sonora, 10 m, rocks, 23 Dec. 1978, A. Kerstitch, 1 male.— Punta Doble, Sonora, 25 m, under rock, 25 June 1983, Alex Kerstitch, 1 specimen (dissected). — Isla Candelero, Sonora, 20 m, among rocks, 2 Jan. 1984, Alex Kerstitch, 1 spec- imen, California Academy of Sciences.— Bahia San Carlos, Sonora, 20 m, rocks and sand, 28 Dec. 1982, A. Kerstitch, 1 specimen.— Bahia Bocochibampo, Guay- mas, 5 m, 19 June 1978, A. Kerstitch, 4 ovigerous females in addition to holo- type.— Bahia San Gabriel, Isla Espiritu Santo, Gulf of California, shoal, among coral, 14 Feb. 1940, Velero III sta. 1110-40, 2 females.— Bahia San Gabriel, Isla Espiritu Santo, 2 m, coral, 15 March 1949, Velero IV sta. 1737-49, 36 specimens, 1 of them male.— Zihuatanejo, Guerrero, shore, among rocks, 11 June 1979, R. C. Brusca, 4 specimens.— Bahia Santa Lucia, Acapulco, Guerrero, 2-8 m, mud A NEW SPECIES OF THOR FROM WESTERN MEXICO 31 and sand, 1-2 Feb. 1954, Velero IV sta. 2596-54, 15 specimens, 2 of them males. — San Lorenzo Rocks, Acapulco, 0—4 m, rocks, 30 Jan. 1954, Velero IV sta. 2591- 54, 15 specimens.—Islas Tres Marias, Mexico, 1927, 4 specimens, United States National Museum (USNM).—Bahia Pinas, Panama, 4—7 m, coral, 29 Jan. 1935, Velero ITI sta. 444-35, 1 ovigerous female (USNM).—Bella Vista, Panama, shore, rock, 2 Feb. 1935, Velero III sta. 445-35, many specimens (USNM). Except as noted, all specimens are in the collection of the Allan Hancock Foundation, University of Southern California. Size range. — Total lengths 8-18.4 mm, carapace lengths 1-5.9 mm. Males with prehensile third pereopods: carapace lengths 1—2.4 mm. Ovigerous females: car- apace lengths 1.6-5.9 mm. Color in life.—Mottled with brown and white, closely resembling algae in the habitat. Lines of chromatophores on carapace. Tail fan with white stripe across proximal dorsal surface. Legs banded with brown and white. Very small individ- uals translucent. Etymology.—The specific epithet means “‘dwelling in algae,” referring to the common habitat of the shrimp. Remarks. —Thor algicola resembles seven other species of Thor in lacking su- praorbital spines. Unlike 7. paschalis, it has a proximal tooth on the stylocerite. In 7. intermedius Holthuis, there is only one pair of minute dorsal spines on the telson; 7. algicola usually has five pair. Thor marguitae Bruce has three—four meral spines on the third pereopod; 7. algicola has only one spine. Thor dobkini Chace has one or two spines on the distal half of the flexor margin of the merus of the first pereopod, T. algicola has at most two spinules on the proximal margin of the merus of the first pereopod. Thor amboinensis (De Man), a commensal of cnidarians, can be distinguished easily from T. algicola in life by its distinctive pattern of large spots, as well as by the angular anterolateral margin of its carapace and the relatively shorter dactyls of its third-fifth pereopods. Of these related species, only 7. amboinensis has been reported from the eastern Pacific (Abele and Patton 1976). Thor algicola is most closely related to T. floridanus Kingsley and T. manningi Chace. Neither of these species has two spinules on the merus of the first pereopod. In T. algicola, the distal claws of the dactyl of the third pereopod of the female are stronger and the claw on the flexor margin tends to be broader than in the other two species, the merus of the third pereopod does not bear more than one spine, the distal margin of the antepenultimate segment of the third maxilliped often bears a stout tooth, the telson often bears five, not three or four, pair of dorsolateral spines, and the cornea and eyestalk are relatively smaller, not equal in length to more than half the length of the entire carapace. However, both the spinules of the merus of the first pereopod and other distinguishing features tend to be missing or difficult to see in smaller animals. The major difference between Thor algicola and the two most nearly related species is size. Chace (1972) gave a carapace length for females of T. floridanus of up to 2.3 mm and TJ. manningi of at most 2.5 mm. The majority of ovigerous females of T. algicola are over 3.0 mm long, with 16 of those examined over 4.0 in carapace length. Males also are larger: Chace (1972) reported the males of T. floridanus as measuring 1.3—1.6 mm and the majority of functional males of T. manningi in the carapace length range of 0.9-1.4 mm, while in T. algicola, func- 32 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES O5+ Sex not determined Functional male =| Ovigerous female 20+ L] Female, not ovigerous Number of Animals Carapace Length in Millimeters Fig. 4. Size distribution of specimens of Thor algicola, by carapace lengths. tional males range from 1.0—2.6 mm in carapace length, with only one animal with a length less than 2.5 mm. The distribution of all of the specimens of 7. algicola is given in Fig. 4. Like 7. manningi, T. algicola may be an imperfect protandric hermaphrodite, with males occurring only in the smallest size ranges. Sexing of the majority of small specimens was not attempted in this study, so males without prehensile third pereopods were not distinguished from females at smaller sizes. Chace (1972) and Rios and Carvacho (1982) reported 7. manningi from the eastern Pacific. I examined the specimens from Mexico which Chace (1972) as- signed to 7. manningi. I consider these specimens to fall within the variation for T. algicola. I suspect that the specimens taken by Rios and Carvacho also belong to T. algicola, although I have not had the opportunity to examine material from their area of study. It is difficult to distinguish specimens of 7. manningi, known from the Carib- bean and western Atlantic, from smaller individuals of 7. algicola, the eastern Pacific species. The two populations of shrimp, however, have been isolated from each other since the closing of the Panamic seaway, about 3.1—3.5 million years ago (Glynn and Wellington 1983). The differences in the sizes of the animals and slight differences in morphology suggest genetic change on either side of the Pan- amic land mass. Further studies of the genetics, behavior and color patterns of these shrimp would be useful in determining how closely related these seeming “sibling species’ are to one another. Acknowledgments I thank Fenner A. Chace, Jr., Smithsonian Institution, for helpful comments and the loan of specimens; A. J. Bruce, Northern Territories Museum of Arts and Sciences, for the loan of specimens of Thor paschalis, and Helen Finney, Texas A&M University, for preparing the illustrations. A NEW SPECIES OF THOR FROM WESTERN MEXICO 33 Literature Cited Abele, L. G., and W. K. Patton. 1976. The size of coral heads and the community biology of associated decapod crustaceans. J. Biogeogr., 3:35—47. Chace, F. A., Jr. 1972. The shrimps of the Smithsonian-Bredin Caribbean Expeditions with a summary of the West Indian shallow-water species (Crustacea: Decapoda: Natantia). Smiths. Contr. Zool. No. 98. 179 pp. Glynn, P. W., and G. M. Wellington. 1983. Corals and coral reefs of the Galapagos Islands. Berkeley: Univ. Calif. Press. 330 pp. Rios, R., and A. Carvacho. 1982. Caridean shrimps of the Gulf of California. I. New records, with some remarks on amphiAmerican distribution. Pac. Sci., 36(4):459-465. Wicksten, M. K. 1983. A monograph on the shallow water caridean shrimps of the Gulf of California, Mexico. Allan Hancock Monogr. Mar. Biol. No. 13. 59 pp. Accepted for publication 29 May 1986. BIOLOGY OF THE WHITE SHARK Memoir #9 Papers from a symposium held by the Southern California Academy of Sciences. Contents include material on shark distribution, ecology, age and growth, visual system, hematology, cardiac morphology, feeding, temperature, heat production and exchange, and attack behavior. Make check or money order payable to Southern California Academy of Sciences, and mail to: SOUTHERN CALIFORNIA ACADEMY OF SCIENCES 900 Exposition Blvd. Los Angeles, CA 90007 I would like to order copies of the “Biology of the White Shark” papers at the price of $22.50 per copy. Enclosed is my check for $____________ . Please ship my order to me as follows: Name Address CO hl a i ee ict ei ARG Y —pe 287.4)1 0) Bull. Southern California Acad. Sci. 86(1), 1987, pp. 34-40 © Southern California Academy of Sciences, 1987 Population Genetics of an Introduced Species: Bairdiella icistius in the Salton Sea Richard Beckwitt Department of Biology, Framingham State College, Framingham, Maine 01701 Abstract.—The bairdiella, Bairdiella icistius (Jordan and Gilbert), in the Salton Sea are the progeny of a successful introduction made in the 1950’s. They form a central link in a productive and heavily utilized sportfishery. Genetic variability at 56 enzyme and protein loci was studied by means of starch and polyacrylamide gel electrophoresis. Eight loci were polymorphic, and all polymorphic loci were close to Hardy-Weinberg equilibrium. Average heterozygosity per locus was 0.043, which is within the typical range for marine fish species. There appeared to be fewer rare alleles than expected in a population at equilibrium. The Salton Sea habitat is harsh and changing. The amount of genetic variability present in the bairdiella is an indication of the ability of this population to adapt to environ- mental change. Any introduction of a species into a new habitat constitutes an experiment on the significance of the founder effect. The introduction of the bairdiella, Bairdiella icistius (Jordan and Gilbert), into the Salton Sea is an example of a successful founder event. From an initial population of 67 individuals, the population rapidly increased into the millions. The case of the bairdiella is particularly amenable to study because the population has been well documented since its inception. The date, number, and point of origin of the founders is well known, and the course of development of the population has been followed regularly up to the present. The bairdiella is a small fish of the family Sciaenidae. The fish is abundant in the Gulf of California, where it is fished commercially (Berdegue 1956). Two introductions of bairdiella were made into the Salton Sea. In 1950, 57 fish were transplanted from San Felipe, Baja California, Mexico, and in 1951, an additional 10 fish were taken from the same location (Whitney 1961). These 67 fish represent the total known introduction. Whitney estimated that juvenile fish that were spawned from the first planting reached reproductive maturity within two years, and that the population of bairdiella was several million by 1954. The bairdiella is currently one of the most abundant species in the Salton Sea. It not only supports a major sportfishery itself, but is the main forage fish for the larger introduced game fishes (e.g., orangemouth corvina, Cynoscion xanthulus, Walker et al. 1961). The Salton Sea contains a simplified and truncated food web, with a few species dominating. The total biomass in the lake is large, and supports an active sportfishery. The number of angler hours and the catch per unit effort are among the highest of any inland body of water in California (Black 1983). Since the Salton Sea was formed in 1904-1907, the level of the lake has fluc- tuated considerably. The salinity has increased to a present level of 38 parts per 34 GENETICS OF BAIRDIELLA 35 Table 1. Enzymes, electrophoretic buffers, tissue source, and number of loci scored for Bairdiella icistius. Buffer systems: POUL = discontinuous Borate-Tris-citrate, ““Poulik’’ (Selander et al. 1971); TC 7 = Tris-citrate-EDTA, pH 7.0 (Ayala et al. 1973); TC 8 = Tris-citrate, pH 8.0 (Selander et al. 1971, electrode buffer diluted 1/5); TBE 9.2 = Tris-borate-EDTA, pH 9.2 (Ayala et al. 1973); EBT = Tris-borate-EDTA, pH 8.6 (Whitt et al. 1976); Acryl = acrylamide gel (see Methods). Tissue sources: B = brain, E = eye, H = heart, L = liver, M = muscle. Locus E:C. Tissue Enzyme abbreviation number Buffer source Adenylate kinase Ak-A 2.7.4.3 INC 7/ Alcohol dehydrogenase Adh-A 1.1.1.1 EBT Aldolase Ald-A 4.1.2.13 INC 7/ Aspartate aminotransferase S-Aat-A 2.6.1.1 INC 7/ M-Aat-A Creatine kinase Ck-A Dol oP INC 7/ Ck-B Esterase Est-1 — Acryl Est-2 Est-3 Est-4 Est-5 Est-6 Est-7 Est-8 Fumarase Fum-A 4.2.1.2 TBE 9.2 General proteins Gp-1 — Acryl Gp-2 Gp-3 Gp-4 Gp-5 Gp-6 Gp-7 Gp-& Gp-9 Gp-10 Gp-11 Gp-12 Gp-13 Gp-14 Glucosephosphate isomerase Gpi-A 5.3.1.9 POUL Gpi-B Glucose-6-phosphate dehydrogenase G6pdh-A 1.1.1.49 TC 8 G6pdh-B Glyceraldehyde-3-phosphate dehydrogenase Gapdh-A 1.2.1.12 EBT Gapdh-B Glycerol-3-phosphate dehydrogenase G3pdh-A 1.1.1.8 TC 8 G3pdh-B Hexokinase Hk-A 2 L-Iditol dehydrogenase Iddh-A il. Isocitrate dehydrogenase S-Icdh-A 1 M-Icdh-A Lactate dehydrogenase Ldh-A 1.1.1.27 POUL Ldh-B Ldh-C Malate dehydrogenase S-Mdh-A Ie leSy/ ew S-Mdh-B TC 8 14 TC 8 2 INC 7 —_— ~) —- —) ARR [ole ol ee Me i We Ie fay my fa foe 9) HL FOE) OE) I RR RK I a fa og to Om eI eh 36 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES Table 1. Continued. Locus EC. Tissue Enzyme abbreviation number Buffer source Malic enzyme S-Me-A 1.1.1.40 TC 8 M Mannosephosphate isomerase Mpi-A 5255168 TC8 H Peptidase Pep-A — EBT B Pep-B L Pep-C L Pep-D L Phosphoglucomutase Pgm-A Dahan POUL M Pgm-B L Phosphogluconate dehydrogenase Pgdh-A 1.1.1.43 TC7 L Superoxide dismutase Sod-A TetSstet TBE 9.2 L Xanthine dehydrogenase Xdh-A [PT 37/ Tey, L thousand (ppt) total dissolved solids, and there are continuing inputs of pesticides and fertilizers from the surrounding agricultural land. The latter inputs. coupled with high temperatures in summer, result in periods of anoxia and fish kills. Several fishes are still reproducing in the Salton Sea at this time, but it is not clear what the time course or significance of continued environmental change will be. There are presently several plans under consideration for the management of the Salton Sea, as a result of which salinity could remain constant or rise as high as 88 ppt (Black 1983). Salinities in excess of 50 ppt are considered to be harmful to the continued reproduction of fishes in the Salton Sea. The Salton Sea is a harsh and changing environment. Knowledge of the genetic structure of the bairdiella can help predict the likely consequences of continued environmental degradation on a major resource. One possible consequence of a founder event or any population bottleneck is the loss of genetic variation from the population. It is this reservoir of genetic variability that provides the material upon which selection operates, as the population adapts to environmental change. Allozyme electrophoresis can be used to examine genotypes and allele frequencies for a large number of loci that encode for enzymatic or structural proteins. Even if these loci are not themselves significant for adaptation to environmental changes. they are the best estimate of variation in the whole genome, including those loci that can respond to selection. Methods Both starch and polyacrylamide gel electrophoresis were used to examine en- zyme or protein variation. A total of 24 enzyme or protein stains were used. which resolved the products of 56 loci. The enzymes examined, the electrophoretic conditions, the number of loci resolved and their abbreviations are given in Table 1. Starch gel enzyme electrophoresis followed Selander et al. (1971) with the following exceptions: The stains for ADH, AK, CK, FUM, GAPDH, LDH, and MPI were from Allendorf et al. (1977). Peptidase staining and nomenclature followed Frick (1983). The ALD stain was from Shaw and Prasad (1970). the AAT stain from from May et al. (1979), and the ME stain was from Ayala et al. (1973). All stains that used G6PDH as an intermediate were modified after Buth GENETICS OF BAIRDIELLA 37 Table 2. Allele frequer.ries, sample sizes and mean heterozygosities per locus (H) for eight poly- morphic loci of Bairdiella icistius from the Salton Sea. Alleles are named in terms of increasing anodal mobility. Locus and sample size Est-3 Est-6 Gp-12 S-Icdh-A Ldh-C S-Mdh-B Pep-D Pgm-B Allele 100 102 102 102 101 100 101 99 1 0.140 0.039 0.069 0.020 0.149 0.385 0.574 0.338 2 0.860 0.794 0.931 0.980 0.851 0.615 0.426 0.662 3 0.167 H 0.242 0.342 0.128 0.039 0.254 0.476 0.491 0.450 and Murphy (1980). The agar overlay method of Brewer (1970) was used for all stains that included an enzyme intermediate. Starch gels were 10% (w/v) Sigma starch. Polyacrylamide gel electrophoresis was used to resolve esterases and general proteins. Esterases were run on 6% (w/v) acrylamide in the TBE buffer of Maniatis et al. (1982) and stained after Selander et al. (1971) using beta-napthyl acetate as the substrate. General proteins were run on split gels (4-8% w/v) using the dis- continuous buffer system of Laemmli (1970) without SDS, and stained with Coo- massie Blue. Data analysis was done with the aid of the BIOSYS-1 package of computer programs (Swofford and Selander 1981). Fish were collected by gill-net from two locations on the north side of the Salton Sea (North Shore Marina and Bombay Beach), with the cooperation of the Cal- ifornia Department of Fish and Game. Fish were packed on dry ice and transported to the laboratory, where they were stored whole at —20°C until used for electro- phoresis. Results There is substantial genetic variability in the Salton Sea population of bairdiella. Eight loci are polymorphic: Est-3, Est-6, Gp-12, S-Icdh-A, Ldh-C, S-Mdh-B, Pep-D, and Pgm-B. Allele frequencies for these loci are given in Table 2. The average heterozygosity per locus is H = 0.043 + 0.016 (S.E.). To put this in perspective, Winans (1980) reviewed the literature on natural populations of 82 species of fishes, for which the mean H was 0.048. Data from two other sciaenids, Genyonemus lineatus and Seriphus politus, can also be compared at 33 loci studied in common (Beckwitt 1983). The H estimates are 0.030, 0.043 and 0.046 for Genyonemus, Seriphus, and Bairdiella respectively. All polymorphic loci are close to Hardy-Weinberg equilibrium (Table 3), implying that inbreeding or differential survival of genotypes is not now a significant factor. Although the average heterozygosity per locus is typical for a fish, the pattern of allele frequencies is not. There are only two loci at which there are rare alleles (frequency less than 0.05). Typically this class is more numerous (Nei 1975). In a sample of 100 diploid individuals, the probability of finding an allele is 0.99 if its true frequency is 0.05; if the true frequency is 0.01, the probability is 0.87. Thus, all but the rarest alleles have been well sampled. The average number of all alleles per locus is 1.16. Using an analysis similar 38 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES Table 3. Exact test of goodness of fit to Hardy-Weinberg equilibrium. Common homo- Common/rare Locus zygotes heterozygotes Other genotypes Probability Est-3 75 22 3 0.339 Est-6 64 34 4 1.000 Gp-12 89 12 1 0.380 S-Icdh-A 98 4 0 1.000 Ldh-C 73 26 2 1.000 S-Mdh-B 39 45 16 0.673 Pep-D 31 54 16 0.419 Pgm-B 42 47 10 0.656 to Bryant et al. (1981), one can estimate the expected number of alleles per locus in a population at equilibrium. This estimate is highly dependent on the value used for N.v (N. = effective population size, v = mutation rate per gene per generation). This value can not be directly estimated from the data, but is likely to be about 0.1 (Nei 1975). Under those conditions, the expected number of alleles in a sample of 100 fish is 1.57. Although this estimate is not likely to be very accurate, it does suggest that alleles could have been lost during the founder event. One can also compare the pattern of heterozygosity per locus with the expected pattern for a large population at equilibrium. When the data for the bairdiella are compared to the theoretical distribution obtained by Fuerst et al. (1977), there is no significant difference (Kolomogorov-Smirnov test, P > 0.2). One cannot ex- clude the null hypothesis that the Salton Sea population is at equilibrium. The data are consistent with the known facts of the introduction of the bairdiella into the Salton Sea. The founding population was relatively large (67 individuals) and the bottleneck was of short duration—there were at least 1,000,000 fish in reproductive condition within 3 years. Under these conditions, one would expect the new population to retain nearly all of the heterozygosity of the parent pop- ulation, but to lose a large number of rare alleles. Discussion The founder effect has been discussed at some length in the literature of the- oretical population genetics. It is often cited as an important mechanism for speciation (see Carson and Templeton 1984 for review). When a founder popu- lation is subject to a severe bottleneck in numbers, there is a chance for the loss or alteration of genetic variability. Nei et al. (1975) discussed the genetic conse- quences of a population bottleneck. They showed that there will be a decrease in heterozygosity that is dependent not only on the size of the bottleneck, but also on the rate of population growth after the bottleneck. They also showed that the loss of neutral alleles from the population is more dependent on the size of the bottleneck and less so on its duration. These results have since been extended and generalized by Chakraborty and Nei (1977), Sirkomma (1983), and Maruyama and Fuerst (1984). There have been few explicit tests of the genetic consequences of a population bottleneck. Several studies have observed decreased genetic variability in periph- eral or relict populations, and inferred a bottleneck or founder event as the cause GENETICS OF BAIRDIELLA 39 (Avise and Selander 1972; Kat 1982; Larruga et al. 1983). One striking example is the elephant seal, hunted to near extinction and now without any genetic variation demonstrable by enzyme electrophoresis (Bonnell and Selander 1974). Introduced species have been studied as examples of founder events. Bryant et al. (1981) studied the face fly, an introduced pest. They found a non-significant decrease in heterozygosity at 14 electrophoretic loci between North American and European populations, but a loss of up to 50% of the alleles in the introduced populations. Turner (1984) examined artificial refugium populations of the desert pupfish, and showed little difference from the parent populations, in terms of heterozygosity or number of alleles. The bairdiella of the Salton Sea harbor considerable genetic variability, equal to most natural populations of marine fishes. Thus, there is considerable potential for adaption to environmental change. However, the possibility that rare alleles are lacking from the Salton Sea population may be important. If the bairdiella are called upon to adapt to radically changing physical factors, some rare allele could be vital. Such speculation can only be verified with data from the parent population. Acknowledgments I would like to thank personnel of the VANTUNA Research Group, particularly M. Matsui and S. Smith, as well as G. Black of the California Department of Fish and Game for help in collecting fishes. M. Matsui also contributed her considerable knowledge of the natural history of the Salton Sea. I would also like to thank D. Buth for advice on enzyme nomenclature. This study was performed under con- tract from Southern California Edison Company, and is a contribution of the VANTUNA Research Group, Occidental College, Los Angeles, California. Literature Cited Allendorf, F. W., N. Mitchell, N. Ryman, and G. Stahl. 1977. Isozyme loci in brown trout (Salmo trutta L.): detection and interpretation from population data. Hereditas, 86:179-190. Avise, J. C., and R. K. Selander. 1972. Evolutionary genetics of cave-dwelling fishes of the genus Astyanax. Evolution, 26:1-19. Ayala, F. J., D. Hedgecock, G. S. Zumwalt, and J. W. Valentine. 1973. Genetic variation in Tridacna maxima, an ecological analog of some unsuccessful evolutionary lineages. Evolution, 27:177- 191. Beckwitt, R. 1983. Genetic structure of Genyonemus lineatus, Seriphus politus (Sciaenidae) and Paralabrax clathratus (Serranidae) in southern California. Copeia, 1983:691-696. Berdegue, J. 1956. Peces de importancia comercial en la costa nor-occidental de Mexico. Comision para el Fomento de la Piscicultura Rural, Mexico. Black, G. F. 1983. The Salton Sea and the push for energy exploitation of a unique ecosystem. Paper presented at Annual Meeting, California-Nevada Chapter, American Fisheries Society, Ana- heim, California. Bonnell, M. L., and R. K. Selander. 1974. Elephant seals: genetic variation and near extinction. Science, 184:908-910. Brewer, G. J. 1970. An introduction to isozyme techniques. Academic Press, New York. Bryant, E. H., H. van Dijk, and W. van Delden. 1981. Genetic variability of the face fly, Musca autumnalis de Geer, in relation to a population bottleneck. Evolution, 35:872-881. Buth, D. M., and R. W. Murphy. 1980. The use of nicotinamide adenine dinucleotide (NAD)- dependent glucose-6-phosphate dehydrogenase in enzyme staining procedures. Stain Technol- ogy, 55:73-76. 40 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES Carson, H. L.,and A.R. Templeton. 1984. Genetic revolutions in relation to speciation phenomena: the founding of new populations. Ann. Rev. Ecol. Syst., 15:97-131. Chakraborty, R., and M. Nei. 1977. Bottleneck effects of average heterozygosity and genetic distance with the stepwise mutation model. Evolution, 31:347-356. Frick, L. 1983. Anelectrophoretic investigation of the cytosolic di- and tripeptidases of fish: molecular weights, substrate specificities, and tissue and phylogenetic distrubtions. Biochem. Genet., 21: 309-322. Fuerst, P. A., R. Chakraborty, and M. Nei. 1977. Statistical studies on protein polymorphism in natural populations. I. Distribution of single locus heterozygosity. Genetics, 86:455—483. Kat, P. W. 1982. The relationship between heterozygosity for enzyme loci and developmental homeostasis in peripheral populations of aquatic bivalves (Unionidae). Am. Nat., 119:824— 832. Laemmli, U.K. 1970. Cleavage ofstructural proteins during the assembly of the head of bacteriophage T4. Nature, 227:680-685. Larruga, J. M., V. M. Cabrara, A. M. Gonzalez, and A. Gullen. 1983. Molecular and chromosomal polymorphism in continental and insular populations from the south-western range of Dro- sophila subobscura. Genetica, 60:191—206. Maniatis, T., F. E. Fritsch, and J. Sambrook. 1982. Molecular cloning: a laboratory manual. Cold Spring Harbor Laboratory, New York. Maruyama, T., and P. A. Fuerst. 1984. Population bottlenecks and nonequilibrium models in population genetics. I. Allele numbers when populations evolve from zero variability. Genetics, 108:745-763. May, B., J. E. Wright, and M. Stoneking. 1979. Joint segregation of biochemical loci in the Sal- monidae: results from experiments with Sa/velinus and a review of the literature of other species. J. Fish. Res. Bd. Canada, 36:1114—1128. Nei, M. 1975. Molecular population genetics and evolution. Elsevier/North Holland, New York. , 1. Maruyama, and R. Chakroborty. 1975. The bottleneck effect and genetic variability in populations. Evolution, 29:1-10. Selander, R. K., M. H. Smith, S. Y. Yang, W. E. Johnson, and J. B. Gentry. 1971. Biochemical polymorphism and systematics in the genus Peromyscus. I. Variation in the old-field mouse (Peromyscus polionotus). Studies in Genetics VI. Univ. Texas Publ. No., 7103:49-90. Shaw, C. R., and R. Prasad. 1970. Starch gel electrophoresis of enzymes—a compilation of recipes. Biochem. Genet., 4:297-—320. Sirkomma, S. 1983. Calculations of the decrease of genetic variation due to the founder effect. Hereditas, 99:11—20. Swofford, D. L., and R. B. Selander. 1981. BIOSYS-1:a FORTRAN program for the comprehensive analysis of electrophoretic data in population genetics and systematics. J. Hered., 72:281-283. Turner, B. J. 1984. Evolutionary genetics of artificial refugium populations of an endangered species, the desert pupfish. Copeia, 1984:364—-369. Walker, B. W., R. Whitney, and G. W. Barlow. 1961. The fishes of the Salton Sea. California Dept. of Fish and Game, Fish Bulletin, 113:77-91. Whitney, R. 1961. The bairdiella, Bairdiella icisitus (Jordan and Gilbert). California Dept. of Fish and Game, Fish Bulletin, 113:105-151. Whitt, G. S., W. F. Childers, J. B. Shaklee, and J. Matsumoto. 1976. Linkage analysis of the multilocus glucosephosphate isomerase system in sunfish (Centrarchidae, Teleostei). Genetics, 82:35—42. Winans, G. 1980. Geographic variation in the milkfish Chanos chanos. I. Biochemical evidence. Evolution, 34:558—-574. Accepted for publication 5 June 1986. Bull. Southern California Acad. Sci. 86(1), 1987, pp. 41-53 © Southern California Academy of Sciences, 1987 The Chiton Fauna of Cocos Island, Costa Rica (Mollusca: Polyplacophora) with the Description of Two New Species Antonio J. Ferreira We regret to announce the death of Antonio J. Ferreira on 19 May 1986. Inquiries and requests for his scientific work may be sent to the Department of Invertebrate Zoology and Geology, California Academy of Sciences, San Francisco, California 94118. Abstract. —Six species of chitons are here recognized at Cocos Island, Costa Rica, including two new species, an [schnochiton and an Acanthochitona. The reported presence on the island of Chiton goodallii and Acanthochitona hirudi- niformis has not been corroborated. Cocos Island lies at 5°32'N, 87°04'W, some 500 kms (300 miles) southwest of Punta Arenas, Costa Rica, 600 kms (350 miles) northeast of the Galapagos Islands. At the eastern edge of the Indo-Pacific, in the path of the North Equatorial Countercurrent, and within the reach of currents from western American shores, the fauna of Cocos Island is of particular interest. Although Cocos is considered to be part of the Panamanian Province (Briggs 1974), its marine fauna has a considerable admixture of Indo-West Pacific species. With respect to mollusks, Hertlein (1963) found that among 88 species recorded at the island, 88% are Panamic, 6% Indo-Pacific, and 6% endemic; Emerson and Old (1964), Montoya (1983), and Shasky (1983, 1985), reported comparable observations. This report on the chiton fauna of Cocos Island is based mostly upon material obtained by Donald R. Shasky (DRS) (Apr. 1983; Mar. 1984; May, 1985) and myself (AJF) (Nov. 1984), from aboard the schooner VICTORIA; additional material was found in the collections of the California Academy of Sciences, San Francisco, California (CAS), U.S. National Museum of Natural History, Wash- ington, D.C. (USNM), and the Los Angeles County Museum of Natural History (LACM). Six species of chitons are here recognized at Cocos Island, Costa Rica, two new to science: Chiton stokesii Broderip, 1832 Placiphorella blainvillii (Broderip, 1832) Stenoplax boogii (Haddon, 1886) Lepidozona rothi Ferreira, 1983 Ischnochiton victoria Ferreira, spec. nov. Acanthochitona shaskyi Ferreira, spec. nov. The reported presence of two other species at Cocos Island—Chiton goodallii Broderip, 1832, and Acanthochitona hirudiniformis (Sowerby, 1832)—are con- sidered to be either misidentifications, or represent such rare and unlikely occur- 41 a 42 SOUTHERN CALIFORNIA ACADEMY OF SCIENCES rences that they cannot be accorded permanent status in the chiton fauna of the island. Systematic Treatment Polyplacophora Gray, 1821 Order Neoloricata Bergenhayn, 1955 Suborder Ischnochitonina Bergenhayn, 1930 Family Ischnochitonidae Dall, 1889 Ischnochiton Gray, 1847a Type species: Chiton textilis Gray, 1828, by subsequent designation (Gray 1847b). Ischnochiton victoria Ferreira, spec. nov. Figs. 1-7 Diagnosis: Very small chitons, up to 3.5 mm long. Valves brightly colored in reddish tones, often with touches of brown or blue; round-backed, not carinate, not beaked; posterior valve remarkably inflated, mucro central, inconspicuous, postmucro convex, at sharp slope. Articulamentum white; sutural laminae sharp, small, separated by wide sinus; teeth small, sharp; slits 9/10-1-9/10. Girdle scales small (up to 100 um) with 12-14 striations. Radula major lateral teeth tricuspid. Type material and locality: Holotype (CAS 061096), entire specimen, off Roca Sucia, near Wafer Bay, Cocos Island, Costa Rica (5°33’N, 87°02’W) at 24 m; 17 loose valves (CAS 061097; LACM 2124; D. R. Shasky Colln.) designated as paratypes, 16 found at same locality as holotype, in sand, at 18-34 m, | also found in sand, 1.5 km NE of Manuelita Id., off Cocos Island, at 91-95 m. Description: Holotype about 3.5 mm long, elongate, round backed; valves very thin, not carinate, not beaked, posterior edges straight. Tegmentum in reddish tones mottled with white, particularly on valve v, with dark brown and blue markings on pleural areas. Anterior valve broken, 1/3 missing. Sculpture of an- terior valve and lateral areas of intermediate valves limited to 6-8 coarse con- centric rugosities (Figs. 1 and 2); lateral areas hardly elevated; central areas uni- formly sculptured with minute, roundish pits, 7-8 um in diameter, close together (Fig. 3). Posterior valve very elevated, inflated (Fig. 4 and 5); mucro inconspicuous, about central; postmucro sloping sharply, convex, with vague concentric rugosi- ties. Articulamentum translucent, colors of tegmentum showing through; sutural laminae short, very thin, sharp; sinus very wide; relative width of sinus on valve vill (width of sinus / width of sutural laminae), 1.2; insertion teeth short, sharp; slits 8-1-10. Girdle dorsal surface covered with imbricate, flat scales, up to 110 um long, with 12-14 striations (Fig. 6A); girdle ventral surface with transparent, rectangular scales, 30 < 15 wm (Fig. 6B), arranged in columns. Radula (Fig. 7) 1.3 mm long comprising some 35 rows of mature teeth; median tooth 11 um wide at anterior blade, narrowing posteriorly; first lateral teeth about 38 um long, widely concave at outer border, with prominent knob at antero-lateral corner; spatulate teeth, simple, 55 wm long; major lateral teeth with tricuspid head, 25 um wide; outer marginal teeth 30