3108 HARVARD UNIVERSITY Library of the Museum of Comparative Zoology The Great Basin Naturalist VOLUME 45, 1985 EDITOR: Stephen L. Wood Published At Brigham Young University, by Brigham Young University TABLE OF CONTENTS Volume 45 Number 1 - 31 January 1985 Spatial patterns of plant communities and differential weathering in Navajo National Monument, Arizona. Jack D. Brotherson, William E. Evenson, Samuel R. Rush- forth, John Fairchild, and Jeff'rey R. Johansen 1 Cryptogamic soil crusts: seasonal variation in algal populations in the Tintic Mountains, Juab County, Utah. Jeffrey R. Johansen and Samuel R. Rushforth 14 Aquatic parameters and life history obser\'ations of the Creat Basin spadefoot toad in Utah. Peter Hovingh, Bob Benton, and Dave Bornholdt 22 New species of Astragalus (Leguminosae) from Mesa County, Colorado. Stanley L. Welsh 31 A fourth species of Oreoxis (Umbelliferae). Stanley L. Welsh and Sherel Goodrich 34 Insect communities and faunas of a Rocky Mountain subalpine sere. David J. Schimpf and James A. MacMahon "^"^ Nutrients in Carexexserta sod and gravel in Sequoia National Park, California. Raymond D. Ratliff" 61 Mites (excluding chiggers) of mammals of Oregon. John O. Whitaker, Jr., and Chris Maser 67 Food of cougars in the Cascade Range of Oregon. Dale E. Toweill and Chris Maser. ... 77 Factors influencing nesting success of burrowing owls in southeastern Idaho. Richard S. Gleason and Donald R. Johnson 81 Note on the diet of long-billed Curlew chicks in western Idaho. Roland L. Redmond and Donald A. Jenni 85 Tundra vegetation of three circjue basins in the nortliern San Juan Mountains, Colorado. Mary Lou Rottman and Emily L. Hartman 87 Use of biomass predicted by regression from cover estimates to compare \ egetational similarity of sagebrush-grass sites. L. David Humphrey 94 A new combination and a new variety in Artemisia tridcntata . Sherel Goodrich, E. Durant McArthur, and Alma H. Winward 99 Understory response to tree harvesting of singleleafpinxon and I' tab juniper. Richard L. Everett and Steven H. Sharrow 105 Acjuatic birds of the White River, LHntah Countv, Utah. Benjamin B. Steele and Stephen B. N'ander Wall ' 113 Patterns of macroiuvertebrate colonization in an intermittent Rock>' Mountain stream in Utah. J. Vaun McArthur and James R. Barnes 117 Checklist of the mosses of (Jrand Teton National Park and Teton (]ount\, Wxoming. John R. Spence 124 Ecological investigation of a suspected spawning site of C'olorado scjuawfish on the Yampa River, Utah. N'incent A. Lamarra, Marianne C. Lamarra, and John G. Carter .... 127 Dillcicntial effects of cattle and sheep grazing on high mountain meadows in the Straw- berry Valley of central Utah. J. B. Shupe and Jack D. Brotherson 141 Unusual social feeding and soaring b\ the Conunou l^axcu (Co/ri/s corax) . Clhutou M. White and Merle- Tanner-White 150 rliicc additional cases ol picdalion b\ niagpies on small uiamiiials. Kerr\' P. Reese .... 152 Number 2 - 30 April 1985 Utah flora: Saxifragaceae. Sherel Goodrich I55 Utah's rare plants revisited. Stanley L. Welsh and L. Matthew Chatterley 173 New records and comprehensive list of the algal taxa of Utah Lake, Utah, USA. Samuel R. Rushforth and Lorin E. Squires 237 Host-parasite studies o( Trichophrya infesting cutthroat trout {Salmo clarki) and long- nose suckers {Catostomus catostomus) from Yellowstone Lake, Wyoming. R. A. Heckmann and T. Carroll 255 New synonymy and new species of bark beetles (Coleoptera: Scolytidae). Stephen L. Wood 266 New Nevada entities and combinations in Eriogonum (Polygonaceae). James L. Reveal 276 Growth and reproduction of the flannelmouth sucker, Catostomus latipinnis, in the Upper Colorado River Basin, 1975-76. Charles W. McAda and Richard S. Wydoski 281 Burrowing Owl foods in Conata Basin, South Dakota. James G. MacCracken, Daniel W. Uresk, and Richard M. Hansen 287 Addendum to the distribution of two herptiles in Idaho. Timothy D. Reynolds and William F. Laurance 291 Nesting and predatory behavior of some Tachysphex from the western United States (Hymenoptera: Sphecidae). Nancy B. Elliott and Frank E. Kurczewski 293 Pollinators of Astragalus monoensis Barneby (Fabaceae): new host records; potential impact of sheep grazing. Evan A. Sugden 299 Vegetational and geomorphic change on snow avalanche paths. Glacier National Park, Montana, USA. David R. Butler 313 Effectiveness of the seed wing of Firms flexilis in wind dispersal. Ronald M. Lanner. . . . 318 Habitat relationships of the blackbrush community {Coleogyne ramosissima) of south- western Utah. James Callison and Jack D. Brotherson 321 Size selection of food by cutthroat trout, Salmo clarki, in an Idaho stream. William D. Skinner 327 Aspects of the biology of the flathead chub (Hybopsis gracilis) in Montana. William Gould 332 Number 3 - 31 July 1985 Quaternaiy paleontology and paleoecology of Crystal Ball Cave, Millard County, Utah: with emphasis on mammals and description of a new species of fossil skunk. Timothy H. Heaton 337 First record ofClimacia californica (Neuroptera: Sisyridae) and its host sponge, Ephyda- tia mulleri (Porifera: Spongillidae), from Idaho with water quahty relationships. William H. Clark 391 Poa L. in New Mexico, with a key to middle and Southern Rocky Mountain species (Poaceae). Robert J. Soreng 395 Dwarf mistletoe-pandora moth interaction and its contribution to ponderosa pine mortal- ity in Arizona. Michael R. Wagner and Robert L. Mathiasen 423 Occurrence of anisakid larvae (Nematoda: Ascardidia) in fishes from Alaska and Idaho. Richard Heckmann and Terry Otto 427 Soil algae of cryptogamic crusts from the Uintah Basin, Utah, U.S.A. John Ashley, Samuel R. Rushforth, and Jeffrey R. Johansen 432 In memoriam: William Wallace Newby (1902-1977). William H. Behle 443 Symbos cavifrons (Artiodactyla: Bovidae) from Delta County, Colorado. Jerry N. McDonald 455 Comparisons of prescribed burning and cutting of Utah Marsh plants. Loren M. Smith and John A. Kadlec ^62 New species and records of North American Pityoplitlwnis (Coleoptera: Scolytidae), Part IV: the Scriptor group. D. E. Bright 467 New species and new records of North American Fityophtlwrus (Coleoptera: Scolytidae), Part V: the Juglandis group. D. E. Bright 476 Second nesting record and northward advance of the Great-tailed Crackle {Qiiiscahis mexicanus) in Nevada. Jennifer A. Holmes, David S. Dobkin, and Bruce A. Wilcox 483 New species of Talinum (Portulaceae) from Utah. N. Duane Atwood and Stanley L. Welsh 485 Types of Nevada buckwheats {Erio'^onum : PoK gonaceae). James L. Reveal 488 Annotated key to Eho^onum (PoK gonaceae) of Nevada. James L. Reveal 493 High rates of photosvnthesis in the desert shrub Chnj.sothamniis naiiseosus ssp. albi- caulis. Tim D.' Davis, N. Sankhla, W. R. Andersen, D. |. Weber, and B. N. Smith 520 F'ood habits of the western whiptail lizard {Cnemidophorus ti' L. Welsh 548 New species of Astragalus (Leguminosae) from southeastern Utah. Rupert C. Barneby and Stanley L. Welsh ' 551 New Sclerocactus (Cactaceae) from Nevada. Stanley L. Welsh and Kaye Hugie Thorne 553 Succession in pinyon-juniper woodlands following wildfire in the Great Basin. Susan Koniak 556 Use of radio transmitter implants in wild canids. Jeffrey S. Green, Richard T. Golightly, Jr., Susan Lyndaker Lindsey, and Brad R. LeaMaster 567 Number 4 -31 October 1985 Life history of the cui-ui, Chasmistes cujus ('ope, in P\raniid Lake, Ne\ada: a review. William F. Sigler, Steven Vigg, and Mimi Bres 571 Helminth parasites of the white-tailed jackrabbit, Lcpus towuscndi , from northwestern ("olorado and southern Wyoming. Larry M. Shults and Lora C Rickard 604 Thermal ecology and activitx' patterns of the short-horned lizard {Phnpiosoma douglassi) and the sagebrush lizard {Sccloporus ek, Okla- homa, a western < )/ark lootlulls stream. ( .'. Stan lodd and Kenneth W. Stexxart 721 Checklist of vascular plants for the Bighorn Canyon National Recreation Area. Robert W. Lichvar, Ellen I. Collins, and Dennis H. Knight 734 Presettlement vegetation of part of northwestern Moffat County, Colorado, described from remnants. William L. Baker and Susan C. Kennedy 747 Winter preference, nutritive value, and other range use characteristics of Kochia pros- trata (L.) Schrad. James N. Davis and Bruce L. Welch 778 Age, growth, and food habits of tui chub, Gila bicolor, in Walker Lake, Nevada. James J. Cooper 784 New variety of Yucca harrimaniae (Agavaceae) from Utah. Elizabeth Neese and Stanley L. Welsh 789 Revision of the Phlox austromontana (Polemoniaceae) complex in Utah. Stanley L. Welsh 79]^ Index 793 HE GREAT BASIN NATURALISl Diume 45 No. 1 31 January 1985 Brigham Young University MUS. COMP. ZOOL I IPPARY MAY ' 3 1985 HARVARD UNlVEfUMTY GREAT BASIN NATURALIST Editor. Stephen L. Wood, Department of Zoology, 290 Life Science Museum, Brigham Young University, Provo, Utah 84602. Editorial Board. Kimball T. Harper, Chairman, Botany; James R. Barnes, Zoology; Hal L. Black, Zoology; Stanley L. Welsh, Botany; Clayton M. White, Zoology. All are at Brig- ham Young University, Provo, Utah 84602. Ex Officio Editorial Board Members. Bruce N. Smith, Dean, College of Biological and Agricul- tural Sciences; Norman A. Darais, University Editor, University Publications. Subject Area Associate Editors. Dr. Noel H. Holmgren, New York Botanical Garden, Bronx, New York 10458 (Plant Taxonomy). Dr. James A. MacMahon, Utah State University, Department of Biology, UMC 53, Lo- gan, Utah 84322 (Vertebrate Zoology). Dr. G. Wayne Minshall, Department of Biology, Idaho State University, Pocatello, Idaho 83201 (Aquatic Biology). Dr. Ned K. Johnson, Museum of Vertebrate Zoology and Department of Zoology, Uni- versity of California, Berkeley, California 94720 (Ornithology). Dr. E. Philip Pister, Associate Fishery Biologist, California Department of Fish and Game, 407 West Line Street, Bishop, California 93514 (Fish Biology). Dr. Wayne N. Mathis, Chairman, Department of Entomology, National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560 (Entomology). Dr. Theodore W. Weaver III, Department of Botany, Montana State University, Boze- man, Montana 59715 (Plant Ecology). The Great Basin Naturalist was founded in 1939 and has been published from one to four times a year since then by Brigham Young University. Previously unpublished manuscripts in English of less than 100 printed pages in length and pertaining to the biological natural his- tory of western North America are accepted. Western North America is considered to be west of the Mississippi River from Alaska to Panama. The Great Basin Naturalist Memoirs was es- tabUshed in 1976 for scholarly works in biological natural history longer than can be accom- modated in the parent publication. The Memoirs appears irregularly and bears no geographi- cal restriction in subject matter. Manuscripts are subject to the approval of the editor. Subscriptions. The annual subscription to the Great Basin Naturalist for private individuals is $16.00; for institutions, $24.00 (outside the United States, $18.00 and $26.00); and for stu- dent subscriptions, $10.00. The price of single issues is $6.00 each. All back issues are in print and are available for sale. All matters pertaining to subscriptions, back issues, or other busi- ness should be directed to Brigham Young University, Great Basin Naturalist, 290 Life Sci- ence Museum, Provo, Utah 84602. The Great Basin Naturalist Memoirs may be purchased from the same office at the rate indicated on the inside of the back cover of either journal. Scholarly Exchanges. Libraries or other organizations interested in obtaining either journal through a continuing exchange of scholarly publications should contact the Brigham Young University Exchange Librarian, Harold B. Lee Library, Provo, Utah 84602. Manuscripts. See Notice to Contributors on the inside back cover. 2-85 650 74822 ISSN 017-3614 The Great Basin Naturalist Published at Provo, Utah, by Brigham Young University ISSN 0017-3614 Volume 45 31 January 1985 No. 1 SPATIAL PATTERNS OF PLANT COMMUNITIES AND DIFFERENTIAL WEATHERING IN NAVAJO NATIONAL MONUMENT, ARIZONA' Jack D. Brotherson,- William E. Evenson,' Samuel R. Rushforth,- John Fairchild,' and Jeffrey R. Johansen .Abstract.— Vegetation patterns in Navajo National Monument, Arizona, were studied over a five-year period from 1977 to 1981. Twelve distinct plant community types occur within the boundaries of the park. These commu- nities are characterized and the dominant plant species of each are recorded. The relationships of parent material, soils, and moisture to plant communities are also discussed. It appears that discrete communities occupy soils of dif- ferent characteristics, particularly with respect to amount of weathering of parent material. Patterns in vegetative cover across the landscape have been reported in the hter- ature since the time of Darwin and before (Weaver and Clements 1938). Such patterns are generally considered to be the result of a long history of development under the in- fluence of both past and present environmen- tal factors (Walter 1973). Several theories have been proposed to explain these patterns. These include the monoclimax ideas of Clem- ents (1936), the individualistic and continuum approaches of Gleason (1939) and Curtis (1955), the vegetational mosaic and environ- mental pattern concepts of Billings (1952) and Whittaker (1953), and the functional fac- torial approach described by Major (1951). Floristic homogeneity is usually associated with imiformity in climate and soil. Con- versely, high habitat diversity often leads to vegetation patterns that are also highly di- verse. Such diverse patterns are usually ex- plained on the basis of climatic, orographic, historic, or edaphic differences (Walter 1973). Unfortunately, broad generalizations are usually of little help in studying the vege- tation ecology of localized areas, since major differences in climate, soils, and topography are not usually present in small geographic areas. Local differences in vegetation can more likely be explained on the basis of mi- crohabitat and historical differences. Differential weathering is a potent force in altering conditions at the microhabitat level (Foster 1971, Birkeland 1974). Two impor- tant types of weathering are usually recog- nized. Mechanical weathering, which is largely caused by the action of wind and wa- ter on the substrate, breaks up preexisting rock into smaller fragments. Chemical weath- ering acts on fragments of all sizes to rear- range the mineralogical composition of the original rock. In addition to these, biological weathering, including that which takes place in the presence of living plants, has also been Shidy completed as part of Contract PX7029-9-0570 from the Southwest Region of the National Park Service. Department of Botany and Range Science, Brigham Young University, Provo, Utah 84602. Department of Physics and .\stronomy, Brigham Young University, Provo, Utah 84602. 'Utah Division of Vvildlife Resources, 1596 West North Temple, Salt Lake City, Utah 84116. Bacteriology and Public Health, Washington State University, Pullman, Washington 99164-4340. Great Basin Naturalist Vol. 45, No. 1 LEGEND ■ MAIN HIGHWAY N DItT ROAD I />-^V. IIVEt e TOWN . ■ RUINS SITi 0 5 10 20 30m.l»s , UTAH ,^ ARIZONA nnehesto I Fii;. 1. Map sliowiiiy; tlu- localitv of Navajo National Monument in northeastern Arizona. shown to be important (Foster 1971, Jenney 1958, Wilding and Drees 1969, and Zinke 1962). The present study was undertaken to de- scribe the plant communities and assess the effects of weathering processes on vegetation in Navajo National Monument. We report below a large diversity of vegetation found within this small geographical area, resulting in 12 major community types whose compo- sition reflects the different substrates pro- duced through differential weathering of the same geological formation. Study Site Navajo National Monument is in north- eastern Arizona, USA (Fig. 1). The mon- ument headquarters are located on the Shonto Plateau near the head of Betatakin Canyon. The monument is made up of three separate units, each of which contains a large "cliff dwelling" of the Anasazi culture. The three units occur in pinyon-juniper slickrock country with many deep-cut canyons and high-walled sandstone cliffs that often reach heights of 380 m above the streambeds. Each unit occurs in a separate canyon. Betatakin and Keet Seel are two of these canyons and belong to the major Tsegi Canyon complex. Nitzin Canyon (Inscription House) is 32 km west of these and is a branch of Navajo Can- yon. A profile diagram of Betatakin Canyon, showing distribution of community types, is illustrated in Figure 2. Figure 3 is a set of vegetation maps of the three segments of the monument. Climatically the area is in the cold temper- ature desert region of the Colorado Plateau. The total annual precipitation ranges from a January 1985 Brotherson et al.: Plant Communities /bOO 7400 >7300 Pinyon- Juniper-Shrub r°° Pinyon- Jumper-Shrub u,*^-"*---^-^ (U 7100 ■§7000 / A'-*^ 0)6900 Li- \ / ^6800 \ Douglas Fir -o ) ,| J 6700 6600 /^Hanging Garden Pinyon- Jumper-Grass Aspen Oak «| Betatakin Creek Scale in Feet 6500 0 1000 2000 3000 7500 7400 7300 7200 7100 7000 6900 6800 6700 6600 6500 Profile diagram of Betatakin Canyon. This profile is taken along the line shown in Figure 3. low of 17 cm to a high of 48 cm, with an av- erage of 29 cm (Brotherson et al. 1978). The period of greatest precipitation is late sum- mer and early fall. Rainfall in the area is of- ten spotty and localized, with cloudbursts oc- curring more often than general rains. Temperatures fluctviate greatly during the seasons, with lows of -23 C in winter and highs of 39 C in summer. The frost-free sea- son ranges from 107 to 213 days, with an av- erage of 155 days. Navajo Sandstone is the major geological formation in all three segments of the mon- ument and is the only exposed rock formation at Keet Seel and Nitzin canyons. At Betata- kin Canyon, the Kayenta Formation is peri- odically exposed in the lower reaches. In all three canyons, Navajo Sandstone cliffs form sheer walls that extend above the creek beds 200 to 250 m. Talus accumulations occur at the base of these cliffs in each canyon. The canyon bottoms are filled with deep deposits of alluvium (Quaternary fill). In Keet Seel and Nitzin canyons these alluvial deposits are deeply eroded. In Betatakin Canyon the allu- vial deposits are less extensive and not as thick. Erosion is a predominant feature of the monument and of the surrounding area. Methods One hundred twenty-nine study plots were selected within the 12 plant community types of Navajo National Monument (Broth- erson et al. 1978). A minimum of four plots was established in each community type. Each study plot measured 10 x 10 m (0.01 ha) and was randomly located within the com- munity being studied. The plots were sub- sampled by 20 regularly placed 0.25 m^ quadrats on a grid in each plot. Observations recorded in each plot included cover of plant species and species groups and physical char- acteristics, including exposure (aspect), slope, and soil depth. Cover was estimated using Daubenmire's (1959) cover classes for all species encoun- tered as well as for the five categories: total Uving cover, cryptogam cover, litter, rock, and bare ground. Plant community composi- tion by life form was obtained by recording species in six life form categories: trees, shrubs, grasses, forbs, annuals, and crypto- gams. Percent cover for all categories was subsequently calculated. Composite soil samples were taken to a depth of 20 cm from the corners and center of each 0.01 ha plot. This depth was consid- ered adequate on the basis of Ludwig's (1969) study of several foothill communities in Utah, which demonstrated that the surface decime- ter of soil yielded 80% of the mineral concen- tration information useful in correlations with plant data. Holmgren and Brewster (1972) also found in a study of desert commu- nities that greater than 60% of the fine roots (those most likely to absorb soil minerals) were concentrated in the upper 20 cm of the soil profile. Soil depth was determined with the use of a 1 m penetrometer. Great Basin Naturalist vol. 4D, i\o. i lilil Aspen ^ Atriplex-Grass Hj Douglas Fir g Mixed-Weed FJ/] Pinyon-Juniper-Grass rn Pinyon-JuniperSage I I Pinyon-Juniper-Mixed Shrub [^ Slickrock - Scattered shrub ^ Streamside Tree ^ Talus Slope Beiatakin Canyon - 200 ha. Fi^. 3. Vee;etati()ii maps ot tlie three units of Navajo National Monument. Keet Seel Canyon - 1 7 ha. MM -rV*^ l(l// >|'ft Inscription House - 1 7 ha. Soil samples were analyzed for texture (Bouyoucos 1951), pH, soluble salts, organic matter, and mineral composition. Soil reac- tion was measured with a glass electrode pH meter. Total soluble salts were determined with a Beckman electrical conductivity bridge. A 1:1 wt/vol soil-water paste (Russell 1948) was used to determine pH and total soluble salts. Organic matter was determined by weight loss after 24 hours at 450 C in a muffle furnace. Soils were extracted with 1.0 normal ammonium acetate for the analysis of calcium, magnesium, potassium, and sodium (Jones 1973). Zinc, manganese, iron, and cop- per were extracted from the soils with DTPA (Lindsay and Norvall 1969). Soil phosphorus was extracted with sodium bicarbonate (Ol- sen at al. 1954). Total nitrogen analysis was carried out using macrokjeldahl procedures (Jackson 1958). Ion concentrations were de- termined using a Perkin-Klmer Model 403 atomic absorption spectrophotometer (Isaac andKerber 1971). Plant nomenclature follows McDougall (1973). Prevalent species of the various plant communities are reported. The prevalent species list includes as many species as the average number of species per 0.01 ha sam- pling area examined (Warner and Harper 1972). Diversity values were computed using the formula: B = l/2pj-, where B is the di- versity index and p; is a measure of the rela- tive abundance of a species in a given habitat (Levins 1966, MacArthur 1972). Cluster analysis (Sneath and Sokal 1973) was applied to similarity index values (Ru- zicka 1958) using unweighted pair-group clustering (UPGMA). The UPGMA method computes the average similarity of each unit to the cluster using arithmetic averages. It is widely used and has been found to introduce less distortion than other methods (Kaesler and Cairns 1972). Gradient analysis was used to assess the in- fluence of weathering on soil types within the plant communities. Thirteen soil parame- ters were selected and treated as gradients. The abiotic properties evaluated were pH, soluble .salts, organic matter, calcium, magne- sium, potassium, sodium, zinc, iron, copper, manganese, phosphorus, and nitrogen. There were 12 values for each gradient, one for January 1985 Brotherson et al.: Plant Communities each of the communities. The resulting gradients were divided into five equal seg- ments. Each plant community was assigned to one of the gradient segments according to its mean for that factor and given a rating from one to five corresponding to the gradient segment. The community ratings were summed across all gradients to yield an index of overall position with respect to all 13 gradients. Linear regression analyses (Cochran and Snedecor 1976) were applied to the biotic and abiotic factors to determine the degree to which these factors were associated. As- pect data were transformed according to Beers et al. (1966) to allow aspect to be used as an independent variable in statistical anal- ysis. Northeast aspects were given a value of 2.0, southwest aspects a value of 0.0, and in- termediate aspects varied from 2.0 to 0.0 in both directions. Results and Discussion Plant Communities Twelve community types were encoun- tered ill the monument (Brotherson et al. 1978) (Table 1, Figs. 2 and 3). Two of these, oak and pinyon-juniper-mixed shrub, were foimd in all segments of the monument. Six of the remaining 10 community types were found in two of the monument areas, and four of the types were restricted to a single monument segment. Betatakin Canyon and its surroimding areas included more variabili- ty of vegetation types than the other seg- ments of the monument. This was partially due to the favorable moisture regiun-s cre- ated by the canyon geology and Navajo Sandstone hydrology. Plant communities are described below. /. Talus slope community. This community was in the monument at the base of large Navajo Sandstone cliffs. Because of differen- tial weathering patterns of the sandstone, it was often found in pockets above canyon bottoms. This community was characterized by having many large and small rocks in its soil. The soil varied in depth from a few inches to several feet. Floristically the com- munity was highly diverse and contained all life forms but was dominated by shrubs and grasses. The species Atriplex canescens, Bou- teloiia gracilis, and Sporobolus cryptandrus were especially important. //. Atriplex-grass community. This com- munity was found in two segments of the monument but was most highly developed at Inscription House on deep deposits of allu- vial fill in the canyon bottom. Vegetatively this community type was similar to the talus slope type but was less diverse and showed only half as many prevalent species. The two most abundant species were Atriplex canes- cens and Oryzopsis hijmenoides. Heavy over- grazing has contributed to severe wind and water erosion in this community type. ///. Pinyon-juniper-grass community. This community was found only in the bottom of Betatakin Canyon. It occurred at the base of sandstone cliffs on both northern and south- ern exposures, in places generally occupied by the talus slope community. It differed from tlie talus slope type in that the soils Table 1. Vegetative coninuinities of Navajo National Monument and the percentage of land area each monument segment. ipied within Monument Segment Vegetation type Betatakin Keet Seel Inscription House Pinyon-jimiper-sage Pinyon-juniper-grass Aspen Douglas fir Hanging garden Oak Pinyon-juniper-mixed shnib Mixed weed Talus slope .\triplex-grass Slickrock-scattered shrub Stream side tree 49.9 3.1 0.7 0.4 Trace 2.4 43.5 Trace ° Trace Trace 4.8 59.6 13.5 9.3 0.2 12.6 0.1 26.0 11.3 25.8 35.0 1.8 (Note: Trace means <0.1%.) 6 Great Basin Naturalist Vol. 45, No. 1 were finely textured and lacked a rocky com- ponent. It was floristically diverse and had a high degree of cryptogam development. Im- portant species were Bonteloiia gracilis, Stipa cotnata, Artemisia tridentata, Poa longiligula, Juniperus utahensis, and Muhlenbergia pungens. IV. Slickrock-scattered shrub community. This commimity was well developed in two of the monument's segments (Table 1). It characteristically exhibited large areas of ex- posed and unweathered rock. This commu- nity type is associated with Navajo Sandstone in much of southern Utah and northern Ari- zona. Slickrock vegetation grows in pockets in the sandstone created by differential weathering where fine textured materials col- lect. These pockets vary in size from a meter to a hundred or more meters across and con- tain a variety of vegetative forms. This com- mimity had the highest species diversity of all the monimient communities. Important spe- cies were Bouteloua gracilis, Onjzopsis hij- menoides. Ephedra viridis, Cercocarpus in- tricatus, Gutierrezia sarothrae. Yucca angustissima, Vulpia octiflora, Opuntia polij- cantha, and Muhlenbergia pungens. The type showed little or no grazing influence due to the fact that it usually occurred on the tops of rocky rises, bluffs, and cliffs where it was inaccessible to sheep and cattle. V. Pinijon-juniper-mixed shrub commu- nity. This type was found in all segments of the monument and occupied the greatest per- cent of the land base. Geographically, it oc- curred between the slickrock and pinyon- juniper-sage commimities. The type exhibited large areas of unweathered sandstone, though not to the extent found in slickrock sites. Soil development in the type was also more ex- tensive than in slickrock. A major dis- tinguishing factor in the vegetation of this community was the presence of several major shnib species in the understory that were ab- sent or rare in other pinyon-juniper related communities. The most important species were Cercocarpus intricatus, Pinus edulis, Cowania mexicana, Fendlera rupicola, Junip- erus utahensis, and Poa longiligula. VI. Pinyon-juniper-sage community. This type was represented in the monument only at Betatakin, where it occurred adjacent to the pinyon-juniper-mixed shrul) tvpe on the plateau back from the canyon rim. It con- tained the oldest and largest pinyon and juni- per trees. Soils of the area were uniformly distributed with very few rocks and/ or slick- rock areas. Average living cover was the highest of any of the pinyon-juniper types. Pinyon and juniper contributed exclusively to the overstory canopy, whereas the understory varied but was dominated mainly by Arte- misia tridentata. VII. Douglas fir community. This type was found in the Betatakin and Keet Seel seg- ments of the monument. It occurred on northern and northeastern exposures, grow- ing in a narrow band along the base of high sandstone cliffs where water seepage oc- curred. Floristically it was of intermediate di- versity in comparison to other community types in the monument. The overstory and understory were dominated by trees and shrubs, with only 3 of the 14 prevalent spe- cies being grasses or forbs. There was a large amount of exposed rock and bare ground in this community, which occurred on steeper slopes than any other community type except hanging gardens. Pseudotsuga menziesii, Poa longiligula, Pinus edulis, Quercus gambelii, Symphoricarpos vaccinoides, and Aster are- nosus were the most important vascular spe- cies. Moss cover was also significant. VIII. Oak community. This community was found mainly in the Betatakin and Keet Seel parts of the monument (Table 1). In these areas, it was found in the canyon bot- toms below Douglas fir. It occupied the banks of the streams and extended away from them toward the cliff bases, growing mostly on deep deposits of alluvial fill. The canopy was dense and dominated exclusively by Quercus gambelii. The understory varied and included Symphoricarpos vaccinoides, Bromus tectortim, Mahonia repens, Smilacina stellata, TJmlictrum fendleri, and Clematis li- gustici folia. There was a deep litter layer throughout much of the type that was best developed at Betatakin Canyon, where it of- ten reached a depth of 25 cm. IX. Aspen community. This community was found at Keet Seel and Betatakin, al- though it was important only in Betatakin Canyon. The community occupied the cen- tral portion of the canyon along the stream and its banks. It was surrounded bv the oak January 1985 Brotherson et al.: Plant Communities community, and together they dominated the riparian canopy of the canyon bottom. The aspen community was of average floristic di- versity and was dominated in the understory by shrubs and forbs. The most abundant spe- cies were Betula occidentalis, Cornus stolo- nifera, Eqidsetum hyernale, Acer negundo, Symphoricarpos vaccinoides. Clematis ligiis- ticifolia, and Smilacina stellata. X. Mixed weed community. This type was foimd only in Keet Seel Canyon, where it oc- cupied a considerable portion of the canyon bottom. It consisted of a mixture of in- troduced and native weeds and was domi- nated almost exclusively by annuals. The most important species were Salsola iherica, Bromiis tectorum. Sisymbrium altissimum, Cleome serrulata, and Epilobium horne- mannii. It occurred on deep deposits of allu- vial fill and appeared to be a remnant of past overgrazing. XL Hanging gardens. This is an unusual commimity type occurring in the massive sandstone cliff regions of southwestern North America. This community was present in Be- tatakin and Keet Seel canyons. The garden at Betatakin was in the cliff dwellings and was associated with two seepage areas where wa- ter slowly emerged from the rocks. No runoff water was present. The hanging garden at Keet Seel was more mesic. Here the vascular plants grew attached to a seepage wall at the base of the northern part of the cliff dwell- ings. At the wettest portion of the garden a small seep emerged to form a puddle at the base of the hanging garden. With the excep- tion of Rhus toxicodendron, the gardens were dominated exclusively by forbs, including Mimulus eastwoodiae, Aquilegia micrantha, Mentezelia albicaulis, Galium aparine, Smila- cina racemosa, and Habenaria sparsiflora. XII. Stream.side tree community. This type was found at Inscription House in Nitzin Canyon. It was dominated by Populus fre- montii and occurred along the stream bank at the cliff base below the ruin. Understory plants were almost nonexistent due to the high activity of domestic sheep and cattle in the area. The animals concentrated in the area along the stream for both feeding and shading. The soil base of the commimity con- sisted of deep deposits of alluvial fill. Wide- spread erosion occured here due largely to extensive overgrazing. Percent Similarity 0 10 20 30 - Slickrock - Pinyon-Juniper-Grass ■ Talus slope - Atrlplex-Grass - Douglas Fir - Pinyon-Junlper-Mixed Shrub - Pinyon-JuniperSage - Mixed Weed • Aspen ■ Oak Hanging Garden Streamside Tree Fig. 4. Cluster dendrogram showing similarities of plant communities in Navajo National Monument. Ecology Indices of similarity (Ruzicka 1958) were computed for each community in relation to all other communities on the basis of species cover values. The similarity indices were low, ranging from 0% to 31%, with an average of 4.3%. In 52% of the comparisons the percent similarity was zero, whereas only 13 of a pos- sible 66 comparisons showed similarities ex- ceeding 10%. Such low similarities indicate that the community types were highly dis- tinct. The similarity indices were clustered to produce a graphical representation of rela- tionships between different community types (Fig. 4). Cluster patterns provide insight into community relationships within the mon- ument since those communities most closely related share floristic and environmental ele- ments. However, the most striking aspect of our cluster analysis was the dissimilarity be- tween community types. Biotic and abiotic environmental data were lumped by commimity and .summarized (Table 2). There was a great deal of variation in these data, both within and between com- munities. Assuming approximate uniformity in chemical and physical properties of un- weathered Navajo Sandstone parent material, differences in the properties of soils between sites should indicate the influence of weath- ering. To assess this hypothesis, mineral con- centrations in the soils of the plant commu- nities were compared to those in the sandstone (Table 3). Some minerals, including Great Basin Naturalist Vol. 45, No. 1 Table 2. Means and standard deviations of biotic and abiotic community characteristics of each vegetation type. Gradients Talus Atriplex P-J-Mixed Community type slope grass P-J-Grass Slickrock shrub BlOTlC FACTORS Percent hving cover 19.8 12.4 42.1 51.6 15.4 ±10.8 ±9.9 ±38.1 ±4.0 ±6.2 Percent tree cover 0.0 0.0 5.7 1.3 0.0 ±0.0 ±0.0 ±6.2 ± 2.3 ±0.0 Percent shnib cover 17.4 39.1 19.7 .33.3 26.3 ±3.3 ±25.5 ±17.4 ±10.4 ±16.6 Percent grass cover 41.9 16.9 .52.1 33.0 36.5 ±12.8 ±8.2 ±18.9 ±6.0 ±1.5 Percent forb cover 21.2 4.5 13.8 9.4 13.7 ±4.9 ±8.8 ±2.7 ±2.5 ±2.2 Percent annual cover 1.7 31.5 0.0 3.1 0.0 ±2.4 ±28.3 ±0.0 ±2.4 ±0.0 Percent cryptogam cover 0.7 0.1 12.1 5.1 5.5 ±1.0 ±0.2 ±15.3 ±,3.2 ±3.3 Diversity 6.0 3.2 7.0 8.2 2.6 ±1.5 ±1.4 ±1.9 ±1.5 ±1.2 Mean # of species/stand 14.8 6.5 20.3 19.8 14.9 ±4.9 ±1.9 ±2.5 ±.5.4 ±2.9 Gross habitat factors Soil depth (cm) 49.4 51.7 ,54.2 .33.7 20.8 ±31.0 ±8.3 ±19.3 ±11.5 ±7.2 Aspect 0.7 0.9 0.7 0.2 1.6 ±0.9 ±0.1 ±1.0 ± 0.3 ±0.4 Percent slope 24.2 .39.1 20.0 18.3 28.0 ±10.6 ±8.6 ±22.9 ±11.6 ±14.1 Percent litter cover 6.1 11.4 7.0 4.7 25.9 ±1.8 ±6.8 ±5.4 ± 3.4 ±8.8 Percent rock cover .38.1 16.9 10.1 0.3 15.7 ± 16.9 ±5.2 ±17.5 ± 0.5 ±14.9 Percent bare ground 34.1 47.7 .33.5 21.0 ,30.8 ± 14.3 ± 10.9 ±12.3 ± 6.9 ±15.1 Soil factors Percent sand 85.6 85.3 81.7 89.3 83.0 ±1.8 ±1.7 ±8.4 ±0.6 ±,3.6 Percent silt 7.2 8.9 9.3 4.0 12.1 ±1.9 ±1.1 ±4.9 ± 0.0 ±2.8 Percent cla\ 7.2 6.2 8.7 7.0 4.9 ±0.8 ±0.8 ±3.8 ± 0.0 ±1.0 Percent tines 14.4 15.1 18.0 11.0 17.0 ±1.8 ±1.8 ±8.7 ±().{) ± 3.6 pH 7.8 7.9 7.3 7.8 7.4 ±0.1 ±0.1 ± 0.3 ±0.1 ±0.2 Soluble salts 274.4 236.7 205.0 201.0 197.8 ±33.6 ±68.1 ± 129.7 ± 13.8 ±52.2 Percent organic matter 3.2 0.7 .3.1 1.6 3.9 ±0.7 ±0.2 ±2.2 ±0.5 ±1.3 N'itrogiMi (ppm) 0.04 0.03 0.03 0.02 0.05 ±0.01 ±0.01 ±0.01 ±0.01 ±0.02 Phosphorus (ppm) 12.0 9.5 9.3 9.6 8.2 ±2.9 ±2.0 ±1.2 ±1.2 ±4.1 Calcium (pjim) 5708.6 3911.0 .3331.3 1729.7 2674.9 ± 2668.0 ± 832.6 ± 3563.5 ±57.6 ± 1515.3 Magnesium (ppun 120.0 412.6 1(X).0 55.3 ,526.6 ± 45.5 ±61.3 ±39.9 ±6.4 ± 124.4 Potassium ippm) 170.0 280.1 128.3 63..3 102.1 ±69.2 ±115.7 ±44.8 ±7.6 ±30.4 January 1985 Brotherson ET AL.: Plant Communities 9 Table 2. Continued Gradients P-J- Douglas Mixed Hanging Stream Sage fir Oak Aspen weed garden side tree 57.4 25.8 26.2 11.7 66.8 86.0 0.5 ±16.7 ±11.9 ±24.0 ± 13.9 ±.34.6 ±4.1 ±0.8 64.3 1.0 0.0 24.9 0.0 2.9 0.0 ±18.1 ±1.9 ±0.0 ±24.8 ±0.0 ±4.1 ±0.0 22.3 18.8 .36.1 25.7 1.0 2.8 1.8 ±8.3 ±17.1 ±36.7 ±20.6 ±1.7 ±2.1 ±2.9 13.3 47.8 3.2 1.7 8.5 3.5 0.0 ±11.9 ±20.8 ±4.0 ±3.2 ±9.3 ±4.9 ±0.0 1.2 23.5 5.4 34.8 ,58.2 74.1 0.0 ±1.3 ±18.1 ±6.1 ±25.0 ±50.7 ±9.9 ±0.0 0.0 0.1 1.3.8 0.0 32.3 0.0 6.6 ±0.0 ±0.2 ±29.1 ±0.0 ± ,56.0 ± 0.0 ±11.6 6.6 0.5 0.0 0.0 0.0 22.1 0.0 ±5.9 ±0.7 ±0.0 ±0.0 ±0.0 ± 13.8 ±0.0 2.3 3.8 2.8 3.6 2.5 2.1 1.4 ±1.0 ± 1.3 ±0.8 ±0.7 ±0.6 ±1.0 ±0.5 8.6 13.3 10.8 11.3 7.7 7.0 3.3 ±3.0 ±0.7 ±4.3 ±2.7 ±1.5 ±2.8 ±3.2 .32.4 47.9 77.9 37.5 66.6 9.3 33.3 ± 12.9 ± 13.6 ±21.3 ±51.7 ± ,57.7 ± 5.5 ±57.7 1.1 1.0 1.4 1.2 0.7 0.2 1.0 ±0.6 ±0.5 ±0.7 ±0.8 ±0.7 ±0.2 ± 0.9 7.9 .30.6 12.6 10.0 5.3 85.0 23.0 ±7.0 ±2.7 ±4.7 ±10.8 ±4.6 3:7.1 ±.5.3 36.1 33.6 91.3 70.8 16.4 9.3 .34.0 ±20.4 ±17.6 ±9.4 ± 18.3 ±7.6 ±11.8 ±4.9 0.9 22.0 0.1 0.6 0.0 13.1 0.0 ±1.2 ±12.6 ±0.3 ±1.6 ±0.0 ±0.3 ±0.0 43.3 20.2 0.8 3.6 21.3 0.0 64.5 ±16.6 ±14.7 ±1.5 ± 3,2 ±22.1 ±0.0 ±7.0 83.6 83.5 76.1 83.9 84.7 83.5 85.0 ±2.5 ±3.8 ± 3.8 ± 5.3 ±1.5 ±0.7 ±1.7 11.4 9.9 14.4 10.4 9.0 9.0 6.0 ±2.3 ±3.0 ±2.9 ± 4.3 ±1.7 ± 0.0 ±1.0 5.0 6.6 9.7 ,5.9 6.3 7.5 9.7 ±0.8 ±1.1 ±2.0 ±1.6 ±0.6 ±0.7 ±1.2 16.4 16.5 24.1 16.3 15.3 16.5 15.7 ±2.5 ±3.8 ±4.0 ±5.5 ±1.5 ± 0.7 ±1.5 7.2 7.4 6.7 7.2 7.5 7.7 8.0 ±0.2 ±0.3 ±0.4 ±0.2 ±0.1 ±0.1 ±0.2 1.36.1 233.9 ,376.9 286.0 ,349.7 517.0 568.3 ±.36.3 ±47.5 ± 75.2 ±74.8 ± 176.1 ±281.4 ±171.7 2.5 3.4 9.1 5.2 2.9 4.1 2.1 ±1.4 ±1.6 ±3.8 ±4.1 ±1.1 ±1.1 ±0.4 0.05 0.05 0.20 0.10 0.06 0.04 0.04 ±0.01 ±0.02 ±0.08 ±0.08 ±0.01 ±0.02 ±0.01 5.5 11.6 23.4 12.4 15.9 10.3 13.4 ±1.1 ±3.0 ±6.7 ±3.5 ±2.0 ±5.5 ±1.0 1627.5 .3027.8 4.3.34.7 3,588.0 2,336.3 8455.0 .5650.7 ±310.3 ± 1428.2 ± 1245.4 ± 19,53.8 ±615.6 ± 264.5 ± 925.7 645.6 83.3 226.4 108.9 86.7 184.0 272.0 ±33.5 ±36.6 ± 70.3 ±84.1 ±6.7 ±0.0 ± 72.0 88.4 121.9 272.5 181.3 ,303.3 152.5 710.0 ±16.7 ±44.8 ±105.1 ±68.9 ±75.2 ±116.7 ±208.0 10 Great Basin Naturalist Vol. 45, No. 1 Table 2 continued. Gradients Community type Talus slope Atriplex grass P-J-Grass Slickrock P-J- Mixed shrub Sodium (ppm) Iron (ppm) Manganese (ppm) Zinc (ppm) Copper (ppm) 12.1 10.8 10.3 9.7 9.8 ±.3.4 ±0.9 ±2.8 ±1.4 ±3.0 1.0 2.3 1.8 0.7 4.6 ±0..3 ±0.2 ±0.5 ±0.1 ±2.1 1.9 4.4 1.9 1.6 4.3 ±0..5 ± 0..5 ±0.2 ±0.2 ±1.8 0.4 1.0 0.3 0.2 1.3 ±0.2 ±0.2 ±0.1 ±0.1 ±0.4 0.3 0.4 0.3 0.2 0.4 ±0.1 ±0.1 ±0.1 ±0.0 ±0.1 Zn, Na, and Ca, were usually lost through weathering due to mobility and leaching. On the other hand, K, Mg, Fe, Cu, Mn, P, and N were apparently concentrated in the upper levels of the soil profile through weathering. This weathering pattern occurred in 9 of the 12 communities. Our gradient analysis of soil factors re- sulted in three groups of communities (Table 4, Fig. 5). The first group was composed of the slickrock, pinyon-juniper-grass, and talus slope commimity types. The soil of the slick- rock community appeared the least changed when compared to the parent material. This was as expected, since slickrock sites were as- sociated with large areas of exposed sand- stone. The soils of the pinyon-juniper-grass and talus slope communities also tended to show minor changes from parent material. These types were found at the base of sand- stone cliffs, where they receive constant in- put of generally unweathered material from the cliffs above. A second group included communities with soils of intermediate weathering. Within the group, soils of the streamside tree type appeared to be most weathered, and soils of the mixed weed and Douglas fir types were least weathered. Soils of the oak community were the most weathered of any occurring in the monument (Fig. 5). These soils were the most acid, low- est in percent sand, and highest in percent fines, organic matter, litter cover, iron, cop- per, manganese, phosphorus, and nitrogen (Table 2). Oak communities within the mon- ument apparently contribute to the weath- ering process and enhance soil fertility through the addition of organic matter. Soil acidity (Fig. 6) in the 12 community types was positively correlated with organic matter (p<0.001). This result is consistent with the findings of Zinke (1962) that soils are more acid where influenced by the cano- py. Increased acidity from organic matter breakdown likely accelerates the weathering Table 3. Mean concentrations of minerals in Navajo Sandstone parent material together with indicated deviations from these concentrations in soils of the various plant communities. Mineral Mean concentration in parent material (ppm) Number of communities with mean concentration in soils Less than parent material Greater than parent material Zn Na Ca K Mg Fe Cu Mn P N 1.54 23.0 4631.0 111.0 53.0 .54 .14 .69 3.6 Trace January 1985 Brotherson ET AL.: Plant Communities 11 Table 2. Continued Gradients P-J- Douglas Mixed Hanging Stream Sage fir Oak Aspen weed garden side tree 9.1 18.0 14.7 12.1 12.0 118.8 48.0 ±0.6 ±13.1 ±4.2 ±2.7 ±1.3 ± 107.1 ± 19.0 6.7 3.0 10.6 2.8 2.0 0.8 1 2 ±1.5 ±1.6 ±8.7 ±2.1 ±0.1 ±0.4 ±0.2 4.2 2.9 12.9 3.5 3.6 3.8 3.9 ±1.3 ±0.9 ±8.4 ±1.4 ±0.4 ±0.1 ±0.7 1.2 0.6 0.8 0.7 0.8 0.6 0.5 ±0.2 ±0,4 ±0.5 ±0.4 ±0.1 ±0.1 ±0.1 0.5 0.4 0.8 0.4 0.3 0.2 0.3 ±0.1 ±0.1 ±0.3 ±0.1 ±0.1 ±0.0 ±0.1 process considerably (Birkeland 1974). In ad- dition, laboratory and field work have dem- onstrated that chelating agents formed from biological processes in the soil can also bring about significant amounts of weathering when in contact with mineral soil (Schalascha et al. 1967). Indeed, such biotic weathering may exceed that brought about by hydrolysis alone imder some circumstances (Birkeland 1974, Schatz 1963, Jackson and Keller 1970). Physical weathering is important in all community types in the monument. In addi- tion, chemical and biological weathering are important in some communities (particularly those in groups II and III of Fig. 5). Biotic weathering appears to be most important in the oak commimity type. These weathering processes constitute a potent force in altering conditions of the habitat and creating the va- riety of environmental conditions suitable to the various plant communities growing there. The plant communities themselves partici- pate in the weathering process as biomass be- comes established. Spatial patterns of plant communities in Navajo National Monument are due to differ- ent microhabitats created by weathering pro- cesses on Navajo Sandstone as well as avail- ability of water. Water availability influences community distribution in three major ways. First, water has been the primary factor in sculpting canyon morphology, which has, in turn, created a variety of microhabitats. Sec- ond, where water is more abundant, physical weathering of parent material is accelerated. Third, soil water differentials have contrib- uted to local variability in vegetation patterns. These effects are especially well illustrated at Betatakin Canyon (Fig. 2). On the canyon rims, runoff is rapid, little water is retained, soil development is poor, and desertic condi- tions prevail. Deep in the canyon, water tends to run into the soil rather than on the surface, accumulating in low areas, produc- ing more mesic conditions. Seep lines at the base of steep canyon walls also provide local Table 4. Relative position of plant community soils along 13 measured gradients. Index (column 14) indicates overall position with respect to parent material. The larger the index, the greater total change from parent material. Soluble Community type Ca Mg P N Zn Mn Fe Cu Na K pH salts O.M. Index Slickrock 1 2 1 1 1 1 1 1 1 1 1 14 Pinyon-juniper-grass 2 2 1 1 1 1 1 1 3 1 2 IS Talus slope 3 2 1 2 1 1 1 1 1 2 2 19 Mixed weed KS 1 3 1 3 1 1 1 2 2 3 2 Douglas fir 2 2 1 2 2 2 1 1 3 2 2 23 Atriplex-grass 2 4 2 1 4 1 2 1 2 1 2 1 25 Aspen 2 1 2 3 3 2 2 1 1 3 2 3 26 Hanging garden 5 2 2 1 2 1 1 5 1 2 5 3 27 Pinyon-juniper-mixed 28 shrub 1 5 1 1 5 2 2 2 1 1 2 1 2 Pinvon-juniper sage 1 5 1 1 5 2 4 3 1 1 3 1 1 29 Stream side tree 3 2 ■ 3 1 2 2 1 1 3 5 1 5 2 30 Oak 2 2 5 5 3 5 5 5 1 2 5 3 5 48 12 Great Basin Naturalist Vol. 45, No. 1 10 15 Slickrock- Scattered Shrub Pinyon-Juniper Grass — Talus Slope 20 - 25- 30 35- 40- 45 - 50 Mixed Weed Douglas Fir Atriplex-Grass Aspen Hanging Garden Pinyon-Juniper Mixed Shrub Pinyon-Juniper Sage Streamside Tree - Oak 3 Fig. 5. C^oniiminity positions on composite gradient of thirteen soil factors. areas of increased moisture, providing habi- tats for the unu.sual Douglas fir and hanging garden communities. Literature Cited Beehs. T. VV., F. E. Dress, and L. C, We.nsel. 1966. .\s- pect transformation in site productivits research. j. Forestry 64:691-692. Bii,i.i\(,s, \V. D. 1952. Tlie environmental complex in relation to plant growth and di.stribntion. Ouar. Hev. Biol. 27:2.51-265. Bn^KELAM), F. W. 1974. Pedology, weathering, and geo- morphological research. Oxford Univ. Press, Lon- don. 285 pp. BoiYcorcos, C;. J. 1951. .X recalihration of the ludro- meter metl'iod for making mechanical analvsis of soils. J. .\gron. 4.]:4:34-43S. 8.0- v^ • 7.8 \ • • 7.6 • ^v 7.4 • •\ \ 7.2- • •^ \ 7.0- ■ \ 6.8- \ V fi R X 6 1 2 3 4 5 6 7 8 9 1C Percent Organic Matter Fig. 6. Relationship Ijetween pH and percent organic matter in soils of plant conimnnities in Navajo National Mommient. Brotherso.n. J. D., G. Nebeker, M. Skoug.^rd, and J. Fairchild. 197S. Plants of Navajo National Mon- ument. Great Basin Nat. .38:19-30. Clements, F. E. 19.36. Nature and structure of the cli- max. J. Ecol. 24:252-284. Cochran, W. G., and G. VV. Snedecor. 1976. Statistical methods. Iowa State Univ. Press. Ames, Iowa. 593 pp. Curtis, J. T. 1955. A prairie continiuun in Wisconsin. Ecology 36:558-566. Daubenmire, R. 1959. A canopy-coverage method of vegetational analysis. Northwest Sci. 33:43-46. Foster, R. J. 1971. Physical geology. Charles E. Merrill Publishing Co., Columbus, Ohio. 526 pp. Gleason, H. a. 19.39. The individualistic concept of the plant association. Am. Mid. Nat. 21:92-110. Holmcren, R. C, a.\d S. F. Brewster. 1972. Distribu- tion of organic matter reserve in a desert shrub community. USDA Forest Service Research Pa- per INT-130. Intermountain Forest and Range Experiment Station, Ogden, Utah. Isaac:, R. A., and J. D. Kerber. 1971. .\tomic absorption and flame photometry: techniques and uses in soil, plant, and water analysis. Pages 17-38 in L. M. Walsh, ed., Instrinnental methods for analysis of soils and plant ti.ssue. Soil Sci. Soc. .-Xmer. Wis. Jackson, M. L. 1958. Soil chemical analysis. 'Prentice- Hall, Inc., Englewood Cliffs, New Jersey. 498 pp. Jack.son, T.A., and W. D. Keller. 1970. A comparative study of the role of lichens and "inorganic" pro- cesses in the chemical weathering of recent Ha- waiian lava flows. .\mer. J. Sci. 269:446-466. Jenney. H. 1958. Role of the plant factor in the pedo- genic huictions. Ecol. 39:5-16. Jones, J. B. 1973. Soil testing in the United States. Comm. Soil Sci. Plant Anal. 4:.307-.322. Kaesleh, R. L., and j. Cairns. 1972. Cluster analysis of data from limnological survevs of the upper Poto- mac River. Am. Midi. Nat. 88:56-67. Lkvlns, R. 1966. The .strategy of model building in ecol- ogy. Am. Sci. 54:421-431. Llndsay. W. L., and W. a. Xoiuall. 1969. Devel- oiMuent of a DTPA micronutrient soil test. January 1985 Brotherson et al.: Plant Communities 13 Agron. Abstracts, p. 84. Equilil)iiiim relationships of Zn++, Fe++, Ca+ + , and H+ with EDTA and DTPA in soil. Soil Sci. Soc Anier. Proc. a3:62-68. LuuwiG, J. A. 1969. Environmental interpretation of foothill grassland communities of northern Utah. Unpublished dissertation. Univ. of Utah. Salt Lake City, Utah. 100 pp. MacArthur, R. H. 1972. Geographical ecology: patterns in the distribution of species. Harper and Row, New York. 251 pp. Major, J. 1951. A Rinctional, factorial approach to plant ecology. Ecology .32;392-412. McDouGALL, W. B. 1973. Seed plants of northern Ari- zona. Mu.seum of Northern Arizona, Flagstaff. 594 pp. Olsen, S. R., C. V. Cole, F. S. Wata.nabe, a.nd L. A. Dean. 1954. Estimation of available phosphorus in soils by extraction with sodium bicarbonate. USDA Cir. 939. Russell, D.A. 1948. A laboratory manual for soil ferti- lity students, .3d ed. Wm. Brown Co., Dubuque, Iowa. 56 pp. RuziCKA, M. 1958. .Anwendung mathematisch-statisti- cher Methoden in der Geobotanik (Synthetische Bearbeitung von Aufnahmen). Biologia, Bratis- lava 13:647-661. Schalascha, C. B., H. Appelt, and A. Schatz. 1967. Chelation as a weathering mechanism. I. Effects of complexing agents on the solubilization of iron from minerals and granodionite. Geochim. et Cosmochim. Acta. 31:587-596. Sf:uATZ, A. 1963. Chelation in nutrition, soil micro- organisms and soil chelation. The pedogenic ac- tion of lichens and lichen acids. J. Agric. Food Chem. 11:112-118. Sneath, p. H. a., and R. R. Sokal. 1973. Numerical tax- omony-the principles and practice of numerical cla.ssification. W. H. Freeman and Co., San Fran- ci.sco, California 573 pp. Walter, H. 1973. Vegetation of the earth- in relation- ship to climate and the ecophysiological condi- tions. The English Universities Press Ltd., Lon- don and Springer-Verlag Inc., New York. 2.37 pp. Warner, J. H., and K. T. Harper. 1972. Understory characteristics related to site quality for aspen in Utah. Brigham Young Univ. Sci. Bull. Biol. Ser. 16(2): 1-20 Weaver, J. E., and F. E. Clements. 1938. Plant ecolo- gy. 2d ed. McGraw-Hill Book Co., Inc. New York. 601 pp. Whittaker, R. H. 1953. A consideration of climax theo- ry: the climax as a population and pattern. Ecol. Monographs 23:41-78. Wilding, L. P., and L. R. Drees. 1969. Biogenic opal in soils as an index of vegetative history in the Prairie Peninsula. Pages 96-103 in R. E. Berg- strom, ed.. The Quaternary of Illinois. Univ. of Il- linois, Coll. Agri. Spec. Publ. 14. 179 pp. Zinke, p. J. 1962. The pattern of individual forest trees on soil properties. Ecology 43:130-133. CRYPTOGAMIC SOIL CRUSTS: SEASONAL VARIATION IN ALGAL POPULATIONS IN THE TINTIC MOUNTAINS, JUAB COUNTY, UTAH Jetfiev R. Johansen' and Samuel R. Rushforth' .\bstr.\ct.- The soil algae in the Tintic Mountains, Juab County, Utah, was studied over a one-year period in 1982 and 1983. Fluorescence microscopy was used to measure algal density in samples directly from the field. A total of .30 algal taxa was observed, blue-green algae being most abundant both in terms of density and number of species. Algal densitv showed peaks in late fall and late spring. Minima were present in September 1982 and July 1983. Sev- eral weak correlations between algal density and climatic data existed. In general algae correlated positively with precipitation and negatively with temperature. A combination of low precipitation and hot temperatures was likely responsible for the low density observed in July. The Chrysophyta followed slightly different trends than the other algal groups, having minima in early October 1982 and late August 1983. Field observations indicated that the de- gree of algal crusting varied noticeably over a period of one year, with highest abundance of hummocking in spring. Heavy summer thunderstorms destroyed algal crusting during July and August of 1983, though absolute density of al- gae increased during this time in response to the extra moisture. Soil algae have received considerable study during the last half century (Melting 1981, Starks et al. 1981). Arid land soil algae are an important component of desert ecosystems for several reasons. Algae in association with lichens and mosses form cnists that stabilize the soil surface against wind and water ero- sion. Improved infiltration of rainwater in CRisted soils further reduces erosion by less- ening the amount of nmoff (Brotherson and RiLshforth 1983). Fixation of atmospheric ni- trogen by blue-green algae, both free living in the soil and associated with soil lichens, has been well documented. Vascular .seedling development tends to be facilitated in areas where crust development is pronounced (St. Clair et al. 1984). Flori.stic studies of soil algae have primari- ly been conducted through the use of enrich- ment and unialgal cultures (Melting 1981). Recent studies of soil algae in the Great Basin and Colorado Plateau have been performed using distilled water wetting and/or standard culture methods (Anderson and Rushforth 1976, Ashley and Rushforth 1984, Johan.sen, Rushforth, and Brotherson, 1981, Johansen, Javakul, and Rushforth 1982). Studies of soil algal biomass have been done using several different methods, including chlorophyll a content, dilution counts, enrichment counts, plate counts, and direct counts (Melting 1981). Direct counts of soil algae are difficult to make, since algae are often scarce in uncul- tured samples and identification of many spe- cies without prolonged culture and study of life cycles is often nearly impossible. How- ever, direct counts are more indicative of natural conditions because biomass and diver- sity estimates are based upon data from com- munities that have not been modified by cul- turing. Fluorescence microscopy can greatly aid in examination of soil algae. Although the value of this technique was pointed out more than 30 years ago (Tchan 1952), it has not been widely used. In the present study, direct counts using fluorescence microscopy were used to measure seasonal changes in abun- dance of arid land soil algae. We recognize that some .sacrifice in taxonomic resolution was made in exchange for ecological data based on a less modified algal community. Mi The study site was located in the Tintic Mountains, 7 km west of Tintic Junction, Juab County, Utah, TIOS R4W S35. Several well-developed algal crusts were 1' ted on a hill crest, 1950-m elevation, in liperus 'Department of Botany and Range Stiince, Brigham Young Univ 14 January 1985 JoHANSEN, Rushforth: Cryptogamic Soil Crusts osteosperma-Artemisia tridentata vascular plant commiuiity. The soil at the locality was examined and classified using standard refer- ences (Soil Conservation Service 1972, Soil Survey Staff 1951, 1975). The soil was classi- fied as Aridic Calcixeroll, fine clayey, mixed, mesic. The surface horizon was a dark greyish brown, sandy clay loam. Other soil characteristics are listed in Table 1 . Samples were collected nine times over a period of one year beginning September 1982 (Table 2). Five samples were collected each time, except 31 October 1982, when 10 samples were collected and processed. One sample collected 15 June 1983 was lost; thus only 4 samples were available for that collection. All samples were collected from an area approximately 10 m^ in size. Exact sample localities were subjectively chosen in crusted areas between shrubs. For each collection pe- riod, the upper 2 cm of five adjacent algal crust hiunmocks were collected and returned to the laboratory for immediate analysis. Samples were homogenized with a metal spa- tula. Six 1-gm subsamples were taken from each of the samples. One subsample was re- served for direct microscopic analysis. The other five were oven dried at (105 C) and weighed again to calculate percent soil mois- ture (Table 2). The subsample reserved for microscopic analysis was wetted immediately prior to ex- amination. The soil was placed in a vol- umetric cylinder, and distilled water was added to a total volume of 10 ml. The sample was then agitated vigorously by hand for 60 seconds. A 1-ml subsample was pipetted from the solution and placed in a dilution tube containing 4 ml of distilled water. This final dilution was used for direct examination un- der a Zeiss KA microscope with dark field and fluorescence optics. A counting chamber with a depth of 0.18 mm and field length of 10 mm was constructed as suggested by Tchan (1952) and used in all counts. Fluores- cing (living) algal cells and colonies were counted under 400X magnification. All algae in 10 transects across the chamber were counted for each sample. Algae were identi- fied to species when possible, or to genus or division when necessary. Some algae were not identifiable and were classified as un- known coccoid algae or unidentifiable algae. Moist plate cultures using Bold's Basal Me- dium (Bold and Wynne 1978) were started at the initiation of the study to aid in species identification. Temperature and moisture data were taken from weather records for Eureka, Utah, the nearest weather station (National Ocean- ic and Atmospheric Administration 1982, 1983). The average daily temperature and precipitation for the two weeks previous to each collection date were calculated (Table 2). An importance index for all living algal taxa was calculated by multiplying density by percent presence (Ross and Rushforth 1980). This method is often used in studies of terres- trial vascular vegetation (Warner and Harper 1972) and has the value of considering both absolute density data as well as frequency of occurrence. Species with importance values greater than 1.0 were designated prevalent species and were used in subsequent analyses of variance. Multivariate analysis of variance adapted for a fixed-effect, unbalanced design follow- ing the methods of Bryce et al. (1980) was used to analyze differences between treat- ments (months) and blocks (species). The variance of each taxon was related to the mean. To satisfy the homogeneity of variance Table 1. Soil properties for the Aridic Calcixeroll at the Tintic Mountain site. Soil for tests taken from A I ho- rizon, 0-28 cm depth, 3 October 1982. Property Sand 55.6% Silt 20.4% Clay 24.0% Gravel 0.9% Lime content 2.0% Organic matter 8.4% Pore space 35.8% Bulk density 1.7 g/cm-^ Soluble salts 240 ppm NH4 100 ppm P 155 ppm NO3 7 ppm Ca 3.4 meq/liter Mg 1.2 meq/liter Na 1.0 meq/liter HCO3 and CO3 3.9 meq/liter CI 0.5 meq/liter K 0.2 meq/liter pH, saturated paste 7.4 16 Great Basin Naturalist Vol. 45, No. 1 assumption of analysis of variance, a log (x+1) transformation was used (Bartlett 1947). In each analysis of variance, standard- ized residuals were plotted against normal scores. In every case the probability plot thus generated was subjectively judged to be nor- mal or close to normal. The Duncan multiple range test was used to determine significance of differences between means when analysis of variance showed significance (Duncan 1955). Unless otherwise stated, the alpha val- ue used for this test was 0.01. Shannon-Wiener diversity indices were calculated (Shannon and Weaver 1949, Pat- ten 1962) for each subsample for each collec- tion date. Similarity indices for all 49 sub- samples collected on the nine collection dates were determined following the methods of Ruzicka (1958). Such indices were also calcu- lated for the nine collection dates using arith- metic averages of subsamples. Similarity in- dices were then clustered (Sneath and Sokal 1973) to determine the relationships between collection periods. Correlation analysis (Snedecor and Cochran 1980) was performed to determine relationships between algal abundance and precipitation and temper- ature data. Results and Discussion Description of the Flora A total of 30 algal taxa was identified dur- ing this study (Table 3). Fifteen of these were blue-green algae, 9 were chrysophytes (in- cluding diatoms), 3 were green algae, and 3 were placed in the categories flagellates, un- known coccoid algae, and unidentifiable Table 2. Soil moisture for each collection date and mean hij^h and low temperature and precipitation for two-week period previous to collection date. Date Temperature Precipitation Soil moisture (degrees C) (cm) (g/kg) 24-IX-1982 8-20 2.4 .37.4 .3-X-1982 fS- 1 H 10.7 205.6 .31-X-1982 0-11 4.4 240.9 29-XI-1982 (-)(> 4 1.0 275.1 14-1-19^3 (-)(i- a 0.0 178.0 1.5-V-19a3 1-16 2.4 .35.9 1.>VI-19H3 6-24 1.6 1.2 29-VlI- 198.3 1,5-28 2.1 19.2 26-VII1-198.3 13-26 4.7 25.9 algae. The latter 3 categories were necessary since some of the algae encountered with the fluorescent microscope could not be placed with confidence into known divisions. Algal taxa in the Tintic locality with an importance index greater than 1.0 included Microcoleus vaginattis (importance value = 351.56), Nostoc species (258.92), uniden- tifiable algae (162.36), unknown coccoids (161.05), Phormidium minnesotense (141.97), unknown coccoid Chlorophyta (118.72), un- identified Chrysophyceae (64.27), Navicida mutica (58.16), Hantzschia amphioxys (56.64), Anahaena cf. variabilis (33.01), Sy- nechococcus aeruginosas (26.37), TolypotJirix tenuis (10.15), unknown Chlorosarcinales (5.81), unknown Chroococcaceae (4.25), and Pinnularia borealis (2.17). Multivariate analysis of variance of preva- lent species density data showed that signifi- cant (p<.001) differences existed between taxa. Duncan's multiple range test showed that Microcoleus vaginatus was significantly more abundant than all other species or cate- gories. Nostoc species were more abundant than all other less common taxa. Uniden- tifiable algae, unknown coccoid algae, Phor- midium minnesotense, and unknown coccoid Chlorophyta formed a group, of which each was more abundant than less common taxa. Chrysophyte cysts, Navicula mutica, Hantz- schia amphioxijs, and Anahaena cf variabilis formed a second group, of which each was more abundant than less prevalent taxa. The Tintic algal flora is similar in most re- spects to that we have observed from other localities in the Great Basin and Colorado Plateau (Anderson and Rushforth 1976, Ash- ley et al. in press, Johansen et al. 1981, 1982, 1984). It is dominated by filamentous blue- green algae, particularly Microcoleus vagi- natus, Nostoc species, and Phormidium spe- cies. In addition, diatoms are more diverse and often more abundant than green algae. One conspicuous difference between our soil crusts and several others we have examined was the absence of an obvious moss and li- chen component. Some lichens were present in our crusts, but they were rare. This is likely due to the fact that the study area is regularly grazed, and grazing is known to cause severe damage to the moss and lichen January 1985 Johansen, Rushforth: Cryptogamic Soil Crusts components of cmsts (Anderson et al. 1982, have studied, analysis of variance showed Brotherson et al. 1983). that diversity was significantly different be- tween months. Duncan's multiple range lest Diversity and Similarity showed that the diversity in collections from cooler months (especially January and June) Shannon-Wiener diversity indices varied was significantly higher than diversity in between 2.285 and 3.399 for the individual warmer months (especially August and subsamples. Average Shannon-Wiener diver- September). sity for each collecting period ranged be- Stand similarity on the basis of species den- tween 2.690 and 3.229. These figures are sim- sity data was high, averaging 68% for the ilar to or higher than other Shannon-Wiener winter stands and 50% for the remaining values for soil algal studies we have made in stands. When these data were clustered, onlv other regions of the Great Basin (Johansen winter stands (31 October, 29 November, 14 and St. Clair in review, Johansen et al. 1982, January, and 15 May) formed a discrete 1984). The rather low range of diversity for group. Stand similarity on the basis of species these samples was borne out by species rich- presence or absence was even higher, with an ness figiu-es. Total number of taxa observed average similarity of 74%. When these data per collecting period varied between 14 and were clustered, the same winter months 19. Even though variability in diversity was formed a group, except that the samples from rather low in comparison to other systems we early October and July were included. Table 3. Taxa present at the Tintic Mountain site together with importance values (PFl) and density for each col- lection date. Unit value for density is 1000 organisms/g dry weight soil. Taxa PFI 9/24 10/3 10/31 11/29 1/14 5/15 6/15 7/29 8/26 Cyanophyta Anabaena cf vuncihUis 33.01 20 15 23 1.55 7 96 15 52 CInoococcus pallidiis O.OS 13 Chroococnis species 0.02 3 Gloeocapsa aeruginosa 0.73 57 15 Clococapsa punctata 0.01 1 Ghu'otlurc linearis var. coniposita 0.56 I 10 3 12 2 Merismopcdia punctata 0.01 3 Microcoleus caginatus 351.56 240 1.53 241 275 249 502 514 281 6.39 Nostoc species 258.92 131 267 234 296 188 235 710 48 305 Oscillatoria gcniinata 0.18 10 Phonuidiuin )ninncsotcnsc 141.97 19 84 138 213 121 266 270 50 98 Synccliococcus aeruginosas 26.37 123 30 36 6 22 36 82 20 23 Tohjpothrix tenuis 10.15 33 10 16 38 18 12 11 Tolijpothrix species 0.33 2 14 Unknown Chroococcaceae 4.25 1 15 47 47 3 7 Chlorophyta Oocijstis species 0.02 4 Chlorosarcinales 5.81 3 2 56 36 37 Unknown coccoids 118.72 15 58 124 164 137 124 304 56 99 Chrysophyta Achnanthes species 0.02 3 Caloneis barillum 0.01 1 Hantzschia amphioxt/s 56.64 81 26 43 87 52 84 46 44 40 Navicula mutica 58.16 71 23 64 87 80 66 75 45 28 Xavicula species 0.02 2 1 Pinnularia borealis 2.17 7 6 8 4 6 8 4 3 Pinnularia species 0.01 1 Unknown pennate 0.01 2 Chrysophyceae &4.27 33 48 64 54 69 67 182 36 28 Other unidentifiable algae Flagellates 0.03 1 3 Unknown coccoids 161.05 15 100 32 256 175 231 445 67 201 Unidentifiable algae 162.36 72 153 226 274 173 137 167 35 1.50 18 Great Basin Naturalist Vol. 45, No. 1 Climatic Factors Our original hypothesis was that precipi- tation and total soil algal growth would be positively correlated during all seasons. How- ever, when we ran correlation analyses, we discovered that precipitation and algal growth were almost always negatively corre- lated, although values were not significant. When we discovered this, we divided total algae into several categories for separate analyses. These categories included Cyanoph- yta, Chlorophyta, Chrysophyta, and other al- gae. The negative correlation we noted for total algal growth held for all of these cate- gories whether the period of precipitation prior to collection date used in the analysis was for 3, 7, 14, 21, or 28 days. In view of these negative correlations, we computed the net increase or decrease in abimdance for each algal group for each col- lection interval. We then correlated these data with climatic data. The results of this correlation of incremental algal growth with precipitation were low, but positive for all algal groups (Table 4). As a whole, our data indicate that algal growth was not related in a linear manner to precipitation. It seems probable that when a minimum amoimt of moisture is present in the soil, algal growth will not be limited and moisture beyond this threshhold does not en- hance growth and may in fact reduce algal density. Stokes (1940) found that soil algae in New Jersey grew best at 40%-60% soil mois- ture and that growth was strongly curtailed in saturated soil. On the other hand, low moisture is often a limiting factor for algal growth in desert regions (Lynn and Cameron 1973). For instance. Brock (1975) found that the depletion of soil moisture to -7 bars be- gan to deter the growth of Microcoletis in desert crusts. Incremental growth of algal groups was also correlated with air temperature. All cor- relations were negative, and the largest coefficients were obtained when median tem- perature rather than maximum temperature data were used (Table 4). Generally, the best correlations occurred when temperature data for the three-day period prior to collection were used rather than data for longer peri- ods. This indicates rapid negative algal re- sponse to high air temperatures. Abundance of blue-green algae was more negatively correlated with temperature than all other groups. This was surprising since many species of blue-green algae are known thermophiles and have been demonstrated to tolerate high temperatures in the laboratory Table 4. Correlation coefficients for incremental algal growth versus climatic factors. Climatic factors include mean daily precipitation (PRECIP), mean maximum temperature (MAX-T), and mean midtemperature (MID-T). Means were coiuputed for the 3- , 7- , 14- , 21- , and 28-day periods prior to the collection dates. Significant correla- tions are asterisked (a<.()5°, a<.01°°). Algal group Climatic factor 3 7 Days 14 21 28 Cyanopiivta PRECIP \L\.\-r MIIJ-T .033 -.315° -.429°° -.077 -..322° -..382°° .026 -.342° -.386°° .063 -..301° -..321° .317° -.,306° -.324° (JILOHOPH^TA PHEC:iP MAX-T MII)-T .181 -.283 -.37,5° .0.39 -.249 -.297° .041 -.2,36 -.287 .066 -.175 -.208 .118 -.199 -.234 CiniYSOI'HVIA PKECIP MA.\-T MM)-T .312° -.260 -.285 -.1,37 -.199 -.231 -.189 -.240 -.275 -.206 -.193 -.217 -.016 -.218 -.248 Othkh ai.(.ai: PHE( IP MAX-T MHi-T .(X)() -.248 -..309° .113 -.257 -.278 .236 -.199 -.210 .251 -.124 -.116 ..331° -.123 -.118 Total alc.m; PHE( IP MAX-T MII)-T .129 -.254 -..3.34° .060 -.2,34 -.266 .169 -.190 -.226 .212 -.120 -.133 ..332° -.130 -.147 January 1985 2200 FIGURE 1. 2000 CYANOPHYTA ^ 1800 5 1600 ■3 « 1200 o O 800 1 CO 600 1— 400 1 200' JoHANSEN, Rushforth: Cryptogamic Soil Crusts FIGURE 2. OTHER UNKNOWN ALGAE 19 J 1600 ] FIGURE 3. CHLOROPHYTA O) 250 (0 E « S 200 O) o O ,50 (0 >< 300 FIGURE 4. CHRYSOPHYTA Figs. 1-4. Mean density and standard deviation of algal groups during collection year: 1, Cyanophyta. 2, Other al- gae of unknown division. '.3, Chlorophyta. 4, Chrysophyta. Note the lows 31 October and 26 August in Chrysophyta, although other groups had minima on 24 September and 29 July. 20 Great Basin Naturalist Vol. 45, No. 1 Fig. 5. Precipitation and daily teiiiperatnre range for the Tintic Mountain site. Vertical lines represent sample times. (Booth 1946, Castenholz 1969). Chrysophytes demonstrated the smallest correlation values, indicating that they were least affected by high air temperature. To visualize algal growth throughout the collection period, mean density and standard deviation for each algal group were plotted against time (Figs. 1-4). In general, algal growth of all groups peaked in late fall and again in June. All groups showed a marked decrease in cell density in July and recovery in August. Tlie Jvily decrease was apparently due to high temperatures combined with low precipitation (Fig. 5). Even though a storm occurred immediately prior to the July col- lecting date, apparently the flora did not have time to respond to the increased mois- ture. Lynn and Cameron (1973), when study- ing soil algae of the Curlew Valley, found that algal growth was minimal between mid- July and mid-September. The July and Sep- tember collections in our study were likewise low in algal density. The recovery in August is perhaps unusual and likely due to the ma- jor storms that occurred mid-Julv to late Au- gust (Fig. 5). Growth curves were remarkably similar for all algal groups (Figs. 1-4). The greatest deviation from the typical pattern was dem- onstrated by chrysophytes (Fig. 4), which did not show the August recovery. Furthermore, the yearly minimum in Chrysophyta followed a hundred-year storm in late September (Jo- hansen in press). Correlation analyses also demonstrated differences between chryso- phytes and the other algal groups (Table 4). It was the only group without any significant negative correlations with temperature, which may indicate that these algae are not as adversely affected by high temperatures. As a final note, we observed that the de- gree of hummocking or pinnacling of algal crusts at the study site varied throughout the year. Throughout the fall of 1982 crusts were abundant and well developed. After snow- melt in May 1983 we noted that crusts ap- peared even more abundant and well devel- oped. At this time, algal hummocks were evident even in jeep roads adjacent to the site. By July crusting in roadways had been destroyed, though cmsts elsewhere were still abundant. In late August the crusts at our study site were severely damaged and eroded by powerful rain storms, although algal num- bers in the soils remained relatively high. These observations may indicate that the du- rability and longevity of algal crusts is much less than we had previously thought. Acknowledgments We thank the members of the Department of Microbiology, Brigham Young University for their assistance during our use of the transmission fluorescence microscope. Blaine Metting, Larry St. Clair, Jack Brotherson, and Bill Evenson reviewed the manuscript prior to publication and provided helpful suggestions. Literature Cited Anderson, D. C, K. T. Harper, .^nd S. R. Rushforth. 1982. Recovery of cryptogainic soil crusts from grazing on I'tah winter ranges. J, Range Manage. .35:.355-.359. Anherso.n, D. C:.. .v.m) .S. H. Hismiohth. 197(i. The cryptogam flora of desert soil crusts in southern Utah, U.SA. Nova Hedwigia 28:691-729. Asiii.KV. J., .x.N-n .S. R. Rlshforth. 1984. Growth of .soil algae on topsoil and processed oil shale from the Uintah Basin, Utah, I'SA. Reclam. Reveg. Res. .3:49-6.3. January 1985 JoHANSEN, Rushforth: Cryptogamic Soil Crusts 21 Ashley, J., S. R. Rushforth, and J. R. Johansen. In press. Soil algae of cryptogamic crusts from the Uintah Basin, Utah, USA. Great Basin Nat. Bartlett, M. S. 1947. The use of transformations. Bio- metrics 3:.39-52. Bold, H. C, and M. J. Wynne. 1978. Introduction to the algae, structure and reproduction. Prentice- Hall, Inc., Englewood Cliffs, New Jersey. 706 pp. Booth, W. E. 1946. The thermal death point of certain soil-inhabiting algae. Proc. Montana Acad. Sci. 5/6:21-2,3. Brock, T. D. 1975. Effect of water potential on a Micro- coleus (Cyanophyceae) from a desert crust. J. Phycol. 11:316-320. Brotherson, J. D., and S. R. Rlshforth. 1983. In- fluence of cryptogamic crusts on moisture rela- tionships of soils in Navajo National Monument, Arizona. Great Basin Nat. 43:73-78. Brotherson, J. D., S. R. Rushforth, and J. R. Johansen. 1983. Effects of long-term grazing on cryptogam crust cover in Navajo National Mon- ument. J. Range Manage. 36:579-581. Bryce, G. R., D. T. Scott, and M. W. Carter. 1980. Estimation and hypothesis testing in linear mod- els—a reparameterization approach to the cell means model. Communications in Statistics— Theor. Meth. A9(2):131-150. Castenholz, R. W. 1969. Thermophilic blue-green al- gae and the thermal environment. Bacteriol. Rev. 33:476-504. Duncan, D. B. 1955. Multiple range and multiple F tests. Biometrics 11:1-42. Johansen, J. R. In press. Response of soil algae to a hun- dred-year storm in the Great Basin De.sert, USA. Phykos. Johansen, J. R., A. Javakul, and S. R. Rushforth. 1982. The effects of burning on the algal communities of a high desert soil near Wallsburg, Utah, USA. J. Range Manage. .35:598-6(X). Johansen, J. R., S. R. Rushforth, and J. D. Brotherson. 1981. Subaerial algae of Navajo Na- tional Monument, Arizona. Great Basin Nat. 41(4):43.3-439. JoH,\NSEN, J. R., AND L. L. St. Clair. Iu rcview. Crypto- gamic soil crusts: recovery from grazing near Camp Floyd State Park, Utah, USA. Johansen, J. R., L. L. St. Clair, B. L. Webb, and G. T. Nebeker. 1984. Recovery patterns of cryptogam- ic soil crusts in desert rangelands following fire disturbance. Bryologist 87:238-243. Lynn, R. I., and R. E. Cameron. 1973. Role of algae in crust formation and nitrogen cycling in desert soils. US/IBP Desert Biome Res. Memo. 7.3- 40(3):2.3.4.6.-l-26. Mettinc, B. 1981. The systematics and ecology of soil algae. Bot. Review 47(2): 195-3 12. National Oceanic and Atmospheric Administration. 1982. September-December, Utah 1982. Cliina- tological Data, Utah 84(9-12). 1983. January-August, Utah 1983. Climatologi- cal Data, Utah 85(1-8). Ross, L. E., AND S. R. Rushforth. 1980. The effects of a new reservoir on the attached diatom commu- nities in Huntington Creek, Utah, USA. Hvdro- biologia 68:157-16.5. RuzicKA, M. 1958. Anwendung mathematiseh-statisti- scher Methoden in der Geobotanik (synthetischc Bearbeitung von Aufnahmen). Biologia, Bratisl. 13:647-661. Shannon, G. E., and W. Weaver. 1949. The mathe- matical theory of communication. Univ. of Il- linois Press, Urbana. 117 pp. Sneath, p. H., and R. R. Sokal. 1973. Numerical tax- onomy. W. H. Freeman and Co., San Francisco. 573 pp. Snedecor, G. W., and W. G. Cochran. 1980. Statistical methods. 7th ed. Iowa State Univ. Pre.ss, .Ames, Iowa. 507 pp. Soil Conservation Service. 1972. Soil survey labora- tory methods and procedures for collecting soil samples. U.S. Dept. Agric. SSIR I. U.S. Govt. Printing Office, Washington, D.C. 63 pp. Soil Survey Staff. 1951. Soil survey manual. U.S. Dept. Agric. Handb. 18. U.S. Govt. Printing Office, Washington, D.C. 1975. Soil taxonomy, a basic system of soil classi- fication for making and interpreting soil surveys. U.S. Dept. Agric. Handb. 4.36. U.S. Govt. Print- ing Office, Washington, D.C. 754 pp. Starks, T. L., L. E. Shubert, and F. R. Trainor. 1981. Ecology of soil algae: a review. Phycologia 20(l):6.5-80. St. Clair, L. L., B. L. Webb, J. R. Johansen, and G. T. Nebeker. 1984. Cryptogamic soil crusts: en- hancement of seedling establishment in disturbed and undisturbed areas. Reclain. Reveg. Res. 3:129-1.36. Stokes, J. L. 1940. The influence of environmental fac- tors upon the development of algae and other mi- croorganisms in the soil. Soil Sci. 49:171-184. Tchan, Y. T. 1952. Study of soil algae. I. Fluorescence microscopy for the study of soil algae. Proc. Linn. Soc. London 77:265-269. Warner, J. H., and K. T. Harper. 1972. Understory characteristics related to site (juality for aspen in Utah. Brigham Young Univ. Sci. Bull., Biol. Ser, 16(2): 1-20. AQUATIC PARAMETERS AND LIFE HISTORY OBSERVATIONS OF THE GREAT BASIN SPADEFOOT TOAD IN UTAH Peter Hovinuh', Boh Benton^'^, and Dave Bomholdt^'^ .\bstr.\ct.- The distrihution and lireeding hahitats of the Great Basin spadefoot toad {Scaphiopu.s intermontanus) were investigated in the Bonneville Basin of western Utah. The permanent springs and man-made reservoirs used for breeding were largely found below the 16()0 m elevation. The pH's ranged between 7.2 and 10.4 and the total dis- .solved solids between 170 and 4800 mg/i. The springs were less alkaline than the rain-filled reservoirs. The lack of aquatic vegetation was a common feature of the reservoirs and most of the springs. Observations of breeding without rain are noted as well as the lack of breeding with rain. The snout-vent lengths of adult spadefoots are greater in the Bonneville Basin than in other parts of the Great Basin. Utilization of permanent water sources and stimuli for emer- gence and breeding, as well as the larger adult size of S. intermontanus in the Bonneville Basin, are discussed in rela- tion to the diverse precipitation patterns, the sparseness of the water sources, and the Holocene history of the Great Basin. Spadefoot toads have been extensively studied in California, Arizona, New Mexico, Texas, and Oklahoma. Very little information is available for the Great Basin spadefoot toad {Scaphiopus intermontanus). Tanner (1931) found S. intermontanus widely dis- tributed in Utah. Most observations in Utah have been within the Colorado River drain- age (Tanner 1931, Hardy 1938, Wood 1935, Wright and Wright 1949). Observations of S. intermontanus in the Great Basin ecosystem are less frequent. Tanner (1931) found spade- foots common along the Wasatch Front and reported observations near Candy and Callao in Utah, near the Nevada border. Synder (1920), Linsdale (1938), and La Rivers (1942) reported the records for Nevada. Scaphiopus intermontanus has several unique features that are not found in other spadefoot species: (1) breeding is reported to occur without rainfall for stimulus (Lin.sdale 1938), (2) a large number of chorusing adults is not es.sen- tial for breeding (Wood 1935, Blair 1956), and (3) permanent water can be utilized for breeding (Bragg 1961). This study describes the breeding habitat of S. intermontanus in the Bonneville Basin of the Great Basin in western Utah, which is bounded by the ancient shoreline of Lake Bonneville, a lake that desiccated some 11,000 years ago (Currey 1980). Lake Bonne- ville filled the valleys in western Utah to a height of about 1552 m above sea level, with up to 350 m of water, and covered 5,000,000 hectares. The lake existed at the high level for about 3000 years and filled the valleys at lower levels for 10,000 years. We found that most breeding sites of S. intermontanus oc- curred in areas that were inundated by Lake Bonneville and that many of these sites were associated with permanent springs. This study extends some of the earlier observations that are unique to S. intermontanus, de- scribes the breeding habitat, and interprets these observations in terms of the Holocene hi.storv of the Bonneville Basin. Methods and Materials Chemical analysis was performed by the Uintex Corporation under contract to the Bu- reau of Land Management and reported in "Water Inventory for Tooele Range Environ- mental Statement" for the Salt Lake District Bureau of Land Management (March 1981). 'Department of Biochemistry. University of Utali. Salt Luke Citv. Utah 84 U2. Reprint requests should be sent to 721 Set end Av UtahH4ia3 •Sail Lake Dislncl Office, Bureau of Land Management, 2370 South 230() West, Salt Lake Citv. Utah 84119. 'Present address; U.S. Fish and Wildlife Service. 2078 Administration Building. 1745 West 1700 South, Salt Lake Cilv. Utah 84104. 'Present address: MmeraK \l.n.a,;.-,n.ni s,.rvice. Atlantic COS Region. 1951 Kidsvell Drive. Vienna. Virginia 22180. 22 January 1985 H( ET AL.: Utah Spadefoot Toad 23 Under this contract conductivity (corrected to 25 C), pH, dissolved oxygen, alkalinity, ar- senic, nitrate, phosphate, suspended dissolved solids, total dissolved solids (residues at 185 C), and coliform bacteria analyses were per- formed. Duplicate field pH measurements in- cluded calibration at each site, with standard buffer solutions of pH 4, 7, and 10 at each site. Conductivity was measured by the YSI- 33 portable conductivity meter in both the Uintex Corporation contract and in the springs and water sources outside Tooele County. Climatic information was taken from the United States Climatological Records (U.S. Weather Bureau 1979-1981). Snout-vent (SVL) measurements were made on tadpoles and adults. Tadpoles were caught with a single net sweep. Measure- ments of SVL of museum specimens were taken from the collections of Utah Museum of Natural History, University of Utah; Mu- seum of Vertebrate Zoology, University of California at Berkeley; and the Monte L. Bean Life Science Museum, Brigham Young University. Data for the study reported here were largely collected during the years of 1979, 1980, and 1981. The Bureau of Land Man- agement-contracted water inventory in Tooele County occurred in 1980 and con- sisted of three trips to the water sources be- tween June and September if water was found on the preceding visits. Photographs of the water sources were taken at each visit. Description of the Bonneville Basin The Bonneville Basin, contained within the Great Basin of the Intermountain Region, is typical of the basin and range topography common to western Utah and Nevada. The higher mountains vary from 3050 and 3980 m above sea level and, except for the Wasatch Mountains on the eastern margin, generate few perennial streams that reach the basin floor. The basin floors range from 1284 m (historic high level of the Great Salt Lake) to about 1550 m above sea level. Most of the in- ternal mountain ranges do not have per- manent streams that reach the valley floors. Annual average rainfall varies from 12 to 31 cm in the valley floors. In some years the valleys may receive less than 5 cm of rain (Hood and Waddell 1968, 1969, Stephens and Sumsion 1978, Hood, Price, and Waddell 1969, Bolke and Sumsion 1978, Stephens 1977). Evaporation from large bodies of open water is estimated to vary between 107 and 127 cm per year (Hood and Waddell 1968, Hood, Price, and Waddell 1969). During the three study years of 1979, 1980, and 1981, over 50% of the rain from April to October occurred in May, with very little precipi- tation in June and July (Fig. 1). The Bonne- ville Basin is characterized as receiving spring rains (Kay 1982). Air temperatures vary between -24 C and -35 C in January and February to the high values of 39 C and 42 C in summer. Freezing temperatures can be expected from October through April. A large portion of the Bonne- ville Basin has 120 frost-free days annually. Soil temperatures at Salt Lake City (1290 m above sea level) vary from 0 C to 35 C at a depth of 10 cm and 2 C to 25 C at a depth of 100 cm (U.S. Weather Bureau 1979-1981). The Bonneville Basin valleys contain four major vegetative types: (1) salt desert, (2) shadscale {Atriplex con ferti folia), (3) sage- brush {Artemisia tridentata)-gra.ss, and (4) pinyon-juniper. Greasewood {Sarcobatiis ver- miculatus) occupies much of the valley floor and areas adjacent to springs and is some- times considered a wetlands indicator spe- cies. SagebRish-grass occurs in the eastern Bonneville Basin and in Nevada, and is not widely distributed in the Bonneville Basin. Pinyon-juniper lower elevational limits occur between 1600 and 1800 m above .sea level (West et al. 1978). Atriplex is the most preva- lent vegetative type that occupies the desic- cated valleys of the Bonneville Basin and is undergoing evolutionary change by chromo- some polyploidy and interspecific hybridiza- tion (Stutz et al. 1979). The general terrestri- al ecology of the Bonneville desert was described by Fautin (1946). Within the Bon- neville Basin eight mountain ranges occur that contain boreal a.ssociations. Results Bonneville Basin aquatic resources.- In the south central portion of the Bonneville Basin, 169 aquatic sites were investigated (Fig. 2). Most of the sites were observed in 24 Great Basin Naturalist Vol. 45, No. 1 2,0p LO- 0 ^ 0) e i^2.0 \ j_j >.o 1981 1.0 0 ;..j 1 jii L - 1 i y APR. ' MAY ' JUNE ' JULY ' AUG. 1 SEPT. 1 Fig. 1. Precipitation records for Dugway, Utah (40° 11' North and 112° 56' West) during the active season of the spadefoot toad (S. internionUiiui.s). Data modified from tlie U.S. W'eatlier Bureau Climatological Records for 1979, 1980, and 1981. 1980, but some .sites in Whirlwind and Skull valleys were observed for three years. Of these 169 .sites, 51 were used by S. inter- montanus at least one of the three years. Browns Spring was utilized only one year in five years of observations. The sites varied in size from small seeps (0.08 m- by 10 cm deep) to large reservoirs with over 1200 m' of water. The aquatic resources utilized by spade- foots consisted of man-made reservoirs (57%) and springs (43%). The man-made reservoirs formed water .sources where none previously existed and can be characterized as having widely fluctuating volumes of water. It is conunon for a reservoir to be full one year iind empty the next year due to the nature of the precipitation patterns. Even when filled during the spring and early summer, water is usually evaporated by autumn. The runoff- filled reservoirs pick up chemicals from the soil and wind-blown particles (including sa- line dust) over the watershed, or from the bentonite lining of the reservoirs (Stephens and Sumsion 1978). The springs utilized by the spadefoots were temporary or permanent. The portion of the springs utilized by spadefoots were the a.ssociated reservoirs (59%), water troughs (27%), streambeds (5%), and wetlands (42%). The wetlands utilized consisted of the distal end of a natural spring (14%), small seeps (14%), or a spring that was dug out (14%). Of- ten .spadefoots bred in the reservoir, water trough, and spring at a given site— thus ac- counting for more than 100% values. If a January 1985 HOVINGH ET AL.: UtaH SpADEFOOT ToAD 25 series of reservoirs were associated with a spring, the spadefoots bred in the most distal reservoir. Although many of the utilized springs had permanent water, usually the wa- ter levels declined by summer to the point a littoral zone could not form. Successful breeding sites were characterized by the ab- sence of aquatic plants in some portion of the aquatic resource. The elevational distribution of the water sources is shown in Figure 3. Of the total wa- ter sources that were utilized by S. inter- montamis, 74% of the sites were below 1550 m elevation or the height of former Lake Bonneville. Only 27% of the nonutilized sites were below the 1550 m elevation. Less than 14% of the utilized water sources were with- in the pinyon-juniper region (above 1600 m). The highest elevation spring (2012 m) was utilized by chorusing spadefoots that never bred. The pH of the water sources utilized by the spadefoots varied from 7.2 to 10.4, with most of the sources having a pH of 8 to 10. Although 84% of the springs contained water with pH of less than 8, the wetlands of these springs often contained water with pH great- er than 8. Most of these water sources con- tained less than 1000 mg/1 of total dissolved solids, although Browns Spring contained up to 4800 mg/1. Under evaporating conditions, phosphate, nitrate, and pH behaved in an unpredictable manner, sometimes increasing and sometimes decreasing in concentration. Alkalinity and total dissolved solids tend to concentrate in a linear manner. No chemical concentrated in direct proportion to the volume— that is, if the volume decreased by a hundredfold, the chemical concentration in many instances did not even increase by tenfold. Often dead tad- poles could be observed in dried-up reservoirs. Adult Emergence and Breeding.— Al- though rainfall or the low-frequency sounds of rain falling on the ground are considered stimuli for emergence and breeding for S. couchi and possibly for S. multiplicatus (Dim- mitt and Ruibal 1980b), the stimuli for emer- gence and breeding of S. intermontanus in the Great Basin is unknown. Breeding occurs in April, May, and early June in the Bonne- ville Basin, where the spring rains can be Fig. 2. Breeding sites of the spadefoot toad (S. inter- montanus) in the Bonneville Basin of the Great Basin. The light shaded area is the extent of former Lake Bon- neville according to Synder, et al. (1964). Vertical series of nnmbers refer to the township number, and the hori- zontal series of nnmbers refer to the range. The diame- ter of each circle represents approximately 5 km. Open circles: water sources not utilized by spadefoots. Closed circles: water sources utilized by breeding spadefoots. characterized as generalized and gentle (compared to the localized, torrential rains of late summer). In 1981 five rainfalls were ob- served from the middle of April (which stim- ulated the breeding) to the end of May (Fig. 1). At White Rocks seep in Skull Valley, the breeding occurred in the middle of April and 26 Great Basin Naturalist Vol. 45, No. 1 1200-1600 1600-2000 2000-2600 Elevation Above Sea Level (m) Fig. 3. Distribution of water sources witli respect to elevation above sea level. Solid pattern: ephemeral stream- filled reservoirs that were utilized by .S. intermontanus. Stippled pattern: ephemeral stream filled re.servoirs not utilized by spadefoots. Open pattern: springs utilized by breeding spadefoots. Vertical line pattern: springs not utilized by spadefoots. again at the end of May after a flash flood .scoured the seep. In Tule Valley spadefoots bred in South Tule Spring in April and at the end of May in nearby Painter Spring. It would seem that, if the mid-April rain .stimu- lated the spadefoots to breed, then they would be .stimulated to breed four additional times before the end of May. Conversely, at the end of May one .spring contained at least two chorusing spadefoots. No rain had fallen for the previous .several days, and the chorusing activity occurred for at lea.st three coasecutive evenings without any additional rainfall. Although the general initial breeding seems to be stimulated bv rainfall at most sites, exceptions are noted. Tadpole observations.— Tadpole growth rates at six different springs are .shown in Fig- ure 4. The initial growth rates varied from 0.53 to 0.66 mm SVL per day at the six springs. If there was alnmdant water in June, the tadpole growth rate slowed down (see White Rock Spring in Fig. 4). Most metamor- phosis occurred by early July. The sizes of the toadlets at metamorphosis varied from 16 to 38 mm SVL. Tadpoles were found in four springs during August and September. Slower initial growth rates at Henry Spring water trough (Fig. 4) may be responsible for the occurrence of tad- poles in August. Chemical analysis of the springs in Figure 4 varied, with total dis- solved solids from 200 mg/1 at White Rocks to 2070 mg/1 at Eight Mile Spring. Henry Spring was similar to Eight Mile Spring. The population that was observed at Browns Spring in September was probably frozen be- fore metamorphosis could take place. At the time of metamorphosis in August at Henry Spring, numerous dead toadlets were ob- served near the water. Adult size.— At Painter Spring water trough in Tule Valley adult spadefoots were actually observed breeding in 1981. The January 1985 HOVINGH ET AL.: UtaH SpADEFOOT ToaD 27 APRIL MAY JUNE JULY AUGUST F,^. 4. Tadpole (S. intennon,an.s) growth rates .n d.fferent springs. The tadpole length (snout^^^^^^^^^ length w measwed. A, White Rocks seep; x, Henry Spring; V, ^hite Rocks Sprmg OE^htN^^l^^ ^ Spr,ng; and •, Sonth Tule Spring. Metaniorphosn^g tadpoles (with four legs and tad) °^ '"^^^^^^^^^ ,, the b? Tan. The tadpoles were caught with a single sweep of the net, measured, and the averaj,e sue figure. 28 Great Basin Naturalist Vol. 45, No. 1 average size (SVL) of the nine individvials was 67 mm ± 6.7 (range, 57 to 77 mm). Be- cause of the large size of these spadefoots compared to values in the literature, the col- lections representing 10 locations in the Bon- neville Basin as well as all the other S. inter- montamis in the Great Basin were mea.sured. Statistical data are shown in Table 1. The specimens from the Bonneville Basin were larger (58 ± 5.3 mm) compared to those specimens from the Colorado River Basin of Utah (52 ± 4.8 mm), Idaho (49 ± 3.9 mm), Nevada (49 ± 4.1 mm), or Washington, Wyoming, Oregon, and Arizona. Both the Bonneville Basin (z = -1-6.7) and the Hum- boldt River Basin in Nevada (z = -4.9) have populations of spadefoots whose mean size exceeds the 99% confidence intervals of the total population, indicating that the size dif- ference is real. Discussion Scaphiopiis intennontamis utilizes every type of water source for breeding in the Bon- neville Basin as long as the total dissolved sol- ids are less than 5000 mg/1. Breeding can oc- cur in permanent reservoirs that contain an abundance of vegetative growth, but under these conditions tadpoles may not survive to metamorphosis. The highly successful breed- ings took place in water sources that either desiccated during the summer or had a large draw-down of water (in which case a littoral zone of vegetative growth was lacking), or breeding occurred in stream beds scoured by flash floods. Only 8% of the water sources were entirely natural. Humans, through range improvements, have created new habi- tat in the Bonneville Basin for the spadefoots and this new habitat greatly reflects typical breeding habitat of the genus Scaphiopus. Some of these new water sources are over 15 km from any existing water sources and, to utilize these sources for breeding, the spade- foots must disperse overland. Although hu- man-made reservoirs are widely utilized, dur- ing the 1979 limited rainfall (Fig. 1), permanent springs were utilized in Whirl- wind Valley because the reservoirs were without water. Subsequent years (1980, 1981) the reservoirs were utilized and the springs were vacated. The lack of large water sources and the spareness of the water sources in general would not be conducive to large numbers of breeding adults. The stimulus by which spadefoots emerge and breed has been the low-frequencv sounds of rain falling on the ground and not mois- ture per se (Dimmitt and Ruibal 1980b). Linsdale (1938) was the first to note that spadefoots may breed without the stimulus of rainfall and suggested such several other stimuli as flash flooding and other chorusing spadefoots. Observations in the Bonneville Basin confirm that not all breeding occurs Table 1. Variations in snout-vent length (SVL) of adult spadefoot toads (Svaphiopiis intcrniontanus) in the (Jreat Basin. Z is the measure of confidence interval and equals the deviation from the mean divided by the standari! deviation of the subpopulation mean. If Z is between -2.58 and +2.58, then there is a 99"'o confidence that the normal distribution of the total population contains the subpopulation mean. Area Number Mean size (SVL-mm) Standard Deviation) Total Utah Bonneville Basin Colorado River Drainage Anr/ONA (C^oconiiio C'ountv) Idaho Nevada Humboldt Drainage Western Soutlicn. Oni;(.()N W'ashinctcjn Wyoming (+ Daggett C^ounty, Utah) 254 51 5.1 114 53 5.4 4.1 28 58 5.3 6.6 77 52 4.8 2.1 8 54 3.2 1.1 26 49 3.9 -2.1 87 49 4.1 -4.5 55 48 6.7 -4.9 11 52 5.6 0.7 21 49 4.7 -1.9 12 50 3.3 -0.7 5 48 3.7 -1.3 5 52 4.7 0.4 January 1985 HOVINGH ET AL.: UtAH SpADEFOOT ToAD 29 with rainfall, although rainfall may be the main stimulus. Conversely, if rainfall is the main stimulus, not all comparable rainfalls stimulate breeding. Both of these observa- tions may be explained by the postbreeding dispersion of the spadefoots and the scarcity of water sources for breeding. If, for instance, the spadefoot is 5 km from a water source when it hibernates, the first rainfall may cause emergence. The travel time for the spadefoot to reach a water source may be from several days to a month, if a water source is to be reached during the breeding season. Consequently, a delayed response be- tween rainfall and breeding may occur. Until the historic development of water sources for range management, the water sources in many valleys (west of the Cedar Mountains, Whirlwind Valley, and Puddle Valley) were nonexistent. Furthermore, the rainfall patterns in the Great Basin are di- verse, with precipitation peaks in winter (California), spring (central Nevada, Idaho, and Bonneville Basin), and summer (Colorado Plateau) (Kay 1982). These conditions have probably existed in the Great Basin for 4500 years, since the establishment of the present ecosystem (Wells 1983). If spadefoots were to adapt to the cold desert ecosystem in the Great Basin, they would have to have adapt- ed to sparse aquatic breeding habitat, to utili- zation of flowing spring snowmelt streams, and to spring breeding (tadpoles would not metamorphose in time for July and August breeding). Upon desiccation of Lake Bonneville, new terrestrial habitat was formed. Atriplex evolved with many new polyploid forms and with hybridization (Stutz et al. 1979) and was fully adaptive to this new basin environment. Pinyon-juniper {Piniis monophyUa and Jiinip- erits osteospenna invaded the region from tlie southern refugium (Wells 1983), and the monotypic subalpine bristlecone pine {Finns longaeva) forest disappeared from the 1660- m elevation level about 10,000 years ago. During the pluvial era, the subalpine forest and the sagebrush approached the lake level (Wells 1983, Currey and James 1982). During the pluvial era, S. intermontanus was isolated on some island ranges or in sand dime regions lake the Escalante Desert in southern Bonneville Basin. Sand dunes were habitat for the relict diploid Atriplex canes- cens (Stutz et al. 1975) and coiild have fur- nished the necessary aquatic habitat for spadefoots adjacent to Lake Bonneville. A second po.ssibility is that the spadefoots were in the Mohave refugium along with the pi- nyon-juniper and migrated north in post- pluvial times. The significance of the larger snout-vent length of the Bonneville Basin spadefoot pop- ulation could be (1) genetic isolation of small populations during the 3000-year Lake Bon- neville period; (2) adaptations for longer pe- riods of hibernation (Jones 1980, Dimmitt and Ruibal 1980a), which would be required during the more xeric conditions that existed in the Bonneville Basin (Currey and James 1982); (3) more food energy going to increase size instead of reproductive potential because spadefoots may not breed every year, appar- ently because of water scarcity; or (4) re- duced adult predation in the Bonneville Ba- sin, which creates older populations. Scaphiopus intennontanits is not expected to acquire new responses to environmental conditions in the Great Basin in general and the Bonneville Basin in particular in view of the numerous restraints placed on the habitat with respect to precipitation, water sources, and biogeographical changes during the Holocene era. Although S. intennontaniis gives the appearance of becoming more am- phibianlike in its behavior (not responding to precipitation and breeding in permanent wa- ter), it certainly has adapted to humanly de- veloped typical Scaphiopus habitat in the Bonneville Basin. No other amphibian was found in the 151 water sources inventoried with the exception of the Western Spotted Frog {Rana pretiosa), which utilized the per- manent portion of the Tule Valley springs. Acknowledgments The authors wish to thank Dr. R. Ruibal (Department of Biology, University of Cali- fornia at Riverside), Dr. D. Currey (Depart- ment of Geography, University of Utah), and Dr. J. S. Gates (Water Resources Division, U.S. Geological Survey, Salt Lake City) for critically reviewing the initial manuscripts; and Dr.' P. Kay (Department of Geography, University of Utah) for assistance with the 30 Great Basin Naturalist Vol. 45, No. 1 statistical analysis of the adult spadefoot specimens. Further thanks are given to Dr. Harry W. Greene (Museum of Vertebrate Zo- ology, University of California at Berkeley), Dr. D. Cox (Monte L. Bean Life Science Mu- seum, Brigham Young University, Provo, Utah), and Dr. J. Legler (Utah Museum of Natural History, University of Utah) for mak- ing available the specimens of spadefoot toads. Literature Cited Blair. W. F. 1956. Mating call and possible stage of speciation of the Great Basin spadefoot. Texas Journal of Science 8:236-2.38. BoLKE, E. L., AND C. T. SuMSio.N. 1978. Hydrologic re- connaissance of the Fish Springs Flat area, Tooele, Juab, and Millard counties, Utah. State of Utah Dept. Nat. Res., Division of Water Rights Tech. Publ. no. 64. Brac;g,' a. N. 1961. A theory of the origin of spade-foot- ed toads deduced principally by a study of their habits, .\ninial Behavior 9:178-186. (arrey, D. R. 1980. Coastal Geomorphology of Great Salt Lake and Vicinity. Utah Geological and Min- eral Survey, Bulletin 116:69-82. CuRREY, D. R., AND S. R. James. 1982. Paleoenviron- ments of the northeastern Great Basin and north- eastern basin rim region: a review of geological and biological evidence. Pages 27-52 !7i D. B. Madsen and J. F. O'Connelly, eds., Man and en- vironment in the Great Basin. Society for Ameri- can Archeology, Paper 2. DiMMiTT, M. A., AND R. RuiBAL. 1980a. Exploitation of food resources bv spadefoot toads (Scaphiopus). Copeia 1980:854-862. 1980b. Environmental correlates of emergence in spadefoot toads (Scdpliiopiis). J. Herpetoiogv 14:21-29. Fautin, R. VV. 1946. Biotic connnunities of the northern desert shrub biome in western I'tah. Ecological Monographs 16:251-310. Hardy, R. 19.38. .\n annotated list of reptiles and am- phibians of Carbon County, L'tah. Utah Acad. Sci., Arts, Letters 15:99-102. Hood, J. W., D. Pric:e, and K. M. Waddell. 1969. Hy- drologic reconnaissance of Rush Valley, Tooele County, Utah. State of Utah Dept. Nat. Res., Di- vision of Water Rights Tech. Publ. 23. Hood, J. W., and K. M. Waddell. 1968. Hydrologic re- connaissance of Skull Valley, Tooele County, Utah. State of Utah Dept. Nat. Res., Division of Water Rights Tech. Publ. 18. 1969. Hydrologic reconnaissance of Deep Creek Valley, T(X)ele and Juab counties, Utah, and Elko and White Pine counties, Nevada. State of Utah Dept. Nat. Res., Division of Water Rights Tech. Publ. 24. Jones, R. M. 1980. Metabolic consequences of accelerat- ed urea synthesis during seasonal dormancy of spadefoot toads, Scaphiopus coticlii and Scci- phioptis mtiltiplicatuf;. J. Experimental Zool. 212:255-267. Kay, p. a. 1982. A perspective on Great Basin paleocli- mates. Pages 76-81 in D. B. Madsen and J. F. O'Connelly, eds., Man and environment in the Great Basin. Society for .\merican .\rcheology. Paper 2. La Rivers, I. 1942. Some new amphibian and reptile records for Nevada. J. Entom. and Zool. .34:5.3-68. Linsdale, J. 1938. Amphibians and reptiles in Nevada. Amer. Acad. .-Vrts and Sci. (Daedalius) 73:197-257. 19.38. Environmental response of vertebrates in the Great Basin. Amer. Midland Nat. 19:1-206. Stephens, J. C. 1977. Hydrologic reconnaissance of the Tule Valley Drainage Basin, Juab and Millard coimties, Utah. State of Utah Dept. Nat. Res.. Di- vision of Water Rights Tech. Publ. 56. Stephens, J. C, and C. T. Sumsion. 1978. Hydrologic reconnaissance of the Dugway Val- ley-Government Creek Area, West-Central L^tah. State of Utah Dept. Nat. Res., Division of Water Rights Tech. Publ. 69. Stutz, H. C, J. M. Melby, and G. K. Livingston. 1975. Evolutionary studies of Atriplex: a relic gigas diploid poulation of .\triplex canescens. .\mer. J. Botany 62:2.36-245. Stutz, H. C, C. L. Pope, and S. C. Sanderson. 1979. Evolutionary studies of Atriplex: adaptive prod- ucts from the natural hybrid, 6N .\. tridentata and 4N A. canescens. .\mer. J. Botany 66:1181-1193. Synder, C. T., G. Hardman, and F. F. Zdenek. 1964. Pleistocene lakes in the Great Basin. U.S. Geolog- ical Survey. Synder, J. O. 1920. Scapliioptis in northern Nevada. Copeia 1920:83-84. Tanner, V. M. 1931. A .synoptical study of Utah am- phibian. Utah Acad. Sci., .\rts, and Letters 8:159-198. U.S. Weather Bl-reau. 1979-1981. Climatologic data. Utah. Vol. 81-83. National Climatic Center. .'Vsheville, North Carolina. Wells, P. V. 1983. Paleobiogeograph\ of montane is- lands in the Great Basin since the last glacio- pluvial. Ecological Monographs 53:341-382. WE.ST, N. E.. R. J. Tai'sch, K. H. Rea, and P. T. TuELLER. 1978. Phytogeographical variation within Juniper-Pinyon woodlands of the Great Basin. Great Basin Nat. Mem. 2:119-1.36. Wood. W. F. 1935. Encounters with the western spade- foot, Scapliiopus haminoiulii, with a note on a few albino larvae. Copeia 1935:100-102. Wricht, a. H., and a. W. Wright. 1949. Handbook of frogs and toads. Comstock Publi.shing Co., Ithaca. New York. NEW SPECIES OF ASTRAGALUS (LEGUMINOSAE) FROM MESA COUNTY, COLORADO Stanley L. Welsh' .Abstract. Named and described is Astragalus dehafuaeus Welsh from Mesa Countv, C^olorado. Botanical investigations in central western Colorado during May 1984 yielded several unique taxa, especially endemics from that portion of the Colorado Plateau. The endem- ic plant taxa are associated with the peculiar habitats available on the raw geological sub- strates in the region. The Mancos Shale For- mation and other fine-textured strata support a phalanx of specially adapted taxa. Thus, it is to be expected that other formations with peculiar physical and chemical properties should support additional rare plants that have been overlooked. Growing on a varicolored, fine-textured, seleniferous and apparently saline portion (Atwell Gulch Member) of the Wasatch For- mation in the De Beque vicinity is an Astrag- alus that is beyond the descriptions of known species in that region (Fig. 1). The plants have white flowers, grow in small to large clumps, and have thinly cartillaginous, inflat- ed pods. Clearly these plants are allied to those taxa in Astragalus section Preussiani. The plants key to the couplet dealing with A. eastwoodae and A. preussii in that section (Bameby 1964). The pods are similar in tex- ture to those of A. eastwoodae, but are erect- ascending initially as in preussii, although they are ultimately spreading or even de- scending in pressed material. The white flow- ers are shared by neither. The pods are pro- portionately narrower than in those of the allied taxa. The surface of the pods is minute- ly scabrid-pubescent, becoming almost or quite glabrous in age. This feature occurs sometimes in the allied taxa. Flower number is mostly 7-9 (11) per raceme in the material from De Beque, not 3-7 as in A. eastwoodae. In A. preussii the flower number varies from few to many. The calyx is conspicuously shorter in the De Beque material than in A. eastwoodii (6.3-8 not 10-12.2 mm long). The De Beque milkvetch lacks the strong scent of selenium characteristic for many spe- cies of the section. However, the plant might still be a selenophyte. It grows with the strongly odoriferous selenium indicator, A. flavus Nutt., a common inhabitant of the Wasatch Formation in the vicinity. Astragalus debequaeus Welsh sp. nov. Af- finis Astragalo sectio Preussiano praesertim A. eastwoodae in leguminibus et habitu gen- erali sed in floribus plus numerosis et aibis calycibus brevioribus legumine ambito et dis- positio et caulibus plus numerosis differt. Plants perennial from a branching caudex, clump-forming, mainly 2-10 dm across, aris- ing from a woody taproot; stems 14-30 cm long, decumbent and curved-ascending; stip- ules 3-6 mm long, ovate to triangular, free; leaves 2-10 cm long, the leaflets 13-21, ellip- tic to oblanceolate, obtuse to rounded, glabrous, flat or somewhat folded, the termi- nal one not confluent with the rachis; pe- duncles 4.5-8.8 cm long, ascending; racemes 3-5.5 cm long, little elongating in fruit; flow- ers spreading to ascending in anthesis, 17-21 mm long; bracts 2-2.5 mm long, ovate- acuminate; pedicels 1-3 mm long; bracteoles 1 or 2, reduced or lacking; calyx 2.3-8 mm long, the tube 5-6 mm long, short-cylindric, stramineus to greenish, sparsely black stri- gose, the teeth 1.3-2 mm long; flowers 17-21 mm long, white, spreading to ascending at anthesis, the banner not strongly arched, but folded along the margins below the apex of the blade; pods ascending, stipitate, the stipe 2-2.5 mm long, the inflated body oblong- to 'Life Science Museum and Departuient of Botany and Range Science, Brigham Young University, Prove, Utah 84602. 31 32 Great Basin Naturalist Vol. 45, No. 1 Fii.. 1. Mlu.cnhs drhr.,uurus Welsh: A. ll.hi. „t t.uU.ng plant. B. DotuU ol l.uU. C. Flower cklail. D liiflori'sii'iite. January 1985 Welsh: New Colorado Astragalus 33 lance-ellipsoid, 15-23 mm long, 6-11 mm thick, the valves thinly leathery and straw colored, imilocular, scabrid-pubescent, be- coming glabrous; ovules 18-24. Type.— USA Colorado. Mesa Co.: Wasatch Formation, T9S, R97W, S26 (SE/SE), ca 12 km S of De Beque, pinyon-juniper and mixed desert shrub, at 1647 m elevation, 16 May 1984, S. Welsh, B. Welsh, & R. Kass 22792 (Holotype: BRY; Isotvpes: POM, CAS, UT, UTC, NY, US, MO, COLO, RM, ISC, CS). Additional specimens.— Colorado. Mesa Co., same provenience and date, S. Welsh, B. Welsh, & R. Kass 22793 and 22802 (BRY). Mesa County., T9S, R97W, SW 1/2 of S23, 5.7 mi S of 1-70, 3.2 mi S of pavement end on De Beque Cutoff Road, Atwell Gulch Mem- ber of Wasatch Formation, 12 June 1984 J Anderson 84-15 (BRY). References B vhm:by, R. C. 1984. Atlas of North American Aslraca- Ins. Mem. N. Y. Bot. Garden 13:1-1188. Hahkington, H. D. 1954. Manual of the plants of Colo- rado. Sat^e Books, Denver. 666 pp. Welsh, S. L. 1978. Utah flora: Fahaceac (Lemnninosael. Creat Basin Nat. 38:225-.367. A FOURTH SPECIES OF OREOXIS (UMBELLIFERAE) Stanley L. Welsh' and Sherel Goodrich- .\bstr.\ct.- Described as a new species is Oreoxis twtteh Welsh & Goodrich from Utah. The genus Oreoxis is regarded in contem- porary treatments (Mathias and Constance 1944, Harrington 1954) as consisting of three species, O. alpina (Gray) Coult. & Rose, O. hmnilis Raf., and O. bakeri Coult. & Rose. The three species are reported by Harrington (1954) and occurring at elevations of 2898 to 3965 m in Colorado. Goodrich (1984) notes that O. alpina grows at 2440 to 3475 m and O. bakeri at ca 3660 m in Utah. Thus, we did not expect to find an Oreoxis at ca 1464 m on sandstone in the Courthouse Pasture vicinity northwest of Moab in Grand County, Utah. The elevation at the site where the plants grow is almost 1000 m below the lowest re- ported occurrence of the genus in Utah. Special attention was given to the plant because of its vertical displacement from other known taxa of the genus. A detailed de- scription was prepared from the collection taken upon discovery. That description was then compared to those of other species in the genus, but especially to O. alpinus, with which it is evidently allied. The Courthouse Pasture plants differed in the copious, per- sistent leaf bases and peduncles, more glandular herbage, fewer leaflet segments, and broader leaflets (Fig. 1). The magnitude of the differences dictates that the plants be recognized at specific rank, as follows: Oreoxis trotteri Welsh & Goodrich sp. nov. Siuiilis Oreoxis alpino sed in foliis basis et pe- dunculis persistentibus plantis plus glandu- losis toliolis segmentis paucioribus et foliis segmentis ultimo latioribus differt. Plants pulvinate-caespitose, forming clumps to 30 cm wide, 4-8 cm tall, scabrous and more or less glandular, from a branching caudex, this clothed with a thatch of per- sistent terete leaf bases and peduncles; leaves all basal, bipinnate, with ca 4 opposite pairs of sessile, lateral, primary leaflets, the upper pairs and those of the smaller leaves some- times once-pinnate and then trifid or pinnati- fid; petioles 1-3.5 cm long; blades 1.5-2.3 cm long, oblong in outline, the lowest pair of pri- mary leaflets 3.5-5 mm long, the ultimate segments 1-3.5 mm long, 1-3 mm wide, el- liptic to cuneate-ovate; peduncles 4-7.5 cm long; umbel solitary; involucre lacking; rays 5-7, 3-5 mm long, involucels of 4-7 linear- subulate bractlets 2-3.5 mm long, distinct or essentially so; pedicels obsolete or to ca 1 mm long; calyx teeth ca 1 mm long, green or purplish; petals and stamens yellow; styles 1-1.2 mm long; fruit 2.8-4.8 (5) mm long, the ribs with low, corky wings to 0.7 mm wide. Type.-USA. Utah, Grand County, T24S, R20E, S21, ca 1 km SSE of historic stage sta- tion and 20 km NW of Moab, in mixed juni- per and desert shrub community, on Navajo Sandstone, at ca 1464 m, 30 April 1984, S. L. Welsh and D. Trotter 22729 (Holotype BRY; 10 isotypes to be distributed). Additional specimens: Utah, Grand County, same local- ity as above, 30 May 1984, D. Trotter s.n. (BRY, fruit only). The plants grow in crevices on the joints in Navajo Sandstone, sheltered by pillowlike outcrops in the sandstone. The crevices are more mesic than the surrounding sandstone, which acts like a funnel in concentrating wa- ter that flows from its surface into the crev- ices. The rounded outcrops tend to shade the 'Life Sticntc Museum and Department of Botany and Range Science. Brigham Young University. Provo, Utah 84602. 'USDA Forest Service. Intcrmountain Forest and Range Experiment Station. Ogden, Utah '84401; stationed in Provo, Utah, a. the Shrub Scic 34 fanuary 1985 Welsh, Goodrich: A Species of Oreoxis Fig. 1. Oreoxis trotteri Welsh & Goodnch. A, Hab.t. B, Habitat. C, Fruit. D, Foliar detail and glandularity. E, Leaf^F, Inflorescence detail. G, Floral detail. 36 Great Basin Naturalist Vol. 45, No. 1 plants for a portion of the day, decreasing further the stress from drought. The species is named for Daryl Trotter, of the Bureau of Land Management in Moab, Utah, whose knowledge of the area and will- ingness to help others with its exploration are hereby acknowledged. References Goodrich, S. 1984. Utah flora: Apiaceae (Umbelliferae). Unpublished manuscript. Brigham Young Univ. 9.3 pp. Harrington, H. D. 1954. Manual of the plants of Colo- rado. Sage Books, Denver. 666 pp. NUthias. M. E., and L. Constance. 1945. Oreoxis Raf. North American Flora 29B:168-169. INSECT COMMUNITIES AND FAUNAS OF A ROCKY MOUNTAIN SUBALPINE SERE David J. Schinipfl2 and James A. MacMahon' Abstract.- Insect faunas and communities are characterized for herbaceous and tree canopy layers in meadow aspen, and spruce/fir stages of a northern Utah sere. A greater percentage of species were in fhysanoptera in hofli aspen strata, and a greater percentage of individuals were in Lepidoptera in aspen canopv. Our sites were quite similar to a wide variety of other terrestrial sites in their distribution of species or individuals among orders or metamorphosis categories. Insects/m^ peaked in the aspen stage, but declined in the herbaceous layer with succession. Insects/plant biomass in the herbaceous layer increased with succession. Insects/m^ and insects/foliar biomass were higher in aspen canopies than in conifer canopies. Insect species/m^ peaked in the aspen stage. This statistic was comparable in meadow and aspen understory, and lower in conifer understory. Insects/m^ in the tree canopies were similar to the values in their respective understories. Insect species/plant biomass increased in the herbaceous layer with succession, but decreased in tree canopies with succession. Species evenness in both strata increased with succession. Adult body length was greatest for meadow species, least for conifer species. Adult body length per individual was greatest in aspen. Life cycle complexity was greatest in aspen. Insects on trees were more likely to have complex life cycles than those on herbs. Studies of insects have been infrequent in the hterature of ecological succession (Price 1975). Some have followed through time the insects associated with a discrete nonauto- trophic resource (e.g., Park 1931, Savely 1939, Mohr 1943, Coombs and Woodroffe 1963, Payne 1965). Others have considered the insects of vegetation only in certain strata (e.g., Smith 1928, Martin 1966, Schowalter et al. 1981) or temporal subunits (e.g., Murdoch et al. 1972, Hurd and Wolf 1974, Butt et al. 1980, Purvis and Curry 1980, Force 1981, Hawkins and Cross 1982) of the full sere. Some have been restricted to certain insect orders (e.g., Strohecker 1937, Southwood et al. 1979, Boomsma and Van Loon 1982). Studies more inclusive in space, successional time, and taxonomic coverage (e.g., Shelford 1913) have seldom been very quantitative. Germane to this topic are some studies not ostensibly concerned with succession. These include monitorings of arthropods colonizing vegetation that had been defaunated (Sim- berloff and Wilson 1969) and the studies of insect communities associated with various kinds of vegetation (e.g., Whittaker 1952, Janzen 1973, Werner 1983). We report on a study of the insect commu- nities associated with the above-ground por- tions of three stages of a subalpine forest sere. This work was performed as part of a test of certain of the hypotheses about .successional trends in ecosystem properties proposed by Odum (1969). In particular, we address Odum's hypotheses about species varietv (richness), species evenness, stratification and spatial heterogeneity, body size, life cycle complexity, and stability. All of these attri- butes should, if Odum is correct, increase during succession. Our results for vertebrates have been published (Andersen et al. 1980, Smith and MacMahon 1981) for some of these, as well as other hypotheses. We also present findings on other characteristics of these insect assemblages, including their re- semblance to those reported for various geo- graphic areas or vegetation categories. Study Area Sampling of insects took place on the Utah State University School Forest in the Wasatch Mountains of extreme northern Utah. In the subalpine zone a succession from herbaceous summer-dry meadow through as- pen {Populus tremuloides) grove to coniferous forest {Picea engelmannii and Abies lusio- carpa) is common. Forested stages have un- derstories dominated by herbaceous angio- sperms, with minimal shrub cover. The na- ture of this site and sequence is depicted and 'Department of Biology and Ecology Center, UMC 53, Utah State University, Logan, Utah 84322. •Present address: Department of Biology, University of Minnesota, Duhith, Minnesota 55812. 37 38 Great Basin Naturalist Vol. 45, No. 1 detailed in Schimpf et al. (1980), but the tem- poral relations of these stages should be dis- cussed here. All three stages are thought to be persistent; the meadows are shrinking from slow centripetal vegetative colonization by aspen, which is followed by conifer estab- lishment. The aspen invasion is thought to be allogenic, permitted by climatic warming since the "Little Ice Age" ended 200 to 300 years ago. Aspen establishment autogenically permits conifer invasion through modifica- tion of the environment near the soil surface. Thus the sere may be atypical in the relative- ly equal ages and long-standing spatial prox- imity of the three stages. A typical pioneer stage, that is, a short-lived community fol- lowing disturbance, is absent. Some sampling took place in 1976, but most of our results are based on samples taken, in 1977 and, except for the tree cano- pies, 1978. The 1977 sampling followed a winter with very low snowfall, whereas the other two years were more typical in this re- gard (Schimpf et al. 1980). Methods Field Sampling— Herbaceous Layer Sampling of the herbaceous layer of vege- tation began 1 June and continued weekly through 20 September 1977. In 1978 weekly sampling began 27 June, ending 15 Septem- ber. Some late season samples yielded no in- sects and are not presented in our results. Samples were obtained using a D-Vac suction .sampler (Southwood 1978a). The reliability of this method has been documented (Tormala 1982). Within each serai stage, a net-lined aluminum frame cage was rapidly set over the vegetation. This was done at 20- pace intervals along a randomly chosen com- pass direction in the meadow and conifer stages. In the spatially more restricted aspen stage, the cage was placed at randomly cho- sen coordinates. The cage used for the aspen and conifer understories entrapped the in- sects above 0.25 m^ of ground. The cage used in the meadows covered 0.50 m-, the larger size being practical in the absence of coarse woody litter. From 5 to 15 such samples were taken weekly in each stand of a stage; care was used to avoid sampling the same spot more than once. Two stands were sam- pled in the meadow and conifer stages in 1977 and the meadow and aspen stages in 1978. Three aspen stands were sampled in 1977 and a single conifer stand in 1978. The vegetation surfaces and net interior were completely vacuumed with a narrow hose D-Vac device through a zippered access in the net. The operator continually manipu- lated the vegetation during the vacuuming process, which lasted from 1 to 5 min. The insects were drawn into a nylon-organdy col- lection net and chloroformed within 30 min. Samples were refrigerated until they could be sorted and counted. Field Sampling— Tree Canopy Aspen branches of randomly chosen trees were sampled during the same weeks in the same stands in which understory sampling oc- curred. Muslin bags 79 by 86 cm were thrust over most of the leafy portions of branches while the bag mouth was kept open on a wire hoop on a 3.5 m handle. Immediately after placement, egress was prevented from the terminal 45 cm of the bag by closing its mouth with drawstrings. The branch was clipped next to the bag mouth with an ex- tendible priming shear. Ladders were used to gain access to branches up to 7 m above ground. Conifer canopy samples were taken every third week from 7 June to 12 August 1977. Samples were taken in two fir-dominated plots, two spruce-dominated plots, and two plots where spruce and fir co-dominated. In each plot six trees more than 15 cm diameter at breast height were randomly selected for sampling. Each tree was divided into vertical thirds, and one sample was taken from each third every time that individual was sampled. Samples came from a randomly chosen com- pass sector of the canopy. If no branches oc- curred within a particular vertical third of the tree, no other trees were substituted. Sampling sites were reached by use of movMitain-climbing equipment and tech- niques. This included belaying the sampler in a sling and strap to give sufficient mobility. Branches sampled were alwavs located above the sampler's position to reduce mechanical disturbance of the branch and its insects. January 1985 ScHiMPF, MacMahon: Insect Communities Branches were enclosed in muslin bags 1.6 m deep fastened by velcro strips to a 25 x 100 cm wire frame with aim long handle. After drawing the bag over the end of the branch, the bag was quickly closed about 30 cm from the mouth with a drawstring. The branch was sawn off, and the bag removed from the frame and dropped to the ground. The sample bag was put into a heavy plastic bag for 30 min with cotton balls soaked in CCI4. The sample bag was opened and the branch was vigorously shaken within it before it was removed and cut up for drying. After drying, the cut branch parts were inspected for in- sects, which were added to those originally collected. Subcortical insects were not sampled. Laboratory Procedures Arthropods were sorted from samples un- der magnification. In contrast to some recent taxonomies (e.g., Daly et al. 1978), we in- clude CoUembola within Insecta. Most in- sects were identified at least to family, then assigned to a binomial species or numbered morphospecies within the family. Some could not be so identified with confidence (e.g., Mi- crolepidoptera). Well-preserved representa- tive adults were incorporated into a synoptic reference collection. Many voucher speci- mens were verified by the following tax- onomists: D. C. Lightfoot (Orthoptera), J. D. Lattin (Hemiptera), P. W. Oman (Homop- tera), W. J. Hansen (Diptera), G. E. Bohart, and G. R. Ferguson (Hymenoptera). The number of individuals in each species was recorded for each sample. Orders were assigned to a metamorphosis type: ametabo- lous, hemimetabolous, paurometabolous, or holometabolous. Adult body length was esti- mated for each species by measuring head- anal length of the largest reference specimen to the nearest mm with a rule. Treatment of Data Dry mass of leaves was determined for each sample from the tree canopies. Data from individual samples were aggregated by computer to generate data for least-squares linear regressions of insect species number vs. leaf biomass and number of insect individuals vs. leaf biomass for each week. E.stimates of leaf biomass/m2 (Williams 1977) were en- tered into the regression models to obtain in- terpolated estimates of mean number of spe- cies/m- and mean number of individuals/ni- in aspen or conifer canopies during each week. D-Vac sample data were combined by computer to derive numbers representing I.O m2 of ground sampled (sample pairs in mead- ows, quartets in forest understories). Not all possible combinations were created, with the number of pairs or quartets equaling the number of samples. These aggregated data were used to estimate mean number of spe- cies/m2 and the J' index of evenness (Pielou 1975). Number of individuals/m- was esti- mated by doubling or quadmpling the num- ber in a single sample. It is possible that the use of the larger cage in the meadow biased the results by capturing insects more effi- ciently. This could bias the estimates of abun- dance or species richness. Species richness es- timates could also be biased if the species- area relationship does not have a slope equal to 1.0 over the 0.25 m- to 0.50 m- range. To test for these biases, a set of samples was taken in the same meadows with the smaller cage as well as the normal set with the larger cage on 18 July 1978, a day when insect abundance was reasonably high. The esti- mates of individuals/m2 and species/m- ob- tained from large samples were statistically compared to those from small samples by means of a t-test. Total number of species/m^ in the aspen and conifer stages was estimated by summing the estimates from the understory and the canopy. We justify this on the basis of ob- served, but unquantified, limited faunal over- lap between the two strata in each forested stage. Because species/m^ values in tree can- opies do not represent integral numbers of canopy samples, combining of canopy and understory samples for the direct counting of species was not possible. Spatial homogeneity of the insect commu- nity of the herbaceous layer was assessed by the resemblance among samples taken in the same stage on the same day. Absolute per- cent similarity (Goodall 1973) was computed for each possible pair of samples on the basis of number of individuals in each species. The 40 Great Basin Naturalist Vol. 45, No. 1 effect of cage size was first tested by com- paring similarity among large vs. small cage samples taken on 18 July 1978 with a t-test after transforming the similarity coefficients to arcsin values. The seasonal course of the means of the following statistics was plotted as second or- der least-squares regressions vs. time; number of individuals/ m 2, number of species/m-, J' evenness, and absolute percent similarity. Because samples from the canopy were from foliage equivalent to varying amoimts of ground surface, analysis of their species evenness by J' was problematical. Instead, relative evenness was inferred from domi- nance-diversity plots (see, e.g., Murdoch et al. 1972) for all canopy insects captured dur- ing 1977. For similar reasons, it was not ap- propriate to calculate a composite measure of diversity, i.e., one including both richness and evenness components, for the canopies. This prevented us from calculating a compos- ite diversity measure for the combined strata of our forested stages. To be consistent with our other analyses, we did not compare com- posite diversities in the herbaceous stratum alone. Even if we could calculate composite diversity measures, they might not provide any more useful information than the mea- sures we present. We agree with Dritschilo and Erwin (1982) that the use of diversity in- dices is often redundant or misleading, as thev found for carabid beetle communities. A summary of all samples taken in each stage and stratum was used to judge resem- blance between the study site and other loca- tions for which insect summaries have been published. Relative percent similarity (Good- all 1973) was calculated between each of our five stages/strata and these other areas. Thiswas calculated on four different bases: faunal ordinal similarity (proportion of all species in each order), community ordinal similarity (proportion of all individuals in each order), faunal metamorphosis similarity (proportion of all species in each metamor- phosis category), and community metamor- phosis similarity (proportion of all individuals in each metamorphosis category). Results and Discussion Approximately 80,000 insects were ob- tained by the sampling procedures and used for most of the analyses which follow. This total is composed of about 18,000 from the herbaceous layer in 1977, 13,500 from the canopy in 1977, and 48,500 from the her- baceous layer in 1978. As will be seen below, the greater number in 1978 reflected greater insect abundance, not an increase in sampling effort. Taxonomic Composition The distribution of the catch among orders is listed in Table 1, which is based on some Tablk 1. Ordinal composition of inst'cls cans^lit in eacli stratinn/snccessional sta<4e from 197(i throuiili 1978. Figures are the percentages contributed by each order to the total number of families, species, or individuals. Families Aspen Aspen Conii.r ( lonifer Aspen Meadow understory canopy underst(.i\ canopy Meadow understorv Coleoptera 15.8 12.2 13.5 7.5 16.0 10.9 3.2 Collembola 2.6 3.1 5.4 3.8 1.2 1.2 1.5 Diptera 27.6 .34.7 28.4 40.0 28.4 21.5 26.1 Hemiptera 11.2 4.1 5.4 (12 6.2 10,1 2.9 llomoplera 8.6 10.2 13.5 S..S 9.9 n.i 7.6 Hyinenoptera 23.0 29.6 24.3 23.8 27.2 38.5 42.2 I-epidoptcra .5.3 1.0 4.0 2.5 2.5 2.9 0.2 Neuroptera 2.6 1.0 0.0 2.5 4.9 0.8 0.2 Odonata 0.0 0.0 1.4 0.0 0.0 0.0 0.0 Orlhoplera 1.3 1.0 1.4 0.0 0.0 0.7 0.5 Psocoptera ().() 0.0 0.0 1.2 1.2 0.0 0.0 Thysanoptera 2.0 3.0 2.7 3.8 1.2 ■7 ■-> 15.6 Trichoptcra 0.0 0.0 98 0.0 0.0 1.2 0.0 0.0 T.nM .MMbtll 152 74 80 81 585 410 January 1985 ScHiMPF, MacMahon: Insect Communities 41 11,000 insects caught in 1976 or 1978 by the same procedures in addition to the 80,000 just mentioned. Inspection shows a lack of any obvious successional trends in fauna! composition when orders are weighted by their number of families. At this higher tax- onomic level, the various stages and their strata exhibited broad resemblance to one an- other. The three orders with the greatest numbers of families were always Diptera, Hymenoptera, then Coleoptera. This se- quence differed from those in aspen or white spruce {Picea glauca) stands in Alaska (Wer- ner 1983). Neither our study nor Werner's sampled subcortical insects, so Coleoptera may be imderrepresented. If orders are weighted by their number of species, a conspicuous rising and falling of Thysanoptera was observed during succes- sion. Hymenoptera always exceeded Diptera, except in the conifer understory, where they were nearly equal. Diptera composed a greater proportion in the understory than in the canopy for both forested stages. Weighting of orders by number of individ- uals reveals the great predominance in aspen canopy of Lepidoptera, a group with rela- tively low abimdance elsewhere in the sere. These are mostly leaf-mining Gracilariidae, principally PlujUocnistis populiella Cham- bers. Diptera were a reduced proportion of the aspen canopy commimity. Again, Diptera exhibited greater representation in forest un- derstories than in the corresponding canopies. Faunal ordinal similarity of our five insect assemblages, i.e., those of different strata/successional stages, to tho.se from other areas is high in most cases (Table 2). Aspen understory and canopy have the lowest mean similarity values; in most cases this is prob- ably due to the high proportion of Thysanop- tera species in these two faunas (Table 1). The mean similarity for all comparisons with other sites (72%) is surprisingly close to the mean similarity among our five stages/strata (80%). Even the similarity of faunas as dis- parate to a subalpine sere as those of tropical areas, often 70% or more (Table 2), .suggests that local relative species richness of insect orders may be little influenced by habitat conditions. At this level of analysis, terrestrial insect assemblages show minimal variation, perhaps due to inherent differences among orders in speciation or extinction rates (Fow- ler and MacMahon 1982). Faunal metamorphosis similarity (Table .3) is even higher than ordinal similarity (Table 2), the former having an overall mean of 91% and a mean of 96% among our stages/ strata. Variation among sites in species richness of a particular order tends to be compensated for by other orders with the same metamorphosis type. All faunas are dominated by two of the four metamorphosis types, paurometabolous and holometabolous. Table 1 conti inued. Species Individuals Aspen Conifer Conifer Aspen Aspen Conifer Conifer canopy understory canopy Meadow imderstory canopy understory canopy 4.6 3.6 14.7 1.3 0.3 0.1 0.2 2.0 1.4 2.7 0.4 8.6 5.5 0.1 3.9 1.0 16.4 33.8 21.9 5.5 26.0 0.9 17.6 7.3 3.6 5.3 7.2 8.6 0.7 0.2 1.8 16.2 11.4 12.9 11.5 42.3 52.9 29.1 45.4 21.4 49.1 32,4 33.0 8.6 13.6 7.2 4.1 31.4 1.8 2.7 7.5 0.3 0.0 40.9 0.4 3.3 0.0 0.9 1.8 0.2 ().() 0.0 0.1 1.3 0.4 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.4 0.0 0.0 0.2 0.0 0.0 0.0 0.0 0.0 0.9 1.1 0.0 ().() 0.0 0.0 0.1 11.0 4.9 0.4 24.3 1.0 21.4 26.4 1.5.9 0.0 0.0 0.7 0.0 0.0 0.0 0.0 0.0 281 225 279 55,419 6.508 16,607 6,024 6.726 42 Great Basin Naturalist Vol. 45, No. 1 Community ordinal similarity (Table 4) av- erages 47%, much less than faunal ordinal similarity. Mean similarity of our subalpine communities to one another (59%) is notice- ably greater than this. Local habitat condi- tions evidently affect communities more than they affect faunas. The aspen canopy has es- pecially low similarity to other assemblages, likely due to its very high proportion of Lepidoptera (Table 1). All of the other as- semblages are from lower strata (except coni- fer canopy), but it is unclear whether we should hypothesize that lepidopterous leaf miners are more common in deciduous cano- pies than in lower strata. This is because these other assemblages resulted from sam- pling procedures that would not collect immature leaf miners. Community metamorphosis similarity (Table 5) averages 66%, more than commu- nity ordinal similarity and less than faunal metamorphosis similarity. Variation among sites in abundance of a given order tends to be compensated for by other orders with the same type of metamorphosis. As compared with faunas, ametabolous insects are more common and holometabolous ones less com- mon. Metamorphosis similarity among our subalpine communities averages 82%, again quite a bit above the general mean for all comparisons. Metamorphosis categories are few in num- ber and so similarly distributed in various in- sect assemblages that they seem to tell little Table 2. Relative percent similarity of insect faunas to those of each stratum and stage of a subalpine sere. Com- parisons are based on the proportions of the total number of species in Coleoptera, Diptera, Hemiptera, Homoptera, Lepidoptera, Thvsanoptera, and all other orders combined. For sources of data from other areas, see Appendix 1. Relative percent similarity to Meadow Aspen understorv Aspen canopy Conifer understory Conifer canopy 94 Cold desert 1 86 Aspen canopy 86 Aspen understorv 87 British trees 91 Cold desert 1 90 C^onifer canopv 80 Conifer imderstorv 80 Meadow 84 Tropical 4 90 Meadow 88 Tropical 9 79 Cold desert 1 78 Cold desert 1 83 Tropica! 8 89 Tropical 9 86 Deciduous 2 78 Meadow 76 Conifer imderstorv 82 Conifer canop\ 89 British trees 86 British trees 75 British trees 73 Conifer canopv 82 Meadow 88 Deciduous 2 82 Temperate field 74 Temperate field 72 British trees 80 Tropica! 12 86 Canada 82 Conifer 72 Deciduous 2 72 Tropica! 1 1 SO Deciduous 2 85 Tropical 7 understorv 81Tropicar2 71 Conifer canops 71 Tropical 9 80 .Aspen understory 85 Temperate field 81 Tropical 7 70 Tropical 8 71 Tropica! 5 79 Tropical 7 85 South .\frican 80 Aspen canopv 70 Canada 69 Temperate field 78 Cold desert 1 trees 84 Tropical 5 80 Tropical 5 68 South African 68 Deciduous 2 78 Tropical 9 84 Tropical 2 78 Tropical 4 68 Tropical 7 68 Tropical 10 76 Aspen canop\ 82 Conifer understory 78 Aspen tmderstorv 68 Tropical 4 67 Tropical « 76 Temperate field 81 Tropical 4 78 South African 67 Tropical 9 66 Tropical 7 76 South African 81 Tropical 1 trees trees 78 Tropical 12 66 Tropical 1 1 66 South African 75 ('anada 80 Tropical 8 78 Tropical 8 66 Tropica! 1 trees 65 Deciduous 1 75 Hot desert 79 Hot desert 76 Canada 65 Hot desert 64 Tropical 4 72 Tropical 2 78 Tropical 3 76 Tropical 1 1 64 Tropical 12 64 Tropical 1 72 Tropical 13 78 Deciduous 1 76 Deciduous 1 63 Tropical 5 63 C:anada 72 Tropica! 1 77 Tropical 12 76 Tropical I 62 Deciduous 1 63 Tropical 3 72 Tropical 5 76 U.S. + Canada 74 Tropical 10 62 Tropical 10 62 Tropical 2 70 Deciduous 1 75 Tropical 13 73 Tropical 3 .59 Tropical 2 60 Tropical 12 69 Tropical 3 75 Tropical 1 1 71 Tropical 6 59 Tropical 3 58 Tropical 6 63 Tropical 1 1 74 Tropical 10 71 Hot desert .58 Tropical 13 .58 Hot desert 62 Tropical 10 74 Tropical 6 69 Tropical 13 53 I'.S. + Canada 56 Tropical 13 62 Tropical 6 74 New York State 67 U.S. + Clanada .53 New York State .53 U.S. + C;anada 61 U..S. + Canada 73 Aspen canopy 71 \spcn understor\- 64 New York State 51 Tropical 6 .50 New York State- 60 New York State ,52 world 1 .36 world 1 40 world 1 44 world 1 61 world 1 77 mean 66 mean 66 mean 7-3 mean 80 mean January 1985 ScHiMPF, MacMahon: Insect Communities 43 about differences in community structure (sensu MacMahon et al. 1981) or faunal com- position. Even feeding categories may show similar patterns on different species of trees in widely separated areas (Moran and South- wood 1982). Insect Abundance Herbaceous Layer.— The effect of cage size on estimates of number of insect individ- uals/m- is summarized in Table 6. In each of the two meadows tested, samples taken with the large cage gave notably greater estimates of insect abundance than did those from the small cage, but the differences were not sta- tistically significant. Inspection of summaries of individual samples revealed that most of the differences in means could be accounted for by high numbers of Aphididae or Thripi- dae in a few of the samples taken with the large cage. Because these are nonflying or weakly flying organisms, it seems doubtful that they escaped more readily during place- ment of the small cage. We conclude that most of the difference in means is due to sampling error, but acknowledge that the large cage may be marginally more efficient at trapping strong fliers such as large Diptera or Hymenoptera. The latter groups probably made up a small portion of the total assem- blage in the field. Insect density in the herbaceous layer de- creased through most of the summer of 1977 Table 3. Relative percent similarity of insect faunas to those of each stratum and stage of a suhaipine sere. Comparisons are based on the proportions of the total number of species in ametabolous, hemimetabolous, paurometabolous, and holometabolous orders. For sources of data from other areas, see Appendix 1. Meadow Aspen understory Relative percent similarity to Aspen canopy Conifer understory Conifer canopy 99 Deciduous 2 98 Tropical 3 98 Tropical 8 98 Conifer understory 97 Aspen understory 97 Aspen canopy 97 Tropical 7 97 British trees 94 Tropical 9 94 Conifer canop\- <^n Tropical 4 94 South African trees 94 Cold desert 1 93 Tropical 5 93 Tropical 2 93 Tropical 12 92 Tropical 1 91 Tropical 10 90 U.S. -I- Canada 89 World 2 89 Tropical 11 88 Tropical 6 87 New York State 86 Hot desert 85 Temperate field 84 Deciduous 1 84 Canada 79 Cold desert 2 71 Tropical 13 91 mean 100 Aspen canopy 99 Tropical 7 97 Meadow 97 Tropica] 9 97 Conifer understory 96 Deciduous 2 96 Tropical 3 95 Tropical 8 95 Tropical 2 95 Tropical 12 94 British trees 91 Conifer canopy 91 Tropical 4 91 South African trees 91 Tropical 6 91 Cold desert 1 90 Tropical 5 89 Tropical 1 88 Tropical 10 87 Deciduous 1 87 U.S. -I- Canada 86 Tropical 1 1 86 World 2 84 New York State 84 Hot desert 82 Temperate field 82 Canada 81 Cold desert 2 73 Tropical 13 90 mean 100 Aspen understory 98 Meadow 98 Tropical 7 97 Aspen canopy 97 Meadow 97 Aspen understory 97 Conifer understory 96 Tropical 3 97 Tropical 9 ' 96 Tropical 8 96 Deciduous 2 96 Tropical 3 95 Tropical 2 95 Tropical 12 95 Tropical 8 94 British trees 92 Conifer canopy 91 South African trees 91 Cold desert 1 91 Tropical 4 91 Tropical 6 90 Tropical 5 89 Tropical 1 88 Tropical 10 87 U.S. -I- Canada 87 Deciduous 1 86 Tropical 1 1 86 World 2 85 New York State 84 Hot desert 82 Temperate field 82 Canada 81 Cold desert 2 73 Tropical 13 94 mean 96 Deciduous 2 96 British trees 96 Tropical 7 94 Conifer canopy 94 Tropical 9 93 South African trees 93 Tropical 12 93 Cold desert 1 92 Tropical 2 92 Tropical 4 92 Tropical 5 91 Tropical 1 90 Tropical 10 89 U.S. -I- Canada 88 Tropical 11 88 World 2 87 Tropical 6 87 New York .State 86 Hot desert 84 Temperate field 84 Canada 83 Deciduous 1 78 Cold de.sert 2 71 Tropical 13 90 mean 99 Cold desert 1 99 Tropical 4 98 Tropical 5 97 South .African tree* 97 Tropical 1 97 Tropical 10 96 Tropical 8 96 U.S. + Canada 96 British trees 95 Tropical 3 95 Deciduous 2 95 World 2 95 Tropical 1 1 94 Meadou' 94 Conifer understory 93 New York State 92 Hot desert 92 Tropical 7 92 Aspen canopy 91 .\spen understory 91 Temperate field 90 Canada 89 Tropical 9 89 Tropical 12 88 Tropical 2 83 Tropical 6 79 Deciduous 1 73 Cold desert 2 67 Tropical 13 91 mean 44 Great Basin Naturalist Vol. 45, No. 1 in the meadows (Fig. la), but rose until the third week of July (week 8) under aspen and conifers (Fig. lb, Ic), declining thereafter. The earlier peaking of the abundance curve in the meadows probably resulted from the earlier melting of snow in this stage as com- pared with the others; our sampling schedule was most likely too late in this year of early snowmelt to include the period of rising abundance in the meadows. Abundance un- Table 4. Relative percent similarity of insect communities to those of each stratum and stage of the sere. Comparisons are based on the proportions of the total insect catch in Coleoptera, Diptera, Hemiptera, Homoptera, Lepidoptera, Thysanoptera, and all other orders combined. For sources of data from other areas, see Appendix 1. Meadow Relative percent similarity to .\spen understory Aspen canopy Conifer understory Conifer canopy 82 Conifer understory 72 Tropical 2 70 Cold desert 2 66 Salt marsh 1 65 Deciduous 1 64 C^onifer canopy 64 Aspen imderstory 62 Tropical 1 2 60 Tropical 13 60 Salt marsh 4 60 Deciduous 1 1 59 Aspen canopy 59 Salt marsh 5 58 Deciduous 10 57 Deciduous 7 .56 Salt marsh 2 55 Pine 53 Pine heath 52 Grass bald 52 Tropical 7 51 Salt marsh 1 49 Deciduous 13 48 Tropical 9 46 Spruce /fir 1 44 Tropica! 3 43 Tropical 1 1 42 Hemlock 42 Deciduous 6 42 Tropical 4 42 Tropical 8 41 Deciduous 8 .39 Tropical 10 37 Tropical 6 .36 Tropical 1 35 Deciduous 12 32 Deciduous 5 31 Alpine drv 31 Deciduous 3 31 Heath bald .30 Deciduous 2 .30 Tropical 5 26 Irish meadow 25 Spruce/fir 2 22 Deciduous 9 18 Alpine wet 47 mean 85 Tropical 12 83 Deciduous 1 80 Deciduous 10 79 Deciduous 11 75 Salt marsh 4 73 Conifer imderstory 72 Salt marsh 5 72 Salt marsh 2 71 Deciduous 7 71 Pine heath 70 Tropical 2 69 Grass bald 69 Salt marsh 3 68 Deciduous 13 67 Tropical 13 66 Tropica! 7 65 Spnicc/fir 1 64 C:o!d desert 2 64 Meadow 62 Pine 62 Salt Marsli 1 61 Deciduous 6 61 Tropical 8 60 Hemloclv 58 Deciduous 8 57 Deciduous 4 .56 Deciduous 12 .55 Tropical 1 •53 Deciduous 5 .52 Tropica! 9 49 Tropica! 3 49 Alpine drv 47 Heath bald 47 (Conifer canopy 46 Deciduous 3 46 Deciduous 9 45 Spruce /fir 2 43 Tropical 1 1 .39 Tropical 10 39 Aspen canopy .36 Alpine wet .3.3 Tropical 6 27 Tropical 5 21 Deciduous 2 20 Irish meadow 58 mean 59 Meadow 56 Conifer understor\- 49 Conifer canopy 45 Deciduous 13 43 Deciduous 1 41 Cold desert 2 41 Deciduous 10 41 Deciduous 11 40 Tropical 7 39 Deciduous 7 39 Tropical 2 .39 Aspen understorv 38 Tropical 12 38 Tropical 8 37 Grass bald 37 Tropical 13 35 Pine heatli 34 Pine 33 Spruce/fir 1 .30 Hemlock .30 Deciduous 8 .30 Salt marsh 1 .30 Salt marsh 2 .30 Salt marsh 3 .30 Salt marsh 4 .30 Salt marsh 5 30 Tropical 9 28 Deciduous 6 26 Tropical 3 25 Tropical 1 1 24 Deciduous 12 24 Tropical 4 23 Tropical 10 23 Deciduous 5 22 Tropical 1 21 Deciduous 3 20 Deciduous 2 19 Tropical 6 17 Tropical 5 16 .\lpine drv 15 Heath bald 13 Deciduous 9 11 Spruce/fir 2 1 1 Alpine wet 8 Irish meadow 30 mean 82 Meadow 70 Deciduous 7 '3 Aspen imderstory 68 Tropical 6 72 Deciduous 1 67 Cold desert 2 67 Salt marsh 4 67 Tropica! 12 64 Salt marsh 2 63 Deciduous 10 63 Deciduous 1 1 61 Tropical 2 61 Salt marsli 3 60 Tropical 13 56 Aspen canopy 56 Deciduous 7 55 Tropical 7 55 Salt marsh 1 .54 Deciduous 13 54 Salt marsh 5 54 Conifer canopy 53 Pine heath 53 Tropical 8 52 Pine 52 Grass bald 49 Spruce /fir 1 48 Tropical 9 45 Hemlock 44 Deciduous 6 42 Deciduous 8 40 Deciduous 12 40 Tropical 1 40 Tropical 4 39 Tropical 3 36 Deciduous 5 .33 Tropical 1 1 33 Deciduous 3 32 .\lpine dr\- .30 Heath bald .30 Tropical 10 28 Deciduous 9 28 Alpine wet 26 Spruce/fir 2 25 Tropical 6 23 Deciduous 2 22 Irish meadow 19 Tropical 15 48 mean 66 Grass bald 66 Pine 64 Meadow 63 Tropical 3 62 Tropical 1 62 Pine heath 61 Deciduous 7 61 Tropical 11 60 Tropical 10 60 Deciduous 10 60 Cold desert 2 .59 Deciduous 6 59 Deciduous 1 1 .56 Tropical 9 55 Deciduous 5 54 Conifer understory 54 Tropical 4 .53 Tropical 12 52 Tropical 1 51 Deciduous I 51 Deciduous 13 51 Salt marsh 5 51 Hemlock .50 .\lpine drv 49 Salt marsh 4 49 Spruce /fir 1 49 .\spen canopv 48 Heath bald 47 Deciduous 9 47 Aspen understory 47 Tropica! 7 46 Deciduous 12 43 Deciduous 3 41 Spruce/fir 2 40 Deciduous 2 .39 Tropical 5 39 Tropical 13 38 Tropical 8 36 Salt marsh 2 .35 Salt marsh 3 31 Salt marsh 1 22 Irish meadow 20 Alpine wet 51 mean January 1985 ScHiMPF, MacMahon: Insect Communities 45 der aspen was comparable to that in mead- ows, whereas that under conifers was decid- edly lower. Sizable variation among stands occurred within all three stages. Peak abundance in 1978 was from two to six times the 1977 levels (Fig. Id, le. If) and occurred during the first two weeks of Au- gust. The later peak dates of this year corre- spond to the later disappearance of snow, re- flected in the much later onset of sampling as compared with 1977. Seasonal change was more pronounced than in 1977. Abundance Table 5. Relative percent siinilaritv of insect eomniunities to those of each stratnm and statue of the sere. Comparisons are based on the proportions of the total insect catch in ametabolous, hemimetabuloiis. paurometabolous, and holometabolous orders. For sources of data from other areas, see Appendix 1. Meadow Aspen understory Relative percent similarity to Aspen canopy Conifer understory Conifer canopy 94 Conifer understory 91 Salt marsh 4 91 Tropical 2 91 Salt marsh 2 90 Cold desert 2 89 Salt marsh 3 88 Tropical 13 83 Salt marsh 1 78 Deciduous 1 76 Aspen understory 76 Deciduous 2 72 Tropical 9 70 Conifer canopy 67 Aspen canopy 64 Tropical 12 62 Tropical 7 60 Grass bald 60 Deciduous 1 1 60 Deciduous 7 59 Salt marsh 5 58 Pine 58 Deciduous 10 55 Tropical 8 55 Pine heath 53 Tropical 6 53 Tropical 3 52 Tropical 4 48 Deciduous 13 47 Tropical 1 1 46 Spruce/fir 1 44 Tropical 10 44 Tropical 1 42 Deciduous 8 41 Hemlock 40 Tropical 5 39 Deciduous 6 35 Deciduous 12 32 Alpine dry 31 Deciduous 5 31 Deciduous 3 31 Heath bald 30 Irish meadow 25 Spruce /fir 2 23 Deciduous 9 22 Alpine wet 59 mean 97 Deciduous 1 94 Conifer canopy 91 Aspen canopy 88 Tropical 12 ' 84 Grass bald 84 Deciduous 1 1 83 Deciduous 7 82 Pine 82 Deciduous 10 82 Salt marsh 5 82 Tropical 7 81 Conifer understory 80 Cold desert 2 80 Tropical 9 80 Tropical 2 79 Pine heath 77 Salt marsh 4 76 Meadow 76 Tropical 8 74 Tropical 6 73 Tropical 3 73 Tropical 4 72 Deciduous 13 70 Salt marsh 2 68 Deciduous 2 68 Salt marsh 3 68 Spmce/fir 1 67 Tropical 1 1 67 Tropical 13 66 Deciduous 8 65 Tropical 1 64 Hemlock 64 Tropical 10 63 Deciduous 6 62 Salt marsh 1 59 Deciduous 12 55 Alpine dry 55 Deciduous 5 55 Deciduous 3 55 Heath bald 47 Deciduous 9 46 Alpine wet 45 Spruce/fir 2 43 Tropical 5 27 Irish meadow 70 mean 96 Conifer canopy 96 Cold desert 93 Grass bald 92 Tropical 12 92 Deciduous 7 92 Deciduous 1 1 92 Salt marsh 5 91 Pine 91 Aspen understory 89 Deciduous 1 89 Deciduous 10 86 Pine heath 81 Deciduous 13 80 Tropical 8 77 Tropical 4 77 Cold desert 2 76 Tropical 7 75 Tropical 9 75 Deciduous 8 74 Tropical 3 73 Conifer understory 73 Salt marsh 4 73 Hemlock 72 Spruce/fir 1 71 Deciduous 6 71 Tropical 2 70 Tropical 6 69 Tropical 1 67 Meadow 66 Salt marsh 2 65 Deciduous 12 65 Alpine dry 65 Deciduous 5 64 Deciduous 3 64 Salt marsh 3 64 Heath bald 62 Tropical 1 1 61 Tropical 13 59 Deciduous 2 59 Tropical 10 58 Salt marsh 1 56 Deciduous 9 55 Alpine wet 49 Spruce/fir 2 37 Tropical 5 22 Irish meadow 72 mean 96 Salt marsh 4 94 Meadow 90 Tropical 2 89 Salt marsh 2 87 Salt marsh 3 86 Tropical 13 81 Aspen understory 81 Salt marsh 1 79 Deciduous 1 77 Conifer canopy 74 Tropical 9 73 Aspen canopy 71 Deciduous 2 70 Tropical 12 66 Grass bald 66 Deciduous 1 1 66 Deciduous 7 65 Salt marsh 5 64 Pine 64 Deciduous 10 63 Tropical 7 61 Pine heath 57 Tropical 8 55 Tropical 6 55 Deciduous 13 54 Tropical 3 54 Tropical 4 49 Tropical 1 1 49 Spruce/fir 1 48 Deciduous 8 47 Hemlock 46 Tropical 10 46 Deciduous 6 46 Tropical 1 41 Deciduous 12 41 Tropical 5 38 Alpine dry 38 Deciduous 5 38 Deciduous 3 37 Heath bald 29 Deciduous 9 28 Alpine wet 26 Spruce/fir 2 26 Irish meadow 96 Aspen canopy 94 Aspen understorv 93 Deciduous 1 90 Tropical 12 89 Grass bald 89 Deciduous 7 89 Deciduous 1 1 88 Salt marsh 5 87 Pine 87 Deciduous 10 83 Pine heath 79 Cold desert 2 78 Deciduous 13 77 Tropical 7 77 Conifer understorv 77 Tropical 8 76 Tropical 9 76 Salt marsh 4 75 Tropical 2 74 Tropical 3 74 Tropical 4 71 Deciduous 8 71 Tropical 6 70 Meadow 70 Hemlock 69 Spruce/fir I 69 Salt marsh 2 68 Deciduous 6 67 Salt marsh 3 66 Tropical 1 65 Tropical 13 63 Tropical 1 1 63 Deciduous 2 62 Deciduous 12 61 Deciduous 5 61 Salt marsh 1 61 Alpine dry 60 Deciduous 3 60 Heath bald 60 Tropical 10 .52 Deciduous 9 52 .Alpine wet 46 Spruce /fir 2 38 Tropical 5 23 Iri.sh meadow 71 mean 46 Great Basin Naturalist Vol. 45, No. a A DVAC/MEADOW/l/1977 ° 9VAC/MEADOW/2/1977 b A DVAC/ASPEN/1/1977 D DVA_C/ASPE;N/2/1977 O DVAC/ASPE:N/3/1977 D-;^S ..gfl^^Q^O^^- 800- d A DVAC/MEADOW/1/1978 D DVA_C/MEAD_0_W/2^1978 600- ^'~-^ /' '= X 400- 200- /•/' A\ '\ / / \ ' 0- e A DVAC/ASPEN/V1978 n DVA_C/ASP_EN/5/1^9_78 C A DVAC/C0NIFER/V1977 ° DVA_C/CqNIFER/2/197_7 DVAC/CONIFER/1978 0 2 4 6 8 10 12 14 16 18 20 0 2 4 6 8 10 12 14 16 18 20 0 2 4 6 8 10 12 14 16 18 20 SAMPLE WEEK SAMPLE WEEK SAMPLE WEEK Fig. L Mean number of individuals/m^ in meadows, aspen understories, and conifer understories as functions of sample week (second order least-squares regressions). Plotted points are weekly means for 1977 (a-c) and 1978 (d-f). Two regression curves are presented for DVAC/MEADOW/2/1978 samples in Fig. Id. The more steeply peaked of the curves is based on all field data collected, including a mean value of 1372 individuals/m^ on sample week 8 (point not shown). This anomalously high value was the result of a concentration of .3477 aphids within one of the eight DV.AC samples taken that week. When this value was replaced with 35 aphids, the average value for the other seven DV.AC samples taken that week, the mean density was reduced to 527 individuals/m^ (point shown on plot), and the less steeply peaked regression curve was obtained. level differed most between years in the meadows, and least in the two forest under- stories. This interyear trend was reasonably consistent with tliat of the herbaceous stand- ing crop, which was 2.3 times as great in the 1978 meadows as during the 1977 drought; the ratios for a.spen and conifer understory standing crop are 1.5 and 1.1, respectively (data from Appendix 2). Abundance variation among stands within stages was sizable again in 1978 but did not obscure the serai trend. On the basis of this more typical mete- orological year (Schimpf et al. 1980), we con- clude that insect abundance/m- in the her- baceous layer declines markedly during succession. Tahi.e 6. Klfect of saiiipliuii with large (()..5() m^) vs. small |().25 nr) D-VAC cage on estimates (mean number of individuals/nr), species density (mean iiumlier of species/m^), and sample sii lute percent similarity). Eight samples were laktii witli fadi ca-^e in each meadow l.S |iil\ U)7S. Meadow Cage Species density Small 228 Large 254 Small .331 Large 480 .10 Sample similarity po ^ 16.6 .10 17.0 11.8 .05 15.3 •Probability that the mean from the larRe laRe is greater due to chance alone. tBawd on similuritv coefficients transformed to arcsin. January 1985 ScHiMPF, MacMahon: Insect Communities 47 700 600 500- 400 A CANOPY/ASPEN/1/1977 ° ?A!iOP^A.SP^N/2/1_977 O CANOPY/ASPEN/3/1977 300 200 0 2 4 6 8 10 12 14 16 18 20 SAMPLE WEEK Fig. 2. Mean number of individuals/m^ in aspen can- opies as functions of sample week, 1977 (second order least-squares regressions). Plotted points are weekly means estimated from abimdance vs. biomass regressions. The trend in 1978 herbaceous standing crop along the sere was the same as that of insect abundance, though the ratios between stages differed. Meadows had 2.9 times the standing crop of aspen understory but only about 2.1 times the peak in.sect density; the meadow: conifer ratios were 10.0 for plants and 2.8 for insects (plant data from Appendix 2). Thus, insect abundance increases during succession if expressed per unit of above- groimd herbaceous biomass rather than per m-. Conifer imderstories had only about half as many plant species/m- as meadows or aspen imderstories (Appendix 2). Thus, our abun- dance trend is consistent with the findings by Pimentel (1961) and Root (1973) of greater concentrations of insects on collards growing in monocultures than on those growing in di- verse plant communities. This agreement must be qualified by the fact that the host plant species differ among our three stages. The representation of plant families and life cycle lengths is much the same in all stages (Schimpf et al. 1980), yet the stages could still differ in the physical, chemical, or archi- tectural (sen.su Lawton 1983) suitability of their plants for iasects. Maiorana (1981) pro- posed that shade itself promotes a greater abundance of in.sect herbivores per unit plant biomass, and Schowalter (1981) suggested that .shade-induced physiological stress on some plants would lessen their biochemical defenses against herbivory. Greater herbivore abundance would be expected to attract more para.sites and predators. Finally, unfa- vorable weather has greater demographic im- pact on insects in subalpine meadows than on those in subalpine forests (Ehrlich et al. 1972). Tree Canopy Layer.— Abundance in as- pen canopies declined dramatically through- out the 1977 sampling season (Fig. 2), but still remained at a level higher than that of the aspen understory (Fig. lb). With canopy leaf biomass/m- being about 4.4 times that of understory aboveground biomass/m^ (Appen- dix 2), the number of insects per unit of her- baceous biomass began the season at a higher value in the canopy, with the values in the two strata converging as the season progressed. The inclusion of leaf-miners in the aspen canopy analysis was responsible for the great- er abundance in that stratum than in the un- derstory, where insects within plants were not sampled. However, inspection suggested that leaf-miners were a meager component of the understory community, so the differences were not sampling artifacts. Abundance in conifer canopies showed the opposite trend, rising substantially during the sampling period (Fig. 3). At all times the number of individuals/m- here was notice- ably less than in aspen. Abundance in coni- fers was well below that in aspen over a wide range of foliar biomass (Table 7). By the sec- ond week of August, both conifer strata had similar abundance levels (Figs. Ic, 3). Be cause tree leaf biomass/m- was some 119 times that of conifer understory herb bio- mass/ma (Appendix 2), insect concentration on herbaceous tissue was about two orders of magnitude greater in the understory than in the canopy above. The biological basis for the low numbers of insects in conifer canopies is not known. Our data are consistent with the observation by 48 Great Basin Naturalist Vol. 45, No. 1 50 :2 45 00 40- < Q > 35- 30 -1 A CANOPY/CONirER/1/1977 D C_A_NqP>/CONIFER/2/1977 /4 ^/'d / °.''' /-''' -''7 ' 1 1 1 1 1 1 £ 20 CD 3 15 SAMPLE WEEK Fig. .'3. Mean number of individuals/in- in conifer canopies as functions of sample week, 1977 (second or- der least-squares regressions). Plotted points are weekly means estimated from almnciance vs. biomass regressions. 700 600- 500- 400 300 200 A MEADOW/1977 D AS_P_EN/J977__ o C9NIFER/1977 SAMPLE WEEK Fig. 4. Mean numl)er of individuals/m- i of the .sere as functions of sample week, 197 der least-squares regre.ssions). Plotted point sums of canopy and understory means. n each stage 7 (second or- s are weeklv Moran and South wood (1982) that narrow- leaved angiosperm trees harbored many few- er insects than broad-leaved species did. Avian predation patterns seem unlikely to in- fluence the seasonal trends in canopy insect abundance. Though insect abimdance rose in conifers and fell in aspens during 1977, avian bioma.ss density and consumption rate was greater in the conifers (Smith and MacMahon 1981). Greater avian foraging pressure could contribute to the lower insect abundance in conifers as compared with aspen, but we are unable to estimate the importance of this factor. Total Abundance per Stage.— Seasonal trends of total abundance per stage of succes- sion are presented in Figvue 4. In 1977 den- sities were highest in the aspen. Likewise, T.MUJ-: 7. Comparisons of the individuals-foliar biom; to those of conifer canopy. Statistical significance of tlie of least-squares linear regressions was deteMniued w illi a to chance alone. Werner (1983) found greater insect abun- dance in an Alaskan aspen stand than in a white sprtice stand. His nimibers/m- consid- erably exceed ours. If our abundances are ex- pressed per unit herbaceous biomass (under- story + tree leaves, Appendix 2), then levels in meadow and aspen were similar to each other and about 20 times that in conifer. Species Richness Herbaceous Layer.— The effect of cage size on estimates of number of species/m- is presented in Table 6. As was the case with abtmdance, sampling with the large cage yielded estimates in both stands that were greater than those from the small cage. ass anc spccics-f, )liar bionu ss relatio iships 0 aspen canopx • differ Iwo-t, ■ncc betwc ilcd t-tcst. en aspen P = pro!) uid couift liulitN Ih -r ill iiit It tlic in •rcept or slope cans difier tine lii(li\ iduals vs. g. foliar biom.i^s Mean intercept P Mean slope species vs. g. foliar biomas.s Aspen 24.12 Conifer 3.36 Aspen 1.14 Conifer 0.()096 .001 Mean intercept Aspen Conifer 5.86 4.71 Mean slope .05 Aspen Conifer 0.0025 .001 January 1985 ScHiMPF, MacMahon: Insect Communities 49 ^ A DVAC/ASPEN/l/1977 D DVAC/ASPE:N/2/1977 A O pVAC/ASPEN/3/1977 Q A DVAC/CONIFER/l/1977 a DVAC/CqNIFE_R/2y(^l977 A DVAC/ASPEN/4/1978 ° PY*p/*SPEN/5/1978 A DVAC/CONIFER/1978 T T T T 0 2 4 6 8 10 12 14 16 18 20 0 2 4 6 8 10 12 14 16 18 20 0 2 4 6 8 10 12 14 16 18 20 SAMPLE WEEK SAMPLE WEEK SAMPLE WEEK Fig. 5. Mean mmiher of species/ni- in meadows, aspen understories. and eoniter underst(jries as tiniclions ol sample week (second order least-sqnares regressions). Plotted points are weekly means for 1977 (a-e) and 1978 (d-f). though not statistically different. The effect of cage size was much less in the case of spe- cies richness than in the case of abundance. There seems to be no reason to qualify ovir comparisons among stages in this regard. However, it must be recognized that the spe- cies richness values we obtained may not be those tliat would have been obtained by sam- pling a full 1.0 m- plot, depending on vegeta- tion patchiness. Species/ m- in the herbaceous layer peaked between the second and fourth weeks of July 1977 in most of the stands (Fig. 5a, 5b, 5c), declining strongly thereafter in all of them. In 1978, seasonal change was more pro- nomiced, with peaks occurring during August (Fig. 5d, 5e, 5f). Peak richness was about twice as great in 1978 meadow and conifer stands as in 1977, but only slightly greater in aspen. This muted interannual difference in aspen is confounded by the fact that, unlike the other two stages, different aspen stands were sampled in different years. Individual aspen groves were too small to sample for consecutive years without affecting 1978 re- sults. The change in aspen locations in 1978 seems to have had less effect on abundance than on species richness (Fig. lb, le). Comparison of regression lines shows that in 1977 species/m^ rose shghtly from mead- ow to aspen, then declined markedly into the conifer. In 1978 there was a conspicuous de- cline from meadow to aspen, with little fur- ther decline into conifer. Averages of peaks for the two years show meadow and aspen understory with similar species richness. Cli- max conifer understory species richness was two-thirds that of these earlier stages. As was the case with abundance, species richness increased with succession if ex- pressed per unit plant biomass instead of per m2. The ratio of peak insect species/m- to grams peak herbaceous aboveground stand- ing crop/m^ (Appendix 2) was about 0.3 for meadow in both years, 0.8 and 0.6 under as- pen, and 1.0 and 1.9 under conifers. Note that the plant and insect peaks were not nec- essarily simultaneous, a phenomenon also noted by South wood et al. (1979). This great concentration of insect species on herbage with successional progression is not explained by differences in plant species richness (see 50 Great Basin Naturalist Vol. 45, No. 1 70 60- CN 50 A CANOPY/ASPEN/1/1977 ° CANqP\/ASPEN/2^1_977 ° CAN?.PY/ASPEN/3/1977 a: 40- LJ CD <. 2 ''^. Z 30- ^■■■.. " \ A \-..0 < \. ^n; ^ 20^ ^^ 0 2 16 18 20 SAMPLE WEEK Fig. 6. Mean niunbor ot species/ m- in aspen canopies as functions of sample week, 1977 (second order least- squares regressions). Plotted points are weekly means es- timated from species vs. biomass regressions. also Usher 1979). The 1977 and 1978 ratios of in.sect .species • plant species- ' • m- were 1.9 and 3.3 in meadow, 2.2 and 2.3 under aspen, and 2.1 and 4.3 under conifer. Although the close agreement of our 1977 ratios supports the general positive correlation between plant and insect .species richnesses (Murdoch et al. 1972, Southwood et al. 1979), the dif- ferent 1978 ratios suggest that other factors contribute to in.sect .species richness. More in- sect .species were present per plant species in 1978, and the ratio differed widely among stages. Increased in.sect species richness irr 1978 cannot be attributed strictly to greater amoimts of phytomass, since in the ca.se of the conifer understory phytomass was nearly the .same in both years (Appendix 2). Plant growth forms and life cycle categories differ in the richness of their associated insect faunas (Lawton 1982, Lawton and Schroder 1977, Strong 1979, Strong and Levin 1979), but the stages we are comparing in this stra- tum differ little in form or life cycle length (Schimpf et al. 1980). The greater concentra- tion of insect species on the conifer under- .story may be due to a more favorable phys- ical environment or a more readily coloni/.able set of host plant species. 70 n A CANOPY/CONIFER/1/1977 ° CANQP^CONlFER/2/1_97_7 60- 50- 40- 30- 20- ,-Q''' 10- ^^^'^^^-'^' ^^-^,0--"' 0- -K" 0 2 4 6 8 10 12 SAMPLE WEEK Fig. 7. Mean number of species/m- in conifer cano- pies as fimctions of .sample week, 1977 (second order least-squares regressions). Plotted points are weekly means estimated from species vs. biomass regre.ssions. Tree Canopy Layer.— Species richness in the aspen canopy was fairly steady through the 1977 sampling period (Fig. 6) despite the precipitous drop in abundance at the same time (Fig. 2). The number of species/m^ was roughly comparable to that in the understory (Fig. lb). Species richness in the conifer can- opy (Fig. 7) was about two-thirds as great as that of the aspen canopy and similar to spe- cies richness in conifer understory (Fig. 5c). With the lack of stratification in .species richness and much greater plant species rich- ness in the understories vs. the canopies (Ap- pendix 2), insect species • plant .species ' • m- was greater in the canopies than in their un- derstories. It would .seem unlikely that all the in.sect species on 1 m- would occur on each plant species on that plot, and the existence of a .smaller number of individuals of many in.sect species than the number of plant spe- cies invalidates this po.ssibility. Therefore, within 1 m- the average herbaceous species harbored fewer insect species than did each tree species. The generality that there is a greater number of insect species per tree spe- cies than per herb species for a given amount of geographic range (Strong and Levin 1979) can thus be extended to a smaller spatial .scale for one point in time. January 1985 ScHiMPF, MacMahon: Insect Communities 51 The greater number of insect species/m2 on tree species vs. herb species may be partly due to the greater standing crop of tree bio- mass. This can be factored out by dividing in- sect species • plant species- ' • m- by the ter- minal standing crop of tree leaves or herb biomass (Appendix 2). Conifer canopy insect species richness is divided by one, not two plant species, since most 1 m- plots are super- posed by only one tree species. These ratios are crvide approximations, for they assume no faimal overlap among herb species in the same 1 m- plot. Midseason insect species • plant species' • m- • lOg-' was 0.4-0.5 in as- pen understory, 1.1-1.3 in conifer understory, 1.1 in aspen canopy, and 0.06-0.07 in conifer canopy. On this basis aspen seems no better for the "packing" of insect species than the average conifer understory species, and per- haps worse tlian understory species of both stages if faunal overlap among plant species is taken into account. Spruce or fir were con- spicuously depauperate in comparison to any of tlie angiosperms. The difference in species richness between aspen and conifer would be even greater on a larger spatial scale (Table 7). This deserves wider investigation, for others (Futuyma and Gould 1979, Neuvonen and Niemala 1981) have noted low numbers of species of Lepidoptera or Tenthedrinoidea on gymnosperms as compared to angiosperm trees in the same area. Hendrix (1980) pro- posed that ferns are less hospitable than an- giosperms to insect herbivore species; perhaps gymnosperms are also entomologically de- pauperate. Evergreenness per se may not be important (Faeth et al. 1981). Moran and South wood (1982) suggested that frequent Ribbing of leaves against one another was re- sponsible for depauperate insect assemblages on narrow-leaved species of Solix; such an ex- planation cannot be invoked for the stiff- leaved fir and sprvice. Furthermore, aspen leaves strike each other with great frequency, yet that species carried far more insects than the conifers, even if leaf-miners are ignored. Total Species Richness per Stage.— Seasonal trends of 1977 combined-strata spe- cies/m- are shown in Figure 8. Species rich- ness clearly rose, then fell during succession, with the aspen level about twice that of the preceding or following stage. Whether one considers the herbaceous layer, tree canopy. 70 60 50 A MEADOW/1977 D ASPEN/1977 _ O CONIFER/1977 40- 30 < X ;< 20 0 2 4 6 8 10 SAMPLE WEEK Fig. 8. Mean niimbor of species/ m^ in each stage of the sere as functions of sample week, 1977 (second order least-squares regressions). Plotted points are weekly sums of canopy and understory means. or combined strata, species/m- does not in- crease throughout succession. This disagrees with the tabulated trend in Odum (1969) but agrees with the preclimax peak in species richness predicted in the text of that paper. Evenness Herbaceous Layer.— During 1977 even- ness peaked in midseason in meadow stands and one aspen stand, but showed mid.season minima in the other forested stands (Fig. 9a, 9b, 9c). No successional trends in midseason J' are evident. Early or late in the sampling period there was a successional increase in evenness. In 1978 evenness was greatest in as- pen and conifer (Fig. 9d, 9e, 9f). Values in all stages were generally higher in 1977 than in 1978. This difference may be an artifact of the greater insect abundance in 1978, with consequent more likely inclusion of rare spe- cies in samples (Hurlbert 1971), and thus is perhaps not biologically meaningful. The overall pattern is one of increasing insect community evenness during succession, in agreement with the prediction of Odum (1969). This conclusion must be tempered by recognition of the mathematical inadequacies 52 Great Basin Naturalist Vol. 45, No. 1 0.9 0.8 0.7 0.6- 0.5- 0.4- 0.3- 0.2- 0.1- 0- 0.9 0.8 0.7 0.6- 0.5 0.4- 0.3 0.2- 0.1 0 A DVAC/MEADOW/1/1977 B a a P VAC/ME ADO W/2/1 977 A DVAC/ASPEN/1/1977 D DVA_C/AS_PEN/2/19_77 b ° PV.*C/ASPEN/3/.1.977 A DVAC/MEADOW/1/1978 d D pVA_C/MEAD_0_W/2/1978 :v&» D a\ ~-~-^ _,^ A A A A A Ol D VAC/CON IFER/1/1977 pvA_c/cq^ 1 1 1 _IFER/2/197_7 ■~1 1 1 1 1 -| A A ; □ D A . D 1 A DVAC/ASPEN/4/1978 ° 9VA?/*SPi:N/5/19_78 -| r 1 1 1 1 r — I 1 A DVAC/CONIFER/1978 0 2 4 6 8 10 12 14 16 18 20 0 2 4 6 8 10 12 14 16 18 20 0 2 4 6 8 10 12 14 16 18 20 SAMPLE WEEK SAMPLE WEEK SAMPLE WEEK Fig. 9. Mean evenness (J') of sample aggregates totaling LO ni- in nieadov\s. aspen undei stories, and conifer nn- derstories as finictions of .sample week (second order least-squares regressions). Plotted points are w eeklv means for 1977 (a-c) and 1978 (d-f). of evenness mea.sures, inchidint^ J' (Routledge 1983). Table 8. Variance and coefficient of variati midseason intrastand sample similarity matrices formed to arcsin. Coefficient Stand Variance of variation 1977 week 8 Meadow 1 41.02 ()..35 Meadow 2 30.96 0.27 Aspen 1 92.92 0.73 Aspen 2 74.79 0.90 Aspen 3 93.24 0.52 Conifer 1 3.5.08 0.25 (:()niter2 iX).07 0.00 1978 week fi Meadow 1 4,5.15 0.27 Meadow 2 2H..32 0.2.3 Aspen 4 51.89 0.41 Aspen .5 4().ttl 0.26 Conifer 66..36 0.55 Tree Canopy Layer.— In Figure 10, we present dominance-diversity plots. We inter- pret the trajectories of these to reflect the differences in evenness between conifer and aspen stages. Aspen exhibits greater relative abundance by its most common species than does conifer, with lesser relative abundance of the less common species in aspen. The rel- ative abundance of species in the conifer are less disparate, so we conclude that species evenness in conifer canopies is higher than that in aspen canopies. Note that this ap- proach does not yield a numerical evenness value and is an integration of a full season's samples. Total Evenness per Stage.— Although the analyses of the understories cannot be nu- merically combined with those of their re- spective canopies, the similar trend in both strata allows us to conclude that evenness in- January 1985 ScHiMPF, MacMahon: Insect Communities 53 0 10 20 30 40 50 60 SEQUENCE OF SPECIES Fig. 10. Dominance-diver.sity plots for a.spen and co- nifer canopies, 1977. Species are plotted by the percent- age of the total season's captiue they compose (log scale) in decreasing order of relative alnnidance. creases during succession, concordant with the prediction of Odvmi (1969). Sample Similarity— Herbaceous Layer suits from sampling a spatially more homo- geneous taxocene. This is because pairs of samples from the same kind of patch will have high .similarity and those from different kinds of patches low similarity. Both vari- ances and coefficients of variation indicate that the meadow was the most consistently spatially homogeneous stage in midseason of both years (Table 8). Although the greater variance under coni- fers may be an artifact of the lower insect abundance there (Wolda 1981), it is con- sistent with the greater habitat heterogeneity of the forest floor. As others have docu- mented (Anderson et al. 1969, Knight et al. 1977, Young and Smith 1979), the conifer un- derstory is a mosaic of patches differing greatly in illumination, temperature, date of snowmelt, throughfall precipitation, and plant species composition. Even closely re- lated plant species may support rather differ- ent insect assemblages (Seifert 1981). We have evidence that the vegetation of our meadows is more homogeneous, in that the mean number of plant species/ m- is high (Appendix 2), but the total number of plant species is low (Reese 1981). It should be rec- ognized that horizontal pattern may exist at several spatial .scales, and that our results have meaning only with reference to the size Samples taken with the large cage were slightly more similar than those taken with the small cage (Table 6). Average similarity and spacing of our samples, among samples from the herbaceous layer was low in all stages in 1977 (Fig. 11a, lib, lie); no consistent seasonal pattern was ob- served, and no obvious successional trend was detected. Similarity levels in 1978 were much the same as those in 1977, and seasonal and intrastage spatial variation obscured any successional trends (Fig. lid, lie, llf). If particular insect commimities occur in patch- es, variance in similarity will be higher than in cases where the same mean similarity re- Total Heterogeneity per Stage Despite lacking information on horizontal heterogeneity in the canopies, we agree with Odum's (1969) hypothesis that pattern diver- sity increases with succession. We do so on the basis of the weak successional increase in horizontal patchiness in the herbaceous lay- er's insect communities and the obvious Table 9. Adult body length of all species and all individuals caught in a stratum of a successional stage during one sampling season. X = mean, SD = standard deviation. Stage/stratum Species un\ veighted Species we ighted b> / alnindance 1977 1978 1977 . 1978 X±SD(mm) X±SD(mm) X±SD(mm ) X±SD(mm) Meadow 4.4±4.1 4.1 ±3.9 3.0 ±2.4 2.2±1.6 .•\spen understorv 3.2±2..3 3.2 ±3.3 3.5±1.6 3.3 ±1.9 Aspen canopy 3.5 ±3.3 3.6 ±2.0 4.6±2.1 5.0 ± 1.6 Conifer understorv 2.6±1.7 3.2 ±3.3 2.0 ±1.0 2.0±1.9 Conifer canopv 2.4±1..5 - 1.4 ±0.7 - 54 Great Basin Naturalist Vol. 45, No. 1 A DVAC/C0NIFE:R/1/1977 n DVAC/CqNIFER/2^)977 ii- d A DVAC/MEADOW/1/1978 30- D DVA_C/MEAD_0_W/2/1978 25- 20- J^^-^^^ 15- ^?^°=--fl?r~-\^^ 10- '"-•-, 5- 0- e A DVAC/ASPEN/V1978 D DVA_C/ASPEN/5/1978 - D A -^ 'a -I — 1 — 1 — 1 — r^A — I — I — 1 A DVAC/CONIFER/1978 0 2 4 6 8 10 12 U 16 SAMPLE WEEK 20 0 2 4 6 SAMPLE WEEK 6 8 10 12 14 16 18 20 SAMPLE WEEK Fig. II. Mean absolute percent similarity of species composition of samples in meadows, aspen understories, and conifer understories as fimctions of sample week (second order least-squares regressions). Plotted points are weekly means for 1977 (a-c) and 1978 (d-f). strong vertical stratification of the forested stages. Adult Body Length Mean adult body length declined during succession if species are not weighted by their abundance (Table 9). This accords with the principle that larger organisms, ceteris paribus, will be less influenced by the more. Table 10. Distribution among metamorphosis categories of all spei successional stage during one sampling season. pronounced fluctuations of the physical envi- ronment of the meadow, due to their lower surface:mass ratios. Ceteris paribus cannot be invoked in this case, since longer insects tend to be proportionately thinner (Rogers et al. 1976, Schoener 1979). Because the relation- ship between insect length and surface area is unknown, we do not know whether sur- face:inass declines with increasing length. Species in the aspen canopy were somewhat ; and all individi ilht i)f a Stage/stratum /year Ameta- bolous Hemimeta- bolous Paurometa- bolous Holometa- bolous Meadow 1977 1 Aspen understorv 1977 2 .\spen canopy Conifer understorv 1977 1977 0 2 Conifer canopy Meadow 1977 1978 1 ■7 Aspen understorv 1978 1 Aspen canopy Conifer understorv 1978 1978 2 3 27 72 24 74 27 72 31 6H 26 72 26 73 23 76 29 69 25 71 January 1985 ScHiMPF, MacMahon: Insect Communities 55 larger than those in aspen iinderstory, where- as the conifer strata showed the reverse rela- tionship. When the species from all stages, strata, and years were averaged (un- weighted), a mean of 3.4 mm resulted, re- markably close to the 3.5 mm reported for samples from Massachusetts by Schoener and Janzen (1968). If species are weighted by abundance, adult body length increased from meadow to aspen, then dropped greatly into the conifer (Table 9). The largest insects were found in the aspen canopy, and the smallest ones in the conifer canopy. The aspen canopy had a large plurality of insects of above-average length (leaf miners), evidenced by the weighted mean's far exceeding of the un- weighted mean. The other stages/ strata had weighted means similar to or smaller than the unweighted means, suggesting numerical pre- dominance of species of medium or small size. Wliether weighted or unweighted, or- ganism size does not increase during succes- sion, thus contradicting Odum's (1969) pre- diction. The insects of the conifer climax are the smallest, regardless of stratum. Werner (1983) found heavier insects in white spRice stands than in aspen stands in Alaska. His data are not directly comparable to ours because they represent actual, not necessarily adult, sizes and because he did not sample so as to extract immature leaf miners. Leaf Cycle Complexity In Table 10 insect metamorphosis cate- gories are ordered from least (ametabolous) to greatest (holometabolous) complexity of the life cycle. The faunas of all stages, strata, and years are similar when analyzed in this way. When species are weighted by abun- dance, life cycle complexity increased from meadow to aspen, then dropped in conifer. In both aspen and conifer, the tree canopy had a greater proportion of insects with com- plex life cycles than did the understory. Life cycle complexity does not increase into the climax as Odum (1969) predicts, but we can partially support his hypothesis in the sense that trees, the successionally later growth form, harbor insects with more complex life cycles than those on herbs. Odum (1969) ac- tually predicted that during succession life cycles would get longer as well as more com- plex. Although we did not address life cycle length. Brown and Southwood (1983) found that this increased during succession. Stability In addition to these planned comparisons, the occurrence of a drought in 1977 permit- ted us to address another of Odum's (1969) postulates, that concerning stability (resis- tance to external perturbations). Inter- estingly, species richness did not differ much between drought and postdrought (1978) years in the herbaceous layer (Fig. 5a-c, vs. 5d-f). Abundance changed most in the mead- ows and least in the conifer understory (Fig. la. Id vs. Ic, If). Effects on evenness were more complex, in that meadows not only changed in time of peak evenness, but also reversed the form of seasonal evenness rela- tionships (Fig. 9a vs. 9d); in aspen understory Table 10 continued. Percv ent of i ill individuals Ameta- Heminieta- Paurometa- Holometa- bolous bolous bolous bolous 9 0 67 24 2 0 66 32 0 0 49 51 1 0 74 25 2 0 51 46 9 0 79 12 1 0 86 14 0 0 41 59 4 0 73 22 56 Great Basin Naturalist Vol. 45, No. 1 the form changes are mixed (Fig. 9b vs. 9e), but conifer understory changed httle (9c vs. 9f). The nature of the changes observed is the same as those noted by Smith and MacMahon (1981) for the avifauna, and Andersen et al. (1980) for small mammals, on larger plots of the same sere. We concur with Odum (1969) that stability increases with succession, at least in the herbaceous layer. General Discussion We have used our analysis of successional trends in the insect component of commu- nities to address "the strategy of ecosystem development" discussed by Odum (1969). The results of this analysis are mixed (Table 11). Similar mixed results were obtained when other taxocenes were analyzed for our sere (Andersen et al. 1980, MacMahon 1980, Smith and MacMahon 1981) as well as for a different one (Witkowski 1979). It is impor- tant to remember, when assessing the success of our attempts to test Odum's ideas, that Odum's original postulates apparently de scribed changes for the total ecosystem, not subunits such as the insect assemblages we deal with. There appears to us to be no easy intellectual nor mathematical solution to the problem of equating organisms varying from microbes to trees for purposes of a synthesis of general successional trends. With our piecemeal approach, if trends are the same for every ecosystem subunit studied, the trend should be true for the ecosystem as a whole. In the cases where different subunits of the same sere show different successional Tahli 1 1. Comparison of six of Odmirs (1969) 24 expected trends in the development of ecosystems to the succes- sional trends in the insect component of a snbalpine sere reported herein. Numbers and descriptions of ecosytem at- tributes and developmental stages are taken from Odum (1969). Ecosystem attributes Developmental stages Mature stages Accept/Reject Rationale Community structure 8. Species diversity- Low High Reject Highest in aspen variety component [Fig. 81 9. Species diversity- Low Higii Accept J' highest in conifer e' of stand F his Tables 4 and 6, understory of stand J 1 = S. ahemiflora (all stations) 2 = Bogue Banks 3 = Junciis (all stations) 4 = Distichilis (all stations) 5 = S. patens (all stations) mean of six species his Tables 4 and 6, understories of stands H,H' "long list" 1 = mean Tables 4a.b (primary flat) 2 = mean Tables 4a, b (secondary) 3 = Table 4a (primary-hill) 4 = Table 4c (abandoned pasture) 5 = Table 4c (primary riparian) 6 = Table 4c (primarv-hill) 7 = Table 4e 8 = Table 4f 9 = Table 4d (corn field) 10 = Table 4d (primary I) 11 = Table 4d (primary II) 12 = Table 4g 13 = mean Tallies 4ii,i U.S. -I- C^anada World 1 World 2 Evans and Miudoch (1968) Southwood (1978b) Dalvetal. (1978) Appendlx 2. Estimated means of plant parameters for stages of the subalpine sere. Data adapted from Williams (1977) and Reese (1981). Stage Herbaceous species/m^ Herbaceous 1977 terminal stanc g/m2 ing crop, 1978 Tree leaves, g/m2 Meadow Aspen Conifer 13.5 14.5 7.2 74 39 15 171 58 16 0 173 1739 NUTRIENTS IN CAREX EXSERTA SOD AND GRAVEL IN SEQUOIA NATIONAL PARK, CALIFORNIA Raymond D. Ratliff Abstr^vct.— Nutrients in soil covered by Carcx exserta sod and in adjacent unvegetated gravel areas were com- pared at Siberian Outpost, Sequoia National Park, California. The comparisons were part of a study to learn if Carex cxseiia meadov\' can be reestablished and if herbaceous cover on gravel areas can be increased. Grazing capacity and aesthetic appeal of denuded areas would be improved by better vegetative cover. The sod had higher concentrations of calcium, copper, iron, magnesium, manganese, nitrogen, potassium, and zinc than did the gravel areas. And it had a higher soil pH and percent organic matter. Sod and gravel did not differ in concentrations of phosphorus and sul- b\r. The differences were as might be expected between climax and badly deteriorated (or early serai) situations, and the results suggest that fertilization may be a useful treatment. Carex exserta (short-hair sedge) meadows are found throughout the Sierra Nevada of Cahfomia. Altitudinally, they extend from the subalpine into the alpine zone (Jackson and Bhss 1982). At Siberian Outpost— an un- glaciated subalpine valley in Sequoia Nation- al Park— stands of Carex exserta (Fig. 1) vary in form from sod steps (Klikoff 1965) to nearly continuous sod on gentle slopes. In and around Siberian Outpost are found ex- pansive areas of coarse-grained granitic gravel. Small plants found in the gravel pro- vide little cover. The sod can withstand considerable use, but continued overuse or trampling will break and eventually destroy it. Sheep are known to have grazed in Sequoia National Park during the late 1800s and early 1900s (Vankat 1970). In parts of Siberian Outpost, pedestaled remnants attest to overgrazing as a cause for destruction of Carex exserta sod. More than 15 cm of sod and top soil have been lost in some places, and little recovery is evident. However, the sod appears to be establishing or reestablishing in other places. Grazing may not be the only reason for the areas of gravel. Sand and gravel may have originated from debris-laden outwash below glacial moraines or may have been deposited in Pleistocene lakes (Benedict and Major 1982). Retreat of glaciers with warmer, dryer summers may have changed the vegetation the areas were capable of supporting. Rapid percolation to deep layers may make precipi- tation largely unavailable for on-site plant growth. High winds may keep some areas clear of snow. And gravels and sands from weathering of the granite rocks may be de- posited over vegetation in some areas about as soon as it becomes established. Thus, some of the gravel areas in and similar to those of Siberian Outpost may be com- pletely natural— representing early serai stages. Nevertheless, where the Carex exserta sod has obviously been destroyed, an eroded stretch of gravel and sand remains. That and the presence of large tree stumps in gravel and up to 290 m from the present forest bor- der suggest that some areas of gravel, such as found in Siberian Outpost, may once have supported considerably more vegetation than now. If the present gravel areas were largely a Carex exserta meadow at one time, can that vegetation be reestablished, or can cover of vegetation now occupying the gravel areas in Siberian Outpost and elsewhere be increased? Observed differences between sod and gravel areas should reveal how loss of Carex exserta sod alters the nutrient status and should sug- gest nutrients that might be added to gravel areas. On the other hand, if the gravel areas represent natural serai stages, knowledge of how they differ in nutrients will increase un- derstanding of serai and climax communities. 'Pacific Southwest Forest and Range Experiment Station. Foresi U.S. Department of .■\i;riculture. 2081 East Si. Fresno. California 61 62 Great Basin Naturalist Vol. 45, No. 1 -#^.a;.S^ l-"iU. I. \i Caliloiin.i ipcr) aiul piistiiif i^iiirx cxscrta st... l..^, , ,M.,,, X.itu.iial Park. January 1985 Ratliff: Carex Nutrients 63 Information on the soil texture, pH, and organic matter of Carex exserta sod is avail- able (Ratliff 1982), but data on its nutrient composition and that of gravel areas are scant. Such information is needed in the se- lection of programs to revegetate back- country areas that have been overused, and to rehabilitate trails and campsites. This paper compares soil organic matter, pH, and nutrients in Carex exserta sod with those in gravel in Siberian Outpost, Sequoia National Park. Methods Siberian Outpost at 36°28'N, 118°17'W (U.S. Geological Survey 1956) lies 11.3 km south of Mount Whitney at 3293 m elevation within the "Boreal Plateau erosion surface" (Matthes 1950, 1962), between Rock Creek on the north and Big Whitney Meadow on the south. Siberian Pass Creek drains the area from east to west. Calamagrostis breweri meadows are found at the bottom areas of Siberian Outpost. The otherwise gravelly slopes have a few species of forbs such as Oreonana dementis, Cahjp- tridium wnbellatum, Eriogonum incaniim, and Lupinus culbertsonii. They have been classified as an Eriogonum-Oreonana de- mentis association (Benedict 1983). Stipa oc- cidentalis and Sitanion Injstrix are occasion- ally abimdant. A grid with a ground distance of 45 m be- tween intersections (grid points) was super- imposed on aerial photographs of Siberian Outpost. Fifty grid points without Carex ex- serta sod (henceforth referred to as gravel), and 30 grid points with Carex exserta sod (henceforth referred to as sod) were ran- domly selected. Fewer grid points were allo- cated for the sod because it occupied less area than the gravel. Three soil cores (17 cm long and 4.7 cm in diameter) were extracted at each grid point and combined to make one sample. Material larger than 2 mm was sepa- rated and discarded. The fraction less than 2 mm was kept for laboratory analysis. In addi- tion, intact cores were collected at 15 and at 6 arbitrary grid points in the gravel and in the sod, respectively. Acidity of each soil sample was determined with a satm-ated paste and an electronic pH meter. Percent soil organic matter was esti- mated by gravinietry and dry combustion. Bulk density and gravel content were esti- mated from the intact cores by gravimetry. Available facilities and funds limited the nutrient analyses that could be made to those common for crop agriculture. Total nitrogen was determined by the ammonia electrode modification of Kjeldahl method (Page et al. 1982). Available phosphorus was determined by the sodium bicarbonate method (Page et al. 1982). Amounts of calcium, copper, iron, magnesium, manganese, potassium, sulfur, and zinc were determined at a commercial laboratory by standard procedures (Reise- nauer 1976). Parametric tests were not appropriate for the data because they either did not conform to the normal distribution, or the variances were not equal, or both. Therefore, the non- parametric rank-sum test with the normal ap- proximation was used (Steel and Torrie 1960). The hypothesis that the sets of values from the sod and gravel belong to a common population was rejected when chance proba- bility of the rank-sum for the sod samples was 5% or less. Presence of a common population was rejected for all the nutrients except phos- phorus and sulfur (Table 1). Because the samples were obtained at arbi- trary grid points, bulk density and gravel data were not random sets. Those data, there- fore, could not be statistically analyzed but are presented for reader information. Results and Discussion Soil from the sod samples contained more nutrients than did soil from the gravel. The natural openings in the sod were not sam- pled, and some of the nutrient differences be- tween the sod and gravel may be due to con- centration of nutrients in the sod at the expense of the openings within it. Therefore, on an area basis, the differences in nutrients may not be as large as indicatd by the mean values (Table 1). Nevertheless, the gravel samples were relatively low in nutrients for plant growth. Nitrogen ordinarily ranges from 0.02% to 0.5% of soil. In prairie soils, 0.1% to 0.3% ni- trogen is usual (Allison 1957). Nitrogen con- tents of alpine mine spoils and topsoil were 64 Great Basin Naturalist Vol. 45, No. 1 0.06% and 0.13%, respectively, on the Bear- tooth Plateau (Brown and Johnston 1976). Soil of the sod at Siberian Outpost, therefore, was relatively high in total nitrogen— about 0.2%. The gravel, on the other hand, had only about 0.04% nitrogen. With 27 ppm and 24 ppni in the sod and gravel, respectively, available phosphorus (Table 1) at Siberian Outpost should be ade- quate for herbaceous plant growth. For pas- ture and range, no response to added phos- phorus is likely with more than 10 ppm available phosphoms in the soil (Reisenauer 1976). High soil acidity, however, generally lowers phosphorus availability. Ammonium acetate extractable potassium was 0.006% in the sod and 0.001% in the gravel (Table 1). Nitric acid extractable po- tassium in the sod (0.02%) was twice that in the gravel (0.01%). By either extraction pro- cess, exchangeable potassium content of the sod bordered on a deficiency (Reisenauer 1976). Potassium content of the gravel ap- peared clearly deficient and was lower than that reported for mine spoils (Brown and Johnston 1976). Potassium is more readily released when concentrations of calcium and/ or magnesium are high. The concentration of exchangeable calcium in mine spoils (Brown and Johnston 1976) was higher than that in either the gravel (0.7 meq/100 g) or the sod (1.2 meq/100 g ) at Siberian Outpost. Also, although higher than in the mine spoils, mag- nesium concentrations in the sod (0.5 meq/ 100 g) and in the gravel (0.2 meq/100 g) at Siberian Outpost were lower than in topsoil on the Beartooth Plateau, Montana (Brown and Johnston 1976). Soils in Siberian Outpost are low in sulfur content. The average concentration of sulfur (1.25 ppm or 0.0001%) in the sod and gravel is equivalent to only 1.4 kg'ha-^ to a 10-cm depth. That concentration is 100 times less than the minimum expected. Amounts of manganese, iron, copper, and zinc in the gravel, though lower than in the sod, appeared adequate for plant growth. However, their amounts were near critical levels for agricultural crops— especially in the gravel. Critical levels suggested for these nu- trients (Reisenauer 1976) were 1.0 ppm (man- ganese), 5.0 ppm (iron), 0.2 ppm (copper), and 0.5 ppm (zinc). Soil organic matter and soil pH (Table 1) were higher in the sod (6.3% and pH 4.2) than in the gravel (1.7% and pH 4.0)-P < 0.001. Carex exserta sites studied earlier (Rat- liff 1982) had higher values (7.3% organic matter and pH 5.1) for the top 20 cm of soil. Soil bulk density averaged 1.1 g*cm-^ in the sod and 1.7 g*cm'^ in the gravel. With the difference in organic matter contents, lower bulk density in the sod would be expected. Gravels composed 37% of the soil mass in the gravel and 24% of the soil mass in the sod. Tahli-; 1. Soil ()i<4anii tioiial Park, California. latter, pH, and nuti Cciri'x cxscria sod and gravel in ,Siberian Outpost, Secjuoia Xa- Sod Gravel Variable Low Average Low Average High Organic' pll Nulricnt \ V Kl> Ca Mg S Mn Fc Cu Zn 4.2 4.0 1258.0 6.7 31.0 128.0 160.3 48.6 0.5 5.4 53.0 0.4 2.9 6.3 9.8 4.2 4.5 2040.07 2951.0 27.30 56.8 55.77 95.0 191.87 257.0 241.60 440.9 65.66 97.3 1..37 3.0 12.78 26.2 102.79 133.0 0.74 1.4 7.28 17.6 1.2 1.7« 3.6 4.0' 201.0 378.52^ 13.0 24.50 8.0 I4.58'> 42.0 94.68^ 40.1 143.4t>' 7.3 25,54' 0.5 1.13 0.5 1.1.5' .5.7 12.42^ 0.1 0.31'' 0.5 1.46» a AveraKe value significantly different from short-hair sedge sod (P < 0.001) by the rank-sum test with normal approximation. " .\innionium acetate cxtractal)le. e Nitric acid exiraclalilc. .3.8 4.5 1223.0 37.0 35.0 179.0 200.4 48.6 2.0 3.7 37.0 1.1 6.6 January 1985 Ratliff: Carex Nutrients 65 Tlie organic matter content of the gravel was only 0.7% lower than in alpine zone top- soil on the Beartooth Plateau (Brown and Johnston 1976) and was greater than that of most Aridisols with a high sand content (Soil Survey Staff 1975). Therefore, the sparse veg- etation in the gravel was unlikely to have produced tlie existing level of organic matter. Denser vegetation at an earlier time could have been responsible for it, however. Loss of the sod by erosion would remove most of the organic matter built up in the soil system and leave the gravel as a pavement. Because or- ganic matter markedly improves the cation- exchange properties of soils, nutrients also would have been removed by erosion. Conclusions This study found substantial differences be- tween soils of Carex exserta sod and gravel in nutrient and organic matter contents and acidity. The differences were as expected be- tween climax and badly deteriorated soils. If the gravel in Siberian Outpost once sup- ported Carex exserta meadow, then loss of sod resulted in removal of 40% of the cal- cium, 58% of the copper, 88% of the iron, 61% of the magnesium, 91% of the manga- nese, 81% of the nitrogen, 74% of the potas- sium, and 80% of the zinc. Losses of those magnitudes can be hypothesized as a con- sequence of the destruction of Carex exserta sod. Alternatively, if the gravel represents a natural serai stage, succession to a Carex ex- serta climax would require nutrient accumu- lation and conservation as the sod develops. Brown and Johnston (1979) suggested fer- tilization with relatively high rates of nitro- gen (up to 168 kg'ha-^) for rehabilitating dis- turbed alpine areas. They also suggested lime applications on alpine disturbances where pH is below 5.5. Given the low pH of Sibe- rian Outpost soils, fertilization with lime may make more phosphorus available. To increase cover of gravel areas or to rehabilitate disturbed Carex exserta sites in Siberian Outpost and similar areas elsewhere (within a present rather than a successional or geological time frame), stimulation by fer- tilization may be a useful treatment. The dif- ferences found between sod and gravel in nu- trient contents suggest including nitrogen. potassium, calcium, magnesium, and sulfur in a fertilizer mix. Acknowledgments I thank the staff of Sequoia and Kings Can- yon national parks, who permitted me to do this study in Siberian Outpost, and Stanley E. Westfall, who collaborated on all phases of the study. Literature Cited Allison, F. E. 1957. Nitrogen and soil feitilits. Pages 85-94 in Soil, USDA Yearbook of Agriculture. U.S. Govt. Print. Off., Washington, D.C. Benedict, N. B. 1983. Plant associations of suhalpine meadows. Sequoia National Park, California. Arc- tic and Alpine Res. 15:383-396. Benedict, N. B., and J. Major. 1982. A physiographic classification of subalpine meadows of the Sierra Nevada, California. Madrono 29:1-12. Brown, R. W., and R. S. Johnston. 1976. Revegetation of an alpine mine disturbance: Beartooth Plateau, Montana. U.S. Forest Service, Intermountain Forest and Range Experiment Station, Res. Note lNT-2()6. 8 pp. ' Brown, R. \V., and R. S. Johnstcjn. 1979. Revegetation of disturbed alpine rangelands. Pages 76-94 in D. A. Johnson, ed.. Special management needs of al- pine eco.systems. Society for Range Management, Denver, Colorado. Jackson, L. E., and L. C. Bliss. 1982. Distribution of ephemeral herbaceous plants near treeline in the Sierra Nevada, California, USA. Arctic and Al- pine Res. 14:3.3-43. Klikoff, L. C. 1965. Microenvironmental intliieuce on vegetational pattern near timberline in the cen- tral Sierra Nevada. Ecol. Monogr. 35:187-211. Matthes, F. E. 1950. Sequoia National Park: a geologi- cal album. Univ. of California Press, Berkeley. 136 pp. 1962. Francois Mattlies and the marks of time: Yosemite and the High Sierras. Sierra Club, San Francisco, California. 189 pp. Page, A. L., R. H. Miller, and D. R. Keeney, eds. 1982. .\gronomv 9. Methods of soil analysis. Part 2. Chemical and microbiological properties. 2d. ed. American Society of Agronomy, Inc., and Soil Science Society of America, Inc., Madison, Wis- consin. 1159 pp. R.\tliff, R. D. 1982. A meadow site classification for the Sierra Nevada, California. U.S. Forest Service, Pacific Southwest Forest and Range E.xperiment Station, General Technical Rept. PSW-6(). 16 pp. Reisenauer, H. M., ed. 1976. Soil and plant-tissue test- ing in C:alifornia. Univ. of California Bull. 1879. 54 pp. Soil Survey Staff. 1975. Soil taxonomy: a basic system of soil classification for making and interpreting Great Basin Naturalist Vol. 45, No. 1 c • IIS Tfological Survey. 1956. Kern Peak Quadrangle, soil surveys. USDA, Soil Conservation Service. U.S. Geolog^^.^ Geological Survev, USDI, Washing- Agriculture Handbook 436. 754 pp. ton D C 1 pp. STEEL. R G. D.. ANU J. H. Torh:e^ I960. Pnn-Pj- ^f^ ^^^J' y l. 1970. Vegetation change in Sequoia Na- procedures of statistics, with speml references to va. . ^,^^j^^^ ^^^^^ California. Unpublished dissertation, the biological sciences. McGraw-Hill. New ork. ^^^^ ^^ California, Davis. 481 pp. MITES (EXCLUDING CHIGGERS) OF MAMMALS OF OREGON' John C). Whitaker, Jr.,' and Chris Maser' .\bstr.\ct.- New information on the ectoparasitic mites of the following species of mammals from Oregon is pre- sented: (1) pika—Ochotomi; (2) squirrels— Ewfamias, Spennophilus, Tamiascitirus; (3) grasshopper mouse— Oni/c/j- omiis; (4) woodrats— Neoforjifl; (5) muskrat— Ondafra; (6) jumping mice— Z«pi/s,- and (7) weasels— Miisfe/a, Spilogale. New records of species for the state and new host records are also given. Finally, a cross-referenced list of all known mites of wild mammals from Oregon is presented. Over the past several years, we have writ- ten a series of papers on the ectoparasites of mammals from Oregon that include data on the mites of mammals from the following genera: Zapus (Whitaker 1979), Arborhrius and Lagurus (Whitaker and Maser 1979), Aplodontia (Whitaker et al. 1979), Neuro- trichus and Scapanus (Whitaker et al. 1979a), Sorex (Whitaker et al. 1980), Dipodomys and Perognathus (Maser and Whitaker 1980), Myotis, Antrozous, Eptesiciis, Lasionycteris, Lasiurus, Pipistrellus, and Plecotus (Whitaker et al. 1983b), Microtus and Clethrioncnnys (Whitaker and Maser 1984), and Thomomys (Whitaker et al., in press). Still in preparation are papers on the ectoparasites of the Town- send chipmunk (Eiitamias toivnsendi) and of the genus Peroinyscus. One purpose of this paper (Part I) is to present new data on the ectoparasitic, pho- retic, and other mites regularly associated with mammals for which we have small sample sizes. The second purpose (Part II) is to present a cross-referenced list of all known ectoparasitic and phoretic mites (excluding chiggers) from the wild mammals of Oregon and their host associations. Part I. Mites of Selected Mammals FROM Oregon Mammals for which new data are present- ed in this paper were from various parts of the state, as given in the species accounts. Mites were preserved in 70% ethanol, cleared and stained in Nesbitt's solution containing acid fuchsin, and mounted in Hoyer's solu- tion. Cover slips were ringed with Euperal. Identifications were verified by specialists as necessary. Mites that constitute new records from mammals from Oregon are listed below. Numbers in parentheses are the number of hosts with mites and the number of mites found. Lagomorpha Pika, Ochotonidae: Ochotona princeps. Pika (4 from Linn Co.). Mite: Echinonyssus utahensis (1/1). RODENTIA Squirrels, Sciuridae: Eutamias amoenus. Yellow-pine chipmunk (2 from Union County). Mites: Andwlaelaps fahrenholzi (2/4), and Proctolaelaps sp. (1/4). Spennophilus heecheiji. Beechey ground squirrel (3 from Curry County). Mites: Andwlaelaps fahrenholzi (2/2) and Haemogamasus occidentalis (1/1). Sperniophilus heldingi. Belding ground squirrel (3 from Malheur County, 1 from Klamath County). Mites: Andwlaelaps fahrenholzi (4/23), Dermacartis reticulosus (3/25), and Cheijlctus linsdalei (1/1). Spermophilus lateralis. Mantled ground squirrel (2 from Harney County, 1 from Deschutes Coimty). Mites: Androlaelaps fahrenholzi (2/4) and Dermacarus sperm- ophilus (1/21). The latter species was described by Fain and Whitaker (1978a). Spermophilus toivnsendi. Townsend ground squirrel (10 from Malheur County). Mites: Androlaelaps fahren- holzi (7/40), Macrocheles sp. (1/2), Dermacarus sperm- ophilus (1/2), and Eunjparasitus sp. (3/3). Spermophilus washingtoni. Washington ground squir- rel (3 from Morrow County). Mites: Androlaelaps fahrenholzi (1/25) and Macrocheles sp. (2/2). 'This paper represents a partial contribution of the Oregon Coast Ecological Survey, Puget Sound Museum of Natural History, University of Puget Sound, Tacoma, Washington 98416. ■Department of Life Sciences, Indiana State University, Terre Haute, Indiana 47809. ' U.S. Department of the Interior, Bureau of Land Management, Forestry Sciences Laboratory, :3200 Jefferson Way, Corvallis, Oregon 97.33L 67 68 Great Basin Naturalist Vol. 45, No. 1 Tamiasciurus douojasi. Chickaree (6 from Benton. Deschutes, Coos, and Linn counties). Mites: Haenw^a- masiis reidi (5/16), Eidaelaps stabuUnis (1/1). and £(/- rijpanisitus sp. (1/1). Native mice and woodrats, Cricetidae: RcUhrodontomijs megalotis. Western harvest mouse (1 from Jefferson County, 1 from Malheur County). Mites: Androhielaps fahrenhohi (.3/2), Eubraclnihielaps debilis (2/1). Peromy.scus numicidatus. Deer mouse (134 from Lin- coln, Lane, Coos, Curry, Linn, Deschutes, Jefferson, Harney, Malheur, and Union counties). Included here is information on ectoparasites of 1.34 deer mice. Since we have not published this information, we give it here even though the sample size is larger than for most spe- cies included in this section. Mites: Ghjcijphagiis liy- piidaei (1/1), Androlaelaps casaUs (1/1), A. fahrenhohi (197/50), Echinomjssus obsoktus (10/2), £. "obsoletus" variant (1/1), £. utahensis (11/9), Eubrachylaelaps circii- laris (6/2), E. debilis (72/25), Euhiehips stahidaris (2/2), Huemo'^ammm reidi (3/3), Laehips kochi (1/1), Anoet- idae (1/1), Cijrtolaelaps sp. (2/2), Eitryparasitus sp. (27/11), Proctokielaps sp. (2/2). Onychomijs leucogaster. Northern grasshopper mouse (2 from .Maihevir County). Mites: Androlaelaps fahren- holzi (2/45>, Haemogamasiis onychoinydis (2/49), Is- ihyropoda anmtus (1/1), Ornithonysstis bacoti (1/2), Hadfordia subuliger (1/1), Myocoptes musculinus (1/1), CAycijphagns hypiidaei (1/1), Euryparasittis sp. (2/.32), Proctokielaps sp. (2/4), Bakvrdania sp. (2/79), and Klee- mania sp. (2/9). Xeotoma cinerea. Bushy-tailed woodrat (11 from Lane, Lincoln, Malheur, and I'nion counties). Mites: Androlaelaps fahrenholzi (2/6), Haemogamasus reidi (3/10), Etilaelaps stabuluris (3/5), Haemogamasus pon- tiger (2/2), Echinonyssus neotomae (1/1), E. utahensis? (2/2), and Euryparasitus sp. (2/2). \eotoJua fiiscipes. Dusky-footed woodrat {4 from Cur- ry, Klamath, and Lane covmties). Mites: Androliulaps fahrenholzi (1/2) and Haemoganuisus reidi (2/2). Xeotoma lepida. Desert woodrat (3 from Malheur County). Mites: Androlachips fahrenholzi (3/6). Voles, Arvicolidae: Ondatra zibethieiis. Muskrat (12 from Coos and Mal- heur counties). Mites: Laehips midtispinosa (12/317), VAbethaearus ondatrae (8/1213), Listrophorus faini (5/81), L. ondatrae (4/40), L. americanus (4/94), L. f«/iV/i(.v (6/46), L. dozieri (6/129), and Radfordia zibeth- ir«/i.s(l/3). The muskrat is unique in having sexcrai species all in the same genus (IJstropliorns), often in large numbers on the same individual (see Bauer and Whitaker 1981). Mites of the genera YAbethacarus and Listrophorus are tiny and cling to individual hairs. Thus they are often overlooked or taken in relatively small numbers, al- though they are abundant on nearlv everv muskrat. Only six of the muskrats in this stuch were examined thoroughly for these mites. Jumping mice, Zapodidae: y.apiis trinolatus. i^icific jumping mouse (28 from Lin- toln, l^me, Coos, and Liim counties). Miles: Androlae- laps fahrenholzi (14/129). Ilaeniogamasus reidi (4/6), //. occidcntalis (1/1), Dennararus muyorkensis (2/11), Bak- erdania sp. (I/l). and Euryparasitus sp. (5/9). Whitaker (1979) reported the following ectoparasites on this host from Oregon: Androlaelaps fahrenholzi, Xeotrondncula microti. X. harpcri. and Radfordia cuingi. Weasels, Mustelidae: Mustela erminea. Short-tailed weasel (8 from Lane, Coos, and Curry counties). Mites: Haemogamasus occi- dcntalis (3/5), H. reidi (1/1), Eidaelaps stabularis (1/2), Eubrachylaelaps debilis (1/1), Echinonyssus obsoletus (1/1), E. utahensis (1/3), Ghjcyphagus hypudaei (1/12), Orycteroxenus soricis (1/1), Euryparasitus sp. (1/1), and Anoetidae (1/1). Mustela frenata. Long-tailed weasel (4 from Coos. Malheur, and Union counties). Mites: Androlaelaps fahrenholzi (2/8), Echinonyssus cynomys (3/.34), £. lon- gichelae (3/14), E. thomomys (3/9), £. obsoletus (1/24). and Lynxacarus mustelae (1/16). Mustela vison. Mink (1 from Lincoln County, 1 from Coos County). Mites: Lynxacarus nearcticus (2/875-1- ). Echinonyssus staff ordi (1/1), £. obsoletus (1/1), and Pseudoparasitus sp. nymph (1/2). Spilogcde putorius. Spotted skunk (12 from Lincoln. Coos, and Curry coimties). Mites: Androlaelaps fahren- holzi (1/1), Eulaelaps stabularis (3/18), Haemogamasus reidi (8/10.5), Xenoryctes latiporus (1/4), Echinonyssus staff ordi (1/1), Pygmephorus designatus (1/1), £((- cheylctia bishoppi (1/1), and Euryparasitus sp. (1/1). New st.\te records for mites: Laelapidae: Haemogamasus onychomydis. Echino- nyssus neotomae, E. staffordi, E. thomomys, Eu- brachylaelaps circularis, Laelaps multispinosa. Gly- cyphagidae: Dermacarus reticulosus, D. newyorkensis, Xenonicics latiporus. Cheyletidae: Chcyletus linsdalei. Myobiidae: Radfordia subuliger. .Mycoptidae: Myocoptes musculinus. Listrophoridae: Listrophorus americanus. L. dozieri, L. faini, L. ondatrae, L. lalidus. Lynxacarus mustelae, L. nearcticus. X\:\V MOST HECORl^S FOR PARASITIC: MITES; Ochotona princeps-Ecliinonyssus utahensis. Eut(nnias amoenus-Androlaclaps f(direnholzi. Spermophilus heecheyi-Haemogamasits occidcntalis. Spermophilus beldingi-.\ndrolaelaps fahrcidiolzi, Der- macarus reticulosus, Chcyletus lin.\dalci. Spermophilus tounsendi-Dcnuacarus spermophilus. fipermophilus tcashingtoni-Androlaclaps fahrenholzi. Tamiasciurus douglasi-Eulael(i))s stahidaris. Pcromyscus maniculat us- Echinonyssus obsoletus. E. obsoletus "variant." Onychomys leucogaster-Haemogaiiiasus onychomydis. Radfordia subuliger, Myocoptes iiiiisculiniis. C.hjcy- phagus hypudaei. Xeotoma cinerea-Eulaelaps st(d>ularis. Hcuinogamasus pontigcr, Echinonyssus utaliensis. Xeotimui fuscipes- Androlaelaps f(dircnholzi. y.apus Irinotatus-Haenuigamasus occidcntalis. Mustela erminca-Haemoganuisus occidcntalis, H. reidi, E.ubrachylaelaps debilis, Echinonyssus obsoletu.'Sy E. utahensis, Ghjcyphagus hypudaei, Oryctero.xenus so- ricis. Most of these parasites are a result of host transfer during predation. Mustela frenata luhinonyssus cynomys, E. long- ichelae. E. thomomys. E. obsoletus. Echinonyssus spp. January 1985 Whitaker, Maser: Mite Parasites of Mammals 69 are a result of liost transfer during; predatioii; particu- larly interesting are the first three species, which are in- dicative of predation on pocket gophers. ^pilooolc putorius-Andwlaclaps fahrcnholzi, Eit- lachips stcibtikiris, Hacmogamosiis wkli. Xenortictcs lati- ponis. Pii<^))tephortis desi. totcnscndi C.lauconnjs sabrin us Lagurus curtatus \ticrotus longiraudus M. iiumlanus M. oregoni M. rieluirdsoni M. townsendi Seotonta linerea S. fuseipes \. lepida ( hnidioini/s Ictieogaster I'irognatlnts pan us Viroiniiseus luanieulalus Sfiennoplidtis beeeheiji S. beldingi S. lateralis S. townsendi ■S. washingtoni X Zapus trinotatus 30.x Carnivora X Mustela frenata X Spilogale putorius X Androlaelaps geomi/s (Strandtmann. 1949) 34 Rodentia X Thomoniys bulbil orus 35 T. mazama 35 Echinoniissus affinis (|ameson. 195011) Rodentia 38 Eutamias townsendi 39 38 Echinoni/ssus arcuatus (Koch, 1839) Rodentia 18. 23. 37 Peromijscus maniculatus 12 39 (probably a misidentification) Echinom/ssus cijnomijs (Radford, 1941) ■ 36 Carnivora 38 Mustela frenata Echinonijssus nr cijtiomijs X 32 Rodentia 12 Claucomijs sabrinus 32 12 Echinonijssus fernuralis (Allred. 12 1957) X Rodentia Thomounjs bulbil orus 35 T. talpoides 35 Echinontjssus hilli (Jameson. 1950b) 14 Rodentia Dipodomijs ordi 21 36 36 Perognathus parvus 21 Echinonijssus incomptis (Eads & 38 38 Hightower, 1952) Rodentia 12 10 Dipodonnjs ordi 21 38 12, 38 Echinonifssus isahellinus (Oudemans. 38 1913) ■ Insectivora Sorex V a grans 12 33 33 Rodentia Clethrionoiniis califoruiciis 34 34 C. gapperi 34 34 Lagurus curtatus 33 21 Microtus longieaudus 12. 13 X M. niontanus 12. 13. 34 39 M. riehardsoni 34 32 M. townsendi 34 12, 33 Echinonijssus latiscutatus (de 12, 34 Meillon & Lavoipierre. 1944) 12,34 Rodentia 34 Mus musculus 2 34 Echinoni/ssus longichelae (Allred & 34 Beck, 1966) X Rodentia X Glaucomijs sabrinus (nest) 32 X Thomomijs mazama 35 12, X T. talpoides 13. 35 12. 21 C^AHNIVORA 12 Mustela frenata X X Echinonijssus longichelae "variant" X Rodentia \ Thomomijs muzama 35 \ T. talpoides 35 January 1985 Whitaker, Maser: 1 Mite Parasites of Mammals Echinoniissus neotonuic (Eads & Eubrachylaelaps dchilis Jameson, Hightower, 1951) 1950a RODENTIA Insectivora Xeotonui cincrca X Sorcx trowbridgei 38 Echinonifssiis obsolcttts (Jameson, Rodentia 1950b') Eutamias townsendi 39 Insectivora Microtus montanus 34 Scapanus tounscndi 13 M. townsendi 34 Sorex trowhridf^ei 13 RoDENTIA Peromyscus maniculatus X Clethrionoiiiys californicus 34 Carnivora Microtiis kncnsencli 34 Mustela enninea X Peiomiisciis moniculatiis X Eulaelaps stabularis (Koch, 1836) Carnivora Marsupialia Miistela enninea X Didelphis lirginiana 14 M. frenata X Insectivora M. vison X Scapanus orarius 36 Echinomjssus obsoletus "variant" S. townsendi 36 Insectivora Sorex trowbridgei 38 Scapanti.s orarim 36 S. vagrans 38 Sorex henclirei 38 Rodentia S. pacificus 38 Arborimus alhipes 33 S. trowhridgei 38 Clethrionomys californicus 34 S. vagrans 38 Eutamias townsendi 39 S. yaquinae RoDENTIA Peromiiscus nuinicidatus 38 Microtus longicaudus M. montanus 34 12,34 X Echinomjssus staffordi M. oregoni 34 (Strandtmann & Hunt, 1951) M. townsendi 34 Carnivora. Neotoma cinerea X Miisteki frenata X Peromyscus maniculatus 12, X Spdogalc putohus X Tamiasciurus douglasi X Ecliinont/ssiis thomomi/s (Allred & Carnivora Beck, 1966) Mustela ermineu X RoDENTIA Sp do gale putorius X Tlionioim/s maziuna 35 Haemogamasus amhulans (Thorell, T. talpoidcs 35 1872) T. townsendi 35 (recorded from refs. 12, 19, 33, 36, Carnivora 38, 40; probably all incorrect Mustela frenata X and should be referred to Eclunonijssiis triacantluis (Jameson, other species) 1950bj RoDENTIA Dipodonuis sp. D. ordi Echinoni/sstis utahensis (Allied & 16 21 Haemogamasus kcegani (Jameson, 1952) Insectivora Scapanus townsendi 17 Beck, 1966) Haemogamasus Upon yssoides Lagomorpha Ewing, 1925 Ochotona princeps X Rodentia RoDENTIA Microtus longicaudus 12 Dipodonujs ordi 21 M. montanus 12 Lagurus curtatus 33 Haemogamasus mandschuricus Neotonui cinerea X Vitzthuni, 1930 Peromijseus nuinieulatus 12, 13, X Rodentia (originally reported as £. Onychomys leucogaster 12 obsoletus) Haemogamasus nidi Michael, 1892 Tlioinomys talpoides 35 Marsupialia Carnivora Didelphis virginiana 14 Mustela ermineu X (probably misidentification) Eubrachijlaelaps circularis (Ewing, 1933) ' RoDENTIA Haemogamasus occidentalis (Keegan, 1951) Peromiiscus crinitus Eidnachiilaelaps crowd Jameson, X Insectivora Neurotrichus gihbsi 19,37 37 1947 ' Scapanus orarius RoDENTIA S. townsendi 19,37 Microtus montanus 12 Sorex bendirei 38 Onychomys leucogaster 12 S. pacificus 38 71 72 Great Basin Naturalist Vol. 45, No. 1 S. tnnchridgei 38 Carmvora S. vagrans 38 Mustcla erminea X S. i/aqtiinae 38 Spilogale piitorius X RODENTIA ("Originally listed as amhulans. Arborimus alhipes 33 reidentified by Redington) Clethrionotmjs californicus 34 Haemogamasiis thomomysi Eutamias townsendi 39 Williams, Sniilev, & Redington Microtus longicatidiis 34 1978 M. montanus 34 RoDENTlA M. oregoni 34 Tlwrnomi/s mazama 35 M. richardsoni 34 T. talpoides 35 M. townsendi 19,. 34 Haemogamasiis sp. Spewurphihis heecheyi X Rodentia 37 7,(ipiis trinotiitiis X Aplodontia nifa CAKMVoav Haemogamasiis n. sp. Mitstclo cwiincd X Rodentia M. frcnatu 19 Clethrionomijs californicus 34 Haemoganui.sus on iiclioni ijdi.s (Ewing, 19.3.3) ROUENTIA Microtus oregoni M. townsendi Hijpoaspis miles (Berlese, 1892) 34 34 Dipodomijs ordi Omichomijs Iciicogastcr Perognathus parvus 21 X 21 Rodentia Microtus townsendi Ischijropoda armatus Keegan, 1951 Rodentia .34 77iOHio?»ii/.s mdzauKi T. talpoides T. toicnsendi 3.5 35 35 Dipodomijs ordi Onychomi/s Iciicogdstcr Pcrogiuithus paniis 21 X 21 Haemogamasiis pontigrr (Beilese, Isclniropodd furmani Keegan, 1956 Rodentia 19()4) RoDENTIA Dipodomijs ordi 21 Glaticont ijs sdhiiii us 32 Laclaps alaskensis Grant. 1947 Rodentia S'eotcmui cincrcii X \. Icpidd 12 I.agurus curldtus 12 Hac'))iog(H)iasus reidi Ewing, 1925 Microtus monlanus 12. .34 Insectivora M. ricluirdsoni 34 Scapanus orarius 37 Omjchomijs leucogdstcr 12 S. townsendi 37 Thomomys mazama 35 Sorex hendirei 38 Laclaps kocl'ii Ondemans, 1936 .S. twu'hridgci 38 Insectivora S. lagrans .38, 40 Neiirotrichus gihhsi .■36 S. ijdcpiinac 38, 40 Rodentia RoDENTlA Microtus lougicdudus 1 2°.. 34 Aplodontia rufd 37 M. montanus 12°.. 34 Arhohnutfi dU)ipcs 3 M. oregoni 34 (HctlihonoDiijs ciilifornicii.s .54 M. toivnsendi 34 C. gappcri 34 Peromyscus mauiculatus X Dipodoinijs ordi 21 1 'idcnlificd as l.dclups Eiittiniias lounseiuli 39 pdcliyj)us) C.ldticoniijs .sy;/j;iiii/.s 32 Laclaps multispinosa (Banks. 1910i Ldgunis ciirldttis 12° , 26, .3.3 Rodentia Mivrottis longicdudiis 12° , 26, 34 Onddtra zilxiliicus X M. montanus .34 Pdtrinyssus liuhhardi ijauifson. M. rirhardsoni 34 1949) M. townsendi 34 HoDi ntia Scotoma cinerea X Ajilodontid rufd 37 .V. fuscipcs X LlSTROl'HOlUUAE Penmnjsciis maniniUiliis X Aplodonlochirus horcdiis I'"ain c\ Tcntiidscinnis doiiglasi X Hvland, 1972 77i(»Hi()()ii/.s nuizama 35 Roi.lNTIV T. talpi>id(s 12° , 35, .36 Aplodontid rufd 37 y.apiis princcps 12° ,26 C'.comiiHctius ))crogndtlii Fain ^: /.. trinolalns X Wli'itaker. 1980 Also "l''()irst Tii'o Mouse" Rodentia (=Arl)oriunis longicatidiis) ■26 Pcrognallius iiun us 9. 21 I January 1985 Whitaker, Maser: Mite Parasites of Mammals 73 Gcunnilicliiis tcxcn.si.s Fain, Whitaker, Scliwan. & Lukoschus, 1981 RODENTIA Dipodomiis onii Lcporacanis stihihigi Fain, Whitaker, & Lukoschus, 1981 21 Lagomorpha Siilrihi^iis Iwclinxini Listiophorus iniiciicdiuis l-ladtortl. 11 1944 RODE.NTIA Oiuhitni zihcthictis Lisiwphoitis dozicii Radford, 1944 RoDENTIA X Onddtiv zihcthicu.s X IJstrophorus faiui Dubinina. 1972 RoDENTlA Onddtni zihcthictis X Listivphonis iiicxicdiuis Fain, 1970 lNSECTIV()a\ Scapdiuix touiiscndi Sorcx hcndirci 36 38 RoDE.NTIA Clethrionoiiujs cdUfoiniciis 34 Lcigiinis curtatm Microttts montanus 33 34 M. oregoni M. richardsoni 34 34 M. toivnsendi Listroplionis onddtive Fain, Kok, & Lukoschus, 1970 34 RODEXTIA Onddird z.il>ctlticiis X Listrophcni.s idhdus Banks, 1910 RoDENTlA Ondatra zibcthicus X Liinxacanis mustclae (Megnin, 1885) Carnivora Mitsteki ficnata Liinxacanis ncarctictis Fain & Hyland, 1973 X Carnivora Mustela visou X Quasdistroplwnis niicroticohis Fain, Whitaker, & Lukoschus, 1978 RODE.NTIA Arhoiiimis albipes A. longicaiidiis 10, 33 10, 33 Macronyssidae Cniptoniissiis desidtoriiis Radovskv, 1967 ' Chiropter.\ Eptesicus fusciis Myotis californicus M. hicifiiotis M. volans 41 41 41 41 Macwnyssiis crosbiii (Ewing & Stover, 1915) Chiroptera Eptcsictis fuscu.s 41 Mijotis cdliforn iciis M. evotis 41 41 M. hicifugus 41 A/, vuhins M . ijinniincnsis Macronijssiis loiigisclnsus (Furiuan, 1950) Chiroptera Plecotus townsendi Macwnijssus niacrodactiihis Radovskv & Beck, 1971 Chiroptera Lasionyctciis noctiidgans Ornidwnyssus hdcoti (Hirst, 1913) (host not given) Rodentia On ychoin ys leucogastcr Steatonyssus antrozoi Radovsk)' & Furman, 1963 Chiroptera Antrozotis pallidiis Steatonyssiis emarginatiis Radovskv & Furman, 1963 Chiroptera PipistrcUtis hcspcnis Steatonyssiis ftinuaiii Tipton & Boese, 1958 Chiroptera Lasiiinis cincrciis Steatonyssiis occidentalis (Ewing, 1933) Chiroptera Epfcsiciis fiisciis macronyssid n. sp. Chiroptera Eptesicus fiisciis Lasionycteris noctivagans My Otis californicus M. evotis M. hicifugus M. cohin.s M. yiiDuincnsis Plecotus loicnscndi Myobiidae Acanthoplithiriiis ciiuddtus eptesicus Fain & Whitaker, 1978 Chiroptera Eptesicus fuseus Aeanduiphthirius sp. nr gracilis Fain & Whitaker, 1978 Chiroptera Myotis volans Acanthophthirius oregonensis Fain & Whitaker, 1978 Chiroptera Pipistrellus hespenis Aniorpliacarus hengererorum Jameson, 1948 Insectivora Sorex pacificiis S. vagrans S. yaquinae Amorphacarus sorieis Fain & Whitaker, 1978 Insectivora Sorex bendirei 25, 41 41 8,38 74 Great Basin Naturalist Vol. 45, No. 1 Eadiea neurotrichtis Liikoschiis, Klompen, & Whitaker, 1980 Lnsectivora Neurotrichus gihbai Eadiea scapanus Fain & Whitaker, 1975 Lnsectivora Scapamis townsendi Eutalpacanis peltcittis Jameson, 1949 Insectivora Neiirotrichus gibl)si Pwtoimjohia brevisetosa Jameson, 1948 Lnsectivora Saipanus toicmendi Sorex bendirei S. pacificus S. troiihrklgei S. vagrans S. ijaifuinac RODENTIA Clethrionouiijs calif nniicus Pteracartts aculeus Dusliahek ^: Lukoschus, 1973 Chiroptera Eptesicus ftisciis Pteracartts sp. nr mintittts dattbetxUmi Dusbabrk. 1973 Chiroptiha Mijotis lohiiis Rudfordia arborimiis Fain 6< Whitaker, 1975 Lnsectivora Satp(tmis totiivsoidi Rudfordia eiviiigi (Fox, 1937) Rodentia YMptts triiiotatits Rudfordia hylaiuli Fain iy Lukoschus, 1977 Rodentia Microttis orcgoni Rudfordiu sttbttliger F\\ ins;, 1938 Rooentia Omjchomtjs Icticogaster Rudfordiu zibcthicdlis (Radford, 1936) Rodentia OruUitra zibrllticiis Myckoptidae Mtjocoptes juponcnsis Radford, 1955 Rodentia CIc'tlirioiioiiitjs talifoniiciis Microlits ricliurdsoni Mtjocoptes tiiiisctilititis (Kocli, 1844) Rodentia ()n|/(7ioHii/.v Icticogastcr NOTOEDRIDAE S'otocdres {Bakcracartis) sp. (ImnoiTERA Eptesictis ftisctis Mtjotis volaiis A/, ijtimaiiciisis PIccotiis lottiisctidi 20 6,36 36 6, 33 Pygmephoridae Pygmephorus designattis Mahunka, 1973 Insectivora Sorex troicbridgei 38 Rodentia Cleth rionom i/s californictis 34 Carnivora Spilogale ptitoriiis X Ptjgmephortis horridus Mahunka, ' 1973 Insectivora Scapantis oraritis 36 Sorex troicbridgei 38 S. tjaqiiinae 28, 38 Ptjgmephortis johttstoiti Smiley & Whitaker, 1979 Insectivora Netirotrichtis gibbsi 28 Ptigmephorus httterlotighac Smilev ■ & Whitaker, 1979 Insectivora Scapantis touitscndi 28 Pilginephortts itiahintkai Smiley & ■ Whitaker, 1979 Insectivora Scapantis toictisetuli 28 Ptigmephorus scalopi Malumka. ' 1973 Insectivora Scapantis oraritis 28 Ptjgmephortis tvhitukeri Mahunka. 19" 3 Insectivora Scupanus oraritis 28 Spinturnicidae Spiutiirnix amcriccmtis (Banks, 1902) Chiroptera Antrozoiis pallidas 41 Eptesictis ftisctis 41 Mtjotis ecotis 27, 41 M. hicifugtis 12. 41 M. volans 41 M. iiiiDuniciisis 27 41 Spintiirnix hakeri Rudnick, 1960 Chiroptera Eptesictis ftisctis 41 Myotis volans 41 Spinturnix globostis (Rudniek. 196()> Chiroptera Myotis i()/f;/is 41 Spinturnix orri Rudnick. 19(S() Chiroptera Antrozoiis pallidum 27 41 Acknowledgments The following assisted with the identi- fication of mites: Alex Fain, Institut de Medecine Tropicale Prince Leopold, Ant- werpen, Belgium (several groups of mites); S. D. Herrin, Center for Environmental Studies, Brigham Young University, Provo, Utah (mites of genus Ediinoni/ssus); E. W. Jame- son, Department of Zoology, University of January 1985 Whitaker, Maser: Mite Parasites of Mammals 75 California, Davis (myobiid mites); G. W. Krantz, Department of Entomology, Oregon State University, Corvallis (macrochelid mites); E. E. Lindquist, Research Branch, Biosystematics Research Institute, Agricul- ture Canada, Ottawa, Ontario (cyrtolaelapid and otlier mites); F. S. Lukoschus, Depart- ment of Zoology, Faculteit der Wiskunde en Natuuruietenschappen, Nijmegen, The Neth- erlands (several groups of mites); B. Mc- Daniel, Entomology-Zoology Department, South Dakota State University, Brookings (several groups of mites); F. J. Radovsky and J. A. Tenorio, Bernice P. Bishop Museum, Honolulu, Hawaii (macronyssid mites); R. L. Smiley, Systematic Entomology Laboratory, Beltsville, Maryland (pygmephorid and cheyletid mites); Nixon Wilson, Department of Biology, University of Northern Iowa, Ce- dar Falls (several groups of mites). The manuscript was read and improved by: Steven Cline (Department of Forest Sci- ence, Oregon State University, Corvallis), Donald Grayson (Department of Anthropolo- gy, University of Washington, Seattle), Mur- ray Johnson (Puget Sound Museum of Natu- ral History, University of Puget Sound, Tacoma, Washington), and Dale Toweill (De- partment of Fisheries and Wildlife, Oregon State University, Corvallis). Martha Smith (Indiana State University, Terre Haute) aided in preparation of mate- rials and in keeping records. Sarah Berry (USDA Forest Service, Forestry Sciences Laboratory, 3200 Jefferson Way, Corvallis, Oregon) typed the various drafts of the manuscript. Errors in identification are ours, since we have made tentative identifications and then sent only representative specimens for veri- fication or reidentification by specialists. Literature Cited 1. Bauer, C. A., and J. O, Whitaker, Jr. 1981. Ecto- parasites of muskrats from Indiana with special emphasis on spatial distribution of coexisting mites of the genus Listroplwrus. Amer. Midi. Nat. 105:112-123. 2. Ewing, H. E. 1922 (1923). The dermanyssid mites of North America., proc. U.S. Natl. Mus. 62:1-26. 3. Ewing, H. E. 1933. New genera and species of par- asitic mites of the superfamily Parasitoidea. Proc. U.S. Natl. Mus. 82:1-14. 4. Fain, A. 1967. Diagnoses d'Acariens nouveaux par- asites de rongeurs ou de singes (Sarcoptiformes). Rev. Zool. Bot. Afr. 76:280-284. ■5. Fain, A. 1969. Lex deutonymphes hypopiales vi- vant en association phoretique sur lex mammiferes (Acarina: Sarcoptiformes). Bull. Inst. Roy. Sci. Na- tur. Belg. 45:1-262. 6. Fain, A., and J. O. Whitaker, Jr. 1975. Two new species of Myol:)iidae from North America mam- mals (Acarina). Bull. ."^nn. Soc. r. Beige Entomol. 111:57-65. 7. Fain, A., and J. O. Whitaker, Jr. 1978a. Notes on two phoretic hypopi (Acarina: Glycyphagidae) on mammals, including a description of Dermacanis spcrmopliihis sp. n. J. Parasitol. 64:512-514. 8. Fain, A., and J. O. Whitaker, Jr. 1978b. Two new myobiid mites from western North America (Acaria: Mvobiidae). J. Parasitol. 64:895-899. 9. Fain, A., and J. O. Whitaker, Jr. 1980. Geomtj- lichus perogrtathi sp. n. (Acari: Listrophoridae) from Perognatluis spp. in the United States. J. Par- asitol. 66:839-840. 10. Fain, A., J. O. Whitaker, Jr., and F. S. Lukoschus. 1978. Qiiasilistwpliorus microticohis gen. n. et sp. n. (Acari: Listrophoridae) from North American microtine rodents. J. Parasitol. 64:1097-1099. 11. Fain, A., J. O. Whitaker, Jr., and F. S. Lukoschus. 1981. Leporacanis (Leporacoides sijlvilagi) n. sp. (Acari, Listrophoridae) parasite d"un lievre ameri- cain. Bull. Ann. Soc. r. Beige Entomol. 117:214. 12. Hansen, C. G. 1964. Ectoparasites of mammals from Oregon. Great Ba.sin Nat. 24:75-81. 13. Herrin, C. S. 1970. A systematic revision of the genus Hirstionijssiis (Acari: Mesostigmata) of the nearctic region. J. Med. Entomol. 7:391-437. 14. Hopkins, D. 1980. Ectoparasites of the Virginia opossum {Didelpliis virginiana) in an urban envi- ronment. Northw. Sci. 54:199-201. 15. Jameson, E. W., Jr. 1950a. Eiibniclnjlaclaps dcbilis, a new laelapid mite (Acarina: Laelaptidae) para- sitic on the deer mouse, Peromijsciis maniciihitus (Mammalia: Cricetidae). J. Parasitol. 36:62-64. 16. Jameson, E. W., Jr. 1950b. Notes on mites of the genus Neoichoronysstis, with description of a new subgenus and three new species of the subgenus Hirstionyssus (Acarina, Laelaptidae, Macronys- sinae). Proc. Entomol. Soc. Wash. 52:161-172. 17. Jameson, E. W., Jr. 1952. Eidiaemogamanus keeg- ani, new species, a parasitic mite from western North America (Acarina: Laelaptidae, Haemo- gamasinae), Ann. Entomol. Soc. Amer. 45:600-604. 18. Jellison, W. L. 1945. A new mite, Laelaps ciplodon- tiae, from Aplodontia. J. Parasitol. 31:373-374. 19. Keegan, H. L. 1951. The mites of the subfamily Haemogamasinae (Acari: Laelaptidae). Proc. U.S. Natl. Mus. 101:20.3-268. 20. Lukoschus, F. S., J. S. H. Klompen, and J. O. Whitaker, Jr. 1980. Eadiea netirotrichiis, n. sp. (Prostigmata: Myobiidae) from Xeurothchus gibh- sii (Insectivora: Talpidae). J. Med. Entomol. 17:498-501. 21. Maser, C., and J. O. Whitaker, Jr. 1980. xMites found in the fiu- of a small .sample of heteromyid rodents from Oregon. Northw. Sci. 54:279-280. 22. Pence, D. B., and J. P. Webb, Jr. 1977. Notes on hypopi of two Dermacanis species (Acari: Astig- mata: Glycyphagidae) from the Douglas squirrel, Tamiascittnts doiiglasii. J. Med. Entomol. 14:175-179. 76 Great Basin Naturalist Vol. 45, No. 1 23. Radford, C. D. 1951. Two new genera of parasitic mites (Acarina: Laelaptidae and Listroplioridae). Parasitology 41: 102- 104. 24. Radovskv, F. J. 1967. The Macronyssidae and I^ielapidae (.Acarina: Mesostigmata) parasitic on hats. L'niv. California Puhl. Entomol. 46:1-288. 25. Hadovsk), F. J., and D. P. Furman. 1963. The Xortli .\nierican species of Stcatonij.i.sus (Acarina: Dernianvssidac). Ann. Entomol. Soc. .Amer. .56:268-276. 26. Hedington. B. D. 1970 (1971). Studies on the mor- phology and taxonomy of Haemogmmmis rcidi Ew- ing, 1925 (.\cari: Mesotigmata). .\carologia 12:643-667. 27. Rudnick. A. 1960. A revision of the mites of tlie familv Spinturnicidae. l'niv. Cahfornia Fulil. Ento- mol. 17:157-284, 28. Smiley, R. L., and J. (). Whitaker, Jr. 1979. Mites of the genus riigmeplwrus (Acari: Pygmephoridae) on small mammals in North America. Acta Zool. Acad. .Scient. Hungaricae 25:383-408. 29. Smiley, R. L., and'j. (). Whitaker, Jr. 1981. Studies on the idiosomal and leg chaetota.w of the Chey- letidae (.\cari) with deseriptions of a new genus and four new species, Interuat. J. .\carol. 7:109-127. .30. Strandtmann, i{, W , 1949. The blood-sucking mites of the genus UncmuhicUips (.Acarina: Laelaptidae) in tlie United States, J. Parasitol. .35:325-352. 31. Whitaker, J. ()., Jr. 1979. Origin and evolution of the external parasite fauna of western jumping mice, genus Z(ij)ii.s. Amer. Midi. Nat. l()l;49-6(). 32. Whitaker, J. ().. Jr., E. A. Lvons. .M. A. Smith, and C;. Ma.ser. 1983a, Xest inhiihitauts and ectopara- sites of northern tl\ing s((uirrels. (.'.Iduamiiis sahii- niifi (Shaw), from nortlieastern Oregon. Nortlnw Sci. 57: 291-295. Whitaker. J. O.. Jr., and C. .\laser. 1979. Parasitic mites of voles of the genera Aiborimiis and Lti- gunis in Oregon. Northw. Sci. 53:224-227. Whitaker. J. O., Jr., and C. Maser. 1984. Parasitic mites of voles of the genera Micwtus and Clctli- rioniwnis. Northw. Sci. 58:142-150. Whitaker, J. O., Jr., C. Maser, and R. E. Lewi.s. In Press. Ectoparasitic mites (excluding chiggers), fleas, and lice from pocket gophers, genus T/io- iiiDDiiis. from Oregon. Northw. Sci. Whitaker, J. O., Jr., C. Maser, and R. J. Pedersen. 1979a. Food and ectopara.sitic mites of Oregon moles. Northw. Sci. .53:268-273. Whitaker, J, O., Jr., C:. Maser. and W. M. \\allace. 1979b. Parasitic mites of the mountain beaver (Aplodontid nifa) from Oregon. Northw. Sci. .53:264-267. Whitaker, J. O., Jr., G. L. Tieben. and C. .Maser. 1980. Mites (excluding chiggers) from the fur of five species of western Oregon shrews. Northw. Sci. .54:26-29. Whitaker, J. ().. Jr.. W , J. Wrenn. and C. .Maser. In preparation. Mites and hce from tlie hn- of 231 Townsend clupmunks. l-jitaniias toicnscndi. from Oregon. Whitaker, J. ().. Jr.. and N, Wilson. 1974. Host and distril)uti()n lists of mites uVcarit, parasitic and phoretic. in the liair of wild mammals of North America, north of Mexico. Auier. .Midi, Nat. 91:1-67. Whitaker, J. O., Jr., C. E. Yunker, and C. .Maser. 1983b. .Acarine ectoparasites (mites^ of bats of Ore- gon. Northw. Sci. .57:97-106. Williams. R. W. 1946. The laboratorv rearing of the tiopical rat mile. Lipomissus Ixutoi iHirstl. J. itol. :252-256, FOOD OF COUGARS IN THE CASCADE RANGE OF OREGON' Dale E. Toweill- and Chris Maser' .\nsTi\.\CT.— Animal and nonanimal items were identified in the digestive tracts of 61 cougars (Felis concolor) collected between 1978 and 1984 from the western slopes of the Cascade Range in Oregon. Forty-two (69%) of the cougars were taken bv hunters in Decembei- and January, 18 (30"(i) were killed at other times of the year because of their proximity to livestock, and one animal was illegally killed in November. Black-tailed deer {Odocoilcus Iwmionu'i columbi(inus) was the most common prey item, although domestic sheep {Ovis ciries), porcupines (Ercthizon dorsatitni). and a variety of small mammals were also recorded. Masticated grass was the most common nonanimal item. Tlie cougar was placed under jurisdiction of the Oregon State Game Commission in 1967 because of a suspected decline in num- bers in the 1950s and early 1960s (Oregon State Game Commission 1967). Prior to 1967, tlie cougar had been subjected to the bounty system established by the Oregon Territorial Government in 1843. The bounty system was repealed by the Oregon Legislature in 1961 (Kebbe 1961), but the cougar was not pro- tected until it was declared a game animal in 1967. It was then protected until 1971, ex- cept for individuals that were killing livestock. The first controlled hunt for cougars in Oregon took place in 1971; 13 were killed. Information on food habits from those 13 ani- mals (3 from the Cascade Range) and from 12 taken in 1972 (none from the Cascade Range) was reported by Toweill and Meslow (1977). The purpose of this paper is to present data on foods of cougars from the Cascade Range of Oregon and to compare these data with other data from Oregon (Maser and Rohwe- der 1983, Toweill and Meslow 1977) and elsewhere. Methods Most of the animals examined during this study were taken legally by hunters during the annual controlled season in December and January; one cougar illegally killed in November and confiscated by Oregon State Police was also examined. Successful hunters were required to present their cougars to personnel of the Oregon Department of Fish and Wildlife within 48 hours, at which time biological data were collected and ownership of the pelt was validated. Additional cougars, killed to protect livestock (primarily domes- tic sheep), were obtained throughout the year. Sex, weight, and physical measurements were recorded for each animal either by per- sonnel of the Oregon Department of Fish and Wildlife or by us. Digestive tracts and repro- ductive organs were removed, labeled, and frozen for later analysis. Each complete di- gestive tract was examined as three separate elements: stomach, small intestine, and colon. Because some animals were eviscerated in the field by hunters, many of whom brought in only stomachs, more stomachs than colons were available for analysis. Weight of stom- ach contents was recorded to the nearest gram. Endoparasites were also preserved. Complete digestive tracts normally repre- sented at least two meals: one in the stomach and one in the colon, with elements from both often found in the small intestine (Maser and Rohweder 1983). Items from stomachs and from colons were recorded separately. Nonanimal items, particularly fragments of vegetation, were identified to provide insight about the habitats in which meals were ingested. ■ of the Oregon Departn , who helped i 'This paper is dedicated to the memory of Ronald Rohweder, an i n cougars of Oregon. Ron was electrocuted 1 July 19S4. -Oregon Cooperative Wildlife Research Unit, Department of Fisheries and Wildlife, Oregon State University, Corvallis, Oregon 97.3.31. 'U.S. Department of the Interior. Bureau of Land .Management. Forestry Sciences Laboratory, 3200 Jefferson Way, Corvallis, Oregon 97.331 77 Great Basin Naturalist Vol. 45, No. 1 Results and Discussion Digestive tracts of 61 cougars containing food items were examined. Information, con- sidered to represent 115 separate meals, was obtained from 61 stomach samples and 54 co- lon samples. All cougars examined in this study were collected from the western slopes of the Cas- cade Range in Oregon between 1978 and 1984, 40 (66%) from Douglas County, 19 (13%) from Lane Covmty, and 1 each from Ciury and Hood River counties. Because of himt unit boimdaries and because animals taken to control livestock predation were killed near human habitations, all cougars were taken at elevations below approx- imately 600 m (2000 ft). Cougars taken throughout the year were examined, although most were taken during the December-January controlled season. Numbers of animals in this sample by month killed were as follows: January— 14, March— 6, April-3, May-1, June-1, July-2, Sep- tember—4, October— 1, November— 2, and December— 27. Data were recorded as legal hmiter kills (42) and others (19). The sex ratio of cougars in our sample was essentially even, 33 males and 28 females. Of these, 20 males and 22 females were taken during the December- January period, and 13 males and 6 females were taken at other times of the year. Weight of stomach contents ranged from less than 1 g (traces of hair) to 3.97 kg (9 lbs). Black-tailed deer was the food item most commonly found in cougars in winter. The dominance of deer in the diet seems a con- stant throughout North America, as is evi- denced by studies conducted in Oregon (Ma- ser and Rohweder 1983, Towcill and Meslow 1977), Washington (Schwartz 1943 in Young and Goldman 1964), Idaho (Hornocker 1970), California (Dixon 1925), Utah and Nevada (Robinette et al. 1959), Utah (Ackerman et al. 1984), Arizona and New Mexico (Hibben 1937), British Columbia (Spalding and Les- owski 1971), and throughout the western United States (Sperry in Young and Goldman 1964). Hippoboscid flies {Lipoptena deprcssa pacified), a common ectoparasite of black- tailed deer, were recovered from the stom- achs of four cougars. Absence of Roosevelt elk {Cervus elaphus roosevelti) in the diet of cougars from the Cascade Range is puzzling. Despite their size, elk are commonly reported in the diet of cougars where ranges of the two species overlap (Ackerman et al. 1984, Hornocker 1970, Maser and Rohweder 1983, Robinette et al. 1959, Toweill and Meslow 1977). Roosevelt elk are relatively common throughout Douglas and Lane counties, al- though they are most often found at eleva- tions above 600 m (2000 ft) and away from human habitations. We suspect that the lack of elk in the diet of these cougars was a result of the small sample size and low elevations at which the cats were taken. Depredation of domestic sheep by cougars occurred all year at low frequency but seemed most common during spring. All oc- currences of domestic sheep in our sample came from cougars killed because of live- stock damage. The high frequency of occur- rence of sheep in the diet of these cougars re- sults from the bias introduced by the inclusion of cougars killed as a result of live- stock depredations. Our data indicate that these cougars had usually fed on sheep for at least two consecutive meals. Porcupines were recorded in the cougar diet with some regularity, and many of the cougar carcasses examined showed evidence of contact with porcupines in the form of embedded quills. The importance of porcu- pines in the cougar diet has been noted by Maser and Rohweder (1983) and Young and Goldman (1964). Deer, porcupines, and grass have been the most commonly reported ^staples of the cougar diet (Dixon 1982). The domestic dog {Canis f(uniUaris) found in the stomach of one cougar may represent feeding on carrion. Fly larvae, found among clumps of dog hair, indicated that the dog had l)een dead several hours before it was consumed by the cougar. Other animals found in the diet of cougars during this study included mountain beaver {Aplodontia rufa), muskrat {Ondatra zihethi- cus), beaver {Castor canadensis), northern fly- ing squirrel {Claucomijs sahrinus), dusky- footed woodrat {\cotoma fiiseipes), Trow- bridge shrew {Sorex trowhridii,ei), and hare {Lepus sp.) and may represent opportunistic feeding (Table 1). Such small mammals have January 1985 TowEiLL, Maser: Oregon Cougar Diet 79 also been found in other studies of cougar food habits but usually total less than 10% of the diet by frequency of occurrence and con- siderably less in terms of relative biomass consumed. Exceptions have been noted, how- ever, in Arizona (Hibben 1937) and southern Utah (Ackerman et al. 1984). We assumed that fish bones, found in the stomach of one cougar reportedly caught in a leg-hold trap, represented trap bait since only skull bones were found. Further, the bones were associated with litter (Douglas-fir {Pseudotsuga menziesii) needles, twigs, peb- bles, and soil) typical of ground cover within a closed-forest stand. Masticated green grass was found in many of the digestive tracts (Table 1), often in large amounts. Although of negligible food value (evidenced by its resistance to diges- tion), such grass may have served to purge some of the tapeworms commonly found in cougar intestinal tracts. We have observed, in the course of field work, recently passed scats consisting almost entirely of masticated grass with entwined tapeworms. Vegetation associated with food items sug- gested that cougars fed most commonly among closed-forest habitats during the win- ter. Needles of Douglas-fir were found in 62% of the stomachs and 89% of the colons of cougars killed in December and January and were associated with plant remains from pon- derosa pine {Pinus ponderosa), vine maple {Acer circinatum), western redcedar {Thuja plicata), Oregon grape {Berberis nervosa), and fern {Polystichum sp.). Douglas-fir needles were also commonly found in association with food of cougars killed other than during December and January (24% of stomach sam- ples and 63% of colon samples). Several plant species characteristic of dry, open canopy vegetation types, not recorded from cougars taken during winter, were found in cougars collected at other times of the year: oak {Querciis garryana), snowbrush ceanothus {Ceanothus velutinus), and Pacific poison oak {Rhus diversiloba). Soil, pebbles, and dry wood, evidence of a cougar having cleaned up a kill, were also found. Conclusion Black-tailed deer, porcupines, and grass are staples of a cougar's diet in the Cascade Range of Oregon, as elsewhere in North America. Cougars prey on domestic sheep when available and may take a number of species of smaller mammals (and perhaps car- rion) as available. Data suggest that most feeding by cougars was done in closed-cano- py vegetation types during winter and that Table 1. Items identified from dii^estive traets of 61 couii;iiis collected from the Oregon. Data presented as number and frequency (in parentheses) of occurrence. .estern Cascade Range of Stomach" Colon" Food items Wi nter Other Winter Other Black-tailed deer 27 (64) 3 (16) 23 (61) 4 (25) Domestic sheep*^ 6 (14) 7 (37) 3 (8) 5 (31) Domestic dog 0 1 (5) 0 0 Cougai'^' 8 (19) 3 (16) 12 (32) 3 (19) Porcupine 1 (2) 1 (5) 1 (3) 2 (13) Mountain beaver 1 (2) 0 1 (3) 1 (6) Muskrat 1 (2) 0 1 (3) 0 Beaver 1 (2) 0 0 0 Northern flying squirrel 0 0 1 (3) 0 Dusky-footed woodrat 1 (2) 0 0 1 (6) Trowbridge shrew 1 (2) 0 0 0 Hare 0 0 1 (3) 0 Unidentified mammal 0 0 0 1 (6) Unidentified fish'' 1 (2) 0 0 0 Masticated grass 4 (10) 2 (11) 8 (21) 6 (38) Number examined 42 19 38 16 "Winter; tal 0.05). However, there was a .significant difference in brood size between 1976, a year with normal precipitation, and 1977, when a drought occurred (F = 3.77, P < 0.05). Of the 22 mortalities confirmed (2 nest- lings, 15 juveniles, and 5 adults), 6 resulted from collisions with motor vehicles. Most of the others were recovered from the vicinity of nest sites, and the cause of death was not identified. Two nestlings were found dead within a burrow, which was excavated after the nesting pair disappeared. We assume that badgers were the major predator on nestlings prior to emergence. A badger was observed entering one of the nest burrows containing an unknown number of voung, onlv one of which later fledged. Food availability affected the productivity of the study population. Ord kangaroo rats {Dipodomys ordii), deer mice (Peromyscus manicukitus), and Great Basin pocket mice {Perognathus parvus) composed 40% of the biomass taken by burrowing owls in 1976 and 32% of that taken the following year (Table 1), a reduction in proportion that is highly significant (Z = 13.3, P < 0.001). This J January 1985 Gleason, Johnson: Burrowing Owls 83 dietary change probably represents a func- tional response to the decrease in density of these species on the INEL in 1977 (Table 2). Based on analysis of castings recovered from nesting sites, there was a significant positive relationship between the biomass of rodents in the diet of individual nesting pairs and the size of their respective broods in 1976 (r = 0.56, N = 10, P < 0.05) but not in 1977. Discussion Most of the burrows on the INEL probably resulted from attempts by badgers to capture Townsend ground squirrels {Spennophilus townsendii). However, nesting pairs also uti- lized badger-excavated burrows of Ord kan- garoo rats near Howe, a location where no ground squirrels occurred. Given the wide distribution of prey suitable for badgers, we assume that large portions of the study area lacked burrowing owls because of factors other than the availability of suitable nesting sites. Although fratricide may have caused some mortality before fledging, we observed no agonistic encounters between siblings. Most mortality occurred after fledging, when the young may be particularly vulnerable to star- vation because of their inexperience in cap- turing prey. Rodents rather than invertebrates repre- sent a more reliable energy source during pe- riods of food-stress because individual cap- tures provide greater biomass. However, the use of Jerusalem crickets is energetically fa- vorable because of their large size and ease of capture. This insect, which is common throughout the arid portions of the western United States (Essig 1936), is usually active above ground only at night, although some are found at the surface on cool, cloudy days (La Rivers 1948). We have no information on annual changes in the density of Jerusalem Table 1. Diet of burrowing owls on the INEL site based on casting analysis, 1976 and 197 1976 1977 N %N Percent biomass N %N Percent biomass Mammals 340 12.6 85.6 160 4.9 61.5 Dipodomijs oidii 59 2.2 21.7 44 1.4 23.7 Micwtits montamis 89 3.3 22.6 56 1.7 20.8 Perognathus panus 84 3.1 8.4 39 1.2 5.7 Peromijsciis mankuhitiis 84 3.1 9.6 14 0.4 2.3 Thomomijs talpoidcs 23 0.9 23.1 6 0.2 8.8 Miis muscuhis 1 <0.1 <0.1 1 <0.1 <0.1 Unidentified passerine .\mphibians Scapliiopus intcruioiitdi .\rachnids Scorpionidae Solpugidac Insects Gryacrididae Acrididcie Cicadidac Carabidae Silphidae Scanibaeidae Tenebrionidae Fonnicidae Vnid. Hipnenoptera Unid. Coleoptera Total <0.1 <0.1 520 19.4 244 9.1 276 10.3 1824 67.8 491 18.2 78 2.9 10 0.4 69 2.6 156 5.8 267 9.9 125 4.6 21 0.8 48 1.8 559 20.8 4.2 2.4 1.8 9.6 6.6 0.3 <0.1 <0.1 0.3 0.5 0.5 <0.1 <0.1 1.1 480 14.8 251 7.8 229 7.1 2595 80.1 1451 44.8 209 6.5 22 0.7 84 2.6 222 6.9 15 0.5 33 1.0 7 0.2 552 17.0 <0.1 0.1 3.6 2.2 32.3 28.4 1.3 <0.1 0.2 0.6 0.1 <0.1 <0.1 1.6 3238 84 Great Basin Naturalist Vol. 45, No. 1 T.\BLE 2. Availability (N/lOO trap days) and utilization of major prey species on the INEL, 1976 and 1977. 1976 1977 % Biomass in diet Species Spring Fall Spring Fall 1976 1977 Peromyscus maniculahis Perognathus parvus Dipodomifs ordi 64 4 17 87 9 33 .56 6 25 26 1 16 3.0 0.3 3.7 1.4 1.9 1.6 Total 8.6 3.3 crickets on the INEL during this period, but their biomass in the diet increased from 6.6% in 1976 to 28.4% in 1977. This food source served to buffer the effects of a reduction in rodent density in 1977. Jerusalem crickets were an important dietary component of this nesting population in an earlier study (Glea- son and Craig 1977), as well as that of other nesting populations (Thomsen 1971, Green 1983). We suggest that the density of Jerusa- lem crickets as well as that of rodents should be monitored in future investigations of nu- merical and functional responses of bur- rowing owl populations in the western United States. Acknowledgments We thank O. D. Markham, Department of Energy, INEL, for his encouragement and cooperation as well as Karen Gleason for field assistance. W. F. Barr, University of Ida- ho, identified the invertebrates in castings. This research was funded by the Division of Biomedics and Environmental Research, De- partment of Energy. We also thank L. C. Stoddart, Utah State University, who gener- ously shared his data on rodent abundance within the study area. Literature Cited BiTTs, K. O. 1971. Observations on the ecology of bur- rowing owls in western Oklahoma. A preliminary report. Proc. Okla. Acad. Sci. 51:66-74. Essk;, E. O. 1936. Insects of western North America. Macmillan, New York. Gleason. R. L.. and T. H. Crak;. 1979. Food habits of Inurowing ouls in southeastern Idaho. Great Ba- sin Nat. 39:274-276. Green, G. A. 1983. Ecologv of breeding liurrowing ou is in the Cohuiibia Basin, Oregon. Unpublished thesis. Oregon State Univ., Corvallis. 50 pp. Harniss, R. O., .and N. E. West. 1973. Vegetation pat- terns of the National Reactor Testing Station, southeastern Idaho. Northwest Sci. 47:.30-43. Henny. C. J., AND L. J. Blus. 1981. Artificial burrows provide new insight into burrowing owl nesting biology. Raptor Research 15:82-85. La Rivers, I. 1948. A svnopsis of Nevada Orthoptera. Amer. Midi. Nat. 39:652-720. Martin, D. J. 1973. Selected aspects of burrowing owl ecology and behavior. Gondor 75:446-456. Murray, G. A. 1976. Geographic variation in the clutch sizes of seven owl species. Auk 602-613. TiioxrsEN. L. 1971. Behavior and ecology of burrowing ouls on the Oakland Municipal .Airport. CxMidor 73:177-192, W'Enc.woon, [. A. 1976. Burrowing owls in south-central Saskatchewan. Blue Jay .34:26-44. NOTE ON THE DIET OF LONG-BILLED CURLEW CHICKS IN WESTERN IDAHO Roland L. Redmond' and Donald A. Jeiini' .\BSTa-\c T.— The diet of Long-billed Curlew chicks is described for the first time. Five insect orders and one arachnid order were identified from nine stomach contents samples. Grasshoppers and carabid beetles were domi- nant pre\- items. Chicks of precocial, nidifugous shorebirds are generally difficult to study because they are very mobile and hard to locate. As part of a larger study of Long-billed Curlew {Nii- meniiis americanus) behavioral ecology, we collected preliminary data on chick food habits. Owing to the difficulty of obtaining samples without sacrificing any chicks, the data are limited. Nonetheless, they are unique for Long-billed Curlews and rare for any nidifugous species. The study area was a short-grass rangeland (ca 21,000 ha) lying between the Payette, Boise, and Snake River valleys in western Idaho (Redmond et al. 1981). We collected stomachs from six recently depredated chicks (aged 14-46 days) during June and July 1978. These were stored frozen until the contents could be removed in the lab. In addition, we took samples of stomach contents from three live chicks weighing at least 300 g (aged 34-44 days) during July in 1977 and 1978. As an emetic, we introduced a 1% solution of antimony potassium tartarate directly into the proventriculus (0.4 cc/100 g body mass, modified after Prys-Jones et al. 1974). Chicks were then held in a closed box until they regurgitated a sample (ca 10 min). Prior to release, each chick was fitted with a radio transmitter, and its subsequent movements and growth monitored for a minimum of four days posttreatment. All chicks survived this period with no visible ill effects from the emetic. Stomach contents were placed in 70% ETOH and then sorted according to tax- on using a dissecting microscope, reference collections, and appropriate keys. We identified five insect orders and one arachnid order from the nine stomach con- tents samples. Orthoptera were found in all nine, Coleoptera in eight, and Hemiptera, Lepidoptera (larvae), and Arachnida were found in just one each. Grasshoppers were the only orthopterans and carabid beetles the only coleopterans that we identified. As such, these two groups appeared to be important prey for Long-billed Curlew chicks. Grasshopper eruptions on the study area generally began in late May, coincident with the annual peak of Long-billed Curlew hatching, and continued into August. By late June each year, grasshoppers were locally abundant, and they became more con- spicuous through July. Given the collection times of our stomach contents samples, the dominance of grasshoppers was expected. A similar prevalence of carabid beetles, how- ever, was surprising. These beetles appeared to be far more abundant earlier in the season (April-May). We suspect that, although grasshoppers might have been more numer- ous in June and July, their mobility reduced their overall vulnerability to Long-billed Curlew chicks. Conversely, a relatively slow- moving black beetle (Barrs 1979), once en- countered, would be easy prey. Because our ability to identify all material from the stomachs was limited by the nature of the samples (see Custer and Pitelka 1975), our data almost certainly underestimate the diversity of prey types taken by Long-billed Curlew chicks. We conclude that these chicks, like their parents (Sadler and Maher 1976, Bicak pers. comm.), take a wide variety of prey according to what is most available and vulnerable on the rangeland. We thank George Kirsch for identifying material from stomach contents and Tom Bi- cak for assistance with field work as well as 'Department of Zoology, University of Mc 85 Great Basin Naturalist Vol. 45, No. 1 comments on the manuscript. Research was supported by the U.S. Department of Interi- or Bureau of Land Management, through Contract YA-512-CT7-5 awarded to D. A. Jermi. Literature Cited B\w^s M. A. 1979. Patterns of movement of radioactive ' carabid beetles. Oecologia 44:125-140. Custer T. W.. a.nd F. A. Pitelka. 1975. Correction fac- tors for digestion rates for prey taken In Snow Buntings {Plectwphenax nivalis). Condor 77:210-212. PrYS-IoNES, R. p., L, SCHIFFERLK .^ND D, VV . MacDonald. 1974. The use of an emetic m ob- taining food samples from passerines. Ibis 116:90-94. Redmond, R. L.. T. K. B.cak, and D. A. Jenni. 1981. An evaluation of breeding season census techniques for Long-billed Curlews {Numenius americanus). Stud, .\vian Biol. 6:197-201. Sadler D. A. R., and W. J. Maher. 1976. Notes on the Long-billed Curlew in Saskatchewan. Auk 9.3:382-384. TUNDRA VEGETATION OF THREE CIRQUE BASINS IN THE NORTHERN SAN JUAN MOUNTAINS, COLORADO Mary Lou RottmaiV and Emily L. Haitman' Abstract.— The vegetation of three alpine cirque basins in the northern San Juan Mountains of southwestern Col- orado was inventoried and analyzed for the degree of specificity shown by vascular plant communities for certain types of habitats identified as representative of the basins. A total of 197 vascular plant species representing .31 fami- lies was inventoried. Growth forms of all species were noted and a growth form spectrum for all of the communities was derived. The caespitose monocot and erect dicot growth forms are the most important growth forms among the communitv dominants. The most common growth form among all species is the rosette dicot. Since the early 1900s a number of ecologi- cal studies have been undertaken in the al- pine tundra of the Front Range of Colorado (Cooper 1908, Holm 1923, Cox 1933, Osburn 1958, Marr 1961, Willard 1963, Komarkova 1976, Flock 1978, May and Webber 1982). However, the alpine tundra of central and southwestern Colorado remains relatively un- known (Langenheim 1962, Johnson 1969). A floristic study was done by Michener (1964) in the subalpine and alpine zones of the Needle Mountains in the southern San Juan Moimtains. Another major study in the south- em San Juan Mountains, an analysis of snow- pack augmentation by Steinhoff and Ives (1976), included the alpine zone as well as the forest ecosystems below. There is no pub- Hshed botanical work on the alpine vegeta- tion of the northern San Juan Mountains. In this study the vegetation of three alpine cirque basins in the northern San Juans was inventoried and analyzed for the degree of specificity shown by individual species and aggregations of species for certain types of habitats. Study Area The San Juan Mountains are a dis- continuous section of the Southern Rocky Mountains situated along the Continental Di- vide in southwestern Colorado. They are lo- cated between 106 and 108 degrees west lon- gitude and 36 degrees 30 minutes and 38 degrees 15 minutes north latitude (Atwood and Mather 1932). Sharp pinnacles, rounded crests, serrate ridges, and broad upland ero- sional surfaces characterize the alpine zone of these mountains. The elevation of the San Juans ranges from 1524 m in the southwest corner to 4358 m at the summit of Uncom- pahgre Peak. They are composed largely of Tertiary volcanic tuffs and lavas that lie un- conformably over metamorphic sedimentary and volcanic intrusive rocks of Precambrian age as well as sediments of Paleozoic, Meso- zoic, and early Cenozoic age (Casadwall and Ohmoto 1977). Broad regional ice fields and transection glaciers occurred during the Pleistocene, producing cirques, basins, tarns, hanging valleys, and broad U-shaped valleys. Today periglacial features such as active pat- terned ground, active rock glaciers, and per- sistent ponds indicate the occurrence of spo- radic or discontinuous permafrost (Ives and Fahey 1971, Barsch 1978). Three alpine cirque basins, representative of the northern San Juan Mountain tundra, were studied. American Basin, Hinsdale County, is characterized by a well-developed moist turf mantle interrupted by areas of bedrock outcrops, talus deposits, and pat- terned ground features. The vegetation in this basin reflects a more moist climatic re- gime than in the other two basins, as evi- denced by the predominance of moist mead- ows, absence of dry meadows, and minimal occurrence of fell-fields. The elevational range of this basin is 3536-3962 m. 'Department of Biology, University of Colorado, Denv , Colorado 80202. 87 Great Basin Naturalist Vol. 45, No. 1 Burns Basin, San Juan County, has a north- west-southeast orientation. The convex slopes forming the perimeter of the main ba- sin present an interesting contrast of moisture regimes. The southwest-facing slope is char- acterized by a dry turf alternating with fell- fields and imvegetated talus. The northwest- facing slope consists of a series of tiers of massive moist and wet ledges with adjacent moist meadows. Midsection of the basin con- sists of moist and wet meadows. The eleva- tional range of this basin is 3634-3932 m. Both Bums and American basins have rock glacier complexes composed of tongue and lobate units. Stony Basin, San Juan County, is formed of three broad turf-mantled steps, each sepa- rated by a bedrock escarpment. Because of its topographic position, the basin is contin- ually buffeted by strong wind resulting in a more severe climatic regime than is found in the other two basins. Islands of dry meadows and fell-fields interrupt the moist meadow of the upper two steps of the basin. The latter have an abundance of frost-associated fea- tures including frost boils, patterned ground, ephemeral ponds, and rock debris islands. The lower step is characterized by a shallow lake and hummocky wet meadow. The eleva- tional range of Stony Basin is 3764-3926 m. Methods Three field seasons, 1981-1983, were in- volved in the study. One hundred ninety-nine habitats and their associated vascular plant commimities were sampled. The abundance of each species was estimated using the stan- dard abimdance classes of Daubenmire (1968). As data from multiple samples of a particular habitat type accumulated, the in- ventoried species and their abundance ratings were analyzed on the basis of constancy of occurrence between samples (Mueller-Dom- bois and Ellenberg 1974). The terms used in describing the species within a community, dominant, secondary, frequent, occasional, and rare, are defined as follows. A dominant species is very abimdant (abundance class) and occurs with a high constancy (80.1%- 100%). A frequent species is frequent in abundance and has an intermediate con- stancy (40.1%- 60%). An occasional species occurs in scattered instances, and its presence or absence is inconsequential to the character of the commmiity in which it may occur. It is occasional in abundance and has a low con- stancy (20.1%-40%). A rare species is both rare in abundance and rare in constancy (1.0%-20%). The term habitat type in this study refers to all of the area (sum of discrete units) that support one plant community. A habitat type may be considered as the abiotic equivalent of the stand-type in that the habitat type is a synthetic unit whose characteristics are ob- tained by combining all the samples of a par- ticular habitat. The habitat types defined are: dry ledge, moist-wet ledge, rock crevice, talus slope, patterned ground, fell-field, dry meadow, moist meadow, wet meadow, shrub tundra, and krummholz. Nomenclature fol- lows Kartesz and Kartesz (1980). Results In the following description of habitat types and associated communities, only the dominant and secondary or more important species are listed. Ledge Habitat Type The rock ledge habitat consists of the bed- rock outcrops which, as a result of jointing and weathering, form benches and channels where windborne fines are deposited and or- ganic debris is accumulated to form a sub- strate for vegetation. The orientation of the bedrock and amount of protection provided by rock overhang are controlling factors in the microenvironment of this habitat type (Younkin 1970). For purposes of vegetation analysis, the ledge habitat is divided along a moisture gradient and includes dry ledges and moist-wet ledges. Two communities are found in the dry ledge habitat and three in the moist-wet ledge habitat. Geinn rossii var. turbinattim and Selagi- nella densa are frequent species in one of the dry ledge communities. Other important spe- cies include Aquilegia coerulea, Cystopteris fmgilis, Draha crassa, PotcntiUa subjuga var. subjiiga, Silene acaulis var. subacauUs, and Smelowskia cah/rina. The other dry ledge January 1985 ROTTMAN, HaRTMAN: COLORADO TuNDRA VEGETATION 89 community is characterized by Carex hay- deniana, Cerastiitm earlei, Dmba crassa, Fes- tuca brachyphyUa, Potentilla nivea, P. rubri- caulis, Saxifraga fkigellaris ssp. platysepala, and Trifolium naniim. One of the moist-wet ledge communities is dominated by Geiim rossii var. turbinatum, with Carex albonigra, C. heteroneura var. chalciolepis, Erigeron melanocephahis, and Saxifraga cespitosa ssp. delicatiila as impor- tant associates. Another moist-wet ledge commimity has Salix reticulata ssp. nivalis as dominant. Associated with this species are Carex albonigra, Salix arctica, Saxifraga ad- scendens ssp. oregonensis, and S. debilis. Ledges with snowmelt water running through fissures in the rock constitute the third possible moist-wet ledge habitat. These are dominated by Cardamine cordifolia at the base of the ledges and Salix reticulata ssp. ni- valis on the benches. Caltha leptosepala, Carex nova. Primula parryi, Sediim rhodan- thum, and Sibbaldia procumbens are present in this community. Rock Crevice Habitat Type Tliis is a restrictive habitat type that only includes soil-filled crevices on solitary boul- ders or contained crevices on rock outcrops or headwalls in the basins. The community concept is loosely applied to the species in these habitats. The only relationship that the singly-occurring species have to one another is their presence on the same boulder or bed- rock substrate. Certain species are frequently found in rock crevices: Androsace septentrio- mdis, Artemisia scopulorum, Claytonia me- garhiza, Draba crassa, D. crassifolia, Festuca brachyphyUa, and Oreoxis bakeri. Talus Slope Habitat Type Extensive talus slopes occur throughout the basins. White (1981) defines talus as an accu- mulation of rock debris of various sizes trans- ported from the source areas by gravity, rain- wash, snowmelt water, or snow avalanching to a site below. Fine material may be present in the interstices of the rock material; these fines provide a suitable substrate for vegeta- tion. Although dominance within the talus slope commimity is poorly defined, certain species show a high constancy for this habi- tat: Angelica grayi, Aquilegia coerulea, Ceras- tium earlei, Claytonia megarhiza, Polemo- nium viscosum, Senecio amplectens var. am- plectens, S. amplectens var. holmii, S. soldanella, and S. werneriifolitis. Patterned Ground Habitat Type Washburn (1956) defines patterned ground as, "a group term for the more or less sym- metrical forms, such as circles, polygons, nets, steps and stripes, that are characteristic of, but not necessarily confined to, a mantle subject to intensive frost action." The forms of patterned ground found in the study basins are nonsorted circles (frost boils and frost hummocks), sorted polygons, and debris is- lands. The centers of the sorted forms have from little to no vegetation because of the as- sociated frost action that keeps the soil suffi- ciently disturbed to prevent plant coloniza- tion (Johnson and Billings 1962). The communities found in patterned ground habi- tats vary with the specific type of patterned ground form. Frost boils are another example of a habitat for which the community con- cept must be loosely applied. Several species are repeatedly found associated with frost boils: Cerastitim earlei, Draba crassifolia, Geum rossii var. turbinatum, Oreoxis bakeri, and Stellaria umbellata. Frost hummock areas are traversed by water-filled channels in the wet meadows. The dominant species on the elevated portion of the hummock is Carex nigricans. Secondary species are Carex pseudoscirpoidea and Festuca brachyphyUa. Dominant species in the saturated areas at the base of the hummocks are Carex aquatilis and C. vernacula. Caltha leptosepala and Carex nova are secondary species in these areas. Sorted polygons are characterized by lichenized rocks in the borders. The central fines of the polygons support a Carex nigri- cans-Sibbaldia procumbens community. Fre- quent associates in this community are Arte- misia scopulorum, Erigeron melanocephalus, and Juncus drummondii. Debris islands, a sorted form of patterned ground, occur as re- petitive units on talus debris. The vegetation on these islands suggests a successional devel- opment from talus slope to meadow. Depend- ing upon the serai stage of development. 90 Great Basin Naturalist Vol. 45, No. 1 dominant species may be either Salix reti- culata ssp. nivalis and Silene acaulis var. sub- acaulis, or Senecio amplectens var. holmii and S. soldanella. Fell-field Habitat Type The fell-field habitat type is characterized by a high proportion of weathered rock ma- terial. Soils are coarse textured, with little or- ganic material and only rudimentary profiles. Fell-fields occur on windward sites, with little or no snow cover, thus exposing the plants and soil to severe dessication. The longest growing season in the tundra occurs in this habitat type. The high diversity found in fell-field communities is contributed pri- marily by the frequent and occasional spe- cies. One of the fell-field communities is dominated by Silene acaulis var. subacaulis and Geum rossi var. tiirbinatum, with Minu- artia obtusiloba, Potentilla diversi folia, Se- laginella densa, and Trifolinm nanum as sec- ondary species. Another fell-field community has Carex elynoides and Geum rossii var. tur- binatum as dominants. Secondary species in- clude Festuca brachyphylla, Minuartia ob- tusiloba, Selaginella densa, and Trifolium nanum. Carex elynoides and Trifolium na- num are dominants in the fell-field commu- nity occurring on slopes and ridge tops. Asso- ciated with these species are Festuca brachyphylla, Geum rossii var. turbinatum, and Silene acaidis var. subacaulis. Dry Meadow Habitat Type The dry meadow habitat type occurs on exposed windy slopes high in the basins, where strong winds create snow-free condi- tions throughout much of the winter. The vegetation in these sites reaches anthesis early in the season, thus completing the growth cycle before vegetation in more pro- tected areas reaches maturity. The most ex- tensive dry meadow community is dominated by Carex elynoides. Secondary species in- clude: Festuca brachyphylla, Geum rossi var. ttirbinatum, Hymenoxys grandiflora, Poa rupicola, and Trisetum spicatum. A minor and highly restricted dry meadow commu- nity is dominated by Kobresia myosuroides. Carex ebenea, C. heteroneura var. chalciole- pis, C. pseudoscirpoidea, Luzula spicata, and Trisetum .spicatum are secondary species in this commmiity. Moist Meadow Habitat Type This is perhaps the most widespread habi- tat type in the study basins and is most repre- sentative of the tundra in the northern San Juan Mountains. The moist meadows are situ- ated on the lee slopes and in topographical concavities protected from the winter cli- mate by snow accumulation that may remain until mid-July. The moist meadow may be re- garded as a complex of several communities, each with a distinct spatial occurrence within the complex. A Deschampsia caespi- tosa-Geum rossii var. turbinatum community is found in lower sites on basin slopes and in concavities. Associated with these species are: Artemisia scopulorum, Carex albonigra, C. nova, Oreoxis bakeri. Polygonum bistor- toides, and Saxifraga rhomboidea. In flat areas at midslope, a Carex nigricans- Sibbaldia procumbens community occurs with Carex vernacula, Erigeron melanoce- phalus, Juncus drummondii, Oreoxis bakeri. Polygonum viviparum, and Ranunculus ma- cauleyi as secondary species. The third com- munity in the complex, dominated by Salix reticulata ssp. nivalis, is present on the high- est moist meadow sites in the basins. Second- ary species include Artemisia scopulorum, Erigeron simplex, Salix arctica, Sibbaldia pro- cumbens, and Sile7ie acaulis var. subacaulis. Wet Meadow Habitat Type Wet meadows are situated on relatively flat surfaces below late-lying snowbanks, in catchment areas in the basins, and adjacent to ponds, lakes, and streams. Frequently dis- sected by rivulets, these areas are often asso- ciated with the presence of sporadic per- mafrost (Johnson and Billings 1962, Ives 1974). As a result of permafrost in the sub- strate and the runoff, the wet meadows are saturated throughout the growing season. Where shallow, standing water is present, a Caltha leptosepala-Cardamine cordi folia- dominated community occurs. The secondary species are Carex nova, Juncus drummondii, Pedicularis groenlandica. Primula parryi, Se- dum integrifolium, and Trifolium parryi. A January 1985 ROTTMAN, HaRTMAN: COLORADO TuNDRA VEGETATION 91 second commimity, dominated by Caltha lep- tosepaki and Juncus dmmmondii, is charac- teristic of better drained sites. Associated spe- cies include: Carex aqiiatilis, Festuca brachij- phijlla, Geiirn rossii var. tiirbinattnn, Primida parryi, Sibbaldia prociimbens, and TrifoHum parnji. Shrub Tundra Habitat Type The shrub tundra habitat type is made up of shrub thickets of Salix brachijcarpa or S. planifolia and associated vegetation. A minor constituent of the basins in this study, this habitat type is hmited to moist depressions and drainage areas. The moist areas are dom- inated by SaJix planifolia. Secondary species in this commimity are: Carex ebenea, C. het- eroneura var. chalciolepis, C. nova, C. pseu- doscirpoidea, and Geum rossii var. turbina- tum. A Salix brachycarpa-dominiLted community occurs on well-drained slopes. A drier environment is reflected in the associ- ated species: PhaceUa sericea, Polemonium viscosiim, Pontentilla diversifolia, Sedum lan- ceolatiim, and Trifoliinn namim. Krummholz Habitat Type Timberline elevations range from approx- imately 3535 to 3720 m in the study basins. The ecotonal area characteristically has rep- resentative species from both the alpine and subalpine zones. Krummholz conifer species, Abies lasiocarpa and Picea engehnannii, are dominant and exert a primary influence on the svirrounding environment and vegetation. The associated species are highly variable from one krummholz habitat to another; however, a list of the more frequent species serves to illustrate the ecotonal nature of the commimity: Aquilegia coerulea. Arnica cordi- folia, Dugaldia hoopesii, Minuartia obtusi- loba. Polygonum bistortoides, Ribes mon- tigeniim, Silene acaulis var. siibacaulis, and Thalictnim fendleri. Discussion of protection from wind, amount and nature of weathering, debris transport, and slope as- pect determine the occurrence of specifc habitat types and associated communities in terms of the dominant and secondary species. Minor variations are seen in the occurrence of occasional species. Rare species have a higher occurrence in the rock-predominating habitats, where competition appears to be less than in a closed turf meadow. A consistent tendency toward dwarfing of species is noted in the Carex nigricans- Sibbaldia prociimbens and Salix reticulata ssp. niua/is-dominated moist meadow com- munities. This appears to correlate with a pattern of late snow release. As noted by Owen (1976), plants of the same species un- der conditions of higher elevation and later snownielt mature and flower at a smaller size than plants not under these conditions. The Kobresia myosuroides-dominsLted dry meadow, which has long been recognized as the climatic climax of the Front Range (Cox 1933, Osburn 1958, Bamberg 1961, Marr 1961, Willard 1963), is highly restricted in its occurrence in the northern San Juan Moun- tains and is replaced in importance by a Carex elynoides-domiimted dry meadow community. Carex Indicators Some members of the Cyperaceae, Carex albonigra, C. arapahoensis, C. heteroneiira var. chalciolepis, and C. pseudoscirpoidea, have rather broad ecological tolerances that enable them to grow in both dry and moist habitats. Carex nova and C. vernacula occur in moist as well as wet habitats. Other Carex species are more specific in their moisture re- quirements or tolerances and are useful in- dicators of substrate moisture conditions. In- dicators of dry substrates are: Carex elynoides, C. perglobosa, and C. phaeoceph- ala. Indicators of moist substrates are: C. nel- sonii, C. nigricans, C. norvegica, C. pyre- naica, and C. nardina var. hepburnii. Carex aqiiatilis is an indicator of wet substrates of- ten characterized by standing water. Species Growth Forms A total of 197 vascular plant species repre- sentative of 31 families was inventoried in the study basins (Rottman 1984). Basin orien- Each species found in the study basins was tation, variability of moisture regimes, degree assigned to a growth form category. The 92 Great Basin Naturalist Vol. 45, No. 1 growth forms of May and Webber (1982) were used. Based primarily on the nature of the shoot habit, these growth forms include: caespitose monocot, single-shooted monocot, erect dicot, rosette dicot, mat dicot, cushion dicot, erect shnib, and dwarf shrub. The caespitose monocot is a tufted grami- noid growth form. Graminoid sods are able to modify tlie microenvironment to a greater extent than any other growth form or pattern of spatial distribution (Billings 1974). When compared to other growth forms, the per- centage of caespitose monocot species is rela- tively small (9%); however, this growth form contributes the greatest percentage of domi- nants in the commimities analyzed. The single-shooted monocot growth form is represented primarily by grasses and sedges (25 species, 12% of total). Although this cate- gory has more species than the caespitose monocot, it is never dominant within a commimity. The highest percentage of species (38%) and the second highest percentage of domi- nants in the communities are erect dicots. Since both the root and shoot systems of erect dicot plants require less space for later- al spread, this growth form is compatible with extremely rocky habitats where only a minimal amount of soil is available. The importance of the rosette dicot growth form (26%) is reflected in the fact that approximately 70% of the species with ubiquitous occurrences in the study basins are in this category. This growth form is found in all the habitat types studied and appears to be equally abundant in both meadow and rock habitats. The mat (6%) and cushion dicot (2%) growth forms are minor in occurrence. Both of these growth forms are considerably more important in tundras to the north, where their greater occurrence is correlated with an increased wind factor. The erect and dwarf shnib growth forms are another minor seg- ment of the growth form spectrum account- ing for a combined 7% of the total species. Community Specificity The specificity of communities for particu- lar haliitat types was evaluated on the basis of conununity dominants. Although habitat types are physically discrete and recogniz- able, a one-habitat type/one-commimity con- cept is not applicable. A similar finding is re- ported by Douglas (1972). It was found that virtually all habitat types have more than one potential community with different domi- nants and that some dominants are repetitive, by themselves or in combination, in different habitat types. Literature Cited Atwood. W. a., and K. F. Mather. 1932. Physiography and Quaternary geology of the San Juan Moun- tains. Colorado'. U.S. Ge'ol. Surv. Prof. Pap. 166. Ba.mberc;, S. a. 1961. Plant ecology of alpine tundra areas in Montana and adjacent Wyoming. Un- published thesis. Univ. of Colorado, Boulder. 163 pp. Bars( H. D. 1978. Rock glaciers as indicators for dis- continuous alpine permafrost. An example from the Swiss Alps. Pages 349-352 in National Re- search Council of Canada, Ottawa, Proceedings of the Third International Conference on Permafrost. BiLLi.N(;s, E. D. 1974. .\rctic and alpine vegetation: plant adaptations to cold summer climates. Pages j 404-443 in J. D. Ives and R. C. Barry, eds., Arc- tic and alpine environments. Methuen, London. Casadvall, T., and H. Ohmoto. 1977. Sunnyside Mine, Eureka Mining District, San Juan County, Colo- rado: geochemistry of gold and base metal ore deposition in a volcanic environment. Economic Geology 72:1285-1.320. Cooper, W. S. 1908. Alpine vegetation in the vicinit\ of Longs Peak. Bot. Gaz. 45:319-337. Cox, C. F. 1933. .Alpine plant succession on James Peak, Colorado. Ecol. Monog. 3:299-372. Daubenmire, R. 1968. Plant communities: a textbook of plant svnecologv. Harper and Row, New York. 300 pp.' DoicLAs, G. W. 1972. Subalpine plant comnumities of the western North Cascades, Washington, .\rctic and .\lpine Research 4:147-166. Fi.ocK, J. W. 1978. Lichen-bryophyte distribution along a snow-cover soil-moisture gradient, Niwot Ridge, Colorado, .\rctic and .\lpino Research 10:31-47. Hoi.M, T. 1923. The vegetation of the alpine region of the Rocky Mountains in Colorado. Memoirs of the National Academv of Sciences 19:1-45. IvEs, J. D. 1974. Permafrost.' Pages 159-194 in J. D. Ives and R. G. Barry, eds., Arctic and aliiinc Environ- ments, .Methuen, London. IvEs, J. D,. AND B. D. Fahey. 1971. Permafrost occur- rence in the Front Range, C^olorado Rocky Moun- tains, United States of .Vnu'iica. J. Glaciologv 10:105-111. Johnson, K. L. 1969. Alpine vegetation and soils of Mesa Seco Plateau, San Juan Moimtains, Colorado. Un- published dissertation. Univ. of Illinois. 225 pp. Johnson. P. L.. and W. D. Bh.i.incs. 1962. The alpine vegetation ot tlic Be.utooth Plateau in relation to January 1985 ROTTMAN, HaRTMAN: COLORADO TuNDRA VEGETATION 93 cryopedogenic processes and patterns. Ecol. Monog. 32:105-135. Kartesz, J. T., AND R. Kabtesz. 1980. A synonym ized checklist of the vascular flora of the United States, Canada, and Greenland. Volume II. The biota of North America. Univ. of North Carolina Press, Chapel Hill. 498 pp. KoMARKOVA, V. 1976. Alpine vegetation of the Indian Peaks area, Colorado Rocky Moimtains. Unpub- lished dissertation. Univ. of Colorado, Boulder. 655 pp. Lan(;enheim, J. H. 1962. Vegetation and environmental patterns in the Crested Butte area, Gunnison County, Colorado. Ecol. Monog. 32:249-285. Mark, J. W. 1961. Ecosystems of the East Slope of the Front Range in Colorado. Univ. of Colorado Studies, Series in Biology 8:1-134. May, D. E., and P. J. Webber. 1982. Spatial and tem- poral variation of the vegetation in its productiv- ity on Niwot Ridge, Colorado. Pages 35-62 in J. C. Halfpenny, ed.. Ecological studies in the Colo- rado Alpine: a festschrift for John W. IVlarr. Univ. of Colorado Institute of .\rctic and Alpine Re- search. Occasional Paper 37. MicHENER, M. J. 1964. The high altitude vegetation of the Needle Mountains of southwestern Colorado. Unpublished thesis. I'niv. of Colorado, Boulder. 77 pp. Mueller-Dombois, D., and H. Ellenberg. 1974. Aims and methods of vegetation ecology. John Wiley, New York. 547 pp. OsBURN, W. S., Jr. 1958. Ecology of winter snow-free areas of the alpine tundra of Niwot Ridge, Boul- der County, Colorado. Unpublished dissertation. Univ. of Colorado, Boulder. 76 pp. Owen, H. E. 1976. Phenological development of her- baceous plants in relationship to snowmelt date. Pages 323-342 in H. W. Steinhoff and J. D. Ives, eds.. Ecological impacts of snowpack augmenta- tion in the San Juan Mountains, Colorado. San Juan Ecology Project, Final Report, Fort Collins, Colorado State Univ. Publications. RoTTMAN, M. L. 1984. A floristic analysis of three alpine basins in the northern San Juan iMountains, Colo- rado. Unpublished dissertation. Univ. of Colo- rado, Boulder. 311 pp. Steinhoff, H. W., and J. D. Ives. 1976. Ecological im- pact of snowpack augmentation in the San Juan Mountains, Colorado. San Juan Ecology Project, Final Report, Fort Collins, Colorado State Univ. Publications. 489 pp. Washburn, A. L. 1956. Unusual patterned grovmd in Greenland. Geo). Soc. of America Bull. 67:823-865. White, S. E. 1981. Alpine mass movement forms (non- catastrophic) classification, description and sig- nificance. Arctic and Alpine Research 13:127-1.37. Willard, B. E. 1963. Phytosociology of the alpine tun- dra of Trail Ridge, Rocky Mountain National Park, Colorado. Unpublished dissertation. Univ. of Colorado, Boidder. 245 pp. YouNKiN, W. E., Jr. 1970. A study of the vegetation of alpine rock outcrops in northern Colorado. Un- published thesis. Colorado State Univ., Fort Col- lins. 109 pp. USE OF BIOMASS PREDICTED BY REGRESSION FROM COVER ESTIMATES TO COMPARE VEGETATIONAL SIMILARITY OF SAGEBRUSH-GRASS SITES L. David Hiiniphiey' Abstract.— Regressions between vegetational cover, estimated with a two-tiered, gridded sampling frame, and biomass were used to obtain predicted biomass values from cover values. Comparisons of eight sites based on pre- dicted bioma.ss data and comparisons of the sites based on cover data indicate that predicted biomass data may better identify differences among sites. Also, some suggestions are made regarding the methods of using cover-bio- niass regressions to obtain predicted biomass values. Biomass data are generally considered to most precisely represent the relative impor- tance of species in a community (Whittaker 1975). However, biomass sampling is often time consuming, especially when sampling a large number of plots is required. As an alter- native, many methods of cover estimation have been used. Methods of weight estima- tion (Pechanec and Pickford 1937, Wilm et al. 1944, Tadmore et al. 1975) and relative weight estimation (Hutchings and Schmautz 1969) have also been used. However, weight estimation methods are inherently dependent on the skill of the observer and results may vary. Cook et al. (1948) estimated units of cover, determined biomass per unit, and ob- tained biomass estimates by multiplying mass per imit of cover by number of units of cover for a species. Payne (1974) presented linear regressions between cover and biomass for many herbaceous species. Anderson and Kothmann (1982) presented a method of cal- culating mass from estimated cover based on linear regression between cover and mass. Such methods can give data that approximate biomass data but require much less time than extensive biomass sampling. Following methods similar to those of An- derson and Kothmann (1982), I used regres- sions between biomass and cover to obtain predicted biomass values from cover values for species on eight sagebrush {Artemisia tri- clentata) -grass sites. Cover estimates u.sed in the regressions were obtained by a method that is more consistent and depends less on 'Biology Department, Idaho State University, Pocatello, Idaho 8:3209. Cm the skill of the observer than does weight es- timation or many other cover estimation methods. The use of a more precise and con- sistent cover estimation method should result in better cover-biomass regressions. The methods by which regressions were calcu- lated differed most notably from those of An- derson and Kothmann (1982) in that sizes of biomass samples used in my regressions were representative of the range of sizes of cover values commonly encountered in cover sam- pling. Also, some suggestions are made re- garding the methods of using cover-biomass regressions to obtain predicted biomass val- ues. Using the predicted biomass values for these eight sites, I present an assessment of advantages of predicted biomass data over cover in comparing vegetation among sites. I compared vegetational similarity among the eight sites based on predicted biomass values to vegetational similarity among the same sites based on cover values. This comparison indicated that predicted biomass values may often better identify differences among sites than can cover data. Mi The results presented here are based on data collected on eight sites in southeastern Idaho for a study of postfire succession in sagebrush-grass areas (Humphrey in press). Each sample site consisted of a 100- x 50-m | plot. On each site, ten 50-m lines were estab- lished parallel to the 50-m axis of the site. t Address: 2998 S. Waterworks Koad, Biiford, Georgia 30518. 94 January 1985 Humphrey: Plant Ecology 95 These lines were chosen at random from a population of 100 lines that were at 1-m in- tervals along the 100-m axis. On each 50-m line, there were five sample points at 10-m intervals for a total of 50 sample plots on each site. A 1.0- X 0.5-m sampling frame consisting of two tiers, superimposed 10 cm apart and gridded off in 1-dm^ divisions, was used for cover estimation. The superimposed grids made it possible to sight vertically for esti- mating cover in each 1 dm^. The frame was placed at each sample plot and leveled by use of adjustable legs and level bulbs on the frame. Cover for each species was estimated to the nearest 1/4 dm" (or 1/8 dm" in the case of very small plants). This method is more precise and less dependent on skill or bias of the observer than estimation by cover class (e.g., Daubenmire 1959) or other esti- mation methods that use single-tiered sam- pling frames because the superimposed grids make it possible to more accurately sight each dm^. Floyd and Anderson (1982) de- scribed the use of a similar frame for cover estimation by point sighting. The point sight- ing method should also give reliable data and is less time consuming than the cover estima- tion method used in this study. On one of every 10 cover plots on each site, the current year's growth of each species was clipped, oven dried, and weighed to de- termine aboveground biomass. In this way, biomass samples of several sizes representa- tive of the range of sizes of cover samples commonly encountered was obtained. Thus, the regressions could be assumed to be valid for the entire range of cover values encoun- tered. (Alternatively, representative biomass samples for each species could be taken (An- derson and Kothmann 1982), but these sam- ples should encompass the range of sizes of cover samples commonly encountered.) Lin- ear regressions between cover and oven dry biomass were calculated from biomass values and corresponding cover values for the clipped plots, with biomass as the indepen- dent variable. (Biomass was chosen as the in- dependent variable because cover was con- sidered to be an estimated value that is dependent on the more precisely determined biomass.) Individual regressions were calcu- lated for single species or for groups of mor- phologically similar species using pooled data for the eight sites. An equation for inverse prediction from regression equations (Zar 1974) was used to obtain a predicted biomass value from each cover value for each species. If the Y intercept was significantly different from zero, a regression forced through the origin was done (Snedecor and Cochran 1980), and this regression was used to obtain predicted biomass values. Similarity index values for each site paired with each of the other sites (Bray and Curtis 1957) were calculated with both cover values and predicted biomass values. These sim- ilarity index values were used to compare the way these two data sets described differences among the eight sites. The sign test (Snedecor and Cochran 1980) was used to determine if the two sets of similarity index values were significantly different, and the nature of dif- ferences between the two sets of similarity index values was examined. Results and Discussion Good cover-biomass regressions were ob- tained for several individual species and for several groups of two or more species that were judged to be morphologically similar (Table 1). Some of these groups are obviously morphologically similar, such as the group Chnjsothamnus viscidiflorus (green rabbit- brush), C. nauseosus (gray rabbitbrush), and Giitierreza sarothrae (broom snakeweed), the group Artemisia tridentata (big sagebrush), A. tripartita (threetip sagebrush), and the group Balsamorhiza sagittata (arrowleaf balsam- root), B. macrophylla (bigleaf balsamroot), and Wyethia amplexicaulis (mule's ear). Other groups are perhaps less obviously sim- ilar, although they are similar enough to pro- vide good cover-biomass regressions. The an- nual species in group 8 in Table 1 are morphologically similar in that they are all small, slender annuals. Agropyron dasysta- chyum (thickspiked wheatgrass) and A. spica- turn (blue-bunch wheatgrass) were grouped together because, on these sites, A. dasysta- chyiim usually grew in quite dense stands, making its growth habit similar to that of A. spicatum. If desired to reduce the number of regressions calculated, or the number of bio- mass samples taken, it appears that single 96 Great Basin Naturalist Vol. 45, No. 1 regressions can often be done for groups of species similar in morphological character- istics such as height, leaf size, and leaf den- sity. Relationships between cover and biomass for the same species may vary with differ- ences in site characteristics such as produc- tivity, mesicness, and other species present. But a regression should be consistent for the same species on other sites in addition to those sites where biomass samples were ob- tained, if the data used to calculate the re- gression cover the range of differences in site characteristics of those other sites, and if the regression explains much of the variance in Y (has a high r-^). Thus, if these conditions are met, the same regressions could be used for different sites and different studies, if the species are in roughly the same seasonal stage of development (for example, peak biomass) and the same cover estimation method is used. In this way, the amount of time-con- suming biomass sampling needed to obtain predicted biomass values can be reduced even further. Regressions have been used in a similar manner to obtain biomass estimates for trees (Whittaker 1966, Dabel and Day 1977). The percent similarity values among sites based on predicted biomass data were signifi- cantly different (at the 1% level) from those based on cover data by sign test comparison. Predicted biomass data indicated lower sim- ilarity among sites, i.e., greater distinctions among sites over all. The difference between the medians of the two sets of similarity in- dex values was rather small, but the max- imum and minimum values of the two sets show that predicted biomass data also in- dicated greater extremes of high and low sim- Table 1. Results of cover-biomass regressions for species or groups of iiiorphologicall\- similar species. Co\er data was in cnr; biomass data was in g. For species or species groups where the Y intercept of the regression was signifi- cantly different from zero, the results of the regression forced through the origin are presented. Slopes of all regre.s- sions listed are significant (P < 0.05). Y is cover; X is biomass (independent variable). (Nomenclature of all plants follows Hitchcock and Cronquist 1973.) Group Species Regre.ssion J.2 n 1 Phlox lonoifoha Y = 0.30 X (forced through origin) 0,80 9 2 LitJios]>cnniim nulcnilc. Helianthclhi uiufloia Y = ()..339 -1- 0.197 X 0,94 14 3 Crepis aciiiiiinata. Achillea luillcfoliiini Y = -0.158 + 0.632 X 0,91 11 4 Balsamorh iz(i sagittciUi, B. macwphi/lla, Wi/cthid (iniplexicaiilis Y = 0.083 + 0.357 X 0,99 10 5 Penstcnwu sp)). Y = 0.844 + 0..325 X 0.78 13 6 Tragopogon diihiiis Y = 0.184 + 0.517 X 0.84 8 7 Lupinus spp. Y = 0.272 + 0.286 X 0.92 7 8 Poh/gonum doughisii, Caiiophijtuni cliff iisiuii. ('olloniid liucdiis, LapptiUi rcdoiLskii Y = 1.34 X (forced througii origin) 0.94 14 9 . Cirsiiiui (irifusc Y = 0.789 + 0.114 X 0.95 8 10 Hnninis tcctoniiii Y = 1.024 X (forced tiirough t)rigin) 0.86 13 1 1 Agropnion ildsii^ldcln/iiiii. Y = 0.114 + 0,488" X 0.80 23 A. s-picdliiiii 12 Poo ncvddensis, P. sandbergii Y = 0,188 + 0.186 X 0.77 14 13 Pod pratcnsis V = 0.597 + 0.439 X 0.95 7 14 A rtc'Di i.sid Iriden Uita, A. tripdititd Y = 2.744 -1- (),185X 0.77 17 15 Chnjsothdmnus viscidiflorus. C. naiiseosiis, Cuticrrczd soroth rdc Y = 0,294 + 0,.366 X 0.92 16 16 Amclanchier dlnifolid, Sijmphoricdrpos orcoph ilti.s Y = 0,968 + 0,2.35 X 0.80 10 17 Piirshid tridentata Y = 0.275 + 0.235 X 0.99 8 January 1985 Humphrey: Plant Ecology 97 Table 2. The median and the niaxinuuii and inini- miiin values of the set of similarity index values based on cover and based on predicted biomass. Cover Predicted biomass Median: Maximum: Minimum: 34.1 56.4 13.5 32.1 58.8 10.8 ilarity among sites (Table 2). In other words, similar sites appeared more similar, and dis- similar sites appeared less similar. The differ- ence between each similarity index value based on predicted biomass and the corre- sponding value based on cover (Fig. 1) illus- trates more clearly that predicted biomass data indicated less similarity among sites over all and a greater range of similarities among sites. At low similarities, predicted biomass data indicated lower similarity between sites than did cover; at moderate similarities, it in- dicated lower similarity than did cover in most cases; while at relatively high sim- ilarities it tended to indicate greater sim- ilarity between sites than did cover. This study suggests that predicted biomass data may often describe distinctions among sites better than cover data. For differences to exist between comparisons of sites based on the two types of data, differences in the relationship of biomass to cover must exist among species, and the species with different ratios of biomass to cover must be unevenly distributed among sites. In this study, pre- dicted biomass data tended to indicate great- er differences among sites because predicted biomass data emphasized species with higher ratios of biomass to cover, and many species that varied more in abundance among sites had higher ratios of biomass to cover. These were tall herbaceous species (groups 2, 9 in Table 1) and shrubs (groups 14, 16, 17 in Table 1). Similarly, predicted biomass data indicated higher similarity than did cover be- tween some pairs of sites that were both dominated by the same species (of shrubs) that had higher ratios of biomass to cover. Predicted biomass data can provide more ac- curate information on the relative impor- tance of species in a commimity than can cover data alone, and it appears that this greater accuracy may often result in greater ability to identify differences among sites. «12 H Z < -I at 16 -1 1 1 1 DIFFERENCE (in units of "/o Sim.) Fig. 1. Differences between similarity index values based on cover and based on predicted biomass. Differ- ences between paired similarity index values (biomass - cover) arranged in order of decreasing percent similarity based on biomass data. Acknowledgments I thank J. E. Anderson and J. H. K. Pech- mann for reviewing drafts of the manuscript and Thelma Richardson for providing statis- tical advice. Literature Cited Anderson, D. M., and M. M. Koth.mann. 19S2. A two- step sampling technique for estimating standing crop of herbaceous vegetation. J. Range Manage. 35:675-677. Bkay, J. R., and J. T. Curtis. 1957. An ordination of the upland forest communities of southern Wiscon- sin. Ecol. Monogr. 27:325-349. Cook, C. W., L. E. Harris, and L. A. Stoddart. 1948. Measuring the nutritive content of a foraging sheep's diet under range conditions. J. Anim. Sci. 7:170-180. Dabel, C. v., and F. p. Day, Jr. 1977. Structural com- parisons of four plant communities in the Creat Dismal Swamp. Virginia. Bull. Torrev Hot. Club 104:3.52-360. Daibenmire, R. F. 1959. A canopy-coverage metiiod of vegetation analysis. Northwest Sci. 33:43-63. Floyd, D. A., and J. E. Anderson. 1982. A new point interception frame for estimating cover of vege- tation. Vegetatio .50:18.5-186. 98 Great Basin Naturalist Vol. 45, No. 1 Hitchcock, C. L., and A. Cronquist. 1973. Flora of the Pacific Northwest. Univ. of Washington Press, Seattle. 730 pp. Humphrey, L. D. In press. Patterns and mechanisms of plant succession after fire on Arteiuisia-gra.ss sites in southeastern Idaho. Vegetatio. Hitc;hings, S. S., and J. E. Schmaltz. 1969. A field test of the relative weight-estimate method for deter- mining herbage production. J. Range Manage. 22:408-411. Payne, G. F. 1974. Cover-weight relationships. J. Range Manage. 27:403-404. Pechanec. J. F., AND G. D. Pk:kford. 1937. A weight estimate method for determination of range or pasture production. J. Amer. Soc. Agron. 29:894-904. S.nedecor, G. VV., and W. G. Cochran. 1980. Statistical methods, 7th ed. Iowa State Univ. Press, Ames. 507 pp. Tadmor, N. H., a. Brieghet, I. Noy-Meir, R. W. Benjamin, and E. Eyal. 1975. An evaluation of the calibrated weight-estimate method for mea- suring production in annual vegetation. J. Range Manage. 28:65-69. Whittaker, R. H. 1966. Forest dimensions and produc- tion in the Great Smokv Mountains. Ecology 47:103-121. 1975. Communities and ecosystems, 2d ed. Mac- millan Publishing Co., Inc., New York. .385 pp. WiLM, H. G., D. F. Costello, and G. E. Klipple. 1944. Estimating forage yield by the double sampling method. Amer. Soc' Agron. J. 36:194-203. Zar, J. H. 1974. Biostatistical analysis. Prentice-Hall, Englewood Cliffs, New Jersey. 620 pp. t A NEW COMBINATION AND A NEW VARIETY IN ARTEMISIA TRIDENT AT A Sherel Goodrich,' E. Diirant McArthur,' and Alma H. Winward- Abstract.- The combination Artemisia tridcntata Nutt. ssp. spicifonui.s (Osteihout) Goodrich & McArriiur comb, nov. is made. This high elevation taxon was originally described at the species level and more recently has generally been treated as a form of A. tridentata ssp. vaseijana. The subspecies designation is supported by its parallel nature to the other A. tridentata subspecies and by its relatively widespread and locally abundant populations. Also, a new variety of A. tridentata ssp. laseijana is proposed. Artemisia spicifonnis Osterhout was de- scribed (Osterhout 1900) from specimens col- lected at North Park, Jackson County (Lari- mer County on the label of Osterhout's type specimen, 2011), Colorado. Artemisia roth- rockii Gray was described (Brewer et al. 1876) from specimens collected at Monache Meadows, Tulare County, California. Plants of these taxa are marked by large heads with about 10-18 flowers, by narrow spicate pan- icles, and by apically dentate or lobate leaves and often with some entire leaves, especially on the flowering stalks. They are mostly found at subalpine and alpine elevations. Both are members of the endemic North American subgenus Tridentatae (McArthur et al. 1981). Hall and Clements (1923) reduced A. roth- rockii to a subspecies of A. tridentata Nutt., and they reduced A. spicifonnis to a synonym or a minor variation of A. tridentata ssp. roth- rockii. They listed a distribution for this com- plex that included Washington to California and east to Wyoming and Colorado. How- ever, they mentioned that the specimens from Colorado including the type specimen of A. spiciformis are more gray or white and more densely cinereous, whereas the type specimen of A. rothrockii is viscid and less densely cinereous. They listed two specimens from Colorado that had partial features of typical A. rothrockii, and fiuther mentioned that most Sierra Nevada plants of the com- plex were cinereous-canescent and scarcely viscidulous. They maintained that perplexing combinations render impossible the recogni- tion of A. spicifonnis. We agree that these two taxa are similar, but Ward (1953), Beetle (1960), Shultz (1983), and Welsh (1983) have treated A. rothrockii and A. spiciformis separately. Ward (1953) maintained A. rothrockii at the species level and included in this taxon only plants of the Sierra Nevada and San Ber- nardino Mountains of California. He suggest- ed that A. rothrockii is composed of three races that possibly have arisen independently and include considerable variation in pubes- cence, stature, purple pigment in the in- volucres, reduction in the inflorescence, leaf form, and habitat. He suggested that A. cana Pursh, and either or both A. tridentata and A. arbuscula Nutt. have contributed to the chromosome complement of A. rothrockii. This taxon (Sierra Nevada and San Ber- nardino Mountain A. rothrockii) is polyploid (4x-8x) (Ward 1953, McArthur et al. 1981). Ward (1953) suggested that more work might show A. spiciformis as a species or as a form of A. rothrockii. With the information avail- able at the time, however, he gave it consid- eration as a local hybrid that is reasonably fertile due to its presumed tetraploid nature. He suggested A. cana and A. tridentata as the parents and stated that he had not seen it ex- tend its range beyond the areas in which both putative parents occur. Beetle (1960) main- tained A. rothrockii at the species level, as did USDA Forest Service, Intermountain Forest and Range Experiment Station, Shrub Sciences Laboratory, Provo, Utah 84601; and adjunct faculty. Depart- ment of Botany and Range Science, Brigham Young University. The authors acknowledge assistance from the Pittman-Robertson W-82-R Wildlife Restora- tion Project. USDA Forest Service, Intermountain Region, Ogden, Utah 84401. 100 Great Basin Naturalist Vol. 45, No. 1 Fig. 1. Distribution oi Artciiiisiii liiilriitald Type locality indicatod In tlic star. .pinf. BHY. OCDF. SSLP, and HM. Ward, but he included in the taxon plants of Colorado, Wyoming, and California. In A. rothrockii he included plants with leaves more or less viscid and not densely canescent that give the whole plant a dark green ap- pearance. He pointed out the similarity of large heads in the type of A. spicifomiis to those of A. rothrockii, but he included other plants with much smaller heads in his con- cept of A. s^picifomiis, and he reduced A. spicifonnis to A. tridentata ssp. vaseyana (Rydb.) Beetle f. spiciformis (Osterhout) Beetle. He listed the range of f. spicifomiis as throughout the range of A. tridentata ssp. va- seyana. He reported fre(}uent confusion with A. rothrockii, and he mentioned that leaves of A. rothrockii are thicker and often more deeply lobed, and the involucral bracts are dark and often purple. Beetle also suggested that A. cana and A. tridentata have had a part in the development of A. spicifonnis. Shultz (1983a, 1983b) also treated A. roth- roc ku as a .speci and included in this taxon only plants of California, as did Ward. She described A. rothrockii as a root-sprouting shrub with dark green, very resinous leaves, and with up to 20 flowers per head. She com- mented on frequent confusion in C'alifornia between A. rothrockii and probably hybrids of A. cana ssp. bolanderi (Gray) Ward and A. tridentata ssp. vaseyana. However, .she main- tained imi(jue anatomical features (gelatinous fibers and niunerous resin ducts) for A. roth- rockii as well as noting that the leaves were January 1985 Goodrich et al.: Artemisia 101 darker green and viscid. She also treated A. spicifonnis as a hybrid of A. cano and A. tri- dentota, as did Ward. Based on the work of Ward and Shultz, we accept A. rothrockii as a plant of California. That being the case, we feel that A. spici- fonnis warrants recognition as more than a localized hybrid. Plants of A. spicifonnis are too widespread (Fig.l), and they form homo- genous stands that are too large and often too far removed from populations of one or both of the putative parents to be mere local hy- brids (McArthur and Goodrich, in press). Welsh (1983) recognized A. spicifonnis as a taxon with relationships running to both A. cmia and A. tridentata, but he maintained this as a species and not as a hybrid. We do not dispute the probability of a hybrid origin, nor do we dispute the occurrence of inter- mediate plants where plants of this taxon and plants of either or both A. cana and A. tri- dentata come together. We think the large- headed plants of Utah and Wyoming that match Osterhout's type of A. spicifonnis and other specimens from Colorado are worthy of taxonomic recognition, and we include them as a subspecies of A. tridentata to bring them in line with other taxa of the complex. We point out that the addition of ssp. spicifonnis brings to five the number of subspecies de- scribed for A. tridentata. Four (ssp. triden- tata, wyomingensis Beetle & Young, va- seijana, and spiciformis) are widely distributed but separate on the basis of mois- ture and elevational gradients and on mor- phological features (Beetle and Young 1965, McArthur 1983, Winward 1983). The other one, ssp. parishii (Gray) H. & C, is similar to and perhaps synonymous with ssp. tridentata. Artemisia tridentata ssp. spicifonnis, with probably hybrid and polyphylic origins, has similarities with ssp. wyomingensis. In the latter case, ssp. tridentata, vaseyana, and A. nova A. Nels. are thought to be involved (Beetle and Young 1965, Winward 1975, McArthur 1983), whereas in the former the putative parents are ssp. vaseyana and A. cana. Like A. rothrockii (sensu Ward and Shultz), ssp. wyomingensis is polyploid (4x-6x; McArthur et al. 1981). However, ssp. spicifonnis is diploid (2x) and tetraploid {4x). Utah populations from the Wasatch Plateau and Uinta Mountains are diploid {n = x = 9) (McArthur et al. 1981, McArthur and Sander- son, unpublished manuscript). A population in the Wyoming Range of west central Wyoming is tetraploid (McArthur et al. 1981). Because the whole of subgenus Triden- tatae has a similar genome (replicated various times in polyploids) (McArthur et al. 1981), and hybridization is possible and in places common in the subgenus (Ward 1953, McAr- thur et al. 1979), ssp. spicifonnis could and probably did arise independently several times at both the diploid and tetraploid lev- els. Apparently, genetic composition and se- lective pressures have led to a stabilized, well-adapted taxon. For Artemisia tridentata ssp. spicifonnis we accept extreme variability in size, thick- ness, and lobing of the leaves, and in the pig- mentation of the involucres, and some vari- ability in the inflorescence (spiciform to narrow-paniculate). We place the greatest emphasis on number of flowers per head (10-18) and in the size of the heads (5-7 mm long, 4-5 mm wide at maturity). These fea- tures seem diagnostic, and they place plants of similar morphology and ecology together. Plants with fewer flowers per head (4-11) and smaller heads that have been referred to as f. spicifonnis belong to A. tridentata ssp. vaseyana. We feel that A. tridentata ssp. spicifonnis is closer to A. rothrockii in head size, number of flowers, and in the tendency to root-sprout than it is to A. tridentata ssp. vaseyana, which does not root-sprout. The variability in features of ssp. spici- fonnis may be a result of several independent origins, as we suggested earlier. This seems reasonable, considering the suggested hybrid origin for the taxon. The plants of California that Shultz reported as possible hybrids are suggested as crosses between A. cana ssp. ho- landeri and A. tridentata ssp. vaseyana. Arte- misia cana ssp. bolanderi is mostly known from California and Oregon, and A. cana ssp. viscidula Osterhout is the logical putative parent throughout much of the range of ssp. spicifonnis. Apparently plants with features of ssp. spicifonnis have not been found in the range of A. cana ssp. cana. Perhaps the name ssp. spiciformis is best reserved for plants in which A. cana ssp. viscidida and A. triden- tata ssp. vaseyana are the logical parents, but we do not propose such reservation at this 102 Great Basin Naturalist Vol. 45, No. 1 time. The probably hybrid origin raises the question whether ssp. spicifomiis should be included in A. cana or in A. tridentata. We have included it with A. tridentata because most of the persistent leaves are lobed and because an additional taxon of A. tridentata at the upper elevation range for the species makes for a logical and complete elevational and mesic gradient series for the species. In addition to morphological features, phenology of plants of ssp. spicifonnis may provide additional evidence that they are unique. Plants of ssp. spicifonnis from high elevations of the Wasatch Plateau, Utah, and from the Wyoming Range, Wyoming, have been transplanted at low elevations at Provo and Ephraim, Utah. These plants come into full flower as early as the last of May and the first two weeks of June. Plants of A. triden- tata ssp. vaseijana and A. cana ssp. viscidida have also been transplanted at these loca- tions. They flower at the transplanted sites about the same time or later than they do in their more montane environments. This is about September or October or even into November. In its native habitat, ssp. spici- fomiis flowers from the first of August through September, but its phenology at the transplanted sites indicates a basic difference from either A. cana ssp. viscidida or A. tri- dentata ssp. vaseyana. Rvdberg (1916) described Arteynisia va- seijana and designated a specimen from Washington (Vasey, 480) as the type. Hall and Clements (1923) examined the type spec- imen and listed 9-10 flowers per head. They discussed this taxon as a variation of A. tri- dentata ssp. rothrockii. Ward (1953) also ex- amined the type specimen and listed 9-11 flowers per head. He discussed this as a form of A. tridentata ssp. tridentata but explained that it differed by having larger heads, larger leaves, and reduced inflorescences. He attri- buted the reduced inflorescences to environ- mental modification but indicated the large head size to be genetically controlled. He also listed several other collections from Washington, Oregon, and Idaho with 7-11 flowers per head. Beetle (1959) included A. vaseijana as a ssp. of A. tridentata. We agree with this combination. Ward (1953) pointed out some ecological and cytological variations in this group. He mentioned plants of intermountain areas (areas between mountains) "with rounded shape with the inflorescences mingled with the vegetative shoots and only partly exceed- ing them." We believe these plants are those Beetle and Young (1965) named as A. triden- tata ssp. ivyomingensis, the same tetraploid plants discussed by Ward (1953) that usually grow on poorer, rockier soils, and that are smaller and slower growing than diploid plants. He compared these plants with a form of sagebrvish of "timbered or mountainous areas in which the plants are very uniform in size, usually about 2 feet in height, and of a rather spreading, flat-topped habit of growth, with the inflorescence extending upward like plumes above the rest of the bush." He fur- ther reported that the two forms retained their different growth rates under uniform conditions in a garden. He discussed these two forms simultaneously but separately from A. vaseyana, which he went on to discuss as another form with larger heads. For many years these montane, flat-topped plants with only 4-6 flowers per head have passed for A. tridentata ssp. vaseyana. We believe they constitute a separate taxon. This small-head- ed, few-flowered phase is widespread in most of the western states, whereas those matching the type of typical A. tridentata ssp. va- seyana are most common in the upper eleva- tional sagebrush areas of Washington, Ore- gon, and Idaho. The morphological differences are striking enough in the ex- treme, but there is no clear-cut boundary, and we do not dispute that intermediate plants are often encountered. However, each of the phases form large, pure stands, and the geographic ranges are not entirely the same. Hironaka et al. (1983) recognized the two different phases, and designated separate habitat types based on the two phases. The differences will continue to be recognized, and names for both seem to be appropriate. The following name is proposed for plants with smaller and fewer flowered heads and wider inflorescences: Artemisia tridentata ssp. vaseyana var. paiiciflora Winward & Goodrich var. nov. Similis A. tridentata ssp. vaseyana var. va- seyana sed in capitulis parvis floribus paucis et inflorescence latis differt. January 1985 Goodrich et al.: Artemisia 103 Type: Utah, Utah County, T7S R4E corner of sections 28, 29, 32, and 33, 9.7 km 85° east of Springville, Uinta N.F., Wasatch Moun- tains, Left Fork Hobble Cr., 1582 m, sage- brush-grass community in a Gambel oak -big- tooth maple zone. Goodrich, Winward, McArthur, and Lewis 21492. Holotype BRY!, Isotypes ID, MO, NY, RENO, RM, SSLP, UC, UT, and UTC. Paratype: Utah, Utah Coimty T8SR3ES1 SEV4 of NW1/4, 5 km 112° east of Springville, 1524 m, sagebrush-grass community. Goodrich, Winward, McArthur, and Lewis 21490. BRY, CS, MONTU, OGDF, OSC, and WTU. The following key provides a summary for the taxa of A. tridentata listed above except for A. tridentata ssp. parishii (mostly of southern California), which is like A. triden- tata ssp. tridentata except for usually arach- noid and short-villous achenes. 1. Unevenly topped shrubs with the inflorescence and vegetative twigs inter- mingled, the flower stalks rarely over twice as long as the subtending vegeta- tive twigs; plants of valleys and lower mountains; fluoresces reddish brown in alcohol (Winward and Tisdale 1969) 2 — Evenly topped shrubs; flowering stalks well elevated above the leaves, mostly over twice as long as the subtending vegetative twigs; plants of low to high elevations in mountains; fluoresces bluish cream in alcohol 3 2(1). Plants mostly over 1 m tall, often with a discernible main trunk, diploids; length/width ratio of leaves 4.0 or greater A. tridentata ssp. tridentata — Plants mostly less than 1 m tall, often quite branched from near the base and then without a readily discernible main trunk, mostly tetraploids; length/width ratio of leaves less than 4.0 A. tridentata ssp. wijomingensis 3(1). Head with 10-18 flowers; plants tending to root-sprout, of subalpine elevations, often in openings in aspen and spruce or fir woods, or where drifting snow ac- cumulates, flowering in August and September in native habitats and in early Jime when transplanted at low elevations (1370 m), diploids and tetraploids, possibly of hybrid derivation involving A. tridentata and A. cana A. tridentata ssp. spicifonnis — Heads with 4-11 flowers; plants not root-sprouting, of medium to high eleva- tions in mountains, flowering in September, October, and November when transplanted at low elevations 4 4(3). Heads with 7-11 flowers, over 1.5 mm wide; inflorescence narrow and spici- form with relatively few heads A. tridentata ssp. vaseijana var. vaseijana — Heads with 4-6 flowers, less than 1.5 mm wide; inflorescence paniculate with numerous heads A. tridentata ssp. vaseijana var. paiici flora Literature Cited Beetle, A. A. 1959. New names within the section Tri- dentatae oi Artemiski. Rhodora 61:82-85. 1960. A study of sagebrush, the section Tridcn- tatae of Artemisia. Wyoming Agric. Expt. Sta. Bull. 368. 86 pp. Beetle, A. A., and A. Young. 1965. A third subspecies in the Artemisia tridentata complex. Rhodora 67;405-4()6. Brewer, W. H., S. Watson, and A. Gray. 1876. Bot. Cal. Vol. 1, Bigelow and Co. Univ. Press, Cam- bridge, Mass. 628 pp. Hall, H. M., and F. E. Clements. 1923. The phyloge- netic method in taxonomy; tlie Nortli American species of Artemisia, Chrysothamnus, and Atri- plex. Carnegie Inst. Wash. Piibi. 326:1-355. HiRONAKA, M., M. A. FOSBERG, AND A. H. WiNWARD. 1983. Sagebrush-grass habitat types of southern Idaho. Bull. No. 35. College For., Wildlife, and Range Sci. Univ. of Idaho, Moscow. 44 pp. McArthur, E. D. 1983. Taxonomy, origin, and distribu- tion of big sagebrush {Artemisia tridentata) and allies (subgenus Tridentatae). Pages .3-13 in K. L. Johnson, ed.. Proceedings of the first Utah Shrub Ecology Workshop. College of Natural Re- sources, Utah State Univ., Logan. .50 pp. McArthur, E. D., A. C. Blauer, A. P. Plummer, and R. Stevens. 1979. Characteristics and hybridiza- tion of important intermountain shrubs. III. Sun- flower family. USDA Fore.st Service Res. Pap. INT-220. Intermountain Forest and Range Ex- periment Station, Ogden, Utah. 82 pp. 104 Great Basin Naturalist Vol. 45, No. 1 McArthur, E. D., and S. Goodrich. In press. Aite- inma trklcntaUi Ssp. Spicifonnis: distribution and taxonomic placement. In E. D. McArthur and B. L. \\'elch, compilers. Proceedings— Symposium on the Biology of Artemisia and Clirijsotlitimniis. USDA Forest Service Gen. Tech. Rep. INT-.In- termountain Forest and Range E.xperiment Sta- tion, Ogden, Utah. Mc;Arthur. E. D., C. L. Pope. .\.\d D. G. Freem.\.n. 1981. Ghromosomal studies of subgenus Tiidcn- latiw of Artemisia: evidence for autopolyploidv. Amer. J. Bot. 68:589-605. OsTERHoiT, G. E. 1900. New plants from Colorado. Bull. Torr. Bot. Club 27:506-508. Rydberg, p. a. 1916. N. Amer. Flora. 34(.3):28.3. Shui.tz, L. M. 1983a. Origin and distribution of Arte- misia rothroekii (Asteraceae, Anthemidaea). Amer. J. Botany 70(5)Part 2:129. SiirLTZ, L. M. 198.31). Systematics and anatomical stud- ies of Artemisia subgenus Tridentatae. Unpub- lished dissertation. Ciaremont College, Clare- mont. California. 169 pp. \V.\Rn, G. H. 1953. Artemisia section Scrij)hi(liinn in North America: a cvtotaxonomic studv. C^ontr. Dudley Herb. 4:155-205. Welsh, S. L. 1983. Utah Flora: Compositae (.Astera- ceae). Great Basin Nat. 43:179-.357. WiNWARD, A. H. 1975. Evolutionary development of the Artemisia tridentata taxa. Page 163 in H. C. Stutz. ed.. Proceedings, .symposium and work- shop, u ildland shrubs. Brigham Young Univ., Provo, Utah. 168 pp. 1983. Using sagebrush ecology in wildiand man- agement. Pages 15-19 ()i K. L. Johnson, ed., Pro- ceedings of the first shrub ecology workshop. College of Natural Resources, Utah State Univ., Logan. 50 pp. WiNWARD, A. H., AND E. W, TisDAi.E. 1969. A suuplified chemical method for sagebrush identification. Univ. of Idaho Forest, Wildlife., and Range Expt. Sta. Note 11. 2 pp. UNDERSTORY RESPONSE TO TREE HARVESTING OF SINGLELEAF PINYON AND UTAH JUNIPER Richard L. Everett' and Steven H. Sharrow- Abstract— Fifteen widely separated sites within tlie pinyon-juniper woodlands of the Great Basin were cleared of trees. Understory response was recorded for 2 to 4 years. The array of postharvest plant assemblages were classified into one of four phytosociological groups using discriminant analysis. Pre- and postharvest plant assemblages from the same site appeared in the same phytosociological group, which indicates postharvest response could be predicted from the preharvest plant assemblage. Initial postharvest response appears cyclic in nature, and cycles are controlled bv both residual plants and the rapid immigration of shrub species. Perennial plant density generally declined fol- lowing tree harvest, but the fewer remaining plants produced significantly more cover than in preharvest stands. Wood harvesting in pinyon-juniper wood- lands of the Great Basin has a long and color- ful history. In the late 1800s, the harvesting of trees for charcoal used in the smelting of ore and as heating fuel had decimated tree populations for an 80-km radius around many mining camps. An estimated 1600 to 2000 ha of woodland were cleared annually to fuel this burgeoning mining industry (Young and Budy 1979). Concurrently, livestock numbers, including the nmnerous draft animals used in wood harvesting, were increasing and in- discriminant burning of woodlands was com- mon (Tausch et al. 1981, J. A. Young, pers. comm.). These perturbations left an array of depleted disclimax understory communities (Tueller 1973) upon which the current wood- lands would become established. Trees rees- tablished rapidly in the Great Basin (cur- rently 7.1 million ha, Tueller et al. 1979), and understory cover and productivity was re- duced (Tausch et al. 1981, West 1983). Rising fossil fuel costs have recently in- creased the demand for cord wood (Meeuwig and Cooper 1981). Thus, widespread tree harvesting is once again a major use of these woodlands. With appropriate management based on a knowledge of probable understory response, wood harvesting could provide a cost effective means of utilizing the wood re- source and increasing the forage base. Understory response to tree harvest is closely linked to the type and number of re- sidual plants (MacMahon 1980, Clary 1974; Dryness 1973, Arnold 1964). Grass and shrub production increases following tree harvest- ing in the Southwest if these plants are nu- merous and capable of capturing released re- sources (Clary 1974, Arnold and Schroeder 1955). But floristically impoverished sites with low site potential for understory pro- duction can remain static for several years following tree removal (O'Rourke and Ogden 1969). Our study tests the hypothesis that post- harvest response in fully stocked singleleaf pinyon {Piniis monophyUa)-lJ tah juniper {Juniperus osteosperma) woodlands of the Great Basin is "site specific cyclic." We hy- pothesize that a linear replacement series of plant forms does not often occur following tree harvest. If mid-to-late successional un- derstory plants or their immigrating seeds are available to the site, they control the charac- ter of postharvest response and negate early successional stages. Rapid reinvasion of trees in the central core of the woodland also trun- cates midsuccessional shrub dominance. This hypothesis is limited to the midelevation of the woodland and may not be viable at either upper or lower ecotones. Methods In 1977, 1978, and 1979 study sites were established in 15 singleleaf pinyon-Utah juni- per stands across the Nevada portion of the Great Basin (Fig. 1). Stands selected for tree 'Interniountain Forest and Range Experiment Station. I'SDA Forest Service, Ogden, Utali 84401 -Rangeland Resources Department, Oregon State University, Corvallis, Oregon 97331. 105 106 Great Basin Naturalist Fiu. 1. Location of study sites in the pinvon-jnniper woodhuKis of the C;,eat Bas.n. Woodland d.stnhut.on patterns takrntn.niTueUeretal. {1979). January 1985 Everett, Sharrow: Pinyon-Juniper Ecology 107 harvest were fully stocked with trees and had minimal imderstory. Plots (0.1 ha) were se- lected for site uniformity in microtopography and understory composition. Trees greater than 1 m in height were hand cut, sectioned, and removed from the plots with minimal distrubance to the soil surface. A series of five parallel transects 20 m in length were laid out on 5-m intervals parallel to one another across the slope. Tree cover was estimated using line intercept (Canfield 1941) along each transect. A series of frames (50 X 50 cm) were laid down at every meter mark, and plant species cover and density were recorded. Understory response on the harvested sites and on adjacent uncut areas were monitored for 2 to 4 years following tree removal. Predominant preharvest under- story species, tree cover (%), elevation, as- pect, and year of harvest are given for each site in Table 1. Discriminant analysis was used to group the array of pre- and postharvest plant as- semblages (30 total) based on nine phytoso- ciological parameters. Phytosociological pa- rameters used were areal species richness, "evenness" in cover among plant forms (Bril- louin's [H] measure of diversity divided by maximum diversity in distribution of cover possible [J = H/H max] : Poole 1974), aggre- gation of perennial plants (Morisita's Index of Aggregation : Poole 1974), total plant cover, and proportion of cover by shrub, perennial grass, perennial forb, annual grass, and an- nual forb. Postharvest data were run using the preharvest groupings to test if post- harvest plant assemblages could be predicted from preharvest data. Pre- and postharvest data sets from the same site are not independent. Thus, data was compared by t-tests of differences. Multi- variate matched pair t-tests of differences (Timm 1975) were conducted for closely re- lated parameter pairs, woody and herbaceous cover, annual and perennial cover, total cov- er and evenness in cover distribution, and perennial plant density and perennial plant aggregation. The test statistic used was Hot- teling's t at the 0.05 significance level. A correlation matrix of the phytosociolo- gical variables was used to identify parame- ters that varied together (P = 0.05) in pre- and postharvest plant assemblages. Propor- tion of species common to pre- and post- harvest plots were included with those varia- bles already mentioned. Results and Discussion Lack of climatic effects on plant RESPONSE TO TREE HARVEST.— Precipitation during the final year of postharvest measure- ment was less than occurred during the year of preharvest measurement. Mean precipi- tation from six official weather stations clos- est to the 15 study sites averaged 28.5 cm Table 1. List of studv sites, dominant nnderstoiv, percent tree cover, elevation, aspect, and harvest year. Study Percent Preharvest understory site cover Elevation Aspect Year Annual forb domlnant Pluicelia httmilis 4 36 2030 N 78 CoUinsia panifloni 9 26 2220 E 79 CoUinsid pciniflorti 15 56 2200 E 79 Shrub domlnant Purshia tridcntatu 1 60 2300 E 77 A rtemisia arbuscti la 2 49 2200 N 77 Artemisia arbusciihi 8 41 2070 W 79 Ribes vehitinum 3 64 2100 E 77 Perennial grass domlnant Poa sandbcroii 5 56 2040 E 78 Pod .S((iu//>t')gil 6 52 2060 N 78 Poa s(indbcr>iii 7 58 2190 E 78 Poa .saiulbcn^ii 11 54 2300 W 79 Poa siindbcrgii 12 54 2280 S 79 Poa sandbergii 13 43 2200 N 78 Vcstuca idahoensis 10 28 2340 N 79 Perennial forb dominant Penstemon pack ijph yUus 14 50 2200 N 79 108 Great Basin Naturalist Vol. 45, No. 1 (September to August) the year of preharvest measurement and 21.1 cm the year of final postharvest measurement. The initial pre- harvest plant cover (X = 2.71 dm^/m^) was not significantly different (P = 0.1) from plant cover on nonharvested plots (X = 2.57 dnf/m^) the final year of postharvest mea- surements. Understory response to tree re- lease was not confounded by precipitation levels following tree harvest. Plant response to tree harvest.— Semi- arid pinyon-juniper woodlands are character- istically low to moderate in species richness, but this varies with site heterogeneity (Har- ner and Harper 1976). There was no increase in areal species richness following tree har- vest. Preharvest plant assemblages averaged 14 species (± 4 SD), and postharvest plant assemblages averaged 11 species (± 6 SD). On 10 of 15 sites, species numbers were lower after tree harvesting. Tlie proportion of species common to pre- and postharvest plant assemblages averaged 57% (± 21 SD) among sites. The proportion was lowest on the more depauperate sites (1, 3, and 15) and highest when preharvest cover was more evenly divided among several plant forms (sites 8, 9,' 10, 11, and 14). Discriminant analysis of plant RESPONSE.— The array of pre- and post- harvest plant assemblages (30 total) were classified by discriminant analysis into (1) an- nual forb, (2) shrub, (3) perennial grass, and (4) perennial forb phytosociological groups. Pre- and postharvest plant assemblages were intermingled within the groups. The four groups were significantly (P = 0.001) differ- ent from each other. Aside from plant form dominance, groups differed significantly (P = 0.001) in areal species richness, perennial plant aggregation, perennial plant density, total cover, and "evenness" in the distribu- tion of cover among plant forms. Figures 2A and B separate pre- and postharvest plant as- semblages for ease of interpretation. The first discriminant function (axis) ordi- nated plant assemblages along a gradient of increasing perennial grass cover and evenness in distribution of cover among plant forms. The second discriminant function (axis) ordi- nated plant assemblages along a gradient of increasing species richness, shrub cover, and perennial forb cover. The third discriminant function (axis) emphasized increasing total cover, species richness, and cover of annual forbs. These discriminant functions explained 80% of the variance in the data. Postharvest response was not unidirection- al in phytosociological change. Vectors drawn from pre- and postharvest plant as- semblages varied in length and direction (Fig. IC). Annuals invaded shmb sites, shrubs in- Table 2. Differences between pre- and postharvest phytosot iological pai anieter pairs iniuit \ariate paired t-test of difference). Phvtosociological Pail 1 Pair 2 Total Evenness Preharvest cover in cover .\nnuals Perennials group Site {dm-/m-) (-1 to +1) (dnrVm-) Annual forb 4 -5.66°° -0.27°° -42.26°° 13.56°° Annual fori) 9 0.78°° 0.13°° 0.78 3.67 .Vnnual fori) 15 2.92°° 0.12 1.3.94°° 0.70 Shrub 1 3.5° 0.47°° 9.28°° 8.28 Shmb 2 4.9°° 0.33° 9.69 9.75° Shrub 3 1.83°° 0.05 0 3.35 Shrub 8 1.03°° 0.04 0.161 4.97t Perennial grass 5 0.61 ().0 1: Morisita's Index) in all pre- and postplant a.s.semblages, with the exception of the regular dispersion of plants (I = 0) on preharvest sites 3, 4, and 15. At least two scales of aggregation were visible on most sites. Understory and tree cover were heterogeneous across the ground sur- face, causing general patchiness. Within a given patch, understory was generally ex- cluded from the duff areas adjacent the tree. January 1985 Everett, Sharrow: Pinyon-Juniper Ecology 111 resulting in understory aggregation in the interspace. Perennial plant density was low in pre-and postharvest plant assemblages: X = 11.45 plants/m^ vs. X = 11.31 plants/m^, respec- tively. Perennial plants increased on sites ini- tially dominated by annuals (sites 3, 4, and 15) and perennial forbs (site 14), and on a single grass site (site 13), but decreased on all others. Physical damage during tree harvest- ing, altered microclimate following tree har- vest, and postharvest grazing effects (sites 5, 6, and 7) probably contributed to this loss of perennial plants. Correlation matrix of phyto- sociOLOGiCAL PARAMETERS.— As perennial grass and forb cover increased, so did species richness, evenness in cover distribution, and perennial plant density (Table 3). High shrub cover was associated with high plant aggre- gation, low evenness, and low grass cover. The proportion of species common to pre- and postharvest plant assemblages was in- versely related to total cover, aggregation. and the shrub-annual forb cover. On sites subject to immigrating shrubs or annual forbs, a sharp change in species composition occurred but species numbers remained less than in grass or forb dominated understory. Diversity of understory increased spatially (aggregation) on shrub sites. Diversity in- creased floristically (species richness) and structurally (evenness) on grass sites. Conclusions Postharvest response was cyclic and could be predicted from preharvest plant assem- blages barring outside perturbations. Post- harvest response was best explained by Eg- ler's (1954) "Initial Floristics" model where residual plants and rapidly immigrating spe- cies dominate response. Predictability of postharvest response with its high proportion of preharvest species (57%) is consistent with this hypothesis. Predictability of response is increased under this system. Unfortunately, some plant forms may be excluded from the Table 3. Correlation matrix of phytosociological parameters. Preharvest SRI PD TC EV AG itS ttG ttF ttAG 77AF 1. 0.66°'^ 0.0 0.49 -0.11 -0.05 0.37 0.14 0.39 0.36 1. -0.07 0.56° -0.18 -0.18 0.52' 0.48 0.07 -0.43 1. -0.06 -0.13 -0.03 -0.27 -0.18 -0.07 0.36 1. -0.28 -0..53° 0.64' 0.42 0.29 -0.09 1. 0.52° -0.21 -0.17 -0.05 -0.33 1. -0.66' 1. -0.36 0.45 1. -0.25 0.28 -0.07 1. -0.49 -0.28 -0.28 0.015 1. Postharvest SR PD TC EV AG 77S ttG 77F 77AG 77 AF ttCS 1. 0.72 0.14 0.48 -0.49 -0.48 0.48 0.58° -0.23 -0.05 0.26 1. 0.19 0.32 -0.27 -0.28 0.57' 0.43 -0.16 -0.37 0.35 1. 0.0 -0.07 0.19 -0.44 0.17 -0.48 0.32 -0.52° 1. -0.5,9° -0.41 0.39 0.23 0.34 -0.18 0.26 1. 0.22 -0.45 -0.29 0.15 0.29 -0.41 1. -0.63° 1. -0.27 0.13 1. -0.18 0.09 -0.15 1. -0.36 -0.37 -0.19 -0.20 1. -0.30 0.72° 0.18 0.23 -0.47 1. SR = areal species richness; PD = plant density; TC = of shnib, perennial grass, perennial forb, annual grass, and am "° denotes significant correlation coefficient (P < 0.05). tal cover; EV = evenness; AG = a al forb. wCS = proportion of specit 1; wS, ttG, ttF, wAG, wAF = proportional cover to pre- and postharvest assemblages. 112 Great Basin Naturalist Vol. 45, No. 1 postharvest plant assemblage. When tree har- vesting is done to increase the forage base for livestock or wildlife, the desired species should already be in the understory. Under- storv species richness was reduced following tree harvest on 10 out of 15 sites as suggested by Loucks (1970) for more mesic forests. Un- derstory diversity increased spatially on shrub sites (aggregation) and increased flo- ristically (species richness) and structurally (evenness) on grass sites. Total plant cover increased following tree harvest on all sites not subjected to severe livestock grazing. Preharvest understory in the fully stocked stands was severely sup- pressed. Mean increase in postharvest cover was 1.48 dm^/m- and represented a 200% in- crease in cover over preharvest conditions. Literature Cited Arnold. F. 1964. Zonation of understory vegetation aronnd a juniper tree. Journal of Range Manage- ment 17:41-42. .'Vb.niold, J. F., AND \V. L. Sc:hroeder. 1955. Juniper con- trol increases production on the Fort Apache In- dian Reservation. Station Paper 18. USD.\ Forest Service, Rocky Mountain Forest and Range Ex- periment Station, Fort Collins, Colorado. 35 pp. Canfikld, R. H. 1941. Measurement of grazing use by the line interception method. Journal of Forestrv 42:192-194. Ci.AHV, W. P. 1974. Response of lu liiateous vegetation to felling of alligator juniper, joiunal of Range Management 27:387-389. IXrness, C. T. 1973. Early stages of plant succession fol- lowing logging and binning in the western Cas- cades of Oregon. Ecology 54:57-69. Kc.i.i.H, F. F. 1954. Vegetation science concepts. I. Ini- tial floristic composition a factor in old-field \eg- etation development. Vegetatio 4:412-417. Erdman, J. -\. 1970. Pin\on-juniper succession after nat- ural fires on residual soils of Mesa Verde, Colo- rado. Brigham Young University Science Bulletin, Biological Series. 11(2). 24 pp. II \RNER, R. E., AND K. T. Harper. 1976. The role of area lieterogeneity and favorability in plant species di- versity of pinvon-juniper ecosystems. Ecology 57:12.54-1263.' Klecka, W. R. 1975. Discriminant analysis. Pages 4:34_467 in H. Nie, C. Hull, J. Jenkins, K. Ste'in- brenner, and D. Bent, eds.. Statistical package for the social sciences. McGraw-Hill, New York. Loucks, O. L. 1970. Evaluation of diversity efficiency, and community stability. American Zoology 10:17-25. .MacMahon, J. A. 1980. Ecosystems over time: succes- sion and other types of change. Pages 27-58 in R. H. \\'aring, ed.. Forests: fresh perspectives from ecosystem analysis. Proceedings Biology Collo- quium, 40th. Oregon State University, Corvallis. Meel wic, R. O., and S. V. Cooper. 1981. Site quality and growth of pinyon-juniper stands in Nevada. Forest Science 27(3):593-601. ORoiRKE, J. T., and p. R. Ogden. 1969. \'egetative re- sponse following pinyon-juniper control in .\ri- zona. Journal of Range .Management 22:416-418. Poole, R. W. 1974. .\n introdiiction to quantitati\e ecology. McGraw-Hill, New York. 532 pp. Tausch, R. J.', N. E. West, and \. A. Nabl 1981. Tree age and dominance patterns in the Great Basin pinyon-juniper woodlands. Journal of Range Management 34:259-264. Terborch, J. 1973. On tlie notion of favorableness in plant ecology. American Naturalist 107:481-500. TiMM, N. H. 1975. Multivariate analysis with appli- cations in education and psychology. W'adsworth Publishing Company, Beln'iont, California. 689 pp. TuELLKR. P. T., C. Beeson. R. J. Tausch, N. E. West, AND K. H. Re\. 1979. Pin\on-juniper woodlands of tlie (ireat Basin: distribution, flora, vegetal cover. l'SD.\ Forest Service Research Paper INT- 229. Intermountain Forest and Range Experi- ment Station, Ogden, Utah. 22 pp. YoiNc. J. A., AND J. D.Bi DV. 1979. Histniii al use of Ne- \adas pinNon-junijier woodland. |i)urn .05) between middle temporary and lower temporary during the first five weeks (basket sampling) but were higher at LT (p < .05) during the following nine weeks (Sur- ber sampling). The permanent sites had high- er (p < .05) densities than any intermittent site throughout the study. Commimity Structure Horn's (1966) measure of community over- lap was calculated for the numbers of species at UP and LP at the start and end of each sampling method (Table 2). The value of overlap, R^, approaches 1.0 as the amount of overlap increases. The low overlap at the be- ginning of the experiment (R„ = .43) reflects the differences that had developed between the permanent sites while separated during drought conditions for approximately 1.5 to 2.0 years. Following the connection by water after the runoff event, LP and UP showed a linear increase in community overlap as LP became more like UP. Colonization Patterns of Selected Organisms The general pattern across taxa was that colonization began later at MT and LT than at UT. Once colonization was initiated, den- sities rapidly increased until MT and LT were similar to UT. This general pattern was variable between groups of organisms. Based on their densities throughout the study period and their feeding and dispersal strategies, the following taxa were selected for detailed analyses: Baetis spp. (Baetidae: Ephemeroptera), Neothremma sp. (Limne- philidae: Trichoptera), and four genera of Ephemeroptera in the family Heptageniidae, Cinygmula spp., Heptagenia spp., Epeorus spp., and Rhithrogena spp. Baetis and Neo- thremma are both in the collector feeding functional group (Merritt and Cummins 1978), but they exhibit distinct differences in mobility. Baetis is an excellent swimmer, ac- tively entering the drift (Corkum et al. 1977, Corkum 1978a,b). Neothremma, a sand case- building caddisfly, may be passively dis- tributed (Minshall and Winger 1968). The heptageniids are all in the collector-gatherer feeding functional group. They are all dorso- ventrally flattened and adapted for crawling and are not considered good swimmers (Cor- kum 1978a). Baetis were the first organisms to colonize the temporary sites (Table 3). The upper per- manent site baskets had more Baetis than all downstream site baskets (p < .05), and UT had more than all sites below it, including LP (p < .05). The lower permanent site had the highest density of Baetis during the basket sampling (1333/m-), but, over all dates, LP was lower than UP and UT (p < .05). By the end of the Surber sampling, UP, MT, and LP had similar numbers of Baetis (p > .05) and each of these sites had higher numbers of Baetis than either UT or LT (p < .05). Neothremma density was greater at UP than at any lower site (p < .05) in both the basket and Surber sampling periods (Table 3), although the lower permanent site was colo- nized before any temporary site. The lower permanent site had more Neothremma than all stations above it (p < .05) except UP (p > .05) by the end of the basket sampling pe- riod. At the conclusion of the study a com- parison of Neothremma density between sites showed UP>UT>MT>LT. Density was higher at LP than LT (p < .05), but it was not different from any other site (p > .05). Table 2. Horn's (1966) measure of community over- lap (Rq) between Upper Permanent and Lower Per- manent sites at the start and end of basket and Surber sampling efforts. Sampler and period Ro° Beginning basket sampling Ending basket sampling Beginning Surber sampling Ending Surber sampling .43 .57 .64 .74 complete similarity or overlap 120 Great Basin Naturalist Vol. 45, No. 1 ^ 0.30 \5 60 DAYS 75 90 105 Fig. 1. Spring-summer hydrograph of Upper Per- manent (UP), Middle Temporary (MT), and Lower Per- manent (LP) sites on Stewarts Creek. The upper temporary site had the fastest colonization by Heptageniidae (Table 3). During the basket sampling period there were no differences between any site except UT and LT (p > .05), where UT was higher. The numbers of heptageniids increased over time at all the temporary sites. Middle tem- porary had more heptageniids than all sites (p < .05) except LP (p > .05), which had more than all sites during the Surber sampling. Drift Catastrophic drift, caused by high flow conditions, should be reflected in the density of organisms in the diift taken during the as- cending limb of the spring hvdrograph (An- derson and Lehmkulil 1968). The patterns for total species, Neothremma, Baetis, and Hep- tageniidae are shown in Figure 1. At LP, drift (Fig. 1) was lowest during the ascending portion of the hydrograph (Fig. 2). The numbers of species increased in the drift over time (p < .05) at the lower drift station (Fig. 1). Each group studied; i.e., Baetis, Neothremma, and Heptageniidae in- creased in numbers at the lower station be- tween the 9th and 12th sampling weeks. It was during this same interval (9th-12th weeks) that the largest reduction in discharge occurred (Fig. 2). At the lower site, total species, total organ- isms, and the density of Baetis, Neothremma, and heptageniids were positively correlated with discharge during the ascending part of the stream hydrograph (Table 4). These same groups continued to increase when discharge was decreasing (p < .05). The upper drift net site had negative but not significant corela- tion coefficients (p > .05) for both the as- cending and descending comparisons. Discussion The large amount of litter that accumu- lated in the channel during the dry period was washed out during the ascending limb of the spring hydrograph; however, leaf pack accumulations did remain in the temporary sections. We therefore expected that organ- isms that feed on detritus and are good dis- persers would disperse throughout the inter- mittent area given sufficient time. Baetis are active drifters and good swimmers; they also feed by collecting detritus. Baetis were the most similar in their distribution and abun- dance at all study sites (Table 3) by the end of this study. If an organism is a passive disperser, then we would expect a pattern of high densities nearer to the source of colonizers, with de- creasing densities further from that source. Neothremma showed such a pattern. The higher densities at LP could be due to changes in flow conditions, i.e., increased Table .3. Mean densities (number/m^) oi Baetis, Xeothicimiui. and Heptageniidae for starlni^ Itaskct sanipli SBS, ending basket samplers = EBS, starting Surber samplers = SSS. and ending Sini)er samplers = KSS ti sampling stations. all UP UT Sampling Taxa SBS EBS SSS ESS SBS EBS SSS ESS Baetis Neothremma Heptageniid 256 722 122 67 356 11 217 1835 456 244 4667 311 11 0 22 769 333 233 89 445 56 278 343 422 January 1985 McArthur, Barnes: Utah Macroinvertebrates 121 settling with reduction in flow, or upstream migration from lower adjacent permanent sections. The increases in heptageniid density at LP and MT, larger than that of UT and LT, have two possible explanations. First, the heptage- niids might have migrated upstream from be- low LP. Second, heptageniids may have quickly moved through the temporary sites in response to poor food or habitat conditions. A pattern similar to Neothremmo would be expected if heptageniid drift were passive. If the organisms were moving to better feeding conditions, then they should congregate wherever a food source was found. Being poor swimmers but adapted for crawling, heptageniids may have moved up from LP through LT or down from UP through UT. Since there were no Heptageniidae found in the drift imtil after the sixth week, we have assumed a crawling migration for Heptage- niidae into the temporary sites. Baetis and Heptageniidae density at MT during July and August was greater than for the other temporary sites. Since MT is an ap- proximate midpoint in the study sites, then the high densities may be a fimction of the overlap in organisms migrating up from LP and down from UP. The higher densities at middle temporary may also be due to the study terminating before equilibrium den- sities could be established at UT and LT. Baetis drift easily and were shown to be dis- tributed with some equality throughout the study sites. Therefore, an increase in Baetis at any one site after sufficient time indicates a selection for a preferred habitat. Even though there was a significant in- crease in discharge during the early weeks of the study, drifting organisms did not follow a similar pattern. The numbers of organisms leaving UP increased over time, but did not increase in proportion to the changes in dis- TOTAL SPECIES MAY JUN JUL AUG Fig. 2. Changes in tlie densities (niunhei/nr^) of taxa, Neothremnui, Baetis, and Heptageniidae in drift below the Upper Permanent (UP) and above Lower Permanent (LP) study sites in Stewart's Creek. charge. It was during the period of reduction in flow that drift increased most rapidly. This supports the work of Minshall and Winger (1968). This response may be a function of one or both of the following: (1) There is no net increase in drifting organisms but only a relative increase in density as the discharge decreases and (2) organisms are drifting as a behavioral response to these discharge changes. If No. 1 is correct then, after adjust- ing the drift densities for discharge and get- ting numbers/ time, the ratio of any pair of data points should be unity. If the numbers of organisms/ m^ are actually increasing because more are actively entering the drift, then the ratios between adjusted pairs should increase Table 3 continued. site MT LT LP SBS EBS sss ESS SBS EBS SSS ESS SBS EBS SSS ESS 0 0 0 922 33 278 467 144 367 511 147 800 0 0 0 33 0 244 567 44 456 200 178 878 0 256 11 1333 422 223 622 78 267 100 500 822 122 Great Basin Naturalist Vol. 45, No. 1 Table 4. Correlation coefficient.s obtained from analysis of the relationships between either the ascending or de- scending limb of the hydrograph and total taxa, total numbers, Baetis, Xeothremina, and Heptageniidae. p < .0.5 Taxa richness Densities Baetis Xeothremina Heptageniidae Site .•\scend Descend .\scend Descend Ascend Descend Ascend Descend .\scend Descend UP LP -.17 -.35 .84° -.59 -.58 -.58 .91 -.77° -.66 -.17 .95° -.18 -.70 -.48 -.02 -.81° -.50 -.16 .40 -..37 over time and the slope of the regression of these adjusted numbers against time should not equal zero. The slope of each curve was significantly different from zero (t > critical value, p < .05), indicating a behavioral re- sponse of the macrobenthos. The benthic sampling procedures used in this study were not selective for any one col- onization vector, i.e., vertical (from hy- porheos), aerial (ovipositing), downstream drift, and upstream migration. We felt that the influence of vertical movement and aerial colonization would be uniform throughout the study site, whereas drifting and upstream movement would show a response to dis- tance. This response would depend on the distance from the colonization source and the dispersal ability of the organisms (Waters 1964, McLay 1970, Elliott 1971a). Some re- searchers have demonstrated significant up- stream movement (Hultin et al. 1969, Bishop and Hynes 1969, Elliott 1971b). The dis- tances traveled upstream were up to 600 cm/night (Elliott 1971b). At that rate an or- ganism could hypothetically move upstream the length of the temporary section in 50 d. Although this is possible, the results of the drift measurements show that there are suf- ficient colonizing organisms in drift to ac- count for the colonization patterns we de- scribed for Baetis and NeutJirenima. The established stream communities, UP and LP, were the primary sources of poten- tial colonizers. Active and passive drift or- ganisms leaving UP would first reach UT. Once settling out, an organism may stay or leave depending on the suitability of the hab- itat, availability of food, space, and flow conditions. Literature Cited A.NDEHSON, N. H., A\u D. .Vl. Lkhmkiiii.. 1968, Caty strophic drift of insects in a vvoodlantl streaii Ecology 49:198-206. Bishop, J. E., a.nd H. B. N. Hynes. 1969. Upstream movements of the benthic invertebrates in the Speed River, Ontario. |. Fish. Res. Bd. Canada. 26:279-298. CoRKUM, L. D. 1978a. The influence of density and be- havioinal type on the active entry of two mayfly species (Ephemeroptera) into the water cohuun. Canadian J. Zool. 56:1201-1206. 1978b. Is benthic activity of .stream invertebrates related to behaviomal drift? Canadian J. Zool. .56:2457-2459. CoRKUM, L. D., p. J. PoiNTINC, A.ND J. J. H. ClBOROWSKI. 1977. The influence of current velocity and sub- strate on the distribution and drift of two species of mayflies (Ephemeroptera). Canadian J. Zool. .55:1970-1977. CiMMiNS, K. W. 1962. An evaluation ot some technicjues for the collection and anahsis of iienthic samples with special emphasis on lotic waters, .\mer. .Midi. Nat. 67:477-504. Elliott, J. .\1. 1971a. The distances travelled h\ drifting invertebrates in a Lake District stream. Oeco- logia 6:350-379. 1971b. Upstream movements of benthic in- vertebrates in a Lake District stream, J, .\nim. Ecol. 40:235-252. FiENBERC, S. E. 1978. The analysis of cross-classified cat- egorical data. NHT Press, Cambridge, Massachusetts. Horn, H, S. 1966. Measurement of "owrlap" in com- parative ecological studies. Amer. Nat. 1()():419_424. HiLTiN, L., B. SvENSso.N, A.ND S. Ulfstrani). 1969. Up- stream movements of insects in a south S\\ edish small stream. Oikos 20:55.3-.557. Ki.NNEDY, H. D. 1955. Colonization of a previously bar- ren stream section by aquatic invertebrates and trout. Prog. Fish-cult.' 17:119-122. Lahimobe, R. W., W. F. Childers, and C. HECKBorrE. 1959. Destruction and re-establishment of stream fish and invertebrates affected In drougiit. Trans. Amer. Fi.sh. Soc. 88:261-285. Mc l.AV, C. 1970. .\ theor\- concerning the distance tra\- elled bv animals entering the drift of a stream. ). Fish. Res. Bd. Canada. 27:.359-370. Mkbritt, R. \V., A.ND K. W. CiMMiNS. 1978. An in- troduction to the acjuatic insects of North .Ameri- ca. Kendall Himt Pub. Co., Dubuque, Iowa. MiNsiiMi.. C. W.. A.ND P. V. WiNcKB. 1968. The effect ot reduction in stream flou on inverteiirate drift, fkolog)- 49:580-582. Patbk K, R. V959. Aquatic life in a new stream. W ater and Sewage Work 531-5.35. RoBV, K. B., |. b. \kwb(U.d. and I). C, Kr\i\n, 1978, Effectiveness of an .irtiticial siilistiatc tor sam- January 1985 McArthur, Barnes: Utah Macroinvertebrates 123 pling niacroinvertehrates in small streams. Fresli- wat. Biol. 8:1-8. Scott, D. T., M. W. Carter, G. C. Bryce, and B. L. Joiner. 1974. RUMMAGE: a general data analy- sis system. Invited paper at the Amer. Soc. Qiial. Cont. and Eng. section of Amer. Stat. Soc, Rich- mond, Virginia. Sheldon, A. L. 1977. Colonization curves: application to stream insects on semi-natural substrates. Oikos 28:256-261. Str,\hler, a. N. 1957. Quantitative analvsis of water- shed geomorphologv. Trans. Amer. (leophvs. U. 38:913-920. TowNSEND, C. R., AND A. G. HiLDHEW. 1976. Field ex- periments on the drifting, colonization and con- tinuous redistribution of stream benthos. J. .Xnim. Ecol. 45:759-772. Ulfstrand, S., L. M. Nilsson, and A. Stercer. 1974. Composition and diversity of benthic species col- lections colonizing implanted substrates in a Swedi.sh stream. Ent. Scand. 5:115-122. Waters, T. F. 1964. Recolonization of a denuded stream bottom area bv drift. Trans. Amer. Fish. Soc. 93:311-315. Williams, D. D. 1977. Movements of benthos during the recolonization of a temporary stream. Oikos 29:306-312. Williams, D. D., and H. B. N. Hynes. 1976. The recolo- nization mechanisms of stream lienthos. Oikos 27:265-272. CHECKLIST OF THE MOSSES OF GRAND TETON NATIONAL PARK AND TETON COUNTY, WYOMING John R. Spence' .\bstr\ct.- a preliminary checklist of the mosses of Grand Teton National Park and Teton Covuitv. comprising 106 species, is presented, the following nine species are reported as new to Wyoming; Sphagnum s(iii(iriosiiiu Crome, Atrichum imdiilatiim (Hedw.) P.-Beauv., TortcUa Immilis (Hedw.) Jenn.. Bnjiim bicolor Dicks.. Fohlia obtusifolia (Brid.) L. Koch. CalUeroon sannentosum (Wahlenb.) Kindb., EiirJiynchiiim ore^auum iSull.) Jaeg,. Flaoiothcciwn pilifenim (Sw. ex C.J. Hartm.) B.S.C. and Lcscuraca stcnophijUa (Ren, & Card.^ Kindh. Although the vascailar plant flora of Grand Teton National Park and vicinity, in western Wyoming, is well known, basic floristic infor- mation on the bryophytes are scarce. Early collections in the state of Wyoming by C.L. Porter led to two papers on the mosses and liverworts, with 54 mosses listed from Grand Teton National Park (Porter 1933, 1935). More recently. Hong (1977) added many new liverwort species to the state flora and Churchill (1979, 1982) added several species of mosses new to Wyoming. Since the reports by Porter, no additions to the moss flora of the park have been report- ed. This paper is based on my own collec- tions from the park from 1978 through 1981, as well as collecting trips by Dr. E. Lawton and Mr. F. J. Hermann. Also, literature re- ports from various monographs have been included. The physiography of Grand Teton Nation- al Park and Teton County is rugged, with ex- tremes of relief greater than 2000 m, provid- ing a wide range of habitats available for brvophytes. The climate of the area has been characterized by Reed (1952), Shaw (1958), and Spence (1981). Major vegetation commu- nities include valley floor coniferous forest and sagebiaish steppe, marshes and lakeshore vegetation, riverine gallery forests, subalpine coniferous forest, subalpine meadows, alpine meadows, and alpine fell-fields (Reed 1952, Loope and Gruell 1973, Sabinske and Knight 1978, Spence and Shaw 1981). Bryophyte di- versity appears to be highest in moist east- facing canyons in the Teton Range, but most areas of the park have only been slightly col- lected. This checklist reports 106 species of mosses, representing about one-half the flora of the state. The large numbers of species Porter (1935) reported from Yellowstone Na- tional Park to the north of Grand Teton Na- tional Park suggests that further collecting will greatly increase the known flora. The checklist is arranged phylogenetically by family, using Corley et al. (1981), with the following exceptions: the Ditrichaceae are maintained as distinct from the Dicranaceae; the Polytrichaceae follow Smith (1971); the Scouleriaceae are separated from the Grim- miaceae according to Churchill (1981); and Platydictya jungermonnioides is maintained { = Amhleijstegium jungenyiannioides (Brid.) A. J. E. Smith in Corley et al. 1981). For North American endemics synonymy follows Lawton (1971). Collectors (including herbaria collections in which specimens are deposited) are noted as follows: P = Porter (RM); H = Hermann (variously distributed at nu- merous herbaria); L = Lawton (WTU); S = Spence (UBC); M = Manville (Grand Te- ton National Park herbarium at Moose, Wvoming). No attempt has been made to verify literature reports. Those species marked with an asterisk (") are apparently new reports for Wyoming, and were not re- ported in Porter (1935), Lawton (1971), or anv monographs and taxonomic works cited in this paper. Chec:klist of the Spec:ies FlUiS.lUW; S976. Sphagnaccac S}>h(i^nuiii nisscnii W Dsiiiii Crome S974. Departiiu'iit of Botany, UiiiviTsIt) B.C:.. Cainda V(iT2Bl 124 January 1985 Spence: W-i Mosses 125 TetrapliidiKfac Tctraphis prlhiruhi H.hIw. H2.5571; S973, PolytiicluKcae AtrirluiDi .sc/iii/ni! Avist. H25548,25765. "A. inuhihitnm (Hedw.) P.-Beauv. L1736. Fohitiuliastrum alpiuuw (Hedw.) G.L.Sniitli S291. /'. /,/r;//,MMitt.)C.L. Smith S352. rohllriiliuui coniiitiiiie Hedw. S318. /'. jmiiixiiniim Hedw. H2556(). /'. pilifrnim Hedw. H25559: S357. Hvixhauniiaceae Biixhdiniiia viridis (DC.) Movi'4. & \estl. H25577. Fissidentaeeae Fissiclens (idiantoides Hedw. LI 7.30. F. osmimdoide.s Hedw. P12()l; L17.3(). Dihichateae C'crdfddoi) juirpincti.s (Hedw.) Hiitl. S75. Distirhiiim capilhwcum (Hedw.) B.S.G. S.3.34. Dieraiiaeeae Dicnnumciski cindtd (Hedw.1 Lindh. ex Milde (Porter 1935). D. crispiila (Hedw.) Milde H25.5.54; S:3.33. Dicivniim rhahdocarpiiiii Sull. (Peterson 1979). /). sropariiim Hedw. H2576fi; S979. Pottiaeeae Bniocriithwphiillum ifciinirostniin (Hedw.) Chen H2557.5; S995. Dcsnuilodon latifoliiis (Hedw.) Brid. S.358. 'Toitclla hiiimli.s (Hedw.) Jenn. H25566. T. tortuo.sa (Hedw.) Limpr. P1166.1169. Toiiuhi norvegica (Web.) Wahlenb. ex Lindb. H25549. T. niialis (Hedw.) Gaertn., Meyer & Scherb. S354. Grininiiaceae C.iiininid (ili)rstris (Web. & Mohr) Schleicli. ex Hornsdi. H25.5.56.25584; L1728; S986. (•;. riatior Brucli ex Bals. & De Not. H25553. (;. iKiitiiuimiii Schimp. L1738,1759. G. ()((//(,s (Hedw.) Lindb. (Porter 19.35). (;. Inrhoplujlla Grev. LI 7.34, 1765. RdcoDiitiiiiiH canesccns (Hedw.) Brid. L1764. K. fdsririildrc (Hedw.) Brid. (Porter 1935). R. sudctiruw (Funck) B. & S. S992a. Schislidiiiiii «o«.s-.s!-.(( Sull. & Lesq. H25543. S. dpocdi-ptim (Hedw.) B. & S. H25544; L1729. S. rivuldir (Brid.) Podp. H255S3. Seouleriaceae Scoulcrid (upiaticd Hook. LI 7.53; H25582. Brsaeeae 'BnjuiiihirolorDivk^. H25545. B. cdcspiticiuni Hedw. S971. B. aipilldw Hedw. S313. B. lisdc De Not. S.337. B. psvudotriijuctrum (Hedw.) Gaertn., Mever & Seherli. H25574,25579; LI756. B. stnwtrkldtm C. Miiell. S.359. B. tiiihindtum (Hedw.) Turn. L1777,1782. B. ulioiiumim (Brid.) B.S.G. M3799. B. lici^clii Spreng. S996b. U'ptohniinn piiriformc (Hedw.) Wils. L17.33,1751; H25567. Poldid amldhisicd (Hohn.) Broth. Shaw 2704 (Shaw 19811. r. cdiiiptotidclwld (Ren. & Card.) Broth. Shaw 2692 (Shaw 1981). P. cnidci (Hedw.) Lindb. LI 737; H25568; S343. /'. dniininotidii (C. Muell.) Andr. Shaw 2695 (Shaw 1981). P. niitdns (Hedw.) Lindb. H25573; S980. "P. ohtusifolia (Brid.) L. Koch S989,993a. /'. proli<:,cni (Kindb. ex Limpr.) Lindb. ex .Xrnell Shaw 2702 (Shaw 1981). P. tundidv ]. Shaw Shaw 2709 (Shaw 1981). P. wdldvnln'roii (Web. & Mohr.) Andr. H25564. Roellid rocllii (Brotli. ex Roell) Andr. ex Crum S.329. Mniaceae Mniunt diizouicutn Amann L177(). .\/. spinulo.snm B.S.G. S980e. M. thonisonii Sehimp. S985a. Plagioiiiniiiiii cuspiddtum (Hedw.) T. Kop. M3802. RliizonniiiiDi nuionifoUiDii (Horik.) T. Kop. H25562,2558(); S97(). R. piincfdtum (Hedw.) T. Kop. M3788. Aulacomniaceac Aiddconiniuiu ditdio^iinuid (Hedw.) Schwaegr. (Porter 19,35). A. palustre (Hedw.) Scliwaegr. S969, Bartramiaceae Bartrauiid idiyphylld Brid. L1761. Phdonotis fontdud (Hedw.) Brid. L1766; S999. Orthotriehaceae Ainphidiiiiii ldj)p(>nictn)i (Hedw.) Schimp. Cain 467,5(UBC). Ortliotricluim dlpvstrc Hornsch. ex B.S.G. L1757. O. citptddtum Brid. (Porter 19.35). O. Idcvigdtum Zett. H255.50,2.5551; L1731, 1760,1768. O. nipestre Sclileich. ex Schwaegr. L1767. Fontinalaceae Dkhehpmi fdlcdtum (Hedw.) Myr. S972. Fontiudlis antipyreticd Hedw. S976b. F. hypnoidea C.j. Hartm. M3787. Climaciaceae Climdcium dincricdDinu Brid. L17.32. C. deudwklcs (Hedw.) Web. & Molir S314. Leskeaceae Lcscurdcd incuivdtd (Hedw.) Lawt. L1783. L. rddk-o.sd (Mitt.) Monk. H25558,25764; L1744-8. °L. fitenophylhi (Ren. & Card.) Kindb. Flowers 38.33A (UBC). ■ Ambleystegiaceae Ambleystegium serpens (Hedv\'.) B.S.G. (Porter 1935). °CdUieroon sdnncntostiiu (Wahlenb.) Kindb. S996. C. strdiiiincum (Brid.) Kindb. S981. Crdtoncuion contnuikitinu (Hedw.) Roth L1778. C. fdkiuum (Hedw.) Spruce H25563; S978. Drvpdnoclddus aduncus (Hedw.) Warnst. S978b. D. cxdimuUiius (B.S.G.) Warnst. S978a. D. uminatiis (Hedw.) Warnst. H25.5.52; L1726, 1754,1762,1781. Hyarohypnuiii hcstii (Ren. & Brvhn ex Ren.) Holz. L1769;S998. //. ochidccum (Turn, ex Wils.) Loeske Cain 4.547 (Jamieson 1976). PIdtydictyd jdii'^ciDidniuoidcs (Brid.) Crum H25572. Braclntlicciaceae Brd,liiitlirridi>i alhwdns (Hedw.) B.S.G. H25546. B. crytltionluzon B.S.G. (Porter 19.35). B. fcndlcri (Sull.) Jaeg. L1772.1774. B. friokUnn (C. Muell.) Besch. H25578. B. hviildic B.S.G. (Porter, 19.35). B. sdlebwsum (Web. & Mohr) B.S.G. H25547. 126 Great Basin Naturalist Vol. 45, No. 1 B. sUirkei (Brid.) B.S.G. S982. B. velutintim (Hedw.) B.S.G. L177.3. ° Eurhijnchium oreganum (SiiU.) Jaeg. M3797,.3792. Plagiotheciateae hoptenjoium pulchelhnn (Hedw.) Jaeg. (Ireland 1969). Plagiothccium dcnticuhitttm (Hedw.) B.S.G. H25565. °P. piliferum (Sw. ex C.J. Hartm.) B.S.G. M3796. Hypnaceae Hiipnum cupressiforme Hedw. S183. H. lindbcrnii Mitt. L1727. H. rcvolutum (Mitt.) Lindb. H2,5557; S335,356. Acknowledgments Help in determining difficult collections was provided by W. B. Schofield, T. Mcin- tosh, and H. Ochi. Dr. E. Lawton and F. J. Hermann generously provided information and species lists from collecting trips in the park. Literature Cited Churchill, S. P. 1979. Mosses of the Great Plains 10. Additions to Nebraska and the Black Hills of South Dakota and Wyoming. Brvologist 82:72-7,5. 1981. A phylogenetic analysis, classification and synopsis of the genera of the Grinimiaceae (Mus- ci). Pages 127-144 in V. A. Funk and D. R. Brooks, eds.. Advances in cladistics: proceedings of the first meeting of the Willi Hennig Society. New York Botanical Garden. 1982. Mo.s.ses of the Great Plains VIII. Additions. Bryologist 85:218-221. CoRLEY, M. F. v., A. C. Crundwell, R. Dull, M. O. Hill, and A. J. E. Smith. 1981. Mosses of Europe and the Azores: an annotated list of species, with synonyms from the recent literature. J. Brvologv 11:609-689. Ho.NG, W. S. 1977. An annotated checklist of the Hepati- cae of Wyoming. Bryologist 80:480-491. Irel.^.nd, R. R. 1969. A taxonomic revision of the genus riagioihecium for North .America, north of Mexi- co. National Museums of Canada, Publications in Botany, No. 1. Jamieso.v, D. W. 1976. A monograph of the genus Hij- giohi/pniim Lindb. (Musci). Unpublished dis- sertation. Univ. of British Columbia. Lawto.n, E. 1971. Moss flora of the Pacific Northwest. Hattori Botanical Laboratory, Nichinan, Japan. LooPE, L. L., AND G. E. Gruell. 1973. The ecological role of fire in the Jackson Hole area, north- western Wyoming. Quat. Res. 3:425-443. Peterson, W. 1979. A revision of the genera Dicrantim and Orthodicranum (Musci) in North America north of Mexico. I'npublished dissertation. Univ. of Alberta. Porter, C. L. 1933. The bryophytes of Wyoming. Part I. Hepaticae. Bryologist 36:5-8. 1935. Bryophvtes of Wyoming. Part II. Hepaticae (concluded) and Musci. Brvologist .38:101-114. Reed, J. F. 19.52. Vegetation of'jackson Hole Wildlife Park, Wyoming. Amer. Midi. Nat. 48:561-582. Sabinske, D. W., and D. H. Knight. 1978. Variation within the sagebrush vegetation of Grand Teton National Park, Wvoming. Northwest Sci. 52:19.5-204. Shaw, \. J. 1981. .\ taxonomic revision of the prop- aguliferous species of Pohlia (Musci) in North America. J. Hattori Bot. Lab. 50:1-81. Shaw, R. J. 1958. Vascular plants of Grand Teton Na- tional Park. Amer. Midi. Nat. 59:146-166. Smith, G. L. 1971. Conspectus of the genera of Poly- trichaceae. Mem. New York Bot. Garden 21:1-83. Spence, J. R. 1981. Comments on the cryptogam vegeta- tion in front of glaciers in the Teton Range. Bryo- logist 84:564-.568. Spence, J. R., and R. J. Shaw. 1981. A checklist of the alpine vascular flora of the Teton Range, Wvo- ming, with notes on biology and habitat prefer- ences. Great Basin Nat. 41:232-242. ECOLOGICAL INVESTIGATION OF A SUSPECTED SPAWNING SITE OF COLORADO SQUAWFISH ON THE YAMPA RIVER, UTAH Vincent A. Lamaira', Marianne C. Lamarra', and John G. Carter .'Vbstract.— On 5 July 1981, 13 adult Colorado squawfish were found in spawning condition at river mile 16.5 in the Yampa River, a major tributary to the Green River. An investigation was undertaken to quantitatively describe this section of the river to gain insights on the spawning requirements of this endangered species. The substrate at the suspected spawning .site was cobble with large interstitial spaces devoid of organics, silts, or clays. It appeared that larvae of several fish species utilized these cobble areas and the associated voids. Diurnal sampling indicated that larval drift occurred between 0100 and 0125 hours. Substrate size also appeared to be a dominant factor in fish distribution. Feeding intensities of these fish corresponded to macroin vertebrate drift. Spawning by the endangered Colorado squawfish {Ptychocheilus liicius) has not been documented in nature. On 1 July 1981, the Colorado River Fisheries Project [CRFP] of the U.S. Fish and Wildlife Service captured 13 adult Colorado squawfish in spawning condition in one localized area of the Yampa River, known as Cleopatra's Couch, 16.5 miles from the Green River confluence. This gave us the opportunity to intensively sample the immediate area in an effort to locate the site and to document conditions in a natural spawning habitat. This survey was under- taken 24-26 July 1981. The objectives of this study were: 1. To characterize this river ecosystem at the sus- pected squawfish spawning site. 2. To document the interactions of the abiotic envi- ronment uith the biotic community. 3. To document the interactions among and be- tween trophic levels. 4. To compare the structure and hmction of the eco- system at this site with that of other similar sites in adjacent tributaries where squawfish were not located. The above objectives were accomplished by intensively surveying the area where spawning squawfish were observed (mile 16.5) and comparing these data to another similar Yampa River site (mile 18.0) appar- ently not used for spawning during 1981. Furthermore, these data are compared to a structurally similar area in the White River, a major tributary of the Green River, where successful squawfish spawning has been sus- pected but not documented. The results of this study are important for recognizing po- tential squawfish spawning sites and for the future maintenance of the physical and bio- logical integrity of this important Yampa River site in the face of future development within the upper Green River Basin. Materials and Methods Field Two field seining sites were established (mile 16.5 and 18.0), and at each of these samples were taken from several different habitat types, e.g. riffles, eddies, and runs. Five seining sample locations were chosen for river mile 16.5 and seven for river mile 18.0. All fish samples were collected with seines (4 x 30 ft, with 1/4-in mesh). Upon capture, all fish were identified, counted, and measured for total lengths. Fish less than 100 mm were slit abdominally and placed in a 10% formalin solution. The diges- tive tracts of fish greater than 100 mm were removed and preserved in 10% formalin. A benthic sample (modified Surber sampler) was collected at each seine haul location. Insects floating in the water column were sampled with drift nets (27.5 cm diameter, 1.0 mm mesh size) at each seine sample loca- tion. These nets were anchored to the river bottom and set in the water column approx- imately 1 m upstream from the area to be seined. Each net was set for approximately 4 hours prior to fish collection and removed 'Ecosystem Research Institute, 975 South State Highway, Logan. Utah 84321. 127 128 Great Basin Naturalist Vol. 45, No. 1 just prior to seining. Drift material was pre- served in 10% formalin. Physical data at each seining sample loca- tion (water temperature, water depth, cur- rent velocity, and dominant substrate type) were also determined. A sediment core sample was also collected at each seining sample location using a 4.6-cm-diameter cor- ing device. These interstitial sediment sam- ples were transported to the laboratory and analyzed for size fractions and organic content. Three cross-river transects at river mile 16.5 and two at river mile 18.0 were estab- lished. Water depth, current velocity (0.6 dis- tance from the bottom), and dominant sub- strate were recorded at transect points every 2 m across the stream. At five equidistant points on each transect, benthic macroin- vertebrates and chlorophyll a (periphyton) samples were collected. In areas deeper than 1 m, SCUBA was used to sample. All chloro- phyll samples were stored on dry ice for transportation. Laboratory In die laboratory fish stomachs were re- moved, and the contents were placed in vials containing 70% ethanol. In fishes with dis- tinct stomachs, such as Ictalurids and some Cyprinids, only the contents of the stomach were taken. In fishes with poorly defined stomachs, such as Catostomids, the anterior portion of the gut from the esophagus to the first bend or distinct constriction of intestine was taken for analysis. Total displaced vol- umes of stomach contents were measured with a graduated volumetric tube to the nearest 0.01 ml. Contents were then identi- fied, counted, and measured to the nearest millimeter. Drift net and benthic samples were floated in a sugar-water solution and poured through a 0.25 mm sieve to separate the organic and inorganic portions. Insects and fishes were then manually sorted from all other organic materials. These items were preserved in 70% ethanol, identified, and counted. The remain- ing debris was dried at 105 C for 24 hours and weighed on a Sartorius balance. All ben- thic insects were additionally measured to the nearest millimeter. Interstitial substrate samples collected in the field were dried at 105 C for 24 hours. The sediments were then sieved into five size fractions (see footnote in Table 1), which were each weighed on a Sartorius toploading balance. The 0.25-mm size fraction was re- dried for 24 hours at 105 C, cooled in a des- iccator, and weighed to the nearest 0.01 mg on an analytical balance. These samples were then ashed in a preheated muffle furnace at 550 C for 20 minutes, cooled, and reweighed. Table 1. Physical and hiolos^ical data sumiiiaiy tor five transects on the V; and 18.0. Figures are an average of river cross-sections with standard deviation. between miles 16..5 Sediment Doininaut as a percent Chi a Depth NelocitN substrate Transect (mg/m-) (m) (m/sec) (cm) A B mean 12.2 0.9 0.63 9.0 42.6 6.0 ' SD 11.9 0.2 0.2 3.4 35.5 7.9 mean - SD 31.fi 1.4 0.3 0.3 -0- 6.0 15.1 0.9 0.2 0.4 -0- 14.7 ^ mean •^'» SD 4.6 0.3 2.4 7.5 (i7.() 20.0 — O.I 1.0 -0- 11.3 4.2 mean •^'' SD 30.0 0.4 2.1 5.8 70.0 18.0 2.3 0.1 0.7 2.9 lO.S 5.6 mean 7.8 2.3 0.6 6.3 27.2 0.2 "* SD 13.1 1.7 ().(i 7.9 43.2 0.5 mean 5 SD 8.8 0.3 3.5 9.0 74.2 13.4 4.6 0.2 1.6 3.4 14.5 9.7 'Sediment size fratlions (mm): A = >14.7 B = 14.7 - 4.0 C = 4.0 - 0.5 D = 0.5 - 0.25 E = <0.25 January 1985 Lamarra et al.: Squawfish 129 POSTBUNOrF Fig. 1. Discharge levels in the Yanipa River chiriiig 1980 'iven. 1 — JDTT 1 Turr — I — ^m — i — urr and 1981. Sample times tor the CHFP and this stndv are The weight loss on ignition was calculated as organic content. Periphyton biomass was determined by scraping a flat area 4.6 cm in diameter from each rock collected. This scraped material was extracted for chlorophyll in 250 mis of 90% acetone. Core samples containing sand or silt were placed directly into 250 ml of acetone for extraction. Chlorophyll a was de- termined with a Turner Model III fluorome- ter (American Public Health Association 1980). Results Transect Sites Steep canyon walls always occur on at least one shore of the river between miles 16.5 and 18.0. Cobble-covered islands and Table 1. Continued size fraction Total Macroinve tebrates of the total Total detritus Total biomass C D E Organics (gms/m-) (#/m-) (gms/ni") 6.2 24.8 20.2 2 9.2 69.2 1.0 4.1 25.0 26.2 .03 16.9 48.1 1.0 15.3 .3:3.7 45.3 0.6 15.6 186.5 1.0 25.6 25.3 28.3 0.8 20.7 203.2 1.3 10.5 2.5 -0- -0- 1.4 762.0 4.5 6.4 0.7 -0- -0- 1.6 879.6 9.0 7.7 4.0 0.3 -0- 2.7 2.33.3 8.1 4.2 2.0 0.6 -0- 3.3 24.7 2.1 10.2 .38.6 23.8 0.3 12.2 53.3 1.8 9.1 26.8 .30.6 0.3 16.7 28.9 2.6 7.0 1.2 0.8 -0- 1.1 302.9 13.8 4.7 1.3 0.5 -0- 0.9 193.0 7.3 130 Great Basin Naturalist Vol. 45, No. 1 Cleopatras Couch River Mile 16.5 SH7 Fig. 2. Locations ot studv sitt represents transect locations. miles 16.5 and 18.0 on the V; represents seint il locations. T shorelines had been recently exposed by re- duced flows (Fig. 1). Beaches within zones of deposition were composed of fine sand or or- ganic silt-mud deposits. Backwaters were of- ten marshy with loose organic substrates. At river mile 16.5 and 18.0, transects and .seining sites were established in areas that re- flected the maximiun number of habitat types foimd in the river (Table 1). Although a compari.son of the water depth, water veloc- ity, and substrate size for these transects in- dicated that Tran.sects 1, 3, and 5 represented mns or riffles of various sizes and Transects 2 and 4- were indicative of deep runs or pools, it was difficult to rigidly classify the habitats. However, the riffle transects tended to have higher average velocities and were shallower in depth, whereas the run-pool transects were deeper with lower velocities. A comparison of the biological data in- dicated the highest average chlorophvll a values (31.6 ± 6.7 mg/vcr) were found in Transect 2 and the lowest (7.8 ± 5.9 mg chla/m^) in Transect 4. Maximum average bioma.sses (N = 5) of macroinvertebrates were found in Transect 3 (10.8 ± 4.07 gms/m^) and Transect 5 (13.79 ± 3.27 gms/m^) and corresponded to the highest water velocities. January 1985 Lamarra et al.: Squawfish 131 A comparison between Transect 2 (pool) and Transect 3a-3b (riffle) indicated marked differences in these two habitat types. The major abiotic differences between these two transects were the dominant and interstitial substrate sizes. Transect 2 had sand and silt as the dominant substrate, with interstitial sub- strates less than 0.5 mm in size. At Transect 3a-3b, the dominant substrate was cobble, with 88% of the interstitial substrates greater than 4 mm. The cobble and absence of silt on this transect provided substantial interstitial voids. CRFP (1982) observed Colorado squawfish moving from the area of Transect 2 into and out of the cobble area of Transect 3b in ap- parent spawning behavior (hashed area in Fig. 2). Dominant substrate type and inter- stitial characteristics at Transect 3a-3b may be critical in the apparent preference of this habitat by spawning Colorado squawfish. At river mile 18.0, a physically similar habitat was found (Transect 5 in Table 1). However, this transect had much higher velocities (>4.5 m/sec) as compared to Transects 3a- 3b (2.1-2.4 m/sec), which may have pre- cluded this area from being used for spawn- ing activities during 1981. Seining Sites Five locations at river mile 16.5 were sam- pled intensively for fish, larval fish, macroin- vertebrate drift, and benthic macroinverte- brates (Fig. 2). Each of the five seining locations were sampled five times at four- hour intervals. At river mile 18.0, seven loca- tions were sampled twice over an eight-hour period. A comparison of the physical and bio- logical data collected at the seining locations (Table 2) indicates that the ranges of these parameters fall within the ranges observed on the transects (Table 1). Nine species of fish were captured during this survey (Table 3). Three species are endemic, two are native, and the remaining five are nonnative. Among the endemics, the most abundant were round- tail chubs (39% of the total). One fish tenta- tively identified as a humpback chub was captured. Redside shiners were the most abundant nonnative (15%), followed by red shiners (13%). The ratio of endemics and na- Table 2. Physical and liiological data collected at tive seining sites from river mile 16.5 on the Yampa River 24-26 July 1981. Area Dominant Substrate Macroinvertebrates 16.5 SH, SH, Location Time sampled Depth Velocity substrate organics Density Biomass RM SH (hours) (m'-') (meters) (m/sec) (cms) (gms/m-) (#'s/m-) (gms/m-) 1620 2030 (K)0() 0910 1435 1640 2100 (K)20 0935 1510 1700 2100 (K)40 0955 1535 1715 2125 0100 1010 1605 1745 2140 0125 1030 1630 0.44 0.02 2.5 0.26 0.88 0.23 7.5 1.05 0.38 1.07 0.75 0.06 0.78 0.14 0.75 1..55 0.67 0.30 7.5 0.25 50 312 1.252 18.43 5.952 132 Great Basin Naturalist Vol. 45, No. 1 1 T 1500 2000 0100 0600 1100 1600 TIME Fig. 3. Distribution of fish captured at five seining sites at river mile 16.5 on tlie Yanipa Hiver over a 24- hour period 24-26 July 1981. tives to nonnatives was 2:1 by density. The distribution of these fish between seining sites and over time was not constant (Fig. 3). Diur- nal movements observed at mile 16.5 were primarily by roundtail chubs. The temporal distribution of fish captured at the lower end of the cobbled-riffle area (SH2 and SH3) was the lowest observed at any site (Fig. 3). It was apparent that fish were not moving into or out of this fast riffle area. A comparison of all seining sites be- tween 1620 and 1745 hours indicates that maximum density of fish was found at SH4 (0.44 fish/m^ sampled). Subsequent seine hauls at this site showed continuously re- duced densities over the 24 hours sampled. A comparison of the other sites (Fig. 3) in- dicates the opposite pattern at the upper site (SH5), with no change in the intermediate seining location (SHI). It was believed that the pattern observed resulted from fish mov- ing out of the deep pool (SH4), past the shal- low run (SHI), and towards SH5, which was located at the lower stretch of a riffle area. Macroinvertebrate and larval fish drift samples were taken at each seining site. A comparison of the drift entering (SH4) and leaving (SH2 -I- SH3) the riffle area, where squawfish were at their highest observed con- centrations, indicates that substantial num- bers of macroinvertebrates and larval fish originated within this cobble habitat (Fig. 4). For example, the diurnal pattern of drift (macroinvertebrates and larval fish) reached a Tahle 3. Incidence of food July 1981. tish ■aptured on the Vanipa Ri\er by seii le 16.5 and 18.0 24-26 Species Young of the year Fish species code w/food w/o food Bhiehead sucker BH 1 0 CatostoDius (li.s((>lx>lus Fiannelmouth sucker FM 5 1 Cutostoiniis latipin n is Fathead minnow FH 0 0 Phnephalus pivmclas Red shiner RS 0 0 S'otropis liitrcDsis Hedside shiner RD 0 0 Riclt(ir(ls(>uiii\ halted Ins iioundtail chub RT 36 4 C'.ila rohtistd .Sand shiner SS 0 0 S'otropis stnnniucits Speckled dace SD 2 0 Rliiuirhthtisosciihis Channel catfish CC 1 0 Icfdhinis })iiiHtiittis January 1985 Lamarra et al.: Squawfish 133 maximum between 2125 and 0100 hours. No larval fish and only 500 macroinverte- brates/hour were captured entering the area, whereas over the same time period five larval fish/hour and over 2000 macroinverte- brates/hour were captured leaving this riffle area. Although drift nets were located at each seining location, larval fish were collect- ed only in nets below this cobble area. Five species of larval fish were identified, includ- ing: speckled dace, roundtail chub, channel catfish, flannelmouth sucker, and carp. This may indicate that this cobble area habitat may have been used for spawning or as a nur- sery area by species other than squawfish. In earlier studies, squawfish larvae were collect- ed between river mile 12.1 and 0.1 during 24, 25, and 26 July (Haynes and Muth 1982). These larval fish (9.0-13.0 mm) corresponded closely in age with the dates when adult fish were observed to be spawning at river mile 16.5 (Tyus et al. 1981). Feeding habits were determined by com- paring benthic and drift samples to fish stom- achs for the locations and times fish were captured (Strauss 1979). Feeding intensity was determined by using a percentage of stomach volume filled. In Table 4, a com- parison is given between the major macroin- vertebrate components in the drift or benthos to the major components of each species diet (dominant four fish species only). These data indicate, and it is reflected in Strauss (1979) Electivity Index (Table 5), that the two na- tives, roundtail chub and speckled dace, have a much richer diet relative to available food when compared to redside and red shiners. The latter two introduced species may be op- portunistic and appear to eat food items in the proportion that they are available. Feed- ing intensity as determined by percentage fullness indicates that red shiners had a signif- icantly greater percentage of stomach con- tents (62.6 ± 4.5 %) when compared to all other species except sand shiners (49.4 ± 8.9 %). Analysis of variance shows that full- ness of all other species did not differ signifi- cantly. Comparing all fish species combined with time (Fig. 5) indicated that fish cap- tured between 1950 and 2140 hours were sig- nificantly fuller (56.2 ± 3.6 %) compared to those of the other four time periods (41.7 ± 2.4 %). This time period corresponded to maximum invertebrate drift (Fig. 4). Discussion An objective of this study was to determine the unique physical and biological features that made Yampa River mile 16.5 attractive to spawning Colorado squawfish. Ideally, the data presented here should have been collect- ed during early July, when spawning fish were observed at the site, but logistical prob- lems prevented this. However, comparative data show similar river conditions between the suspected spawning time of 5 July and this survey period, 24-26 July. During a Table 3. Continued ilts Total fish captured %of total Subadiilts/Adi YOY SA/A Grand total CRFP w/food vv/o food (1982) 8 0 1 8 9 3% 4% 6 0 6 6 12 4% 5% 1 0 0 1 1 <1% 6% 36 2 0 38 38 13% 16% 40 4 0 44 44 15% 8% 69 5 40 74 114 39% 32% 10 2 0 12 12 4% 3% 50 8 2 58 60 20% 21% 0 0 1 0 1 <1% 2% 134 Great Basin Naturalist Vol. 45, No. 1 Below(SH2fSH3) Above(SHi4) Below(SH2+SH3) Above(SH4) 1500 2000 0100 0600 1100 160 0 TIME FiR. 4. Densities of clritt iiiacroiiiveitehiates (r:'s/houi> and lai\al tislies (ir's/lunir) captured at seleett sites (SH) at river mile 1(15 on tlie Yanii)a River. Data eolleeted 24-2(i JnK 19S1. January 1985 Lamarra et al.: Squawfish 135 ^ 1631-1745 1950-2140 — T 0-0125 TIME 1 I 0910-1030 1435-1630 Fig. 5. Feeding intensity (expressed as a percentage of fullness of stomach volume) of all fish compared over time at Yanipa River mile 16.5. Fish captured between 1950 and 2140 hours were significantly hiller than fish capt\ired at other times. Data collected 24-26 July 1981. CRFP (Tyus et al. 1981) survey, at Transect 2, the width of the river was estimated at 70 m compared to 54 m measured for this sur- vey. Both sample periods occurred during postrunoff, with an intervening river stage decrease of approximately 0.6 m and a corre- sponding flow reduction of 180 cfs (480 cfs to 300 cfs). The biological community in this study was similar to that documented in previous studies of the Yampa River. The dominant macroinvertebrates (Ephemeroptera: Rhithrogena, Baetis; Plecoptera: Isogemis; Trichoptera: Hydropsche, given in order of abundance) were also previously reported by Bailey and Alberti (1952) and more recently by Carlson et al. (1979) and Annear (1980). The density of macroinvertebrates also corre- sponded to the levels reported by Annear (1980). The fish species composition of the Yampa River within Dinosaur National Park appears to have changed dramatically in recent years. Carlson et al. (1979) noted that Holden (1973) reported bonytail chub, largemouth bass, bluegill, sunfish, and walleye within this stretch of river. However, in 1979 these spe- T.ABLE 4. Percentage composition of macroinvertebrates in the diet of fishes (A) and in benthic or drift samples (B) collected in the Yampa Hiver near river mile 16.5 24-26 July 1981. (A) Percentage in each fish species diet Speckled Redside Red Roundtail All Orders shiner shiner chub dace species Ephemeroptera 92.7 92.6 78.6 87.5 85.8 Plecoptera — 00.1 — 00.0 Trichoptera 00.1 — 05.1 04.2 03.1 Coleopteia Diptera __ __ — 02.3 00.6 05.6 01.5 13.2 05.8 07.8 Hymenoptera 01.3 04.3 02.9 — 02.2 Lepidoptera __ — — 00.2 00. 1 Hemiptera 00.4 01.6 00.1 00.1 00.4 Hydracrina — — 00.1 — 00.0 (B) Percentage in each sample type Orders Benthic Drift Ephemeroptera 94.26 99.58 Plecoptera 00.87 00.09 Trichoptera 04.25 00.22 Coleopteia 00.04 00.01 Diptera 00.12 00.08 Odonata — — Lepidoptera 00.08 — Hemiptera 00.09 00.00 Momoptera 00.02 00.00 Arachnida — ()().()() Megaloptera — ()().()() Oligochaeta 00.19 — 136 Great Basin Naturalist Vol. 45, No. 1 cies were not found, but plains killifish and unfavorable to nonnatives. Longitudinal data sand shiners were collected (Seethaler et al. collected on the White River, a major tribu- 1979). Data from this study is most similar to tary to the Green River, (Lamarra and Carter that of Carlson et al. (1979). A comparison 1981) show a high correlation between the with the most recent collections (Colorado presence of endemic fishes and coarse sub- River Fishery Project 1982) indicates a high strates (riffle areas). Conversely, sandy areas degree of similarity except for: (1) the ab- with an absence of riffle habitats in the lower sence of squawfish and the presence of a pos- stretches of the White River appear to be sible humpback chub in our collections and avoided by the endemic community. Valdez (2) the higher densities of redside shiners in and Clemmer (1982) also report a similar our collections. relationship between substrate size and It has been noted by Crosby (1975), Lani- roundtail and humpback chub. Although no gan and Berry (1979), Carlson et al. (1979), longitudinal data were collected in this study, and Valdez and Clemmer (1982) that the pro- a low proportion of nonnatives to endemics portion of nonnative to endemic fishes in- and natives (1:2) was found in the Yampa creases as the major tributary streams ap- River site, suggesting a similar correlation, proached their confluences with the Green Fish species distribution, food selection, and River. Lanigan and Berry (1979) attribute the feeding intensities were determined by means higher proportion of nonnatives to a decrease of a diurnal sampling scheme for fish, drift in habitat diversity, whereas Valdez and insects, and larval fishes. These data have Clemmer (1982) have suggested that riverine provided insights into the structure and func- habitats occupied by native fishes are often tion of this ecosystem. It is interesting to note Table 5. Calcvilated liiioar Electivit\- Index (Strauss 1979) for the dominant four fish species in the Yanipa Ri\er near river mile 16.5. Eleetivitv comparisons are made for benthic (A) and drift (B) organisms. Data collected 24-26 July 1981. Redside (A) Benthic Speckled Red Roundtail All Orders shiner shiner chub dace species Ephemeroptera -.016 -.017 -.157 -.068 -.085 Plecoptera -.009 -.009 -.008 -.009 -.009 Trichoptera -.042 -.043 .008 -.001 -.012 Coleoptera .000 .000 .000 .023 .{K)6 Diptera .055 .014 .131 .057 .077 Hymenoptera .013 .043 .029 .000 .022 Lepidoptera -.001 -.001 -.001 .()()() .000 Hemiptera .003 .015 .000 .()()() .003 Homoptera .000 .000 .000 .000 .000 Annelida -.002 -.002 -.002 -.002 -.002 Hydracrina .000 .()()() .001 .000 .(M)0 Redside (B) Drift Speckled Red Roundtail All Orders shiner shiner chub dace species Ephemeroptera -.069 -.070 -.210 -.121 -.1.38 Plecoptera -.(K)l -.001 .(K)0 -.001 -.001 Trichoptera -.001 -.002 .049 .040 .029 Coleoptera .000 .000 .()()() .02.) .006 Diptera .055 .014 .131 .057 .077 Odonata .000 .000 .()()() .000 .(K){) Hymenoptera .013 .043 .029 .000 .022 Lepidoptera .000 .000 .()()() .002 .001 Hemiptera .003 .()\5 .000 .000 .00.) Homoptera .000 .()()() .000 .000 .000 Arachnida .000 .0(M) .()()() .000 .(H)0 Megaloptera .000 .000 .()()() .000 .000 Hydracrina .000 .000 .(K)l .(KM) .(H)() January 1985 Riffle Lamarra et al.: Squawfish 137 Pool Legend White River Yompa River Size Fractions (mm) Size Fractions (mm) Fig. 6. Distributions of interstitial sediments (percentage of total weight) for representative riffle and pool tran- sects in the White River as compared to representative pool riffle transects in the Yampa River. The smallest two size fractions that are causing armoring in the White River sediments are 25%-.30% of the total sediments. These fractions in tlie Yampa River are only 4% of the total. that larval fishes were captured only in the vicinity where squawfish were collected and that their movement corresponded to macro- invertebrate drift (2125-0100 hours). Five species of larval fishes emerged from this cobble site. Furthermore, inspection of the data collected at each seining site indicates a potential movement of fish from this same area at approximately the same time (Fig. 4), suggesting that the habitat was utilized by species other than Colorado squawfish. An important characteristic of this cobble area was the interstitial voids that contained little or no organics. The difference in drift above and below tliis area strongly suggests that larval fishes utilize these voids. Beames- derfer and Congleton (1982) observed the northern squawfish {Ptychocheilus orego- nensis) spawning in similar cobble and gravel sites in the St. Joe River, Idaho, where there was an absence of small substrates (sand) with adhering eggs up to 15 cm below the sub- strate surface. Prewitt et al. (1982) indicated that other similar habitats exist in the White River. However, their analysis of physical habitat and stream flow requirements for spawning squawfish considers only substrate size, not interstitial particles or voids. In the White River, the armored sediments (Fig. 6) prevent interstitial voids from developing, thus making the cobble habitat at Yampa River mile 16.5 unique in comparison to ob- served sites in other tributaries. It should be noted that Haynes and Muth (1982) did col- lect larval squawfish at Yampa River mile 17.8 on 14 August 1982. These fish were esti- mated to be 25 days old, indicating another spawning site above river mile 16.5. The physical factors that regulate the dis- tribution of organisms in stream environ- ments are varied. Such factors as nutrients (Hynes 1970), turbidity (Mann et al. 1972), temperature (Sprules 1947), light (Westlake 1966), and water velocity (Mclntire 1966) -f 1 1.0 SUBSTRATE (cm) .30 QRD 1 1.0 SUBSTRATE (cm) 10 Fi«. 7. Distribution ot tisli siil.stial, '!"■' li lictwocn ri\i'r miles 16.5 and 18.0 2 t-2() Inlv UJ.SI. Sec Tai)lc .? lor snccu's codes pvi have been shown to have profound effects on the river biocenosis. In this study, certain physical factors were also shown to be impor- tant in determining the distribution of the aquatic community. A comparison of the ma- jor trophic levels (primary producers, detritus levels, macroinvertebrates, and fish) indicates that distribution mav be related to substrate size (Fig. 7). The same general pattern ap- peared for all of the major groups of organ- January 1985 Lamarra et al.: Squawfish 139 isms: avoidance of sand (0.25 cm) and selec- tion for either small substrate (<0 .075 cm) or larger substrate (2.5 to 10 cm). Exam- ination of the interactions between these biotic components and substrate size appears to be consistent with the findings of Cum- mins and Lauff (1969). The distribution of fish across the various substrate sizes differed by species. The en- demic fish appeared to avoid sand and gravel and select from intermediate to large sub- strates. Within the nonnative species, sand shiners appeared to prefer environmental conditions associated with the smallest sub- strate, whereas red and redside shiners fa- vored intermediate (2.5 to 7.5 cm) substrates. All species avoided sand. The mechanisms causing the above observed distributions of endemic and nonnative fishes are unknown; however, the similar distributions of the fish, periphyton, and macroinvertebrates with substrate size indicate important interactions. As noted previously, the density of the riv- er biocenosis changed markedly over differ- ent substrate types; however, as Carlson et al. (1979) have stated, the distribution of organ- isms may be more influenced by com- binations of the effect of physical and biotic characteristics than by physical factors alone. Our data indicate these types of inter- relationships are important in the Yampa River. For example, no statistical relationship was found between the biomass of macroin- vertebrates and the biomass of periphyton or detritus; however, a significant relationship was found between water velocity and mac- roinvertebrate biomass (r^ = -H.45;p = .01). This relationship was believed to be indirect because the dominant benthic macroinverte- brate functional group was found to be fine particulate collectors or filter feeders (98% of the total biomass) that may be responding to areas of high food availability (high flow). An attempt has been made in this report to quantitatively describe the physical and bio- logical environment at a suspected spawning site for Colorado squawfish. The Yampa Riv- er at mile 16.5 was dominated by loose cobble substrate in association with large sandy pools. This site was also found to be free of small interstitial particles and organic material. This area appeared to be heavily used by several species of larval fishes and actively feeding adults. The combination of well-washed coarse substrate (cobble), abun- dant food (both drift and benthos), and adja- cent areas with slow, uniform laminar flow may be critical factors in determining pre- ferred reproductive sites of Colorado squawfish. Conclusions 1. A dominant abiotic factor appeared to be substrate type and associated inter- stitial sediment size fractions. a. Yampa River mile 16.5 was domi- nated by cobble substrate with in- terstitial voids containing little or no organics. b. Associated with these cobble areas were sand-substrate pools. c. Periphyton, macroinvertebrates, and fishes were found in lowest densities in or on sandy substrates. d. Highest densities of periphyton, macroinvertebrates, and fishes were found over substrates smaller than boulders (except sand). e. Roundtail chubs were dominant among the endemic fishes and pre- ferred cobble substrate. f. Nonnatives were dominated by redside shiners and red shiners, and both species were most abun- dant over cobble substrates. g. Sand shiners were the only species to demonstrate a preference for silt substrates. 2. Samples showed nocturnal hours to be important periods of biological activity. a. Periods of highest macroinverte- brate drift occurred between 1500 and 0200 hours. b. Drifting larval fish were collected only after 2000 hours, with the peak occurring at 0100-0125 hours. c. Drifting larval fish appear to have originated in a cobble substrate with substantial interstitial voids. d. Maximum feeding intensity of fish captured occurred between 1950 and 2140 hours. 140 Great Basin Naturalist Vol. 45, No. 1 e. Fish were found to move from the sandy-pool area into the lower sec- tions of riffles during the same peri- od of time when maximum feeding intensity was observed. Acknowledgments The authors acknowledge the White River Shale Oil Corporation for funding this study, Chuck McAda of the Colorado River Fish- eries Project for his assistance in getting sam- ples out of the Yampa River, and Dr. Richard Valdez for his review and critique of the manuscript. Literature Cited Amehk AN Fi ULic Health Association. 1980. Standard niethixls tor the examination of uater and waste water. APH.\. Washinuton, D.C. Edition 15. 11.34 pp. Annear, T. C. 1980. .\ characterization of Yampa and Green River ecosystems. Unpublished tliesis. Utah State Univ., Logan. 143 pp. Bailey, C. and R. Alberti. 1952. Lower Yampa River and tributaries study. Federal ,\id Project F-3-R-1. Trout stream studies. 154-180. Colorado Came and Fish Department, Denver. Bea.mesderfer, R. C. and J. L. Con(;leto.n. 1982. Spawning behavior, habitat selection, and early life history of northern squawfish with inferences to Colorado squawfish. Idaho Cooperative Fish- ery Research Unit. Univ. of Idaho. Moscow. Pages 47-127 in Part 3, Colorado River Fishery Project, Final Report. Contracted Studies. Fish and Wildlife Service, Bureau of Reclamation, Salt Lake City, Utah. Carlson, C. \., C. (i. Prewitt, D. E. Snyder. E. ]. Wick, E. L. Ames, and W. D. Fronk. 1979. Fishes and macroinvertebrates of the Wiiite and Yampa rivers. Biological Series I. Bureau of l^and Management, Denver, Colorado. 275 pp. Colorado River Fisheries Project. 1982. Final report; Part 2, Field investigations. Fish and \\'ildiife Service. Bureau ot Reclamation. Salt Lake Citv. Utah. .3(S5 pp. 1982. Final report: Part 3, Conlractetl studies. Fish and Wildlife Service. Bureau oi Keclania- tion. Salt Lake City, Utah. .324 pp. Crosby, C:. 1975. White River fish collections. Utah Di- vision of Wildlife Resources, Regional Offices. Vernal, Utah. Unpublished mimeograph. 10 pp. CiMMiNS, K. W., AND G. H. Lauff. 1969. The influence of substrate particle size on the microdistributicm of stream macrobenthos. Ih tlrol)i()l()gia. 34:145-181. Hay.nes, C. H., and R. T. Mlth. 1982. Identification ot habitat requirements and limiting factors for Col- orado squawfish and humpback chubs. Job Prog- ress Report. State of Colorado. Project SE-3-4. Endangered Wildlife Investigations. 43 pp. HoLDEN. P. B. 1973. Distrilnition, abundance, and life history of the fishes of the Upper Colorado River Basin. Unpublished dissertation. Utah State Univ., Logan. 59 pp. Hynes, H. B. \. 1970. The ecology of running waters. Liverpool Univ. Press, England. Univ. of Toronto Press, Canada. .55 pp. Lamarra, v., a.nd J. Carter. 1981. Investigation of fish distribution, habitat, and food preference in the White River. IHah and Colorado. Final report to Wliite River Shale Oil Corporation from Ecosys- tem Research Institute. Logan, Utah. Lanican, S. H.. and C. R. Berry, Jr. 1979. Distribution and abundance of endemic fishes in the \\'hite River in Utah. Final Report. U.S. Bureau of Land Management, Utah State Office, Salt Lake Citv, Utah. Contract 14-16-0006-78-0925. Mc Intire, C. D. 1966. Some effects of current velocity on periphvton communities in laboratorv streams. Hvdrobiologia. 27:.559-.570. Mann, K. H., R. H. Britton, A. Kowalczewski, T. J. Lack, C. P. Mathews, and I. McDonald. 1972. Productivity and energy flow of all trophic levels in the River Thames, England. Pages 579-596 in Z. Kajak and A. Hillbricht-Ilkowska, eds.. Pro- ceedings of the IBP-U\ESCO Symposium on Productivity Problems of Freshwaters. Polish Sci- entific Publishers, Warzeawa-Krakow. 918 pp. Prewitt. C. C, B. A. Caldwell, and W. Miller. 1982. C'ompletion report. Y'ampa-White Physical Habi- tat Study. Colorado River Fisheries Project, Fort (Collins, Colorado. Pages 59-73 in Final report: White River. Colorado River Fishery Project. Fish and Wildlife .Service. Salt Lake City. Utah. Seethaler, K. H., C W. .Mc.\da, and R. S. Wydoski. 1979. Endangered and threatened fish in the Yampa and Green rivers of Dinosaur .National Monument. Pages 605-612 in R. M. Linn, ed.. Proceedings of the First Conference on Scientific Research in National Parks. U.S. Dept. Inter., National Park Service., Trans. Proc. Ser. 5. Si'RLLEs. W. M. 1947. .\n ecological investigation of stream insects in .\lgon(juin Park, Ontario. I'niv. Toronto Stud. Biol. Serv. 56:1-81. Sthaiss, H. E. 1979. Reliability estimates Un Ivlevs" electixitN- index, the forage ration, and a pro- poseti linear index of food selection. Trans. .\mer. Fisheries Soc. 108:344-352. Tvis, H. M., E. J. \\i( K, AND D. C. Skates. 1981. A spawuiuii; migration of (.'oloratlo s(|u.iwfish [rtiicliochiclus Iticius) in tiu' Yampa and (ireen rivers. Colorado ami Utali. 19SI. 13th Ann. Symp., Desert Fishes Council. De.itli \allev Na- tional Monument, ('alifornia. Valdez, R. .\., and G. H. Clemmer. 1981. Life iiistory and prospects for recovery of the humpback and bonytail chub. Proceedings at symposium on F'ishes of the I'pper ("olorado River S\stfm: jiies- ent and hilure. Western Div. AFS. W estlake, D. F. 1966. The light climate for plants in rivers. Pages 99-119 in R. Bainbridgc, G. C. Evans, and O. Rackman, eds.. Light as an ecolog- ical factor. Brit. Ecol. .Soc. Svmp. 6. Blackwell. Oxford. 4r)2 pp. DIFFERENTIAL EFFECTS OF CATTLE AND SHEEP GRAZING ON HIGH MOUNTAIN MEADOWS IN THE STRAWBERRY VALLEY OF CENTRAL UTAH J. B. Sliupc' and Jack D. Bmtliersoir Abstr.\ct.— Species diversity, niche metrics, cover, frequency, and soil relationships were studied on high moun- tain meadows on adjacent cattle and sheep allotments in Strawberry Valley, Wasatch County, Utah. The cattle allot- ment vegetation was predominantlv Mountain bluebell (Mertcnsia cilkiki), and the sheep allotment vegetation was predoiuinantK Sniallwiug scilge [Cdicx uiicmpteni). Other species of importance on both areas included Letterman needlcgrass [Stijxi IcttcntKinii). Mountain iiromegrass (Bronuis carinatus), and Yarrow {Achillea millefoliuin). Tall forbs were most abundant on the cattle allotment, and low forbs, perennial grasses, and sedges were most abundant on the sheep allotment. Vegetation composition on the two allotments was significantly different. High mountain meadows in Utah have been important to ranchers as forage re- sources since the pioneers entered the state in 1847. The effects of grazing in Utah have been recorded since 1865 and have shown that vegetation composition and productivity are altered by hvestock (Roberts 1930). Re- cent studies conducted near Elk City, Idaho, show that cattle grazing reduces herbage production, changes species composition, and increases bare ground in dry mountain mead- ows (Leege et al. 1981). Livestock graze se- lectively, with sheep being best adapted to browse ranges and cattle showing preference for grass ranges (Stoddart et al. 1975). De- pending on levels of grazing, common use by more tlian one class of livestock can improve the range or accentuate the detrimental ef- fects caused by individual species. Merrill et al. (1968) stated that common use equalized the grazing pressure, and each species ben- efited from the grazing of the other. The purpose of this study was to determine the differential effects of 30 years of contin- uous cattle and sheep grazing upon the vege- tative composition of dry mountain mead- ows. Data from this study are useful for the efficient management of such meadows. Study Site The study sites are in the Strawberry Val- ley west of Strawberry Reservoir approx- imately 30 km southeast of Heber City, Wasatch County, Utah (Township 3S, Range 12W, Section 8-WV4, Section 17-W1/2, Section 2O-W1/2, 81/2, Uintah meridian) (Fig. 1). Eleva- tion of the sites varied from 2789 to 2819 m. The study sites lie at the head of the Mud Creek and Clyde Creek drainages on the lee- UTAH Fig. 1. Map of the studv site location in the Straw- berrvVailev of central Utali. 'Soil Co: ation SerNice, 621 Ins Dnve, Sterling. Colorado HO' ot Botany and Range Sciencf. Brii;ham Yonng t'nn Provo. Utah K4602. 141 142 Great Basin Naturalist Vol. 45, No. 1 20 30 Percent Similarity 40 50 60 70 80 90 100 8-Sheep 3-Cattle 1 -Sheep 3-Sheep 5-Sheep 9-Sheep 4-Cattle 7 -Cattle 2-Cattle 5-Cattle 1 -Cattle 1 0-Cattle 4-Sheep 10-Sheep 6-Cattle 2-Sheep 6-Sheep 7-Sheep 8-Cattle 9-Cattle C : cattle allotment sites S : sheep allotment sites Fi(j;. 2. Cluster diaunim ot soil factors tor the e tttle and sheep allotments ward side of the ridge top separating Utah and Wasatch counties. The dry meadows chosen for study oc- curred in adjacent cattle and sheep allot- ments located within the aspen-conifer forest. Both allotments have been free of common use grazing since 1952 and are now utilized as summer range by livestock and wildlife. Livestock AUM levels are equivalent on the study sites of both allotments. The area has cool summers and cold winters with heavy snows. Annual precipitation averages 610 mm (24 in), of which 60% falls as winter snow (England 1979). The soils are derived from weathered sandstone and silt-shale. Wind was not measured at the sites but blew continually from the west during the hours we were on the study site. Methods Vegetation was sampled by using (0.25m-') quadrats placed every 3 m along 30 m tran- sects. Twenty transects were randomly se- lected across the slope, with 10 in the cattle allotment and 10 in the sheep allotment. Monocot plants were identified following Cronquist et al. (1977) and dicots from Welsh and Moore (1973). Cover and frequency of all plant species and plant life form categories were calculated for both cattle and sheep al- lotments. Cover for each species was esti- mated as suggested by Daubenmire (1959). Species diversity (Shannon and Weaver 1949, Patton 1962) and prevalent species (Warner and Harper 1972) were determined for the cattle allotment, sheep allotment, and both allotments combined. Vegetative differences between the cattle and sheep allotments were determined by use of cluster analysis (Sneath and Sokal 1973), Spearmans" Rank Correlation (Snedecor and Cochran 1968), and Students t-tests. Niche overlap (based on geographical distribution of individuals) was calculated for each possible pair of species within and between the allotments (Colwell and Futuyma 1971). Soil samples from the first 20 cm of the soil profile (Ludwig 1969) were taken from each transect and placed in Zip-lock plastic bags January 1985 Shupe, Brotherson: Effects of Grazing 143 to insure water retention. Soil samples were weighed wet, dried in an oven for 20 days at 50 C, and then reweighed. Percentage mois- ture was determined from the differences in wet and dry weights. Average soil depth was measured with aim penetrometer at the 0-, 15-, and 30-m marks along each transect (Greenwood and Brotherson 1978). Analyses for calcium, copper, iron, magnesium, man- ganese, nitrogen, phosphorus, potassium, so- dium, zinc, organic matter, pH, and texture were conducted by the Soils Analysis Labora- tory, Department of Agronomy and Horticul- ture, Brigham Young University, Provo, Utah. Means, standard deviations, and Students t- tests were used to determine soil differences between cattle and sheep allotments. The study sites were clustered according to the similarity and distribution of soil and vegeta- tion within the allotment meadows (Sneath and Sokal 1973). Results and Discussion Soils in the meadows were loams and sandy loams with a minimum of 1.5 inches of available water-holding capacity. This repre- sents the moisture between wilting point and field capacity (USDA 1975). Although not significantly different, soil moisture values taken in late July had means of 10.5% for the cattle allotment and 13.5% for the sheep al- lotment. These levels are close to the wilting point for most plants and if drying continues will eventually force the plants to become dormant. Of the soil factors tested, only iron, bare ground, percent sand, and percent silt showed significant differences between cattle and sheep allotments (Table 2). Percentage bare ground and sand were greater and iron concentrations and percentage silt lower in the cattle allotment. The cluster of all 20 transects in the cattle and sheep allotments were highly similar (41% or greater), and no definite patterns emerged relative to soil difference between the two allotments (Fig. 2). The differences in concentrations of iron in the allotment soils do not seem important vegetatively. When large amounts of copper, manganese, and zinc occur in acid soils, iron is usually deficient (USDA 1957, Vose 1982). These chemicals have low concentrations in our soils and therefore should not tie up iron. Textural differences (Table 1) between the two allotments were significant though the differences were small (8% for sand and 5.6% for silt). To understand if these differences in- fluenced vegetation patterns between the two allotments, soil textures for all transects were superimposed on the cluster (Fig. 6) of vegetative data. No distinguishable patterns emerged. Further, when percentages of sand Table 1. Soil characteristics of cattle and sheep allotment meadows with their means, standard deviations, and coefficients of variation. Significance levels were computed using the Students t-test. Cattle Sheep Factor X Sd CV X Sd CV Significance level Bare ground (%) 31.0 12.86 41.4 21.8 7.61 34.9 .05 Soil depth (cm) 47.1 7.0 14.9 45.5 9.2 20.2 N.S. Clay (%) 12.1 3.2 26.1 14.5 2.9 20.0 N.S. Sand (%) 52.6 6.1 11.5 44.6 7.3 16.4 .05 Silt (%) 35.3 4.5 12.6 40.9 5.1 12.4 .05 pH 6.1 0.1 1.6 6.2 O.I 1.6 N.S. Organic matter (%) 2.5 0.8 32.0 3.0 0.7 23.3 N.S. Soil moisture (%) 10.5 2.9 27.6 13.4 5.1 38.1 N.S. Calcium (ppm) 881.3 372.0 42.2 1162.7 388.8 33.4 N.S. Magnesium (ppm) 80.0 22.5 28.1 92.4 33.8 36.6 N.S. Sodium (ppm) 18.8 11.6 61.7 18.4 7.9 42.9 N.S. Potassium (ppm) 183.9 118.9 64.7 186.1 48.6 26.1 N.S. Iron (ppm) 66.4 19.8 29.8 93.8 24.8 26.4 .05 Copper (ppm) 0.9 0.2 22.2 1.0 0.2 20.0 N.S. Manganese (ppm) 44.7 11.7 26.2 56.7 19.3 34.0 N.S. Zinc (ppm) 3.8 1.6 42.1 5.2 1.4 26.9 N.S. Nitrogen (%) 0.1 0.03 30.0 0.1 0.04 40.0 N.S. Phosphorus (ppm) 33.6 9.8 29.2 38.2 9.2 24.1 N.S. 144 Great Basin Naturalist Vol. 45, No. 1 and silt from the 20 sites were correlated with the prevalent species, no significant relationships developed. Bare ground affects the amount of mois- ture that soaks into the ground or nms off. With greater amounts of bare ground, more moisture mns off and can therefore increase erosion on the sites. However, no differences in erosion patterns on the two allotments were observed. The bare ground on the sites has apparently been accentuated by the dif- ferential grazing patterns that have occurred over the past 30 years. Cattle tend to select those plants with greater basal areas, thus leaving greater amounts of exposed ground when they are removed from a site. Twenty-nine of 38 plants species identified within the study area were located on the cattle allotment and 37 on the sheep allot- ment (Table 2). Percentage cover and fre- quency for all species are shown in Table 2 and prevalent species in Table 3. The cattle allotment had 13 prevalent species and the sheep allotment 16. The three most prevalent Table 2. Plant species alon12") and low forbs (<12"). Significant differences (P<.05) ap- peared relative to low forbs, tall forbs, and perennial sedge life forms. The sheep allot- ment contained greater amounts of low forbs and perennial sedges, and the cattle allot- Table 4. Life form types with their average cover and frequency values in the cattle and sheep allotments. Signif- icant differences are based on cover values and Students t-tests. Cfc ttle allotment Sheep allotment Significance Plant life form Cover fre (juency Cover frequency level Low forbs ° L5.0 14 27.4 19 .05 Tall forbs" 37.8 17 17.8 13 .05 Perennial grass 16.6 31 19.2 21 N.S. Perennial sedge 3.5 16 23.0 76 .05 Shrub 0.3 2 0.2 2 N.S. • than 12 inches tall, and short forbs are 12 inches or less. 148 Great Basin Naturalist Vol. 45, No. 1 Percent of Niche Overlap 40 50 60 r<^ rC C ^ ^ C 5- Allium acuminatum 24- Madia glomerata 34- Stellaria lamesiana 38- Viola nuttallii 37- Vicia americana 18- Gllla aggregata 28- Polemonium foliosissimum . 1 9- Hackelia flonbun'da . 2- /^goseris glauca . 36- Symphoncarpos orbiculatus . 4- Agrostis variabilis .21- Hieracium scouleri . 1 - Achillea milli folium . 1 0- Carex Microptera . 1 2- Collomia linearis -35- S(/pa lettermanii . 1 3- Deschampsia caespitosa . 29- Polentilla gracilis .31- Ranunculus alismaefolius .20- Helenium hoopseii .25- Melica bulbosa . 1 4- Erigeron speciosus .22- Ligusticum tilicinimum . 3- Agropyron trachycaulum . 9- Bromus carinatus .26- Mertensia ciliata .27- Orthocarpus tolmei . 7- Arabis drummondii .33- S((pa Columbiana . 8- Artemesia ludoviciana .32- Rudbeckia occidentalis .11- Castilleja sulphurea . 6- Artemisia dracunculus .1 5- Festuca pratensis .17- Geranium viscosissimum .23- Lupinus argenteus .30- Poa pratense .16- Gayophytum ramosissimum ment had greater amounts of tall forbs. Pe- rennial grasses were more prominent on the sheep allotment, but differences were not significant. With the soil factors showing little or no differences between allotments, the differ- ences in vegetative composition in the allot- ment meadows appear to have been caused by grazing pressures. Perennial sedges are preferred by cattle and tlius appeared more abundantly on the sheep allotment. Con- versely, sheep prefer forbs that were more abundant on the cattle allotment. As grazing pressures increase, preferred vegetative life forms decrease and may disappear from the allotments used by a specific type of livestock. To more efficiently manage these mead- ows, plant species and life forms should be grazed evenly. This would reduce the effects of preferential grazing (which tends to favor one species over another), which changes competitive relationships among species. This can be achieved by grazing at the prop- er time of year and with the proper mix of livestock species and munbers. An appropri- ate mix or rotation of livestock onto the allot- ment meadows would allow for a more uni- form use of life forms and species to occur without losing any preferred species from the grazing area. Literature Cited CoLWKLi.. W. K.. \\i> 1) |. Im n ^\l\. 1471, On the nu-a- surcment of nulic ImMcith .uk! ox. i lap. IaoIouv 52:567-57(i. Cnt).\yilST, A., A. 11. lk)l.M(.HI N. \. 11. llolMC.Hl \, |. 1.. Revk.m., .\.m) p. K. Ik)i.M(;HK.\. 1977. Intri- January 1985 Shupe, Brotherson: Effects of Grazing 149 10 20 30 Percent Similarity 40 50 60 70 80 90 100 c! 8-Cattle 10-Cattle 5-Cattle 6-Cattle 7-Cattle 9-Cattle , 2-Cattle , 3-Cattle . 8-Sheep . 9-Sheep . 4-Sheep . 5-Sheep . 1 -Cattle . 4-Cattle .10-Sheep . 3-Sheep . 6-Sheep . 2-Sheep . 7-Sheep . 1 -Sheep Fiif. 6. Cluster diatirani of the percentaue similarity ot vegetation on study sites from both allotments mountain Flora. Volume 6, Vascular plants of the Intermountain West— USA. 504 pp. D.AUBENMiRE, R. 1959. A canopy coverage method of vegetational analysis. Northwest Sci. 33:43-46. E.NCi.AND, J. L. 1979. Measured and inferred moisture gradient relationships: a study of montane steppe in central Utah. Unpublished thesis. Brigham Yoimg Univ. Provo, Utah. 36 pp. CIree.nwooi), L. R., .^no J. D. Brotherso.x. 1978. Eco- logical relationships between pinyon juniper and true moimtain mahogany stands in the Uintah Basin, Utah. J. Range Manage. 31:164-167. Kjuecer, W. C, .\nd a. H. WiNWARD. 1974. Influence of cattle and big game grazing on understudy structure of a Douglas fir, Ponderosa pine, Ken- tucky bluegrass community. J. Range Manage. 27:450-453^ Leege, T. a., D. J. HERM.4.N', .AND B. Za.mora. 1981. Ef- fects of cattle grazing on mountain meadows in Idaho. J. Range Manage. 34(4):324-328. LuDwic;, J. A. 1969. Environmental interpretation of foothill communities of Northern Utah. Unpub- lished di.ssertation. Univ. of Utah. 100 pp. Merru.l, L. B., p. O. Reardon, and C. L. Leinweber. 1968. Cattle, sheep and goats . . . mix "em up for higher gains. Texas Agric. Progr. 12:13-14. Patton, B. C. 1962. Species diversity in net phytoplank- ton of Raritan Bay. J. Mar. Res. 20:57-75. Roberts. B. H. 1930. A comprehensive history of The Church of Jesus Christ of Latter-day Saints. Deseret News Press, Salt Lake City, I'tah. Shannon, G. E., and W. Weaver. 1949. The mathe- matical theory of communication. L'niv. of Il- linois Press, Urbana. 117 pp. Sneath, p. H. a., and R. R. Sokal. 1973, Numerical tax- onomy—the principles and practice of nimierical cla.ssification. W. H. Freeman and Co., San Fran- cisco. 537 pp. Snedecor, G. W., and W. G. Cochran. 1968. Statistical methods. Iowa State Univ. Press, .\mes, Iowa. 593 pp. Stoddart, L. a., a. D. Smit[i, and T. W. Box. 1975. Range management. .McGraw-Hill Book Com- panv. 532 pp. USDA 1957. Soil. Pages 111-115 in The yearbook of Agricultine.' l'SD.\ Soil Conservation SERvif:E. 1975. Soil Tax- onomy. Pages 436-754 in Agricultural handbook. VosE, P. B. 1982. Iron nutrition in plants: a world over- view. J. Plant Nutr. 5(4-7):2.3.3-249. Warner, J. H., and K. T. Harper. 1972. Understory characteristics related to site-qualitv for aspen in Utah. Brigham Young Univ. Sci. Bull., Biol. Ser. 16(21:1-20. Welsh, S. L., and G. Moore. 1973. Utah plants— tracheophvta. Brigham Young Univ. Press. Provo, Utah. 474 pp. UNUSUAL SOCIAL FEEDING AND SOARING BY THE COMMON RAVEN {CORVUS CORAXy Clayton M. White- and Merle Tanner-White' .Abstract.— A flock of some 1000 ravens, not associated with night roosts, was seen during the breeding season. Some were feeding on the ground while others soared in the air above them. It is suggested that the large soaring flock acted as information transfers for feeding conditions. The Common Raven (Corvus corax) can be a highly social species. Social aerial soaring, play, and other forms of group interactions are known (Rubey 1933, Lockley 1953, Hew- son 1957, Davis 1967, Dorn 1972). There is lack of agreement over the function of such soaring because of its temporal variation (summer vs. winter), as reviewed by Knight and Call (1980). There is likewise little knowledge as to whether participants of such behavior are breeding or nonbreeding indi- viduals (JoUie 1976). The aforementioned so- cial events frequently involve fewer than 200-300 individuals. However, large num- bers of over 800 birds are known to gather in autumn and winter at overnight communal roosts (Stiehl 1981). Knight and Call (1980) likewise do not mention literatvire reference to large feeding aggregations, which may ac- tually take the form of a social aggregation. Ravens may, however, congregate at refuse dumps, but such groups usually occur in the nonbreeding season or in developing areas with exploitable refuse dumps. Unlike the huge winter flocks of the American Crow {Corvus brachyrhynchos) that gather at dumps, our experience has been that fewer than 100 or so ravens gather at such feeding sites. The following observations involving more than 1000 ravens, but not at communal roosts, may shed some light on these topics. On 11 April 1982 at midday, near the junc- tion of US 191 and U-211, about 26 km north of Monticello, Utah, two different groups of ravens were seen about 1 km apart. The first flock contained about 400 birds, (part count, part estimate) feeding on the ground over about a 3-ha area, and more than 380 were counted soaring in a boil above those on the ground. This soaring flock contained individ- uals from about 50 m to more than 300 m in the air. A second flock of more than 250 was also seen soaring within 1 km northeastward of the former flock. Since none of the birds from the first flock were seen soaring or fly- ing in the direction of the second flock, we are confident that the birds were not double counted. Those feeding on the ground were in a tan- sy mustard {Descurainia)-native grass-mixed small sagebrush (Arteinisia) habitat. From what we could determine, they were feeding on an abundant locust hatch. The habitat of the immediate surrounding area consisted of barren rock structures characteristic of the Utah canyonland country and juniper {Juni- periis) stands. Birds from the first flock that were seen to depart from the main group flew at about 300 m plus in the air in a south- westward direction. We had just seen pairs of ravens still on breeding territories, some with young in the nest, or family groups of recent- ly fledged young (perhaps some year-old groups) about 14-25 km to the west closer to the Canyonlands National Park. The cumulative data from this observation suggest the following information relative to the initially stated questions. Because of the date, these were not from wintering flocks, unless such flocks remain together long after the start of the breeding season, in which case members of the flock would consist of mainly nonbreeding birds. Because we had earlier seen pairs near nesting territories and 'Data gathered while on Agreement 12354-TSA-l to Bechtel Croup, Inc. •Department of Za)logy, Brigham Young University. Prove, Utah 84602. 'College of Physical and Mathematical Sciences, Brigham Young University, Provo, Utah 150 January 1985 White, White: Common Raven 151 also postnesting groups, it is likely that the flocks we saw were nonbreeding or certainly postbreeding individuals. The presence of the foraging flock below the soaring birds sugges- ted that soaring not only may have a social or play function, but also an information trans- fer fimction. The soaring flock may provide information on food sources as suggested for the idea of "centers of information" in soar- ing vultures and eagles (Ward and Zahavi 1973, Sherrod et al. 1977). If the latter expla- nation is reasonable, then some of the birds may be assumed to be members of pairs with- in commuting distance of the food source. Because of the density of breeding pairs over the distance that such communication could be effective, it seems unlikely that more than 100 of the individuals could have been from breeding territorial birds. Lastly, if these flocks constituted nonbreeding social flocks in the sense of those described by Stiehl (1978), their roosting sites would need to be on cliffs, within the juniper forest, or perhaps on some distant electric power pylon as re- cently observed by K. Steenhof (pers. comm.). Literature Cited response to sonic l)an^. Brit. Davis, P. 1967. Raven's Birds 60:370-371, DoRN, J. L. 1972. The conmion raven in Jackson Hole, Wyoming. Unpublished thesis. Univ. of Wyo- ming, Laramie. Hewson, R. 1957. Social flying in ravens. Brit. Birds 50:432-434. JoLLiE, M. 1976. Species interrelationships of three cor- vids. Biologist 58:89-111. K.NicHT, R. L., AND M. W. Call. 1980. The common raven. U.S. Bur. Land Manage., Tech. Note 344. LocKLEY, R. M. 1953. Aerial assembly of ravens in De- cember. Brit. Birds 46:347-348. RuBEY, W. W. 19.33. Flight maneuvers of the raven. Bird Lore .35:14.3-145. Sherrod, S. K., C. M. White, and F. S. L. Williamson. 1977. Biology of the bald eagle on Amchitka Is- land, Alaska. Living Bird 15:14.3-182. Stiehl, R. B. 1978. Aspects of the ecology of the com- mon raven in Harney Basin, Oregon. Unpub- lished dissertation. Portland State Univ., Port- land, Oregon. 1981. Observations of a large roost of common ravens. Condor 83:78. Ward P., and A. Zahavi. 1973. The importance of cer- tain assemblages of birds as "information cen- tres" for food finding. Ibis 115:417-5.34. THREE ADDITIONAL CASES OF PREDATION BY MAGPIES ON SMALL MAMMALS Kerry F. Reese' .-Vbstract.— Three acts of predation by Black-billed Magpies {Pica pica hudsonia) on small mammals indicate that magpies will not onlv attack prev opportunistically encoimtered at close range, as suggested by recent literature, but will also pursue prey detected at long distances. The brief accounts by predation by the two North American magpies, Black-billed {Pica pica hudsonia) and Yellow-billed {Pica nuttalli), appear to occur when the birds op- portunistically discover mammals at close distances while actively foraging (Blackburn 1968, Goulden 1975, Boxall 1982). Goulden (1975) and Boxall (1982) described inter- actions between the birds and their mamma- lian prey (one observed kill each). In this note I describe three observations of Black- billed Magpies attacking and killing small mammals and the defensive movements of the prey. On 25 February 1979 I was observing mag- pies at a feeding station baited with meat and located approximately 80 m from my vehicle. The station was at the edge of a small stand of trees adjacent to a large plowed field west of Richmond, Utah. Over 30 cm of snow cov- ered the field. At 0955 h a vole (Microtus sp.) appeared on the snow approximately 20 m from my vehicle and was highly conspicuous as it moved about. A magpie flew from a tree near the feeding station 80 m away and land- ed 2 m in front of the vole. The vole ran to- ward the field's edge while the magpie fol- lowed by hopping along and flying 1-2 m at a time. Wlien the magpie was within 6-7 m of the vehicle, the bird noticed me and de- parted. As it was leaving, a second magpie flew from the direction of the trees and was about to land behind the vole, but saw me and left. Tlie vole ran back across the snow away from the tnick and, when it was 15-20 m away from me, a magpie, perhaps the first, landed 0.3 m behind it. The vole turned, stood on its hind legs, and lunged at the bird. The vole charged three times in this manner, and on each charge the bird leapt into the air with wings flapping. After the third charge, the magpie jumped to the vole's side and pecked once at its head and neck. The vole was still and the bird struck 3-4 swift, hard pecks at its head. The magpie picked up the vole, flew back toward the trees, and landed on a fence post. The bird proceeded to peck the vole several times and then carried the carcass into the trees out of sight. The entire incident lasted about two minutes. At 1045 h the same day, another vole ap- peared 35-40 m from me in the same field. I had observed the vole on the snow for no more than 15 seconds when a color-banded first-year male magpie, which had been rest- ing in a tree near the feeding station, flew 40 m to the vole and landed directly in front of it. The vole quickly turned and tlie magpie pecked at its head. When the vole was still, the bird carried it into the trees. No more than 30 seconds elapsed from the time the magpie landed luitil it departed with the vole. On 17 March much of the snow had melt- ed, particularly along the edges of the road, where there was a mosaic of snow, tufts of dead, standing grass, and exposed, plowed soil. At 1538 h, a magpie was chased from the feeding station by a more dominant bird | and landed on a fence post along the road 30 I m in front of the vehicle. At 1541 h, the mag- pie left the post, flew 15 m across the road, landed and almost simultaneouslv struck with its beak at a vole exposed on a patch of snow. I had not seen the vole and was unaware of 'Department of Fisheries and \V ildlife, Utah State L'ni' versitv of Idiiho. Moscow. Idaho 8.3H43. l.ojjan. L'tah W.322, Present address: Colleije of Forestry. U ildhte and Kani;e .Sck 152 January 1985 Reese: Magpie Predati 153 how long it had been visible. The vole ap- peared luihurt and ran into a clump of dead grass. Circling the grass twice, the magpie pecked into it several times, but I was unable to see if the vole was struck. The vole sud- denly charged tlie bird, which jimiped into the air as described previously. The vole ran toward an exposed plow furrow 2-3 m away, but was intercepted halfway there by the bird landing in front of it. Pecking the vole's head approximately 15 times, the magpie killed it, picked it up, and dropped it. I left the vehicle to flush the bird away in order to identify the species of vole, but the magpie picked it up again and flew into the trees 70-80 m away. These observations and those in the liter- ature suggest that Black-billed Magpies re- spond opportunistically to the presence of small mammals. My observations indicate that magpies will also attempt to kill small mammals when detecting them from up to 80 m away, not just when encountering prey in close proximity. These birds made kills in the immediate vicinity of an ample food source, the feeding station. Perhaps magpies prefer prey over carrion when possible. All three voles took escape and defensive measures, but to no avail. Voles were highly conspicuous on the snow and, once above the surface, had limited access to refuge from predators. In all three cases the birds (at least two different ones) carried the voles to shel- ter, presumably to eat them as Boxall (1982) reported. None ate the voles in the open as Goulden (1975) described, even though both days were svmny with no wind. Boxall (1982) comments on the scarcity of reports of predation by magpies on small mammals in North America. I believe that such events are rare and primarily fortuitous for the birds. The events that I observed may have been due to the presence of voles in a conspicuous and vulnerable setting, not a normal predator-prey situation. In over 500 h of observing magpies over two winters, these were the only acts of predation I witnessed. These observations were made possible through research supported by the Frank M. Chapman Memorial Fund and the Ecology Center and Department of Fisheries and Wildlife, Utah State University. Literature Cited Blac KBURN, C. F. 1968. Yellow-billed Magpie drowns its prey. Condor 70:281. Boxall, P. C. 1982. Further observations of predation by Black-billed Magpies on small mannnals. J. Field Ornithol. 53:172-173. Goulden, L. L. 1975. Magpie kills a ground squirrel. Auk 92:606. NOTICE TO CONTRIBUTORS Manuscripts intended for publication in the Great Basin Naturalist or Great Basin Natural- ist Memoirs must meet the criteria outhned in paragraph one on the inside front cover. They should be directed to Brigham Young University, Stephen L. Wood, Editor, Great Basin Natu- rahst, 290 Life Science Museum, Provo, Utah 84602. Three copies of the manuscript are re- quired. They should be typewritten, double spaced throughout on one side of the paper, with margins of at least one inch on all sides. Use a recent issue of either journal as a format, and the Council of Biologij Editors Style Manual, Fourth Edition (AIBS 1978) in preparing the manuscript. An abstract, about 3 percent as long as the text, but not exceeding 200 words, written in accordance with Biological Abstracts guidelines, should precede the introductory paragraph of each article. All manuscripts receive a critical peer review by specialists in the subject area of the manu- script under consideration. Authors may recommend one or two reviewers for their article. Accepted manuscripts less than 100 printed pages in length will automatically be assigned to the Great Basin Naturalist. Those larger than 100 printed pages will be considered for the Memoirs series. Illustrations and Tables. All illustrations and tables should be made with a view toward having them appear within the limits of the printed page. The original illustrations or glossy photoprints of them should accompany the manuscript. Illustrations should be prepared for reduction by the printer to either single-column (2% inches) or double-column (5V2 inches) width, with the length not exceeding 7V2 inches. Costs Borne by Contributor. Contributors to the Great Basin Naturalist should be prepared to donate from $10 to $40 per printed page toward publication of their article (in addition to reprint costs). Excessive or complex tables requiring typesetting may be charged to the author at cost. Authors publishing in the Great Basin Naturalist Memoirs may be expected to contrib- ute $40 per printed page in addition to the cost of the printed copies they purchase. No re- prints are furnished free of charge. Reprint Schedule for the Great Basin Naturalist 100 copies, minimum cost for 2 pages, $26. Each additional 2 pages, $6. Each additional 100 copies, $4 for each 2 pages. Examples: 300 copies of 10 pages = $82; 200 copies of 13 pages = $86. Great Basin Naturalist Memoirs No. 1 The birds of Utah. By C. L. Hayward, C. Cottam, A. M. Woodbury, H. H. Frost. $10. No. 2 Intermountain biogeography: a symposium. By K. T. Harper, J. L. Reveal et al. $15. No. 3 The endangered species: a symposium. $6. No. 4 Soil-plant-animal relationships bearing on revegetation and land reclamation in Nevada deserts. $6. No. 5 Utah Lake monograph. $8. No. 6 The bark and ambrosia beetles of North and Central America (Coleoptera: Scolytidae), a taxonomic monograph. $60. No. 7 Biology of desert rodents. $8. TABLE OF CONTENTS Spatial patterns of plant comnuinities and differential weathering in Navajo National Monument, Arizona. Jack D. Brotherson, William E. Evenson, Samuel R. Rushforth, John Fairchild, and Jeffrey R. Johansen 1 Cryptogamif soil crusts: seasonal variation in algal populations in the Tintic Mountains, Juab County, Utah. Jeffrey R. Johansen and Samuel R. Rushforth . 14 Aquatic parameters and life history observations of the Great Basin spadefoot toad in Utah. Peter Hovingh, Bob Benton, and Dave Bornholdt 22 New species of Astragalus (Legiuninosae) from Mesa Countv, Colorado. Stanley L. Welsh ■ ' 31 A fourth species of Oreoxis (Umbelliferae). Stanley L. Welsh and Sherel Goodrich 34 Insect communities and faunas of a Rocky Mountain subalpine sere. David J. Schimpf and James A. MacMahon .37 Nutrients in Citrex exserta sod and gravel in Sequoia National Park, California. Raymond D. Ratliff 61 Mites (excluding chiggers) of mammals of Oregon. John O. Whitaker, Jr., and CJhris Maser 67 Food of cougars in the Cascade Range of Oregon. Dale E. Toweill and Chris Maser .. 77 Factors influencing nesting success of burrowing owls in southeastern Idaho. Richard S. Gleason and Donald R. Johnson 81 Note on the diet of long-billed Curlew chicks in western Idaho. Roland L. Redmond and Donald A. Jenni 85 Tundra vegetation of three cirque basins in the northern San Juan Mountains, Colorado. Mary Lou Rottman and Emily L. Hartman 87 Use of biomass predicted by regression from cover estimates to compare vegetational similarity of sagebrush-grass sites. L. David Humphrey 94 A new combination and a new variety in Artemisia tridentata. Sherel Goodrich, E. Durant McArthur, and Alma H. Winward 99 Understory response to tree harvesting of singleleaf pinyon and Utah juniper. Richard L. Everett and Steven H. Sharrow 10,5 Aquatic birds of the White River, Uintah Countv, Utah. Benjamin B. Steele and Stephen B. Vander Wall '. 113 Patterns of macroinvertebrate colonization in an intermittent Rocky Mountain stream in Utah. J Vaun McArthur and James R. Barnes 117 (checklist of the mosses of Grand Teton National Park and Teton County, Wyoming. John R. Spence 124 Pkological investigation of a suspected spawning site of Colorado squawfish on the Yampa River, Utah. Vincent A. Lamarra, Marianne C. Lamarra, and John G. Carter 127 Differential effects of cattle and sheep grazing on high mountain meadows in the Strawberry Valley of central Utah. J. B. Shupe and Jack D. Brotherson 141 Unusual social feeding and soaring by the Common Raven (Co/ii/.v corax). Cla\ ton M. White and Merle Tanner-White 1.50 Three additional cases of predation by magpies on small mammals. Kerry P. Reese ... 152 HE GREAT BASIN NATURALISl iume 45 No. 2 30 April 1985 Brigham Young Universit GREAT BASIN NATURALIST Editor. Stephen L. Wood, Department of Zoology, 290 Life Science Museum, Brigham Young University, Provo, Utah 84602. Editorial Board. Kimball T. Harper, Chairman, Botany; James R. Barnes, Zoology; Hal L. Black, Zoology; Stanley L. Welsh, Botany; Clayton M. White, Zoology. All are at Brig- ham Young University, Provo, Utah 84602. Ex Officio Editorial Board Members. Bruce N. Smith, Dean, College of Biological and Agricul- tural Sciences; Norman A. Darais, University Editor, University Publications. Subject Area Associate Editors. Dr. Noel H. Holmgren, New York Botanical Garden, Bronx, New York 10458 (Plant Taxonomy). Dr. James A. MacMahon, Utah State University, Department of Biology, UMC 53, Lo- gan, Utah 84322 (Vertebrate Zoology). Dr. G. Wayne Minshall, Department of Biology, Idaho State University, Pocatello, Idaho 83201 (Aquatic Biology). Dr. Ned K. Johnson, Museum of Vertebrate Zoology and Department of Zoology, Uni- versity of California, Berkeley, California 94720 (Ornithology). Dr. E. Philip Pister, Associate Fishery Biologist, California Department of Fish and Game, 407 West Line Street, Bishop, California 93514 (Fish Biology). Dr. Wayne N. Mathis, Chairman, Department of Entomology, National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560 (Entomology). Dr. Theodore W. Weaver III, Department of Botany, Montana State University, Boze- man, Montana 59715 (Plant Ecology). The Great Basin Naturalist was founded in 1939 and has been published from one to four times a year since then by Brigham Young University. Previously impublished manuscripts in English of less than 100 printed pages in length and pertaining to the biological natural his- tory of western North America are accepted. Western North America is considered to be west of the Mississippi River from Alaska to Panama. The Great Basin Naturalist Memoirs was es- tablished in 1976 for scholarly works in biological natural history longer than can be accom- modated in the parent publication. The Memoirs appears irregularly and bears no geographi- cal restriction in subject matter. Manuscripts are subject to the approval of the editor. Subscriptions. The annual subscription to the Great Basin Naturalist for private individuals is $16.00; for institutions, $24.00 (outside the United States, $18.00 and $26.00); and for stu- dent subscriptions, $10.00. The price of single issues is $6.00 each. All back issues are in print and are available for sale. All matters pertaining to subscriptions, back issues, or other busi- ness should be directed to Brigham Young University, Great Basin Naturalist, 290 Life Sci- ence Museum, Provo, Utah 84602. The Great Basin Naturalist Memoirs may be purchased from the same office at the rate indicated on the inside of the back cover of either journal. Scholarly Exchanges. Libraries or other organizations interested in obtaining either journal through a continuing exchange of scholarly publications should contact the Brigham Young University Exchange Librarian, Harold B. Lee Library, Provo, Utah 84602. Manuscripts. See Notice to Contributors on the inside back cover. 6-85 650 77142 ISSN 017-3614 The Great Basin Naturalist Published at Provo, Utah, by Brigham Young University ISSN 0017-3614 Volume 45 30 April 1985 No. 2 UTAH FLORA: SAXIFRAGACEAE Sherel Goodrich' .\bstract.— a revision of the Saxifragaceae is presented for the state of Utah. Included are 41 taxa in 12 genera. Keys to genera and species are provided, along with detailed descriptions, distributional data, and comments. Litliophragma ojabra Nutt. in T. & G. var. lamuloso (Suksd.) Goodrich is the only new combination proposed. No new taxa are proposed. This is another in a series of works leading to a definitive treatment of the flora of Utah. The Saxifragaceae family is of rather small size in the state. Several taxa are cultivated as ornamentals, and some species of Ribes are grown for their imii. For the most part the taxa seem to be well marked, and separation of taxa is rarely compounded by hybridiza- tion in the native species of this family. The arabic numerals following the discussion of each taxon indicate the number of Utah spec- imens examined by me in the preparation of this treatment. The Roman numerals indicate the number of specimens I have collected from Utah. Acknowledgments Appreciation is expressed to the curators of the following herbaria in Utah: Brigham Young University, Provo; Forest Service Her- barium, Ogden; Garrett Herbarium, Univer- sity of Utah, Salt Lake City; Intermountain Herbarium, Utah State University, Logan. I appreciate the loan of specimens from each of these herbaria. Saxifragaceae Saxifrage Family Perennial herbs or shrubs; leaves basal, al- ternate, or opposite, with or without stipules; flowers perfect, regular, solitary to many and racemose or cymose; sepals, petals, and sta- mens borne on a floral cup or hypanthium; hypanthium saucer or cup shaped or tubular, sometimes small or essentially lacking; sepals 4 or 5, often appearing as lobes of the hypan- thium, petal-like in Ribes; petals 4 or 5, dis- tinct, alternate with the sepals; stamens (4), 5, 8, 10, or more; ovary superior, partly inferi- or, or inferior; fruit a capsule, follicle, or ber- ry. This is an extremely variable family. As treated herein it includes segregates some- times regarded as belonging to the Grossula- riaceae (Ribes) and Parnassiaceae (Parnassia). Our woody species with opposite leaves have been included in Hydrangiaceae and in Philadelphaceae. 'Intermountain Forest and Range Experiment Station, Forest Service, U.S. Department of Agriculture, Ogden, Utah 84401. Present address: Vernal Ranger District, Ashley National Forest, Vernal, Utah 84078. 155 156 Great Basin Naturalist Vol. 45, No. 2 1. Plants shrubs, woody well above ground level 2 — Plants herbaceous, sometimes with woody caudices at or below ground level 7 2(1). Leaves alternate, lobed and toothed; petals shorter than the sepals, not over 4 mm long, often similar in texture and color to the sepals; ovary inferior; fruit a berry Ribes — Leaves opposite, entire or toothed; petals longer than the sepals, over 4 mm long (except in Fendlerella), of contrasting texture and color from the sepals; ovary not completely inferior; fruit a capsule 3 3(2). Stamens numerous (more than 10); petals 4 (or sometimes more in cultivated plants) Philadelphus — Stamens 8 or 10; petals 4 or 5 4 4(3). Leaves toothed, with petioles 2-10 mm long, or rarely sessile; petals 5, 4-13 mm long; stamens 10 5 — Leaves entire, sessile, or the petiole 1-2 mm long; petals 4 and 13-20 mm long, or 5 and only 2-4 mm long; stamens 8 or 10 6 5(4). Flowers rather numerous, paniculate; ovaries and sometimes the petals stellate- pubescent; filaments often petaloid and bifid apically; plants introduced, cultivated Deutzia scahra Thunb. — Flowers few, in small cymes; ovaries sericeous-canescent; petals glabrous; filaments hardly petaloid, sometimes dilated basally, not bifid apically; plants indigenous; leaves sericeous-canescent Jamesia 6(4). Petals 4, 13-20 mm long; stamens 8; styles 4; leaf blades 9-30 mm long, the lower surface without pustulate hairs; shrubs ca 1-2 m tall, of Grand and San Juan counties Fendlera — Petals 5, 2-4 mm long; stamens 10; styles 3; leaf blades 4-12 mm long, the lower surface sometimes with pustulate hairs; shrubs to ca 1 m tall, of other distribution Fendlerella 7(1). Leaves all basal, distinctly and often abruptly petioled, not lobed more than V2 the distance to the midrib; flowers on naked scapes or the scapes with a solitary bract; stamens 5 or 10 8 — Leaves not all basal, or sometimes so in depauperate plants but then the blades not distinctly petioled or else lobed more than V2 the distance to the midrib; flowers often not scapose; stamens 10 12 8(7). Flowers solitary and terminal on long scapes with a solitary bract; petals 6-14 mm long; leaves not toothed, sometimes cordate, otherwise not lobed; fertile stamens 5 Pornassia — Flowers not solitary; scapes without bracts; petals 2-4 nmi long; leaves either toothed or lobed or both and sometimes cordate as well; fertile stamens various 9 9(8). Leaves peltate, the blades 5-40 cm wide, cupped in the center; petioles and scapes to 10 din long or more PeltipInjUum — Leaves not peltate, the blades to 10 cm wide, not cupped in the center; petioles not over 2.5 dm long; scapes less than 7.5 dm long 10 10(9). Flowers in simple, very narrow, elongate, spikelike, ebracteate racemes; petals parted or divided into filiform segments; leaves toothed and lobed; stamens 5; plants from rhizomes, not clothed at the base with persistent leaf bases Mitella April 1985 Goodrich: Utah Flora, Saxifragaceae 157 — Flowers not in spikelike racemes, or if so then bracteate; petals entire; leaves not both toothed and lobed and stamens 5, or if so then plants from woody caudices and clothed at the base with persistent leaf bases 1 1 11(10). Leaves crenate-toothed and lobed; stamens 5; stipules fused to and decurrent on the petioles; plants often of dry rocky places Heuchera — Leaves subentire, crenate, or very coarsely dentate but not lobed; stamens 10; stipules lacking or free of the petioles; plants mostly of dry meadows or wet places Saxifraga 12(7). Leaves parted or divided to the midrib the basal ones abruptly constricted into slender petioles 0.5-10 cm long; petals deeply lobed or cleft, with 3-7 lobes; plants from slender bulblet-bearing rhizomes and fibrous roots .. Lithophragma — Leaves entire, toothed, or lobed, but not divided more than V2 the distance to the midrib or if so then sessile or nearly so; petals entire; plants not or rarely from bulblet-bearing rhizomes 13 13(12). Leaf blades of basal and stem leaves crenate-toothed (the teeth more than 10), also shallowly lobed, reniform or orbicular; bracts of inflorescence smaller than the leaves, but similarly toothed; petals pink to deep red; floral cup 4-7 mm long, often reddish or purplish; plants of the Bear River Range Boykinia — Leaf blades entire, toothed, or lobed (the teeth or lobes 8 or fewer); upper leaves and bracts entire; petals yellow or white, sometimes with purple mark- ings; floral cup 1-3 mm long, greenish or purplish; plants variously distributed Saxifraga Boykinia Nutt. Nom. Cons. Caulescent, glandular herbs from woody branched caudices and thick scaly rootstocks; leaves petioled, basal, and alternate, with membranous stipules; flowers perfect, regu- lar, borne in compact, few-flowered, brac- teate cymes; hypanthium calyxlike; sepals 5; petals 5; stamens 10; ovary ca V2 inferior; styles 2, free or connate below; fruit a cap- sule; seeds several. Boykinia jamesii (Torr.) Engl. James Saxi- frage. [Saxifraga jamesii Torr.; Telesonix jamesii (Torr.) Raf.; Therefon hencheriforme Rydb.; Boykinia heitcherifonnis (Rydb.) Ro- send.] Perennial caulescent herbs, 6.5-20 cm tall; caudices clothed with broad marcescent leaf bases; stems hirsute and stipitate-glandu- lar; petioles 1.5-5.5 cm long, glandular; leaf blades 12-60 mm wide and about as long, reniform or orbicular, truncate or cordate at the base, crenate and more or less shallowly lobed; bracts similar to the upper leaves but smaller and less toothed, and the upper ones usually entire; hypanthium 4-7 mm long, campanulate, glandular to pilose-glandular, often reddish or purplish; sepals 3-4.5 mm long, glandular; petals subequal to the sepals or shorter, pinkish or reddish. Crevices, often in limestone at 2680 to 2990 m in the Bear River Range, Cache Co.; Idaho and Montana to Colorado; and southern Nevada (Spring Mts.); 6 (0). Our plants are referable to var. heucheriformis (Rydb.) Engl. Fendlera Engelm. & Gray Shrubs with opposite, nearly sessile leaves without stipules, deciduous; flowers perfect, rather showy; hypanthium calyxlike; sepals 4; petals 4; stamens 8, the filaments flattened, lobed at the apex; ovary inferior at the base, 4-loculed; styles 4; stigmas minute; fruit a capsule, over V2 superior, septicidal; seeds few in each locule. Fendlera rupicola Gray Cliff Fendlerbush. [F. tomentosa Thornb.]. Much branched shrubs about 1-2 m tall; bark of twigs longi- tudinally ridged and grooved, reddish or strawcolored, turning gray; leaves opposite or appearing fasciculate-opposite, 9-30 mm long, 2-7 mm wide, lance-linear, linear, ellip- tic, or less often ovate, entire, sometimes slightly revolute, sparingly strigose on both sides, the midrib prominent, grooved above, ridged beneath; flowers solitary or 2-3 158 Great Basin Naturalist Vol. 45, No. 2 together at the ends of short branches; hy- panthium 2-3 mm long; sepals 3-5 mm long, to 8 mm in fruit, persistent, strigose beneath, tomentose-villous above; petals 13-20 mm long, constricted to a narrow claw, the blade to 11 mm wide, white; stamenal filaments ca 6-8 mm long, to 2 mm wide at the base, 2- lobed at the apex, the lobes 2-3 mm long, the anther shorter or longer than the lobes; styles 4, appearing as 2 at anthesis, glabrous or with multicellular hairs; capsules 8-15 mm long. Blackbnish and pinyon-juniper commimities at 1372 to 1707 m in Grand and San Juan counties; west central Colorado and south- eastern Utah through Arizona and to Texas; 11 (0). Fendlerello Heller Shmbs with opposite, sessile, or nearly ses- sile leaves, lacking stipules; flowers perfect, in small compound cymes; hypanthium ca- lyxlike; sepals 5; petals 5; stamens 10, the fil- aments dilated below the narrow apex; ovary about V2 inferior, 3-loculed; styles 3; fruit a capsule, septicidal; seeds 1 in each locule. Fendlerella utahensis (Wats.) Heller Utah Fendlerella. [Wliipplea utahensis Wats.]. Sprawling or ascending, much branched shmbs to 1 m tall; bark of twigs strigose, whitish, exfoliating in milky or translucent stripes or flakes; leaves 4-12 mm long, 1-6 mm wide, linear-oblanceolate, linear, elliptic, or less commonly ovate, entire, slightly revo- lute, strigose, the hairs sometimes pustulate, especially on the lower surface; flowers in small compound cymes; hypanthium incon- spicuous at first, finally to 2 mm long in fruit, turbinate-campanulate; sepals 1-1.5 mm long; petals 2-4 mm long, white; staminal fil- aments dilated and petaloid just below the narrow apex, white; styles ca 1.5-2 mm long; capsules 3-4 mm long. Sagebrush, pinyon, juniper, and mountain brush communities, mostly on sandstone and sandy soil at 1480 to 2135 (2745) m in Garfield, Millard, Uintah, Utah, and Washington counties; north- western Colorado to Arizona and west to California; 17 (0). Heiichera L. Perennial scapose herbs from scaly, some- what woody branched caudices or rootstocks; leaves basal, stipulate; flowers paniculate or racemose or nearly spicate, bracteate, per- fect, regular; hypanthium calyxlike; sepals 5; petals 5, small, entire, usually clawed; sta- mens 5; ovary partly inferior, 1-loculed; styles 2; fruit a capsule, opening between the 2 more or less divergent stylar beaks; seeds many. ROSENDAHI., C. O., F. K, Bl TTKRS. AMI (). Lakela. 1936. A nion()tantlia joncsiana (2) Cnjptantha ochrolcuca (1) Ciiscuta warncri (2°) Cijrladrnia huniilis var. joucsii (1) C\jinoptvrus hcckii (2) C'tjnioplcrus higjiinsii (1) Cipnoptcrus niininuis (2) Dalea flavesccns var. p/j/crj (2) Draha uiagtiirci \ar. /j»r^-ei (2) Echinocereus cngchnainiii var. purpuniis (L) Echinocereus triglnchidialus var. inennis (L) Epilohium nevadense (2) Erigcron cromptistii (1) Erigcron kachinensis (2) Erigeron maguirei var. Diaguirci (1) Erigcron maguirei var. h(uris(>uii (2) Erigeron mancus (2) Erigeron prosehiticus ( 1 ) Erigeron sionis (2) Erigeron iinterni(mnii (2) Eriogonuni anunophihini (1) Eriogonum aretioides (1) Eriogonum crompiistii (2) Eriogonum humiiagans (2) Washington Grand. San Juan Kane Kane, Washington Garfield, Wayne San Juan Sanpete (?), Utah Uintah Uintah Wayne Grand, San Juan Iron Sanpete, Sevier Garfield, Iron, Kane, Piute Grand Kane, Washington Enierv, Sevier Millard Garfield, Wayne Garfield, Iron Washington Garfield, Kane Uintah Beaver, Millard. Tooele Carbon, Enierv Grand Enierv Garfield Millard Emery, Garfield. CJrand Wayne Kane Garfield, Iron, Kane Garfield. San Juan Box Elder. Welier Washinuton Millard. W ashiiiat Cache San Juan Emery Wayne Grand, San juan Iron, Kane Kani-. Wasliiniiton Ducluvsuc Millard Ciarfield (Jarfield San Juan Eriogonum lancifolium (2) Eriogonum loganum (2) Eriogonum natum (1) Eriogonum smithii (1) Eriogonum soredium (2) Festuca dasijclada (2) Gaillardia flava (2) Gf/ff/ caespitosa (2) Glaucocarpon suffrutcscens (1) Hedysarum occidcntatc \ar. canodc (1) Heterotlieca joncsii (2) Htpncnoxtjs dcprcssa (2) Ihpncnoxiis hclcnioidcs (2) Lcpidiiiin Ixirnchiiiuuini (I) Lcpidiuiu iiiDnlanunt vnr. neeseae (2) Lcpidium niontinunn vav. stellac (2) Lepidium ostlcri {!) LesqucrcUa tumulosa (1) Lomatium latilobum (2) Mentzclia argillosa (2) Wijas caespitosus (2°) Oenotltera acutissinui (2) Pediocactus despainii (2) Pediocactus uinkleri (2) Penstcmon atuoodii (2) Penstemon bracteatus (1) Penstcmon conipactus (2) Penstcmon concinnus (2) Penstcmon goodrichii (2) Penstcmon grahamii (1) Penstemon Icptanthus (2) Penstemon nanus (2) Penstcmon parvus (2) Penstemon scariosus var. albifluLis (1) Penstcmon tidestromii (2) rcnslcmon udrdii (2) Vhacclia argillacca (L) I'hacclia iiulccora (2) Priuuda maguirei (1) Psor(dca cpipsihi (2) Psoralen pariensis (1) Psorothamnus pohjadenius var. joncsii (2) Ranunculus acriformis \ar. (icsliialis (2) Sclcrocactus glaucus (L) Sclcrocactus puhispiniis (2) Schrocailiis urightiae (L) Sclaginclla utahensis (2) Carbon, Emery Cache, Morgan Millard Emery Beaver Emery, Sanpete, Wasatch Emery, Grand Wayne Uintah Carbon, Emery Garfield, Kane, Washington Duchesne, Emery, Sevier Emery, Garfield, Sanpete, Sevier Duchesne Garfield Kane Beaver Kane Grand, San Juan Sanpete, Sevier Sevier Daggett. I'intaii Emery Wayne Garfield, Kane Garfield Cache Beaver. Iron. .Millard Duchesne. I'intah Uintah Sanpete Beaver, Millard Garfield, Piute Uintah Juab, Sanpete Sanpete, Sevier Utah Enierx, San |uan, Wa>'ne Cache Kane Garfield, Kane Emery Garfield Duchesne. Uintah. San juau Beaver, Juab. Millard, Sevier, Tooele Emery, Kane, Wayne Washington April 1985 Welsh, Chatterley: Utah's Rare Plants 175 Table 1 continued. Name Distribution Senecio dimorphophijUus var. intermedins (2) Silcne petersonii var. minor (1) Silene petersonii var. petersonii (2) Splmeraleea caespitosa (2) Spluieralcea psoraloides (2) Sphaeromeria eapitata (2) Spliaeromeria ruthiae (2) ^uertio gijpsieola (2) Theh/podiopsis argillacea (1) Thehjpodiopsis harnebiji (2) Tounsendia apriea (1) Trifoliiim (nulersonii var. friseanum (1) Xt/lorhiz-d eromiuistii (2) San Juan, Sanpete Garfield, Iron Garfield, Sanpete Beaver, Millard Enierv, Wayne Garfield Washington Millard Uintah Emery Emery, Sevier Beaver, Millard Kane adding significantly to the knowledge of rare plant distribution of the region. Surveys fund- ed by agencies of the federal government, in- cluding studies of M-X related sites in west- ern Utah, oil-shale lands in the Uinta Basin, and inventories in portions of southeastern Utah have added important data. The con- certed effort of active individual botanists during the past few years has also been pro- ductive. Because of these activities, addition- al information has been generated that makes desirable further discussion of Utah's poten- tially threatened or endangered plant species. No previous discussion has included a sum- mary and update of information concerning those species that have been officially listed from the state. This information is included below. Following the discussion of officially listed taxa, rare plant species being consid- ered for listing are discussed. Taxa that have been considered for listing but were down- graded to category 3 species in the Novem- ber 1983 Federal Register are discussed at the end of the article. Utah's Federally Listed Species Eight taxa known from Utah are currently included on the federal list of endangered and threatened species. Five of the plants are species of cactus. The other three include a milkvetch, a poppy, and a phacelia. These taxa are some of the state's most rare and en- dangered plants. Under the provisions of the Endangered Species Act, protection is provided to any species threatened by (1) the destruction or modification of its habitat, (2) overutilization for commercial or scientific purposes, (3) dis- ease or predation, (4) the inadequacy of exist- ing regulatory mechanisms, and (5) other man-made factors. Two listing classifications are provided. The term endangered is given to a species in danger of extinction through all or a significant portion of its range. The term threatened is applied to species that are likely to become endangered in the fore- seeable future. To ensure the proper protection of threat- ened and endangered species, the Endan- gered Species Act regulates interstate and foreign commerce of protected plants and, perhaps most importantly, provided financial assistance for scientific studies, managerial activities, and land acquisition to ensure the continued existence of rare taxa. A brief discussion of each officially listed species is given below, including status (threatened or endangered), a discussion of rarity, known distribution, habitat, and eleva- tion. Collections deposited at BRY are listed, including the date of collection, county, township and range, collectors, and collec- tion number. Distribution maps are also in- cluded. Figure 1 shows the distribution of of- ficially listed species in the state of Utah by township and range. Arctomecon humilis Gov. Map 1 Family: Papaveraceae. Federal status: endangered. The dwarf bear poppy is known only from Washington County, Utah. It is perhaps the most endangered of the plants in Utah. Popu- lations occur near St. George in areas where the soil has a high gypsum content. Their proximity to a rapidly expanding city and their existence on lands frequented by off- road vehicles places the continued existence of the species in danger. Recently the U.S. Fish and Wildlife Service has developed a re- covery plan for this taxon, and the state of Utah has initiated steps to protect the spe- cies. Vegetative types associated with the species are salt desert or warm desert shrub communities. Substrate is clay gypsiferous soils of the Moenkopi Formation. Elevation ranges from 2500 to 2880 feet. 176 Great Basin Naturalist Vol. 45, No. 2 Washington CouN-n-: T43S. RITW. JW Harrison 32- 124 (1932); T42S, R15W, WP Cottani 7238 (1937); T43S, R15W, A Terril (1961); T43S. R16\\', \D Atwood 1704 (1969); T43S, R15W, SL Welsh & ND Atwood 9695 (1970); T43S, R15W, LC Higgins 4210 (1971); T43S, R16W. D Atwood 6590 (1976); T43S, R16W, D Atwood (1976); T43S, R16W, RK Gierisch 4266 (1978); T43S, R16W, SL Welsh 20388 (1981); T43S. R16W, LC Hig- gins & BT Welsh 13415 (1983); T43S, R16W, LC Hig- gins & BT Welsh 13423 (1983); T43S. R15W, E Neese 12896(1983). Astragalus perianus Barneby Map 2 Family; Faliateae. Federal status: threatened. The Rydberg milkvetch was known origi- nally from Piute and Garfield counties. In 1981 Bameby annotated several collections previously designated as A. serpens to A. perianus, extending the known range of this rare taxon into Iron and Kane counties. The Piute County collections are all from the Tushar Mountains, near Bullion Creek. All the Garfield County collections are from near Mt. Dutton. To ensure protection of the species, the U.S. Forest Service developed a managmient plan that was signed in 1983. Two monitoring stations were established, one near the type locality and one on Mt. Brigham. A recovery plan is currently in the review stage. This high elevation milkvetch occurs in alpine, aspen, grass-sedge meadow, and mixed conifer woodland community types. Substrates include tertiary igneous gravels, volcanic ash, rocky loam, and clay soils. Elevational range extends from 10,000 to 11,000 feet. Garfield CotNiY: T.32S, R.3W R&D Foster 5408 (1977); T.32S, R.3W, JR Murdock s.n. (1975); T.32S, R3W, SL Welsh & JR Murdock 12849, 12849b (1975); T32S, R3\\'. D Atwood 8195 (1981); T32S, R3W, SL Welsh & K Tavlor 14454 (1976); T.32S. R.3W, R&D Foster (1977). Iron'Cointy: T.34S, R7W, R&D Foster 4230 (1977); T.35S. R7W. R Foster 5586 (1977); T,34S. R7W. R&D Foster 4737 (1977). Kank County: T37S, R2W, R&D Foster 4440 (1977). Piute County; T28S. R5W, SL Welsh & J Henioid 18142 (1978); T28S. ]{4\\ . D Atwood 7.397 (1979); T27S. ]{2W. L C Higgins 11.55A (1967i; T28S, R5W. D Atwood 8045 (1981). Echinocereiis engelmannii (Parry) Rumpler var. purjmreus L. Benson Map 3 FiuniK: ( iactaceae. Federiii status: end; lie red. The purple-spined hedgehog cactus is known from St. George, north to Veyo and east to Leeds and Hurricane in Washington County. Habitat for this variety of the hedge- hog cactus includes the blackbrush-ephedra and Mohave Desert vegetation types, where it grows in sandy pockets of the Navajo Sand- stone Formation and other substrates. Eleva- tion ranges from 3100 to 3700 feet. Washin(;t(jn County: T42S. R15W, L Benson 1.3637, POM (1949); T42S, R16W, D Ross 6 (1982); T42S, R15W, D Ross 4 (1982); T41S, R1.3W, D Ross 2 (1982); T41S, R13W, D Ross 3 (1982); T40S, R16W, D Ross 1 (1982); T42S, R15W, J Anderson (1982); T42S, R15W, J Anderson (1982); T42S. R15W, D Atwood 5096 (1973); T42S, R16W. SL Welsh 20.389 (1981); T41S, R13W, D Ross 5 (1982). Echinocereus triglochidiatus Engelm. var. inerynis (K. Schum) Rowl. r I Family: Cactaceae. Federal status: endangered. The spineless hedgehog cactus occurs in San Juan County, Utah, but is also known from Colorado. Two collections of the spe- cies are deposited at BRY and come from the La Sal Mountains. The cactus grows in pin- jj yon-juniper-galleta grass or pinyon-juni- I per-Yucca baccata-hVdckssLge community types. This plant prefers sites with very shal- low, rocky soil, usually less than six inches deep. It is commonly found along edges of sandstone outcrops or exposed sandstone | slabs belonging to the Brushy Basin and Salt I Wash members of the Morrison Formation. A management plan has been developed for this species, and a monitoring station was es- tablished in the La Sals in 1980. Elevation ranges from 5000 to 8000 feet. San Juan CIounty: T27S. U24F. SL W clsh. D Atwood 99.33 (1970); T27S, R23E. H.M Thompson s.n. (1979). Pecliocactus silcri (Engelm.) L. Ben.son Map ,5 Faniiiv: Cactateae. Federal status: endangered. The Siler pincu.shion cactus is known from near Fredonia in Mohave and Coconino counties, Arizona, and from southeast of St. George in Washington Comity, Utah. Habitat for the species includes an Atriplex-ephedra- C^owania vegetative community. It grows in clay soils of the Moenkopi Formation. April 1985 Welsh, Chatterley: Utah's Rare Plants 177 Washington County: T43S, R15W, SL Welsh and AH Barnuin 12712 (1975); T43S, RllW, E Neese 12879 (1983): T43S, R14W, E Neese 12902 (1983). Phacelia argiUacea Atwood Map 6 FaiiiiK: H\di()phyliaeeae. Federal status: endangered. The clay phacelia is known only from Utah County. Originally collected by Marcus E. Jones in 1883 and 1894, this phacelia was not recognized at specific level until 1975. There are no known collections of the plant for the period from 1894 until 1971, when Duane Atwood rediscovered the population in Span- ish Fork Canyon. Two populations of this rare plant exist. The type locality occurs on the east side of Highway 6. Recently an addi- tional and larger population was discovered on the west side of the highway. Habitat for the clay phacelia is an Agropijron community on shale outcrops of the Green River Forma- tion. Elevation is approximately 6600 feet. Utah County; TllS, R8E, ME Jones s.n, (1883); T6S, R12W, ME Jones 5591 (1894); TIOS, R6E, ND Atwood et al. .3091, Holotype (1971); TIOS, R6E, SL Welsh & K Taylor 15277, 15047, 15047A (1977); TIOS, R6E. F Smith, E Neese & J Trent 1722 (1982). Sclerocactiis glaucus (K. Schum) L. Benson Map 7 Family: Cactaceae. Federal status: threatened. The Uinta Basin bookless cactus occurs in Colorado as well as Duchesne, Uintah, and San Juan (?) counties in Utah. Habitat in- cludes salt desert shrub and shrub-grass com- munities on terrace gravels and commonly on clays of the Uinta Formation. Elevation ex- tends from 4700 to 5800 feet. Classification of cacti has long been difficult, and this genus is no exception. The straight-spined Sclero- cactiis glaucus has been known in Utah pri- marily from the Uinta Basin. However, a straight-spined plant was recently collected in San Juan County, east of Hite. Current study of Utah's collections (Welsh, in press) questions the appropriateness of species des- ignation for this straight-spined phase of Sclerocactiis. Possibly straight spines in this genus are not more important taxonomically than spineless phases of other plants scattered through spined taxa elsewhere in the Cac- taceae. It may be that Sclerocactiis glaucus. in Utah, should be placed within a more broadly based concept of another species. Duchesne County: T4S, RIW, UBM, SL Welsh & E Neese 18303 (1979); TllS, R17E, E Neese & S White 8654 (1980); T4S, RIW UBM, S Welsh & E Neese 18306 (1979); T3S, R6W, UBM, SL Welsh 18512A (1979). Uintah County; T7S, R20E, E Neese & BT Welsh 7277 (1979); TllS, R18E, E Neese 4487 (1978); T9S, R19E R Foster 7588 (1979); T8S, R20E, Welsh & Neese 18340 (1979); T9S, R20E, E & J Neese 7584 (1979); T8S, R20E, SL Welsh & E Nee.se 18.324 (1979); TllS, R19E, E Neese & S White 8652 (1980); TllS, R18E E Neese & S White 8655 (1980); T7S, R20E, Neese & B Welsh 7279 (1979); TllS, R17E, E Neese & S White 8653 (1980); T4S, RIW UBM, 7279 (1979); TllS, R17E, E Nee.se & S White 8653 (1980); T4S, RIW UBM, Welsh & Neese 18307, 18308, 18309, 18309A (1979). San Juan County: T35S, R15E, SL, BT, ML Welsh 22187 (198,3). Sclerocactiis wrightiae L. Benson Map 8 Family: Cactaceae. Federal status: endangered. The Wright fishhook cactus occurs in Emery and Wayne counties. The greatest number of collections come from Wayne County, where populations of the plant occur near Factory Bvitte and North Caineville Mesa. Habitat includes salt desert shrub, shrub-grass, and juniper communities on Mancos Shale, Dakota, Morrison, Summer- ville, and Entrada formations. Elevational range extends from 4800 to 6100 feet. Emery County: T24S, R6E, Neese & White 7315 (1979); T24S, R6E, E Neese & K Mutz 11428 (1982); T26S. RUE, Neese & Thorne 7177, 7177A (1979); T23S, R6E, JG Harris 657 (1980). Wayne County: T27S, R9E, KD Heil (1978); T27S, R9E, KD Heil (1978); T29S, RUE. KD Heil (1976); T,30S, R7E, D Atwood & S Good- rich 8632 (1982); T27S, R9E, KD Heil (1978); T27S, R9E, S, E, & M Welsh 16704, 16723 (1978); T27S, R9E, S Welsh, K Tavlor, F Peahodv 13094 (1976). Plants Currently under Review In December 1981 the Utah Native Plant Society appointed a committee to review in- formation accumulated during the past col- lecting season. From that review recommen- dations were made concerning the status of Utah's rare plant species. Taxa were placed in categories based on priorities for listing. Those species considered most threatened were placed in the highest category, and those to which the threat was not as critical were placed in lower categories. In each year since 1981, the Utah Native Plant Society has 178 Great Basin Naturalist Vol. 45, No. 2 Table 2. Utah's rare plants by county. Beaver Cn/f)tanth(i coinpacta Eriogonum sorciUum Lepidium ostlcri Penstenum couriuniis Sck'wcactus pulyispiuus Spluicnilcca nicspitosci Trifolitini (iiKlcrsonii var. frisainiint Table 2 continued. Box Elder Draha i\ui^uirci Cac:he Erigewn civmpiistii EriogoniiDi Joganum Pcnslctuon coiiipdctti'i rrii)iul(i ludiiuiiri Carron Eriogouinii hinciloliuDi Cniptanthd crcutzfcldti HcclysiiniDi occidcntdlc Daccett ( h'nothcrd dcutissinid hiiikci Cymopterus minimus Dalea flavescens var. cpira Eriogonum arctioides Eriogon inn cronqKistii Heterotheca jonesii Hipncnoxi/s Iwlcnioidcs Lepidium montanum var. nceseac Penstcmon dtuoodii Penstcmon l>rdctcdtus Penstcmon jxinus Psorak'O pariensis Ranunculus acriformis var. ac.stiv(dis Sdenc petersonii var. minor Silene petersonii var. petersonii Sphfieromeria capitatu Gra.nd Asclepias cutleri Astragalus ischji Astragalus sahulosus C.njptantha data Ciicladenia Itumilis var. jonesii Erigeron maneus Gaillardia flava LoDidtium Idtilohiim DicnEs.NE Erigeron untermannii Hytnenoxys deprcssa Lepidium harnelnjanuni Penstcmon goodrichii Seldcrocdctus glaucus Emery Astrdg(dus suhcinereus var. hasdtticus Cniptatitha creutzfeldtii Cniptduthd jonesiana Cijeladenid huindis var. jonesii Erigeron maguirei var. maguirei Eriogonum lancifolium Eriogonum smitltii I'estuea dasiiehida (.'.ailhirdid fluid llcdiisariim oceidenldle var. eanone Ihimenoxijs deprcssa Ihpnenoxijs liclcnioidcs Pedioeactits dcspainii Phdcetia indcconi Psorotliinnniis poliiddenius var. jonesi Selerocdclns u right iae Splidcrdlccd psoraldidcs 'rhclijpodiopsis hdrnehiji Tinvnscndia aprica CJARFIELI) Astragalus harncbiji Astragalus pcri: T16S, R6E, ME Lewis 6607 (1980). Hijmenoxys depressa (T. & G.) WeLsh & Reveal Map 66 Family: Asteraceae. Federal designation: threatened, categor\' 2. The type of this species was taken by Fre- mont on his second expedition in the Rocky Mountains, possibly from the Uinta Basin. However, it was not until recently that any collection was reported from there. Most known populations are .scattered throughout Emery County, and Fremont could have col- lected the species when he traversed that area. The literature indicates the plant's dis- tribution extends into Garfield County, but J no specimens at BRY corroborate this. Dur- |j ing the 1982 field sea.son the species was also di.scovered south of Duchesne in Duchesne County. Habitat for the species is the pinyon- juniper or mixed de.sert shrub communities on barren expo.sures and, often, the rimrock April 1985 Welsh, Chatterley: Utah's Rare Plants 193 of cliffs. Elevation extends from 4400 to 8400 feet. DicnESNE County: T6S, R4W, LIRM, Goodricli & At- wood 16806 (1982). Emery County: San Rafael Swell, WD Stanton s.n. (no year); T22S, R13E, BF Harrison 5591 (1931); T19S, RllE, BF Harrison 8128 (1936); T19S. RllE, SL Welsh & ND Atwood 9907 (1970); T19S, R12E, K Despain 240 (1977); T19S, R7E, ME Lewis 4729 (1977); T19S, R7E, S Welsh & S Clark 15417 (1977); T18S, R12E, S Welsh & S Clark 16140, 16165 (1977); T21S, R14E, K Despain 306 (1978); T19S, R12E, K Despain 355 (1978); T19S, R12E, J Harris 126 (1979); T24S, R6E. S White & G Moore 64 (1979); T20S, RIOE, K Despain 530. 542 (1979); T21S, RUE, J Harris 409 (1979); T18S, RllE, J Harris 542 (1979); T22S, R6E, BT Welsh 336 (1980); T22S, R6E, BT Welsh 344, 349 (1980); T22S, R13E, JG Harris 712 (1980); T20S, RIOE, K Des- pain 608 (1980); T24S, R13E, J Harris 786 (1980); T26S, R9E, JC Harris 822 (1980); T26S, R9E, J Harris 842 (1980); T26S, R9E, J Harris 864 (1980); T19S, RllE, JG Harris 891 (1980); T26S, R9E, J Harris 920 (1980); T25S, R9E, J Harris 922 (1980); T20S, RllE, J Harris 961 (1980); T22S, R13E, M Williams & D Atwood 80-123-2 (1980); T26S, R8E, JG Harris 979 (1980); T21S, RUE, J & M Harris 1084 (1981); T26S, R9E, Atwood and Good- rich 8652 (1982); T2.5S, R9E, Atwood and Goodrich 8654 (1982); T22S, R14E, E Neese 13331 (1983). Sevier County: T24S, R5E, San Rafael Swell. J Harris 579 (1979). Hymenoxijs helenioides (Rydb.) Cockerell Map 67 Family: Asteraceae. Federal desi47 (1980). SF.yiKH Cointy: T24S, R5E, Reveal & Welsh 721, Holotvpe (1966); T24S, R5E, SL Welsh et al. 8972 (1969); T24S R5E, E Neese & S White 3017 (1977). Tri folium andersonii Gray ysLr. friscanum Welsh M. Family: Fabaceae. Original citation: Great Basin Naturalist 38: .3.55. 1978. Federal designation; endangered, category 1. This newly described variety was known to occur only on the rocky ridges and limestone gravels of the San Francisco Mountains, near Frisco in Beaver County. However, it was discovered in 1982 in similar habitat areas on the Tunnel Spring Mountains in the Desert Experimental Range in Millard County. It has been regarded as endangered, but that designation should probably be tentative un- til additional information can be gathered. Elevational range of the plant is from 6700 to 8000 feet. Be.^ver Cou.nty: T27S, R1.3W, F Peabody et al. 406, Holotype (1976); T27S, R1.3W, K Ostler & D .\nderson 1206, 1260 (1978); T26S, R13W. K Ostler & D Anderson 1349 (1978); T27S, R13W, SL Welsh (k M Chatterley 19.526 (1980); T27S, R13W, SL Welsh & M Chatterley 19658 (1980). Millard County: T24S, R17\\', S Good- rich 169,53 (1982); T24S, R17W, S Goodrich 16972 (1982). Xijlorhiza cronquistii Welsh & Atwood Map 111 Family: .\steraceae. Original citation: Brittonia 33: 302. 1981. Federal designation: threatened, category 2. Only a single large and uniform population of the Cronquist xylorhiza is known, and this occurs near Horse Mountain in north central Kane County. The plant is confined to the badlands topography of the Kaiparowits For- mation and is associated with a broad-leaved phase of Grai/ia brandcgci. Elevation of the species is approximately 6600 feet. Kane Cov.nty: T38S, R2E, SL & SL Welsh 12819, Holotype (197.5). Pl.\nts no Longer Under Review Plants that have previously been consid- ered in categories 1 or 2 but which were downgraded to category 3 in the November 1983 Federal Register are dicu.ssed below. No distribution maps are included for these spe- cies. All but one of the ta.xa are category 3C plants, meaning that the distribution of the plant is wider than previously supposed or that there are no current threats to the taxa. However, if additional threats to these spe- cies are discovered, the plants could once again become candidates for listing. Angeliea irheeleri Wats. Famih : .\piaceae. Federal designation: categor\' 3C. April 1985 Welsh, Chatterley: Utah's Rare Plants 203 The Wheeler angelica is a rare plant, ap- parently restricted to the northern and cen- tral parts of Utah. It is fairly widespread, one population having been located in each of five separate counties. However, the popu- lations are extremely disjunct, and the reason for this distribution pattern is not known. Though the plant is no longer under review for listing, as additional information about the species is accumulated its status may change. Angelica wheeleri is a streamside or wet meadow species and grows at elevations ranging from lower foothills to approx- imately 10,000 feet. Cache County: T12N, R2E, LC Higgins 10371 (1977). Juab County: T12S, R2E, S Goodrich 15862 (1981). Piute County: T27S, RIW, LC Higgins 4721 (1971). Sevier County: T26S, R2E, LC Higgins (1977). Utah County-: T5S, R.3E, Cottam 3845 (1928). Astragalus chamaemeniscus Barneby Family: Fabaceae. Federal designation: category 3C. The ground-crescent milkvetch was, until recently, considered restricted to east central Nevada, where it grows in Lincoln, White Pine, and Nye counties. However, during the 1981 field season it was collected on the Es- calante Desert in Iron County, Utah. It oc- curs on hillsides and valley floors in deep sands derived from limestone formations, generally in a sagebrush community type. Elevation extends from 4900 to 6500 feet. Only one population is known from Utah, and so the plant is considered rare in this state even though its total distribution does not warrant current consideration for listing. Seven collections from Nevada are deposited at BRY. Iron Counts': T33S, R17W. F Smith & S Jensen, s.n. (1981). Astragalus consobrinus (Barneby) Welsh Family: Fabaceae. Federal designation: category 3C. The Bicknell milkvetch is based on A. cas- taneifowiis var. consobrinus. Collections of the species at BRY are from three counties. However, Barneby 's discussion of the basio- nym includes two additional counties as part of the distribution. This milkvetch grows in alluvial soils of varied compositions, usually on open gravelly knolls and hillsides in sage- brush-grasslands and pinyon-juniper commu- nities. Elevational range extends from 6000 to 8000 feet. Emery County: T23S, R6E, Neese & White 7355 (1979); T20S, R6E, R Foster 8262 (1979); T19S, RUE, D Atwood and B Thompson (1983). Sevier County: T26S, R4E, D Atwood and B Thompson 7597 (1980). Wayne County: T28S, R4E, Riplev & Barneby 8605 NY (1947); T29S, R5E, NH Holmgren et al. 2092 ('l965); T28S, R3E, SL Welsh et al. 1.3088 (1976); T28S. R3E, SL Welsh, et al. 1.3090 (1976); T.30S, R6E, SL Welsh 1.3.3.56 (1976); T28S, R4E, SL Welsh & G Moore 13837 (1976); T27S, R4E, SL Welsh 14973 (1977); T27S, R5E, SL and ER Welsh 21214 (1982). Astragalus henrimontanensis Welsh Family; Fabaceae. Federal designation: category 3C. Distribution of the Dana milkvetch is re- stricted to Garfield County where all but one population occurs on the Henry Mountains. The other is from the Aquarius Plateau. This species grows in quaternary alluvium and colluvium over various geologic strata in a mixed ponderosa pine, pinyon, juniper, and sagebrush community. Elevational range ex- tends from 7400 to 9200 feet. Garfield County; T31S, RIOE, A Cronquist and N Holmgren 9296 (1961); T.33S. R3E, NH Holmgren et al. 2106 (1965); T32S, RIOE, SL Welsh 9817 (1970); T31S, RIOE, B Wood s.n. (1972); T31S, RIOE, E Neese 1697 (1976); T31S, RIOE, E Neese 1738a (1976); T31S, RIOE, E Neese & S White 3087 (1977); T31S, RIOE, E Neese, S White 3415 (1977); T32S, RIOE, E Neese and S White .3459 (1977); T.33S, RUE, E Neese & S White 4096 (1977); T31S, RIOE, E Neese 5061 (1978); T33S, RIOE, E Neese 5128 (1978); T34S, RUE, E Neese & A Tve 9948 (1980). Astragalus lentiginosus Dougl. ex Hook, var. pohlii Welsh & Barneby Family: Fabaceae. Citation: Iselya 2: 1-2. 1981. Federal designation: category 3C. The Pohl milkvetch is known from three localities in Tooele County, one from south of Benmore, another from north of Vernon, and the third from Skull Valley. This unique member of the lentiginosus complex grows in silty gravels of a greasewood-sagebrush com- munity type at an elevation ranging from 4350 to 5500 feet. Tooele County: T9S, R4W, JD Walker s.n., Paratype (1962); T7S, R5W, SL Welsh 16498, 16501, Paratype (1978); T8S, R5W, SL Welsh et al. 16743, Holotype (1978); T4S, R8W, A Tave 1332 (1981). 204 Great Basin Naturalist Vol. 45, No. 2 Astragalus lutosus Jones Family: Fabaceae. Federal designation: category 3C. Populations of the Dragon milkvetch are centered in Utah near the junction of the White River and the Colorado-Utah border in eastern Uintah County. However, the plant is abundant over a wide area and also occurs in Colorado. The species is confined to white shale outcrops of the Green River Shale Formation. Habitat for the taxon is a mixed desert shrub community, and eleva- tional range extends from 5200 to 6950 feet. During the 1982 field season one dozen new collections of the species were made, twice as many as had been previously collected. Two of the recent collections come from Wasatch County, which extends the distribution of this species in Utah. Uintah Cointy: White River, ME Jones, Isotvpe (1908); T12S, R25E, ME Jones, Paratvpe (1908); Ti2S, R24E, E Neese & JS Peterson 6103 (1978); T12S, R24E, L England 1772 (1979); T12S, R24E, L England 1775 (1979); TllS, R24E, SL Welsh 19617 (1980); T12S, R24E, E Neese & F Smith 11387, 11391 (1982); TllS, R24E, E Neese & C Fullmer 11468 (1982); TllS, R25E, E Neese & C Fullmer 11469, 11476 (1982); T12S, R24E, K Thorne & B Neelv 1791 (1982); T12S, R25E, K Thorne et al. 1827 (1982); TllS, R25E, E Neese & C Fullmer 11487 (1982); TIOS, R25E, K Thorne & C Fullmer 1937 (1982); TIOS, R25E, E Neese & C Fullmer 11677 (1982); TIOS, R25E, E Neese & F Smith 12054 (1982). Wasatch County: T5S, RllW UBM, S Goodrich, 11790, 11793 (1982); T5S, RllW UBM, S Goodrich et al. 19733 (1983). Astragalus monumentalis Barneby Family: Fabaceae. Federal designation: categorv 3C. The monument milkvetch occurs in San Juan County from Canyonlands National Park to west of Natural Bridges National Monument and in Garfield County west of the Colorado River, near Hite. The plant grows on exposed rim rock and slick rock of the Cedar Mesa Sandstone Formation in pin- yon-juniper and warm desert shrub commu- nities. Elevational range extends from 4000 to 62(X) feet. (;ahih;i.i) Cointy: T33S, H14K, SI. Welsh 5214 ,1966); T33S. R14E, SI. Welsh & (; .Moore 7122 (1968); T32S. R12E, KD Kaneko 24 (1971); T,33S, R14E, SL Welsh 11027 (1971); T.32'/2S, R15E, A Cronquist 11382 (1976); T33S, R14E, E Neese & S White 2816 (1977); T33S, RUE. SL Welsh 20352 (1981); T33S, R14E, KD Heil 1622 (1983); T.33S, R14E, SL and BT Welsh, M Chatterlev 21953 (1983); T32S. R14E, SL Welsh 21784 (1983). San Juan County: T36S. R16E, BE Harrison 11595 (1950); T30S, R19E, SL Welsh et al. 2888 (1964); T31S, R19E, SL Welsh et al. 2901 (1964); T30S, R19E, SL Welsh et al. 2954 (1964); T30S, R19E, G Moore .307 (1965); T30S, R19E, G Moore 348a (1965); T34S, R15E, SL Welsh 5191 (1966); T35S, R5E, SL Welsh 5211 (1966); T34S. R15E, SL Welsh 5212 (1966); T30S, R19E, SL Welsh 7030 (1968); T.30S, R19E, SL Welsh et al. 8900 (1969); T.33S, R14E, CS Schoener 24 (1971); T36S, R15E, D Atwood 7503 (1980); T30S, R19E. SL Welsh et al. 21113 (1982); T.32S, R18E, SL Welsh 21144 (1982); T33S, RUE, BT, SL & ML Welsh 1428 (1983). Cryptantha johnstonii Higgins Family: Boraginaceae. Federal designation: category 3C. Occurrences of the Johnston catseye are known only from an area in the middle of Emery County where the plant grows on low rolling hills and sandy clay soil of the Carmel Formation. Associated vegetative types in- clude mixed desert shrub and scattered pin- yon and juniper communities. Elevational range of the species extends from 5200 to 5800 feet. Emery Cointy: San Rafael, Harrison 5628 (1931); T19S, RUE, Higgins 1310, Holotvpe, Isotvpe (1968); T20S, R13E, LC Higgins 3520 (1970); T20S,' R13E, LC Higgins ,3522 (1970); T22S, R8E, A Cronquist 11417 (1976); T20S, R9E, NH Holmgren et al. 9111 (1979); T23S, R9E, D Atwood 7211 (1979); T22S, R8E, SL Welsh 20457 (1981); T19S. R12E, D Atwood 9236 (1983). Eriogonuni clavellatum Small Family: Polygonaceae. Federal designation: threatened. categor\' 2. The Comb Wash buckwheat is restricted in distribution to a small area west of Bluff in San Juan County near Comb Wash. It occurs on the clay or sandy clav slopes of the Moen- kopi Formation in the desert shrub commu- nity. Elevation extends from 3300 to 4500 feet. The species has also been di.scovered in Montezuma County, Colorado. San Juan County: Barton Range, Eastwood 132, Tvpe-NY (1895); T42S, R21E, Maguire 5853 (1933); T40S, R20E, Reveal et al. 840 (1967); T40S. R20E, At- i uood 2472 (1970); T40S. R2()E, \\'elsh & Atwood 9977 I 1 1970); T4()S, R2()E. D Atwood 7183 (1979); T41S, R19E, 1 D Atwood 7872(1981). EriogonuDi con/mhosum Benth. var. matthcwsac Reveal Fainilv: i'oK gonaccac. Federal designation: category .3C;. April 1985 Welsh, Chatterley: Utah's Rare Plants 205 The Matthews buckwheat is known from near Zions Canyon in Washington County. Taxonomic problems exist with this variety. It is probable that it should be considered as a morphological subunit of another species in the genus. The taxon occurs on the purplish siltstone and sandy loam soil of the Chinle fonuation at 3800 to 4000 feet elevation. Washington County: T42S, RlOW, Welsh et al. 9509, Isotvpe (1969); T41S, RlOW. R Foster 5663 (1977); T42S. RlOW, JD Brotherson 3124 (1982); T41S, RlOW, J Anderson (1982). Eriogoniirn nanum Reveal Family: Polvgonaceae. Federal designation: category 3B. High elevation talus slopes and limestone outcrops in the subalpine conifer zone are habitat for this species. Elevation ranges from 8,500 to 10,050 feet. Until recently, occur- rences of the dwarf buckwheat were consid- ered restricted to Weber and Box Elder coimties. However, in 1977 the plant was collected on Mt. Bartles, in Carbon County. Current taxonomic studies suggest that the variation attributed to this taxon may stem from ecological responses and that the plant may not be worthy of specific designation. Box Elder County: T8N. RIW, JL & CG Reveal 4608, 4609 (1976). Carbon County: T13S, R14E, S Welsh & S Clark 15905 (1977). Weber County: T8N, Rl, Reveal & Holmgren 665, Type (1964); T5N, RIE, SL Clark 2224 (1972); f 8N, RIW, SL Clark 2.327 (1972). Eriogonum tumuloswn (Barneby) Reveal Family: Polvgonaceae. Federal designation: category 3C. This small mat-forming buckwheat occurs primarily in central Duchesne County. How- ever, several populations exist in Emery Coimty, one population has been reported from Uintah County, and one population has been discovered in Juab County. The plant grows in clay soils and sandstone ledges and is associated with a mixed desert shrub, pin- yon-juniper community type. Elevational range extends from 5700 to 6600 feet. IX:cHESNE CouNTY': T3S, R6W UBM, NH Holmgren et al. 1941 (1965); T4S, R5W UBM, JL & CG Reveal 849 (1967); T2S, R5W UBM, JL Reveal 964 (1968); T3S, R5W UBM, E Neese 4414 (1978); T3S, R5W, Welsh 18516 (1979); TIS, R7W UBM, R Foster 7846 (1979); T4S, R6W UBM, E Neese & SL Welsh 8897 (1980); T3S, R6W UBM, B Albee 5015 (1981); T4S, R5W UBM, K Thome & C Fullmer 1624, 1739 (1982); T3S, R6W UBM, S Goodrich 16749 (1982); T4S, R5W UBM, K Thorne, C Fullmer 1739 (1982). Emery Cou.nty: San Ra- fael Swell, WP Cottam 52,33 (1932); T19S, R13E, LC Higgins & JL Reveal 1297 (1968); T24S, RUE, ND At- wood 1849 (1969); T24S, R9E, JG Harris 969 (1980). Juab County: T17S, R2W, S Goodrich 181,37 (1983). Uintah County: T3S, R21E, S Goodrich 5686 (1976). Giitierrezia sarothrae (Pursh) Britt. & Rusby var. pomariensis Welsh Family: .^steraceae. Federal designation: category 3C. This Uinta Basin endemic is known from Duchesne and Uintah counties but occurs primarily near Vernal. It grows in clay, silt, or sandy clay substrates, on semibarren sites, in desert shrub and scattered juniper commu- nity types. This is a unique but somewhat common phase of G. sarothrae, and much possible habitat occurs within its relatively limited distributional range. Elevations ex- tend from 4900 to 7000 feet. Duchesne County: TIS, R4W UBM, E Neese & SL Welsh 8274 (1979). Uintah County: T4S, R24E, J Brotherson 800, 803 (1965); T4S, R23E, J Brotherson 831 (1965); T4S, R20E, SL Welsh & G Moore 6754 (1967); T6S, R24E, ND Atwood & LC Higgins 5853 (1973); T6S, R23E, E Neese & JS Peterson 6083 (1978); T5S, R21E, E Neese 6325 (1978); T5S, R22E, E Neese & L England 6686, 6689 (1978); T7S, R21E, BT Welsh & G Moore 249 (1979); TIS, R4W UBM, E Neese, SL Welsh 8274 (1979); TIN, RIE UBM, S Goodrich 1,3688 (1979); T4S, R24E, E Neese & M Chatterlev 9887 (1980); T5S, R19E, Neese et al. 11017 (1981); T5S', R25E, E Neese & J Trent 12294 (1982); T6S, R25E, E Neese & J Trent 12315 (1982); T5S, R25E, E Neese & J Trent 12334 (1982). Lomatium iunceum Barneby & Holmgren Family: Apiaceae. Original Citation: Brittonia 31: 96. 1979. Federal designation: category 3C. This clump-forming lomatium was origi- nally known only from the San Rafael Swell in Emery County, but has since been located on Fishlake National Forest lands in the southeastern corner of Sevier County, and along the Waterpocket Fold in both Wayne and Garfield counties. Habitat for the plant is barren clay draws and shaley hills of the Moenkopi formation in desert shrub and pin- yon-juniper communities. Elevational range of the species extends from 5300 to 7700 feet. Emery County: T20S, RUE, LC Higgins & JL Reveal 1268 (1968); T20S, RUE, SL Welsh 10963 (1971); T20S, RUE, S Daines 26 (1971); T20S, RUE, CS Schoener 59 (1971); T22S, R12E, M Wright 68 (1971); T20S, RUE, B Albee 2594 (1975); T22S, RUE, Welsh et al. 14767 206 Great Basin Naturalist Vol. 45, No. 2 (1977); T20S, RUE, SL Welsh 14770 (1977); T21S, R6E, SL Welsh 147a3 (1977); T25S, R6E, SL Welsh 14794 (1977); T20S. R6E, S Clark & K Taylor 2463 (1977); T21S, R6E, ME Lewis 5366 (1978); T22S, R6E, S White 16, 95 (1979); T22S, RIOE, JG Harris 680 (1980); T20S, R6E, D Atwood 7523 (1980); T25S, RIOE, S Goodrich & D Atwood 16597 (1982). Garfield Covst\: T34S, R7E, S Goodrich & SL Welsh 15648 (1981). Sevier County: T26S, R4E, D Atwood 7601 (1980). Way.ne Cou.nty: T32S. R7E, E Neese 11319 (1982); T32S, R7E, S Good- rich & D .\t wood 16.567 (1982). Lygodesmia entrado Welsh & Goodrich Family: .\steraceae. Original citation: Great Basin Naturalist 40: 83. 1980. Federal designation: category 3C. This white-flowered skeleton plant with a characteristic bird's nest appearance is only known from west-northwest of Moab in Grand County and from southwest of Green- river in Emery County. It is found growing on the Entrada Sandstone Formation in a juniper community type at an elevation of 4800 feet. It is related to the more common L. arizonica; however, its features are strik- ingly different from other plants of Lygo- desmia in Utah. Gra.np County: T24S, R21E, JS Allan 132 (1972); T24S, R19E, SL & SL Welsh 16725, Holotype (1978). Emery County: T22S, R14E, K Despain 401 (1978). Parry a rydhergii Botsch. Family: Brassicaceac. Federal designation: category 3C. The Rydberg parrya is known from near Bald Mountain in Summit County, Leidy Peak on the border of Uintah and Daggett counties, and Dead Horse Pass and the King's Peak area in Duchesne County. The plant is restricted to a habitat of rocky talus slopes and a spruce or alpine tundra community type. It is a high elevation species usually oc- curring between 10,500 and 12,200 feet. l>\(.(.i:rr Cioi sty: Yellowstone Pass, (] Lambert and CL Woods, USFS-INT (1926); T2N, R17E. AH Holm- gren 71,35 (1947). Duchesne County: T4N, R4W UBM, Mnrdock 54 (1950); T4N, R4W UBM, SL Welsh, E Neese, D Atwood 189.33 (1979); T5N, R4W UBM. SL Welsh, E Neese. D Atwood 19010 (1979); T4N, R2W, SL Welsh, E Neese, D Atwood 19042 (1979). Summit County: Uinta Mtns, S Watson ,54, Type-NY (1869); Uinta Mtns, ME Jones s.n. (1890); TIS, RUE, Weins 49,59 (1974); T2N. RUE, Ostler 677 (1977); TIS, R9E. Ostler & McKnight 1628 (1978). Ui.ntah Cou.nty: T4N, HIE UBM. Waite 2.52, 297 (1971): TIS, H19F. D At- wood ct al. 79,55(1981). Sphaeralcea leptophylla (Gray) Rydb. var. janeae Welsh Family: Malvaceae. Original citation: Great Basin Naturalist 40: .36. 1980. Federal designation: category 3C. This variety of globemallow has been ob- served only in the area of the type collections at Canyonlands - National Park, but threats seem to be minimal to the plant. It occurs on sandy slopes in a blackbrush community type along the White Rim road. San Juan County: T29S. R18E, SL Welsh 7064, Para- tvpe (1968); T29S. R18E, SL Welsh 7085, Holotype (1968). Yucca toftiae Welsh Family: .\gavaceae. Federal designation: category 3C. Known populations of the Toft yucca oc- cur along or near Lake Powell in south- eastern Utah. Two populations occur in hanging gardens on the east side of the lake in San Juan County. The other known local- ities occur on ridges or along tributaries adja- cent to the western edge of Lake Powell in Kane County. Habitat for the plant is sandy alluvium and sandstone outcrops. Approx- imate elevation is 4300 feet. Kane Cou.nty: T42S, R7E, SL Welsh & G Moore 11779, Paratvpe (1972); T41S, R8E. ES Nixon et al. 11073 (1982).' San Juan County: T41S, R9E. ND At- wood 4112, Paratvpe (1972); T41S. R9E. SL Welsh 119,3,5a, Holotype (i973). Literature Cited Arnett, G. R. 1983. Supplement to review of plant ta.\a for listing as endangered or threatened species. Federal Register 48(229):53640-53670. Greenwalt, L. a. 1976. Endangered and threatened wildlife and plants. Federal Register 41(117): 24,524-24572. Greenwalt, L. A., and Gehringer. 1975. Endangered and threatened wildlife and plants. Federal Reg- ister 40:44412-44429. Lamberton, R. E. 1980. Review of plant taxa for listing as endangered or threatened species, Federal Register 45(242):82480-82569. Rollins, R. C. 1982. Thclifpodiopsis and Srhocnrwnihc (Cruciferae). Contr. Gray Herli. 212:71-102. W elsh, S. L. 1978. Endangered and threatened plants of Utah: a reevaluation. Great Basin Nat. 38:1-18. Welsh, S. L. 1984. Utah flora: Polygonaceae. Great Ba- sin Nat. 44:519-5,57. Welsh. S. L., N. D. Atwood, and J. L. Ri;\kal. 1975. Endangered, threatened, e.xtiiut, endemic, and rare or restricted I'tah vascular plants, (Jreat Ba- .sin Nat. 35:,327-376, Welsh. S, L,. and K, H. Thorne. 1978. Illustrated man- ual of proposed endangered and threatened pl.uits of I'tah, l',S. Fisii and Wildlife Service puhlicatioiL 318 pp. April 1985 Welsh, Chatterley: Utah's Rare Plants 207 FIGURE i. Fig. 1. Distribution by township and range of Utah's federally listed species. 208 Great Basin Naturalist Vol. 45, No. 2 10 15 20 Fig. 2. I)istril)utii)ii 1)\ tow nsliip and laiii^f of Utali's rare plant taxa under review for listi April 1985 Welsh, Chatterley: Utah's Rare Plants 209 Maps 1-4. Distribution of (1) Arctornecon humilis, (2) Astragalus perianus, (3) Echinocereus engehnannii var. pin-pttreus, and (4) E. triglochidiatus var. inermis. 210 Great Basin Naturalist Vol. 45, No. 2 Mans 5-8. Distiihiitioii of (5) Pcdiocactus .silcri, (6) rinnvlia ar^illiicea, (7) Sricwautiis ^i^ldunis. and (8) S. ap' n ii<'hti(i( April 1985 Welsh, Chatterley: Utah's Rare Plants 211 MAP 10 Maps 9-12. Distribution of (9) Asclepias cutleri, (10) A. welshii, (11) Astragalus amptillariiis, and (12) A. banicbyi. 212 Great Basin Naturalist Vol. 45, No. 2 Maps 13- Ui. Distiihdtioii of (13) Astwfialtis htiniillouii. ■wm,„i.stii. (14) A. cicsncticus. (15) .A. cquisolcnsis. and (16) A. April 1985 Welsh, Chatterley: Utah's Rare Plants 213 Maps 17-20. Distribution of (17) Astragalus harhsonii, (18) A. iselyi, (19) A. lentiginosus var. ursinus, and (20) A. montii. 214 Great Basin Naturalist Vol. 45, No. 2 Maps 21-24. Distiihutioii ol (21 ) A.s7(y/-i»/i/.s sahulosiis, (22) A. slrkiliflonis, (23) A. .s»/?ri»u-)r(/.v var. Ixisdlticus. and (24) .A. uncutlis. April 1985 Welsh, Chatterley: Utah's Rare Plants 215 MAP 26 Map Corijph 15-28. Distribution of (25) Castillejci (uiuaricnsis, (26) C. revealii, (27) Cirsiiim virgineusis, and (28) umthd missouriensis var. nicirstonii. 216 Great Basin Naturalist Vol. 45, No. 2 Maps 29-32. D.stnhul.on .,f ,29. Cniptautha hanuhyi. (30) C. romparta. (31^ C. nvutzfcldtii. and (32) C. ./«to. April 1985 Welsh, Chatterley: Utah's Rare Plants 217 Maps 33-36. Distribution of (33) Cnjptantha jonesiana, (34) C. ochroleuca, (35) Cuscuta warneri, and (36) Cycladenia humilis var. jonesii. 218 Great Basin Naturalist Vol. 45, No. 2 I MAP 39 Maps 37-40. I)istiil)uti()ii ol (37) Ciimoptrni.s hcckii. (38) (?:. Iii^insii. (39) C. miniiniia, and (40) DaUa flaicsccns var. e}>i((i. April 1985 Welsh, Chatterley: Utah's Rare Plants 219 Maps 41-44. Distribution of (41) Draha inaguirei var. hurkei, (42) Epilobiiiin nevadense, (43) Erigeron cronqiiistii, and (44) E. hicJiinensis. 220 Great Basin Naturalist Vol. 45, No. 2 Maps 45-48. D.str.hution of (45) Eriocron maguirn var. .u.guircL (46) E. magunei var. hamsoni,. (41) E. ,nancus, and (48) E. pwscliiticu.'i. April 1985 Welsh, Chatterley: Utah's Rare Plants 221 Maps 49-52. Distribution of (49) Erigewn sionis, (50) E. untennannii, (51) Eriogonwn ammophilum, and (52) E. aretioides. 222 Great Basin Naturalist Vol. 45, No. 2 Maps 53-56. Distrilmtion of ( Ic'cmuin. 53) Erioiion,,,,, nom,uistii. (54) /•;. lumihaoans, (55) E. lannfoHtmu and (56) April 1985 Welsh, Chatterley: Utah's Rare Plants 223 MAP 60 Maps 57-60. Distribution of (57) Eriog,outiin natiim. (58) E. smithii. (59) E. soreditim. and (60) Festiica dusiiclada. 224 Great Basin Naturalist Vol. 45, No. 2 Maps 61-64. Distribution of (61) Caillanlia fhna. {(yD C.ili,i cacspitosa. (63) Chnicocnrpon siiffnitcsrens. and (6- Hediisantm occklentulc var. canonc. April 1985 Welsh, Chatterley: Utah's Rare Plants 225 MAP 68 Maps 65-68. Distribution of (65) Heterotheca jonesii, (66) Hymenoxijs depressa, (67) H. helenioides, and (68) ^epidhnn harnebijanum. 226 Great Basin Naturalist Vol. 45, No. 2 Maps 69-72. Distribution of (69) ].c]ml'uuu uiontanin and (72) LrsciucirUa luiuiilosa. ncescac, (70) L. inoutantiw var. stcllac. (71) L. ostlcri, April 1985 Welsh, Chatterley: Utah's Rare Plants 227 Maps 73-76. Distribution of (73) Lomatiwn latilohinn, (74) Mentzclia aigilkmi. (75) \ajas cacspitosus. and (76) Oenothera acutissiina. 228 Great Basin Naturalist Vol. 45, No. 2 MAP 80 Maps 77-80. Distribution hracteahis. of (77) Pcdionu-tm tocii.stis Crow Order Chamae.siphonales Chainacsiphon incnistans C.run. Xcnococcus species Order Oscillatoriales Auahacna azollac Stras. Anahacna cylindrica Lemm. Anabaena flos-aciuae (Lyngb.) Breb. Anahaena osciUarinidcs Bory Anahacna spiroidcs Kleb. Anahacna spiroidcs var. crassa Lemm. Anahacna toruhmi (Carm.) Lager. Anahaena variahili.s Kuetz. Anahaena species Anahaena species Aphanizonienon flos-aijiuw (Lemm.) Rails Liinojiija acru'^inco-cacrulca (Kuetz.) Com. Lynohija (icstuarii (Mert.) Liebm. lAjnoJnja cpiphytica Hier. ex Eng. & Prant. Lynghi/a major Meneg. Lynghya majusculla (Dill.) Harv. Lynghya martcn.siana Meneg. Lynghya species Microcolcus pahidosus Kuetz.) Com. Nodidaria harieyana (Thwaites) Thiuet Nodtdaria sputnigena Mert. Nostoc caeruleuni Lyngb. Nostoc species Oscillatoria agardhii Com. Oscillatoria aniocna (Kuetz.) Com. Oscdhitoria ampJiihui Ag. Oscillatoria angusta Koppe Oscillatoria angu.stissiuia West & West Oscillatoria articulata Card. Oscillatoria limosa (Roth) Ag. Oscillatoria nigro-viridis Thwaites Oscdlatoria sancta (Kuetz.) Gom. Oscillatoria subbrevis Schmidle Oscillatoria sid)tUlissinta Kuetz. Oscillatoria tenuis Ag. Oscillatoria species Oscillatoria species Phormidiuiu anibiguum Gom. Phormidiuw autumnale (Ag.) Gom. 240 Table 1 continued. Great Basin Naturalist Vol. 45, No. 2 Species Phonnidhim incrustatitm (Naeg.) Gom. Phormidiiim iniindatum Kuetz. Phonnidium teniie (Meneg.) Gom. Phormiditiin species Plectonema honjomin) Gom. Rivtdcirio species Schizothrix lacustris .\. Braun ex Kuetz. Spiridina laxa G.M.Sm. Spimlina major Kuetz. Tohipothhx species ro/[/pof/ui.v species Reference 12 3 4 5 9 10 11 12 13 14 14° 14" 14° 14° 14° Chuorophyta Class Chlobophyceae Order Volvocales Carteria cordifonnis (Cart.) Dill. Carteria klebsii (Dang.) France em. Troitz. Carteria stelUfera Nyg. Chlamiidomonas altera Skuja Chlamiidomonas globotia Snow Chlamiidomonas pohjpijrenoidcum Pres. pAidorina elegans Ehr. Pandorina mortim (Muell.) Bory Pandorhui species Plcodorimi iUinoisensis Kofoid Pteromonas cordiformis Lenun. em Fott Pteromonas cniciata Playf. Pteromonas species Sphaerellopsis aulata (Pasch.) Gerl. WislouchieUa planktonica Skv. Order Tetrasporales Gloecijstis ampla (Kuetz.) Lager. ^phaerocijstis schroeteri Chod. Tetraspora hdmca (Roth) .\g. Order Ulotrichales Ciilindrocapsa gcHiuu/Zn W'olle Geminella scalarifonnis G.West Microspora crassior (Hansg.) Hazen Microspora floccosa (Vauch.) Thur. Microspora loefgrennii (Nordst.) Lager. Microspora stagnorum (Kuetz.) Lager. Microspora tumidtihi Hazen Microspora species VIothrix aecpialis Kuetz. I'hlhrix cijlindrictim Pres. VIothrix scalarifonnis G.West VIothrix tencrrima Kuetz. VIothrix teninssima Kuetz. I'lothrix variabilis Kuetz. VIothrix zonula (Web. & Mohr.) Kuetz. VIothrix species Order Chaetophorales Aphanocliaete repens A.Bravm Chaetophora elegans (Roth) Ag. 4 3 4 10 10 10 10 10 10 10 10 10 April 1985 RusHFORTH, Squires: Utah Lake Algae 241 Talile 1 continued. Specie,' Reference 1 2 3 4 5 6 7 8 9 10 11 12 13 (Jo/igro.si'm species Miixonema ( = Stigeocloniiint) species .Sfig«)r/o/i/i(;»i aestivale Hazen Sti'^cocloniiiDi attcnuatiim (Hazen) Coll. Stifi,eo(loniuin stdgnatilc (Hazen) Coll. Stii^eocloniiiDi siibscciimhiiti Kuetz. Stigeocloniiiiii tcuiic (.-^g.) Kuetz. 3 10 Order Oedogoiiiales ('(Y/(><;()iii(//» nipiUifonuc Kuetz. Oc(li>f!,(>niuiii ((ipillifonuc var. dehdnjanum (Clir Ocdo'^oniii»i species Hirn. Order Ulvales hMtcwinorplid niniUi (Roth) Ag. Enteromorpha intestinalis (L.) Crev. Enteromo)-pha prolifera (Dan.) Ag. Enteromorpha species ? (recorded as Entcroiuorclia) Entero)no)-pha species Order Cladophorales Cladophora adlicoma Kuetz. Cliidophora crispita (Roth) Kuetz. Cladophora fracta (Dill.) Kuetz. (ladophora glomerata (Lenim.) Kuetz. Cladophora insignis (Ag.) Kuetz. C'ladophora species RhizocloniuDi hicroghjphicuiii (Ag.) Kuetz. Order Chlorococcales Actinastruni graciliiiuim C.M.Sni. Actinastriim hantz-whii Lager. Actinastrum hantzschii var. elongattim G.M.Sni. Actinastriim hantzschii var. fltiviatile Schr. Ankistrodcsmus convohitus Corda Ankistrodesmus falcatus (Corda) Ralfs Ankistrodcsmus falcatus var. mirahilis (West & West) G.S.West Ankistrodesmus falcatus var. stipitatus (Chod.) Lemm. Ankijra judaiji (G.M.Sm.) Fott. Botniococcus braunii Kuetz. Botryococctis sudcticus Lemni. Characium species Chlorococcum infusionum (Schr.) Meneg. Clostcriopsis longissinta var. tropica West & West Coclastrum microporum Naeg. Crucigcnia quadrata Morr. Crucigcnia tctrapedia (Kirch.) West & West Dictijosphacrium chrenhergianum Naeg. Franceia droescheri (Lemm.) G.M.Sm. Kirchneriella hinaris (Kirch.) Moeb. Lagedu'imia longiscta var. major G.M.Sm. Lagcrhcimia uratisUncicnsis Schr. Micractinium pusillum Fres. Ooci/.sfi.s' horgei Snow Ooci/stis eUiptica W.West Oocystis gigas Arch. Ooci/stis glococtistiformis Borge Ooci/stis lacustris Chod. 242 Great Basin Naturalist Vol. 45, No. 2 Table 1 continued. Reference Species 12 3 4 5 6 7 9 10 11 12 13 14 Oocystis novae-seinliae Willie Ooci/fitis parca West & West Oocysfis piisUla Hans. Oocystis suJimarina Lager. Pedidsfnini lx>n/aniiiu (Turp.) Meneg. Pcdiastrwn duplex Meyen Pcdkistrum duplex var. bidchi/lohuDi .\.Braun Pedkistruiu duplex var. elathnitum (.\.Braun.) Lager. Pediastrum duplex var. graeilimum West & West Pedicistruiii simplex (Meven) Lemm. PedidstruDi simplex var. duodenarium (Bail.) Rabh. Pediastrum tetras (Ehr.) Ralfs. Pediastrum tetras var. tetraodon ((Chorda) Rabh. Pediastrum species Plauktosphaeria i^elatinosa G.M.Sni. {hiadri<^ula elosteriodcs (Bohlin) Printz Quadrii^ula laeustria (Chod.) G.M.Sm. Seenedesmus ahundans var. brevieauda G.M.Sni. Seenedesmus aeuminatus (Lager.) Chod. Seenedesmus areuatus var. plati/disca G.M.Sm. Seenedesmus bijuga (Turp.) Lager Seenedesmus bijuga var. alterans (Rein.) Hansg. Seenedesmus bijugfi var. flexuasus Lenini. Seenedesnuis dimorphus (Turp.) Kuetz. Seenedesmus longus var. naegelii (Breb.) G.M.Sni. Scenedes))ius opoliensis P.Richter Seenedesnuis perforatus Lenini. Seenedesmus quadrieauda (Turp.) Breb. Seenedesnuis quadrieauda var. !on<^ispina (Chod.) G.NLSni. Seenedesmus quadrieauda var. jxinus G.M.Sni. Seenedesmus quadrieauda var. maximus West & West Seenedesmus quadrieauda var. quadrispina (Chodat) G.M.Sni. Seenedesmus quadrieauda var. uestii G.M.Sm. Scenedesm us species Seenedesmus species Sehroederia setigera (Schr.) Lemm. Schroederia species Selenustrum bibraian.um Reinscii Selenastrum graeile Reinsch Selenastrum uestii G.NLSm. Treubaria triappendieulata Bern. Order Zygnematales ('losterium moniliferiim (Borv) Ehr. Closterium venus Kuetz. C'losterium species Closterium species Cosmarium margaritiferum Meneg. Cosmarium oeidiferum Lagh. Cosmarium ralfsii Breb. C'osmarium tetraophthalnuim (Knetz.) Breb. (■osmarium species Desmidium species Doeidium species ('•euieiilaria species Mougeotia species Mougeotia species 10 April 1985 RusHFORTH, Squires: Utah Lake Algae 243 Table 1 continued. Species Reference 12 3 4 5 10 11 12 13 14 Spi/ogi/r« (leciniina (Muell.) Knetz. Spirag|/m formoaa (Trans.) Czur. Spiro^yw poiiicali.s (Muell.) CI. Spirogijra species Statirastmm natator West Statirastnim pciradoxum Meyen Stdurastnim tetracerum Ralfs 7Aig,nn)\a species Class Charophyceae CJiara species Class Chrysophyceae Order Chromulinales Hydntnis foetidus (Vill.) Trev. Division Chrysophyta Order Ochroinonadales Dinohnjon hdvaiicum Imhof Diii()/;/!/()n divcroens Imhof DinoI)iyon sockdc var. (iiticricauiim (Brunn.) Bach. Dinobryon sertularia Ehr. Mallomonas acaroides Perty M(dlon}onas caudata Iwan. MalUmioncis pseodocoromita Pres. Malloiitonas tonsiirata Teil. 10 Class Xanthophyceae Order Mischococcales Chamciop.sis cylindrica (Lamb.) Lemm. Chloiohotrys icguUiris (W.West) Bohl. Ophiocytitiiu cocldearc (Eich.) .\.Braun. Opiiiocytiiini ciispidattim (Bail.) Rabh. Opluocytiuiu nuijtis Naeg. Opliiocytiuni pariuhin) (Perty) .\.Braun. Order Tribonematales Trilmiu'iuti Iwinbycinum (Ag.) Derb. & Sol. TrihonciiHi Diiniis Hazen Tiihonciiui utricidostim Hazen Triboncnid species 3 4 3 4 3 10 Order Vaucheriales Vauchcria borcalis Hirn Vciucheria gemincita (Vauch.) DeCand. Vaucheria sessilis var. clavata (Vauch.) DeCand. Class Bacillariophyceae°° Achnanthes affinis Grun. Aclindnthcs chilensis var. subdequdUs Achudnthvs clevei Gnui. Achiidnthes clevei var. rostrdtd Hust. Reim. 14 13 13 14 13 14 244 Great Basin Naturalist Vol. 45, No. 2 Table 1 continued. Reference Species 1 2 3 4 5 6 7 8 9 10 11 12 13 14 Achnanthes deflexa Reim. 10 14 Achnanthes exigito Grvm. 10 11 13 14 Achnanthcs gibherula Griin. 14° AchnantJies hauckiana Grwn. 9 10 11 13 14 Amphora species 9 Ano)noeoneis costata (Kiietz.) Hust. 6 14 Anomoconeis serians var. bracht/sira (Breb. ex Kuetz.) Hust. Anomoeoneis sphaerophoni (Ehr.) Pfitz. Anoiuoeoneis sphaeropJwra var. ^ucutheri O.Miiell. Anmnoconeis vitrea (Grun.) Ross Anomoconeis species Asterionella fonnoso Hassall Bacilhiria paradoxa Gnielin BaciUaria paxilUfer (Muell.) Hend. ( = B. paradoxa) Biddulphia kievis Ehr. Caloneis amphishaena (Bory) CI. Caloneis bacillaris var. thermalis (Grun.) CI. Caloneis haciUiim (Grun.) CI. Caloneis fenzlii (Grun.) Patr. Caloneis fenzUoides Cl.-Eul. Caloneis lamella Zakr. Caloneis lewisii Patr. Caloneis limosa (Kuetz.) Patr. Caloneis oregonica (Ehr.) Patr. Caloneis permagna (Bail.) CI. Caloneis schttmanniana (Grun.) CI. Caloneis schiimanniana var. fasciata Hust. Caloneis srhiimanniana var. linearis Hust. Caloneis silicula (Ehr.) C\. Caloneis silicula var. limosa (Kuetz.) VanLand. Caloneis ventricosa (Ehr.) Meist. Caloneis ventricosa var. subundulata (Grun.) Patr. Caloneis lentricosa var. truncatula (Grun.) Meist. Campiilodisciis chjpciis Ehr. Campylodiscus hibernicus Ehr. Camptjlodiscus noriciis var. hibernicus (Ehr.) Grun. (= Campylodiscus hibernicus) Chaetoceros elmorei Boyer Cocconema ( = Cijmbclla) species Cocconcis diminuta Pant. Cocconcis discidus (Schuni.) CI. Cocconeis fluviatilis Wall. Cocconcis pcdiculus Ehr. Cocconcis placcntula Ehr. Cocconcis placcntida var. euglypta (Ehr.) CI. Cocconcis placcntula var. lineuta (Ehr.) V.H. Coscinodiscus lacustris Grun. Coscinodiscus species Cyclotclla antiqua W.Sni. Cijclotclla bodanica Eulen. Cyclotclla cointa (Ehr.) Kuetz. Cyclotclla kuetzingiana Thwaitcs Cyclotclla kuetzingiana var. pUnutophora Fricke Cyclotclla meneghiniana Kuetz. Cyclotclla meneghiniana \ar. pumihi (Gnui. ex V.H.) Hast. Cyclotella ocellata Pant. 14 0 11 13 14 11 13 14 11 0 11 13 14 0 11 13 14 0 11 14 0 11 13 14 11 14 0 11 13 14 0 11 13 14 11 13 13 14 11 14 13 14 11 13 14 11 11 13 14 11 13 11 14 14 13 14 11 14 2 14 2 6 10 11 13 14 6 11 13 14 6 6 10 11 13 14 6 14 6 9 10 11 14 6 9 9 10 10 10 10 11 1 2 13 14 14 14 14 f) 9 10 11 11 13 14 14 6 9 10 11 1 11 2 13 14 14 6 10 11 13 14 April 1985 RusHFORTH, Squires: Utah Lake Algae 245 Table 1 continued. Reference Species 1 2 3 5 6 7 9 10 11 12 13 14 CijchtvUa stcUigcra (CI.) Grun. Cticlotclhi stiidtd (Kiietz.) Grun. Cijclotclhi striata var. ambigiia Grun. Ctjchtclki species 1 CycloteUa species 2 Ciilincbotheca gracilis (Breb.) Grun. Cijinatoplcura angulata Grev. Cijmatopleura elliptica (Breb.) W.Sni. Cipmitopk'tira elliptica var. constricta Grun. C.yinatopleura solea (Breb.) W.Sm. Cymatopleura solca var. palffiji (Pant.) Cl.-Eul. Cijmatopleura solea var. regula (Ehr.) Gnin. Cipnhella affinis Kuetz. CAjnihclla ai)iphirephala Naeg. e.\ Kuetz. Cipnhella ci.stula (Ehr.) Kirch. Cijnihella cijmbiformis Ag. Cijmhella ehrenberoU Kuetz. Cijinbella inaecjiiali.s (Ehr.) Rabh. Cipiilnlta lucxicana (Ehr.) CI. Cipnhella niieroeephala Grun. CipnJyella miniita Hilse ex Rabh. Cipnhella miniita i. latens (Krasske) Reiin. Cipnhella ntiniita var. pseiidograeilis (Choln.) Reim. Cipnhella ntiniita var. silesiaca (Blei.sch ex Rabh.) Reim. Cipnhella muelleri Hust. Cipnhella muelleri f. ventricosa (Temp, et Perag.) Reim. Cymhella prostrata (Berk.) CI. Cymhella prostrata var. aeursualdii (Rabh.) Reim. Cymhella protracta Ost. Cymhella piisilla Grim. Cymhella sinuata Greg. Cymhella suhturgida Hust. Cymhella tumida (Breb. ex Kuetz.) V.H. Cymhella tumidiila Grun. ex A.Schmidt Cymhella turgida Greg. Cymhella species Cymhella species Cymhella species 1 Cymhella species 2 Denticula elegans Kuetz. Denticula elegans var. kitfoniana (Grim.) DeT. Denticula elegans f. valida Pedic. Denticula subtilis Grun. Denticula tenuis Kuetz. Denticula tenuis var. crassula (Naeg.) West & West Diatonui anceps (Ehr.) Kirchn. Diatoma hiemale (Roth) Hieb. Diatoma hiemale var. mesodon (Ehr.) Grun. Diatoma tenue Ag. Diatoma tenue var. elongatum Lyngb. Diatoma vulgare Bory Diatoma vulgare var. breve Grim. Diatoma vulgare var. grande (W.Sm.) Grun. Diatoma species Diatomella halfouriana Grev. Diploneis elliptica (Kuetz.) CI. Diploneis nuirginestriata Hust. Diploneis oblongella (Naeg. ex Kuetz.) Ross Diploneis oculata (Breb.) CI. 1 2 10 14 14' 14' 11 11 10 14 14 10 11 13 14 14 10 11 13 14 14 10 11 13 14 11 14 10 11 13 14 10 11 13 14 11 13 14 10 11 13 14 10 14 14 14 11 14 10 11 13 14 10 11 12 13 14 11 13 14 10 11 14 11 13 14 14 11 14 10 11 14 10 11 13 14 10 14 10 10 10 10 11 11 13 14 14 14 14 11 14 14 11 14 11 14 10 11 13 14 10 11 1 2 14 10 11 13 14 10 11 14 11 14 11 13 14 11 13 14 10 11 13 13 14 246 Table 1 continued. Great Basin Naturalist Vol. 45, No. 2 Reference Species Diploneis psciiclocalis Hust. Diploneis pitella (Schiim.) CI. Diploncis smithii (Breh. ex W.Sm.) CI. Diploneis siuithii var. (lilatatd (M.Perag.) Bover Diplonei.s smithii var. pumilo (Grun.) Hu.st. Diploneis species 1 Diploneis species 2 Diploneis species Encynonema (= Cipnhelhi) species Entonwneis alata (Ehr.) Ehr. Entomoneis piiludosa (W.Sm.) Reim. Epithemia adnata (Kuetz.) Breh. Epithemia adnata var. minor (Perag. et Herih.l Pati Epitliemia adnata var. porceUus (Kuetz.) Patr. Epithemia adnata var. prohoseidea (Kuetz.) Patr. Epitliemia argus (Ehr.) Kuetz. Epithemia ar^iis var. alpestris Grun. Epithemia argus var. longieomus (Ehr.) Gnui. Epithemia argus var. pmtracta A.Maver Epithemia hi/ndmanii W.Sm. Epithemia intermedia Fricke Epithemia ocellata (Ehr.) Kuetz. Epithemia sorex Kuetz. Epithemia turgida (Ehr.) Kuetz. Epithemia turgida var. gramdata (Ehr.) Bnm. Epithemia turgida var. westermannii (Ehr.) Grun. Epithemia zebra (Ehr.) Kuetz. Epithemia zebra var. saxoniea (Kuetz.) Grun. Epithemia species Eunotia areus var. bidens Gnm. Eunutia eunata (Kuetz.) Lagerst. Eunotia diodon Ehr. Eunotia ineisa W.Sm. e.x Greg. Eunotia peefinalis (O.F.Muelf.) Rabh. Eunotia peetinalis var. minor (Kuetz.) Rabh. Eunotia tenella (Gnm.) CI. Eunotia species Eragiliiria hieapitata A.Maver Eragilaria brevistriata Grun. Eragilaria brevistriata var. capita ta Heril). Eragilaria brevistriata var. inflata (Pant.) Hust. Eragilaria eapueina var. mesolepta Rabh. Eragilaria eonstruens (Elir.) Grun. Eragilaria eonstruens var. /)inof/« (Ehr.) Grun. Eragilaria eonstruens var. p(/(ni7« Grtm. Eragilaria eonstruens var. trnfrr (Ehr.) Cirun. Eragilaria erotonensis Kitton Eragilaria lapponiea Cwun. Eragilaria leptostauron (Eln.) Hust. Eragilaria leptostauron var. (/,//;/V; (Cirun.) Hust. Eragilaria piniuita Elir. Eragilaria pinnata var. intereedens ((^,run.) Hust. Eragilaria pinnata var. laneettula (.Schum.) Hust. Eragilaria similis Krasske Eragilaria vaucheriae (Kuetz.) Peters, Eragilaria vaucheriae var. eapitellata ((.Wuu.) Patr. Eragilaria vireseens Ralfs Erustulia rhomboides (Ehr.) DeT. 12 3 4 5 6 7 8 9 10 11 12 1.3 14 10 11 10 11 10 11 11 10 11 9 10 1] 9 u 9 11 11 11 11 11 11 10 1] 10 11 10 11 10 11 11 13 14 13 14 13 14 14 13 14 3 14 13 14 13 13 14 14 13 14 13 14 14 13 14 14 14° 14 14 14 13 14 14 14° 14 14 13 14 14 13 14 13 14 13 14 13 14 13 14 13 14 1.3 14 1.3 14 14 14 14 13 14 13 14 14° 13 14 14 April 1985 RusHFORTH, Squires: Utah Lake Algae 247 Table 1 continued. Reference Species 1 2 3 5 6 7 8 9 10 12 13 Fnistiilid vtiloaris (Thwaites) DeT. Goiiiphoncis lierctilcana (Ehr.) CI. Goinphoncma (icuminatiim Ehr. Ginnphonema affine Kuetz. Gomplionema affine var. insionc (Greg.) Andrews Gomphonema anau.statuiu (Kuetz.) Rahli. Goinplioneitui (lugiistatiiin var. intcnucdin Grun. GoDiplioneuia rlcvei Fricke Gimiphoncnid dicliotoniKin Kuetz. Goniplioneina gracile Ehr. em. V.H. Gomphonema infricattim Kuetz. Gomplionema intrieattim var. vibrio (Ehr.) CI. Gomphonema kinceolotiini Ag. i'.omphoncma olivacetim (Lvngb.) Kuetz. Gomphonema olivaecum var. eah-area (CI.) CI. Goniphonema panuhim Kuetz. G<^mpIionema septum Mogh. Gomphonema sphaerophorum Ehr. Gomphonema stihckivahim (Grun.) Grun. Gomphonema subckwatum var. rommutatiim (Grun.) A.Mayer Gomplionema siibclavatum var. ntexieantim (Grun.) Patr Gomplionema teneUtim Kuetz. Gomphonenia tnineattnn Ehr. Gomphonema tnincatitm var. eapitatum (Ehr.) Patr. Goniphonema tnineatum var. tni^^idiim (Ehr.) Patr. Gomphonema ventrieosiim Greg. Gomphonema species Gomphonema species Gi/rosiiima acuminatnin (Kuetz.) Rabh. Gyrosigma attenuatum (Kuetz.) Rabh. Gyrosionia maeram (W.Sm.) Griff. & Henfr. Gy io.sii!,ma speneerii (Quek.) Griff. & Henfr. Gywsiffna speneerii var. eiintila (Grun.) Reini. Gyrosi'^ma strigilis (W.Sm.) CI. Uannaea areas (Ehr.) Patr. Hantz.sehia amphioxys (Ehr.) Grun. Hantz.sehia amphioxys i. eapitata O.Muell. H(n]tz.sehi(i virgata (Roper) Gnm. Mastogloia elliptiea var. danseii (Thwaites) CI. Mastooloia piimila (Grun.) CI. Mastogloia smitliii Thwaites ex W.Sm. Mastogloia smidiii var. laeustris Grun. Melosira distans (Ehr.) Kuetz. Melosira granulata (Ehr.) Ralfs Melosira granulata var. angustissima O.Muell. Melosira islandiea O.Muell. Melosira italica (Ehr.) Kuetz. Melosira roseana Rabh. Melosira varians Ag. Melosira species Meridian eireulare (Grev.) Ag. Meridian circidure var. eonstrietum (Ralfs) V.H. Meridian species Navicula abiskoensis Hust. Navieula aeeepta Hust. Navicula aecomoda Hust. 10 10 9 9 10 9 10 9 10 10 9 9 10 9 10 9 10 8 9 10 9 10 9 10 10 13 13 14 12 13 14 13 13 13 13 13 13 12 13 12 13 13 12 13 13 13 14 13 14 248 Great Basin Naturalist Vol. 45, No. 2 Table 1 continued. Reference Species \avict(1a aniphihola CI. Xdvkiihi (inolica Ralfs \(ivicu](i iin^iista Griin. Xciiinihi arenaria Donk. XaLiciila ancmis Hust. Xdviciild (itomiis (Kiietz.) Gnm. Xaviciila aurkiihita Hust. XaciniUi cniroia Sov. Xaviciild hdcillum Ehr. Xdvicuhi hkephdld Hust. Xdviciild cdpitdtd Ehr. Xdviculd Cdpitdtd var. hiingdikd (Gnui.) Ross Xdvktild Cdpitdtd var. Iiinchiirocmi.'i (Grun.) Pati Xdviciihi cinctd (Ehr.) Ralfs Xdviculd circumtextd Meist. ex Hust. Xdviculd dementis Gnm. Xdviculd cnicictild (VV.Siu.) Donk. Xdviculd cniptocephdid Kuetz. Xdviculd cniptocephdid var. venctd (Kuetz.) Rabh. Xdviculd cuspichitd (Kuetz.) Kvietz. Xdviculd cuspiddtd var. herihaudi Perag. Xdviculd cuspiddtd var. mdjor Meist. Xdviculd difficilliinoidc.s Hust. Xdviculd dulcis Patr. Xdviculd el^incnsis (Greg.) Ralfs Xdviculd elginemis var. wstnita (A.Mayer) Patr. Xdviculd exigud Greg, ex Grun. Xdviculd exigud var. cdpitdtd Patr. Xdvicidd festivd Krasske Xdvicidd gdstrum (Ehr.) Kuetz. Xdviculd gotthindicd Grun. Xdviculd graciloides A.Mayer Xdviculd gregdrid Donk. Xdviculd grimmei Krasske Xdviculd hdlophild (Grun.) CI. Xavicula hdlophild f. tenuirostri.s Hust. Xdviculd heufleri Grun. Xdviculd heufleri var. leptocepludd (Breb. ex Gnin.) Perag. Xdviculd inflexd (Greg.) Ralfs. Xdviculd insockibilis var. dissipdtoides Hust. Xdviculd integrd (W.Sni.) Ralfs Xdviculd Idcustrl'i Xdvkuld hievissimd Kuetz. Xdviculd kmceoldtd (Ag.) Kuetz. Xaviculd kipidosd Krasske Xdviculd nienisculus var. upsdlieiisis (Grun.) Grun. Xdviculd niiniiud Gnui. Xdviculd iniuusculd (Jruu. Xdvictdd iiiolestifontiis Hust. Xdvkuld inuralis Grun. Xdvkuld itditkd Kuetz. Xdvkuld ))ditk(i var. cohnii (Hiisc) (irun. Xdviculd tiiuticd var. nivdlis (Ehr.) Hust. (= Xdviculd nivdiis) 3 4 5 6 7 8 9 10 12 13 14 9 9 10 9 10 10 9 10 9 9 10 10 9 10 9 10 10 9 10 10 9 10 11 11 14 13 14 14 13 14 14 14 14 13 14 13 14 13 14 13 14 13 14 14 12 13 14 12 13 14 13 14 13 14 14 14 14 14 13 14 14 13 13 14 13 13 14 14 14 13 14 14 13 14 13 13 14 11 13 11 1 1 13 13 1 1 11 11 13 I! 13 13 14 April 1985 RusHFORTH, Squires: Utah Lake Algae 249 Table 1 continued. Reference Species 12 3 4 5 6 7 9 10 11 12 13 Xdikiila nuttica var. undiilata (Hilse) Grun. Xdiiciila nivalis Ehr. Xiiiicuhi oblongd (Kiietz.) Kuetz. Savictila odiosa Wall. Xavicida omissa Hust. S'dvicuki pellictdosd (Breb. ex Kuetz.) Hilse Xdvicida pcregrina (Ehr.) Kuetz. Xdviciild pldccnttdd (Ehr.) Kuetz. Xdvindd pldccutuhi f. rostrata ,'\. Mayer Xdvicttld pwtwctd Gnui. Xdvicuki protidctd f. suhciipitdtd (Wisl. et For.) Hust. Xdviculd psciidottiscida Hust. Xdiictdd piipidd Kuetz. Xdvictda piipuld var. capitatd (Skv.) Meyer Xdiiculd piipuld var. cllipticd Hust. Xavicidd pupiila var. uuttdta (Krasske) Hust. Xdvicidd piipuhi var. rectdttgtddris (Greg.) Grun. Xdvicula pijgmded Kuetz. Xdiiculd radiosa Kuetz. Xdvicuki radiosd var. pdrvd Wall. Xdvicidd mdiosa var. tenella (Breb. ex Kuetz.) Gnui. Xdiiculd rcinliardtii (Gmn.) Grun. Ndiiculd icinlididtii var. ellipticd Herib. Ndiicidd rliyncocephdia Kuetz. Navicidd rliyncocephdia var. dinpliiccros (Kuetz.) Grun. Naviculd ihyncocephala var. gemiainii (Wall.) Patr. Xdiiculd sdliudnnn Grun. Naviculd s(dindium var. intennedia (Grun.) CI. Navicula schweteri var. escambia Patr. Navicida sciitelloides W.Sni. ex Greg. Navicula secreta var. apiculata Patr. Navicula secura Patr. Navicula seminidum Gnui. Navicula seminuloides Hust. Navicula septata Hust. Navicula strenzkei Hust. Navicula subbacillum Hust. Navicula subhamulata Grun. Navicula symmetrica Patr. Navicula tenelloides Hust. Navicula tcnera Hust. Navicula tripunctata (O.F.Muell.) Borv Navicula tripunctata var. schizonemoides (V.H.) Patr. Naviculd tusculd Ehr. Navicula viridula (Kuetz.) Kuetz. em V.H. Navicula viridula var. avenacea (Breb. ex Grun.) V.H. Navicula viridula var. rostellata (Kuetz.) CI. Navicula uardii Patr. Navicula wittrockii (Lagerst.) Temp, et Perag. Navicida species Navicida species Navicula species Navicula species Neidium affine (Ehr.) Pfitz. 9 10 9 10 9 10 9 10 10 14 14 13 14 14 13 14 13 14 13 14 14 13 14 13 14 13 13 14 13 14 13 14 13 14 12 13 250 Great Basin Naturalist Vol. 45, No. 2 Table 1 continued. Reference Species 3 4 5 6 7 9 10 11 12 13 14 Neidiiim affine var. amptiirliijnctis (Ehr.) CI. Neidium bisulcntuin var. baicalense (Skv. & Meyer) Reim. Neidium dubitim (Ehr.) CI. Xeidium iridis (Ehr.) CI. Nitzsrliia (icicuhiris W.Sni. xV/tor/nV/ acicularoides Hiist. Nitzschia acuta Hantz. Nitzschia ampJnbia Grim. Nitzschia angustata (W.Sm.) Gnin. Nitzschia angustata var. acuta Griin. Nitzschia apiculata (Greg.) Gnm. Nitzschia capitcUata Hust. Nitzschia claiisii Hantz. Nitzschia communis Rabh. Nitzschia denticula Grun. Nitzschia dissipata (Kuetz.) Grun. Nitzschia duhia W.Sm. Nitzschia epithemioidcs Grun. Nitzschia filiformis (W.Sm.) Hust. Nitzschia fonticola Grim. Nitzscliia frustuhim (Kuetz.) Grim. Nitzschia gracilis Hantz. Nitzscliia granulata Grim. Nitzschia liantzschiana Rabh. Nitzscliia holsatica Hust. Nitzschia hungarica Gnm. Nitzschia inconspicua Grun. Nitzschia intennedia Hantz. e.x CI. & Grim. Nitzschia lacunarum Hust. Nitzschia linearis (Ag.) W.Sm. Nitzschia longissima var. closterium (W.Sm.) V.H. Nitzschia microcepliala Grun. Nitzschia minuta Bleisch Nitzschia obtusa W.Sm. Nitzschia ovalis Arnott Nitzschia palea (Kuetz.) W.Sm. Nitzschia paleucea Grim. Nitzschia perminuta Grun. Nitzschia punctata (W.Sm.) Chun. Nitzschia pusilla (Kuetz.) CJrim. em. Lange-Bert. Nitzschia recta Hantz. Nitzschia nmiana Grun. Nitzschia sigma (Kuetz.) W.Sm. Nitzschia sigmoidea (Ehr.) W.Sm. Nitzschia spcctabilis (Ehr.) Ralfs Nitzschia stagnorum Rabh. Nitzschia subacicularis Hust. Nitzschia subtilis (Kuetz.) Grun. Nitzscliia thermalis Kuetz. Nitzschia thermalis var. minor Hilse Nitzschia tryblionella Hantz. Nitzschia tryblionella var. debilis (.\rnott) .A.Mave Nitzschia trijblionella var. genuina Grun. ( = Nitzscliia tnjblionella var. trijhlioncUa) Nitzschia tnjhlionclla var. levidensis (W.Sm.) Grui Nitzschia tryblionella var. victoriae Grun. Nitzschia valdestriata Aleem et Hust. 11 13 6 10 11 12 13 14 6 9 10 11 13 14 6 14 9 10 11 13 14 6 11 11 13 13 14 6 9 11 12 14 14° 6 10 14 14 6 9 10 12 14 14° 6 9 10 10 14 14 6 9 10 13 14 6 9 13 13 14 14 6 9 9 10 13 14 6 10 12 13 14 9 13 13 14 14 6 6 9 10 13 14 9 10 14 14° 6 14 9 10 14 6 9 10 12 13 14 9 10 10 14 6 9 10 13 14 14° 6 13 13 14 14 9 10 13 14 14 9 13 14 14° 14 14 6 10 13 14 6 ]() 10 13 14 10 13 14 6 10 13 14 14° April 1985 RusHFORTH, Squires: Utah Lake Algae 251 Table 1 continued. Reference Species 12 3 4 5 7 8 9 10 11 12 13 Nitzschia species 1 Nitzschia species 2 Nitzschia species 1 Nitzschia species 2 Nitzschia species 3 Opephora martt/i Herib. Pinnularia ahaujcnsis var. hncahs (Hust.) Patr. Pinnuhiria abaiijensis var. siibiinduhita (A.Mayer ex Hust.) Patr. Pinntikiria acrosphaeria W.Sm. Pinnularia appcndiculata (Ag.) CI. Pinnularia biceps Greg. Pinnularia borcalis Ehr. Pinnularia borcalis var. rcctantj^tilaris Carlson Pinnularia brcbissonii (Kiietz.) Rabh. Pinnularia l)urkii Patr. Pinnularia maior (Kuetz.) Rabh. Pinnularia microstauron (Ehr.) CI. Pinnularia molaris (Grun.) CI. Pinnularia nobilis (hr.) Ehr. Pinnularia obscura Krasske Pinnularia rutfneri Hust. Pinnularia viridis (Nitz.) Ehr. Pinnularia viridis var. minor CI. Plagiotropis arizonica Czar. & Blinn. Plagiotropis vitrea (W.Sm.) Gnm. Plagiotropis vitrea var. scaligera (Gnin. ex CI. & Grun.) Perag. Pleurosigma australe Grun. Flcurosigma delicatulum W.Sm. Rliizosolenia minima Levan. Rhoicosphenia curvata (Kuetz.) Grun. ex Rabh. Rhopatodia gibba (Ehr.) O.Muell. Rhupalodia gibba var. vcntricosa (Kuetz.) H. et M. Perag. Rhopalodia gibberula (Ehr.) Muell. Rhopalodia gibberula var. protracta Grim. Rhopalodia gibberula var. vanheurckii O.Muell. Rhopalodia muscidus (Kuetz.) O.Muell. Scolioplcura pcisonis Grim. Stauroncis anccps Ehr. Stauroneis anceps var. siberica Grun. ex CI. Stauroncis kriegeri Patr. Stauroneis muricUa f. linearis Lund Stauroneis phocnicenteron (Nitz.) Ehr. Stauroncis phocnicenteron var. brunii iM. Perag 6c Herib.) Voigt Stauroncis phocnicenteron f. gracilis (Ehr.) Hust. Stauroneis smithii Grun. Stauroneis wislouchii Poretz. & Anis. Stauroncis species Sl(j)li(m(Htiscus astraea (Ehr.) Grun. Stcpltanodiscus astraea var. minutula (Kuetz.) Grun. Stcphanodiscus carconensis var. pusilla Gnm. Stcpltanodiscus dubius (Fricke) Hust. Stcphanodiscus hantz.schii Giim. StcpliaiuHliscus invisitatus Hohn & Heller. Stcphanodiscus minutus CI. & Moell. 11 11 11 10 11 11 11 11 11 11 11 10 11 9 10 11 9 10 11 10 11 10 9 11 10 11 10 11 11 12 13 14 14 14 14 13 14 14 13 13 14 13 14 14 13 14 13 14 13 14 14 14 14 14 13 14 14 13 14 13 14 13 14 14 14 13 14 13 14 13 14 14 13 14 11 14 11 13 10 14 10 11 12 14 11 12 13 12 13 11 12 252 Great Basin Naturalist Vol. 45, No. 2 Table 1 continued. Species Stephanodisnis niagarae Eh Stephanodiscm suhtiUs V.Goor Stephcmodiscus species Stephonodisctis species 1 Steplumodlscus species 2 Surirelhi angusta Kuetz. SurireUa biseriato Breb. SurireUa linearis W.Sm. Si/rire//« /ine«ns var. consthcta Grun. .S»rire//fl otfl/is Breb. r-, r i S»nre//« or«/-.s var. hnohtncUii (W.Sm.) Cl.-Eul. Surirella ovota Kuetz. S(((ire//f/ ouflto var. pinnafw W.Sm. SurireUa patella Ehr. SurireUa pseudoiaUs Hust. SurireUa robusta Ehr. Si/nre//« roZ^i/sta var. spkndida (Ehr.) V.H. SurireUa spiraUs Kuetz. SurireUa striatula Turp. SurireUa species Sunedra aciis Kuetz. S|/nerfrfl amphicephahi var. m<,vfnm« (Grun.) Hust Sipiedra capitata Ehr. Si/nerf»fl cijchpum Bruts. S(/nprfr« delieatissima var. «ng(/.sfi.s.snm/ Grun. Si/'iedrfl fameUca Kuetz. Si/nfdra fasiculata (Ag.) Kuetz. Si/nerfm fasiculata var. fri/nr«ta (Grev.) Patr. Si/nprfra/i/ifoniHS var. e.vi/i.s Cl.-Eul. Synerfra i?iflr.(/»iens!S Sov. Si/nerfra minuscula Grim. Simedr« pflra.sificfl (W.Sm.) Hust. Synprfm pma.sifia/ var. .s»/;con.sfnrf« (Grun.) Hust Si/n«/r« pulchclla Kuetz. Sipiedra radians Kuetz. Si/n«/ra ri(/)i;K^n.v Kuetz. Syn«/r« r»mpen.S' var. /«»u/i«ri.s (Kuetz.) Gnu Si/riedra nnnpens var. fragilarioides Grun. Sr/nef/r« runipen.S' var. »ieneg/iini«nfl Grun. Synedra rumpens var. .scofirfl Grun. Siinedra socia Wall. Stynt'(/m fcnrrn W.Sm. Sifnedra ulna (Nitz.) Ehr. Synedra ulna var. ro/Umt^/ Oestr. Si/nr(/r« ulna var. rl«n!cn (Kuetz.) Grun. S,/n«/»Y/ ulna var. /ongi.s.si»i« (W.Sm.) Brun^ S,/n(Y/»Y; ulna var. oxyrhynehus (Kuetz.) V.H Si/nrf/m ii/nn var. oxyrhynehus f. ' »(cr/i()r()Mfr«rto (Forti) Hust. Synedra ulna var. spathulifera (Grun.) V.H. S'l/ner/ra »/»!« var. subaeqiialis (Grun.) V.H. Synedra species 7V)r//arw//()pecies Reference 12 3 4 5 7 8 9 10 11 12 13 14 Class Euglenophyceae Division Euglenophyta Order Eiiglenales Eii^lcud clirciihvriiii Klehs Eu{^/('»r/ '^racili.s Klclis Eu^h'iui oxyinis Stliniar. Euglena proxima Dang. Etiolena tripfcris (Duj.) Klebs Euglcna liridi.s Ehr. Eu<:,lcu(i species Lepocimlis salina Fritscli Phaciis chloroplastes Pres. Phactis spiralis All. et Jahn PIuiciis tortus (Lenini.) Skv. Stromhomonas fliividtilis (Lenini.) Defl. Trachelomonas crchca (Kill.) Defl. Division Pyrrhophyta Class Dinophyceae Order Peridiniales Ceratiuni hiriindinella (Muell.) Dujard. Order Glenodiniales Glenodinium dinobnjonis (Wol.) Lind. CIcnodiniuui penardiforme (Lind.) Schi Glenodinium species Cryttophyta Class Cryptophyceae Order Cryptomonadales Cliroomonas species Cryptomonas species Column numbers refer to the following papers: 1. Tanner (1930) 3. Snow (1932) 5. Harding (1971) 7. EPA (1977) 9. Grimes et al. (1980) 11. Grimes and Riishforth (1982) 13. Javakul and Rushforth (1983) "Algal species previously unreported from Utah Lake. "Diatoms are listed in alphabetical order. Literature Cited Boll.\.nd, R. F. 1974. Paleoecological interpretations of the diatom succession in recent sediments of Utah Lake. Unpublished dissertation. Univ. of Utah, Silt Lake City. 100 pp. Grimes, J. .\., L. L. St. Clair, and S. R. Rlshforth. 1980. A comparison of epiphytic diatom assem- blages on living and dead stems of the common grass Phragmites australis. Great Basin Nat. 40(3):223-228. Grimes, J. A., and S. R. Rushforth. 1982. Diatoms of recent bottom sediments of Utah Lake, Utah, USA. Bibliotheca Phycologia 55:1-179. 2. Tanner (1931) 4. Harding (1970) 6. Holland (1974) 8. Whiting (1978) 10. Rushforth et al. (1981) 12. Grimes and Rushforth (1983) 14. Present report 1983. Diatoms of surface sediments of Utah Lake, Utah, USA. Hydrobiologia 99:161-174. Harding, W. J. 1970. A preliminary report on the algal species presently found in Utah Lake. Great Ba- sin Nat. 30:99-105. 1971. The algae of Utah Lake. Part II. Great Ba- sin Nat. 31:125-134. Javakul, A., and S. R. Rushforth. 1983. Diatoms in sediment cores in Utah Lake, Utah, USA. Hydro- biologia 98:159-170. Rushforth, S. R., L. L. St. Clair, J. A. Grimes, and M. C. Whiting. 1981. Phvtoplankton of Utah Lake. Great Basin Nat. Mem'. 5:85-100. S.NOw, E. 1932. A preliminary report on the algae of Utah Lake. Proc. Utah Acad. Sci. 9:21-28. 254 Great Basin Naturalist Vol. 45, No. 2 Tanner, V. 1930. Freshwater biological studies at Utah Lake, Utah. Proc. Utah Acad. Sci. 7:60-61. 19,31. Freshwater biological studies at Utah Lake No. 2. Proc. Utah Acad. Sci. 8:198-203. Squires, L. E., M. C. Whiting, J. D. Brotherson, and S. R. RusHFORTH. 1979. Competitive dis- placement as a factor influencing phytoplankton distribution in Utah Lake, Utah. Great Basin Nat. 39(3):245-252. U.S. Environmental Protection Agency. 1977. Report on Utah Lake, Utah County, Utah, EPA Region VIII. National Eutrophication Working Paper 861. 19 + 65 pp., appendices. Whiting, M. C, J. D. Brotherson. and S. R. RusHFORTH. 1978. Environmental interaction in summer algal communities of Utah Lake. Great Basin Nat. 38(1):31-41. HOST-PARASITE STUDIES OF TRICHOPHRYA INFESTING CUTTHROAT TROUT {SALMO CLARKI) AND LONGNOSE SUCKERS {CATOSTOMUS CATOSTOMUS) FROM YELLOWSTONE LAKE, WYOMING R. A. Heckmann' and T. Clarroll- Abstract.— Trichophnja sp. (Protozoa) on the gills of cutthroat trout (Salmo clarki) and longnose suckers (Catostoinus catostoiiius) was studied using light and electron microscopy and tracer techniques. All cutthroat trout, 14 cm in total length and above, from Yellowstone Lake, Yellowstone National Park, Wyoming, were infested with the suctorian. No trichophryans were foimd on fry or fingerling cutthroat trout. Sixty percent of the examined longnose suckers fioni the same location were infested. Light microscopy disclosed extensive pathology of gill epithelium in longnose suckers infested with Trichophnja that was not observed for infested cutthroat trout. Electron micrographs show damage to immediate host gill cells by both parasites, depicted by a reduction and lack of mitochondria. Both parasites form attachment helices (0.52 X 0.04 fxm), which may originate in the protozoan cell membrane and fimction for maintenance of parasite position on the host cell. There was no uptake of '^C, injected into host fish, via the attachment helices by the parasite that further substantiated the mechanical fimction for the spiral structiue. Protozoan feeding on host tissue may be accomplished by use of necrotic gill tissue and mucus. Trichophnja clarki (Heckmann 1970, 1971) was found on the gills of all adult cutthroat trout {Sabno clarki) examined from Yellow- stone Lake, Yellowstone National Park, Wyoming, during the summers of 1968 and 1969. Trichophyra catostomi (Heckmann 1970, 1971) was present on the gills of 60% of the adult longnose suckers (Catostomus catos- toinus) examined from the same region. Butschli (1889) reported Trichophnja in perch (Perca) and pike (Esox) from Europe and assigned the species name T. pisciiim. Davis (1937, 1942) was the first to report Tri- chophnja in the Northern Hemisphere. He assigned tlie name T. micropteri and T. icta- luri for the gill parasites of smallmouth black bass {Micropterus dolomieui) and channel cat- fish {Ictahirus punctatus), respectively. No name was given for Trichophnja in brook trout (Salvelinus fontinalis). He also was the first to suggest that it may have a pathogenic effect. Chen Chih-leu (1955) and Prost (1952) added to Chinese and European records by assigning T. sinensis to infested white and black Amur fishes and T. intennedia to in- fested salmon-fry {Salmo salar). Lorn (1960) added to the host record for T. intermedia by including brown trout {Salmo trutta) and three other fishes in Czechoslovakia. Culbert- son and Hull (1962) summarized all host rec- ords of Trichophrya and suggested T. piscium be used for all species found in fishes. This suggestion was followed by Sandeman and Pippy (1967), who reported on four salmon- ids of Newfoundland infested with Tri- chophrya. Hoffman (1967) stressed the need for further taxonomic study of trichophryan species and their symbiotic effects. Heck- mann (1970), used the same criteria outlined by Culbertson and Hull (1962) and transmis- sion electron microscopy of the protozoan, described two new species, one in cutthroat v# Fig. L A gill macerate from an infested cutthroat trout from Yellowstone Lake showing numerous sucto- rian ciliates, Trichophn/a clarki, next to gill filaments. (lOOOX) Department of Zoology, Brigham Young University, Provo, Utah 84602. ■Department of Botany and Microbiology, Montana State University, Bozeman, Montana 59715. 255 256 Great Basin Naturalist Vol. 45, No. 2 Fig. 2. A gill section from a cutthroat trout showing a Trichophryan attached closely to the epithelial cells at the tip of the filament (arrow). (400X) trout (T. clarki) and one in longnose suckers (T. catostomi). The objective of this study was to examine the host-parasite relationship of Trichophrija on cutthroat trout and longnose suckers. Three methods were used for study of this problem: light microscopy, electron micros- copy (transmission TEM and scanning SEM), and radioactive tracers. To date there has been no ultrastructural description of the host-parasite relationship or the interface be- tween the ectoparasite and host cells. Meyer (1966) questioned the parasitic nature of T. ictaliiri and stated the main effect may be mechanical interference with respiration. Davis (1967) reported heavy loss among fin- gerling and adult smallmouth bass, raised in hatcheries, due to T. micropteri. These were attached to the gills by a broad base, closely applied to the epithelium, causing hyper- plasia and necrosis of host tissue. Materials and Methods Light Microscopy Fishes ranging in size from 3.5 to 45.7 cm total length, were obtained from several sites in Yellowstone Lake and Yellowstone River. Intact gills, infested with Trichophrija, were scraped and the macerate was examined (Fig. 1). Infested gills were also fixed with 10% for- malin and prepared by standard methods for histological examination (Davenport 1960). Sections were stained with the following: Harris' hematoxylin and eosin, periodic acid Fig. .3. Another section of cutthroat trout gill tissue showing numerous suctorian ciliates attached to the tip of gill filaments (arrow). (lOOX). Schiff (McManus 1956), mercuric brom- phenol blue (Mazia et al. 1953), five-dye stain (Greenstein 1961), and Schiff's reagent (Dav- enport 1960). Electron Microscopy Infested gill macerate and gill filaments were placed in small plastic vials containing 2.5% gluteraldehyde buffered with potassium phosphate (O.IM, pH 7.3). Post fixation was accomplished with 1% osmium tetroxide in the same buffer. Standard methods were used in preparing the tissue for sectioning (Dawes 1971). The dehydrated material was em- bedded in Araldite epoxy plastic and sec- tioned, then poststained with uranyl acetate and Reynold's lead citrate. In an attempt to determine possible differences in the cy- tochemical nature of magnified structures, the staining procedure was varied in the fol- lowing manner: no poststain, uranyl acetate only, and lead citrate only. Fixed gill macer- ate and filaments were sent to Florida State University for examination by scanning elec- tron microscopy. Tracer Study A tracer experiment was conducted with four infested cutthroat trout of approx- imately equal size from Yellowstone River. Each fish was anesthetized with MS 222 and injected intracardially with 5 microcuries of "C-D-Glucose (U). Previous to the injection blood samples were qualitatively checked for April 1985 Heckmann, Carroll: Fish Parasites 257 Fig. 4. This is a section of cutthroat trout gill tissue stained with periodic acid Schiff to emphasize mucopolysaccharides. Note the dark staining granules in the host (H) epithelial cells and similar granules in the cytoplasm of the suctorian ciliate (P). The protozoan could be using mucus as a source of food (40()X) glucose, using chromatography. The fish were sacrificed 1, 2, 4, and 8 hours after in- jection. The gills were removed and washed in physiological saline and fixed in 10% for- malin. Samples of 25 Trichophnja and 0.4 grams of gill filament were analyzed from each fish using liquid scintillation counting (Arnoff 1960, Chase and Rabinowitz, 1962). The suctorians and gill filaments were placed in liquid scintillation vials containing 0.5 and 1.0 ml, respectively, of hydroxide of Hya- mine (Rohm and Haas) for 12 hours to dis- rupt the cell membranes. Scintillation fluid was added and each sample was counted for a period of 20 minutes. The trichophryans were fixed with 10% formalin and washed three times with changes of formalin. The wet film method for autoradiography (Pelc 1947, MacDonald et al. 1948) was used to corroborate the data from liquid scintilla- tion counting. Gills from the injected fish were prepared histologically (Davenport 1960) and sectioned at 20 jum. Duplicate thin sections from each fish were spread on glass slides that were immersed in water along with imexposed film (Fuji plate film; ET2F- 9327). A strip of film was then removed from the plate and placed over the tissue sections. The tissue-film preparation was subsequently removed from the water, air dried, and stored in Ught-tight film boxes. The prepara- tion was developed and stained with Mayer's hematoxylin after 2, 4, 6, and 8 weeks' expo- sure (Shigematsu 1969a, 1969b). It was then observed with a compound microscope. Results Light Microscopy Histology.— The suctorian parasites on cutthroat trout are usually concentrated on the lamellar tips of the gill filament where they are closely attached to the epithelial cells (Figs. 2 and 3). Cutthroat trout samples, with the largest number of trichophryans, had 7.1% of the gill surface covered by the parasite. In one 36 cm trout, there was an av- erage of 31 suctorians per gill filament total- ing about 42,000 organisms. The macro- and micronucleus of T. clarki were Feulgen posi- tive (Schiff's reagent) and the mercuric bromphenol stain showed an intense blue area between the parasite and the epithelial host cell. Periodic acid Schiff staining re- vealed similar particles in both protozoans and the adjacent surface of the gill epithe- hum (Fig. 4). HisTOPATHOLOGY.— Scctious of gills from cutthroat trout infested with T. clarki had no apparent cytological damage (Figs. 2 and 3), whereas longnose suckers inhabited by T. ca- tostomi were definitely affected by the para- site (Figs. 5 and 6). There was definite dam- age to the gill lamellae, characterized by hyperplasia and hemorrhaging of the adja- 258 Great Basin Naturalist Vol. 45, No. 2 $^mf4F' 1 t.''5i"^SSi '.. S <5 Fig. 5. This is a section of longnose sucker gill tissue. The suctorian ciliates are causing pathological damage to the host (arrow and box) tissue that includes hemorr- haging and hyperplasia of the gill lamellae. Necrosis (N) also occurs in the infested tissue with subsequent club- bing of filaments. (4()0X) cent host tissue with subsequent necrosis (Fig. 5,6). Transmission Electron Microscopy The ultrastnictural characteristics of T. clarki and the gill epithelial cells have been described (Heckmann 1970). Sections of the interface between the host epithelial cell and parasite were prepared. A helical structure in the interface attaches the parasite to the gill epithelium (Fig. 7). This structure (referred to as attachment helix) has the following mea.surements: length, 0.52 /xm (range 0.20 to 0.82 jum) and width 0.04 [xm (range 0.03 to 0.06 ^m). The filament making up die helix appeared circular in transverse section (Fig. 8) and revealed an electron transparent center boimd by an opaque ring. Lorn, 1970, described a similar structure for suctorian infested fish concurrent with this initial description (Heckmann 1970). The at- Wk \v^^ Fig. 6. In infested longnose sucker gills, the Trichoph- ryans are not attached to the tips of the gill filaments but locate at the base (arrow) of the structure. (600.\) tachment helix is found only on the side of the protozoan next to the host cell (Fig. 7). It appears to originate as a cleft in the outer wall of the protozoan. The protozoan mem- brane, in the cleft, moves into the space be- tween the host cell and parasite and expands into a long filament (Figs. 9 and 10). The fila- ment then contracts to form the helix. Cy- tochemical evidence from ultrathin sections along with histochemical analyses of .speci- mens suggest that the origin of the helix is the outer wall of the protozoan. The helix was osmophilic when unstained thin sections of oxmiiun-fixed protozoa were viewed. It was mercuric bromphenol blue positive. Lipo-protein material, found in biological membranes, have an affinitv for osmium whereas protein stains blue with bromphenol blue. A .series of electron micrographs shows or- ganelle changes to the epithelial cells of cut- throat trout due to T. clarki. In gill epithelial cells not infested with 7'. clarki, there are nu- merous mitochondria with well-defined April 1985 Heckmann, Carroll: Fish Parasites 259 (t-* ♦^ «: • "' .. ♦ v-*^"' ^ fmi-^^ H 0 P 1 .^J. •* Fig. 7. At the electron microscopic level of magnification, attachment helices (arrows) are visible between fish host cells (H) and the protozoan (P). These structures aid in holding the suctorian ciliate next to the host epithelial cell. (12,000X) cristae (Fig. 11). Infested cells show, in com- parison to normal cells, swollen mitochondria that have fewer cristae and lack the outer en- veloping membrane (Fig. 12). Thus damage is detected as the number of host cell mito-^ chondria decrease and disappear. Similar T. catostomi in damage was observed for longnose suckers. Scanning Electron Microscopy Scanning electron microscopy shows T. clarki to be saucer-shaped, with the convex 260 Great Basin Naturalist Vol. 45, No. 2 f'^- < ' .> ^^ , i^-^' '-V ;1^. ^ 4#^,, ^^■^> * '"*'*-J. T^ -■' ^^**?'; Fig. 8. This electron photomicrograph shows the interface between the fish epithelial tell and the protozoan. Note the spiral, hollow nature of the attachment helices (arrows). (12.5,(X)0.\) surface attached to the gill epithelium. There are fine filaments between the suctorian and the host cells that are probably aggregations of attachment helices (Fig. 13). Tracer Study Liquid scintillation counts and autoradiog- raphy indicate there is no uptake of 'C by T. clarki from the fish host during eight hours following injection. The isotope was still pre.sent in the gill epithelium after two hours and was presumably available to parasites us- ing sustenance directly from the host. Silver grains in the autoradiographic film devel- oped in all four gill samples after four weeks exposure. Samples from fi.sh sacrificed at one- and two-hour intervals after ' right- hand side. described. Uspenskaja (1966) found small for attachment. Scholtyseck and Hammond cytoplasmic extensions ("rootlets") from Mijx- (1966a, 1966b) noted ribbonlike extensions idiuui into the urinary bladder epithelium of (15 mm by 2 jum) from EUncria macrogmeto- Esox lucim. He considered these to function (iytes into host cells and postulated that their primarily as absorption organelles rather than function was ingestion of nutrients. April 1985 Heckmann, Carroll: Fish Parasites 263 Fig. 11. This is a fish gill epithelial cell containing characteristic mitochondria (ni). (10,500x) Acknowledgments U.S. Sport Fisheries and Wildlife. Financial Dr. C. J. D. Brown directed the study; Dr. ^"PP^^t was given by Montana Cooperative Gary Strobel and Dr. Akiyo Shigematsu of- Fisheries Unit, Montana State University Ag- fj^,., .^ ^ r- I.- ncultural Lxpenment Station, Froiect 410, tered technical assistance. Cooperation was „ i c-r t i c • If ,1 XT .• 1 Ti 1 o J raper 157, ournal Series, received from the National rark Service and ^ ^ ^*\>.'^-:r .'.«'*■-:• H / fA l^.; m ' # Fig. 12. When the Trichophryan attaches to the gill filament, the immediate epithelial cells show mitochondrial (m) and organelle degeneration, probably due to the masking effect of the protozoan. (10,500X) 264 Great Basin Naturalist Vol. 45, No. 2 4 Fig. 13. Tins is a scanning electron niicioscopy of a cutthroat trout gill. The Trichophryan is attached to the gill lamellae (box) and appears to be saucer-shaped. (lOOOX) Literature Cited Arnoff, S. 1960. Techniques of radiobiochemistry. Iowa State Univ. Press, .\nies. 228 pp. Bardele, C. F., and K. G. Grell. 1967. Electron- enmikroskopische Beobachtungen sur Nahrung- sauhiahme bei dem Suktor A.c\neia tuberose (Eh- renberg). Z. hu" Zellforsch. 80:108-123. B.\TissE, A. M. 1967a. De developpement des phialo- cvstes chez les acinetiens. C. R. .\cad. Sci. 265(D):972-974. 1967b. Donnees nouvelles dur la structure et le fonctionnenient des centouses tentaculaires des acinetiens. C. R. Acad. Sci. 265(D); 1056-1058. BuTSCHLi, O. 1889. Protozoa. Pages 1842-1945 in Bronn's Klassen und Ordnungen des Thierreichs. C. F. Wintersche. Verlag, Leipzig. Vol. I, Pt. 3. Ghase, G. D., and J. L. Rabinowitz. 1962. Principles of radioisotope methodology. Burgess Publishing Go., Minneapolis, Minnesota. 372 pp. GnEN, G. L. 1955. Tlie protozoan parasites from four species of Ghinese pond fishes. .\cta Hvdrobiol. Sinica. 2:123-164. GuLBERTSON, J. R., AND R. W. HiLL. 1962. Species iden- tification in Tricliophnjd (Suctoria) and the oc- currence of melanin in some members of the genus. ]. Protozool. 9(4):455-459. Davenport, H. A. 1960. Histological and histochemical techniques. VV. B. Saunders Go., Philadelphia, Pennsylvania. 401 pp. Davis, H. s' 19.37. A gill disease of the smailuioutli blackbass. Prog. Fish-Gult. 27:7-1 1. .. 1942. A suctorial parasite of the smallmouth blackbass with remarks on other suctorial para- sites of fishes. Trans. Aiiitr. Microscop. Soc. 61:.30t)-.327. 1967. Gulture and diseases of game fishes. Ihiiv. of (California Press, Berkeley. 332 pp. Dawks, (.1. J. Biological technifjues in electron microsco- py. Barnes and Noble, New York. 193 pp. Gree.nstein, J. S. 1961. .\ simplified five-dye stain for sections and smears. Stain Technol. 36(2):87-88. Heckman.n, R. .\. 1970. Gomparative morphology and host-parasite studies of Trichophrya clarki (sp. n) on cutthroat trout (So/mo clarki). Unpublished dissertation. Montana State Univ. 69 pp. 1971. Parasites of cutthroat trout from Yellow- stone Lake, Wyoming. Prog. Fish-Gult. 33:103-107. Hoffman, G. L. 1967. Parasites of North .American freshwater fishes. Univ. of Galifornia Press. Berkeley. 486 pp. Hill, R. W. 1961a. Studies on suctorian protozoa: the mechanism of prev adherence. J. Protozool. 8(4): 343-360. 1961b. Studies on .suctorial protozoa: the mecha- nisms of ingestion of prev cvtoplasm. J. Pro- tozool. 8(4):,351-.3.59. LoM, J. 1960. Protozoan parasites found in Gzechoslova- kia fishes. I. Myxosporidia, Suctoria. Zool. Li.sty Folia Zool. 10(24):45-59. 1970. Attachment structures in ectoparasitic pro- tozoans of fishes and their possible relation to pathogenicity. Abstracts: Second Int. Gongress of Parasitic. No. 386. MacDonald, a. M., J. GoBB, AND A. K. Solomon. 1948. Radioautograph techniques with ''*G. Science 107-550-552. Mazia, D., p. a. Brewer, and M. .\lfert. 1953. The cy- tochemical staining and measurement of protein with mercuric bromphenol blue. Biol. Bull. 104:57-67. McManus, J. F. A. 1946. Histological demonstrations of mucin with periodic acid. Nature 158:202. Meyer, F. P. 1966. Parasites of catfishes. U.S. Dept. of Interior, Fi.sh and Wildlife Service FDL-5:l-7. Pelc, S. R. 1947. .\utoradiograph technique. .Nature 169:749. Frost, M. 1952. Investigations on parasitic protozoa on the gills of fishes. I. Trichophrija intermedia sp. n. on the gills of salmon frv. .\nn. Univ. Mariae Gurie-Sklodowski. 6( 12):376-.386. RuDZLNSKA, M. A. 1965. The fine structine and huiction of the tentacle in Tokophrija inf\isionitnu ]. Gell Biol. 25:459-477. RuDZLNSKA, M. A., G. J. Jackson, and M. Tiffrai. 1966. The fine structure of Colopoda maupasi with special emphasis on food vacuoles. J. Pro- tozool. 13(3):440-459. RiDZLNSKA, M. a., and K. R. Porter. 1954. Electron mi- croscope study of intact tentacles and disc of To- koplin/a infiisioniim. Experienlia 10(ll):460-468. Sandenl\n, I. M., AND J. H. G. PipPY. 1967. Parasites of freshwater fishes (Salmonidae and Goregonidae) of insular Newfoundland. J. Fish. Rs. Bd. Ganada 24(9): 191 1-1943. Sc:holtyseck, E., .\nd D. M. Hammond. 1966. Spezi- fische Feinstruckteren bei Parasit wirt aks .A-us- druck ihrer Wechselwirkungen am Beispeil von Gocoidicn. Z. f. Parasitenkunde 28:78-94. S( iioltyse( K. E., AND J. V. Ernst. 1966. Fine structure of till' luacrogametes of Eiineria perforans. E. sleidac. /■-'. hovis and £. auhurnensis. J. Parasit. 52(5):975-987. Shic;em.\tsu, A. 1969a. Lecture of autoiailio'^iapliv in bi- ology (IV) MicroautoradiogiapliN-autoradi- ographv techniques for biochemical research field. Radioisotopes 18(5): 160-170. April 1985 Heckmann, Carroll: Fish Parasites 265 1969b. Lectures of autoradioi^raphv in hiolo-y (V) Purriuia striiformis. Phytophathology Microautoradiography-autoradiograpliv tech- 55(1): 1219-1222 , ,■ , -,. „f niques for the biochemical research field. Radu,- Uspenskaja, A. V. 1966. On the mode of nu r.t.on of isotopes 18(5): 194-203. vegetative stages of Myxidiun, Uchcrkuhm Strobel, G a. 1965. Biochemical and cytological pro- (Butschli). Acta Protozool. 4(10):81-S8. cesses associ ated with hydration of luedospores of NEW SYNONYMY AND NEW SPECIES OF BARK BEETLES (COLEOPTERA: SCOLYTIDAE) Stephen L. Wood' Abstract.— New synonymy in Scolytidae is proposed as follows: Camptoccrus opacicoHis (Eggers) ( = Camptocenis aquilus Wood), Cladoctonus corumbensis (Eggers) (= HopUtophthorus hoUiiantis Wood), Cladoctomis interniptiis (Eggers) ( = HopUtophthorus sentiis Wood), Cnenionijx errans (Blandford) ( = Ceratolepis harhatus Schedl), Cnemomjx flavicomis (Chapnis) (= Cnemonyx vianai Schedl), Cnesinus clividmts Schedl (= Cnesiniis dnjogmphus Schedl, Cnesinua laevicoUis Schedl), Cnjptocurus spinipennis Schedl (= Hijloperus caudatus Browne, Hiflopcnis hirornis Browne), Dendrosiniis ater Eggers (= Dendwsinus hirstitiis Schedl), Hylesimis acuJeatus Say ( = Ihjlcsinus imperkdis Eichhoff), Hylesintis rordipennis Lea (= Uijlesinus papiianiis Eggers), Hylesinus Duicmalioni (Stebbing) (= Hylesintis altemans Schedl, Lcperisiniis fraxinoidcs Beeson, Lcpcrisiniis fnixinoidcs Schedl), Hylesinus nilioriniis Eggers (= Trogloditica whiista Schedl), Phloeosinopsoides triseriatus (Schedl) (= Xyleeliinus papuanus Schedl), Phloeothbus scarabaeoides (Bernard) ( = Phlocothbus aiuericunus Dejean), Scolytogenes dancini Eichhoff ( = \igritus simdis Eggers, Nigritus major Eggers, Seolytogenes cryptolepis Schedl), Scolytodes notatus Eggers ( = Hexacohis pseiidobicolor Eggers, Hexaeohis subparaUeus Eggers, Hexacohis pelicerinus Schedl), Scolytopsis punctirollis Blandford ( = Seolytopsis argentinensis Schedl, Seolytus bruchi Schedl, Scolytopsis toba Wichniann), Toinicus piniperdd (Linnaens) (= Blastophagus kluisianus Murayama), Xyleeliinus spathifer Schedl (= Pteleobius kmiatiae Schedl). Species new to science are described as Acanthotomicus ipsimorphus (Costa Rica), Acrantus opimus (Indonesian New Guinea), Bothrosternus hirsutus (Venezuela), Cnesinus diseretus (Venezuela) Cnesinus minor (Costa Rica), Corthylus trunacatus (Peni), Hyhirgus indicus (India), Piiehyeotes minor (Australia), Phloeosinopsoides piimdus (Papua New Guinea), Xylecliinosomus pdosiis (Brazil). During the past several years, a world revi- sion of the genera of Scolytidae has been in preparation. While conducting that study, I have had the opportunity to visit several mu- seums for the purpose of studying type mate- rial. This led to the discovery of a number of synonyms and to the detection of several spe- cies new to science. The above abstract sum- marizes 29 cases of .synonymy from all parts of the world and lists the names and country of origin for 10 previously unnamed species. Tlie new species represent 9 different genera and come from Australia (1), Brazil (1), Costa Rica (2), India (1), New Guinea (2), Peru (1), and Venezuela (2). Of .special interst to American students is conclusive placement in synonymy of Hyle- sinus imperialis Eichhoff and of the nomen nudum, Phloeotribiis americanus Dejean. New Synonymy Camptocerus opacicollis (Eggers) Loginiiiis opacieollis Eggers, 1929, Wiener Ent. Zcit. 46:61 (Holotype, male; Ostbolivia; Eggers Coll.. apparently on loan to Wien Nat. M\is.) Campfoeerus (npnlus Wood. 1972, Bull. Ent. Res. 62:244 (Holotype, male; 12° 49' S 51° W, Brazil; British Mus. Nat. Hist.). \eic synonyiny The male holotype of Loganius opacicollis Eggers was deposited in the Eggers Collec- tion, but it never reached the U.S. National Museum with the Eggers Collection. It was found in Schedl material at the Wien Mu- seum and was compared to a male paratype of Camptocerus acjuilus Wood. Only one spe- cies is represented by this material. For this reason, the junior name is placed in synony- my as indicated above. Cladoctonus corumbensis (Eggers) Hoplitcs eorumbensis Eggers, 1950. Ent. Blatt. 45-46:149. (Holotype: Corumba, Matto Grosso, Brazil; Eggers Collection, apparentlv on loan to Wien Nat Mus.) HopUtophthorus boliviunus \\ Ood, 1961. (ircat Basin Nat. 21:106 (Holotype, female; Route between Boyuilbe and Charagua via Cue\;i. Ingri. etc. Bo- livia; U.S. Nat. Mus.). New synonymy Because the holotype of Hoplites corum- bensis Eggers was missing from the Eggers Collection at the U.S. National Museum, it could not be compared to species sub- 'Lifc Science Museum auil Departiiu' .i;y, BriKhani Youn.i; I', 266 April 1985 Wood: New Bark Beetles 267 sequently named in this genus. When it was found in Schedl material at Wien, it was compared to paratypes of HopIitopJitJwrtis boliviae Wood and found to be identical. For this reason, the name bolivianns must be placed in synonymy as indicated above. Cladoctoniis intcnuptus (Eggers) HopUics intcnuptus Eggers, 1940. Arh. iiiorpli. taxoii. Ent. 7:126 (Holotvpc; Guadeloupe; Fleutiaux Coll.) HopIitophtJiorus sentus Wood, IWil, CJreat Basin Nat. 21:3 (Holotype, female; La Cuchilla, Sevilla, Co- lombia; Wood Coll.). Netc synonymy A male cotype of Hoplites interruptiis Egg- ers was found among the Schedl material at Wien and was compared directly to a male paratype of Hoplitophthonis sentus Wood. The two specimens represent the same spe- cies. For this reason the name sentus is placed in synonymy as indicated above. Cnemonijx errans (Blandford) Ccidtolcpis cmnis Blandford, 1896, Biol. Centr. Amer., Coleopt. 4(6): 127 (Lecotype, male; in ■'Mexican" tobacco refuse; British Mus. Nat. Hist., desig- nated by Wood, 1972, Great Basin Nat. 32:19) Cemtolepis barhatus Schedl. 1954, Dusemia 5:24 (Holo- type, male; Nova Teutonia, Brazil; Wien Nat. Mus.). New synony))iy The holotype of Cerotolepis barhatus Schedl was examined and compared directly to my homotypes C. errans Blandford. Be- cause only one species is represented by this material, Schedl's name is placed in synony- my as indicated above. Cnemonyx flavicornis (Chapuis) Lo'^anius flinicomis Chapuis, 1869, Synopsis des Scoly- tides, p. 53 (Two svntvpes; Cumana; Brussels Mus.) Cnemonyx Lianai Schedl, 1950, Acta Zool. Lilloana 9:289 (Holotype; Valle Hermoso, Dep. Punilla, Cordoba, .Argentina; Wien Nat. Mus.). .Vt'ir synonymy The type and two paratypes of Cnemonijx vianai Schedl in the Schedl material at Wien were placed by Schedl as a synonym of Loga- nius flavicornis Chapuis, but this was appar- ently never published. Because I have exam- ined both syn types of flavicornis and the Schedl holotype, I concur with his decision and place Schedl's name in synonymy as in- dicated above. Cnesinus dividuus Schedl Cncsinus dividuus Schedl, 1938, Rev. Soc. Ent. .Argen- tina 10:22 (Lectolype, female; Tigre, Buenos Aires, Argentina; Wien Nat. .Mus., present designation) Cnesinus dryogmphus Schedl, 1951, Dusenia 2:78 (Lec- totype, female; Nova Teutonia, Brasil; Wien Nat. Mus., present designation). New synonymy Cnesinus laevieollis Schedl, 1951, Dusenia 2:79 (Lecto- type, female; Nova Teutonia, Santa C^atarina, Brasil; Wien Nat. Mus., present designation). New synonymy The "hoiotype-s" cited by Schedl (1979) for his species Cnesinus dividuus, C. dryogra- phus, and laevieollis are all of the same sex and all are syntypes. As indicated above, I here designate those "holotypes" as lecto- types of dividuus, dnjographus, and laevi- eollis respectively. They were compared di- rectly to one another and to my series from Nova Teutonia and were found to represent the same species. The two junior names must, therefore, be placed in synonymy as in- dicated above. It was a common practice of Schedl to designate male and female "holo- types," one of each for his collection and one of each for the collection of institutions sub- mitting the specimens for identification. Hence, the confusion of "holotypes.". Cnjptocurus spinipennis Schedl Ciyptoeuius spinipennis Schedl, 1957, Ann. Mag. Nat. Hist. (12) 10:870 (Holotype, male; Moshi district, Tanganyika; British Mus. Nat. Hist.) Hyloperus caudatus Browne, 1970, J. Nat. Hist. 4:547 (Holotype, male; Gyel Nyaki, Mambilla Plateau, Nigeria; British Mus. Nat. Hist.). New synonymy Hyloperus bieornis Browne, 1970, J. Nat. Hist. 4:546 (Holotype, female; Gyel Nyaki, Mambilla Plateau, Nigeria, British Mus. Nat. Hist.). New synonymy The male holotypes of Cryptocurus spin- ipennis Schedl and Hyloperus caudatus Browne were compared directly to one an- other and obviously are the same species. The female paratypes of spinipennis in the Schedl collection were compared directly to female paratypes of H. bieornis Browne and my paratypes to the holotype of bieornis. The females are identical. Schedl's series was taken from the host, apparently from the same tunnels, and appears to indicate an ac- curate association of the sexes. If this is cor- rect, both of Browne's species are junior syn- onyms of Schedl's name and are placed in synonymy as indicated above. 268 Great Basin Naturalist Vol. 45, No. 2 Dendrosinus ater Eggers Dcndiusiniis ater Eggers, 1930, Ent. Blatt. 26:167 (Holo- type, male; Ostbolivia; U.S. Nat. Miis. Dendrosinus hirsutus Schedl, 1958, Acta Zool. Lilloana 16:38 (Lectotype, female; Santa Fe, Dap. Garay Maclas, Argentina; Wien Nat. Miis., present des- ignation). Xetf siinonymy The description of Dendrosinus hirsutus Schedl is composite. Because of this the "holotype" cited by Schedl (1979:117) is here designated as the lectotype of this species. This lectotype is a specimen of D. ater Egg- ers in which the elytral setae are not fully colored and appear pale. For this reason, Schedl's name must be placed in synonymy as indicated above. Hylesinus aculeatus Say Hylcsinm aculeatus Say, 1824, J. Acad. Nat. Sci. Phila- delphia 3:322 (Syntypes; Missouri; apparently lost). Hylesinus imperiahs Eichhoff, 1868, Berliner Ent. Zeitschr. 12:149 (Syntypes; Wisconsin, Georgia; 1 male labeled Anier. Bor., Ulke Coll. is probably a cotype, labeled "holotype" by Schedl; Wien Nat. Mus.). New synonymy The male of Hylesinus imperiahs Eichhoff in the Schedl collection and labeled by him as the holotype appears to be an Eichhoff specimen obtained by Schedl during World War II from the Stettin Museum. Since other Stettin Museum specimens of Eichhoff now in the Schedl collection appear to be authen- tic, I see no rea.son to doubt the authenticity of this specimen. However, it should be cited as a lectotype, not as a holotype. This .speci- men now makes it possible to remove all doubt from its placement as a synonym of aculeatus as indicated above. Hijlesinus cordipennis Lea Hylesinus cordipennis Lea, 1910, Proc. Rov. Soc. Vic- toria, n.s., 22:144 (Syntypes; Cairns, Queensland, Australia; one syntype Wien Nat. Mus.) Hylesinus papuanus Eggers, 1923, Zool. Meded. 7:133 (Lectotype, male; Insel Yule bei Neu Guinea; Wien Nat. Mus.). New synonyiuy A female syntype of Hylesinus cordipennis Lea in the Wien Mu.seum and the male lecto- type of H. papuanus Eggers were compared directly to my series from Bulolo, New Guin- ea. These specimens all represent the same species. If the Schedl syntype actually does represent Lea's species, then the name papuanus must be placed in synonymy as in- dicated above. Hylesinus macmahoni (Stebbing) Sphacrotrypes macmahoni Stebbing, 1909, Indian For. Mem., Zool. Ser. 1(2): 16 (Two syntvpes; Sangar Scallon, near Takt-i-Suliman .Mountain, 7.000 ft.. Baluchistan, Pakistan; Forest Research Institute. Dehra Dim) Hylesinus alternans Schedl, 1959, Indian For. Rec, n.s.. Entomology 9(8): 172 (Holotype, male; Rawal- pindi, Punjab, India; Wien Nat. Miis.). Neic synonymy Leperisinus fraxinoides Beeson, 1941, Ecology and Con- trol of the Forest Insects of India, p. 287. Nomen nudum Leperisinus fraxinoides Schedl, 19.59. Indian For. Rec, Entomology 10(2):.39 (Paratype, female; Lolab, Putshai. Kashmir, India; Wien Nat. .Mus.). New synonymt/ Two cotypes of Sphaerotrypes macmahoni Stebbing, the male holotype of Hylesinus al- terans Schedl, a female paratype of Leperi- sinus fraxinoides Schedl, and 25 other speci- mens from the Indian states of Jammu, Kashmir, and Punjab were examined. It is quite clear that only one species is represent- ed, although the sexual dimorphism is more conspicuous than in most members of this genus. None of the Indian specimens report- ed by Schedl to have been returned to the Forest Research Institute ever reached their destination. In fact, the available evidence in- dicates that they were never mailed and still reside in the Schedl Collection at Wien. Con- .sequently, the "missing" holotype of fraxi- noides probably never existed and is repre- sented only by the paratype cited above. Whatever that situation might be, it is clear that both of Schedl's names must be placed in synonymy as indicated above. Hyh ndii aggers Hylesinus uilianhena). \ciu .^iinoni/niy The female holotype of Phloeosinopsis triseriatus Schedl was compared directly to the male paratype of Xylechinus papuanus Schedl in the Schedl material at Wien and to a pair of this species from Bulolo, New Guin- ea, in my collection. All represent the same species. Phloeotribiis scarabaeoides (Bernard) Scohjtus scarabaeoides Bernard, 1788, Mem. Hist. Nat. Provence 6:271 (Syntypes?; France; lost. Neo- type, female; Galliae meridionalis in O/eo; Kiel part of Copenhagen Fabricius Coll., designated by Wood, 197.5, Bull. Zool Nomencl. 32:122) Phloeotribiis americanus Dejean, 1837, Cat. Coleopt., ed. 3, p. .331 (Anier. bor; nomen nudimi) Although the name Phloeotribiis american- us Dejean, nomen nudum, has been cited in the literature on several occasions, efforts to locate the specimen on which Dejean's name was based were fruitless until I found it in the Wien Museum (original collection, not part of the Schedl material). It is labeled "Amer. bor., Phloeotribiis americanus Dej." It is an incorrectly labeled specimen of P. scara- baeoides apparently from southern Europe and is not part of the American fauna. Scolijtogenes darwini Eichhoff Scohjtogencs darwini Eichhoff, 1878, Mem. Soc. Roy. Sci. Liege (2) 8:497 Stettiner Ent. Zeit. ,39:387 (Holotype; Hindostan, "Birma" on type; Wien Nat. Mas.) Nigritiis similis Eggers, 1923, Zool. Meded. 7:142 (Lecto- type; Java; U.S. Nat. Mas., designated by Ander- son and Anderson, 1971, Smithsonian Contrib. Zool. 94:30). New synonymy Nigritus major Eggers, 1927, Philippine J. Sci. 33:69 (Lectotype; Surigao, Mindanao, Philippines; U.S. Nat. Mus.) New synonymy Scohjtogcnes rryplolcpis Schedl, 1951, Tijdschr, Ent. 93:5.5 (Holotype; Nakronda, Dehra Dun, Uttar Pradesh, India; Wien Nat. Mus.). New synonymy The holotypes of Scolytogenes darwini Eichhoff and S. cryptolepis Schedl and co- types of Nigritus sitnilus Eggers were com- pared directly to one another and to my specimens. My .specimens were also com- pared directly to the lectotypes of N. major Eggers and N. similis. All represent the same, common, widely distributed species. It occurs in vines from India and Sri Lanka to the Phil- ippines and northern Australia. Scohjtodes notatiis (Eggers) Hcxacohis notntus Eggers, 1940, Arb. Morph. Taxon. Ent. Berlin-Dahlem 7:133 (Holotype, male; Trois Rivieres, Guadeloupe; "Eggers Coll.") tiexacohts psciidobicolor Eggers, 1940, Arb. Morph. Tax- on. Ent. Berlin-Dahlem 7:132 (Holotype, male; Trois Rivieres, Guadeloupe; U.S. Nat. Mus.). New synonymy Hexacohis subparallehts Eggers, 1940, Arb. Morph. Tax- on. Ent. Berlin-Dahlem 7:1.34 (Holotype, Trois Rivieres, guadeloupe; Fleutiaux Collection). New synonymy Hexacohis pelicerinus Schedl, 1952, Dusenia 3:3.58 (Holotype, male; Mexico?; Wien Nat. Mus.). New synonymy Paratypes (or cotypes) of Hexacolus no- tatus Eggers, H. pseudobicolor Eggers, and H. subparallelus Eggers, and the holotype of H. pelicerinus Schedl in the Schedl Collection (Wien Nat. Mus.) were examined and com- pared directly to one another All represent the same common Carribean species. Scohjtopsis puncticollis Blandford Scohjtopsis puncticollis Blandford, 1896, Biol. Centr. Amer., Coleopt. 4(6):123 (Syntypes; Guatemala; Bristish Mus. Nat. Hist.) Scolyptosis argentinensis Schedl, 1937, Rev. de Ent. 7:84 (Lectotype, female; Prov. Tucuman. Argentina; Wien Nat. Mus., designated by Schedl, 1979, Kat. wiss. Samml. Nat. Mus. Wien. Ent. 2:25). New synonymy Scohjtus bruchi Schedl, 19.39, Not. Mus. La Plata 4:170 (Lectotype, male; Misiones orillas del Ignazu, Ar- gentina; Wien Nat. Mus., designated by Schedl 1979:48). .Veir synonymy Scohjtopsis toba Wichmann, 1914, Ent. Blatt. 10:1.36 (Holotype; Santa Sofia, Paraguay; Nat. Mus. Wien). .Veil; synonymy The holotype of Scolytopsis toba Wich- mann and the lectotypes of S. argentinensis Schedl and S. bruchi Schedl, and a pair of my 270 Great Basin Naturalist Vol. 45, No. 2 homotypes of S. puncticollis Blandford were ail compared directly to one another. Be- cause only one species is represented by this material, the Schedl and Wichmann names are placed in synonymy as indicated above. Tomicus piniperda (Linnaeus) Dermestes piniperda Linnaeus, 1758. Systema Natvirae, ed. 10, p. .355 (Syntypes; Europae; presumably at Uppsala). Blastophagus khasianus Murayama, 19.59, Bull. Brooklyn Ent. Soc. .54:75 (Holotype, Shillong, Assam, India; U.S. Nat. Mus.). New .s|/noni/»n/ Murayama named Blastophagus khasianus from a specimen in poor condition that came from a long series taken by C. F. C. Beeson. After examining the Murayama type, the en- tire series of Beeson at the Forest Research Institute, and long series from other parts of Asia and from Europe, it is apparent that kJiasianus represents a very minor variation that does not warrant either specific or sub- specific status. Xijlechinus spathifer Schedl Xtjleehimis spathifer Schedl, 19.55, Rev. Chil. Ent. 4;2,56 (Lectotype; Laguna de Malleco, Pemehne, Chile; Wien Nat. Mus., present designation) Pteleohiiis lomatiae Schedl, 1975, Stud. Neotrop. Fauna 10:2 (Holotype, male; Nahuel Huapi National Park, Argentina; Wien Nat. Mus.). New synonijiny The description of Xijlechinus spathifer Schedl is composite. The specimen cited by Schedl (1979:233) as the "holotype" and la- beled holotype in the Schedl collection is here designated as the lectotype of this spe- cies. This lectotype and the holotype of Pte- hhuis lornatiae Schedl were compared di- rectly to one another and were found to represent the same species. The junior name is placed in synonymy as indicated above. NEW TAXA Acantliotomicus ipsif annus, n. sp. This species is distinguished from mimicus (Schedl) by the slightly larger, stouter body form, by the sculpture of the frons as de- scribed below, by the coarser, closer elytral punctures, and by the deeper, coarser, more /ps-like elytral declivity. Male.— Length 2.2 mm (paratypes 2.2-2.3 mm), 2.8 times as long as wide; color reddish brown. Frons resembling mimicus except lower half of frons much more strongly, trans- versely impressed, upper half less strongly convex; surface more nearly rugose, with rather numerous, coarse, isolated granules. Vestiture similar but coarser. Pronotum LI times as long as wide; similar to mimicus except asperities slightly smaller, punctures on posterior half slightly smaller, not as deep. Elytra L4 times as long as wide; similar to mimicus except strial and interstrial punc- tures slightly smaller, deeper, much closer, usually somewhat confused near base particu- larly near suture. Declivity not quite as steep, more deeply impressed, as in 4-spined Ips; punctures on striae 1 and 2 mostly in rows, others confused; margin armed by four den- ticles positioned exactly as in mimicus but considerably larger, 4 pointed, and about twice as large as 3; lower margin from den- ticle 4 to suture acutely, moderately expla- nate, with crest undulating to form three in- definite cusps somewhat resembling those of some Orthotomicus. Vestiture as mimicus ex- cept slightly finer. Antennal sutures procurved as in mimicus and many other Acanthotomicus and rather similar to Ips concinnus (Mannerheim) and mexicanus (Hopkins) of North America. Type locality.— Santa Rosa National Park, Guanacaste, Costa Rica. Type material.— The male holotype and two male paratypes were taken at the type locality in 1982, by George Stevens. The host was not recorded but could have been Spondias momhin. A male and a female from Bahia, Brazil could be this species but are ex- cluded from the type series. The holotype and paratype are in my collection. This species, supported by mimicus, is re- markable in that it represents the closest structural approach to the Ips-Oiihotomicus group yet found and greatly increases the probability of a neotropical origin of this seg- ment of the Ipini. Acrantus opittius, n. sp. Recent literature treating the genus Acran- tus is chaotic. Representatives of an assem- blage of species from three or more genera April 1985 Wood: New Bark Beetles 271 have been thrown together with httle or no thought given to basic characters, and a ma- jority of the species that actually belong here currently reside in still other, unrelated gen- era. Among material at hand, the species de- scribed here appears allied to mundiilus Brown, although the relationship is not close. It is distinguished from mundidus by the larger size and by niunerous other characters cited below. Male.— Length 2.9 mm, 2.2 times as long as wide; color very dark brown, vestiture pale. Frons shallowly, broadly concave on about central third, gradually transcending to flat- tened or feebly convex on surrounding areas; surface mostly smooth, brightly shining, be- coming subreticulate toward vertex, punc- tures moderately abundant, rather small, shallow but distinct, much smaller in concave area. Vestiture absent, apparently abraded (a few short, erect, scalelike setae on one side.) Pronotum 0.84 times as long as wide; out- line about as in mundidus except transverse impression on anterior fourth stronger; sur- face shining, pmictures shallow, of moderate size, dense, nmning into one another, mar- gins of a few in lateral areas subcrenulate. Vestiture mostly abraded, of erect, short, stout, almost scalelike setae. Elytra 1.4 times as long as wide; sides al- most straight on basal two-thirds, broadly rounded behind; crenulations on bases small, narrow, distinct, about 13 on each elytron; striae narrowly, distinctly impressed, punc- tures deep, close; interstriae about twice as wide as striae, convex, covered by resin but apparently shining, almost smooth, with mi- nute, confused punctures, each with numer- ous confused, narrow, sharp crenulations (each equal to one-fourth width of an inter- striae), these transcend into uniseriate tu- bercles at base of declivity. Declivity steep, convex; interstriae narrower than on disc, with fine, imiseriate rows of tubercles at base, these reduced and almost obsolete toward apex. Vestiture of abundant, erect, small scales in ground cover; each interstriae with a row of longer erect scales, each four times as long as ground scales, about six times as long as wide. Type locality.— Pak Pak on south coast of Bombarai, Vogelkop, Dutch New Guinea. Type materl\l.— The male holotype was taken at the type locality on 4-VI-1959, be- tween 100 and 700 m. The holotype is in my collection. Bothrosternus hirsutus, n. sp. This species is distinguished from truncatus Eichhoff by the finer, much longer elytral setae, and by differences in the pronotal and elytral sculpture cited below. Male.— Length 2.3 mm (paratypes 2.2-2.5 mm), 2.1 times as long as wide; color dark brown to almost black. Frons about as in truncatus except upper area less strongly convex, glabrous on a small area less than one-fourth as extensive; most of surface granulate-reticulate. Pronotum very similar to truncatus except more finely, closely aciculate, grooves much longer, not as deep; rather dull. Vestiture fin- er, more abundant. Elytra similar in outline to truncatus; striae more abruptly impressed, punctures al- most obsolete, strongly reticulate; interstriae three times as wide as striae, strongly retic- ulate, shallow, obscure punctures mostly on margins. Declivity rather steep, broadly con- vex, about as in truncatus. Vestiture of fine, strial and interstrial hair, moderately abun- dant, longest setae near base of declivity equal in length to twice width of an inter- striae; somewhat shorter on lower half of de- clivity. Proepisternal pubescent area large, setae white (in both sexes). Female.— Similar to male except epistoma bearing a transverse, subcarinate elevation as in truncatus; glabrous area on upper frons about half as large as in truncatus and less strongly convex. Type locality.— Rancho Grande, Aragua, Venezuela. Type material.— The male holotype, fe- male allotype, and 10 paratypes were taken at the type locality 9-IV-1970, 1100 m. No. 429, Serjania, by me. Eleven paratypes bear the same data labels except for collection No. 420, taken from Tabebuia twigs. The holotype, allotype, and paratype are in my collection. Cnesinus discretus, n. sp. This species (male) is distinguished from nitidus Eggers (male homotype) by the 272 Great Basin Naturalist Vol. 45, No. 2 stronger epistomal impression, with the up- per, convex area of the frons more coarsely sculptured, by the more strongly impressed and more closely punctured discal striae, and by other characters. Male.— Length 2.2 mm, 3.1 times as long as wide; color very dark brown, elytra red- dish brown. Frons as in nitidus except lower third of frontal area more strongly, transversely im- pressed; surface rather coarsely granulate, some granules at summit of convexity form- ing an obscure, indefinite, irregular carina. Pronotum as in nitidits, punctures slightly closer and a bit more longitudinally strigose. Elytra as in nitidus except discal striae very slightly more strongly impressed, many pimctures confluent (never confluent in Jiiti- dus); declivital interstriae 2 less strongly re- duced in width, tubercles on its upper half much smaller. Type locality.— Rancho Grande, Aragua, Venezuela. Type material.— The male holotype was taken at the type locality on 9-1 V- 1970, 1100 m, from tlie broken twig of an unidentified tree, by me. The holotype is in my collection. Cnesinus minor, n. sp. This species is distinguished from electinus Wood by the much smaller size, by the dif- ferent sculpture of the female frons as de- scribed below, and by other differences cited below. Female.— Length 1.6 mm, 2.5 times as long as wide; color dark reddish brown. Frons as in electinu.s except epistomal cal- lus .shorter (on longitudinal body axis), orna- mental setae uniformly distributed on summit of callus (without a median glabrous area as occurs in electinus); setae on lateral margins reduced in number and in length. Pronotum 1.1 times as long as wide; about as in electinus except punctures not quite as elongate. Elytra 1.6 times as long as wide; similar to electinus except strial punctures larger, inter- .striae only slightly wider than striae; declivi- ty steeper, less strongly impressed; interstrial setae at ba.se of declivity slightly flattened on their distal halves (not at all flattened in electinus). Type locality.— Grecia, Costa Rica. Type material.— The female holotype was taken on 27-XM955, by B. Malkin. The holotype is in my collection. The type series of electinus (from Jalisco, Mexico) is entirely distinct from this species. However, other series from Nayarit and Guerrero are intermediate in size and show some indications of intergradation. If addi- tional specimens are found in Central Ameri- ca that show additional intergradation, the population represented by this species may have to be reduced to subspecific rank. Corthylus truncatus, n. sp. This unique species is unmatched in this remarkable genus. It is distinguished from all other species in the genus by the very stout body form, by the elytra being equal in length to the ponotum, by the truncate, mar- gined, elytral declivity, and by other charac- ters described below. Female.— Length 4.0 mm, 1.8 times as long as wide; color yellowish brown (mature color?). Frons deeply, broadly excavated from eye to eye, from epistoma to vertex; surface al- most smooth, minutely irregular; lateral mar- gins below eye subacute; epistomal margin rather strongly emarginate. Vestiture largely restricted to upper half of concavity, moder- ately abundant, longer toward upper margin; lateral margins below eye to epistomal emar- gination ornamented by a dense, confused row of rather long hair. Antennal club mi- nutely pubescent, very large, rather strongly a.symmetrical; sutures aseptate except possi- bly 1 at extreme anterior margin, showing as shallow grooves, 1 slightly oblique, 2 straight; cirrus very .slender, consisting of about six setae, exceedingly long, left cirrus extending over back ending near right margin of pronotum. Pronotum 0.81 times as long as wide; out- line almost semicircular; transversely very broadly convex, longitudinally almost straight except feebly declivous on anterior third; as- perities, weak, few in number, restricted to median third of declivous area; surface smooth, apparently slightly shagreened, punctures minute, sparse, inconspicuous. Glabrous. April 1985 Wood: New Bark Beetle*^ 273 Elytra 0.98; sides almost straight and paral- lel to declivital margin; posterior margin al- most straight, weakly curved; disc smooth, shining, punctures sparse, minute, apparently confused. Declivity abrupt, subvertical, weakly convex, margin marked by an ele- vated circumdeclivital costa, its crest acute, continuous from suture at base to suture at apex without midulations or denticles; sur- face smooth, finely reticulate, punctures mi- nute, confused, not close, a sparse row of fine granules in position of interstriae 3. Vestiture of short hair, sparse on disc, a bit more con- spicuous on declivity; of variable length. Type locality.— Jungle near Leonpampa, Hwanuco Department, Peru. Type material.— The female holotype was taken at the type locality on 6-XII-1937, 800 m. No. 3811, by F. Woytkowski. The holotype is in my collection. Hylurgus indicus, n. sp. Although several species have been as- signed to this genus in past history, this ap- pears to be only the third that actually be- longs here. It is distinguished from micklitzi Watchl by the smaller size, by the absence of a frontal tubercle, by the vestiture, and by other characters cited below. Female.— Length 3.2 mm (paratypes 3.0-3.3 mm), 3.0 times as long as wide; rather dark reddish brown. Frons resembling micklitzi except much more strongly convex, without a transverse impression just below middle, more coarsely tuberculate; median carina on epistoma of uniform height, without tubercle or tooth at dorsal end; vestiture apparently longer, more abimdant. Pronotum 1.1 times as long as wide; resem- bling micklitzi except more quadrate, sides more nearly parallel, almost straight; punc- tures apparently deeper, closer; vestiture shorter, more abundant. Elytra 1.9 times as long as wide; resem- bling micklitzi except strial punctures more distinct, slightly larger; vestiture with much fewer setae in ground cover, erect setae mostly in rows on both disc and declivity (abundant and strongly confused in micklitzi). Male.— Apparently not represented in material at hand. Type locality.— Kumaon (region), W. Al- mora, U.P., India. Type material.— The female holotype, and three female paratypes were taken at the type locality by H. G. Champion. Other paratypes include two labeled Ranikhet, Ku- maon, U.P. India, 6-VIII-1916, Pinus longifo- lia logs, H. G. Champion, and 1 from U. Gumti Val., W. Almora, U.P., India, from the same host and collector. The current name of the host is Pinus roxburghii. The holotype and four paratypes are in the Forest Research Institute Collection, Dehra Dun, U.P., India. Two paratypes are in my collection. Pachycotes minor, n. sp. This species is distinguished from villosus Schedl by the smaller size, by the much less abundant vestiture, by the much smaller strial punctures, by the smaller, more widely spaced declivital interstrial tubercles, and by other characters cited below. Male.— Length 2.3 mm (paratypes 2.3-2.6 mm), about 2.1 times as long as wide; color very dark brown. Frons impressed (almost flat but not con- cave) on median half from slightly above eyes to below level of antennal insertion; lower half of impressed area shining, coarsely retic- ulate, dull, finely subreticulate and deeply, rather coarsely, somewhat closely punctured in remaining areas; epistomal margin slightly produced on median third; epistomal pro- cesses distinct and almost subtuberculate near median line. Glabrous on shining, coarsely reticulate area; rather coarse, moderately long, hairlike setae in lateral areas, shorter above. Antenna about as in araucariae Schedl, except scape very slightly longer. Pronotum 0.9 times as long as wide; widest on basal fourth, sides convergently arcuate. Surface finely reticulate, rather dull; punc- tures moderately small, distinct but not deep, irregularly spaced by one to four diameters of a puncture. Vestiture short, sparse, most setae about equal in length to diameter of punctures from which they arise. Elytra about 1.3 times as long as wide (spread slightly); sides almost straight and parallel on more than basal two-thirds, broadly rounded behind; basal margins with 274 Great Basin Naturalist Vol. 45, No. 2 individual crenulations recognizable (not cos- tate as in some species); striae weakly im- pressed, punctures small, their centers reticu- late-granulate, spaced in a row by about three diameters of a puncture; interstriae feebly convex, subreticulate, dull, each with a central row of low, poorly formed crenula- tions, crenulations decrease from half width of interstriae at base to subtubercles at base of declivity. Declivity steep, convex; striae more strongly impressed, punctures closer; interstriae more distinctly convex, each bear- ing a row of 7 to 9 moderately coarse, rovmded tubercles to near apex, tubercles spaced by distances about equal to width of an interstriae. Vestiture sparse, consisting on posterior half of small, moderately abundant, amber scales, and rows of erect, rather short, moderately stout hairs; a few hairlike setae may extend to basal half. Female.— Similar to male except frons uniformly convex, without a glabrous, reti- culate-granulate area on lower half; strial pimctures less distinct. Type locality.— Palen Creek, about 96 km (60 miles) south of Brisbane, Queensland, Australia. Type material.— The male holotype, fe- male allotype, and 1 male and 5 female para- types were reared 14 August 1972 from a piece oi Araucariu cutminghamii taken by R. A. Yule that came from the type locality. These specimens emerged with an enormous series Pachy cotes clavatus Schedl and were found by me among that material. Tlie holotype and allotype are in the Aus- tralian National Collection, Canberra; the paratypes are in my collection. Phloeosinopsoides ptimilus, n. sp. This species is distinguished from tri- seriatus Schedl by the much smaller size, by the deeply, extensively excavated male frons, by the much larger, deeper, strial punctures, and by other characters cited below. Male.— Length 1.5 mm (paratypes 1.4-1.5 mm), 2.4 times as long as wide; color reddish brown, vestiture pale. Frons broadly, deeply, subcircularly exca- vated from eye to eye, from epistoma to well above eyes; surface reticulate-subgranulate, punctures small, obscure; vestiture of sparse, coarse, long setae uniformly distributed. Pronotum 0.94 times as long as wide; widest slightly behind middle, sides moder- ately arcuate, rather broadly rounded in front; surface smooth, dull, punctures small, close, their anterior margins elevated into very fine crenulations from base to apex; ves- titure of rather numerous, short, pale, recum- bent scales. Elytra 1.5 times as long as wide; outline about as in triseriatiis; striae slightly im- pressed, punctures very coarse, deep, close; interstriae half as wide as striae, punctures small, uniseriate, their anterior margins ele- vated, thereby causing interstriae to appear subserrate. Declivity steep, convex; details as on disc. Vestiture consisting of uniseriate in- terstrial rows of recumbent (anteriorly) to semirecumbent (declivity) short scales; each scale widest near its apex, about twice as long as wide. Female.— Similar to male except frons convex, surface reticulate-granulate, with very small, shining granules; pronotal crenu- lations distinctly larger. Type locality.— Near Bulolo, Morobe District, New Guinea. Type material.— The male holotype, fe- male allotype, and six paratypes were taken 6-VII1-1972, No. 91, from an unidentified vine by me. The holotype, allotype, and paratypes are in my collection. Xylechinosomus pilosus, n. sp. This species is distinguished from hirsutus Schedl (2.9 mm) by the smaller size, by the presence of some scalelike setae in the elytral ground cover, and by the absence of a small, rounded granule beside each puncture on the pronotum. Male.— Length 2.3 mm (allotype 2.4 mm), 2.2 times as long as wide; color brown. Frons moderately concave from slightly below upper level of eyes to level of antennal insertion; surface shining, obscurely reticu- late, punctures small and very obscure above, larger and more distinct below. Vestiture of fine, short, inconspicuous hair of uniform distribution. Pronotum 0.82 times as long as wide; somewhat like contractus (Chapuis) except punctures much closer, each with its floor (interior) strongly reticulate; spaces between April 1985 Wood: New Bark Beetles 275 pimchires smooth. Vestiture of rather abun- dant, fine, moderately long hair. Elytra 1.5 times as long as wide; about as in Jiirsutus; striae slightly impressed, punc- tures rather small, deep; interstriae about twice as wide as striae, shining, each with rather numerous, small, confused crenula- tions, smaller, rounded, and more numerous at base, reduced to pointed uniseriate tu- bercles at base of declivity. Declivity convex, moderately steep; interstriae each with a row of small, pointed tubercles. Vestiture of rather abundant, fine, long hair; some setae on declivity of slender, pointed scales. Female.— Similar to male except frons ir- regularly convex, punctures more distinct; in- terstrial crenulations and tubercles much smaller; vestiture on pronotum and elytra distinctly shorter. Type locality.— Curitiba, Parana, Brazil. Type material.— The male holotype and female allotype were taken at the type local- ity on 13-1-1969 in Araucaria angustifoUa bark by C. W. and L. O'Brien. The holotype and allotype are in my collection. NEW NEVADA ENTITIES AND COMBINATIONS IN ERIOGONUM (POLYGONACEAE) James L. Reveal' Abstract.— New species and varieties of the plant genus Eriogonum (Polygonaceae) that occur in Nevada are proposed. Eriogonum tiehmii and E. ochrocephahnn var. alcxanderae are endemic to Nevada, and E. leuisii is restricted to extreme northeastern Nevada and adjacent Utah. These entities belong to the subgenus Eucycla. Two varieties of E. umhellatum (of the subgenus Oligogonum) are proposed. The first, var. juniporinum, is found in eastern Nevada and in the desert ranges of southeastern California. The second is var. furcosiim, which is restricted to the Sierra Nevada. One new combination is also proposed: E. nudum var. gramincum. The following new entities are validated for a treatment of Eriogonum in a companion article (Reveal 1985). Eriogonum ochrocephahnn S. Wats. var. alexanderae Reveal, var. nov. A var. ochrocephalo caulibus pubescentibus differ. Low, rounded to spreading cespitose pe- rennial herbs forming open to compact mats 1-8 dm across and up to 2 dm high, with a much branched, woody caudex arising from a stout, gnarled, woody taproot; leaves erect to spreading, numerous, the leaf-blades lanceo- late to narrowly ovate, 1-2 cm long, 0.5-1.5 cm wide, white-tomentose on both surfaces, the petiole 2-5 cm long, tomentose; stems erect, 3-10 cm long, tomentose; invohicres turbinate, (3.5) 4-5 mm long, 2-2.5 mm wide, rigid and tubular, tomentose without, the 5-6 lobes 0.3-0.5 mm long and erect; flowers yellow with yellowish green midribs, 2-3 mm long, the tepals oblong, united about '/3 to V2 their length; stamens exserted, the fil- aments 3-4 mm long, pilose basally, the an- thers yellow, 0.4-0.5 mm long, oblong; achenes light brown, 3-3.5 mm long, the nar- row base tapering to a long, 3-angled beak. Type.— NEVADA, Lyon Co.: Along Ne- vada Highway 3 in Wilson Canyon between Smith and Mason, 12.8 mi NE of Smith and 2.5 mi SE of the junction of Nevada Highway 3 and Norydike Cutoff, on volcanic tuff hills S of the West Fork of the Walker River, asso- ciated with Atriplex, at about 1460 m (4800 ft) elevation, 21 Jun 1978, Reveal et al. 4737. Holotype, US! Isotypes BRY! CAS! DUKE! F! GH! MARY! MEXU! MO! NY! OKL! RENO! RSA! TEX! UC! UTC! and elsewhere. Additional specimens examined.— NE- VADA. Lyon Co.: SW slope of Wassuk Range, 13 Jun 1947, Alexander ir Kellogg 5314 (OKL, UC); 0.5 mi NW on Nevada Highway 22 from rd E along the East Walker River, 6 Jun 1981, Tiehm 6527 (MARY); Aldrich Grade along Nevada Highway 3C N of Fletcher Springs, 6 Jun 1981, Tiehm 6558 (MARY). Pershing Co.: 2.8 air mi N of Trin- ity Peak, Trinity Range, 28 Jun 1980, Tiehm 6133 (MARY); W of Cooper Valley, SE end of Sahwave Mts., 29 May 1983,' Tiehm 6 Tucker 7758 (MARY). Washoe Co.: NE side of Hungry Valley near Hungry Spring, 31 May 1980, Tiehm et al. 5769 (MARY); W of site of Leadville, Granite Range, 30 Jun 1983, Tiehm 8017 (MARY); 1.5 air mi NE of Grass Valley Range, NW end of Granite Range, 6 Jul 1983, Tiehm 8075 (MARY). The var. alexanderae is named for Annie M. Alexander (1867-1950) who discovered this plant during her last botanical expedition to Nevada with Louise Kellogg in 1947. Miss Alexander was then 80 years old. It is a plea- sure to remember this fine Nevada collector by naming this variant in her honor. The var. alexanderae is generally found on the eastern edge of the distribution of var. 'Department of Botany, University of Maryland, College Park, Mar>land 20742 and National Mnsei Washington, D.C. 20560. Research supported by Niitional Science Foundation (.rant BMS7.5-!.30(a. This the Maryland Agricultural Experiment Station. n of Natural Hi Scientific .\rlicli .\3S;57. Contribution (i8I7 of 276 April 1985 Reveal: Eriogonum 277 ochroceplmlum. The latter may have glabrous or glandular stems, but those of var. alexan- derae are always tomentose. To date, the two variants have not been found growing sympatrically. Eriogonum lewisii Reveal, sp. nov. A E. desertoro (Maguire) R. J. Davis foliis brevioribus et angustioribus cum caulibus floccosis nee tomentosis, involucris floccosis apice. Low, roimded perennial herbs forming a compact to slightly spreading mat 1-4 dm across and up to 1 dm high, with a much branched, woody caudex arising from a stout, gnarled, woody taproot; leaves erect to spreading, numerous, the leaf-blades elliptic to ovate, 1-1.5 cm long, 4-6 (7) mm wide, grayish tomentose on both surfaces, becom- ing less so and greenish beneath the tomen- tum with age on the upper surface, the peti- ole 0.8-1.8 (2) cm long, tomentose; sterns erect, 4-8 (10) cm long, floccose; involucres turbinate-campanulate, 2.5-3 mm long, 2-2.5 mm wide, rigid, floccose without with the hairs restricted (by late anthesis) to the 5 spreading, 1-1.5 mm long, lobes; /Towers yel- low with reddish yellow midribs, (2) 2.5-3 mm long, the tepals oblong, united about V4 of their length; stamens exserted, the fila- ments 3-4 mm long, pilose basally, the an- thers yellow, 0.4-0.5 mm long, oblong; achenes light brown, 3-3.5 mm long, the nar- row base tapering to a long, 3-angled, mi- nutely bristled beak. Type.- NEVADA, Elko Co.: White Ele- phant Butte, S of Elk Mtn, on a steep, open, gravelly slope, associated with Cercocarpus and Senecio, sec. 4, T46N, R61E, at 2530 m (8300 ft) elevation, 30 Jul 1976, Reveal h Re- veal 4596. Holotype, US! Isotypes, BRY! CAS! F! GH! MARY! MEXU! MICH! MO! NY! OKL! RENO! RSA! TEX! UC! UTC! Additional collections examined.— NEVADA. Elko Co.: Independence Mts., 12 Aug 1980, Tiehm ir Rirdsey 5193 (RENO). UTAH. Box Elder Co.: Copper Mtn, 13 Jun 1928, Cottam 3089 (BRY, F). This new species of Eriogonum is named in honor of Mont E. Lewis, long a major collec- tor of intermountain plants for the United States Forest Service and an authority on the genus Carex. It was on the basis of a collec- tion he made that I searched the White Ele^ phant Butte area for this unusual buckwheat. It, like E. ochrocephalum var. alexanderae proposed above, belongs to the subgenus Eu- cycla (Nutt.) Kuntze. In Nevada, Lewis's buckwheat may be dis- tinguished by its floccose stems and in- volucres, small leaves, and high elevation habitat. It differs from the lower elevation and more southern Eriogonum desertorum by these features and others, notably the well- defined yet somewhat papery involucral tube that retains its rigid characteristics. In Utah and Idaho the capitate forms of Eriogonum brevicaule var. laxifolium (Torr. & Gray) Re- veal most closely resemble E. lewisii. The va- riety tends to be restricted to the Wasatch Ranges of Idaho and Utah, ranging eastward to southwestern Wyoming. Like E. desertor- um, the var. laxifolium has a well-defined and densely tomentose involucre. In addition, the leaves of this more eastern expression are decidedly longer and narrower than those of E. lewisii. Eriogonum tiehmii Reveal, sp. nov. A E. anemophilo Greene involucris longi- oribus et floribus sparse glandulosis differt. Low, spreading perennial herbs forming a dense compact mat up to 3 dm across and to 1.6 dm high, with a much branched woody caudex arising from a stout, gnarled taproot; leaves erect to spreading, numerous, per- sistent, with a bluish gray hue, the leaf-blades elliptic to oblong, (0.8) 1-2 (2.5) cm long, 5-8 (10) mm wide, entire, grayish to whitish to- mentose on both surfaces, often slightly greenish under the tomentum with age on the upper surface, the petiole 0.5-1.6 (2) cm long, tomentose without, glabrous within, with an expanded petiole base 3-5 mm long and 1-2 mm wide; stems erect, (0.6) 1-1.3 (1.5) dm long, floccose, greenish or reddish under the hairs; involucres turbinate-campan- ulate, 4-5 mm long, 3-4 mm wide, rigid, floccose and reddish without, glabrous within except for a few hairs at the very tip in some, with 5-6 erect to slightly spreading, 1.5-2 mm long, narrowly triangular lobes, the 278 Great Basin Naturalist Vol. 45, No. 2 bractlets linear, 2-4 mm long, minutely grandular and toothed, the pedicels exserted, 4-7 mm long, glandular throughout and espe- cially so near the apex; flowers yellowish white or whitish to cream with reddish mid- ribs and apices or merely reddish to reddish green midribs, often with a greenish yellow base, 2.5-3.5 mm long at anthesis, up to 4 mm long in early fruit, the tepals oblong, the outer slightly broader than inner, both with out-rolled margins, the apices truncate with a slightly emarginate apex in the outer series of tepals, stipitate glandular along the midrib and base without, sparsely glandular within, united 1/5 to 1/4 of the length; stamens ex- serted, the filaments 3-4 (4.5) mm long, pi- lose basally, the anthers pale yellow, 0.4-0.5 mm long, oval to oblong; achenes light brown, 3-4 mm long, the subglobose base ta- pering to a long, 3-angled beak about Va the length of the fruit, the stigma 1-1.2 mm long. Type.- NEVADA, Esmeralda Co.: Silver Peak Range just N of the road from Silver- peak to Fish Lake Valley, 1.2 air mi NNW of Cave Springs, sec. 27, TIS, R37E, 1830 m (6000 ft) elevation, 31 May 1984, Tiehm, Re- veal Williams and Reveal 8534. Holotype, US! Isotypes, BRY! CAS! MARY! NY! RENO! RSA! UTC! and elsewhere. Additional specimens examined.— NE- VADA: Esmeralda Co.: Silver Peak Range, 1.2 air mi NNW of Cave Springs, 18 May 1983, Tiehm 7707 (BRY, CAS, MARY, NY, RSA, UTC). This remarkable species, named for Arnold ("Jerry") Tiehm, may be immediately recog- nized by its large, distinctly lobed involucres, cream-colored flowers, and stipitate-glandu- lar tepals. In this latter feature, Eriogonnm tiehmii is unique. Minute glands are common on the inner surface of tepals in many species of cespitose buckwheats belonging to the sub- genus Eucycla. Tiehm's buckwheat is the only species with well-defined stipitate glands on the outer surface. In Nevada, E. tiehmii is morphologically most similar to E. anemophihim Greene and the cream-colored phase of E. beatleijae Reveal. The scapes of the latter are glandular, not floccose as in E. tiehmii, and the involucre of both established species does not approach the size of E. tiehmii. Eriogonnm nndnm Benth. var. gramineum (S. Stokes) Reveal, comb. nov. Based on E. gramineum S. Stokes, Gen. Eriog. 60. 1936. Type. California, Inyo Co.: Argus Mts., 1897, Purpus 5676. Holotype, UC! Isotypes, GH! K! MIN! P! US! The concept of Eriogonnm nudum var. pubiflorum Benth. in DC. has become in- creasingly restricted in its definition so that now the plant is defined as a northern ele- ment ranging from the central Sierra Nevada of California northward to south central Ore- gon. In Nevada, var. pubiflorum is found only in the extreme northwestern portion of the state, and then it is infrequent. In south- ern California, the long recognized var. pau- ciflorum S. Wats, of the Transverse ranges (and southward to northern Baja California Norte) has not been confused with var. pub- iflorum, but in the southern portion of the Sierra Nevada the distinction is not always readily apparent. Howell (1976) proposed var. westonii (S. Stokes) J. T. Howell for the plants of the Tehachapi region and the west- ern slope of the Sierra Nevada. The defini- tion of that variety, as noted by Howell, can- not be expanded to include the plants of the desert range and eastern slope of Sierra Ne- vada common to Inyo and Mono counties, California. Thus, the var. gramineum is pro- posed. At present, var. gramineum, charac- terized by its yellow pubescent flowers and inflated stems, is not known from Nevada. It is to be sought in the Death Valley region of the state. Eriogonnm umbeUatum Torr. var. furcosum Reveal, var. nov. A var. eUiptico foliis subglabris vel glabris supra. Low, rounded subshrubs up to 4.5 dm high and 8 dm across, infrequently forming a spreading mat to 5 dm across at higher eleva- tions; leaves in loose rosettes, the leaf-blade elliptic to oblong, (0.7) 1-2.5 (3.5) cm long, 3-8 (13) mm wide, densely white tomentose below, thinly floccose or more commonly glabrous and bright green above, the petiole 3-10 (12) mm long; flowering stems erect, slender, 0.5-2 dm long, thinly floccose; in- florescences compoundly umbellate, 0.5-1.5 April 1985 Reveal: Eriogonum 279 dm long, the branches floccose; involucres with tubes 2-3 (4.5) mm long, the usually re- flexed lobes shorter to as long as the tube, floccose without; flowers bright yellow, (5) 6-8 mm long including the stipe. Type.- CALIFORNIA, El Dorado Co.: Along California Highway 89, 2.2 mi S of U.S. Highway 50, on sandy granitic soil, asso- ciated with Arctostaphylos, Artemisia, and juniper-pinyon, 23 Aug 1975, Reveal 3971. Holotype, US! Isotypes, ARIZ! BRY! CAS! DUKE! F! GH! K! MARY! MEXU! MICH! MO! NY! OKL! RENO! RM! RSA! TEX! UC! UTC! and elsewhere. Representative specimens.— CALIFOR- NIA. Alpine Co.: Silver Lake, 2 Sep 1933, Mason 7255 (lA, UC). Amador Co.: Bear Riv- er, 30 Jul 1896, Hansen 1966 (B, MIN). Cala- varas Co.: Dorrington, 7 Aug 1923, Jepson 10058 (JEPS). El Dorado Co.: 5.5 mi S of Meyers, 9 Aug 1941, Wheeler 402 (JEPS). Fresno Co.: Mono Creek Dam, 16 Jul 1935, Everett 6 Johnson 7313 (DS, IDS, LA, MO, OKL, RSA, TAES, UC, UT). Kern Co.: ridge SE of Pine Flat, 28 Jul 1965, Twissehnann 11330 (CAS, UTC). Madera Co.: The Niche, East Fork of Granite Cr., 17 Aug 1958, J.T. Howell 34552 (CAS). Mariposa Co.: above Nevada Falls, Yosemite N.P., 10 Jul 1889, Chestnut 6 Drew s.n. (KANU, UC). Nevada Co.: Puddingstone Ridge, 5 mi E of North Columbia, 24 Jul 1971, True 6903 (CAS). Pla- cer Co.: Cisco, 14 Sep 1938, Rose 38257 (RM). Sierra Co.: Sierra Valley, Sep 1872, Lemmon s.n. (ISC). Tulare Co.: Burnt Ridge, 18 Sep 1962, Twissehnann 7780 (CAS). Tuo- lumne Co.: 0.8 mi E of Dardanelle, 28 Jul 1972, Reveal h Reveal 2813 (BRY, CAS, DUKE, F, MICH, MO, NY, OKL, RSA, TEX, UTC). NEVADA. Washoe Co.: nr Verdi, 19 Jun 1903, Stokes s.n. (RSA). The var. furcosum has been included under the name of the Pacific Northwest expres- sion, var. stellatum (Benth. in DC.) M. E. Jones, or, as this must now be called due to a recent change in the International Code of Rotanical Nomenclature, var. ellipticum (Nutt.) Reveal (Reveal 1983), in recent treat- ments of the genus in California (Reveal & Munz 1968). This expression is primarily found in the Sierra Nevada, and then mainly along the more gentle western slope. As the elevation of this range decreases to the north, the variety crosses the crest of the Sierra Ne- vada and enters into the Lake Tahoe Basin of California and onto the eastern foothills in extreme western Nevada. In southern California, the var. furcosum gives way to var. munzii Reveal in the Trans- verse Ranges and to var. subaridum S. Stokes on the arid eastern slopes of the Sierra Ne- vada. The var. chlorothamnus Reveal in Munz is found along the eastern foothills of the Sierra Nevada bordering the Inter- mountain Region; this variant is isolated from var. furcosum by the Sierran crest. To the north, the new variety is replaced by another compoundly umbellate expression of Eriogo- num umhellatum. This northern expression, which extends into the Siskiyou-Trinity re- gion of California and adjacent southern Ore- gon, is as yet unnamed. Along the Coast Ranges of California is the distinctive var. balniforme (Torr. & Gray) Jeps. This expres- sion is not found in the Sierra Nevada. Eriogonum umhellatum Torr. var. juniporinum Reveal, var. nov. A var. subaridio S. Stokes floribus cremeis differ. Plants forming low shrubs or subshrubs up to 8 dm high and 10 dm across; leaves in loose rosettes, the leaf-blade elliptic, (0.7) 1-2 cm long, (3) 5-10 (12) mm wide, greenish or whitish floccose to glabrous on both surfaces, becoming greenish floccose above with matu- rity; flowering stems slender, erect, 1-2.5 dm long, green and floccose to nearly glabrous; inflorescences compound umbellate, 0.5-1.5 (2) dm long, floccose; peduncles slender, 0.5-5 (6) cm long, floccose; involucres with tubes (2.5) 3-3.5 mm long, the reflexed lobes 1-2.5 mm long, thinly floccose; flowers cream colored or whitish, (4) 5-6 mm long. Type.- NEVADA, White Pine Co.: Along U.S. Highway 50-6 at Sacramento Pass, on the northern end of the Snake Range, about 11 mi northwest of Baker, on sandy soil asso- ciated with juniper-pinyon and Artemisia, at about 2180 m (7150 ft) elevation, 13 Aug 1975, Reveal 6 Reveal 3925. Holotype, US! Isotypes, BRY, CAS, MARY, NY, OKL, TEX. Representative specimens.— CALIFOR- NIA. San Bernardino Co.: Upper Cotton- wopd Canyon, Mid Hills, 30 Aug 1973, Hen- rickson 12727 (RSA); 2 mi from Kingstone 280 Great Basin Naturalist Vol. 45, No. 2 Peak, S slope of Kingstone Mts., 23 Oct 1977, Henrickson ir Prigge 16298 (RSA); above Keystone Spring, New York Mts., 13 Oct 1935, Miinz 13854 (CI, DS, POM, UC, UTC); Mitchells Caverns State Park, Providence Mts., 20 Jun 1973, Thome 6 Tilforth 44033 (RSA, in bud); NW of Pachalka Springs, Clark Mts., 6 Oct 1935, Wolf 7605 (LA, OKL, RSA). NEVADA. Lincoln Co.: Horse Spring Basin, Mormon Range, 28 Jun 1954, GuUion 566 (OKL, UC); Silver Canyon, Mt. Irish, 19 Jim 1938, Jaeger s.n. (POM, not in flower). White Pine Co.: Pole Canyon, Snake Range, 15 Aug 1964, Holmgren ir Reveal 1664 (BRY, CAS, DS, MO, NY, RENO, RM, RSA, UC, UTC); Ward Mtn., S of Ely, 4 Aug 1969, McClintock s.n. (CAS); 1.3 mi E of Robinson Summit, 15 mi NW of Ely, 13 Aug 1975, Reveal I- Reveal 3927 (MICH). The var. juniporinum is closely related to var. suharidum S. Stokes in terms of its habit and habitat but more like var. dichro- cephalum Gand. and var. versicolor S. Stokes as to flower color. The latter two variants are spreading, matted perennials and not at all shrubby, and in general the leaves are tomen- tose, at least on the lower surface in these ex- pressions. The disjunct distribution of var. juniporinum is intriguing not only because the variant occurs in both the Great Basin cold desert habitat in southeastern Nevada and in the warm desert habitat of the Mojave Desert in southeastern California, but also because there is a delay in flowering between the two areas. In Nevada the variety comes into flower in late June and continues to flower into early September. In California, plants begin to flower in July and continue to flower well into late October. It is likely this difference in flowering time is due to the sea- sonality of rainfall. The var. juniporinum is generally a plant of the desert foothills and low passes in Nevada, reaching its upward limits about 2250 m elevation. The var. di- chrocephalum and var. versicolor are found in the mountain ranges across southern Nevada between the two disjunct populations of var. juniporinum. These variants generally occur at higher elevations, or at least at points higher on mountain slopes than var. junipori- num. In California the new variety is found on the higher slopes, but it may be found as low as 1350 m elevation in the Providence Mountains and on the Kingstone Range. Literature Cited Howell, J. T. 1976. Erid'^onuui notes \'II. Mentzelia 1:17-22. Reveal, |. L. 1983. The Deiiiouliii rule and newly man- dated combinations in Eriog,onuiii (Polyg- onaceae). Taxon 32:292-295. 1985. An annotated key to Eiio^oniim (Polyg- onaceae) in Nevada. Great Basin Nat. In press. Reveal, ]. L., and P. A. Mi.\z. 1968. "Eriogonum." Pages .33-72 in P. A. Munz, Supplement to a Cal- ifornia flora. Berkelev: Univ. of California Press. GROWTH AND REPRODUCTION OF THE FLANNELMOUTH SUCKER, CATOSTOMUS LATIPINNIS, IN THE UPPER COLORADO RIVER BASIN, 1975-76' Charles W. McAda^^ and Richard S. Wvdo,ski2.4 Abstract.— Growth rates estimated using the scale annuli of flannelmouth sucker, Catostomus latipinnis, did not differ between fish collected from the Gunnison and Colorado rivers, and the Green and Yampa rivers. However, body condition and fecundity were significantly greater in the former population. Age of first maturity for male and female fish from all rivers was IV; and most fish were mature by age VII. The smallest mature female collected was 405 mm, and the smallest mature male was 391 mm total length. Fecundity ranged from 4,000 ova in fish 450 mm long to 40,000 ova in a 500-mm fish: mean ovum diameter was 2.39 mm. Ripe male flannelmouth suckers were collected from early April through June; ripe females were collected from both study areas during May and early June. The flannelmouth sucker, Catostomus lati- pinnis, is one of the most abundant and wide- ly distributed native fishes in the warm water tributaries and mainstream rivers of the Up- per Colorado River Basin (Tyus et al. 1982). However, its distribution in the Lower Colo- rado River Basin has been substantially re- duced by habitat alteration resulting from channelization and water development (Minckley 1973). Despite its importance as a native species endemic to the Colorado River system, little is known of the biology of the flannelmouth sucker (McDonald and Dotson 1960, Wiltzius 1976, Carlson et al. 1979). In this report we describe the growth, maturity, and fecimdity of the flannelmouth sucker in the major rivers of the Upper Colorado River Basin. Methods Flannelmouth suckers were collected from the confluence of the Yampa and Green riv- ers in Dinosaur National Monument and from reaches of the Colorado and Gunnison rivers near their confluence in western Colo- rado (Fig. 1). Descriptions of sampling sites were provided by McAda and Wydoski (1980), whereas an account of the general physical and ecological features of the large rivers of the Upper Colorado River Basin was provided by Bishop and Porcella (1980). Fish were sampled between April and No- vember 1975 and 1976 by using trammel nets (26-90 m long, 2.5 cm-mesh inner wall, 25 cm-mesh outerwall) and seines (30 m long, 2.5 cm mesh; and 5 m long, 3 mm mesh) and by electrofishing. Collected fish were weighed (g) and measured in total length (mm). Scales from midway between the later- al line and the anterior insertion of the dorsal fin were used for age determination. An age determination for an individual fish was con- sidered to be accurate when agreement on the number of scale annuli occurred between the first and second examinations. A fish was excluded from analyses that involved fish age when no agreement on the number of scale annuli could be reached after a third exam- ination of the scales. Total body length (TL, mm) at time of an- nulus formation was estimated using the equation TL = b^ + biSR + biSR^ + b^SR^ (Carlander 1956), where SR is the radius to that particular annulus X 80 and b^ - bj, are 'The Utah Cooperative Fishery Research Unit is jointly supported by the United States Fish and Wildhfe Service. I'tah Division of Wildhfe Resources, and Utah State University. ■Utah Cooperative Fishery Research Unit, Utah State University, UMC-52, Logan, Utah 84322. 'Present address: United States Fish and Wildhfe Service, 551 25% Road, Grand Junction, Colorado 81505. 'Present address; United States Fish and Wildlife Service, Program Plans, Washington, D.C. 20240. 281 282 Great Basin Naturalist Vol. 45, No. 2 WYOMING Flaming Gorge Reservoir COLORADO KILOMETERS Lake Powell Fig. 1. Locations ot s;i rivers; (2) (Colorado liivci 111 tlie Upper Colcjrado \{\\xy Ba: illi ot the (liinnisoii l-liver; ami (.1 Hi empirical constants derived using multiple regression analysis. Length-weight relation- ships were estimated using the equation Log W = fo, Log TL - bu (Tesch 1971), where W is body weight (g) and /;„ and /; , are empirical constants derived using least-squares analysis. Fish were dissected to determine sex and stage of sexual maturity. Females were con- sidered mature when ovaries contained large, opaque, yellow ova; males were considered mature when the testes were enlarged and white. Fish were considered ripe when sex products were expressed with light pressure on the abdomen. Ovaries from mature fe- males collected between April and June 1975 and 1976 were used to estimate fecundity. Fecundity was estimated gravimetrically in the laboratory, where about 10% of each ova- ry was weighed to the nearest 0.1 g and the individual mature ova from this subsample April 1985 McAda, Wydoski: Flannelmouth Sucker 283 Table 1. Leiiiz;tli-\voisi;lit. hody-staU' and Iciititli-fftuiidily iclatioiisliips tor namu Gunnison, Colorado, Green, and Yanipa rivers, 1975-1976. Lencth-Wkicht Yampa/Green: Log W = 3.13 Log TL - 5.:57 (n = 297, R- = 0.9) Colorado/Gunnison: Log \V = 3.09 Log TL - 5.21 (n = 292, K^ = 0.9) B()nY-Sc:.\LE Yampa/Green Rivers Male: TL = -18.6342 + 4.SS94 SR - 0.0054 Sfi^ _ ().()()04 SR^ (n = 139, R^ = 0.9) Female: TL = -8.7835 + 3.5187 SR + 0.0131 SK^ - 0.0001 .SR^ (n = 137, R^ = 0.8) Colorado/ (Iimnisoii Rivers Male: TL = -8.7835 + 2.9904 SR + 0.0156 SR- - 0.0001 SR'5 (n = 158, R- = 0.8) Female: TL = 46.385 + 0.2820 SR + 0.0443 SR^ - 0.0002 SR'^ (n = 137, R^ = 0.8) Lencth-Fecundity Yampa/Green: Log F = 4.03 Log TL - 6.70 (n = 58, R^ = 0.6) Colorado: Log F = 3.00 Log TL- 3.76 (n = 45, R- = 0.7) Gunnison: Log F = 3.48 Log TL - 5.14 (n = 15, R^ = 0.6) were counted. Tlie fecundity of that fish was then estimated by proportion using total ova- ry weight. Total counts of ova from two fish demonstrated that estimated fecundity dif- fered from actual fecundity by less than 5%. The mean diameter of mature ova from indi- vidual fish was derived from measurements of 30 ova made with an ocular micrometer. The total length-fecundity relationships were de- termined using the equation Log F = b] Log TL - hi) (Bagenal 1967), where F is fecundity and hn and h^, are empirical constants derived using least-squares analysis. Statistical comparisons between length- weight, body-scale, and length-fecundity regression equations were made using analy- sis of covariance (Snedecor and Cochran 1967). Results Age and Growth About 80% of the flannelmouth suckers used in our analyses were collected between April and July. The collection from the Gun- nison River was made in April 1976. Body- scale relationships differed significantly be- tween male and female fish among the study areas (P<0.05: Table 1). However, mean length at annulus formation was similar for the study groups and they were averaged for this report (Fig. 2). Average growth in- crements of all fish were greatest at the for- mation of the third annulus and declined steadily thereafter. Data for the two sexes were combined be- cause there was no significant difference be- tween the length-weight relationships for male and female flannelmouth suckers (P>0.05). No statistical difference was de- tected between length-weight regressions from data on fish from the Gunnison and Col- orado rivers (?>0.05). Flannelmouth suckers from the Colorado and Gunnison rivers were significantly heavier than fish of equal length from the Yampa and Green rivers (P< 0.001: Table 1). Reproduction In the Colorado and Gunnison River col- lections, the smallest mature female was 421 mm long, and all females longer than 490 mm were mature. The smallest matvire male was 391 mm long, and all males 470 mm or longer from the Colorado and Gunnison riv- ers were mature. In the Yampa and Green River collections, the smallest mature male was 393 mm long, and all males were mature at 460 mm: the smallest mature female was 405 mm long and all females were mature at 470 mm (Table 2). Thus, fish began to mature at age IV and most were mature by age VI. 284 Great Basin Naturalist Vol. 45, No. 2 ANNULUS Fii^. 2. Mean total length and growth increments at annuliis formation for male and female flannelmoiith suckers troin the Cunnison, Colorado, Green, and Yampa rivers, 1975-1976. Numbers indicate sample size. Bars indicate one standard de\ iation. The diameters of preserved, mature ova ranged from 1.99 to 3.15 mm (mean = 2.39 mm; n = 49). The relationship between ova diameter and fish length was not significant (P>().()5). Fish from the Yampa and Green rivers pro- duced significantly fewer matiue ova than did fish from tlie Gunnison or Colorado rivers (P<().(X)1: Table 1). Flannelmouth suckers from tlie Gunnison River also produced sig- nificantly fewer ova than did those from the Colorado River. Based upon the regression equations, a fish 450 mm long from the Yam- pa River produced about 9,800 ova, whereas fish of this length from the Gunnison and Colorado rivers produced about 12,700 and 15,900 ova, respectively. Ripe male flannelmouth suckers were col- lected from both study areas when sampling began in early April and on through June. April 1985 McAda, Wydoski: Flannelmouth Sucker 285 Table 2. Relation of tot il length t( sexual matin- ty in tlannt liiiouth suckers from the Y nnpa, (Jreen C^olorado, and Gunnison rivers, 1975- 1976. Colorado and Gunnison Rivers Yam pa and G reen Rivers Total Fem ale Male Female Male length Number Percent Number Percent Number Percent Number Percent (mm) of fish mature of fish mature of fish mature of fish mature 381-390 3 0 4 0 1 0 3 0 391-400 3 0 6 17 1 0 1 100 401-410 1 0 6 33 5 20 3 100 411-420 2 60 7 71 2 0 5 80 421-430 5 0 6 67 7 43 7 100 431-440 2 60 6 100 3 67 19 100 441-450 9 75 11 91 5 40 18 94 451-460 4 80 14 93 6 67 12 92 461-470 5 100 16 100 14 93 11 100 471-480 7 92 14 100 9 100 11 100 481-490 14 100 12 100 5 100 9 100 491-5(K) 15 100 6 100 9 100 1 100 501-510 8 100 5 100 13 100 4 100 511-520 9 100 2 100 10 100 521-530 13 100 1 100 3 100 531-540 7 100 5 100 541-550 5 100 3 100 551-560 7 100 _ _ 561-570 1 100 2 100 571-580 - - 1 100 Ripe females were collected in May and early June, but none were collected after this period. In the Yampa and Colorado rivers, ripe male and female flannelmouth suckers were collected at the upstream points of cobble bars in water about 1 m deep, with a water velocity of about 1 m/sec. Although spawn- ing activity was not observed, the presence of ripe females (collected only over the cobble bars) suggested that spawning probably oc- curred nearby. Flannelmouth suckers repro- duced successfully in both 1975 and 1976 as evidenced by the abundant young of the year (30-40 mm TL), which we readily captured at all study areas by midsummer. Discussion The back-calculated length of flannel- mouth suckers at the time of annulus forma- tion for annuli I and II were similar to those estimated by McDonald and Dotson (1960) for flannelmouth suckers from the upper Green River and by Carlson et al. (1979) for fish from the upper Yampa River; however, our estimates of total body length at the for- mation of annuli III and greater were longer than their estimates. They also observed a de- cline in annual growth increments at age IV, similar to this study, which probably reflects the diversion of energy from growth to re- production at the onset of reproductive maturity. Although annual growth increments did not differ significantly between rivers, the significant difference in the length-weight relationships reflects heavier body weight for a given length of fish from the Colorado and Gunnison rivers. This observation probably reflects a difference in the nutritional status of the fish examined in this study. However, the similarity between back-calculated length at annulus formation during previous years suggests that these differences may not al- ways occur. Similar differences in weight of razorback sucker, Xyrauchen texamis, and fe- cundity of bluehead sucker, C. discobolus, from these study sites were observed by McAda and Wydoski (1980, 1983 [respective- ly]) during the same period. Although this ob- servation probably reflects a difference in the nutritional status of the fish examined in this study, we cannot speculate whether this phe- nomenon represents consistent differences between the two study areas or merely re- flects differences that occurred during the study period. However, the similarity be- 286 Great Basin Naturalist Vol. 45, No. 2 tween back-calculated lengths at annulus for- mation during previous years suggests that these differences may not always exist. Vanicek and Kramer (1969) documented a decline in the growth rate of Colorado squawfish, Ptijchocheilus hicius, and roundtail chub, Gila robusta, in the Green River after the closure of Flaming Gorge Dam, which they attributed to the resultant decrease in water temperature. Although probably a con- tributing factor, the lower water temperature in the Green River cannot be completely re- sponsible for the observed differences in fe- cundity and body weight because flannelmouth suckers in this study were pri- marily collected in the mixing zone of the Green River with the Yampa River, which maintains its historic temperature regime. Acknowledgments We thank Karl Seethaler, George Kidd, Dennis Cox, John Henderson, Larry Maclain, and Craig Reger, who assisted in the field; and Jay King, Calvin Larson, Kevin Priest, Kurt Rudd, and Joy Wydoski, who assisted in the laboratory. Lynn Kaeding, P. W. Esch- meyer, and Ross Bulkley provided valuable comments to early versions of this manuscript. Literature Cited Ba(.i:.\al, T. B. 1967. A short review of fish fecunditv. Pages 89-111 in S. D. Cerking, ed., The biologi- cal ha.sis of freshwater fish production. Blackwell Scientific Publications, Oxford. Bishop, \. B., and D. B. Porcella. 1980. Physical and ecological aspects of the upper Colorado River Basin. Pages 17-49 in \V. A. Spofford, Jr., A. L. Parker, and A: V. Kneese, eds.. Energy devel- opment in the soutliwest, Vol. 7. Resources For the Future, Res. Pap. R-18. Washington, D.C. 4.56 pi). Cahlander. K. D. 1956. Fish growth studies: techniques and role in surveys and management. Trans. North Amer. Wildl.. and Nat. Res. Conf. 21:262-274. Carlson, C. A., C. G. Prewitt, D. E. Snyder, E. J. Wick, E. L. Ames, and W. D. Frank. 1979. Fish- es and macroinvertebrates of the White and Yampa rivers, Colorado. Bur. Land Manage. (Colorado). Biol. Series No. 1. 159 pp. Mc:Ada, C. W., and R. S. Wydoski. 1980. The razor- back sucker, Xyniiiclicn texanii.s. in the Upper Colorado River Basin, 1974-76. U.S. Fish Wild. Serv., Tech. Pap. 99. 15 pp. 198.3. Maturity and fecundity of the bluehead sucker, Catostomus discobolus (Catostomidae), in the Upper Colorado River Basin, 1975-76. South- west Nat. 141:120-12.3. Mc Donald, D. B., and P. A. Dotson. 1960. Fishery in- vestigations of the Clen Canyon and Flaming Corge impoundment areas. I'tah Dept. Fish and Game Info. Bull. 60-63, Salt Lake City. 70 pp. Minckley, W. L. 1973. Fishes of Arizona, .\rizona Game and Fish Dept., Phoenix. 286 pp. Snedecior. G. W., and W. G. Cochran. 1967. Statistical methods, 6tli ed. Iowa State Univ. Press. .\mes. 539 pp. Tescii, F. W. 1971. Age and growth. Pages 98-130 in W. E. Ricker, ed.. Methods for assessment of fish production in freshwaters. Int. Biol. Programme Handbook Number 3. 2d ed. Blackwell Scientific Publications. Oxford. Tvrs. H. M., B. D. Burdick, R. A. Valdez, C. M. Haynes, T. a. Lytle, and C. R. Berry. 1982. Fishes of the Upper Colorado River Basin: Distri- bution, abundance, and status. Pages 12-70 in \\. H. Miller, H. M. Tyus. and C. A. Carlson, eds.. Fi.shes of the Upper Colorado River System: pres- ent and future. West. Div.. Anier. Fish. Soc. 131 pp. Vanicek, C. D., and R. H. Kramer. 1969. Life history of the Colorado s(iua\\fish, rtychocheilus Indus. and the Colorado chub, Gihi rohiistii. in the Green River in Dinosaur National .Monument, 1964-1966. Trans. Amer. Fish. Soc. 98(2): 193-208. \NiLTzius, W. J. 1976. Some historic influences of reser- voirs and irrigation diversions on flows, temper- atures, and fish distribution in the Gunnison Riv- er. Colorado Div. Wildl. Fort Collins. 194 pp. BURROWING OWL FOODS IN CONATA BASIN, SOUTH DAKOTA James G. MacCracken'-^. Daniel W. Uieslc^ and Rielianl M. Hansen' Abstract.— Burrowing Owls [Atliciw cuniculdrid) were studied in a prairie dog town of southwestern South Dakota. Pellets regurgitated h\ Burrou ing C)u Is contained a wide variety of prey remains. Insects, spiders, small manuuals, and vegetation were the most frequent items identified in the pellets. Mammals were consumed most frequently dining spring and early summer. Insects were consumed in large numbers during the entire period of this studv, but they became more frequent in owl pellets during late summer and fall in association with a decline of mammal remains. Some prey items observed aroimd owl nest sites were not foimd in the pellets examined. Possible secondary poisoning of some prey of Burrowing Owls has not produced any change in owl food habits, based on other studies reported in the literature. Published information on Burrowing Owls (Athene cunicularia) pertains mostly to food habits (Robertson 1929, Hamilton 1940, Sperry 1941, Bond 1942, Marti 1974, Gleason and Craig 1979). However, Thomsen (1971), Coulombe (1971), and Martin (1973) also ex- amined Burrowing Owl behavior and ecolo- gy. Little is known about ecology of Bur- rowing Owls in South Dakota except that thev are summer residents in all but the Black Hills area (Whitney et al. 1978). Burrowing Owls are frequently associated with prairie dogs (Cynomys hidovicianus) in southwestern South Dakota, where they use prairie dog burrows as nest sites and escape cover. Aufforth (1981) stated that Burrowing Owl numbers are declining in the Northern Great Plains. Tlie information available on Burrowing Owl population trends in South Dakota suggests, however, that they are stable (N. R. Whitney, pers. comm.). Recent prairie dog poisoning programs may have re- sulted in the direct or secondary poisoning of Burrowing Owl prey when Compound 1080 and strychnine were in use. Poisoning of Bur- rowing Owl prey could have altered the food resources of the study area and owl food habits. Tlie purpose of this study was to examine Burrowing Owl food habits throughout the owl's annual period of residency on the study area. Study Area and Methods The study was conducted in Conata Basin, which is on the Buffalo Gap National Grass- lands in southwestern South Dakota. Conata Basin is a lowland area surrounded by buttes and mesas and is bordered on its northern, eastern, and western edges by Badlands Na- tional Park. The basin supports short-grass prairie dominated by blue grama (Boitteloua gracilis), buffalo grass (Biichloe dacty hides), western wheatgrass (Agropyron smithii), ca- rices (Carex spp.), red three-awn {Aristida longiseta), scarlet globemallow (Sphaeralcea coccineo), wooly Indian wheat [Plantago spinuloso), and plains prickly pear {Opuntia pohjacantha). The study area has been grazed by cattle since 1900, and forage utilization is often in excess of 60% (Uresk et al. 1982). Prairie dogs historically and currently occupy the entire Conata Basin area (— 700 km-) despite poi- soning programs (Merriam 1902). The last major prairie dog poisoning effort occurred in 1979. Follow-up poisoning at a mainte- nance level may be conducted to control prairie dog reinvasion. Prey remains identified from regurgitated pellets were used to estimate Burrowing Owl food habits from April to October 1981. Fresh pellets were collected every two weeks or whenever visits were made to the study 'Department of Range Science, Colorado State University, Fort Collins, Colorado 80523. -Present address: P.O. Box .314.5. Palmer, Alaska 99645. 'USDA Forest Service, Rocky Moimtain Forest and Range Experiment Station, South Dakota School of Mil 57701. and Technology, Rapid City, South Dakota 287 288 Great Basin Naturalist Vol. 45, No. 2 area. The accumulation of prey remains at nest sites also provided additional informa- tion on foods that did not show up in the pel- lets (Thomsen 1971). All owl pellets were oven dried at 60 C for 48 hours, then weighed. Pellets were then placed in fine mesh nylon bags and agitated in warm water in a clothes washer until the pellets fell apart and all soluble material was cleared. The bags were then tumbled dry in a clothes dryer. This procedure was developed by Johnson and Hansen (1979) for the analy- sis of coyote {Conis latrans) feces. Mammal remains were identified by hair characteristics (Moore et al. 1974) and/ or comparison of teeth with reference materials. Feathers were identified to order using char- acteristics described by Day (1966). Refer- ence materials were used to identify birds beyond order when possible. Arthropod re- mains were identified to family by com- parison with insects and spiders collected from the area. Owl food habits were quantified as percent frequency for each two-week collection, based on the number of items in the pellets, not the nimiber of pellets examined. A two- way analysis of variance was used to test for differences among prey categories (mammals, birds, reptiles, and arthropods) and months. Tukey's method was used to determine which factors accounted for any differences. Results Data on Burrowing Owl foods was ob- tained from 145 pellets. The mean dry weight of a pellet was 1.1 ± 0.4 grams. Ar- thropods accoimted for the majority of items in owl pellets during each month of study, whereas mammals were the next most abun- dant prey item. Reptile remains were in- frequent in the pellets, but vegetation was abundant. Vegetation in owl pellets was usu- ally represented by small plant fragments, presumably originating from the stomachs of prey. However, large pieces of grass were also recovered (Table 1). Mammal remains were most frequent in owl pellets in May, June, and July; then they decreased substantially (P < 0.05) in August and September. Arthropod remains became more frequent (P < 0.05) in the pellets in conjunction with the decrease in mammals. No other significant differences were de- tected in Burrowing Owl food habits. Prey remains that were found at nest bur- rows but did not show up in the pellets were Lark Bunting {Calamospiza melanoconjs), great plains toad {Bufo cognatus), chorus frog {Psiiedacris triseriato), unidentified fish, and tiger salamander {Ambijstoma tigrinum). Discussion Burrowing Owls in southwestern South Dakota consume a wide variety of animals and some vegetation. Most other studies have reported similar results (Robertson 1929, Hamilton 1940, Bond 1942, Thomsen 1971, Marti 1974, Gleason and Craig 1981). Al- though Burrowing Owls are reported to be primarily insectivorous (Earhardt and John- son 1970), Gleason and Craig (1979) pointed out that, on a biomass basis, mammals may be more important. Thomsen (1971) found mammals to be more frequent than insects in Burrowing Owl pellets in California. The shift in frequency of mammals and in- sects in owl pellets between the May-June- July and August-September periods was the biggest difference in owl food habits in this study. Diet diversity (H', Shannon-Weiner in- dex, Pielou 1975) did not differ by much be- tween these two periods, 2.6 and 2.4, respec- tively, but diet breadth (Levins 1968) did, 4.7 and 2.1, respectively. Thus, owls consume a wider variety of prey during spring and early summer than during late summer and fall. Marti (1974) also reported a decrease in mammal consumption by Burrowing Owls in August and September. Errington and Bennet (1935) noted an increase in insect con- sumption in late summer and suggested that it was related to the fledging of young owls. Burrowing Owl feathers were frequently found in the pellets examined. Earhardt and Johnson (1977) cited studies that have report- ed Burrowing Owls to be cannibalistic. In this study only one or two owl feathers were encountered in a pellet, and they were usual- ly breast feathers, suggesting that the feathers were ingested while preening. Although plant parts were found in almost every pellet examined, they were primarily small plant fragments from prey stomachs. In April 1985 MacCracken et al Table 1. Mean percent frequency of occurrence' of foods .: Burrowing Owl 289 found in pellets regurgitated by Burrowing Owls in Conata casni, oum.. .^«.xv.v... Month May June July August September Food item Mammals 12.5 7.8 9.9 3.1 12.2 5.8 1.2 5.5 2.5 Peronnjscus spp. 2.7 0.4 1.2 Siikihif^ti.s floridamts 0.5 0.4 0.6 0.5 Micwtiis ochrooastcr 0.5 0.5 5.2 2.1 Spennophihis tridcccmUncatus Perognathus spp. 0.5 0.4 2.0 Geomijs hmsarius 0.6 Mijotis sp. 0.4 CAjnomiis hidovicianus 0.5 Mm muscuhis 1.9 1.2 Unidentifiable Birds 5.1 3.6 4.9 1.9 2.2 1.0 3.8 3.8 7.2 6.7 Athene euuietilarki- 0.5 0.5 Eremophila alpestris 2.6 1.2 0.5 Passerine 0.5 0.4 Unidentifiable 1.5 1.2 0.5 FIeptiles 1.2 Lizard 1.5 0.5 Snake 54.0 50.2 53.9 70.9 63.6 Abthbopods Coleoptera 16.6 7.4 9.5 12.7 11.1 Carabidae ,3.8 Chrysomelidae 0.5 0.6 10.0 Cincindelidae Coccinellidae 0.5 0.5 12.1 0.6 2.6 1.0 Curculionidae 10.5 12.4 3.7 9.6 Histeridae 6.2 0.5 2.1 0.4 2.5 1.4 Scarabidae Silphidae 1.5 11.2 0.6 7.9 2.7 16.9 0.5 12.0 Tenebrionidae 0.5 Larvae Diptera 0.4 Larvae Hemiptera 0.8 Unidentifiable Hymenoptera 0.5 1.9 5.4 2.6 12.9 Forniicidae 2.3 2.3 Ichneunionidae 0.5 Proctotmpidae 1.4 Sphecidae 0.8 Unidentifiable Lepidoptera 1.0 Larvae Orthoptera 10.9 2.7 1.8 16.9 12.5 Acrididae Araneida .3.1 10.4 6.5 2.5 Lycosidae 16.3 18.9 18.5 19.2 15.5 Vegetation 0.5 15.8 3.1 1.2 Grass 18.9 14.8 18.0 15.5 Plant fragments 0.6 Wood 0.6 Optuntia polijacantha seed 12.1 12.1 14.6 12.6 3.7 7.7 7.7 Miscellaneous 14.2 12.6 3.7 Bone PTrrrr cVirII 0.4 : 77 1 f^«™ rt*»llptt: 'Raspfi on number of iiema ict-uv^-.--^ ..— . j ^Based on the occurrence of feathers in owl pellets. See D>scuss.on. 290 Great Basin Naturalist Vol. 45, No. 2 some pellets we did find large pieces of grass and woody material that had been ingested directly by an owl. They could have been in- gested during efforts to capture and kill prey. Thomsen (1971) also frequently found vege- tation in Burrowing Owl pellets. Because some items consumed by Bur- rowing Owls do not show up in the pellets, prey remains at nest sites must also be exam- ined. Thomsen (1971) reported five prey items at nest sites that were not present in Burrowing Owl pellets. This absence could have at least two explanations: (1) it is not possible to find every pellet an owl produces, and (2) some foods or parts thereof may be completely digested. For example, we found that when Burrowing Owls ate great plains toads they typically placed the toad on its back and consumed the viscera and muscles of the legs, leaving the skeleton and other less digestible portions. The similarity of Burrowing Owl food habits among this study and others reported in the literature indicates that possible poi- soning of owl prey did not effect owl food habits. Acknowledgments W. Agnew provided the insect reference collection. R. T. Reynolds, D. Finch, and S. Anderson provided helpful comments con- cerning earlier drafts of this manuscript. Literature Cited Al'ffohth, a. K. 1981. The Burrowing Owl in North Da- kota. North Dakota Outdoors. April 1981. Bo.\n, R. M. 1942. Food of the Burrowinii Owl in west- ern Nevada. Condor 44:18.3. (]()iL()Nn!E, N. H.- 1971. Behavior and population eeolo- gv of the Burrowing Owl, Spcotiito cuniculdiid. in the Imperial Vallev of California. Condor 73:162-176. Day. M. G. 1966. Identification of hair and feather re- mains in the gut and faeces of stoats and weasels. Zool. Soc. London Proc. 148:201-217, E.\KH.\RDT. C. M., .\ND N. K. Johnson. 1970. Size di- morphism and food habits of North .\nierican Owls. Condor 72:251-264. Errincton, p. L., and L. J. Bennet. 19.35. Food habits of Burrowing Owls in northwestern Iowa. \\'ilson Bull. 47:125-128. Gleason, R. L., and T. H. Craic. 1979. Food habits of Burrowing Owls in southeastern Idaho. Great Ba- sin Nat. .39:274-276. Hamilton. W. J. 1940. A note on the food of the west- ern Burrowing Owl. Condor 43:74. Johnson, M. K., and R. M. Hansen. 1979. Coyote food habits on the Idaho National Engineering Labo- ratory. J. Wildl. Manage. 43:951-955. Levins, R. 1968. Evolution in changing environments. Princeton Univ. Press, Princeton, New Jersey. .Marti. C. D. 1974. Feeding ecology of four sympatric owls. Condor 76:45-61. Martin, D. 1973. Selected aspects of Burrowing Owl ecology and behavior. Condor 75:446-456. Merriam, C. 1902. The prairie dog of the Great Plains. Pages 257-270 in United States Department of ,\griculture Yearbook. Washington Government Printing Office, Washington, D.C. Moore, R. D., L. E. Spence, and C. E. Dlgnolle. 1974. Identification of the dorsal guard hairs of some mammals of Wvoming. \\'\ oming Game and Fish Dept. Bull. 14. ' Pielou, E. C. 1975. Ecological diversity. Wiley and Sons, New York, New York. Robertson, J. 1929. Some observations on feeding habits of the Burrowing Owl. Condor 31:38-39. Sperry, C. C. 1941. Burrowing Owls eat spadefoot toads. Wilson Bull. 53:45. ' Thomsen, L. 1971. Behavior and ecology of Burrowing Owls on the Oakland Municipal Airport. Condor 73:177-192. Uresk, D. W., J. G. MacCracken, and A. J. Bjucstad. 1982. Prairie dog density and cattle relationships. Pages 199-201 in Proceedings, 5th Great Plains Wildlife Damage Control Workshop. Lincoln, Nebraska. Whitney. N. R., B. E. Harrell, B. K. Harns, N. HoLDEN, J. W. Johnson, B. J. Rose, and P. F. Sprinc;er. 1978. The birds of South Dakota. South Dakota Ornithologists" Union, Vermillion, South Dakota. ADDENDUM TO THE DISTRIBUTION OF TWO HERPTILES IN IDAHO Timothy D. Reynolds' and William F. l.aiirancf .Vbstract.— Dtie to an error in printing (luality of an earlier article, the distribution maps for the night snake (W(//»ig/r/u; torqiiiitd) and the tiger salamander {Amhystoina tip^riniiui) in Idaho are reprinted. Recently we presented county by county distributions of the night snake {Hypsiglena torquata) and the tiger salamander (Ambijsto- ma tigrinum) in Idaho (Laurance and Rey- nolds 1984). Our objective in that article was to confirm the presence of these species with- in the Idaho ranges indicated by Stebbins (1966, map 5 and 175), and document any range extensions or gaps in distribution. The information was collected from the scientific literature, accounts of historical expeditions into Idaho, and interviews with various uni- versity, state agency, and qualified lay per- sons. The results were presented in two fig- ures and two tables. The figures were intended to graphically illustrate and com- Fig. 1. Range extension and previously documented range for Ambijstoma tigrinum in Idaho compared with the range illustrated in Stebbins (1966). Fig. 2. Range extension and previously documented range for Hypsiglena torquata in Idaho compared with the range illustrated in Stebbins (1966). 'Radiological and Environmental Sciences Laboratory, U.S. Department of Energy. 550 Second Street, Idaho Falls. Idaho 83401. -Museum of Vertebrate Zoology, University of California, 2.59.3 Life Sciences Building, Berkeley, California 94720. 291 292 Great Basin Naturalist Vol. 45, No. 2 pare (1) Stebbins' (1966) ranges for both the night snake and tiger salamander in Idaho, (2) previously documented accounts of these spe- cies in Idaho, and (3) new records of each for the state. Unfortunately, because of a printing error, the figures presenting Stebbins' (1966) ranges were reproduced so poorly in our paper (Laurance and Reynolds 1984) that the fun- damental objectives of that effort were com- promised. Here we again present Stebbins' (1966) ranges, the previously documented ac- coimts, and new records for the tiger sala- mander (Fig. 1) and night snake (Fig. 2) in Idaho. Our conclusions remain the same. First, the tiger salamanders observed and collected from Canyon and Ada counties in SW Idaho represent a significant westward range exten- sion for that species. Second, county by coun- ty documentation of both species within the ranges indicated by Stebbins (1966) is lack- ing. Last, the lack of records for the tiger salamander in several north central counties and for the night snake in the south central counties suggests either a true discontinuity in each population or the absence of diligent attempts to census herptiles in those counties. Additional efforts are needed to fully eluci- date the distribution of these and other herp- tiles in Idaho. We thank the editor of the Great Basin Naturalist for appreciating the fundamental importance of the printing error to our pre- vious article and consequently encouraging us to submit this addendum. Literature Cited Laurance, W. F., and T. D. Reynolds. 1984. Con- firmation and expansion of the reported distribu- tion of two species of Idaho herptiles. Great Ba- sin Nat. 44:313-316. Stebbins, R. C. 1966. A field guide to western reptiles and amphibians. Houghton Mifflin Co., Boston i-xiv -t- 279 pp. NESTING AND PREDATORY BEHAVIOR OF SOME TACHYSPHEX FROM THE WESTERN UNITED STATES (HYMENOPTERA: SPHECIDAE) Nancy B. Elliott' and Frank E. Kurczowski^ .\bstr\c T.— The first piihlislu'd observations on the nesting and predatory behavior of Tarhtfsphex antennatiis Fox, T. occidentalis Pulawski, 7". iciUianisi Bohart, T. ijoh Pulawski, T. alpestris Rohwer, T. chirconis Viereck, T. apricus Pulawski, and T. cockcrcUac Rohwer are presented herein. Also included are host and behavioral data on T. tarsatus (Say), T. apicaUs fiisiis Fox, T. similis Rohwer, T. asliiiieadii Fox, and T. intindiis exserttis Fox. The genus Tachysphex Kohl contains many rather small ground nesting species that uti- lize orthopterous prey. Nesting specifics are important in elucidating phylogenetic rela- tionships in this large genus (Kurczewski and Elliott 1978). Some of the common eastern species have been studied extensively, but many western Tachysphex have never been observed. Biological data heretofore have been published for only 9 of the 46 western species (Krombein et al. 1979, Pulawski 1982). Thus, studies on western Tachysphex are needed to fill gaps in our understanding of the phylogeny and evolution of this large genus. In this paper we bring together informa- tion accumulated on the behavior patterns of Tachysphex species from the western United States, along with prey records for several species based on museum specimens. In- cluded are the first published notes on T. an- tennatus Fox, T. occidentalis Pulawski, T. wiUiamsi Bohart, T. yolo Pulawski, T. al- pestris Rohwer, T. clarconis Viereck, T. apn- eas Pulawski, and T. cockerellae Rohwer. In preparing the paper we have followed the nomenclature used by Krombein et al. (1979) and Pulawski (1982). The species are listed alphabetically by species group as in Krom- bein et al. (1979). PoMPiLiFORMis Species Group Tachysphex antennatus Fox E. J. Kurczewski collected two females with prey at Erie, Pennsylvania, in 1981. Each made low, short flights. Both acridids were determined as nymphal Melanoplus sp. Tachysphex apricus Pulawski A paratype of this species, collected at Tucson, Arizona (W. Benedict), is pinned with a specimen of the phasmid ParabaciUus hespenis Hebard, more than four times its length. We report this record, although Pu- lawski (pers. comm.) questions its authenticity. Tachysphex occidentalis Pulawski A paratype of this species, collected at 5500 ft at Antelope Spring, California, by M. E. Irwin, is pinned with a nymph of the acri- did Schistocerca shoshone (Thomas). Tachysphex tarsatus (Say) Previus authors (Williams 1914, Evans 1970, Alcock and Gamboa 1975) reported fe- males making single-celled nests and storing one, rarely two, large acridid nymphs {Mela- noplus spp., Trimerotropis spp.). The first of two females collected at St. Anthony, Idaho, carried prey on the ground; the other carried prey in a series of short flights. We report the following prey records: Dissosteira Carolina (L.) (Walworth, Wisconsin; 17 July 1913); Melanoplus complanatipes Scudder (1 mi SW Tom's Place, Mono Co., California; 13 Au- gust 1963; C. A. Toschi); 2 Melanoplus sp. nymphs (Erie, Pennsylvania; 13 July, 25 'Department of Biology, Siena College, Loudonville, New York 12211. -Department of Environmental and Forest Biology, College of Environmental Science and Forestry, Syracuse, New York 13210. 293 294 Great Basin Naturalist Vol. 45, No. 2 August 1980; E. J. Kurczewski); Pseudopo- mala brachyptera (Scudder) nymph (Bath, Mason Co., Ilhnois; 3 July 1968; G. C. Gau- mer); Orphidella pelidna (Burmeister), male (Closter, New Jersey; 16 July 1963; M. Sta- tham). The last record is of the first adult prey reported for this species. Tachijsphex tvUliamsi Bohart A female collected at San Francisco, Cali- fornia, in May 1960 by J. A. Powell is pinned with a nymph of Trimerotropis occidentalis (Bnmer) no larger than she. Another wasp from Cornish, Utah (27 July 1973; G. Eick- wort and G. Bohart), identified as either this species or T. tarsatus, is pinned with a nymphal Melanophis sanguinipes (Fabr.) Tachijsphex yolo Pulawski One paratype was collected with a small nymph of Melanophis foediis Scudder (St. Anthony, Idaho; 15 July 1977). The prey was being carried on the ground. Terminatus Species Group Tachijsphex alpestris Rohwer A specimen from Morro Bay, California (J. A. Powell), is pinned with a nymphal acridid, Oeckdeonotus sp. Tachijsphex apicalis fiisiis Fox This species usually nests in sand cliffs (Kurczewski and Snyder 1968, Williams 1914), but Ran and Rau (1918) reported a fe- male attempting to nest in the mortar of a building foundation and another starting a nest from an antlion burrow. One of us (F. E. K.) has observed this species searching in the openings between the siding shingles of a cot- tage. A female was observed at St. Anthony, Idaho (23 June 1979), digging in the soft sand of a previous excavation. When first seen, she was throwing sand out of the entrance. Upon completing it, she left the entrance open, ori- ented in a hovering flight, and flew away. Twice she returned without prey, hovered near the entrance, and flew off. When re- turning to the nest with prey, .she landed nearby, although sometimes up to 1.5 m away, before resuming flight and carrying the prey directly into the open entrance. This wasp averaged 5.5 minutes between eight successive provisioning trips and 2 minutes within the nest between such trips. The unfinished nest contained a single cell at the end of a burrow 7.5 cm long. This cell contained seven nymphal acridids; the female had been carrying an eighth when collected. Prey were placed head inward, venter up- ward or head inward, on the side. The prey were identified as Melanophis sanguinipes (Fabr.). Williams (1914) reported that one female of this species was unable to fly with a large grasshopper. He excavated a single-celled nest with two prey, but Kurczewski and Sny- der (1968) reported that T. a. apicalis con- structs multicelled nests, and stores several prey per cell. Tachijsphex clarconis Viereck Four females were observed at St. An- thony, Idaho, during June 1979. One made 40 digging entries into the nest. The time spent inside increased as her digging progressed to- ward completion (x = 22.1 sec, first 10 en- tries; X = 31.3 sec, last 10), as did the time spent outside leveling sand (x = 12.2 sec, first 10; X = 29.0 sec, last 10). The female in- terrupted digging to chase awav a satellite fly- After removing and distributing the sand, this female walked around the entrance and threw sand back toward it. Periodically she took flight and hovered or perched on plants. After .3.5 min she reentered the nest and re- mained inside for 7.25 min. Then she tempo- rarily closed the entrance, hovered 7.5 cm above the nest for a few seconds, flew in a circle, and landed before flying away. Total observation time was 44 minutes. Prey were usually carried in flight directly to the nest area. A female would deposit her prey near the entrance, open and enter the nest, reappear headfirst, and pull in the grass- hopper by its antennae. One female depos- ited her prey for about 10 minutes while searching for her nest. Then she located and opened the nest and took in the prey as described. The same female was later seen April 1985 Elliott, Kurczewski: Wasp Behavior 295 carrying a rather large grasshopper in a series of short flights. During carriage the acridid's head protnided anteriorly beyond the wasp's. Three wasps stored from three to six grass- hoppers per cell, as indicated by successive provisioning trips. Three acridids were recov- ered from one of the cells. Prey included Melanoplm foedns Scudder, nymph (1), and M. sanguinipes (8). The only parasite observed, besides the miltogrammine fly, was a chrysidid, probably Hedi/cliridium sp., seen digging into one of tlie nests while the wasp was away. TaclujspJwx similis Rohwer Two nests were excavated at Wamego, Kansas, in July 1968. One had a single cell with three prey; the other, four cells contain- ing 3, 3, 5, and 3 prey. Prey used by T. similis at this locality included nymphs of Mela- noplus sp. (8), Memiirio sp. (1), Oplia oh- scurus (Tliomas) (2), and Pseudopomala bra- chijptero (Scudder) (8). Undatus Species Group Tachijsphex ashmeadii Fox This species was studied at Lakin, Kansas, in August 1964 and at St. Anthony, Idaho, in July 1977. We also include several new prey records from other collections. Our observa- tions and previously published data indicate a greater range of variability in this species than is characteristic of many other Tachijsphex. Females started digging nests either from the sand surface or from preexisting depres- sions such as animal tracks. Digging wasps entered the burrow headfirst and threw sand backward with the forelegs while backing out of the nest to distances of 1.6 to 3.2 (x = 2.3) cm. One female made 21 entries from in- ception to completion of her nest; another, 25. The two wasps averaged 56 (4-159) sec- onds inside the burrow and 16 (6-40) seconds outside during digging and sand removal, re- spectively. There were no changes in dura- tion of times as digging progressed. One fe- male, digging in extremely hot sand, frequently left the nest and flew to nearby vegetation. She flew backward out of the nest, throwing the sand behind her, and aver- aged only about half as long on the sand sur- face as had other females. Wasps left their entrances open while hunting. They also left intact the ovoidal- elongate tumuli that accumulated during dig- ging. One such tumulus was 5 cm long, 3 cm wide, and 0.5 cm high. Following completion of the burrow, females walked around their entrances and sometimes reentered. They then made low flights above the site and flew to nearby vegetation to hunt. Females made the most rapid hunting movements of all Tachijsphex we have observed. Periodically they returned without prey and reentered their burrows. One such female returned four times: 5, 29, 33, and 105 minutes after she first left to hunt. Some females dug within the entrances during these returns. Previous records (Williams 1914, Evans, pers. comm.) and many of our own observa- tions indicate that females of T. ashmeadii usually prey on large acridids in the late nymphal or adult stage and transport them to the nest on the ground. However, we have seen females carrying smaller prey (up to 2.2 times the wasp's weight) directly into the nest in flight, while holding the prey's anten- nal bases with the mandibles. Times between provisions ranged from 22 to 178 (x = 106; n = 5) minutes. Females using more than one prey per cell spent an average of 2.5 minutes inside their nests before leaving to hunt again. Nest dimensions for three burrows at La- kin, Kansas, were as follows: burrow length: x = 9.6 (8.9-10.4) cm; cell depth: x = 5.3 (5.0-5.9) cm. These nests were single celled, and each contained. a single grasshopper (weights: 59, 200, and 232 mg). Williams (1914) and Alcock and Gamboa (1975) also reported a single prey per cell, but two single-celled nests in Idaho each contained two prey. In one the egg was on the larger grasshopper, which was taken into the nest last. H. E. Evans (pers. comm.) excavated a nest of this species at Rodeo, New Mexico, and found a single prey bearing an egg in the distal burrow rather than in the cell. 296 Great Basin Naturalist Vol. 45, No. 2 Table 1. Frey oi Tacltiisplwx asliuic adii. Species State of c ollection Source or collector Aoeneotettix deorum (Scudder) KS, WY Williams 1914, Lavisrne and Pfadt 1966 Arphid sp. ID NBE Bnincria sordida (McNeill), nymph ID NBE CordiUacm crenulata (Bniner), adult KS Williams 1914, Krombein et al. 1979 Cordillacris occipitalis (Thomas) WY Lavigne and Pfadt 1966 Mekinoplti.s bivittatus (Say) ? KS FEK Melanoplua lakinus Scudder, nymph AZ P. Ranch Melanopliis sant^iiinipes (Fab.) ID NBE Mekmoplus sp., nymph CO, AZ G. C. Gaumer, G. and K. Eickwort Metotor sp., nymph — Krombein and Burks 1967 Opeia sp., nymph — Krombein and Burks 1967 Oi])huleU(i sp., adult TX R. E. Acciavatti Phlihostwma sp., adult — Krombein and Burks 1967 Twchijrhachijs kiowa (Thomas), male KS, WY Williams 1914. Lavigne and Pfadt 1966 Trimerotmpis sp., nymph KS FEK Trimerotropis bilobatu Rehn & Hebard. , male and nymph ID NBE Trimerotropis paUidipennis (Burm.), ad vdt and nymph TX J. E. Gillaspy Prey collected in Kansas were invariably placed venter upward, head inward, but Al- cock and Gamboa (1975) reported lateral placement from a nest in Arizona. Eggs were laid on either side, transversely between the prey's pro- and mesothoracic legs. Acridids from 11 genera have been preyed upon by this species (Table 1). Brullii Species Group Tachijsphex mitndus exsectus Fox One female was collected with a nymph of a tettigoniid, Conocephahis sp., at Wamego, Kansas, in July 1968 (G. C. Gaumer). Kur- czewski (1979) has described the nesting be- havior of this species, which also occasionally preys on gryllids. JuLLiANi Species Group Tachysphex cockerellae Rohwer A specimen from Napa Co., California (D. & W. Dunenmacher), is pinned with a nymph of the mantid, Litaneutria minor (Scudder). Discussion Krombein et al. (1979) divided the North American species of Tachysphex into five species groups, based on morphological cri- teria. Our studies, especially those on mem- bers of the Terminatus species group, show that behavioral traits may also characterize these groups. Members of the Terminatus group usually nest in flat sand, except for T. a. apicalis and T. apicalis fusus, which nest in sand cliffs (Kurczewski and Snyder 1968). All species construct multicelled nests which are provi- sioned with small, nymphal acridids. Since the prey are small, they are usually carried to the nest in flight, and several are used to stock each cell. Species in this group close the nest while hunting, after elaborately lev- eling the tumulus, except for the subspecies of T. apicalis, whose vertical nesting sites preclude this behavior (Kurczewski and Sny- der 1968). Nesting behavior of T. clarconis, reported here for the first time, supports its placement in the Terminatus group because it shares this set of nesting components. All previously studied members of the Brullii group prey upon orthopterans other than acridids. Tachijsphex akujoi Pulawski preys upon blattids (Pulaw.ski 1974, Elliott et al. 1979), and T. hclfrag,ei and T. mundus prey mostly upon tettigoniids (Krombein et al. 1979, Kurczewski 1979). Tachi/spliex co- qiiilletti Rohwer, the only previously studied North American species of the Julliani group, preys upon mantids (Alcock and Gamboa 1975). Our report of mantids as prey of April 1985 Elliott, Kurczewski: Wasp Behavior 297 T. cockerellae substantiates its placement in this group. The large Pompiliformis group now con- tains 44 North American species (Krombein et al. 1979, Pulawski 1982). Many species in this group also share a suite of similar behav- ioral traits, including the making of single- celled nests, utilizing one or a few large prey per cell, carrying prey on the ground, leaving the nest open while hunting, and not leveling the tumulus (Kurczewski 1964, Williams 1914). Although Krombein et al. (1979) placed T. ashmeadii in the Undatus group, components of its nesting behavior match very closely those described above. This sug- gests that T. ashmeadii has close affinities with the species in the Pompiliformis group. Studies on certain species in the Pompili- formis group, however, indicate considerable variation in behavior. Tachysphex pechumani Krombein demonstrates many of the group's characteristics but makes a Ridimentary tem- porary closure. Based on its morphology and behavior (Kiu-czewski et al. 1970, Kurczewski and Elliott 1978), we have suggested that this species occupies a unique phylogenetic posi- tion intermediate between the Terminatus and Pompiliformis species groups. Tachys- phex kromheini Kurczewski makes single- celled nests but stores several small acridids and tettigoniids, mixed, and carries them to the nest in flight (Kurczewski 1971). Tachys- phex kromheini is the only Nearctic species in the group tliat preys on families other than Acrididae. Should the record of T. apricus preying on phasmids be substantiated, this would further increase the range of prey fam- ilies reported for this group. We also report that females of T. yolo and T. wiUiamsi use rather small prey in comparison with many previously studied members of the group. It is probable that further studies of species in this group will identify affinities in behavior that separate the large group into several smaller ones. Pulawski (1982) noted that, morphologically, the group is less distinct than other species groups of Tachysphex and has suggested the grouping may be artificial. Acknowledgments We thank the following for use of their field notes and/ or prey records: R. E. Accia- vatti, U. S. Forest Service; G. C. Eickwort, Cornell University; H. E. Evans, Colorado State University; G. C. Gaumer, Houston, Texas; and E. J. Kurczewski, Erie, Pennsylva- nia. The following museum curators loaned us wasp specimens pinned with prey: S. Frommer, University of California, Riverside; L. L. Pechuman, Cornell University; R. R. Snelling, Los Angeles County Museum. W. J. Pulawski, California Academy of Sci- ences, determined the Tachysphex species. Prey were determined by A. B. Gurney, USDA, Systematic Entomology Laboratory; M. A. Brusven, University of Idaho; D. C. Rentz, CSIRO, Canberra, Australia; and I. J. Cantrall, Museum of Zoology, University of Michigan. The research in St. Anthony, Idaho, was fi- nanced, in part, by a grant from the Board of Trustees of Hartwick College (NBE); those in Lakin, Kansas, by NSF Postdoctoral Fellow- ship 44138 (FEK), and those in Wamego, Kansas, by NSF Predoctoral Traineeship 10- 307A (NBE). Literature Cited Alcock, J., AND G. Gamboa. 1975. The nesting behavior of some sphecid wasps of Arizona including Bem- bix, Microbemhex, and Philantbus. J. Arizona Acad. Sci. 10:160-165. Elliott, N., F. Kurczewski, S. Clafli.n, and P. Salbert. 1979. Preliminary annotated list of wasps of San Salvador Island, the Bahamas, with a new species of Cerceris (Hymenoptera: Tiphi- idae, Scoliidae, Vespidae, Pompilidae, Sphe- cidae). Proc. Entomol. Soc. Washington 81:352-365. Evans, H. E. 1970. Ecological-behavioral studies of the wasps of Jackson Hole, Wyoming. Bull. Mus. Comp. Zoo!. 140:451-511. Krombein, K. V. 1979. Family Larridae. Pages 1617-1650 in K. V. Krombein et al., eds.. Catalog of Hvmenoptera in .\merica North of Mexico. Vol. 2, Apocrita (Aculeata). Smithsonian In- stitution Press, Washington, D.C. Krombein, K. V., and B. D. Burks. 1967. Hymenoptera of America North of Mexico. Synoptic catalog. 2d suppl. USDA Agric. Monogr. 2:1-584. Kurczewski, F. E. 1964. A comparative, ethological study of some Nearctic digger wasps of the genus Tacbijspbex Kohl. (Hymenoptera, Sphecidae, Larrinae). Unpublished dissertation. Cornell Univ., Ithaca, New York. 277 pp. 1971. A new Tacbijspbex from Florida, with keys to the males and females of the Florida species (Hymenoptera: Sphecidae: Larrinae). Proc. Ento- mol. Soc. Washington 73:111-116. 298 Great Basin Naturalist Vol. 45, No. 2 1979. Nesting behavior of Tacliysplux mundiis Fox. (Hymenoptera, Sphecidae, Laiiinaei. Polskie Pismo Entomol. 49:641-647. KuRCZEWsKi, F. E., .-VNO N. B. Elliott. 1978. Nestint^ behavior and ecology of Tachijsphex pecluimani Krombein (Hymenoptera: Sphecidae). J. Kansas Entomol. Soc. 51:765-780. KiRczEwsKi, F. E., N. B. Elliott, and C. E. Vasey. 1970. A redescription of the female and descrip- tion of the male of Tachijsphex perhumdni (Hymenoptera: Sphecidae, Larrinae). .\nn. Ento- mol. Soc. Amer. 63:1594-1597. KuRCZEwsKi, F. E., AND N. F. R. Snyder. 1968. Evolu- tion of cliff-nesting in digger wasps. C:on- servationist 2.3:28-31. Lavicne. R. J.. AND R. E. Pfadt. 1966. Parasites and predators of ^^'voming rangeland grasshoppers. Univ. of Wvoming Agric. Expt. Sta. Sci. Monogr. 3:1-31. PvLAv\'SKi, W. J. 1974. \ revision of the Neotropical Tachysphcx Kohl. (Hym., Sphecidae). Polskie Pis- mo Entomol. 44:.3-80. 1982. New species of North American Tdchijs- phex wasps (Hymenoptera, Sphecidae). Proc. Cal- ifornia Acad. Sci. 43:27-42. Rau, p., and N. Ral. 1918. Wasp studies afield. Prince- ton Univ. Press. Princeton, New Jerse\ . x\ -I- 372 pp. \\ iLLiAMs. F. X. 1914. Monograph of the Larridae of Kansas. Kansas Univ. Sci. Bull. 8:119-213. POLLINATORS OF ASTRAGALUS MONOENSIS BARNEBY (FABACEAE): NEW HOST RECORDS; POTENTIAL IMPACT OF SHEEP GRAZING Evan A. Sudden' .Vbstract.— Important bee specios inhahifini; tlie study area are listed, includiiii; tliose oliserved and collected foratiinresson B. (I'yiohoiuluis) losncscnskii Hadoszkowski 8. VII. 79 14.VII.79 8.VII.79 7.VII.79 7w,2q S 3w S 4w,lq S Iw S MeCAC Hn.IDAK Anthiditiiu [Autliidiimi) chjpvodrnttituin Swenk Anthidiiiui sp. ('(dhinlhidiiiiii foniiosinn {(Bresson) Osiiiia {Monilosjuki) vara Cockerell O. {Xotlio.siiiid) griiidclidc Cockerell (Ispp. 17.VII 79 8.VII 79 14.VII 79 I4.VI1 79 7.VII 79 14.VII 79 14.VII 79 16.VII 79 17.VII 79 2m,2f S 1 0 lf,2m S 2f,lm s If s If s If 0 1 0 4 0 4 0 Anthophoridae Anthophora (Anthophoni) urhana tirhana Cresson Group II— Probable visitor* Apidae Apis ineUifera L. Bombus {Separatol70inhus) nionisoni Cresson Megachilhjae Hoplitis {MoniinictlKi) dUyifrons argciitifwns (Cresson) (Iroip III— Possible visitors Andrenidae Andrena (Tnichandrena) clcodora clcodora (Viereck) A. (Plastrandrciui) j)ruiioniui priinonnti Cockerell Colletidae CoUetcs consors consors Cresson Megachilidae Anthidium (Antliiditim) hanningense Cockerell A. (A.) monnonum Cresson A. (A.) tenuiflorae Cockerell Hoplitis (Monumetha) fulgida platyitra (Cockerell) Osmia (Chenosmia) calla Cockerell O. (Monilosmia) dcnsa densa Cresson O. (Acanthosinia) integra Cresson 304 Great Basin Naturalist Vol. 45, No. 2 Table 2. Unweighted host plant data for bee associates of Astnigaltis uwuocnsis Barnehy. Bee species listed alphabetically by group (see Table 1). Data for Apis mellifera omitted. Sources: Kronibein et al. (1979). Moldenke and Neff (1974), Thorp et al. (1983), personal observations reported in Appendi.\ I of this paper. Papilionaceous No. legume hosts as No. No. No. Astragalus proportion of total: Bee species families genera species species Genera Species Group I Anthidium chipcodcnUiiiiui 4 6 11 2 .50 .67 Bomhits litintii LS 39 39 1 .21 .21 B. nevadensis nccddcnsis 9 16 16 1 .38 .38 B. vosnesenskii 38 117 142 1 .09 .15 C(dhintltidiiiiii fDiinosui)} 6 9 9 1 .44 .44 O.smid c(ini 7 11 18 1 .27 .39 ('. iihudclidc 3 3 3 1 .33 .33 C.KOl P II Authopliow udnuia 3 103 174 1 .10 .11 B. nionisoni Ifi 3.5 35 1 .20 .20 Hoplitis alhifwns iirrsailles. INR.A, Paris. TwioH, D. \\\ 1981. C:hecklist of plants of the Mono iiasin. .Mono Basin Research Group (iontrib. 3. ISBN 0-939714-01-9. Ti I'l DiNo. V. J.. AND N. L. Stanton. 1980. Spatiotem- poral variation in phenology and abundance of April 1985 Sugden: Pollinators of Astragalus 311 floral resources on shortgrass prairie. Great Basin Nat. 40(.3): 197-215. Thorp, R. W. 1969. Ecology and behavior oi Autho- plioid ('(licdidsii (Hvnienoptera: .Apidae). Amer. Midi. Nat. 82(2):321-:337. TnoHP, H. \V., 1). S. HoRNiNC, Jr., and L. L. Dunning. 1983. Bumblebees and cuckoo bumblebees of California (Hymenoptera: Apidae). Bull. Califor- nia Insect Survey v. 23. 79 pp. Wilms, J. C. 1973. A dictionary of the flowering plants and ferns. VIII ed. Cambridge Univ. Press. Appendix I. New host records for bees associated with Astaiodlii.s inonocnsis Barneby. Location for all records: Mono Craters region. Mono County, California. Afitragalus monocnsis records included in text. Table 1. f = female, q = queen, m = male. Host plant abbreviations: C.p.v. = OinisothuDinus pdiryi vulcanictis (Greene) Hall & Clem. (Asteraceae), H.v. = Hulsca vestifa Gray (Asteraceae), L.d. = Lupimis duninii Eastw. (Fabaceae), P.f. = Phacclki frigida Greene (Hydrophyllaceae), E.u. = Eriou iitnbclkitum Torr. (Polygonaceae), C.u. = CdliiptmUum umbeUatiim (Torr.) Greene (Portulacaceae), M.c. = Mimiilu.s coccineus Congd. (Scrophulariaceae). Species Host plant Number, sex, Date or caste Andrcna (Trachandrena) rlcodow clcodom Viereck A. (Plastmndrcna) prunorum j)niui>nnu Cockerell .\nthophoridae Antlioplioid iirlnimi iiiIhiiui Cresson Apidae Bomhus (Pijrolw)nbus) hiintii Greene B. {^cpdiafobondius) niorrisoni Cresson E.u. E.u. E.u. (Pijwbombus) vosni'scnskii Radoszkowski Colletidae Colletes consors consors Cresson H.v. L.d. C.p.v. H.V. P.f. C.p.v. L.d. P.f. P.f. P.f. P.f. L.d. L.d. L.d. C.p.v. C.p.v. P.f. P.f. L.d. C.p.v. P.f. P.f. 28.VI.'81 2f 1.VII.-81 4f 28.VI.'81 2f 26.VI.'79 If 7.VII.79 If 8.VIII.79 If 8.VII.79 Iw 24.VII. 79 2w.lq 8.VIII.79 Iw 21.VIII.-81 Iw 13.VII.79 3w,lq 14.Vn.79 3w,lq 24.VII.79 Iw 8.VIII.79 2w 24.VII.79 4w 8.VIII.79 Iw 21.Vin.'81 4w 8.Vni.79 3w 21.VIII.'81 Iw 14.VII.79 Iw 24.VII.79 Iw 8.Vni.79 Iw 21.VIII.'81 Iw I3.Vn.79 3f 14.VII.79 3f Megachilidae Anthiditim (Anthidiuin) banningense Cockerell A. (A.) monnontim Cresson A. (A.) tenuiflorae Cockerell P.f. L.d. P.f. H.v. H.v. L.d. 14.VII.79 If 4.VII.'80 Im 4.VII.'80 If 8.VII.79 If 14.VII.79 If 14.VII.79 If 312 Appendix I continued. Great Basin Naturalist Vol. 45, No. 2 Species CoUanthidiiim formosiim (Cresson) L.d. C.p.v. Hoplitis (Momimetha) albifrons aroentifrons (Cresson) ^^•<-- Osmia {Chenosmia) ccilki Cockerell L.d. O. iMonilosmia) dcnsci demo Cresson P-t- L.d. O. (Nothosmio) ghndeliae CockereW ^1-c. O. (Acanthosmia) integra Cresson ^^c. Host Number, sex, plant Date or caste 24.VII.79 If 21.VIII.81 Im 1.5. VII. -80 If 8.VI.-79 5f 24.VII.-79 If 13.VII.79 If 9.VI.-80 If 9.VI.'80 If VEGETATIONAL AND GEOMORPHIC CHANGE ON SNOW AVALANCHE PATHS, GLACIER NATIONAL PARK, MONTANA, USA David R. Butler' .\bstr.ac:t.— Sl\ suhalpine snow avalanche paths studied in 1975 were revisited in the summer of 1983, with the purpose of examining geomorphic and vegetational change that may have occurred during this eight-year period. Repeat photography and field reconnaissance were used to assess vegetational and geomorphic change. Vegetational responses to avalanches were apparent on several of the avalanche paths, generally bv an increase in brush cover. Geomorphic changes were not apparent, suggesting that avalanches need not be geomorphically effective to initiate substantial vegetative disruption. Long-term records of geomorphic and veg- etational change in subalpine and alpine en- vironments of North America are scarce, be- cause of the limited period of historical settlement and problems of inaccessibility. Recent studies have reported geomorphic change on alpine debris slopes based on peri- ods of observation ranging from 7 to 15 years (Gardner 1979, 1982, 1983a, Luckman 1981). Subalpine slopes, however, have been rela- tively ignored in long-term studies. In 1975, Butler examined the general vege- tative conditions and geomorphic processes active on 12 subalpine avalanche paths in Glacier National Park, Montana, USA (Butler 1979). The intent of the study presented here is to describe the geomorphic and vegeta- tional changes that took place on six of these snow avalanche paths during the eight-year interval from 1975 to 1983. The occurrence and synchroneity of large-scale destructive avalanches were of particular interest. Sites MV3, MV4, SN5, SN6, SN7, and SN8 were revisited in 1983 (Fig. 1). MV3 and MV4 had also been revisited briefly in 1981. These paths all impinge on either a highway (MV3 and MV4 in the McDonald Creek Val- ley), a heavily used foot trail (SN6, SN7, and SN8 in the Snyder Creek Valley), or a popu- lar backcountry campsite (SN5 in the Snyder Creek Valley), and thus major avalanche events are noted by National Park personnel. These six avalanche paths were selected for further study because: (1) some had tree-ring and/ or highway and trail crew maintenance records of avalanche frequency and magni- tude (MV3, MV4, SN5, and SN7); (2) some were encountered enroute to the above paths and had similar characteristics of vegetation and site conditions (SN6 and SN8); and (3) time limitations restricted revisitation to the most easily accessible paths. Detailed site de- scriptions of the six revisited paths may be found in Butler (1979). Tree species present on the drier ava- lanche paths of the Snyder Creek Valley in- clude Abies lasiocarpa and a few Psetidotsuga menziesii, whereas the paths of the McDon- ald Valley support Betula papijrifera and Picea engehnannii. Flexible-stemmed shrubs and small trees, able to withstand avalanche impact pressures of up to about 10 t m-2, in- clude Acer glabrum, Alnus spp., Sorbus scopulina, and Crataegus douglassii. Acer and Alnus are most common (Butler 1979). Small- er, flexible berry bushes (e.g., Vaccinium) are also common. Methods Repeat Photography Repeat photography allows qualitative as- sessment of geomorphic and vegetational changes that may occur over a given interval. Photographs taken during field work in 1975 were repeated in 1981 for MV3 and MV4 and on all six sample paths in 1983. Photo- graphs were taken looking up-path from the farthest down-path point (in the case of SN5, 'Department of Geography, Oklahoma State I'l itv, Stillwater. Oklahoma 74078. 313 314 Great Basin Naturalist Vol. 45, No. 2 HIGHWAY TRAIL CAMPSITE 2 miles ^ 0 0 12 3 kilometers /A N (T Fi<4. 1. Map ot tlu' study a the avalaiulu' patlis niciitiom Montana and tlu- (Canadian-USA horcl tlu" contial portion ol C he text. The inset map si heavy bnish cover dictated that photographs be taken about 100 m from the path, on the opposite valley side). Figure 2 illustrates a typical subalpine avalanche path (SN8) as it appeared in 1983. Dendrogeomorphology Butler (1979) tagged 30 trees in the runout zone of path SN7 in 1975. Relocation of these trees and assessment of their condition would indicate whether a large-scale destruc- tive avalanche had occurred during the inter- val between visits. Additional tree-ring dat- ing of high magnitude avalanche events had been planned for paths SN5, SN6, and SN8. Access to Snyder Creek drainage was closed, however, on 24 June 1983, because of the presence of two families of grizzly bears (Ursus arctos horribilis). The drainage was April 1985 Butler: Avalanche Path Vegetation 315 .^^4Yt>^fVv^ Fi 0.02 83 Number aquatic adults 0.377 0.142 < 0.005 73 Nmnber aquatic larvae 0.027 _ > 0..50 100 Number C. stolonifera 0.241 — > 0.05 50 Number E. doddsi 0.489 0.239 < 0.001 92 Number Formicidae 0.172 _ > 0.20 73 Mean length all food 0.381 0.145 < 0.005 _ Mean length aquatic larvae 0.472 0.223 < 0.001 _ Number taxa 0.342 0.117 < 0.01 _ Range of length 0.286 0.082 < 0.02 - later analysis. All items in the stomachs were measured mider a dissecting microscope and identified to the lowest taxonomic level prac- tical (terrestrial animals other than Formi- cidae were lumped together, but most aquat- ic forms were taken to genus or species). To enable the use of correlation analysis, data were transformed to logarithmic scale prior to statistical analysis (Zar 1974). Results Over 30 taxa, ranging in length from 2 to 70 mm, were recovered from stomachs of the 48 trout (see Appendix). The strong correla- tion for total number of prey and trout length (Table 1) suggests that overall food consumption is dependent on trout size. This is also true for consumption of EphemereUa doddsi larvae, the most frequently occurring aquatic insect larvae found in the stomachs of the cutthroat trout sampled. However, con- sumption of total number of organisms of ter- restrial origin, number of berries from red- stem dogwood {Corniis stolonifera), number of ants (Formicidae), and number of aquatic insect larvae were not related to trout size. Some of the berries from redstem dogwood did appear partially digested, indicating that trout may be gaining .some nutritional value from them. It is possible that correlation analysis was redundant for the number of ants and num- ber of organisms of terrestrial origin and for the number of E. doddsi and number of aquatic insect larvae. Although no hidden relationship was detected for the ant com- ponent, the relationship between trout length and £. doddsi would have been missed had one relied .solely on the aquatic larvae variable. Mean length and the range of length (long- est minus shortest) of aquatic insect larvae in- gested by cutthroat trout were also related to trout length. As cutthroat trout from Pali- sades Creek grow, it appears that they rely on larger food, but still take smaller food items (Fig. 1). Cutthroat trout from Palisades Creek did not consume prey shorter than 2 mm, whereas almost 94% had a minimum prey length of at least 3 mm. With increasing trout length, the number of different food items increased. Apparently, as they grow, cutthroat trout in Palisades Creek do not shift entirely to larger prey or to a narrower range of taxa. Since certain variables (e.g., number of aquatic larvae and number of organisms of terrestrial origin) were shown to be unrelated to trout length, one cannot automatically as- sume the positive relationships found for other variables and trout length are allome- tric, with the possible exception of total food item number. Since large prev were found in the stomachs of small trout, although in- frequently, it is not possible to assume that the relationship between mean prey length and trout length is strictly a function of trout (mouth) size. These observations are based on a small sample of trout, taken from only one season; consequently, it is difficult to extrapolate these findings to the .same population during seasons having different benthic invertebrate abundances and diversities. More research into the.se variables appears warranted. D ISCUSSION Prey size as a factor affecting the feeding of fish has been suggested for salmonids (Ware 1972) and strongly indicated in the April 1985 40 30- 20- 10- Skinner: Trout Food Selec:tion 329 ID Fig. 1. Histogram of minimum lengths of food con- sumed and their frequency of occurrence in the 48 cut- throat trout stomachs sampled. ecology of other fishes (e.g., Lepomis sp. in Werner and Hall 1974). Ware (1972), in labo- ratory experiments, has shown reaction dis- tances by rainbow trout of lake origin to in- crease with increasing prey size. Ware (1973) included prey size in a model formed for pre- dicting prey vulnerability to predation by rainbow trout, and he found the intensity of predation should closely parallel the change in prey size. As shown, the mean length of food taken by cutthroat trout from Palisades Creek dur- ing the smiimer increased with trout length. Because small trout did have the ability to consume comparatively large prey, the ques- tion of why they don't take more of them arises. Smaller trout may spend too much energy in capturing, handling, and ingesting large food. If so, this would preclude them from taking great numbers of large prey, yet not entirely prevent an occasional large in- vertebrate from being taken. It may also be that trout are intimidated by large prey and avoid them. Large bodies may be representa- tive of predators, and tliis may explain why there is a linear relationship between food size and trout size. A third possible cause could be social interaction within the trout population. Large trout may be dominant over smaller trout, enabling first choice of valuable food items (e.g., large prey) to be given to large individuals. Ringler (1979) found brown trout in an ar- tificial stream to consume small prey, even when larger prey were available. Increasing the size of prey and the assortment of prey types without reducing the range in prey length, as observed for cutthroat trout in the present study, may improve the energy gain for an individual fish. Overall energy expense should increase with body size; hence, large trout, especially drift feeders, would be at a disadvantage if they restricted their intake. Bisson (1978) postulates that smaller prey should at least repay the energy spent in their consumption by trout. Prey lengths of about 2 mm have been sug- gested as minimum detection levels for rain- bow trout, below which prey cannot be dis- tinguished from other drifting material (Bisson 1978). The results of this study on cutthroat trout support this postulate. How- ever, since lengths of members of the prey populations in Palisades Creek were not taken during the present study, it is difficult to speculate on the lowest possible lengths of invertebrates available. Benthic samples taken approximately two weeks earlier re- vealed large abundances of invertebrates of 1 to 3 mm sizes, indicating that smaller prey may have been present. This minimum length of detection hypothesis suggests that only prey larger than about 2 mm be used in estimating the available cutthroat trout food in a stream. The strong correlation between the num- ber of E. doddsi larvae and trout length is most interesting. This species is not a notable drifter in Palisades Creek (unpublished obser- vations by author), and it may be possible that the majority of E. doddsi were taken from the stream bottom. If so, then as cut- throat trout get larger, they tend to feed more from the stream bottom than the drift. Tippets and Moyle (1979) observed a similar trend in the feeding of rainbow trout from the McCloud River, because larger trout con- tained food items more common to the ben- thos than the drift. Simpson and Wallace (1978) suggest that terrestrial organisms form a major portion of the diets of cutthroat trout in Idaho. McMas- ters (1970) sampled cutthroat trout during July from a similar size stream in south- eastern Idaho and found 33% to have fed upon terrestrial organisms. Of the trout sam- pled in the present study, 83% fed upon ter- restrial organisms (Table 1). It is also inter- esting that so many trout (73%) had taken 330 Great Basin Naturalist Vol. 45, No. 2 ants. Jenkins (1969) made use of marked ants in an experimental manipulation of drifting food and the feeding activities of rainbow and brown trout. The use of a prey organism in such manipulative studies should be predi- cated upon its natural occurrence in the diet of the fish being studied. Results from the present study indicate that ants are a com- mon item in cutthroat trout diets; hence, ad- aptation of Jenkins's (1969) study to cutthroat trout should be feasible. The occurrence of berries from redstem dogwood {Comas stolonifera) in the cutthroat trout stomachs indicates an omnivorous trait of this species. Since half the trout sampled contained this food item, it seems easy to conclude that a generalist attitude exists to- ward feeding by cutthroat trout. However, optimal foraging theory (Krebs 1978) in- dicates that it may be more advantageous for trout to consume the vegetable than the in- vertebrate portion of their diet, depending upon caloric or nutrient value of the respec- tive items and the energy spent in obtaining them. Since the number of berries from red- stem dogwood apparently was not related to trout size, it is not reasonable to say that larger fish may benefit more than smaller fish by feeding on this item. The data from the present study indicates that food size selection is occurring in a pop- ulation of stream cutthroat trout. Size selec- tion may be an important mechanism in the feeding of this species; however, it must be noted that, although the r values were signifi- cant—in some cases p < 0.001— the r^ values were relatively low (0.145 for mean length of all prey and 0.222 for mean length of aquatic larvae). This suggests that other factors, pos- sibly of equal importance, are operating on this interaction. For instance. Otto and Sjorstrom (1984) suggest morphology of stonefly larvae and the role of cerci and an- tennae in modifying their predator-prey rela- tionship with first-year brown trout. Irvine and Northcotte (1984) observed rainbow trout fry in artificial stream channels to feed significantly more from groups of live Simuli- um sp. larvae than dead Simiiliuni sp. larvae, a .situation attributed to .some invertebrate behavior that could not be exliibited by dead Simulium sp.. An interaction may also exist among prey size, morphology, and behavior. Acknowledgments Field assistance was generously provided by Dennis E. and Rhoda T. Skinner, and lab- oratory work was aided by Rhoda T. and Kimberely A. Skinner. Dr. James D. Mclver identified the Formicidae and Dr. Karl E. Holte identified the Corniis stolonifera; both are faculty at Idaho State University. Con- structive criticism of the manuscript was pro- vided by Drs. G. Wayne Minshall of Idaho State University and James R. Barnes of Brig- ham Young University. Appendix Assortment of food items found in the stomachs of 48 cutthroat trout taken from Palisades Creek, Bonneville County, Idaho. Eplienieioptera Baetidae Baetis sp. larvae and adult Siphlomiridae Amck'tus sp. Heptageniidae Epcorus sp. Cinij^mtila sp. Ephemerellidae EphemerelUi colonidcusis E. doddsi larvae and emerging adults E. incrmis Plecoptera Perlidae Acwncuria sp. Ferlodidae Mc^^anijs sp. Skinihi sp. larvae and adult Neniouridae 7Mp(id(i sp. C^hloioperlidae Trichoptera Linniephilidae S'colhrcimiKi sp. (dossossoniatidae CUossossoiud sp. R>ae()philidae Hijdcophihi sp. Hydropsychidae Arctopsyche sp. April 1985 Skinner: Trout Food Selecti 331 Odonata C;oiiiphidae Dipteia Tipulidae Antocliii sp. Tipiild sp. Di(i(im>t(i sp. Siinulidae larvat' and pupae Chirononiidae C^oleoptera Elmidae larvae and adult Dvtiscidae Terrestrial Formicidae Hynienoptera Diptera others Terrestrial vegetation Corniis ■stolonifcra Fishes Cottidae Cottiis bairdi Literature Cited .\llan, J. D. 1983. Predator-prey interactions in streams. Pages 161-191 in J. R. Barnes and G. W. Min- shall, eds.. Stream ecology: testing of general eco- logical theory. Plenum Publ. Corp., New York. 1978. Trout predation and the size composition of stream drift. Limnol. Oceanogr. 23:1231-1237. 1981. Determinants of diet of brook trout {Salvc- liniis fontinalis) in a mountain stream. Can. J. Fish. Aquat. Sci. 38:184-192. BissoN, p. .\. 1978. Diel food selection by two sizes of rainbow trout {Salnio gaiidneri) in an experimen- tal stream. J. Fish. Res. Bd. Canada 35:971-975. Elliott, J. M. 1967. The food of trout (Sobno tnitta) in a Dartmoor stream. J. App. Ecol. 4:59-71. 1973. The food of brown trout and rainbow trout {Salmo tnitta and Salmo gairdneri) in relation to the abundance of drifting invertebrates in a mountain stream. Oecolog. Berlin 12:329-347. Fleener, G. C. 1952. Life hi.story of the cutthroat trout, Salmo chirki Ricliaidsoiii in Logan River, Utah. Trans. Amer. Fish. Soc. 81:235-248. CRnFiTii, J. S., Jn. 1974. Utilization of invertebrate drift by brook trout (Scdvclintis fontimdis) and cut- throat trout (S(dmo clarki) in small streams in Idaho. Trans. Amer. Fish. Soc. 103:440-447. Hunt, P. C, .\nd J. W. Jones. 1972. The food of brown trout in Llyn Alaw, Anglesev, North Wales. J. Fish. Biol. 4:333-352. IiwiNE, J. R., AND T. G. NoRTnc;oTTE. 1983. Selection bv young Rainbow trout {Scdmo gairdneri) in simu- lated stream environments for live and dead prey of different sizes. Can. J. Fish. Aquat. Sci. 40:1745-1749. Jenkins, T. M., Jr. 1969. Night feeding of brown trout and rainbow trout in an experimental stream channel. J. Fi.sh. Res. Bd. Canada 26:3275-3288. Krebs, J. R. 1978. Optimal foraging: decision rules for predators. Pages 23-63 in Krebs, J. R. and N. B. Davies, eds.. Behavioural ecology: an evolution- ary approach. Blackwell Sci. Publ., Oxford. McMasters, M. J. 1970. The food habits of trout (Salmonidae) and sculpins (Cottidae) in two mountain streams. Unpublished thesis. Idaho State Univ. 123 pp. Otto, C, and P. Sjorstrom. 1983. Cerci as anti- predatory attributes in stonetlv nvmphs. Oikos 41:200-204. Reed, E. B., and G. Bear. 1966. Benthic animals and food eaten by brook trout in Archuleta Creek, Colorado. Hydrobiology 27:227-237. Ringler, N. H. 1979. Prey selection by drift feeding brown trout {Salmo trutta). J. Fi.sh. Res. Bd. Can- ada 36:392-403. Scott, W. B., and E. J. Grossman. 1973. Freshwater fi.shes of Canada. Bulletin 184, Fish. Res. Bd. Canada, Ottawa. 966 pp. Simpson, J. C, and R. L. Wallace. 1978. Fishes of Idaho. Univ. Press of Idaho, Moscow. 237 pp. Thomas, J. D. 1964. Studies on the growth of trout, Sal- mo trutta, from four contrasting habitats. Proc. Zool. Soc. London 142:459-509. " Tippets, W. E., and P. B. Moyle. 1978. Epibenthic feeding by rainbow trout {Salmo gairdneri) in the McCloud River, California. J. .\nim. Ecol. 47:549-559. Ware, D. M. 1972. Predation by rainbow trout {Salmo gairdneri): the influence of hunger, prey density and prev size. J. Fish. Res. Bd. Canada 29:1193-1201. 1973. Risk of epibenthic prey to predation by rainbow trout {Sahno gairdneri). J. Fish. Res. Bd. Canada 30:787-797. Werner, E. E., and D. J. Hall. 1974. Optimal foraging and the size selection of prey by the bluegill sun- fish {Lepomis maerochirtis). Ecology 55:1042-1052. Zar, J. H. 1974. Biostatistical analysis. Prentice Hall, Inc. 620 pp. ASPECTS OF THE BIOLOGY OF THE FLATHEAD CHUB {HYBOPSIS GRACILIS) IN MONTANA William Gould' Abstract.- Maturo flathead chubs (Hijhopsis gmciUs) were present in mid-July and mid-August collections from the Musselshell River, Montana. The estimated numbers of mature eggs present in eight females were 360-753 per female. The smallest mature female and male collected were 113 and 123 mm in total length, respectively. The male to female sex ratio in collections was about 1:1. Only small differences were detected among the length-weight relationships of males and females and samples taken from various seasons and localities in Montana. Observations on size groups, fish a.s.sociates, and habitat characteristics of flathead chubs are presented. The life histories of several species of Hij- hopsis are poorly known (Lee et al. 1980). One of these is the flathead chub, Hijhopsis gracilis (Cross 1967, McPhail and Lindsey 1970, Brown 1971, Scott and Crossman 1973, Pflieger 1975 and Lee et al. 1980). Most of the published information on the natural his- tory of this species in the United States is contained in a systematic study by Olund and Cross (1961) and a bionomics study by Mar- tyn and Schmulbach (1978). This report pre- sents additional biological information on the flathead chub. Description of the Study Site The collection site was on the Musselshell River (T8N R25E S22) in central Montana. At this location the river is a plains stream having an elevation of about 971 m and a substrate of sand- and silt-covered pebbles. Records taken at the collection site over a four-year period (USGS 1979, 1980, 1981, 1982), showed the pH range was 7.7-8.5, with 90% of the measurements being 8.0 or greater. Average monthly values were 100-1700 mg/1 for suspended solids, 240-830 mg/1 (as CaCO,) for alkalinities, and 4-31 mVs for flows. Flows were usually highest in May or June and lowest in August or Septem- ber. Water temperatures ranged from 0.0 to 23.0 C, with lows occurring from December through February and annual highs from June through August. Flathead chub typically inhabit fluctuating streams with alkaline, tur- bid waters (Olund and Cross 1961, Brown 1971). Methods and Materials Specimens were collected with an 11.0 X 3.7 m seine having an 8-mm-square mesh and preserved in 10% formalin. In addition, col- lections in the Vertebrate Museum of Mon- tana State University (MSU) were examined. The total length (TL) of each specimen was measured to the nearest 1 mm; standard length (SL) was derived from TL in the linear regression SL = 0.2665 + 0.7863 TL, which was obtained from measurements of 65 speci- mens 70-134 mm TL. The weight of each blot-dried specimen was determined to the nearest 0.01 g on a Mettler Model H16 bal- ance. The sex of each fish was determined by examination of the gonads under a dissecting microscope. Egg size was measured with an ocular scale calibrated with a stage microme- ter. The total number of mature eggs in a fish was estimated by using: Mt = where Mj = the total number of mature eggs in the fish's ovaries, Wj = the total weight of the fish's ovaries, M^ = the num- ber of mature eggs in the sample of the ovary, and Ws = the weight of the sample of 'Cooperative Fishery Research Unit, Biology Department, Montana State University, Bozeman, Montana 59717. Cooperators are the U.S. Fish and WildUfe Service, Montana State University, and the Montana Department of Fish. W ildhfc and P.irks. 332 April 1985 Gould: Montana Flathead Chub 333 Table 1. Tlie average diameter (mm) of 10 of the largest eggs in 58 specimens of Hijbopsis i^nicilis collect- ed from the Musselshell River, Montana. Average Collection Total length egg diameter in mm date offish (mm) (numbers of specimens) 31 Mar 81 71-123 0.2(12), 0.3(2) 0.5(1), 0.7(7) 0.8(2) 19 Jul 83 113-155 0.2(4), 0.3(1) 0.4(2), 0.5(3) 0.6(2), 0.8(1) 1.0(1), 1.1(3) 1.2(1), 1.3(1) 1.4(1) 15 Aug 83 117-160 0.2(1), 0.3(1) 0.4(3), 0.5(1) 0.6(3), 1.3(1) 11 Nov 82 114-134 0.2(1), 0.6(2) 0.7(1) the ovary. Two estimates were made of the Mt in each fish. The length-weight relation- ships in specimens were determined by using login w = log a -I- b (log„>L), where w = blot-dried weight in grams, L = total length in millimeters, and a and b were constants derived from the data. Results Reproductive Condition of Females Fifty-eight females were collected prior to, during, and after the presumptive spawning season for an evaluation of their reproductive condition (Table 1). Females with the largest eggs (1.0-1.4 mm in diameter) were found in July and August 1983 collections. These spec- imens were considered mature because of the large size and orange color of the eggs. Ex- amination of 23 females from eight collec- tions at MSU showed that females with eggs of a comparable size were present only in July samples; none were present in August and September collections. The average daily maximum and minimum water temperatures associated with the collection of mature fe- males in this study were 23 and 18 C in July and 25 and 21 C during the first 15 days in August (unpublished data, Montana Depart- ment of Fish, Wildlife, and Parks). The estimated average number of mature eggs (1.0 mm or greater in diameter) in the eight mature specimens collected during this study was about 491 (Table 2). There was no trend for larger females to have more eggs than smaller ones within the size range treated (r = 0.15). The smallest mature female collected (Table 2) was 113 mm TL (89 mm SL). The smallest mature female found in nine museum collections of MSU was 112 mm TL (88 mm SL). The ovaries of the eight mature fish in Table 2 weighed 0.59-0.99 g and were 2.3-5.9% of total body weights. In the 22 im- mature females collected concurrently with the mature fish in this study, ovary weights were 0.10-0.71 g and 0.5-1.8% of total body weights. Reproductive Condition of Males Milt was stripped from 13 males taken in July and August 1983 (Table 3). The smallest ripe male was 123 mm TL (97 mm SL). How- ever, some males larger than this were not ripe in August. Table 2. Estimated numbers of mature eggs in eight Hijbopsis gracilis collected from the Musselshell River, Montana. Estimated total Average numbers of diameter Collection Total length Total weight mature eggs of 10 mature date of fish (mm) of fish (g) from two samples eggs (mm) 19 Jul 83 113 12.99 442/293 1.0 120 16.72 453/521 1.1 122 17.55 508/524 1.1 124 18.50 539/446 1.3 130 20.04 483/441 1.2 136 22.95 6.38/633 1.1 155 36.02 753/360 1.4 15 Aug 83 160 37.72 453/372 1.3 334 Great Basin Naturalist Vol. 45, No. 2 Table 3. The reproductive condition of 18 male Hy- bopsis gracilis collected from the Musselshell River, Montana. Numbers of specimens in parentheses. Collection Total length of males in mm date Ripe Not ripe 19 Jul 83 15 Aug 83 123, 128. 129 132, 133, 143 127. 132, 135 (2), 136 (2), 146 122, 124, 127 132, 140 Sex Ratio The male to female sex ratios of specimens examined were not significantly different from 1:1 (Table 4). This ratio did not change significantly with increases in the size of specimens examined. Length-Weight Relationships The length-weight relationships of a sample of males and females taken 31 March 1981 were calculated separately (Table 5). An F test of the slope and intercept showed the two regressions were not statistically dif- ferent (P = 0.66), so length-weight data from all fish were combined. The length-weight relationships of flathead chubs collected during different seasons and from different localities in Montana were similar (Table 5). Tlie relationship was also similar among different size groups. Fish Associates Fish captured with H. gracilis in the Mus- selshell River were Coiiesius plumheus, Cijpr- inus carpio, Hyhognathus argyritis, H. pla- citus, Carpiodes carpio, Catostomus com- mersoni, C. platyrhynchus, Moxostoma macr- olepidotum, Microptertis dolomieui, and Noturus flavus. Hybognathus placitus and C. commersoni were reported previously by Olund and Cross (1961) as associates of flat- head chubs. Age and Growth The length-frequency analysis of 305 speci- mens 29-127 mm TL collected from the Musselshell River 31 March 1981 indicated three size groups were present. The approx- imate average total length of specimens in each size group was 43 (N = 116), 81 (N = 149), and 116 (N = 40) mm. Attempts to verify these size groups as age groups by ex- amination of scales, opercula, and vertebrae proved unsuccessful. Discussion Most of the characteristics studied in flat- head chubs from Montana were similar to those reported for the species in the midwes- tern U.S. The July and August spawning sea- son for flathead chubs in Montana was the same as has been reported for this species in Kansas (Cross 1967) and Iowa (Martyn and Schmulbach 1978), but it is more extended than the July season suggested by Brown (1971) and reported by Bishop (1975) for Montana and Peace River, Canada speci- mens. The water temperatures recorded dur- ing the spawning season of the flathead chub in Montana were virtually the same as those Table 4. The niunbers of males and females in samples of Hijhopsis gracilis fiom Montana. Total Number Number Sources of len^th Sample of of C; ilculate specimens (mm) size males females X- Musselshell River, this study Cx)llccti()us in MSU nmseuni Composite of above sources Tabular X^ = 3.84 at P = 0.05 with one degree of freedom. 71-171 79-154 71-171 75 53 128 38 25 63 37 28 0.007 0.085 0.016 April 1985 Gould: Montana Flathead Chub 335 Table 5. The length-weight relationships in collections of Hifhopsis v / / I ^ ^ — -^ ^^^ L. californicus c / -Vv\/ / Jy^^ ij // ^^o: ^..-^^ ^ k_ 1/ /■Vi--' '^ ,,^ L. americanjs TO A X / r^ — Q) 12- /I yy S. nuttallii/ y\,y^ audubonii < ^ ■ ^ X X y y^^- idahoensis ^ X X xxxxxxxx X XXXX XX ^ 12 16 20 24 Diastema Length Fig. 4. Plot of Sylvilagus and Lepus dentaries from Crystal Ball Cave (X's) and ranges of variation for all species of leporids presently living in and near Utah and Nevada (circumscribed). Some of the measurements of Recent specimens were made by the author from the Brigham Young University Monte L. Bean Museum mammal collection, andsome were provided by J. A. White(1984, pers. comm.). Thenumberof Recent specimens measured were 31 of S. idahoensis, 22ofS. nuttallii, 33of S. audubonii, l2ofS.floridanus, 40ofL. americanus, 36ofL. californicus, and 29 of L. townsendii. Symbols on the plot margins represent Crystal Ball Cave specimens on which only one of the two plotted measurements could be made. Measurements are in millimeters. cent S. idahoensis specimens to which it was compared (but smaller and distinct from other species of leporids), and the other Crystal Ball Cave specimens are also comparatively large, suggesting that this species may have de- creased in size at the end of the Pleistocene. This species presently lives in the region of the cave and to the north and west (Hall 1951, 1981). Sylvilagus nuttallii Material. — Anterior portion of right den- tary without teeth (BYUVP 5493), left dentary with M/1 (LACM 123658), anterior portion of left dentary with I/l, P/3,/4, M/1 (BYUVP 5578), 4 right P/3's (BYUVP 5717, 5731, 5769, 5794), 4 left P/3's (BYUVP 5773, 5782, 5795, 5810). Discussion. — Sylvilagus is commonly dis- tinguished from Lepus by its smaller size, al- though there is some overlap (namely, S. aquaticus and S. cunicularius are larger than Lepus americanus; J. A. V^hite 1984, pers. comm.). The species of these genera presently living in the region are usually dis- cernible by size, but the species within each genus are usually not (Fig. 4). Of the two species of Sylvilagus presently living in the Snake Range, S. audubonii has a larger mean size and tends to have much more crenulation in the second reentrant angle of P/3 than S. nuttallii (although there is overlap in both characters). Sylvilagus floridanus, which occurs just south of Utah and Nevada, has an even larger mean size than S. audubonii but has little crenulation in the P/3 like S. nuttallii. BYUVP 5493 and LACM 123658 compare well in size with S. nuttallii and S. audubonii (Fig. 4), but none of the nine P/3's oi Sylvilagus ' size from Crystal Ball Cave 354 Great Basin Naturalist Vol. 45, No. 3 have much crenulation in the second reentrant angle of P/3, suggesting that they belong to S. ntittallii rather than S. auduhonii. Although other species could be represented, the evidence suggests that at least the majority of the specimens listed above are of S. nuttallii. Sylvilagus nuttallii presently inhabits the region of the cave and northward, whereas S. auduhonii inhabits the region of the cave and southward. Sylvilagus nuttallii also tends to occur at higher elevations and in more wooded or bushy areas than S. auduhonii, which lives in plains or open country (Hall 1951|. Since Gandy Mountain is presently covered with only sparse bushes and is sur- rounded by open plains, the presence of S. nuttallii and absence of S. audidjonii in the assemblage suggests a replacement of woodland-alpine vegetation by the present desert conditions since the Pleistocene. Lepus cf americanus Material. — Right dentary with /I, P/3,/4, M/1,/2 (BYUVP 5519), anterior portion of left dentarv with P/4, M/1 (BYUVP 5543). A left dentary with /I (BYUVP 5430) falls within the size range of L. americanus and L. californicus. Discussion. — The jaw dimensions and P/3 widths of these specimens are intermediate in size between the Sylvilagus specimens (de- scribed above) and the majority of the Lepus specimens (described below). They fall in and near the range of variation of the smallest L. californicus and largest S. auduhonii speci- mens (Fig. 4), but most of the P/3's of these two species have a highly crenulated second reentrant angle, whereas the P/3 of BYUVP 5519 does not. These specimens are also indis- tinguishable from S. floridanus, but this spe- cies has never been reported living or as a fossil from Utah or Nevada. Lepus americanus does not presently occur in the Snake Range but does occur 160 km (100 miles) to the north and east, mainlv at high elevations (Durrant 1952, Hall 1951). Since the assemblage generally contains more species that presently range north of the cave than south of the cave, it is not at all unreason- able that L. americanus could have inhabited the region of the cave in the recent geologic past. Kurten and Anderson (1980) listed a number of Pleistocene fossil sites where L. americanus has been found south of its present range. Lepus townsendii MATt:RiAL. — Fused dentarv pair with right P/3,/4, M/1,/2,/3, left /I, 'P/3,/4, M/1,/2 (BYUVP 5488), right dentarv with M/1,/2 (BYUVP 5467), right dentarv with /I (BYUVP 5533), left dentary with all teeth (BYUVP 5442), left dentary with /I, P/3,/4, M/1,/2 (BYUVP 5484), 5 left dentaries without teeth (BYUVP 5424, 5429, 5474, 5532, LACM 123657), anterior portion of left dentarv with- out teeth (BYUVP 5439), 7 isolated right P/3's (BYUVP 5733, 5770-5772, 5793, 5796, 5802), 7 isolated left P/3's (BYUVP 5735, 5736, 5780, 5783, 5790, 5791, 5804). A partial left dentary with P/3 (BYUVP 5485), 28 dentaries lacking P/3 (BYUVP 5422, 5427, 5436, 5438, 5448, 5450, 5454, 5456, 5458, 5462, 5473, 5475, 5478, 5483, 5487, 5489, 5493, 5495, 5500-5502, 5524, 5527, 5530, 5531, 5540-5542), 7 isolated right P/3's (BYUVP 5617, 5732, 5745, 5768, 5774, 5778, 5792), and 5 isolated left P/3's (BYUVP 5716, 5775, 5776, 5801, 5809) show characteristics common to both L. townsendii and L. californicus. Discussion. — Lepus townsendii and L. californicus are distinguished from Sylvilagus and L. americanus by their large size. They are distinguished from each other by L. townsendii having a larger mean size (Fig. 4) and having less crenulation in the second reentrant angle of P/3 than L. californicus (Hibbard 1952). Miller (1976) observed L. californicus to have a highh' crenulated P/3 in most, but not all, cases, and Hibbard (1944, 1963) noted that individual variation is very great. My observations and those of J. A. White (1984, pers. comm.) show that many individuals of these species cannot be distin- guished by either size or the amount of crenu- lation in PAS; but statistical analysis can be used to estimate their relative abundance (Grayson 1977). Hibbard (1952) stated that the anterior part of P/3 is narrower in L. townsendii than in L. californicus, but, al- though I noticed variation in the narrowness and roundedness of the anterior P/3"s, it did July 1985 Heaton: Crystal Ball Cave Fossils 355 not correlate with the amount of crenulation in the second reentrant angle of that tooth. BYUVP 5424, 5467, and 5474 have greater alveolar length (P/3-M/3) to diastema length ratios than any Recent leporid specimens measured (Fig. 4), but they fall closest in size, especially based on their large tooth row length, to L. townsendii . Since 11 of the 43 measurable Lepus den- taries are larger than any modern L. californi- ctis specimens measured (Fig. 4), and over half of the large Lepus P/3's from the assem- blage show no crenulation (a very rare condi- tion in L. calif ornicus), it is clear that L. townsendii is well represented. Most of the 29 jaws that could be either L. townsendii or L. californicus are closer to the mean size of L. townsendii, and the isolated 13 P/3's of L. townsendii or L. californicus have slight crenulation in the second reentrant angle, yet are considerably less crenulated than the vast majority of L. californicus specimens. Since only two highly crenulated P/3's clearly be- longing to L. californicus (listed below) were found, most of these 13 P/3's with intermedi- ate crenulation probably belong to L. townsendii. Based on this information I esti- mate that the ratio of L. townsendii to L. californicus specimens from the Crystal Ball Cave assemblage is about 10 to 1. Grayson (1977) stated that L. townsendii is a more northern species and inhabits higher elevations and more grassy habitats than L. californicus, which prefers dryer shrubby ar- eas. With Sylvilagus, the more northern spe- cies is represented in the assemblage, and the more southern species is not. This is also the trend with Lepus. Hall (1981) reported L. townsendii in the area of Crystal Ball Cave but Durrant (1952), in a more detailed map, did not. Both reported L. californicus throughout the Bonneville Basin area. I have seen numer- ous L. californicus around Candy but never a L. townsendii, and J. C. Bates (1984, pers. comm.) reported never noticing any L. townsendii but seeing numerous L. californi- cus. This difference between the fossil and living species at Candy suggests that climatic boundaries have shifted northward in latitude and upward in elevation since the Pleis- tocene. Crayson (1977), using both fossil and Recent data, demonstrated that L. californi- cus increased in number at the expense of L. townsendii during the Recent and that it be- came the more dominant species in the Creat Basin 5,()()0 to 7,()()() Y. B. P. Although the eco- logical and adaptational differences between these two species are not fully understood, Grayson (1977) attributed this change to a post-Pleistocene warming trend. The species shift indicated by the Crystal Ball Cave assem- blage reiterates the data presented by Grayson (1977). Lepus californicus Material — Right P/3 (BYUVP 5781), left P/3 (BYUVP 5734). Twenty-nine dentaries and 13 other P/3's (listed under L. townsendii) show characters found in both L. townsendii and L. californicus. A left dentary with /I (BYUVP 5430) falls within the size range of L. americanus and L. californicus. Discussion. — Lepus californicus differs from L. townsendii in having a smaller mean size (Fig. 4) and a more crenulated second reentrant angle in P/3 as discussed above. The two P/3 specimens listed above have more crenulation than was seen in eight Recent L. townsendii specimens but are typical of L. californicus. The 13 P/3s of either L. townsendii or L. californicus (listed and dis- cussed above) show less crenulation than the vast majority of L. californicus specimens studied, but some of them could represent L. californicus since crenulation in the P/3 is not always present (Miller 1976). Lepus californi- cus is presently the most common lagomorph around Crystal Ball Cave (J. C. Bates 1983, pers. comm.), so its poor representation in the fossil assemblage suggests that it has only re- centlv become abundant there. Order Rodentia Family Sciuridae Mannota flaviventris Material. — Anterior portion of skull with right Ml/, 2/, 3/, left Ml/, 3/ (BYUVP 6528), anterior portion of skull without teeth (LACM 123663), dentary pair with right M/1,/3 (BYUVP 6536), right dentary with P/4, M/1,/3 (LACM 123665), right dentary with M/1,/2,/3 (BYUVP 6543), right dentary with M/2,/3 (BYUVP 6621), right dentary with M/3 356 Great Basin Naturalist Vol. 45, No. 3 (BYUVP 6620), left dentary with /I, M/1,/2,/3 (BYUVP 6477), left dentary with M/1,/2,/3 (LACM 123669). Another 70 partial maxillae (some with teeth), 70 partial dentaries (some with teeth), and approximately 300 isolated cheek teeth (BYUVP 6476, 6478-6518, 6520-6527, 6529-6535, 6537-6542, 6544- 6605, 6607-6619, 6622-6648, LACM 123664, 123666-123668, 123670) are of Mannota and compare favorably with M. flaviventris. Discussion. — Mannota is distinctly larger than other living sciurids (Hall 1981) tut dis- tinctly smaller than the extinct Paenemarmota (Repenning 1962). Mannota flaviventris is distinguished from M. monax by its anteriorly divergent upper tooth rows and from M. cali- gata, M. ohjtnptis, and M. vancouverensis by its smaller size (Hall 1981). Mannota flaviven- tris is also distinguished from these other spe- cies by its less massive dentition, M3/ being longer than it is wide, and M/3 having a trian- gular rather than a quadrangular outline (Lo- gan 1983). Hay (1921) named M. arizonae based on a partial skull from northern Arizona and said it was similar to M. flaviventris. Since this specimen is probably Late Pliocene in age and the validity of the species is uncertain (Kurten and Anderson 1980), it is not consid- ered a candidate for the Crystal Ball Cave specimens, all of which are indistinguishable from Recent M. flaviventris. The presence of M. flaviventris in the Crys- tal Ball Cave assemblage represents a shift in the climate and vegetation of the area because this species now inhabits only much higher elevations in the Snake Range (Hall 1981, Mead et al. 1982) and does not live on or around Candy Mountain (J. C. Bates 1983, pers. comm.). Hall (1946) reported fossil M. flaviventris from several caves far south of the present range of this species. Zimina and Gerasimov (1969) proposed that the marmot greatly expanded its distribution and num- bers under Late Pleistocene periglacial condi- tions for which it was well adapted, but it has since diminished its range significantly. Mar- niota flaviventris is not a cave-dwelling spe- cies, so its great abundance in the Crystal Ball Cave assemblage suggests that it once lived on Candy Mountain in large numbers, strongly supporting the hypothesis of Zimina and Gerasimov (1969). Spermophilus townsendii Material. — Anterior portion of skull with both I/'s (BYUVP 6060), partial skull with right P4/, Ml/,2/ (BYUVP 6255), partial skull without teeth (BYUVP 6462), 7 right den- 'taries with all teeth (BYUVP 6107, 6109, 6141, 6282, 6284, 6326, 7256), and 2 left den- taries with all teeth (BYUVP 6421, 6433). An- other 439 partial maxillae (some with teeth), 562 partial dentaries (some with teeth), and approximately 4,000 isolated cheek teeth compare favorably with S. townsendii. Discussion. — Spermophilus townsendii has the smallest mean size of any North Amer- ican species of Spermophilus and is also slightly smaller than Ammospermophilus leu- curus (Hall 1981). Spermophilus also differs from Ammospermophilus by having distinctly larger masseteric tubercles just anterior to the upper tooth rows (Hall 1981). The three par- tial skulls listed above and many of the partial maxillae without teeth have large masseteric tubercles that distinguish them from Am- mospermophilus. All the specimens listed above compare best in size with S. townsendii, but some of those only referred to this species are probably Ammospennophilus . Kurten and Anderson (1980) listed 13 extinct species of Spermophilus, but the only one close enough in size and age of deposits to the Crystal Ball Cave specimens to be considered is S. taylori, named by Hay (1921) and based on a single specimen from Texas. Kurten and Anderson (1980) considered this a doubtful species and most likely a synonym of S. townsendii. It is therefore not considered here. The presence of a single species of Sper- mophilus at Crystal Ball Cave is a striking contrast to the five possible species recovered from Smith Creek Cave in subequal numbers (Mead et al. 1982). These include S. cf townsendii, S. variegatus, and S. cf lateralis, which still inhabit the Snake Range, and S. cf. richardsonii and S. cf hcldingi, which have been extirpated but still inhabit Utah and/or Nevada (Hall 1981). The reason for this differ- ence may be that Smith Creek Caxe is at the base of 3,673 m (12,050 foot) Nh)unt Moriah and at the edge of the flat open Snake Valley, an area of diverse niches in contact with sev- eral diverse environments even now, and cer- tainly an area across which climatic July 1985 Heaton: Crystal Ball Cave Fossils 357 boundaries crossed many times during the Pleistocene. Candy Mountain, on the other hand, is only a small hill far out in Snake Valley, the area most favorable to S. townsendii (Hall 1946), and is isolated from the main Snake Range by 10 km (6 miles) of flat valley. The abundance of Spermophilus town- sendii fossils at Crystal Ball Cave suggests that this squirrel lived around Candy Mountain in large numbers for a long time, probably since fossils started accumulating in the cave. Dur- rant (1952) said this species is well suited to the western Utah desert and is particularly abundant around springs. Hall (1946) told how S. townsendii was a traditional food for native American Indians. Spermophilus townsendii is not a cave-dwelling animal as is Neotoma, and yet it is over twice as abundant as Neotoma in the assemblage (contrary to my earlier statement that Neotoma was the best represented genus, Heaton 1984). Neotoma has a much more restricted niche than Sper- mophilus and is never found in large num- bers. Since squirrels are very unlikely to ven- ture deep into caves, all the specimens must have been brought in by wood rats and/or small carnivores. It is interesting that fossil deposition occurred so rapidly, even deep in this isolated cave, that an outside species is better represented than the primary cave- dwelling species. J. C. Bates (1984, pers. comm.) reported seeing no squirrels on Candy Mountain and only a few in the sur- rounding valley in the many years he has lived in Candy. This, in contrast to its abundance as a fossil, suggests that S. townsendii reduced its numbers at the close of the Pleistocene in Snake Valley. Ammospermophilus cf. leucurus Material. — Right maxilla without teeth (BYUVP 8295), 2 left maxillae without teeth (BYUVP 8296, 8297). Some of the 439 maxil- lae, 562 dentaries, and approximately 4,000 isolated cheek teeth listed under Sper- mophilus townsendii probably also belong to this taxon. Discussion. — Ammospermophilus is dis- tinguished from Spermophilus by its smaller masseteric tubercles and its less robust lower cheek teeth (Hall 1981). Ammospermophilus leucurus now lives around Candy Mountain, but A. harrisii, A. interpres, A. insularis, and A. nelsoni, the other four extant species, oc- cur only south of Utah (Hall 1981), so the Crystal Ball Cave specimens are referred to A. leucurus although no character could be found to rule the others out. According to Durrant (1952) A. leucurus commonly occurs with S. townsendii but has a more restricted habitat, preferring rocky ter- rains. Ammospermophilus is best adapted for high temperatures (Vaughan 1972), and its low abundance in the assemblage compared to Spermophilus townsendii suggests that it has not inhabited the area as long, at least not in its present abundance. V^ith summers be- coming hotter and drier at the close of the Pleistocene, Ammospermophilus may have increased its numbers at the expense of Sper- mophilus in Recent times. Eutamius minimus Material. — Right dentary with P/4, M/1,/2 (BYUVP 6812), 3 right dentaries with P/4, M/1 (BYUVP 6171, 6210, 6755), left den- tary with all teeth (BYUVP 6190). Discussion. — Eutamius has two pre- molars in each maxilla, whereas Tamius has only one. Eutamius minimus is the smallest species of Eutamius and has a narrower and squarer P/4 than E. dorsalis or E. umhrinus. All the specimens listed above match E. tnin- irnus with respect to the P/4 and are smaller than the other species. Eutamius minimus and E. dorsalis live in the region of Crystal Ball Cave, and E. umhrinus lives higher in the Snake Range and westward into Nevada (Hall 1981). Eutamius minimus was also recovered from Smith Creek Cave (Mead et al. 1982). Eutamius minimus inhabits diverse habitats from deserts to forests, so its presence in the assemblage is not surprising. Eutamius dorsalis Material. — Right dentary with P/4, M/1,/2,/3 (BYUVP 6233), right dentary with P/4, M/1 (BYUVP 6257), 2 right dentaries with M/1 (BYUVP 5974, 6304), 2 left dentaries with M/1 (BYUVP 6129, 6134). Three partial right maxillae with Ml/ (BYUVP 6064, 6288, 6295) and a partial left maxilla with Ml/ (BYUVP 6000) also compare favorably with this species. 358 Great Basin Natur\list Vol. 45, No. 3 Discussion. — Eutamius dorsalis is dis- tinctly larger than E. minimus and slightly larger than £. urn/?rinus (Mead et al. 1982). It has a distinct isolated mesoconid on M/1 that is lacking in E. iimbrinus and is part of an ectolophid in E. minimus (Miller 1976). The M/l's of the six dentaries listed above match E. dorsalis in this character, and the four max- illae listed above match best in size with E. dorsalis but cannot be positively distin- guished from £. umhrinus. Of the larger chip- munks, only E. cf. umhrinus was reported from Smith'Creek Cave (Mead et al. 1982), and I have found only E. dorsalis in Crystal Ball Cave. Their present ranges help explain this difference since E. umhrinus only inhab- its the Snake Range west of Crystal Ball Cave and E. dorsalis inhabits the entire range (Hall 1981). Their ranges show that E. umhrinus is more isolated in areas of high elevation and more commonly absent from the areas once covered by Lake Bonneville. Family Geomyidae Thomomys umhrinus Material. — Anterior portion of skull with both I/'s (BYUVP 6656), anterior portion of skull with left 1/ (LACM 123672), right den- tary with /I, P/4, M/1 (BYUVP 6657), left den- tary with P/4 (BYUVP 8283). Four palates without teeth (BYUVP 6653, 6654, 6664, 6665), 4 right dentaries without cheek teeth (BYUVP 6660, 6663, 6666, 8281), and 8 left dentaries without cheek teeth (BYUVP 6655, 6658, 6659, 6662, 6681, 7009, 7010, 8282) also compare favorably with this species. Discussion. — Thomomys is distinguished from other North American geomyids by the absence of a superficial groove on the anterior face of the upper incisors (illustrations in Hall 1981), and none of the I/'s listed above have this groove. Thomomys umhrinus differs from T. talpoides and T. monticola, the only other species o{ Thomonujs living in Nevada, Utah, or surrounding areas, by having a sphenoidal fissure, by not having the palatine foramina fully anterior to the anterior openings of the infraorbital canals (Durrant 1952), and by the absence of a lingual indentation in the anterior lobe of P/4 (Hall 1946). The two Thomomys skulls from Crystal Ball Cave have the sphe- noidal fissure, and their palatine foramina are fully anterior to the infraorbital canals. The two P/4's also lack the lingual indentation as in T. umhrinus. Mv observations and also those of Hall (1946, Fig. 308-321) indicate that T. umhrinus has a larger mean size than the other two species mentioned (contrary to Bergman's rule), and all the Crystal Ball Cave specimens compare best in size with the larger T. umhrinus. Thomomys umhrinus is the only geomyid currently inhabiting the Snake Range, and it is a southern species, ranging from Nevada and Utah southward into Mexico (Hall 1981). Thomomys hottae and T. townsendii are now considered as subspecies of T. umhrinus (Hall 1981). Thomomys talpoides, which inhabits mountain ranges to the east and west of the Snake Range, has Nevada and Utah as almost its southern boundary and extends northward into Canada. Thomomys talpoides tends to inhabit higher elevations than T. umhrinus as well as higher latitudes. Thomomys cf. , tal- poides was the only geomyid reported from Smith Creek Cave (Goodrich 1965). Hall (1946) pointed out that, although T. umhrinus is usually a lower-elevation species than T. talpoides, T. umhrinus is the only geomyid in the Snake Range and occurs at all elevations (but is less abundant at higher ele- vations than is T. talpoides at similar eleva- tions in other ranges). Hall (1946) attributed this to antiquity of occupancy and proposed that T. umhrinus, having no competitors in the Snake Range, developed populations adapted to higher elevations. Since T. umhri- nus was the species best adapted to the valleys surrounding the Snake Range, no species bet- ter adapted to higher elevations could pass through to their favorable habitat. This could explain why the Crystal Ball Cave assemblage suggests no northward range shift for species of Thomomys as it does for other groups such as lagomorphs. If Hall (1946) is right, the ten- tative assignment of the Smith Creek Cave specimen to T. talpoides (Goodrich 1965) must be in error. Another possibility is that predatory birds transported the specimen to the cave, but this seems imlikely since T. talpoides occurs onK' as close as 75 km (45 miles) to the northwest and 180 km (108 miles) to the east of Smith Creek Cave. Hall (1946) also pointed out that geom\'ids, as individuals, are extremely sedentary, and this could be the July 1985 Heaton: Crystal Ball Cave Fossils 359 cause of their slow invasion and northward retreat compared to other mammals. Family Heteromyidae Perognathus cf. fonnosus Material, — Partial right maxilla with P4/ (BYUVP 6682), 2 right dentaries with P/4 (BYUVP 6859, 6879), 2 right dentaries with M/2 (BYUVP 6711, 6856), left dentary with all teeth (BYUVP 6697), left dentary with P/4, M/1 (BYUVP 6786), left dentary with P/4, M/2 (BYUVP 6115). Discussion. — Perognathus longimembris, P. parvus, and P. fonnosus now inhabit the Crystal Ball Cave area, and the closest other species range more than 250 km (150 miles) to the east and south (Hall 1981). Of the three local species, P. longimembris can be ruled out because its M/3 is distinctly smaller than its P/4 (Hall 1981), and the BYUVP 6697 has the opposite condition. Perognathus parvus and P. formosus are very similar dentally, and the Crystal Ball Cave specimens match well with both of them. Perognathus formosus has a larger mean size than P. parvus, and the Crystal Ball Cave specimens compare best in size with P. formosus, although P. parvus and several other western species cannot be ruled out. Miller (1979) referred all the Perognathus specimens found at Smith Creek Cave to P. cf parvus, but, since the identification was ten- tative at both caves, it does not seem wise to speculate about a possible difference between the two assemblages with respect to this genus. Microdipodops megacephalus Material — Right maxilla with P4/, Ml/, 2/ (BYUVP 6695), right maxilla with P4/, M2/ (BYUVP 6781), right maxilla with Ml/ (BYUVP 6709). Three partial right maxillae with P4/ (BYUVP 6669, 6674, 6797), a partial right maxilla with a partial Ml/ (BYUVP 6759), a right dentary with /I, P/4, M/1 (BYUVP 6693), 2 right dentaries with P/4 (BYUVP 6795, 6860), and a left dentary with P/4 (BYUVP 6708) are oi Microdipodops and compare favorably with M. megacephalus. Discussion. — Microdipodops is similar to Perognathus but can be distinguished den- tally by the molars having a single enamel loop as opposed to the biloph nature of Per- ognathus molars. The P/4's are also distinct in being more hypsodont and having a straight posteriolabial border as opposed to the round and symmetrical nature of the Perognathus P/4's. Microdipodops megacephalus ranges throughout most of Nevada and into neigh- boring states including Utah, and it is cur- rently found around Crystal Ball Cave (Hall 1981). Microdipodops pallidas, the only other species, occurs along the southern Nevada- California border more than 320 km (200 miles) southwest of Crystal Ball Cave (Hall 1981). Microdipodops megacephalus can be distinguished from M. pallidus by the latter's possessing a small notch in the labial side of Ml/, and all the Crystal Ball Cave specimens possessing the Ml/ are clearly M. mega- cephalus. Microdipodops cf megacephalus was reported at Smith Creek Cave (Miller 1979), and all heteromyid taxa recovered were low in abundance as at Crystal Ball Cave. This low abundance is probably due to a low den- sity in life, since even now they are rarely seen in the area. Dipodomys microps Material. — Two right dentaries with /I (BYUVP 6672, 8284), left dentary fragment with P/4 (BYUVP 6676). Nine maxillae with- out teeth (BYUVP 5593, 6667, 6668, 6670, 6675, 6677-6680) and 2 right dentaries with- out teeth (BYUVP 6673, 6683) also compare favorably with this taxon. Discussion. — Dipodomys is distinctly larger than other heteromyid genera. Dipodomys microps is distinguished from other species of Dipodomys by having chisel- shaped lower incisors (anterior face flat) rather than awl-shaped lower incisors (anterior face round), and the incisors of BYUVP 6672 and 8284 are chisel-shaped. The P/4 of D. microps is also distinct in having a larger and more isolated anterior loph than that of D. ordii or D. merriami but not a complete separation of lophs as in D. deserti, and the P/4 of BYUVP 6676 clearly matches D. microps. The re- ferred specimens also match perfectly with Recent D. microps but lack the diagnostic teeth. Of the four species of Dipodomys presently living in Utah and Nevada, D. mi- crops and D. ordii are found in the Snake 360 Great Basin Naturalist Vol. 45, No. 3 Range and D. merriami and D. deserti occur more than 200 km (125 miles) to the south and west (Hall 1981). Dipodomijs microps has a much smaller range than D. ordii, occurring only in Nevada and parts of surrounding states (Hall 1981). The Dipodomijs specimens recov- ered from Smith Creek Cave (Miller 1979) were referred to D. ordii because they had awl-shaped lower incisors. This difference be- tween the two assemblages is difficult to ex- plain because the range differences between these species do not suggest distinct differ- ences in habitat preference. Family Cricetidae Peromyscus manicidatus Material. — Right maxilla fragment with Ml/, 2/ (BYUVP 6703), left maxilla with Ml/,2/,3/ (BYUVP 6782), left maxilla frag- ment with Ml/, 2/ (BYUVP 6771). Thirty-nine Peromyscus dentaries containing one or more molars compare favorably with P. manicidatus and P. crinitus. Discussion. — Of the six species of Per- omyscus that inhabit Utah and Nevada, only P. maniculatus, P. truei, and P. crinitus cur- rently live around Crystal Ball Cave (Hall 1981). Peromyscus manicidatus was captured live inside the cave by the author in 1982 and 1983. Peromyscus fossils from Smith Creek Cave were not identified to the species level (Goodrich 1965, Mead et al. 1982, Miller 1979). Dental characters that distinguish spe- cies of Peromyscus are few and not always reliable. Peromyscus manicidatus and P. truei belong to the subgenus Peromyscus, which has accessory tubercles or enamel loops on the labial side of Ml/ and M2/; P. crinitus belongs to the subgenus Haplonujomys, which lacks these features (Hall 1981). I found this charac- ter to be quite rehable, and the specimens listed above all have prominent cusps on Ml/ and M2/. In further refinement of this charac- ter. Miller (1971, 1976) was able to separate P. maniculatus from all other western species of Peromyscus by the presence of an antero- conule on Ml/ with direct attachment to the anterocone rather than being joined to it by a distinct loph as in P. truei. Specimens listed above fit P. maniculatus in this respect. Spe- cies of the subgenus Haplomyonu/s usually lack the anteroconule entirely (Hall 1981, Miller 1971, 1976). Unfortunately, excessive wear on the teeth erases this character. Of the 40 Peromyscus dentaries containing one or more molars, 39 compare best in size with the smaller P. manicidatus and P. crini- tus, but no character could be found to sepa- rate these species based on dentaries. Miller (1976) found the P/3's of P. maniculatus, P. crinitus, and P. eremicus to be relatively more reduced than P. hoijUi and P. truei. The 8 Crystal Ball Cave Peromyscus dentaries con- taining M/3 tend to have M/3 relatively re- duced as in P. manicidatus, P. crinitus, and P. eremicus, and in size all the 39 dentaries listed above compare best in size with the smaller P. maniculatus and P. crinitus. Peromyscus cf. crinitus Material. — Right maxilla with Ml/,2/ (BYUVP 6780), left maxilla with Ml/,2/ (BYUVP 6769), left maxilla with Ml/ (BYUVP 6715). Thirty-nine Peromyscus dentaries con- taining one or more molars compare favorably with P. maniculatus and P. crinitus. Discussion. — These specimens lack acces- sory tubercles and enamel loops on the two principal outer angles of M 1/ and M2/, so they probably belong to the subgenus Haplomy- omys (Hall 1981). Of the two species of Hap- lomyomys found in Utah, P. crinitus and P. eremicus, the Crystal Ball Cave specimens compare better in size with the smaller P. crinitus (although there is considerable over- lap). Some of the 39 dentaries discussed under P. maniculatus (above) could also belong to this species since no character was found to distinguish them based on dentaries. Per- omyscus crinitus is presently found around the cave, but P. eremicus ranges onlv as close as 225 km (135 miles) to the south (Hall 1981), further suggesting that these specimens are P. crinitus. Peromyscus cf. truei M.\TERIAL. — Left dentarv with M/1 (BYUVP 6718). Disc:ussiON. — Peromyscus truei is the largest species of Peromyscus living in Utah and Nevada (Durrant 1952, Hall 1981), and the M/1 listed above compares well in size with this species and is larger than the mean July 1985 H EATON: Crystal Ball Cave Fossils 361 size of P. eremictis and P. boylii and distinctly larger than any P. maniculatus or P. crinitus M/l's examined. Identification is based only on size since no other character could be found to distinguish M/l's of Peromysciis. This species is found throughout the Great Basin, so its presence in the assemblage is not surprising. Neotoma lepida Material. — Partial skull without teeth (LACM 123671), 2 partial right maxillae with Ml/ (BYUVP 7045, 7065), left maxilla with Ml/ (BYUVP 7154), partial left maxilla with Ml/ (BYUVP 7246). Discussion. — Neotoma lepida and N. cinerea are the only species o( Neotoma that presently inhabit the Snake Range (Hall 1946, 1981). Of three wood rats that I trapped in Crystal Ball Cave and two elsewhere on Candy Mountain in 1982 and 1983, all were N. lepida. 1 did trap a N. cinerea in another cave in the Snake Range 35 km (22 miles) south of Crystal Ball Cave, so they are known to inhabit caves in the area. Miller (1979) re- ported both N. lepida and N. cinerea from Smith Creek Cave but did not comment on their relative abundance. Of these two spe- cies, N. cinerea is much more boreal than N. lepida, having a more northern range and be- ing found at higher elevations (Finley 1958, Hall 1946, 1981). Durrant (1952) and Hall (1981) also reported N. albigida, N. mexicana, and N. stephensi living in Utah but far to the south and east of Crystal Ball Cave. Neotoma cinerea is usually distinguishable from N. lepida by its larger size and deeper anterolabial reentrant angle on Ml/ (Finley 1958). According to Hall (1946), the maxillary alveolar length is always 8.8 mm or less in N. lepida and 9. 1 mm or more in N. cinerea for the Nevada subspecies, and Finley (1958) re- ported only a slight overlap for the Colorado subspecies. The three other Utah species of Neotoma are intermediate in size between N. lepida and N. cinerea, and N. albigula has the Ml/ pattern of N. lepida whereas N. Tuexicana and N. stephensi have the Ml/ pattern of N. cinerea (Finley 1958). Because these are the most diagnostic characters, only maxillae with Ml/ and/or a measurable alveolar length were considered. The Crystal Ball Cave specimens listed above compare best in size with N. lepida, the only species of Neotoma known to presently inhabit the cave. Maxillary alveolar lengths of Neotoma specimens from the cave show a strongly bimodal distribution, suggesting that N. albigtda, N. mexicana, and N. stephensi are not represented since they are intermedi- ate in size between N. lepida and N. cinerea. The shallow anterolabial reentrant angle of the Ml/'s also compares favorably with N. lepida. The scarcity of N. lepida specimens in the assemblage suggests that this species probably has not always inhabited the cave as it does now. Neotoma cinerea Material. — Anterior portion of skull with both I/, Ml/, 2/ (BYUVP 7384), maxilla pair with all teeth except left 1/ (BYUVP 7281), maxilla pair with right Ml/, 2/, 3/, left Ml/, 2/ (BYUVP 7282), maxilla pair with both Ml/, 2/ (BYUVP 7067), maxilla pair with right Ml/, 2/ (BYUVP 7015), maxilla pair with left Ml/ (BYUVP 7213), 9 right maxillae with Ml/,2/,3/ (BYUVP 7136, 7149, 7158, 7167, 7214, 7248, 7254, 7314, 7320), 3 right maxillae with Ml/ (BYUVP 7273, 7316, 7330), 25 partial right maxillae with Ml/ (BYUVP 7014, 7018, 7024, 7038, 7046, 7104, 7114, 7125, 7134, 7138, 7147, 7170, 7177, 7180, 7182, 7197, 7204, 7216, 7242, 7247, 7249, 7276, 7344, 7348, 7349), 10 right maxillae without teeth (BYUVP 7255, 7343, 7353, 7367, 7377, 8286-8290), 7 left maxillae with Ml/,2/,3/ (BYUVP 7095, 7212, ,7250, 7257, 7274, 7376, 7379), 4 partial left maxillae with Ml/,2/ (BYUVP 7101, 7174, 7179, 7324), partial left maxilla with Ml/,2/ (BYUVP 7017), 34 partial left maxillae with Ml/ (BYUVP 7021, 7061, 7062, 7072, 7073, 7087, 7099, 7106, 7133, 7140, 7142, 7144, 7145, 7151, 7162-7164, 7172, 7175, 7183, 7189, 7200, 7205, 7217, 7220, 7225, 7267, 7300, 7317, 7318, 7322, 7351, 7362, 7371), 6 left maxillae without teeth (BYUVP 7171, 7346, 8291-8294). An- other approximately 200 maxillae, 200 den- taries, and 2000 isolated molars compare best with this species. Discussion. — Neotoma cinerea is recog- nized by its large size and deep anterolabial reentrant angle on Ml/ as discussed above. 362 Great Basin Naturalist Vol. 45, No. 3 Neotoma cinerea has the largest mean size of any species oi Neotoma, and all the specimens listed above match Recent N. cinerea in size and have the deep anterolabial reentrant an- gle on Ml/ when this tooth is present. This makes N. cinerea the second best represented species in the Crystal Ball Cave assemblage after Spennophilus townsendii. The fact that N. cinerea is abundant in the assemblage but not found in the cave now, whereas N. lepida is rare in the assemblage but now the only wood rat living in the cave, suggests that a replacement of N. cinerea by N. lepida has recently taken place in the area. The great abundance oiN. cinerea remains at sites 1 and 2 of Crystal Ball Cave also helps substantiate my hypothesis that wood rats were the pri- mary means of transporting fossils, especially of large mammals, into the cave. The domi- nance of N. cinerea over N. lepida in the assemblage suggests that N. cinerea was the primary species involved in this transport. The ecological differences between N. cinerea and N. lepida have significance both to the replacement of the former species by the latter and to the accumulation of fossils in the cave. Finley (1958), in his detailed study of Colorado wood rats, found den sites to be the most limited resource for all species. Since all wood rats prefer the same basic types of den sites, namely rocky crags and caves, multiple species are rarely found coexisting (Finley 1958). This suggests that, when condi- tions at Crystal Ball Cave reached a threshold where they favored N. lepida over N. cinerea, the replacement took place quickly. Neotoma cinerea prefers higher elevations and lati- tudes than iV. lepida, and hot summers in arid regions seem to be a limiting factor for this species (Finley 1958, Hall 1981). The chang- ing conditions that led to the replacement of N. cinerea by N . lepida may have been the increase in temperature and decrease in mois- ture at the close of the Pleistocene, the shift in vegetation caused by it, or both. Regarding food, Finley (1958) stated that N . cinerea prefers soft-leaved shrubs, forbs, and mon- tane conifers, whereas N. lepida prefers xero- phytic shrubs, forbs, cacti, and shrubby trees. Species o[ Neotoma differ somewhat in den preferences and collecting habits. Finley (1958) stated: "Dens of N. cinerea are usually in high vertical crevices in cliffs or caves, whereas those of ... N. lepida are usually in low horizontal crevices or under boulders or large fallen blocks. Dens of [N. ] cinerea usu- ally contain larger accumulations of sticks and bones." That N. cinerea collects more mate- rial, especially bone, is very significant since I consider wood rats as the primary mechanism of fossil deposition at Crystal Ball Cave. This suggests that the rate of bone deposition de- creased when N. lepida replaced N. cinerea, and it helps explain why many elements of the present local fauna are so poorly represented in the assemblage and why all the dated fossils are Late Pleistocene rather than Recent in age. A replacement of N. cinerea by N. lepida parallels the replacement oiSijlvilagiis nuttal- lii by S. auduhonii and Lepiis townsendii by L. californiciis (discussed above) and helps confirm that a warming trend took place in the recent past. Although N. cinerea still lives in the area, it seems to have been driven to higher elevations in the Snake Range. Ondatra zibethicus Material — Palate without teeth (BYUVP 7383), partial right dentarv with anterior f of M/1 (BYUVP 7391). Discussion. — Ondatra is easily distin- guished from other microtine rodents by its large size combined with rootless molars. On- datra zibethicus is now considered the only extant species o{ Ondatra (Hall 1981), and the Crystal Ball Cave specimens are indistin- guishable from this species. A number of fossil species have been named, but there is consid- erable confusion about their status (Miller 1976). All the extinct species considered valid by Semken (1966) and Nelson and Semken (1970) are smaller than O. zibethicus. The Crystal Ball Cave dentary is almost as large as the largest O. zibethicus to which it was com- pared. The M/1 is 7.9 mm long and 2.5 mm wide which best matches measurements taken from Wisconsinan-age O. zibethicus specimens (Nelson and Semken 1970). The palate is slightly smaller than the mean of O. zibethicus but well within its range of variation. Ondatra zibethicus has not been reported living near Candy but occurs as close as 56 km (35 miles) to the northeast and 160 km (100 [uly 1985 Heaton; Crystal Ball Cave Fossils 363 miles) to the south (Hall 1981). Since Ondatra is a reliable indicator of permanent water (Nelson and Semken 1970), the retreat of" Lake Bonneville and loss of perennial streams in the area probably lead to its extirpation from the Snake Range. Microtus cf longicaudus Material.— Two left M3/'s (BYUVP 6940, 6981), 7 right M/3's (BYUVP 8220-8226), 15 left M/3's (BYUVP 7002, 8227-8241). Numer- ous other partial jaws and isolated molars can- not be distinguished from Lagurus but lack characters that would assign them to other species oi Microtus, some of which are likely Microtus since over a third of the microtine M/3's belong to Microtus. Among these are a partial skull with both Ml/, 2/ and the poste- rior incisive foramina (BYUVP 8285) and a right maxilla with Ml/, 2/ (BYUVP 6943). Discussion. — Microtus differs from Lagu- rus , the only other microtine of its size with rootless molars, in having 3 transverse loops on M/3 rather than 4 prisms, some of which are closed triangles, and in having a large semicircular posterior loop on M3/ rather than a simple elongate loop (Hall 1981). The 2 M3/ 's and 22 M/3's from Crystal Ball Cave listed above clearly match Microtus in this respect. There are many species oi Microtus, some of which have distinct dental characteristics and some of which do not. The only two species of Microtus now in- habiting the Snake Range are M. longicaudus and M. montanus, and no character could be found to distinguish them dentally. The inci- sive foramina of M. longicaudus are not con- stricted posteriorly as are those of M. mon- tanus, but they differ from those of Lagurus only in having slightly curved rather than straight external margins. Since only the pos- terior end of the incisive foramina are found on skulls that could be Microtus from Crystal Ball Cave, skulls of M. longicaudus in the collection are indistinguishable from Lagu- rus. Of 13 skulls containing incisive foramina that may be Microtus, 2 have constricted inci- sive foramina as in M. montanus (listed be- low), and 11 compare well with M. longi- caudus and Lagurus. Three other species of Microtus presently occur in Utah but not in the Snake Range: M. pennsijlvanicus and M. richardsoni in the central mountain ranges and Af . mexicanus in the southwestern corner of the state. Micro- tus pennsijlvanicus has a posterior loop on M2/ not found in other species, and this char- acter was only found on one specimen (listed below). Microtus richardsoni is distinctly larger than the other species described here, and none of the microtine specimens from Crystal Ball Cave are large enough to compare with it. Microtus mexicanus is dentally indis- tinguishable from M. montanus and M. longi- caudus, and its incisive foramina are identical to Lagurus and similar to M. longicaudus. The specimens listed above are identical to Recent specimens of M. longicaudus, M. mex- icanus, and more distant ranging species. But since Af. longicaudus presently occurs at Crystal Ball Cave whereas M. mexicanus oc- curs more than 400 km (250 miles) to the southeast (Hall 1981), and because the gen- eral trend in the region is for range boundaries to be migrating northward, the Crystal Ball Cave specimens (except the few discussed be- low) are referred to M. longicaudus. Microtus cf. montanus Material. — Two partial palates without teeth, which include the posterior end of the incisive foramina (BYUVP 8218, 8219). Discussion. — Microtus montanus is the only microtine of its size presently occurring in Utah or Nevada with incisive foramina that abruptly constrict posteriorly and are nar- rower posteriorly than anteriorly. The poste- rior ends of the incisive foramina in these two specimens are too narrow to be M. longi- caudus, M. pennsijlvanicus, M. mexicanus, or Lagurus curtatus. Microtus townsendii and M. oregoni also have incisive foramina like M. montanus, but they both occur only along the Pacific Coast from northern California to southern British Columbia. Since Af. mon- tanus presently occurs in the Snake Range (Hall 1981), the Crystal Ball Cave specimens are referred to it. Microtus montanus tends to occur at higher elevations than other species of Microtus in Utah (Durrant 1952), so its presence in the assemblage suggests that con- ditions at the cave during the Late Pleistocene may have been like those of higher elevations in the Snake Range now. 364 Great Basin Naturalist Vol. 45, No. 3 Microtus cf. pennsylvanicits Material. — Partial skull with right Ml/,2/ (BYUVP 6973). Discussion. — Microtus pennsijlvaniciis is unique in having a rounded posterior loop behind the 4 closed angular sections of M2/. This single specimen from the assemblage has this posterior loop, but the loop is not com- pletely closed off from the preceding triangle as in the Recent specimens to which it was compared. Since the distinguishing character is not fully developed, the specimen is only referred to M. pennsijlvaniciis. This species is n(»t presently found in the Snake Range, but it occurs 190 km (114 miles) east of Crystal Ball Cave in the mountains of central Utah and is a northern species (Hall 1981). Considering the climatic shifts since the recession of Lake Bon- neville, it is not unlikely that it could have inhabited the Snake Range during the Late Pleistocene. Lagurus curtatus Material.— Skull with right I/, M2/,3/, left I/, Ml/,2/ (BYUVP 6899), left dentary with M/1,/2,/3 (BYUVP 6977), left dentary with M/2,/3 (BYUVP 6986), 28 right M/3's (BYUVP 8242-8270), 9 left M/3's (BYUVP 8271-8280). Numerous partial jaws and isolated molars may be L. curtatus but cannot be distin- guished from Microtus longicaudus (as dis- cussed above). Discussion. — The differences between Lagurus and Microtus are discussed above. Lagurus curtatus, the only North American species of Lagurus, is distinguished from Old World representatives by having four instead of five closed triangles on M/3 and cement present in the reentrant angles of the molars (Hall 1981). This species presently occurs in the Snake Range and northward into Canada (Hall 1981). Lagurus specimens are nearly twice as abundant as those of Microtus in the assemblage, but, since no information on their Recent relative abundance or habitat differences could be found, it is difficult to understand the reason for this. Order Carnivora Family Canidae Canis c{. hit runs Material — Lower inci.sor (BYUVP 7459), right C/l (LACM 123675), partial left M/1 (BYUVP 7460). The frontal region of a skull (LACM 123676) and an anterior fragment of a left dentary without teeth (BYUVP 7458) also compare favorably with this species. Discussion. — These specimens are indis- tinguishable from specimens of Recent C. la- trans, generally recognized as the only spe- cies of coyote in the Pleistocene or Recent (Giles 1960). Dentally, C. latrans falls within the wide range of variation of the domestic dog, C. familiaris (Anderson 1968), so the possibility that the Crystal Ball Cave speci- mens are C. familiaris cannot be totally elimi- nated. Nevertheless, C. latrans is presently very abundant around the cave (J. C. Bates 1983, pers. comm.. Hall 1981) and has been recognized from nearby Pleistocene assem- blages that have better stratigraphic control (Kurten and Anderson 1972, Miller 1979), so there is no reason to believe it would not be found in the assemblage. Also, domestic dogs tend to have many more tooth malformations than coyotes (Anderson 1968), and none are seen in the Crystal Ball Cave specimens. Lack of human fossils and artifacts at Crystal Ball Cave makes domestic dogs less likely to be present than at sites that contain such rem- nants of human occupation. Although resi- dents of Candy have domestic dogs that some- times roam on Candy Mountain, the lack of any canid specimens in the assemblage that cannot be referred to native species also sup- ports the conclusion that the Crystal Ball Cave specimens are C. latrans. Canis cf. lupus Material— Partial right Ml/ (BYUVP 7455), left P/1 (BYUVP 7457), posterior end of right M/1 (BYUVP 7456), left M/1 (LACM 123674), iixis (LACM 123710). Discussion. — Identification of these canid fossils is based on their size, being substan- tially larger than C. latrans but considerably less robust than C. dirus. They do, however, fit within the large size range of C. familiaris, so the identification must be tentative. Goodrich (1965) reported C. lupus from Smith ('reek Cave but did not describe the material. Canis lupus has been reported liv- ing in the Snake Range in Recent times (Hall 1981), although man has now reduced its range and numbers considerably. July 1985 Heaton: Crystal Ball Cave Fossils 365 Vulpes vulpes Material — Skull with right Pl/,2/,4/, left P4/, M2/ (BYUVP 8299), posterior portion of right maxilla with Ml/,2/ (BYUVP 7466), par- tial left maxilla with Ml/,2/ (BYUVP 7467), 2 right Cl/'s (BYUVP 7468, 7470), left CI/ (BYUVP 7469), right P4/ (BYUVP 7474), left P4/ (BYUVP 7471), right dentary with M/2 (BYUVP 7461), posterior portion of right den- tary with P/4, M/1,/2 (BYUVP 7463), left den- tary with M/1,/2 (BYUVP 7464), anterior por- tion of left dentary with M/1,/2 (BYUVP 7462), right P/4 (BYUVP 7475), left P/4 (BYUVP 7472). An anterior fragment of a right dentary without teeth (BYUVP 7465) and an anterior fragment of a left dentary without teeth (BYUVP 7476) also compare favorably with this species. Discussion. — Vulpes is distinguished from Urocyon by the configuration of the crest on the top of the skull and the lack of a promi- nent "step" on the posteroventral margin of the dentary. The ventral margin of the den- tary of Vulpes curves upward posteriorly be- ginning at the posterior end of the tooth row, but in Urocyon it remains uncurved well be- hind the tooth row all the way to the "step." Urocyon, which now ranges from the cave site southward and throughout North America, is intermediate in size between V. vulpes and V. velox. Four of the Crystal Ball Cave speci- mens include the posterior dentary and lack the "step" characteristics of Urocyon, and all the Crystal Ball Cave specimens are larger than the largest Urocyon specimen examined but compare well in size and shape to V. vulpes. Vulpes vulpes does not presently occur around Crystal Ball Cave but V. velox and U. cinereoargenteus do (J. C. Bates 1983, pers. comm.. Hall 1981). The presence of the more northern V. vulpes but not the more southern U. cinereoargenteus in the cave assemblage represents a northward shift of the boundary between these two species. The ranges of V. vulpes and U. cinereoargenteus do overlap to a degree now, but in the western United States the overlap is not great, and where it does occur V. vulpes favors the higher eleva- tions and U. cinereoargenteus the lower ele- vations (Hall 1981). Based on range maps in Hall (1981), the range of V. vulpes in the western United States is quite scattered, sug- gesting that it is relectual and that this species is diminishing in numbers there. Urocyon cinereoargenteus has a distinct northern boundary across Utah and Nevada with no remnant populations, suggesting that this species has been making a northward inva- sion. The Crystal Ball Cave assemblage con- firms that U. cinereoargenteus has been ex- panding its range at the expense of V. vulpes. Vulpes velox Material. — Left dentary with P/3 and par- tial M/1 (BYUVP 7477), posterior portion of left M/1 (BYUVP 7479). A partial left dentary with M/2 (BYUVP 7478) also compares favor- ably with this species. Discussion. — Vulpes velox and V. macro- tis are now considered conspecific (Hall 1981). The dentary (BYUVP 7477) lacks the "step" of Urocyon, and the M/1 lacks a small cuspule found on the posterolabial margin of the main cusp of all the Urocyon specimens but none of the Vulpes specimens examined. The Crystal Ball Cave specimens listed above are smaller than U. cinereoargenteus but may be similar in size to the smaller U. littoralis, which is known only from islands along the coast of southern California (Miller 1971). Since V. velox still lives around Crystal Ball Cave (J. C. Bates 1983, pers. comm.), its pres- ence in the assemblage is not surprising. Its low frequency compared to the now extir- pated V. vulpes suggests that it may not have always inhabited the area, may have inhabited it in much smaller numbers, or may have had a different microhabitat causing it to frequent the cave area less than V. vulpes. The ranges of V. vulpes and V. velox presently overlap to a degree, especially in the midwest, but in the western United States this overlap is small (Hall 1981). Although V. velox occurs in the Snake Range now, it is a more southern spe- cies than V. vulpes, so its northern range ex- tentions may be of Recent age. Family Mustelidae Mustela cf frenata Material.— Left Ml/ (BYUVP 7487), right dentary with P/2,/3, M/1,/2 (BYUVP 7483), partial right dentary without teeth (BYUVP 366 Great Basin Naturalist Vol. 45, No. 3 7484), left dentary with M/1,/2 (BYUVP 7488), left dentary with M/1 (BYUVP 7485), partial left dentary with M/1,/2 (BYUVP 7486). Discussion. — All these Mustela speci- mens compare best in size with M. frenato, which presently lives around Crystal Ball Cave. The size range of M. frenata is over- lapped by the smaller but more variable M. erminea (Kurten and Anderson 1980), which also ranges in the cave area (Hall 1981). The specimens could belong to M. erminea since this species is dentally similar to M. frenata. Mustela rixosa is always smaller and M. ni- gripes and M. vison are always considerably larger than the Crystal Ball Cave specimens. Mustela frenata was the most abundant mustelid at Smith Creek Cave, but M. er- minea was also present (Miller 1979). Since all the Crystal Ball Cave specimens fall in the narrow size range of M. frenata, they are re- ferred to this species. Mustela cf vison Material.— Left Ml/ (BYUVP 7482). A ju- venile left dentary without teeth (BYUVP 7491) also compares well with this species. Discussion. — This isolated tooth was com- pared to a variety of Recent mustelids and other small carnivores and found most similar to M. vison. This is North America's largest extant species of Mustela (although the extinct sea mink, M. macrodon, was larger) and is distinctly larger than, but similar in shape to, M . frenata (described above). Mustela vison was recovered from Smith Creek Cave (Miller 1979) and presently occurs 160 km (100 miles) north and east of Crystal Ball Cave (Hall 1981), but it does not currently live in the Snake Range. This species requires lakes or streams to survive (Hall 1946), so its extir- pated nature in the Snake Range may have been due to the recession of Lake Bonneville and/or loss of perennial streams in the area at the end of the Pleistocene. Mustela vison is sometimes confused with M. nigripes since both are of similar size (Kurten and Anderson 1980), and no distinc- tion in isolated molars could be foimd in the literature. Mustela nigripes is currently en- dangered, and no comparative material was available. It has never been reported from western Utah or Nevada, so the Crystal Ball Cave specimens are referred to M. vison, which is known to have lived in the area. Martes americana Material, — Left dentary with M/1 (BYUVP 7480), left M/1 (BYUVP 7481). The anterior portion of a right dentary without teeth (BYUVP 7489) and the posterior portion of a right dentary without teeth (BYUVP 7523) probably also belong to this taxon. Discussion. — Anderson (1970), in her sys- tematic review of the genus Martes, consid- ered M. nobilis (found in four caves in Wyo- ming, Idaho, and northern California) to be a distinct species from M. americana. Of the two, M. nobilis is larger, and its lower car- nasial has a relatively shorter trigonid. The lower canines of M. nobilis sometimes have faint grooves on the external surface not found in M. americana (Anderson 1970). The only other species of Martes presently living in Utah is M. pennanti, the fisher. It is consider- ably larger than M. americana, M. nobilis, and the Crystal Ball Cave specimens. Neither M. americana nor M. pennanti currently live in the Snake Range, but both occur in the mountains of central Utah and northward. BYUVP 7480 is as large as the largest M. americana specimen to which it was com- pared, and, judging from the incisor socket, its incisor was slightly larger. The other speci- mens are the same size as most Recent M. americana specimens. Both lower carnasials match perfectly in shape with M. americana and do not show a relatively shorter trigonid, so they are assigned to M. americana. A right Ml/ of M. nobilis was recovered from Smith Creek Cave (Miller 1979), but M. americana has never been reported. The ecological and chronological separation of these tuo species in the Snake Range is, therefore, problematic. Brown (1971) listed M. americana as one of eight species of boreal mammals that presently range in the Sierra Nevada and the Rocky Mountains but on none of the isolated Great Basin ranges in between. This citing demonstrates that A/, americana did range at least as far east in the Great Basin as the Snake Range before becoming extirpated. Brachyprotoma brevimala, sp. nov. Type. — Anterior portion of skull including a complete palate except the most posterior July 1985 Heaton: Crystal Ball Cave Fossils 367 Fig. 5. (3X). Photographs of the type specimen of Brachyprotorna brevirimla (BYUVP 7490) in palatal and right side ' (smallest) socket of each Ml/ and extending posteriorly to include the entire anterior wall of the braincase (BYUVP 7490, Fig. 5). Only the right P4/ was found in situ, but a right and left Ml/ (previously cataloged as BYUVP 7492 and 8298, respectively) fit perfectly into the sockets of the type specimen, where they have been permanently mounted. The type specimen is of a young adult based on com- plete fusion of the premaxillae, maxillae, nasals, and frontals and on lack of significant tooth wear. Both the skull and isolated Ml/'s were recovered from site 1, channel A, Crys- tal Ball Cave, Millard County, Utah (Figs'. 1 and 3) by Wade E. Miller and party 19 March 1977. The type specimen is housed at the Brigham Young University Vertebrate Pale- ontology Laboratory. Diagnosis. — Brachyprotorna brevimala has a short face and a maxillary tooth formula of I3/-3/, C1/-1/, P2/-2/, M1/-1/ as in B. ob- tusata. Face and maxillary dental measure- ments average 15% smaller than those of B. ohtusata. Brachyprotorna brevimala is distin- guished from B. obtusata by P4/ being trans- versely narrower and having a more posteri- orly directed lingual cusp and by Ml/ being more reduced and distinctly shorter antero- posteriorly. In other known characters B. bre- vimala is equivalent to B. obtusata. Brachyprotoma brevimala has the most re- duced P4/ and Ml/ and the shortest maxilla of any known skunk, and it is for this latter char- acter that the species is named. Description. — The maxillary dental for- mula of I3/-3/, C1/-1/, P2/-2/, M1/-1/ is known among the mustelids only in two genera of skunks, Conepatus and Brachyprotoma (al- though I found one abnormal Recent Spilo- gale putorius specimen with this formula). The Crystal Ball Cave specimen is clearly a skunk (subfamily Mephitinae) based on the presence of only two pairs of upper premolars (mephitines have two or three, all other mustelids have three or four), the small size (only the subfamilies Mustelinae and Mephitinae have such small adult individu- als), the lingual cusp of P4/ extending from the middle of the tooth (as opposed to the more anterior extention in the mustelines), Ml/ be- ing anteroposteriorly shorter labially than lin- gually (mustelines have the opposite condi- tion), and the internal nares extending almost as far anteriorly as the posterior end of the tooth row (they are much more posterior in mustelines). Compared to extant mephitines, the Crys- tal Ball Cave specimen represents an individ- ual of similar size to Spilogale but much smaller than Conepatus and Mephitis. The palate is shorter and wider than that of Spilo- gale putorius, but the interorbital breadth 368 Great Basin Naturalist Vol. 45, No. 3 shows that the type specimen represents a larger individual than the average S. putorius. The P4/ is similar to Spilogale, differing only in having the lingual cusp slightly more poste- rior, but it is proportionally much narrower than the P4/ of Mephitis and Conepatus. The Ml/'s are proximo-distally shorter than any of the living mephitines (especially Conepatus and Mephitis that have large square Ml/'s) and are closest to Spilogale in shape and cusp pattern. The rostrum of the type specimen is shorter than that oi Spilogale, matching that of Conepatus in proportions. The external nare is steep as in Conepatus, but it is relatively small and round as in Spilogale. Both infraor- bital canals are single rather than double or triple, a species-diagnostic character in Conepatus (Hall 1981) but variable in Mephi- tis and Spilogale. In addition to the three extant genera, three Pleistocene skunk genera have been named: Buisnictis, Brachyprotoma, and Os- motherium (Kurten and Anderson 1980). Os- motherium can be ruled out since it is large and very similar to Mephitis (Kurten and An- derson 1980), the living skunk genus that is most distinct from the Crystal Ball Cave speci- men. Both Buisnictis and Brachyprotoma are small and have proportionally short jaws like the Crystal Ball Cave specimen. Buisnictis has been recovered from Late Pliocene de- posits of southwestern Idaho (Bjork 1970) and Middle Pliocene to Early Pleistocene deposits of Kansas and Oklahoma (Hibbard 1941, 1950, 1954), but it has no record in the Late Pleis- tocene or Recent. Buisnictis has a short jaw with crowded premolars, but it differs from the Crystal Ball Cave specimen in having three pairs of upper premolars instead of two (Kurten and Anderson 1980). Based on an illustration by Hibbard (1954), the P4/ o( Buis- nictis has its lingual cusp extending from the anterior part of the tooth as in the mustelines, and the Ml/ is distinctly longer than that of the Crystal Ball Cave specimen. These mor- phologic and age differences show that the Crystal Ball Cave specimen is distinct from Buisnictis. The Crystal Ball Cave specimen matches the genus Brachyprotoma in having short jaws, only two jiairs of upper premolars, P4/ and Ml/ similar in shape and cusp pattern to Spilogale, and in the age of deposits in which they have been recovered. Until recently Brachyprotoma was only known from a few Early Pleistocene to Early Recent age cave deposits in the eastern United States. But during the period of this study, P. M. Young- man (1984, pers. comm.) recovered several Brachyprotoma specimens from two fossil sites in the Yukon Territory of Canada. Al- though no previous Brachyprotoma speci- mens have been reported closer than 1880 km (1130 miles) from Crystal Ball Cave, morphol- ogy clearly allies the Crystal Ball Cave speci- men with this genus. But there are specific differences between the Crystal Ball Cave specimen and other skulls that have been as- signed to the genus Brachyprotoma. To test the amount of variation to be expected within a species of skunk, 73 specimens of Recent Spilogale putorius were measured, 60 from the Harvard University Museum of Compara- tive Zoology and 13 from the Brigham Young University Monte L. Bean Museum. Spilo- gale putorius makes a good standard for the expected individual variation in species of Brachyprotoma, both because Spilogale is probably the most closely related extant genus to Brachyprotoma and because S. puto- rius borders on being divisible into multiple species (although most workers presently con- sider it a single species). Based on the great amount of variation seen between the Crystal Ball Cave specimen and other skulls assigned to the genus Brachyprotoma compared with the amount of variation seen among individu- als of S. putorius, I believe the Crystal Ball Cave specimen warrants the status of a new species. The B. brevimala type is smaller than speci- mens of B. obtusata in most measured charac- ters, averaging about 15% smaller (Table 3). The greatest differences occur in P4/ and Ml/, which are the most varied niiixillary teeth be- tween skunk taxa. The mean length of P4/ in B. obtusata is only 7% greater than in B. brevimala, although the mean width is 22% greater. The lingual cusp of P4/ in B. brevi- mala also points more posteriorly than in B. obtusata, being nearer Ml/ at its lingual tip rather than closer at its base or parallel as in B. obtusata. The Ml/of B. obtusata is 16% trans- versely wider on the average, but the labial anteroposterior length is 30% greater and the lingual anteroposterior length is 59% greater than in the B. brevimala type on the average. July 1985 Heaton: Crystal Ball Cave Fossils 369 Table 3. Skull measurements of Brachijprotoma specimens and mean skull measurements of Brachyprotoma ohtusata and Spilogale putorius. Brigham Young University Vertebrate Paleontology (BYUVP) 7490 is from Crystal Ball Cave, Utah. American Museum of Natural History (AM NH) 12426 and 11772 are from Connard Fissure, Arkansas (Brownl908, Hall 1936). U.S. National Museum (USNM) 8155 is from Cumberland Cave, Maryland (Gidley and Gazin 1938, Hall 1936). Carnegie Museum (CM) 11057A and 20233 are from Frankstown Cave, Pennsylvania (Hall 1936, Peterson 1926, P. M. Youngman 1984, pers. comm.). A skull misidentified as Carnegie Museum (CM) 308 (here listed as Cra. Pit) is from Crankshaft Pit, Missouri (Oeseh 1967, Parmalec ct al. 1969). Starred measurements are based on photos only. Measurements are in millimeters. The coefficients of variability (C. V.) have been multiphed by 100. BYUVP AMNH AMNH USNM CM CM Cra. B. obtumta S. putorius Location of measurement 7490 12426 11772 8155 11057A 20233 Pit Mean C.V. Mean C.V. Width between orbits 17.1 18.0 _ 18.2 — 17.0 17.7 3.63 17.0 7.52 Width between outer molars 19.8 20.5 — 20.1 19.6 20.1 2.25 19.3 6.62 Length of rostrum to internal nare 18.6 17.0* 18,0* — — 17.4 — 17.5 2.88 20.3 8.49 Length of cheek teeth series 11.2 11.0* 12.1 12.2 — 11.7 14.2* 12.2 9.75 14.2 6.80 P4/ anteroposterior length 5.7 5.9 6.2 6.2 6.4* 5.7 6.3* 6.1 4.32 6.0 8.80 P4/ greatest transverse width 3.3 3.8 3.9 3.9 4.0* 3.5 4.5* 3.9 8,. 30 3.7 9.97 Ml/ labial anteroposterior length 3.0 .3.8 4.1 4.0 3.8* 3.9 3.9* 3.9 2.98 4.8 7.96 Ml/ lingual anteroposterior length 2.2 3.0 3.3 3.8 3.5* 3.5 3.6* 3.5 7.94 4.0 9.16 Ml/ greatest transverse width 5.0 5.7 6.0 6.3 5.8* 5.3 5.7* 5.8 5.77 5.7 6.97 Since there is only minor variation in these characters among B. ohtusata skulls (Table 3) but distinct difference between them and the Crystal Ball Cave specimen, and because the differences between the B. brevimala type and specimens of B. ohtusata are far greater than would be expected within a species (based on the variation found among 73 indi- viduals o( Spilogale putorius, the most closely related extant species), erection of a new spe- cies for the Crystal Ball Cave specimen is clearly justified. Discussion. — Brachyprotoma specimens have been previously recovered from the fol- lowing deposits of Early Pleistocene to Early Recent age: Port Kennedy Cave and Franks- town Cave, Pennsylvania (Cope 1899, Peter- son 1926), Cumberland Cave, Maryland (Gid- ley and Gazin 1938), Crankshaft Cave and Brynjulfson Cave, Missouri (Oesch 1967, Par- malee and Oesch 1972, Parmalee et al. 1969), Connard Fissure, Arkansas (Brown 1908), and two sites in northern Yukon Territory, Canada (P. M. Youngman 1984, personal communica- tion). Most of these specimens are lower jaws, and the only seven skulls (or skull fragments) that have been previously reported are Carnegie Museum 11057A and 20233, Ameri- can Museum of Natural History 11772 and 12426, U.S. National Museum 8155 and 11960, and a specimen identified as Carnegie Museum 308 by Oesch (1967) but which does not correspond to that number in the Carnegie Museum catalogs (M. R. Dawson 1984, pers. comm.). Parmalee et al. (1969) illustrated this latter specimen but did not identify it by catalog number. Cope (1899) named Mephitis (Spilogale) ohtusatus for a single small dentary from Port Kennedy Cave, but E. D. Cope died before the completion of this paper, and a footnote stated that "none of the specimens labelled by Prof Cope bear this name." Brown (1908) named the genus Brachyprotoma, and he considered M. ohtusatus to belong to this genus as well as M. fossidens and M. leptops, two species named by Cope previous to the naming of M. ohtusatus. From Connard Fis- sure Brown (1908) reported B. fossidens, B. leptops, and B. ohtusata based on dentaries, and he named B. pristina based on two partial skulls and three dentaries (the skull cataloged as American Museum of Natural History 12426 is the type for the type species of this genus) and B. spelaea based on one dentary. The dentaries Brown (1908) identified as B. fossidens and B. leptops are far too large to belong to the same genus as the small speci- 370 Great Basin Natur.\list Vol. 45, No. 3 mens he identified as B. obtusata, B. pristina, and B. spelaea, and no one since has consid- ered these two species as belonging to the genus Brachyprotoma. Later Hay (1923) named B. putoria from Frankstown Cave. The naming of multiple species of Brachyprotoma in the early publications listed above has been widely criticized by later workers because the amount of variation among specimens is less than that seen within living species. Hall (1936) and Kurten and Anderson (1980) considered the genus Brachyprotoma to be clearly monotypic, with the^only valid species being B. obtusata, the earliest named species that can be applied to the genus. The Brachyprotoma skull from Crystal Ball Cave is the first specimen of Brachyprotoma distinct enough from B. ob- tusata to warrant the erection of an additional species of this genus. Concerning the paleoecology of Brachyprotoma, Kurten and Anderson (1980) stated that this genus has always been associ- ated with boreal faunas, although other skunk genera were also recognized at each site. This matches the "more boreal than present" na- ture of the Crystal Ball Cave assemblage and suggests that the post-Pleistocene climatic shift may have lead, directly or indirectly, to the extinction o{ Brachyprotoma. Since fossils of Brachyprotoma are only found in a few deposits and even then are few in number, this genus probably never had a high density of individuals in life. The Brachyprotoma brevimala type was first misidentified as Spilogale (Heaton 1984), the most similar living genus. Miller (1982) reported cf Spilogale from Crystal Ball Cave, possibly based on this same specimen. Mephi- tis was also mentioned in my preliminary re- port (Heaton 1984), but further examination proved that the anterior right dentary (BYUVP 7489) upon which the identification was based was equally referable to Martes americana, which is represented by addi- tional material. Although both Mephitis mephitis and Spihgale piitorius (^gracilis) now inhabit the Snake Range (Hall 1981), and Spilogale has been recovered from deposits over 12,000 years old in Smith Creek Cave (Mead et al. 1982), their presence is uncon- firmed in the Crystal Ball Cave assemblage. Since Brachyprotoma seems to have lived contemporaneously with other skunk genera, it is interesting to speculate about how their niches varied. All living skunks tend to be nocturnal and omnivorous, so they are rarely tied to specific foods or habitats. Minor niche differences do occur between living North American genera: Spilogale is the most car- nivorous. Mephitis the most herbivorous, and Conepatus the most insectivorous. Spilogale has narrow sharp teeth, Conepatus at the other extreme has very broad teeth, and Mephitis is intermediate but has the longest tooth rows. Brachyprotoma (especially B. brevimala) has pushed the narrowing of the teeth seen in Spilogale to an extreme, con- verging on the carnivorous genus Mustela. This suggests that Brachyprotoma was more carnivorous than any of the living skunks. Why Brachyprotoma lost P2/ and short- ened its tooth rows, paralleling the genus Conepatus, is a mystery. Members of the genus Mustela have longer tooth rows than skunks, so in that respect Brachyprotoma di- verged from Mustela. Brachyprotoma was trending in a direction that is difficult to ex- plain. Brachyprotoma also did not survive the post-Pleistocene changes as did the aforemen- tioned genera (although some species were lost and ranges altered). I propose that these two facts are correlated. Brachyprotoma was probably adapting to a specialized niche that existed during the Pleistocene but disap- peared during the Recent. 1 also propose that this specialization was a feeding habit and/or preference for a particular prey item since the specializations discussed are all dental. No postcranial material has been reported to doc- ument additional specializations, and the most diagnostic skunk characters, scent glands and color patterns, are in the soft anatomy, which is obviously unavailable. With such limited data (about 27 specimens from nine sites), further speculation would be unwarranted. All that can be concluded is that Brachyprotoma was restricted to boreal con- ditions, was widespread in North America, was probably low in density, and did not sur- vive the post-Pleistocene changes. The evolution of the genus Brachyprotoma has been discussed by Kurten and Anderson (1980). They stated that it seems most closely related to Spilogale, but both were probably derived from the Mio-Pliocene genus July 1985 Heaton: Crystal Ball Cave Fossils 371 Promephitis. No intermediate forms are avail- able to show the exact phylogeny, however. Some speculation can be made about the rela- tionship of B. brevimala to B. ohtusata. Brachyprotoma brevimala has gone to a greater extreme in the characters that differ- entiate Brachyprotoma from other skunks (shorter face and narrower teeth) and is there- fore more specialized. Since specialists almost always evolve from generalists, B. brevimala probably evolved from B. obtusata. The fact that B. obtusata has been found in deposits from Early Pleistocene to Early Recent age (Kurten and Anderson 1980) and B. brevimala is known only from a Late Pleistocene to Re- cent deposit also supports this conclusion. Family Felidae Smilodon cf. fatalis Material. — Partial left ectocuneiform (BYUVP 7530), claw (BYUVP 7497). Miller (1982) reported cf Smilodon from Crystal Ball Cave based on a single vertebra (W. E. Miller 1983, pers. comm.), but this specimen is ap- parently lost (possibly due to an explosion that affected the collection). Discussion. — The ectocuneiform is dense, worn, and coated with a calcite crust. The claw is missing the outer plates but is otherwise in good condition. The specimens were compared with Smilodon and Felis atrox, the only two Late Pleistocene cats large enough to be considered, and both compare best with Smilodon (W. E. Miller 1984, pers. comm.). The ectocuneiform was previously referred to Panthera atrox (Heaton 1984), but comparison with actual specimens rather than casts shows that it is clearly Smilodon. The only previous citing of Smilodon in Utah is from the Silver Creek fauna of north central Utah (Miller 1976), but it has been found in Pleistocene assemblages throughout North America. Kurten and Anderson (1980) considered S. fatalis to be the only valid species of Late Pleistocene Smilodon in North America, but it has been known by many other names. Based on this synonymy, the Crystal Ball Cave specimens are referred to S. fatalis al- though they are doubtfully identifiable to the species level. Felis concolor Material.— First right metacarpal (BYUVP 7502), 4 claws (BYUVP 7498-7501). Discussion. — Felis concolor is the only cat of its size presently living in North America, but similar-sized species oiAcinonyx and Ho- motJierium existed during the Pleistocene. Lyiix and other species of Felis (disregarding those often placed in the genus Panthera ) are distinctly smaller than F. concolor, and Smilodon, Panthera atrox, and P. onca are distinctly larger. The Crystal Ball Cave speci- mens were compared with material of Felis species, Acinonyx, and Homotherium at the Los Angeles County Museum and found to match perfectly in size and shape with speci- mens of F. concolor; but they clearly differ from the other felids mentioned. Felis con- color presently lives throughout the Snake Range (Hall 1981), and J. C. Bates (1983, pers. comm.) reported a citing in the Snake Valley near Candy as well as many higher in the mountains. Lynx cf rufus Material.— Right CI/ (BYUVP 7494), right P/4 (BYUVP 7496). The anterior portion of a right maxilla without teeth (BYUVP 7495) is probably also referable to Lynx. Discussion. — Lynx rufus currently inhab- its the area of Crystal Ball Cave (J. C. Bates 1983, pers. comm.), but L. canadensis ranges only as close as central Utah and northward and prefers colder climates (Hall 1981). Lynx canadensis is slightly larger than L. rufus and has considerably larger feet (Ingles 1965). The specimens recovered fall in the size range of both L. rufus and L. canadensis, but they tend to be closer in size to L. rufus. None of the claws recovered could be referred to this genus, so the difference in foot size was not helpful. Since L. rufus presently lives around the cave, the specimens are referred to it. Order Perissodactyla Family Equidae Equus cf scotti Material. — Left cuneiform (BYUVP 7542), right lunar (BYUVP 7544), 2 right scaphoids (BYUVP 7549, 7550), right mag- num (BYUVP 7561), second phalanx (LACM 372 Great Basin Natur\list Vol. 45, No. 3 Table 4. Measurements oiEquus first phalanges from Crystal Ball Cave. All measurements are in millimeters and parallel to the main bone axes. Maximum proximo- Maximum proximal width Minimum medial width Maximum distal width distal Antero- Antero- Antero- Catalog number length Transverse posterior Transverse posterior Transverse posterior BYUVP 7580 76.8 38.1 28.6 27.0 17.7 .35.8 20.8 BYUVP 7581 72.0 38.3 .30.6 24.2 16.8 33.2 20.0 BYUVP 7582 — — — — 16.4 16.1 BYUVP 7583 75.8 .38.6 .30.3 24.8 17.4 33.9 19.5 LACM 123681 74.9 39.7 31.5 27.0 18.0 37.5 21.7 LACM 123682 74.1 — 31.8 25.4 17.6 34.4 20.4 Table 5. Measurements ofEquus second phalanges from Crystal andf)arallel to the main bone axes. all Cave. All measurements are in millimeters Maximum proximo- Maximum proximal width Minimum medial width Maximum distal width distal Antero- Antero- Antero- Catalog number length Transverse posterior Transverse posterior Transverse posterior BYUVP 7587 — — 27.7 — — — BYUVP 7588 47.5 48.7 31.7 42.3 21.7 46.5 25.6 BYUVP 7589 44.2 41.5 28.1 33.2 19.4 35.5 + 23.9 BYUVP 7590 — — 25.4 — — — BYUVP 7591 .39.3 + — 27.0+ — 21.2+ — 25.5 BYUVP 7593 37.0 .39.1 22.8 .30.4 31.1 + 17.1 BYUVP 7594 36.5 31.9 24.0 28.2 15.6 30.4 19.2 LACM 123683 51.8 62.0 .35.2 .52.7 24.1 .57.1 28.8 LACM 123684 45.2 44.0+ .30.8 37.2 21.2 37.1 24.5 LACM 123685 40.6 42.1 26.7 .36.7 19.5 39. 1 25.0 Table 6. Measurements of Equiis third phalanges from Crystal Ball Cave. All measurements are in millimeters and parallel to the main bone axes. Anterior Maximum transverse Maximum antero- Articulation surface Catalog number height width posterior width Transverse Antero-posterior BYUVP 7595 40.6 61.7 44.2 49.2 22.4 BYUVP 7.596 39.4+ — BYUVP 7597 32.7 39.6 ,38.0+ 36.0 13.8 BYUVP 7600 47.0+ 48.7 49.1 + 37.8 14.8 BYUVP 7601 .34.0 37.1 .34.6+ ,34.0 16.6 BYUVP 7602 — — — 34.4 14.0 BYUVP 7603 — — — 41.0 13.6 BYUVP 7605 35.2 41.1 38.0 37.3 13.6 BYUVP 7606 ,37.7 — — 33.7 14.1 BYUVP 7607 40.8 — — 29.2 15.3 BYUVP 7608 39.0 40.7 13.8 BYUVP 7610 — 33.2 27.8 + — — 123683), third phalanx (BYUVP 7595). A juve- nile left P/2 (BYUVP 7623), a partial juvenile first phalanx (BYUVP 7586), a .second phalanx (BYUVP 7588), 3 partial third phalange.s (BYUVP 7596, 7607, 7608), and a distal sesamoid (BYUVP 7622) probably belong to this species also. Phalanx measurements are listed in Tables 4, 5, and 6. Discussion. — Several species of large horses have been recognized from the Late Pleistocene of western North America. The Rancho La Brea asphalt deposits have yielded a single species of large horse (Savage 1951) usuallv referred to E. occidentalis (Merriam 1913, Stock 1963, Willoughby 1974), although the validity of this name has been ({uestioned (Miller 1971). Based on comparative material and measurements made by Willoughby (1974), the large Crystal Ball'Cave horse is distinct from the Rancho La Brea horse in July 1985 Heaton: Crystal Ball Cave Fossils 373 85 E I 31. E. occjdentalis . E. conversidens E. pacificus E. scotti E. niobrarensis 22 26 30 34 38 Minimum Lateral Width 46 50 Fig. 6. Plot oiEquus first phalanges from Crystal Ball Cave (X's) and ranges of variation for some Late Pleistocene North American species (circumscribed). The number of specimens plotted to show the range of variation were 46 of £. conversidens, 9 of £. niobrarensis, 6 of £. occidentalis, 6 of £. pacificus, and 2 of £. scotti. These measurements were taken from Dalquest and Hughes (1965), Gazin (1936), A. H. Harris (1984, pers. comm.), and Harris and Porter (1980). Measurements are in millimeters. having more transversely broad phalanges (Figs. 6, 7, and 8) and carpals with relatively larger articulation surfaces. The Crystal Ball Cave specimens are distinctly larger than E. niobrarensis based on measurements given me by A. H. Harris (1983, pers. comm.) and in Harris and Porter (1980). Harris (1983, pers. comm.) also provided me with measurements of E. pacificus (although the validity of this species has been questioned by Savage 1951) from Fossil Lake, Oregon, and phalanges of this species match well in size with the large Crystal Ball Cave horse but are not as trans- versely broad. Gazin (1936) listed measurements of the type specimen of E. scotti, and, of all speci- mens and data seen, only it has phalanges that are as transversely broad as the Crystal Ball Cave specimens. The second phalanx (LACM 123683) is slightly larger than the E. scotti type but has identical proportions (Fig. 7), and the third phalanx (BYUVP 7595), al- though smaller because it is of a subadult, has the same proportions as the anterior third phalanges of the E. scotti type (Fig. 8). Dalquest (1964) stated that E. scotti was very heavily built, and this would suggest that the foot and toe bones are broad compared with 374 58- ^50- X S42 34 Great Basin Naturalist E. occidentalis Vol. 45, No. 3 E. pacificus E. scotti . conversidens 24 32 40 48 Minimum Lateral Width 56 Fig. 7. Plot of Equus second phalanges from Crystal Ball Cave (X's) and ranges of variation for some Late Pleistocene North American species (circumscribed). The number of specimens plotted to show the range of variation were 26 of £. conversidens, 3 of £. niobrarensis, 8 of £. occidentalis, 4 of £. pacificus, and 2 of £. scotti. These measurements were taken from Dalquest and Hughes (1965), Gazin (1936), A. H. Harris (1984, pers. comm.), and Harris and Porter (1980). Measurements are in millimeters. 62- E. occidentalis (3_^!l — ---^.^^cificus £54- cr\ Ja 1 E. ntobrarensis \^\e. scotti L X 1 /■ ^ X 38 X X X L ^. conversidens 34 42 58 66 Maximum Lateral Width 82 90 Fig. 8. Plot oi Equus third phalanges from Crystal Ball Cave (X's) and ranges of variation for some Late Pleistocene North American species (circumscribed). The number of specimens plotted to show the range of variation were 6 off. conversidens, 5 of £. niohrarensis, 1 of £. occidentalis, 2 of £. pacificus, and 2 of £. scotti. These measurements were taken from Dalquest and Hughes (1965), Gazin (1936), A. H. Harris (1984, pers. comm.), and Harris and Porter (1980). Measurements are in millimeters. July 1985 Heaton; Crystal Ball Cave Fossils 375 other species ofEquus. The large carpals from Crystal Ball Cave mentioned above, espe- cially the cuneiform and magnum, are broad and have much larger articulation surfaces than the Rancho La Brea horse. Based on this limited information in the literature, the largest carpals listed above compare most fa- vorably with E. scotti also. Equus scotti was originally named and de- scribed from Texas by Gidley (1900), and most specimens have been found in that state (Dalquest 1964, Gidley 1903, Johnston 1937). Hopkins et al. (1969) recovered a left metatarsal from the Late Pleistocene Ameri- can Falls Lake Beds of southeastern Idaho that they referred to E. scotti. It is therefore not unlikely that E. scotti lived in Utah. A large horse was represented at Smith Creek Cave by a single vestigial metapodial (Miller 1979), but no attempt was made to identify it to species. BYUVP 7588 is not as laterally broad as LACM 123683 but is too large to belong with the smaller species. The epiphysis is not fully fused, showing that it represents a subadult. It is the only bone from Crystal Ball Cave that matches well with the Rancho La Brea horse, although it is slightly smaller. But, since it may differ by only individual, foot, or age variation from the better represented E. cf. scotti, it is tentatively referred to that species. Equus ? conversidens Material — Right M3/ (LACM 123677), thoracic vertebra (BYUVP 7687), 3 right pisi- forms (BYUVP 7536-7538), left pisiform (BYUVP 7539), 2 right cuneiforms (BYUVP 7540, 7541), 4 right lunars (BYUVP 7543, 7545-7547), partial left lunar (BYUVP 7548), 4 right scaphoids (BYUVP 7551-7554), 4 par- tial left scaphoids (BYUVP 7555-7558), 2 right trapezium-trapezoids (BYUVP 7559, 7560), 2 right magnums (BYUVP 7562, 7563), partial right magnum (BYUVP 7564), left magnum (LACM 123678), 2 partial right unciforms (BYUVP 7565, 7566), proximal tibia epiphysis (BYUVP 7570), distal epiphysis of right tibia (BYUVP 7571), partial distal epiphysis of left tibia (BYUVP 7572), right calcaneum (LACM 123679), left calcaneum (BYUVP 7573), right astragulus (BYUVP 7575), right juvenile as- tragulus (BYUVP 7574), left astragulus (LACM 123680), right navicular (BYUVP 7576), left navicular (BYUVP 7577), left cuboid (BYUVP 7579), right meso-ento (BYUVP 7578), proximal portion of left metatarsal (BYUVP 7567), 2 distal metapodial epiphyses (BYUVP 7568, 7569), 6 first pha- langes (BYUVP 7580, 7581, 7583, LACM 123684, 123685), 3 partial first phalanges (BYUVP 7582, 7584, 7585), 5 second pha- langes (BYUVP 7589, 7593, 7594, LACM 123684, 123685), 4 partial second phalanges (BYUVP 7587, 7590-7592), 5 third phalanges (BYUVP 7597, 7600, 7601, 7605, 7606), 2 par- tial third phalanges (BYUVP 7602, 7603), ju- venile third phalanx (BYUVP 7610), 11 proxi- mal sesamoids (BYUVP 7611-7621). Phalanx measurements are listed in Tables 4, 5, and 6. Discussion. — In addition to the fossils of large horses from Crystal Ball Cave (referred to E. cf. scotti) are numerous bones of smaller horses. Some of these compare well with E. conversidens, the species to which most small Pleistocene North American horse fossils have been assigned, but others do not. Con- siderable time has been spent evaluating the size and morphologic variation among these bones and comparing the results with descrip- tions and measurements in the literature. But both complexities within this collection and disagreements regarding valid species in the literature have prevented positive species identification of these small horse bones. Equus conversidens (Owen 1869) has been considered by some workers to be the only valid species of small Pleistocene horse in North America (Harris and Porter 1980, Miller 1971), and most other named species of small Pleistocene horses have at some time been synonymized with this species (Dalquest and Hughes 1965, Hibbard 1955, Hibbard and Taylor 1960). However, most workers presently recognize at least two spe- cies. Owen (1869) named E. tau at the same time he named £. conversidens. Poor illustra- tions of the type specimens have caused some workers to consider E. conversidens and E. tau synonymous (Hibbard 1955). But Dalquest (1979) and Mooser and Dalquest (1975), after researching the early descrip- tions (the type specimen of E. tau is lost), considered these two species distinct. The teeth that Mooser and Dalquest (1975) as- signed to E. tau are smaller than those of £. 376 Great Basin Naturalist Vol. 45, No. 3 conversidens, and the metapodials are longer and more slender. Skinner (1942) assigned a first phalanx from Papago Springs Cave, Ari- zona, to £. tau because it was much narrower than those of £. conversidens from the same assemblage. But, based on his measurements, this phalanx is narrower transversely than anteroposteriorly, making it doubtful of being horse at all. Hay (1915) named E. francisci, which was synonymized with E. conversidens by Hib- bard and Taylor (1960). But Lundelius and Stevens (1970) reprepared the metatarsal of the type specimen and found it to be distinctly longer and narrower than that of £. conversi- dens. Lundelius and Stevens (1970) therefore considered E. francisci a valid species, and they synonymized E. quinni (based on the similar long metatarsal) and Onager zoijatalis (based on dental similarities) to it. Dalquest (1979) considered E. francisci, as well as E. littoralis, E. achates, and E. qiiinni, to be synonymous with E. tan, and he considered E. conversidens and E. tau the only two valid species of small Pleistocene North American horses. Based on an illustration in Lundelius and Stephens (1970), the M3/ of E. francisci is distinctly wider transversely than that of E. conversidens, although they are of similar anteroposterior length. LACM 123677, al- though quite worn, has the same width and length as the E. franscisci type and has an enamel pattern most similar to it also. Dalquest (1979) synonymized E. francisci with E. tau, but the M3/ of the lectotype of £. tau illustrated by Mooser and Dalquest (1975) is not transversely broad like the E. francisci type and Crystal Ball Cave M3/. Unfortu- nately the only phalanx measurements given in the literature are for E. conversidens, ex- cept the questionable first phalanx assigned to E. tau by Skinner (1942). The only phalanges from Crystal Ball Cave that compare well with measurements of E. conversidens phalanges in the literature are three of the five second phalanges (Fig. 7). The other two second phalanges (BYUVP 7593, 7594) are distinctly smaller than any assigned to E. conversidens yet have com- plete epiphyseal fusion. All nine first pha- langes are from individuals intermediate in size between those represented by the two sets of second phalanges, and all are small compared with the first phalanges assigned to £. conversidens in the literature (Fig. 6). Six of the eleven third phalanges articulate well with the three larger second phalanges yet are smaller than the third phalanges assigned to £. conversidens in the literature (Fig. 8). The other third phalanx (BYUVP 7600) is larger than any assigned to £. conversidens and too large to articulate with any of the second pha- langes under discussion. It is important to consider sexual dimor- phism, individual variation, and variation among different feet of the same individual to see how much variation is expected within a species. Willoughby (1974), in a table of bone measurements from 25 species and races of Equus, listed mean dimensions for both sexes with respect to two characters: metacarpal midwidth and metacaqjal midwidth divided by length. Metacarpals of males had a mid- width of 3.1% to 7.3% greater than females and a midwidth divided by length of 2.3% to 6.9% greater than females. Species with more sexual dimorphism in metacarpal width tended to also have more dimorphism in width relative to length, so male metacarpals tend to be more robust and just slightly longer than female metacarpals. These measure- ments show that sexual dimorphism is not great in Equus and certainly not sufficient to have caused the variability seen among the small Crystal Ball Cave equids. Howe (1970), in a study of Equus {Plesip- pus) simplicidens , showed that individual variation in bone size can be greater than previously thought. Because the large num- ber of specimens at Nebraska's Broadwater Quarry fell into a single size curve with no gaps, he concluded that they all represent a single species, and he synonymized a number of species that had previously been named based on limited material at other sites. Table 5 of Howe (1970) shows that the largest metacarpal and metatarsal lengths and widths average 32% larger tlian the smallest corre- sponding measurements, and none are more than 36% larger. Even with a sample size of 97 to 190, the metapodials measured by Howe (1970) show less variation than do the few second and third phalanges from Crystal Ball Cave. July 1985 H EATON: Crystal BallCavk Fossils 377 Table 7. Measurements of first phalanges oi' Camelops cf. hestemus (C) and Hetniauchenia cf. macrocephala (H) from Crystal Ball Cave. All measurements are in millimeters and parallel to the main bone axes. Maximum proximo- Maximum proximal width Minimum medial width Maximum distal width distal Antero- Antero- Antero- Catalog number ID length Transverse posterior Transverse posterior Transverse posterior BYUVP 7627 c 117 39+ 35+ LACM 123689 C 114 43 35 LACM 123691 c — 32 32 BYUVP 7640 H 23 19 LACM 123690 H 100 28 32 20 15 18 33+ 28 19 35 29 18 — 22 Isolated front and rear phalanges are usu- ally indistinguishable and therefore have an additional degree of variation. Front and rear phalanx measurements were taken from re- cent £. cahallus and E. btirchelU specimens, and the larger measurements for each species averaged 4.2% larger than the smallest corre- sponding measurements with a maximum of 9.4% larger. But even this much variability, in addition to sexual and individual variation, does not adequately account for the great size range among the small Crystal Ball Cave equids. Six measurements of the 5 second pha- langes from Crystal Ball Cave (excluding those referred to E. scotti) show that the largest measurements are 24% to 43% larger than the smallest corresponding measure- ments with an average of 31.5% larger. Eleven measurements of the 9 third pha- langes from Crystal Ball Cave show that the largest measurements are 7% to 122% larger than the smallest corresponding measure- ments with an average of 50.7% larger. Con- sidering the second and third phalanges sepa- rately, each has enough variation to make it marginal whether they could all be assigned to the same species considering sexual, individ- ual, and foot variation. The variation seems even more extreme when one considers that the smallest second phalanges (BYUVP 7593, 7594) are from much smaller individuals than the smallest third phalanx, and the largest third phalanx (BYUVP 7600) is from a larger individual than the largest second phalanx. This is far more variation than can be ac- counted for by the sexual, individual, and foot variation for a single species as discussed above, and it suggests that multiple species of horse smaller than E. cf. scotti are repre- sented at Crystal Ball Cave. Finding a dividing line between two spe- cies in this material is nearly impossible, how- ever. Most of the material could be assigned to a species of horse 15% smaller than the smallest material assigned to E. conversidens, but the two smallest second phalanges (BYUVP 7593, 7594) and the largest third pha- lanx (BYUVP 7600) seem too far from the mean to belong to this supposed species. Un- til more phalanx measurements are available for small Pleistocene horses other than E. con- versidens, it is difficult to determine how many species are represented by the smaller Equus fossils from Crystal Ball Cave and whether most of the material represents an unusually small variety of E. conversidens, a species distinct from E. conversidens such as E. tail and/or E. francisci, or both. Order Artiodactyla Family Camelidae Camelops cf. hestemus Material. — Right scaphoid (LACM 123686), left scaphoid (LACM 123687), left lunar (BYUVP 7624), left magnum (BYUVP 7625), right unciform (BYUVP 7626), distal fragment of metapodial (BYUVP 7629), 2 first phalanges (BYUVP 7627, LACM 123689), proximal portion of first phalanx without epi- physis (LACM 123691), partial proximal epi- physis of first phalanx (BYUVP 7638), 3 sec- ond phalanges (LACM 123692, BYUVP 7630, 7632), 3 proximal portions of second pha- langes (BYUVP 7633, 7634, 7637), 3 partial proximal portions of second phalanges (BYUVP 7628, 7635, 7636), 3 third phalanges (BYUVP 7639, 7641, 7642). Six sesamoids (BYUVP 7644-7649) are probably of Camelops but may represent Bison. Phalanx measurements are listed in Tables 7, 8, and 9. 378 Great Basin Naturalist Vol. 45, No. 3 Table 8. Measurements of second phalanges oiCainelops cf. hesternus (C ) and Hemiauchenia of. macrocephala {H ) from Crystal Ball Cave. All measurements are in millimeters and parallel to the main bone axes. Maximum proximo- Maximum proximal width Minimum medial width Maximum distal width distal Antero- Antero- Antero- Catalog number ID length Transverse posterior Transverse posterior Transverse posterior BYUVP 7630 C 70+ 35 29 24 17 BYUVP 7632 C 67+ 37 31 29 19 BYUVP 7633 C 31 + 31 27 20 BYUVP 7634 C — 34+ 28+ — — BYUVP 7635 c — — 31 — — BYUVP 7636 C 26+ — — BYUVP 7637 C — 34+ 30 — 19 LACM 123692 c 69 37 31 26 19 BYUVP 7631 H 52 28 23 20 14 30- 22 Table 9. Measurements of third phalanges ofCamelops cf hesternus from Crystal Ball Cave. All measurements are in millimeters and parallel to the main bone axes. Catalog number Maximum proximo- distal length Maximum proximal width Transverse Antero-posterior BYUVP 7639 BYUVP 7641 BYUVP 7642 23+ 29+ 33+ Discussion.— Webb (1965, 1974) recog- nized only four valid genera of Late Pleisto- cene North American camels: Titanotylopus, Camelops, Heiniauchenia { = Tanupolama), and Paleolama (in order of decreasing size). Titanotylopus is somewhat common and Camelops is very common in Late Pleistocene assemblages of western North America, but neither has been found in the east (Webb 1974). Hemiauchenia is found in Late Pleis- tocene deposits throughout the Americas (Webb 1974) and is commonly associated with Catnelops (Miller 1979). Paleolama has only been found in Florida, Texas, and southern California in Pleistocene deposits of North America (Miller 1976). Miller (1982) identi- fied Catnelops and Hemiauchenia from Crys- tal Ball Cave. The specimens listed above fall within the range of variation oi Camelops hesternus mea- surements from Rancho La Brea, southern California (Webb 1965) and Selby and But- ton, eastern Colorado (Graham 1981). T. E. Downs (1984 pers. comm.) provided me with 8 first phalanx measurements of Titanotylopus sp., 21 of Catnelops hesternus, and 21 of Hemiauchenia sp. from southern California deposits. Those of Titanotylopus range from 105 to 138 mm in length with an average of 121 mm, those oi Camelops hesternus range from 105 to 125 mm in length with an average of 116 mm, and those of Hemiauchenia range from 91 to 110 mm in length with an average of 94 mm. The two complete first phalanges from Crystal Ball Cave, both of which are of adults based on epiphyseal fusion and bone density, measure 114 and 117 mm in length (Table 7). Although there is some overlap in first pha- lanx length between these genera, the Crystal Ball Cave specimens clearly match best with Camelops. Savage (1951) recognized four valid species o( Camelops: C. hesternus and C. huerfanen- sis, which are larger, and C. sulcatus and C. minidokae, which are smaller; and Webb (1965), in his detailed description of Camelops, supported this system. Based on limb bone measurements given by Savage (1951), C. minidokae was about 14% smaller than C. hesternus. Camelops huerfatiensis can only be distinguished from C. hesternus and C. sulcatus can only be distinguished from C. minidokae based on dental characters (Graham 1981, Savage 1951). Both C. minidokae and C. sidcatus are too small to match the Crystal Ball Cave specimens, and both are known only from pre-Wisconsinan deposits (Kurten and Anderson 1980). Camelops hesternus and C. huerfanensis are very similar and may be conspecific (Hop- July 1985 Heaton; Crystal Ball Cave Fossils 379 kins 1955, Savage 1951). Both are known from the Late Pleistocene, and both are known from Idaho (Gazin 1935, Hopkins 1955, Hop- kins et al. 1969) and Colorado (Cragin 1892, Graham 1981). Camelops hesternus is the only species oiCamelops reported from Utah. A Camelops hesternus skull was recovered from a lava tube 140 km (87 miles) east-south- east of Crystal Ball Cave (Homer 1928, 1929) and dated at 11,075 ± 225 Y. B. P. (Nelson and Madsen 1979). Camelops cf hesternus was reported from the Silver Creek fauna in north central Utah (Miller 1976). Camelops sp. was reported from Smith Creek Cave (Harrington 1934, Stock 1936, Miller 1979), but the only material mentioned is a right navicular (Miller 1979), and no attempt was made to identify it to species. Since the Crystal Ball Cave specimens match measurements of C. hesternus byT. E. Downs (1984, pers. comm.), Graham (1981), and Webb (1965), and since C. hesternus is the only species reported from the state of Utah, the Crystal Ball Cave specimens are referred to this species. But, since the only diagnostic character to distinguish C. hester- nus from C. huerfanensis is a dental feature not applicable to the Crystal Ball Cave speci- mens (Hopkins 1955, Savage 1951), C. huer- fanensis cannot be positively eliminated on the basis of these foot elements. Hemiauchenia cf. macrocephala Material. — Distal right portion of meta- podial (LACM 123688), first phalanx (LACM 123690), partial proximal portion of first pha- lanx (BYUVP 7640), second phalanx (BYUVP 7631). Phalanx measurements are listed in Tables 7, 8, and 9. Discussion. — Two genera of small camels are recognized from the Pleistocene of North America: Hemiauchenia and Paleolama (Webb 1974). Based on illustrations oi Hemi- auchenia (=Tanupolama) macrocephala {=stevensi) by Stock (1928), and Paleolama mirifica by Webb (1974), the metapodials of H. macrocephala are 63% longer but 3% transversely narrower at the distal end than those of P. mirifica. The Crystal Ball Cave metapodial fragment is 12% transversely nar- rower than the H. macrocephala specimens illustrated by Stock (1928) and measurements from the Vallecito Creek site in southern Cali- fornia and Ringold site in Washington State provided by T. E. Downs (1984, pers. comm.). The first phalanges from Crystal Ball Cave fall well within the range o( Hemiauche- nia specimens reported by T. E. Downs (1984, pers. comm.), McGuire (1980), and Schultz (1937). Nothing was available to compare the second phalanx with, but it is from the same size of camel as the other ele- ments. The Crystal Ball Cave specimens clearly match the more narrow-legged Hemi- auchenia rather than the more broad-legged Paleolaina. Webb (1974) synonymized the North American genus Tanupolama with the South American genus Hemiauchenia and recog- nized six valid species. Of these, only H. macrocephala is found in the late Pleistocene of North America. Hemiauchenia macro- cephala represents the synonymy of a number of previously named North American species (Webb 1974), and it is the best-known Pleis- tocene llama (Kurten and Anderson 1980). Since only this species matches the age and locality of the Crystal Ball Cave assemblage, and since the Crystal Ball Cave specimens match specimens from other sites assigned to this species, the four Crystal Ball Cave speci- mens are referred to H. macrocephala. Char- acters separating this species from others of Hemiauchenia are almost entirely dental (Webb 1974), however, and are therefore not apphcable to the Crystal Ball Cave material. Miller (1982) reported Hemiauchenia from Crystal Ball Cave based on the same material reported here. Miller (1979) reported ? Hemi- auchenia sp. from Smith Creek Cave based on a left cuboid, the proximal portion of a scapula, and a juvenile metapodial. Hemi- auchenia is better represented than Camelops at Smith Creek Cave by a ratio of 3 to 1, but Camelops is better represented than Hemi- auchenia at Crystal Ball Cave by a ratio of 7 to 1. This difference seems even more dramatic in light of the selection for smaller bones at Crystal Ball Cave but not at Smith Creek Cave. Although this difference could be ex- plained by slight age differences in these fau- nas, human intervention, or chance preserva- tion, I feel it is more likely due to habitat differences between these two genera of camels. 380 Great Basin Naturalist Vol. 45, No. 3 Kurten and Anderson (1980) stated that ". . . Hemiauchenia had a long stride and was highly cursorial. It was a plains-dweller and probably fed primarily on grass." About Camelops they stated: "Although primarily a grazer, Camelops, with its long neck and legs, was probably an occasional browser." Al- though these two camels are thought to have been plains-dwelling grazers, it is interesting to speculate about their habitat differences. Webb (1974) presented strong evidence that Hemiauchenia gave rise to the mountain- dwelling South American llamas. Camelops, on the other hand, probably resembled the living dromedary camel (Kurten and Ander- son 1980), which prefers flat plains habitats. The fact that Camelops is by far the better represented camel at Crystal Ball Cave, lo- cated in a small outlier surrounded by a flat valley, and Hemiauchenia is better repre- sented at nearby Smith Creek Cave, located in a canyon at the base of a high mountain, suggests that Hemiauchenia preferred higher elevations and/or more rugged terrain than Camelops. Family Cervidae cf. Cervus elaphus Material.— First phalanx (BYUVP 7811). Discussion. — Several cervid phalanges from Crystal Ball Cave are intermediate in size between Cervus and Odocoileus. BYUVP 7811 (60.2 mm long) is the largest of these and is much closer in size to Cervus. In compari- son with the others it is distinctly larger and more robust, yet high bone porosity suggests that it is of a subadult. Navahoceros fricki is another Late Pleistocene cervid found as close to Utah as Arizona and Wyoming, and its size is intermediate between Odocoileus and Cervus (Kurten and Anderson 1980). No char- acter has been described to differentiate pha- langes of Navahoceros and Cervus, and no comparative material of Navahoceros was available to the author. Cervus elaphus was recovered from Smith Creek Cave (Miller 1979) and has been reported living in the Snake Range in Recent times (Hall 1981), so the phalanx is referred to this species. Odocoileus hemionus Material. — Partial right dentarv with P/3,/4, M/1 (BYUVP 7651) and anterior left dentary with P/3,/4, M/1,/2 (BYUVP 7650, probablv from the same individual), partial right dentarv with P/3 (BYUVP 7652), left patella (BYUVP 7934). Of 21 first and 51 sec- ond phalanges of noncamelid artiodactyls, most compare best in size and proportions with Odocoileus. Discussion — BYUVP 7650 and 7651 are of a juvenile and compare best in size and degree of hypsodonty with juvenile individu- als of O. hemionus. The P/4's in these den- taries have three lobes rather than two, a con- dition seen in juveniles of Odocoileus but not Antilocapra. The P/3 of BYUVP 7652 is iden- tical to adult O. hemionus and distinctly larger and less hypsodont than A. americana. The first and second phalanges from Crystal Ball Cave that compare best with Odocoileus have a slightly larger mean size than those of Re- cent O. hemionus living in Utah. This demon- strates that the Crystal Ball Cave specimens are of O. hemionus rather than the smaller O. virginiana (Hall 1981), and it suggests that deer decreased in size at the end of the Pleis- tocene much as did Ovis canadensis (Harris and Mundel 1974). Based on numbers of phalanges, Odo- coileus is the best represented artiodactyl in the Crystal Ball Cave assemblage; but Antilo- capra americana is now the dominant artio- dactyl of the local fauna. Odocoileus sp. was reported at Smith Creek Cave by Goodrich (1965), but no material was found by Miller (1979). Mule deer now live in Smith Creek Canyon (Miller 1979) and sometimes come down to Candy at night to feed in cultivated fields (J. C. Bates 1984, pers. comm.). The replacement of Odocoileus by Antilocapra, suggested by comparison of the Crystal Ball Cave assemblage with the living community, shows that plant communities preferred by deer apparently moved upward in altitude from Snake Valley to higher elevations in the Snake Range at the close of the Pleistocene. Family Bovidae Antilocapra americana Material. — Partial left maxilla with Ml/, 2/,3/ (BYUVP 7656). Discussion. — The M3/ was distinguished from Odocoileus by being veiy hypsodont, transversely narrower, and having a more July 1985 Heaton: Crystal Ball Cave Fossils 381 pointed posterior end as in Antilocapra. It is identical in size and proportions to the largest male specimen of A. americana available for comparison and distinctly larger than the ex- tinct Pleistocene antilocaprids. Since A. americana presently lives around Candy Mountain in small herds, it is not sur- prising to find it in the assemblage. But it is not well represented as a fossil, suggesting that Snake Valley has not always been the treeless desert that it is now. Since Odo- coileus hemionus is the dominant artiodactyl in the fossil assemblage and Antilocapra americana is the dominant living artiodactyl in the area, Antilocapra americana must have become abundant in the area in Recent times and replaced Odocoileus hemionus , probably due to changes in the vegetation. Ovis canadensis Material. — Posterior portion of right den- tary with M/1,/2,/3 (LACM 123695) and pos- terior portion of left dentary with M/3 (LACM 123696, probably from the same individual), left magnum (BYUVP 7780). Discussion. — The molars of LACM 123695 and 123696 are distinctly larger and more robust than living Ovis aries and are even slightly larger than Recent O. canaden- sis. This suggests that the jaws are Pleistocene rather than Recent in age because Harris and Mundel (1974) demonstrated that O. canadensis became reduced in size at the end of the Pleistocene. Bighorn sheep are commonly found in Pleistocene assemblages in the Great Basin (Hibbard and Wright 1956, Stokes and Condie 1961). Even in historic times they have been reported natively in the Snake Range (Durrant 1952, Hall 1946, 1981). Ovis canadensis was temporarily lost from the Snake Range but was reintroduced in the mid- dle 1900s and presently thrives in the higher elevations (Mead et al. 1982). Shortly after this reintroduction, one young ram lived on Candy Mountain for several months (J. C. Bates, 1983, pers. comm.), but this is the only citing known to me for such a low elevation in the area. Ovis canadensis is the best represented un- gulate in the Smith Creek Cave assemblage, and Oreamnos harringtoni is also well repre- sented (Miller 1979). No Oreamnos material has been identified from Crystal Ball Cave, and Ovis is less represented than horse, camel, and deer. This difference between the two assemblages is probably because wild goats and sheep are mountainous animals and would rarely venture into Snake Valley. It may also represent the fact that Smith Creek Cave was a shelter for humans since many Outs fossils found there appear butchered (Miller 1979). Ovis cf. aries Material. — Right metacarpal and 2 first phalanges found associated (BYUVP 8300). Discussion. — These associated bones were found as float near the east entrance of Crystal Ball Cave, and their greasy appear- ance suggests that they are Recent. The length and shape of the metapodial demon- strates that it is of the genus Ovis, and it is slightly longer than the O. aries specimens to which it was compared but distinctly smaller than living O. canadensis. Ovis aries is now a common domestic animal in the area, and many roam on Candy Mountain each winter (J. C. Bates 1984, pers. comm.). Since this species is a Recent introduction from Europe, its presence has little signifi- cance to this study. It does show, however, that the smaller bones of large mammals are still being deposited in Crystal Ball Cave, probably by woodrats since gates on the cave entrances would keep out all but the smallest carnivores. These specimens were found just north of the east entrance, an area where woodrats and their nests are often found. cf Symbos cavifrons Material. — Second phalanx (BYUVP 7923), distal portion of second phalanx (BYUVP 7924), 2 partial second phalanges (BYUVP 7925, 7926), 2 distal portions of sec- ond phalanges (BYUVP 7921, 7922). Discussion. — These short, broad second phalanges compare best among living species to Ovibos moschatus but are slightly longer and narrower. BYUVP 7923 is the most com- plete specimen, missing only one side of the distal extention. It has a length of 42 mm, a proximal transverse width of 27 mm, and a 382 Great Basin Naturalist Vol. 45, No. 3 proximal anteroposterior width of 26 mm. BYUVP 7924 has the same proximal measure- ments as BYUVP 7923, and BYUVP 7925 has a proximal anteroposterior width of at least 26 mm. The distal ends taper in such a way that they are hard to measure. The general shape of these second phalanges shows that they are from an animal more closely related to Ovibos than any other living bovid. Few phalanx measurements of Pleistocene oxen are avail- able, but Nelson and Madsen (1980) and Stokes and Hansen (1937) reported abundant isolated Symbos cavifrons and Bootheriiim bombifrons crania from Lake Bonneville de- posits, and McGuire (1980) reported Eu- ceratherium from a Late Pleistocene deposit in central Nevada. Kurten and Anderson (1980) described Symbos cavifrons as being taller and more slender than Ovibos moschatus, and this de- scription matches the difference between the Crystal Ball Cave specimens and Ovibos moschatus perfectly. Bootherium is smaller than Symbos and is thought by many to repre- sent females or juveniles of that genus (Kurten and Anderson 1980, Nelson and Madsen 1980). Euceratherium was larger and more heavily built than Ovibos (Kurten and Ander- son 1980), and a first phalanx illustrated by McGuire (1980) is far too big at the distal end to match the second phalanges from Crystal Ball Cave. So, although no comparative mate- rial was available, both the description and known range of Symbos cavifrons make the Crystal Ball Cave specimens most referable to that species. Conclusions The Crystal Ball Cave assemblage is the first Late Wisconsinan fauna reported from the state of Utah and represents the closest known terrestrial fossil deposit to Lake Bon- neville. The assemblage differs from most other cave faunas by its fossils being far inside the cave where man and birds probably had no influence on what was deposited. As a re- sult, the assemblage is better than average in representing the proportions of animals that lived in the area, but there are some obvious biases. Neotoma, always an animal of low den- sity, was the second most abundant genus in the assemblage simply because it is one of the few animals that lives in the cave. But, other than cave-dwelling species, the assemblage probably gives a fairly good record of the abundance of most groups, at least those that lived in the immediate vicinity of the cave. The assemblage, for example, contains a ratio of small mammals to large mammals and car- nivores to herbivores that might be expected in a living community. One very strong bias is the size of bones in the assemblage that I have attributed to the limit of bone size that a wood rat can carry. Bones of large mammals were brought in after the carcasses deteriorated, as evidenced by the presence of only small iso- lated elements. This bias tends to make large species less represented in the assemblage than in the living community and very large species unrepresented. Proboscidian fossils have been found in Lake Bonneville deposits (Nelson and Madsen 1980) but not in Crystal Ball Cave, probably because there was no means to transport such large bones inside. It is difficult to say if any other animals besides wood rats contributed to transporting fossils into the cave. No other rodents are known to transport bones as wood rats do. Small carnivores could have done so, but the low abundance of carnivore fossils in the as- semblage suggests that none habitually used the cave as a home. The small size of the original cave entrance would have prevented the entry of any large mammals. Both the distance of the fossils inside the cave and the low abundance of birds compared to mammals suggests that birds did not transport any fos- sils in, and this is one of the main differences between Crystal Ball Cave and Smith Creek Cave (and most other cave deposits). Clearly no inorganic processes such as wind, water, or gravity could have been responsible for the fossil deposits since they are in fine dust in an isolated part of the cave where none of these forces have a magnitude capable of transport- ing bones. Crystal Ball Cave has been accumulating fossils from at least 23,000 years ago to the present. Although some of the fossils are Re- cent, the assemblage as a whole shows dra- matic differences from the present-day local fauna. The poor representation of many mam- mals that currently live in the area may be due to the shift from Neotoma cinerca to N. lepida as the wood rat that inhabited the cave, and it July 1985 Heaton: Crystal Ball Cave Fossils 383 also suggests that the shift to the present cli- mate occurred very recently in the history of the assemblage. Brachijprotoma, Smilodon, several species of Equus, Camelops, Hemi- auchenia, and Symbos (or a closely related genus) are represented in the assemblage, all of which are now extinct. As mentioned ear- lier, there was a widespread extinction of large mammals at the close of the Pleistocene, the cause of which is under debate. This as- semblage does not resolve that problem, but it does demonstrate that a marked climatic shift did take place contemporaneously with the extinctions, and this suggests to me that the extinctions were also a result of this cli- matic shift. Equally as significant as the extinctions are the shifts in species ranges that the Crystal Ball Cave assemblage documents. The pres- ence of Ondatra zibethicus and Mustela cf vison, both of which require perennial water and are extirpated from the area, represent the drying of Lake Bonneville and perennial streams around Candy Mountain. Ochotona princeps and Martes americana were extir- pated from the Snake Range without replace- ment but still live at high elevations in nearby ranges. Marmota flaviventris, Cervus ela- phus, and Ovis canadensis are represented in the assemblage but now inhabit only higher elevations in the Snake Range. In other cases, species now abundant at Candy Mountain are unrepresented or poorly represented in the assemblage, and their more boreal counterparts, now extirpated or rare in the area, are well represented as fos- sils. Among jackrabbits, Lepus calif ornicus is presently the dominant species, but L. townsendii, its more boreal counterpart, is by far the better represented species in the fossil assemblage. Among cottontails, Sijlvilagus audubonii and S. nuttallii make up the present local fauna, but only S. nuttallii, the more northern species, is found in the assem- blage. Lepus americanus, a functional cotton- tail (J. A. White 1984, pers. comm.)andavery boreal animal, is probably represented but is now extirpated from the Snake Range. Scotoma lepida, the only wood rat seen living in Crystal Ball Cave, is rare in the assemblage, but N. cinerea, its more boreal counterpart, is one of the two most abundant fossil species. Vulpes vulpes is well represented in the cave assemblage but extirpated from the area, and Urocyon cinereoargenteus , a more southern fox of similar size, now inhabits the area but is not found as a fossil. Although the Crystal Ball Cave assemblage differs dramatically from the present-day local fauna, it is not atypical of Late Pleistocene assemblages in the region. Figure 9 shows the location of and Table 10 compares the mam- malian taxa recovered from 10 Late Pleis- tocene-Recent cave assemblages within 400 km (240 miles) of Crystal Ball Cave. The most unique feature of the Crystal Ball Cave assem- blage is the presence oi Brachyprotoma since it represents the first citing of the genus from the western United States and the first recov- ery of the new species herein named B. brevi- mala. Ondatra zibethicus was found in Crys- tal Ball Cave but not at the other localities, probably because of this cave's close proxim- ity to Lake Bonneville. Symbos cavifrons may be present at Crystal Ball Cave but absent from the other assemblages for the same rea- son since it is most common in Lake Bon- neville deposits. Some interesting paleoecological informa- tion can be inferred from the differences be- tween the Smith Creek Cave and Crystal Ball Cave assemblages in particular since they are close geographically but located in somewhat different habitats. Several species of Sper- mophilus have been recovered from Smith Creek Cave, but large numbers of a single species have been recovered from Crystal Ball Cave. This can probably be attributed to the greater habitat diversity at Smith Creek Cave, which is at the base of a high mountain. Among camels, Hemiauchenia is better repre- sented at Smith Creek Cave, but Camelops is better represented at Crystal Ball Cave. Al- though based on a small sample size, this sug- gests that Hemiauchenia favored higher and/ or more rugged terrain than Camelops because Smith Creek Cave is located in the main Snake Range and Crystal Ball Cave is located in an outlier in Snake Valley. Of the non-camelid artiodactyls, Odocoileus hemi- onus is the best represented in the Crystal Ball Cave assemblage and Ovis canadensis is the best represented in the Smith Creek Cave assemblage. Oreamnos harringtoni fossils have been found in Smith Creek Cave but not in Crystal Ball Cave. Now Antilocapra ameri- 384 Great Basin Naturalist Vol. 45, No. 3 • Wilson Butte Cave . IDAHO Deer Creek Cave • Silver Creek Site Mineral Hill Cave • Gatecliff Shelter. Smith Creek. 1 Cave •Crystal Ball Cave \^ NEVADA U T A H \v Mormon \ Mountain \ Cave \ Gypum \ Cave* A R 1 Z 0 N A \ r^ Rampart Cave ••Vulture Cave Fig. 9. Map showing the location of 10 Late Pleistocene cave faunas (see Table 10 for a list of the mammalian taxa recovered) and the Silver Creek fossil site described by Miller (1976). Table 10. Comparison of the Crystal Ball Cave fauna with nine other Late Pleistocene/Early Holocene mammalian cave faunas located within 400 km (240 miles) of Crystal Ball Cave. The locations of these caves are shown in Figure 9. 1 Crystal Ball Cave, Utah (Heaton th s report). 2 Wilson Butte Cave , Idaho (Gruhn 1 961, Lundelius et al. 1983). .3 Deer Creek Cave, Nevada (Ziegler 1963). 4 Mineral Hill Cave, Nevada (McGu re 1980). 5 GateclifiT Shelter, Nevada (Crayson 1983, Thomas 1983). 6 Smith Creek Cany on caves, Nevad 1 (Miller 1979, Mead et al. 1982). 7 Mormon Mountain Cave, Nevada (Jefferson 1982). 8 Gypsum Cave, Nevada (Mehringer 1967, Lundelius et a 1. 1983). 9 Rampart Cave, Arizona (Harington 1972, Lundelius et a . 1983). 10 Vulture Cave, Arizona (Mead and Phillips 1981). Species/Localities 1 2 3 4 5 6 7 8 9 10 Sorex sp. X X ? S. vagrans X Notiosorex sp. X X N. crawfordi X X Myotis sp. X X Eptesicus fiisctis X Plecotus townsemlii ? X Antrozous pallidas X cf. X Nothrotherium X X N. shastensis X X Ochotona princeps X X X X X X X Sylvilagus sp. X X X X X X X X X S. idahoensis X X X X X July 1985 H EATON Crystal Ball Cave Fossils 385 Table 10 continued. Species/Localities 1 2 3 4 5 6 7 8 9 10 S. nuttallu X X X X Lcpus sp. X X X X X X X X L. amcricanus cf. L. toicnscndii X cf L. californicus X X cf cf cf Marmota sp. X X X X X X X X X cf X M. JIaviventris X X X X X X X X Spermophilus sp. X X X X X X X X S. armatus X S. townsendii X X X cf S. richardsonii X cf S. beldingi X X cf S. variegatus X X X S. lateralis X cf X Ammospermophihis leiicurus cf. X cf. Eutamias sp. X X X X X X E. minimus X X X X E. umbrinus X cf E. dorsalis X X cf E. amoenus X Thomomijs sp. X X X X X X T. talpoides X cf T. umbrinus X X Perognathus sp. X X X X P. parvus X cf X P. formosus cf. P. intermedius cf. Microdipodops sp. X X X M. megacephalus X X cf Dipodomijs sp. X X X X D. or da cf D. microps X Castor canadensis X Reithrionomys sp. X R. megalotus X Peromyscus sp. X X X X X X X P. maniculatus X X X P. crinitus cf X P. truei cf X Onychomys sp. X Neotoma sp. X X X X X X X X N. lepida X X X N. cinerea X X X X X Ondatra sp. X 0. zibethecus X Cleithryonomys sp. X C. gapperi X Phenacomys sp. X X X P. intermedins X cf cf Microtus sp. X X X X X X X X M. californicus X M. longicaudus cf X cf cf cf M. montanus cf X cf cf M. pennsylvanicus cf Lagurus curtains X X Zapus princcps cf Erethizon dorsatum X X X X Canis sp. X X X X X C. familiaris X C. latrans cf. X X cf C. lupus cf. X cf Vulpes sp. X X X X X X 386 Great Basin Natur.\list Vol. 45, No. 3 Table 10 conintued. Species/Localities 1 2 3 4 5 6 7 8 9 10 V. vtilpes X X X V. velox X X X Urocyon sp. X Ursus sp. X X U. horribilis X Bassariscus sp. X X X B. astutus X X Mustek sp. X X X X X M. erminea X X M. frenata cf. X X M. vison cf. X Maries sp. X X X X M. americana X X X M. nobilis X Taxidea taxtis X X Spilogale sp. X X X X S. putorius X X Mephitis mephitis X Brachyprotoma sp. X B. brevimala X Smilodon fatalis cf. Pant her a air ox ? P. onca X Felis concolor X X X X Lynx sp. X X X X X X L. rufus cf. X X X X X Eqtitts sp. X X X X X X E. sp. (large) X X £. scotti cf. E. occidentalis X E. sp. (small) X X E. conversidens ? Camelops sp. X X X X C. hesternus cf X cf Hemiauchenia sp. X X X ? X H. macrocephala cf Cervus sp. cf ? X X C. elaphus cf X X Oducuileus sp. X X X X X O. hemionus X X Antilocapra sp. X X X X X X A. americana X X X X X Capromeryx minor ? Oreamnos sp. X X O. harringtoni X X O. americanus cf Ovis sp. X X X X X X X X X O. canadensis X X X X X X X X X O. aries cf cf Bison sp. X ? X ? B. bison X Eticerotherium sp. X Symbos cavifrons cf cana is the lx\st represented artiodactyl in Snake Valley, Odocoileus hemionus is the best represented artiodactyl in the Snake Range, Ovis canadensis is found only at high eleva- tions in the Snake Range, and Oreamnos har- rin^toni is extinct. This suggests that these four arti()dact\ is can be placed in the following order of ele\ ation preference starting at the highest: Oreamnos harringtoni, Ovis canadensis, 0(h)eoileus hemionus, and An- tilocapra americana. At the end of the Pleis- tocene, in rough terms, each of these species July 1985 Heaton: Crystal Ball Cave Fossils 387 moved upward in elevation to fill the habitat of the next higher species. The one at the top went extinct; the one at the bottom became abundant. Differences of lesser magnitude between the Crystal Ball Cave and Smith Creek Cave assemblages must be dealt with more carefully because they may represent slight differences in the age of the deposits, biases in the mode of deposition, human in- tervention, or chance preservation. Identifi- cation of more material, especially at Smith Creek Cave, could make comparison of these two assemblages a very valuable paleoecologi- cal study. The Crystal Ball Cave fauna, like many pre- viously studied faunas, shows that a dramatic climatic shift occurred at the end of the Pleis- tocene and caused many species to move northward in latitude and upward in elevation and to become extinct. This shift is particu- larly well expressed in the Crystal Ball Cave assemblage because its close proximity to Lake Bonneville made the drying trend very severe in the area. The Crystal Ball Cave fauna documents the previous ranges and abundances of many taxa that help in recon- struction of details of the last Pleistocene ice age. Acknowledgments This study was supervised by Wade E. Miller, who helped in the collection and iden- tification of specimens and preparation of the manuscript. His insistence that every identifi- cation be backed by thorough research and explanation has made a lasting impression on me. Thanks is also due Kenneth L. Stadtman and Clyde L. Pritchett for providing compara- tive specimens and help in identification. Jer- ald C. and Marlene Bates of Candy allowed access to Crystal Ball Cave on many occasions and provided helpful information on the his- tory and original condition of the cave, as well as the mammals that currently live in the im- mediate area. The Los Angeles County Mu- seum generously loaned the Crystal Ball Cave specimens in their possession for inclusion in this study. Howard C. Stutz identified the plants, Lee F. Braithwaite the gastropods, and Stephen L. Wood the beetle. John A. White provided information valuable for identifying the lago- morphs. Elaine Anderson provided informa- tion helpful in evaluating the Brachyprotoma skull. Phillip M. Youngman provided unpub- lished information and measurements on Brachyprotoma specimens he recovered from Yukon Territory, Canada. Arthur H. Harris provided bone measurements for several spe- cies of horses. Theodore E. Downs gave me information on Pleistocene horses and mea- surements of Pleistocene camels. Jim L Mead identified some of the bovid specimens and provided other helpful encouragement. Wade E. Miller, J. Keith Rigby, Morris S. Petersen, Lehi F. Hintze, and Stephen L. Wood made critical reviews of the manuscript. Thanks is especially due my wife, Julie, for help with collection and curation of specimens, gathering of literature, and prepa- ration of the manuscript. This study was funded by grants from the National Speleological Society and Associated Students of Brigham Young University, by a private donation from Herbert H. Gerisch, and by Brigham Young University research assistantships to the author. Publication costs were paid by the Joseph M. and Jessie K. D. Savage Endowment of the Brigham Young University Monte L. Bean Museum and the Brigham Young University College of Physi- cal and Mathematical Sciences. Literature Cited Anderson, E. 1968. Fauna of the Little Box Elder Cave, Converse County, Wyoming, Univ. Colorado Stud. Earth Sci. Ser. 6:1-59. 1970. Quaternary evolution of the genus Martes (Carnivora, Mustelidae). Acta. Zool. Fennica 130:1-132. BjORK, P R 1970. The carnivora of the Hagerman local fauna (Late Pliocene) of southwestern Idaho. Anier. Phil. Soc. Trans. 60(7): 1-54. Brattstrom, B H 1976. A Pleistocene herpetofauna from Smith Creek Cave, Nevada. South. Califor- nia Acad. Sci. Bull. 75:283-284. Brown, B 1908. The Conard Fissure, a Pleistocene bone deposit in Northern Arkansas: with description of two new genera and twenty new species of mam- mals. Amer. Mus. Nat. Hist. Mem. 9(4):157-208. Brown, J H. 1971. Mammals on mountaintops: nonequi- librium insular biogeography. Amer. Nat. 105(945):467^78. 1978. The theory of insular biogeography and the distribution of boreal mammals. Great Basin Nat. Mem. 2:209-227. Bryan, A L. 1979. Smith Creek Cave. In: D. R. Tuohy and D. L. Rendall, eds.. The archaeology of Smith Creek Canyon, Eastern Nevada. Nevada State Mus. Anthrop. Pap. 17:162-251. 388 Great Basin Naturalist Vol. 45, No. 3 Chamberland, R V , AND D T Jones 1929. A descrip- tive catalog of MoUusca of Utah. Univ. Utah Biol. Ser. Bull. 1(1): 1-203. Cope, E. D 1899. Vertebrate remains from the Port Kennedy bone deposit. Philadelphia Acad. Nat. Sci. J. 11:193-267. Cragin. F W 1892. Observations on llama remains from Colorado and Kansas. Amer. Geol. 9:257-260. CURREY, D. R 1982. Lake Bonneville; selected features of relevance to neotectonic analysis. U.S. Geol. Surv. Open-File Rept. 82-1070:1-29. Dalquest, W. W 1964. Equus scotti from a high terrace near Childress, Texas. Texas J. Sci. 16(,3): 350-358. 1979. The little horses (genus Equus) of the Pleis- tocene of North America. Amer. Midland Nat. 101(l):241-244. Dalquest, W W . and J T Hughes. 1965. The Pleis- tocene horse, Equus conversidens. Amer. Mid- land Nat. 74(2):408-417. Davies, W. E. 1960. Origin of caves in folded limestone. Nat. Speleo. Soc. Bull. 22:5-19. Davis, W. M 19.30. Origin of limestone caverns. Geol. Soc. Amer. Bull. 41:475-628. Durrant, S. D 1952. Mammals of Utah. Univ. of Kansas Press, Lawrence. 549 pp. Durrant, S D . M R Lee. and R M Hansen 19.55. Additional records and extensions of known range of mammals from Utah. Univ. Kansas Mus. Nat. Hist. Pub. 9(2):69-80. FiNLEY, R. B., Jr 19.58. The wood rats of Colorado: distri- bution and ecology. LIniv. Kansas Mus. Nat. Hist. Pub. 10(6):213-552. Gazin, C. L. 1935. Annotated list of Pleistocene Mam- malia from American Falls, Idaho. Washington Acad. Sci. J. 25(7):297-302. 1936. A study of the fossil horse remains from the Upper Pliocene of Idaho. U.S. Nat. Mus. Proc. 83(2985): 281-320. GiDLEY, J. W 1900. A new species of Pleistocene horse from the Staked Plains. Amer. Mus. Nat. Hist. Bull. 13:111-116. 1903. The freshwater Tertiary of northwestern Texas, American Museum expedition of 1899-1901. Amer. Mus. Nat. Hist. Bull. 19:617-635. GiDLEY. J W , and C L Gazin 19,38. The Pleistocene vertebrate fauna from Cumberland Cave, Mary- land. U.S. Nat. Mus. Bull. 171:1-99. Gilbert, G K 1890. Lake Bonneville. U.S. Geol. Surv. Mon. 1:1-438. Giles, E. 1960. Multivariate analysis of Pleistocene and Recent coyotes {Canis latrans) from California. Univ. California Geol. Sci. Pub. 36:369-390. Goodrich, R B 1965. The Quaternary mammalian mi- crofaunal assemblage of Smith Creek Cave, Ne- vada. Unpublished thesis. California State Univ., Los Angeles. 45 pp. Graham, R, W 1981. Preliminary report on Late Pleis- tocene vertebrates from the Selby and Dutton Archeological/Paleontological sites, Yuma County, Colorado. Univ, Wvoming (-ontrib. Geol. 20(l):33-56. Grayson, D. K. 1977. On the Holocene history of some northern Great Basin lagomorphs. J. Mammal. 58(4):507-513. 1983. Small mammals. In: D. H. Thomas, ed., The archaeology of Monitor Valley: 2, Gatecliff Shelter. Amer. Mus. Nat. Hist. Anthrop. Pap. 59(1):99-126. Green. D. J 1961. Introduction to the speleology of Utah and adjacent areas. Nat. Speleo. Soc. Salt Lake Grotto Tech. Notes 2(60):192-201. Gruhn, R 1961. The archaeology of Wilson Butte Cave, south-central Idaho. Idaho State Mus. Occas. Pap. 6. Hall, E. R 19.36. Mustelid mammals from the Pleis- tocene of North America; with systematic notes on some Recent members of the genera Mustela, Taxidea and Mephitis. In: Studies of Tertiary and Quaternary mammals of North America. Carnegie Inst. Washington Contrib. to Paleont. 473; 41-119. 1946. Mammals of Nevada. Univ. California Press, Berkeley. 710 p. 1951. A synopsis of the North American Lago- morpha. Univ. Kansas Mus. Nat. Hist. Pub. .5(10):119-202. 1981. The Mammals of North America, 2d ed. John Wiley and Sons, New York. 1181 pp. Halliday, W R 19.57. The caves of Gandy Mountain, an initial reconnaissance. Nat. Speleo. Soc. Salt Lake Grotto Tech. Notes 2(38):31-37. Handley, C O , Jr 1959. A revision of American bats of the g,eneTa Eudenna and Plecotus. U.S. Nat. Mus. Proc. 110(3417):95-246. Harington.C.R. 1972. Extinct animals of Rampart Cave. Canadian Geog. J. 85(5); 178-183. Harper, K T , D C Freeman, W K Ostler, .\nd L G Klikoff. 1978. The flora of Great Basin mountain ranges; diversity, sources, and dispersal ecology. Great Basin Nat. Mem. 2:81-103. Harrington, M. R. 1934. Ancient horses and ancient men in Nevada. Masterkey 8:16.5-169. Harris. A H . and P Mundel 1974. Size reduction in bighorn sheep (Ovis canadensis) at the close of the Pleistocene. J. Mammal. .55(3):678-680. Harris. A. H., and L. S. W Porter 1980. Late Pleis- tocene horses of Dry Cave, Eddy County, New Mexico. J. Mammal. '61(l):46-65. ' Hay, O. P 1915. Contributions to the knowledge of the mammals of the Pleistocene of North America. U.S. Nat. Mus. Proc. 48:51.5-575. 1921. Descriptions of Pleistocene vertebrata, types of specimens of which are preserved in the United States National Museum. U.S. Nat. Mus. Proc. 59:617-638. 1923. The Pleistocene of North America and its vertebrated animals from the States east of the Mississippi River and from the Canadian Provinces east of Longitude 95. Carnegie Inst. Washington Pub. 322:1-499. Heaton, T H 1984. Preliminary report on the Quater- nary Vertebrate fossils from Crystal Ball Cave, Millard County, Utah. Cur. Res. 1:65-67. HiBBARD. C W 1941. New mammals of the Rexroad Fauna from the Upper Pliocene of Kansas. Kansas Acad. Sci. Trans. 44:26.5-,313. 1944. Abnormal tooth pattern in the lower denti- tion of the jackrabbit, Lepus californicus deserti- co/« (Mearns). J. Mammal. 25:64-66. July 1985 Heaton: Crystal Ball Cavk Fossils 389 1950. Mammals of the Rexroad Formation from F^ox Canvon, Kansas. Univ. Michigan Mus. Pale- ontol. Contrib. 8(6): 113-192. 1952. Vertebrate fossils from Late Cenozoic de- posits of central Kansas. Univ. Kansas Paleontol, Contrib. 2:1-14. 1954. A new Phocene vertebrate fauna from Okla- homa. Michigan Acad. Sci. Arts Let. Pap. 39: 339-359, 1955. Pleistocene vertebrates from the Upper Be- cerra (Becerra Superior) Formation, Valley of Tequixquiac, Mexico, with notes on other Pleis- tocene forms. Univ. Michigan Mus. Paleontol. Contrib. 12:47-96. 1963. The origin of the P/3 pattern oi Sylvilagus, Caprolagus, Oryctolagns a.nd Lepus. J. Mammal. 44(1): 1-15. HiBBARD, C. W, AND D. W. Taylor. 1960. Two Late Pleis- tocene faunas from southwestern Kansas. Univ. Michigan Mus. Paleontol. Contrib. 16:1-223. Hibbard. C. W., and B a. Wright. 1956. A Pleistocene bighorn sheep from Arizona. J. Mammal. 37(1): 105-107. Hill, C. A, 1976. Cave minerals. The Speleo Press, Austin, Texas. 137 pp. Hopkins, M. L. 1955. Skull ofa fossil camelid from Ameri- can Falls lake bed area of Idaho. J. Mammal. 36:27^282. Hopkins. M L . R. Bonnichsen, and D. Fortsch, 1969. The stratigraphic position and faunal associates of Bison (Gigantobison) latifrons in southeastern Idaho, a Progressive Report. Tebiwa 12(1): 1-8. Howard, H. 1935. A new species of eagle from a Quater- nary cave deposit in eastern Nevada. Condor 37:206-209. 1952. The prehistoric avifauna of Smith Creek Cave, Nevada, with a description ofa new gigantic raptor. South. California Acad. Sci. Bull. 51(2):50-54. Howe, J. A. 1970. The range of variation in Equiis (Plesip- pus) simplicidens Cope from the Broadwater Quarries of Nebraska. J. Paleontol. 44:958-968. Humphrey, S. R., and T. H. Kunz. 1976. Ecology of a Pleistocene relect, the western big-eared bat {Ple- cotus townsendii), in the southern Great Plains. J. Mammal. 57(3):470-494. Ingles, L G. 1965. Mammals of the Pacific states. Stan- ford Univ. Press. 506 pp. Jefferson, G. T 1982. Late Pleistocene vertebrates from a Mormon Mountain cave in southern Nevada. South. California Acad. Sci. 81(3);121-127. Johnston. C S. 1937. Notes on the craniometry ofEquiis scottiCkWey. J. Paleontol. 11:459-461.' KURTEN. B., AND E ANDERSON. 1972. The sediments and fauna of Jaguar Cave: H-The fauna. Tebiwa 15(l):21-45. 1980. Pleistocene Mammals of North America. Columbia Univ. Press, New York. 442 pp. Logan, L. E 1983. Paleoecological implications of the mammalian fauna of Lower Sloth Cave, Guadalupe Mountains, Texas. Nat. Speleo. Soc. Bull. 45(1):3-11. LuNDELius. E L , Jr , R W Graham, E Anderson, J GuiLDAY, J. A Holman, D W. Steadman, and S. D. Webb. 1983. Terrestrial vertebrate faunas. Pages 311-353 in S. C. Porter, ed., Late-Quater- nary environments of the United States: Volume I, The Late Pleistocene. Univ. Minnesota Press. LuNDKLius, E. L., Jr.. and M. S. Stevens. 1970. Equus francisci Hay, a small stilt-legged horse, Middle Pleistocene of Texas. J. Paleontol. 44(1): 148-153. Malo'IT. C a. 1938. Invasion theory of cavern develop- ment. Geol. Soc. Amer. Proc. 1937:323. Martin, L D , and A M Nei'nkr 1978. The end of the Pleistocene in North America. Nebraska Acad. Sci. Trans. 6:117-126. Martin. P. S. 1967. Prehistoric overkill. Pages 7.5-120 in P. S. Martin and H. E. Wright, Jr., eds., Pleis- tocene extinctions: The search for a cause. Yale Univ. Press, New Haven. MgGuire, K. R 1980. Cave sites, faunal analysis, and big game hunters of the Great Basin: a caution. Quat. Res. 14:263-268. Mead, J I , and A M Phillips III 1981, The Late Pleis- tocene and Holocene fauna and flora of Vulture Cave, Grand Canyon, Arizona. Southwestern Nat. 26(3):257-288.' Mead. J I., R. S. Thompson, and T. R. Van Devender. 1982. Late Wisconsinan and Holocene fauna from Smith Creek Canyon, Snake Range, Nevada. San Diego Soc. Nat. Hist. Trans. 20(l):l-26. Meiiringer, p. J. Jr. 1967. The environment of extinction of the Late Pleistocene megafauna in the arid southwestern United States. Pages 247-266 in P. S. Martin and H. E. Wright, Jr., eds., Pleistocene extinctions: The search for a cause. Yale Univ. Press, New Haven. Merriam, J. C 1913. Preliminary report on the horses of Rancho La Brea. Univ. California Pub. Dept. Geol. Bull. 7:397^18. Mifflin. M. D., and M. M Wheat 1979. Pluvial lakes and estimated pluvial climates of Nevada. Nevada Bureau Mines Geol. Bull. 94:1-57. Miller, S J 1979. The archaeological fauna of four sites in Smith Creek Canyon. In D. R. Tuohy and D. L. Rendall, eds., The archaeology of Smith Creek Canyon, eastern Nevada. Nevada State Mus. An- throp. Pap. 17:271-329. Miller, W. E. 1971. Pleistocene vertebrates of the Los Angeles Basin and vicinity (exclusive of Rancho La Brea). Los Angeles Co. Mus. Nat. Hist. Bull. 10:1-124. 1976. Late Pleistocene vertebrates of the Silver Creek local fauna from north central Utah. Great Basin Nat. 36(4):387-424. 1982. Pleistocene vertebrates from the deposits of Lake Bonneville, Utah. Nat. Geog. Res. Rept. 14:473-478. Moore, G W . and B. G Nicholas. 1964. Speleology: the study of caves. D. C. Heath and Company, Boston. i20pp. Mooser, O,. and W. W. Dalquest. 1975. Pleistocene mammals from Aguascalientes, Central Mexico. J. Mammal. 56(4):781-820. Mosimann, J E , AND P. S Martin. 1975. Simulating overkill by Paleoindians. Amer. Sci. 63:304-313. Myers. A J. 1969. Geology of the Alabaster Cavern Area. Oklahoma Geol. Surv. Guidebook 15:6-16. 390 Great Basin Naturalist Vol. 45, No. 3 Nelson, M E , and J H Madsen, Jr 1979. The Hay- Romer debate: fifty years later. Univ. Wyoming Contrib. Geol. 18(l):47-50. 1980. Paleoecology of the Late Pleistocene, large mammal community in the northwestern Bon- neyille Basin, Utah. Geol. Soc. Amer. Abs. Progr. 12(6) :299. Nelson, R B 1966. Structural development of northern- most Snake Range, Kern Mountains, and Deep Creek Range, Nevada and Utah. Amer. Assoc. Petrol. Geol. Bull. 50(5): 92 1-951. Nelson, R. S., and H. A. Semken. 1970. Paleoecological and stratigraphic significance of the muskrat in Pleistocene deposits. Geol. Soc. Amer. Bull. 81(12):3733-3738. Oesch, R. D 1967. A preliminary investigation of a Pleis- tocene vertebrate fauna from Crankshaft Pit, Jef- ferson County, Missouri. Nat. Speleo. Soc. Bull. 29(4): 163-185. Owen, R. 1869. On fossil remains of equines from central and south America referable to Equus conversi- dens, Ow., Equus tan, Ow., and Equus arcidens, Ow. , Royal Soc. London Phil. Trans. 159: 559-573. Parmalee, P W , and R D Oesch 1972. Pleistocene and Recent faunas from the Brynjulfson Caves, Missouri. Illinois State Mus. Rept. Inves. 25:1-52. Parmalee, P W , R D. Oesch, and J E. Guilday. 1969. Pleistocene and Recent vertebrate faunas from Crankshaft Cave, Missouri. Illinois State Mus. Rept. Inves. 14:1-37. Peterson, O A 1926. The fossils of the Frankstown Cave, Blair County, Pennsylvania. Ann. Carnegie Mus. 16:249-315. Repenning. C a. 1962. The giant ground squirrel, Paene- marmota. J. Paleontol. 36:540-556. Romer, a. S. 1928. A "fossil" camel recently living in Utah. Science 68(1749): 19-20. 1929. A fresh skull of an extinct American camel. J. Geol. 3:261-267. Savage, D. E. 1951. Late Cenozoic vertebrates of the San Francisco Bay Region. Univ. California Pub. Dept. Geol. Sci. Bull. 28:215-314. Schultz, J. R 1937. A Late Cenozoic vertebrate fauna from the Coso Mountains, Inyo County, Califor- nia. Carnegie Inst. Washington Publ. 487(4): 77-109. Scott, W E , W D McCoy, R R Shroba, and M Rubin 1983. Reinterpretation of the exposed record of the last two cycles of Lake Bonneville, western United States. Quat. Res. 20:261-285. Semken, H A. 1966. Stratigraphy and paleontology of the McPherson Equus Beds (Sandahl Local Fauna), McPherson County, Kansas. Univ. Michigan Mus. Paleontol. Contrib. 20:121-178. Skinner, M. F. 1942. The fauna of Papago Springs Cave, Arizona. Amer. Mus. Nat. Hist. Bull. 80(6): 14,3-220. Smith. G R 1978. Biogeography of intermountain fishes. In K. T. Harper and J. L. Reveal, eds., Inter- mountain biogeography: a symposium. Great Basin Nat. Mem. 2:17-42. Smith, G R , W L Stokes, and K F Horn 1968. Some Late Pleistocene fishes of Lake Bonneville. Copeia 4:807-816. Stock, C 1928. Tanupolama, a new genus of llama ft-om the Pleistocene of California. Carnegie Inst. Washington Pub. 393:29-37. 1936. A new mountain goat from the Quaternary of Smith Creek Cave, Nevada. South. California Acad. Sci. Bull. 35(3): 149-153. 1963. Rancho La Brea. Los Angeles Co. Mus. Sci. Ser. 20:1-83. Stokes, W. L., and K. C. Condie. 1961. Pleistocene bighorn sheep from the Great Basin. J. Paleontol. 35(3):598-609. Stokes, W L , andG H Hansen 1937. Two Pleistocene musk-oxen from Utah. Utah Acad. Sci. Arts Let. 14:63-65. Thomas, D H 1983. Large mammals. In D. H. Thomas, ed. , The archaeology of Monitor Valley: 2, Gate- clifiFShelter. Amer. Mus. Nat. Hist. Anthrop. Pap. 59(1): 126-129. Thompson, R S 1979. Late Pleistocene and Holocene packrat middens from Smith Creek Cave, White Pine County, Nevada. In D. R. Tuohy and D. L. Rendall, eds.. The archaeology of Smith Creek Canyon, eastern Nevada. Nevada State Mus. An- throp. Pap. 17:361-380. Thompson. R S , and J I Mead 1982. Late Quaternary environments and biogeography in the Great Basin. Quat. Res. 17(l):39-55. Valastro, S., Jr., E M Davis, and A G Varela 1977. Univ. of Texas Austin Radiocarbon Dates XI. Ra- diocarbon 19(2):280-325. Vaughan, T a 1972. Mammalogy. Philadelphia: W. B. Sanders Co. 463 pp. Webb, S. D. 1965. The osteology oiCamelops. Los Ange- les Co. Mus. Sci. Bull. 1:54. 1969. Extinction-origin ecjuilibria in Late Ceno- zoic land mammals of North America. Evolution 23:688-702. 1974. Pleistocene llamas of Florida, with a review of the lamini. Pages 170-213 in S. D. Webb, ed.. Pleistocene mammals of Florida. Lhiiv. Presses Florida. Wells, P. V. 1983. Paleobiogeography of montane islands in the Great Basin since the last glaciopluvial. Ecol. Mon. 53(4):341-382. WiLLOUGHBY, D P 1974. The empire of Equus. A. S. Barnes and Company, South Brunswick. 475 pp. ZlEGLER, A C 1963. Unmodified mammal and bird re- mains from Deer Creek Cave, Elko County, Ne- vada. In Deer Creek Cave, Elko County, Nevada. Nevada State Mus. Anthrop. Pap. 11:1.5-22. Zimina, R. P , AND I P Gerasimov 1969. The periglacial expansion of marmots (Mannota) in middle Eu- rope during the Upper Pleistocene. In M. Ters, ed., Etudes sur le Quaternaire dans le Monde. CNRS, Paris, 465-472. FIRST RECORD OF CLIMACIA CALIFORNICA (NEUROPTERA: SISYRIDAE) AND ITS HOST SPONGE, EPHYDATIA MULLERl (PORIFERA: SPONGILLIDAE) FROM IDAHO WITH WATER QUALITY RELATIONSHIPS William H. Clark' Abstract. — The spongillafly, CUmacia californica Chandler, and its sponge host, Ephijdatia mitlleri (Lieberkuhn), are reported from the state of Idaho for the first time. Climacia californica has not previously been reported from E. mulleri. Collections were made in the Burley-Twin Falls area, and detailed water quality data are provided for the first time for spongillafly larvae. The water quality data also expand the known tolerance limits of E. mulleri for water temperature, conductivity, pH, hardness, silica, and residue. On 16 July 1980, six larvae oi Climacia cali- fornica Chandler were collected in the Snake River (near River Mile 653.7) at Heyburn, Minidoka County, Idaho, at an elevation of 1265 m (Fig. 1). The Snake River at this point is part of Milner Lake, due to Milner Dam, which is located approximately 21 km down- stream to the west. The spongillallies were collected on the surfaces of the crustose sponge, Ephydatia mulleri (Lieberkuhn) (Fig. 2). The sponges were found near shore growing on the under- »^'l«3»-'J »(«.'« mm rfe:| mtmC Fig. 1. Snake River at Heyburn, Idaho. Habitat oi Climacia californica and Ephydatia mulleri. Idaho Department of Health and Welfare, Division of En\ Idaho, Caldwell, Idaho 83605. . 801 Reserve Street, Boise, Idaho 83712, and Mt I of Natural History, College of 391 392 Great Basin Naturalist Vol. 45, No. 3 Fig. 2. The sponge Ephijdatia mulleri from the Snake River near Heyburn. Scale: Pen is 13 cm in length side of rocks, bricks, and pieces of concrete and asphalt in approximately 0.25-0.5 m of water (Fig. 1). Materials and Methods Specimens were hand collected by pulling up rocks and debris and preserved in 70% alcohol. Voucher specimens of C. californica are deposited in the Orma J. Smith Museum of Natural History, College of Idaho, Cald- well (CIDA); the Entomology Collection, University of Idaho, Moscow (UI); the De- partment of Biology, Boise State University, Boise, Idaho (BSU); and in the private collec- tions of A. D. Allen and R. C. Biggam. Speci- mens of £. mulleri are deposited in the collec- tions of the O. J. Smith Museum of Natural History, College of Idaho, Caldwell (CIDA) and the Department of Biology, University of New Orleans, New Orleans, Louisiana. Water quality and (quantity data were col- lected and analyzed by the U.S. Geological Survey in accordance with their standard pro- cedures, as well as with "Standard Methods" (American Public Health Association 1971). Results and Discussion The aquatic neuropterous family Sisyridae has not previously been reported from Idaho. Climacia californica was originally described from California (Chandler 1953) and was pre- viously known from only California and Ore- gon (Chandler 1963). The species determina- tion of the Idaho specimens is based on comparisons of the larvae with the illustra- tions of Brown (1952) and Chandler (1953) and on a collection of adults taken by sweeping shoreline vegetation at Echo Lake near the Snake River at Twin Falls, Twin Falls County, by Albert D. Allen 13 July 1976. Additional specimens were collected at this locality by R. C. Biggam 22 July 1976 and b\' R. C. Biggam and L. R. Schoenike 19 August 1982. Twin Falls is approximately 42 river km down- stream from the Heyburn locality. The sponge, E. mulleri, is listed as widely distributed in the Northern Hemisphere by a variety of workers (Gurney and Purfin 1959, Pennak 1953, Penney 1960, and Penney and Racek 1968, to list a few) but has apparently not been reported from Idaho. July 1985 Clark; Idaho Sisyhidae 393 Table 1. Water quality parameters for the Snake River near Burley, Idaho 1974-1981. Standard Parameter Units (N) Mean Deviation Maximum Minimum Water temperature CENT 70 10.20 7.10 23.00 0.00 Flow CFS 69 6519.30 4590.01 21300.00 390.00 Turbidity JTU 44 6.30 4.32 22.00 1.00 Conductivity Micromho 70 447.0 71.08 842,00 351.00 Dissolved oxygen MC/L 67 10.60 1.96 14.50 6.80 DO saturation Percent 61 107.00 16.52 163.00 59.00 COD low level MG/L 33 12.70 13.28 77.00 0.00 COD high level MG/L 31 12.80 8.36 38.00 0.00 pH SU 68 8.20 0.49 8.90 6.30 Carbon dioxide MG/L 42 1.90 1.46 7.20 0.50 Total alkalinity as CACO3 MG/L 47 160.00 20.55 202.00 124.00 HCO3 ion HCO3 MG/L 41 189.00 23.92 230.00 150.00 CO3 ion CO3 MG/L 40 4.00 5.56 22.00 0.00 Total nitrogen as N MG/L 70 0.95 1.10 8.40 0.20 Organic nitrogen as N MG/L 70 0.60 0.93 7.80 0.02 Ammonia total-NH4 MG/L 22 0.06 0.07 0.27 0.00 NH3 + NH4 - N total MG/L 70 0.05 0.05 0.25 0.00 Total Kjeldahl nitrogen MG/L 70 0.63 0.94 7.90 0.05 NO,, and NO3 N total MG/L 70 0.32 0,46 3.10 0.00 Total PO, PO, MG/L 4 0.08 0,05 0.15 0.03 Phosphorus total MG/L 70 0.06 0,03 0.14 0,01 Total organic carbon MG/L 49 2.96 1.32 7,00 1,40 Total hardness as CaC03 MG/L 12 165.83 53. 17 190.00 0,00 Calcium dissolved MG/L 11 47.00 3.44 51.00 42.00 Magnesium dissolved MG/L 11 15.60 1.12 17.00 14.00 Sodium dissolved MG/L 11 19.30 2.80 23.00 15.00 Potassium dissolved MG/L 11 3.40 0..34 4.00 2,90 Chloride total MG/L 11 18.54 3.93 23.00 13,00 Sulfate total MG/L 11 41.80 8.27 58.00 27,00 Silica dissolved MG/L 49 14.80 5.65 25.00 3.00 Arsenic total M-G/L 53 3.40 1.08 7.00 1.00 Cadmium total |xG/L 53 5.40 4.74 12.00 0.00 Chromium total M,G/L 52 2.70 4.36 10.00 0.00 Copper total jjlG/L 53 36.50 48.62 170.00 2.00 Iron total M,G/L 53 259.00 186.27 1100.00 10.00 Lead total jjlG/L 53 55.70 44.08 100.00 0.00 Zinc total fiG/L 53 45.30 65. 19 420.00 0.00 Mercury total M,G/L 53 0.10 0.18 1.10 0.00 Selenium total |xG/L 50 0.30 0.51 2.00 0.00 Fecal coliform bacteria /lOOML 53 30.00 25.56 86.99 0.00 Suspended solids at 150 C MG/L 26 14.90 17.07 87,00 2.00 Residue suspended at 180 C MG/L 14 17.90 16.59 59.00 3.00 Residue dissolved at 180 C MG/L 50 264.00 34.06 338.00 201.00 'Data from U.S. Geological Suney (197,5-1982). Little has been reported concerning the water quality requirements of spongillaflies. Roback (1974) describes the larvae of Sisyri- dae as not especially tolerant of extremes of water chemistry. He goes on to state that "Clijuacia areolaris was found at alkalinity greater than 210 ppm, total hardness greater than 300 ppm, and sulfate greater than 400 ppm." The mean alkalinity at the Burley Snake River station was 160 mg/1 (mg/1 = ppm) with a range of 124-202 mg/1; mean total hardness was 165.83 mg/1 (range 0-190 mg/1); and the mean sulfate concentration was 41.8 mg/1 (range 27-58 mg/1) (Table 1). In addition. Table 1 provides, for the first time, detailed water quality information pertaining to this insect. Harrison (1974) summarized the available water quality data for the sponge, E. mulleri. He listed the following ranges of parameters for which I have comparable data; conductiv- ity 31-100 micromhos/cm; total hardness 60-160 mg/1; pH 6. 1-8.5; residue 42-75 mg/1; 5:0, 0.7-11.6 mg/1; and temperatures of 394 Great Basin Naturalist Vol. 45, No. 3 16-24 C. Data for the above parameters from the Snake River, Biirley station, are as fol- lows: conductivity, X = 447 micromhos/cm (range 351-842); total hardness, X = 165.83 mg/1 (range 0-190); pH, X = 8.2 (range 6.3-8.9); silica, X = 14.8 mg/1 (range .3-25); and temperature, X = 10.2 C (range 0-23). The term residue listed in Harrison (1974) could refer to several forms (Table 1). Sus- pended solids is one of the most commonly used measurements of residue and for the Snake River station had a range of 2-87 mg/1 (X = 14.9). The Snake River data thus expands the known water quality tolerances of E. mul- leri for temperature, conductivity, pH, hard- ness, silica, and residue. Table 1 lists 33 additional water quality parameters that have not been previously re- ported for either C. californica or E. mulleri. Acknowledgments M. A. Porrier provided the sponge identifi- cation. A. Allen provided collection records. J. B. Johnson critically reviewed the manu- script and gave valuable comments. R. C. Biggam assisted with the literature search and provided collection records. Literature Cited American Public Health Association 197L Standard methods for the examination of water and waste- water. Amer. Pubhc Health Assoc, New York. 874 pp. Brown, H. P. 1952. The hfe history oiClimacia areolaris (Hagen), a neuropterous parasite of fresh water sponges. Amer. Midi. Natur. 47:L30-160. Chandler, H P. 1953. A new species oiClimacia from California (Sisyridae, Neuroptera). J. Washington Acad. Sci. 43:182-184. 1963. Aquatic Neuroptera. Pages 234-236 in R. L. Usinger, ed., Aquatic insects of California. Uni- versity of California Press, Berkeley. 508 pp. Gurney. a. B , AND S Parfin 1959. Neuroptera. Pages 973-980 in W. T. Edmondson, ed.. Fresh-water biology. John Wilev and Sons, Inc., New York. 1248 pp. Harrison, F W 1974. Sponges (Porifera: Spongillidae). Pages 29-66 in C. W. Hart, Jr., and S. L. H. Fuller, eds. , Pollution ecology of freshwater in- vertebrates. Academic Press, New York. 389 pp. Pennak, R. W. 1953. Fresh-water invertebrates of the United States. Roland Press Co., New York. 769 pp. Penney, J T 1960. Distribution and bibliography (1892-1957) of the fresh-water sponges. Univ. South Carohna Publ. Ser. III. Biol. 3:1-97. Penney, J T., and A. A. Racek 1968. Comprehensive revision of a worldwide collection of freshwater sponges (Porifera: Spongillidae). Bull. U.S. Nat. Mus. 272:1-184. Roback, S S. 1974. Insects (Arthropoda: Insects). Pages 3l;3-376 in C. W. Hart, Jr., and S. L. H. Fuller, eds.. Pollution ecology of freshwater inverte- brates. Academic Press, New York. 389 pp. US. Geological Survey. 1975. 1974 Water resources data for Idaho. Part 1, Surface water records. Part 2, Water quality records. USGS, Boise, Idaho. 295 and 297 pp. ' 1976. Water resources data for Idaho, water vear 1975. USGS Water-Data Rpt. ID-7.5-1, Boise, Idaho. 684 pp. 1977. Water resources data for Idaho, water vear 1976. USGS Water-Data Rpt. ID-76-1, Boise, Idaho. 634 pp. 1978. Water resources data for Idaho, water year 1977. USGS Water-Data Rpt. ID-77-1, Boise, Idaho. 640 pp. 1979. Water resources data for Idaho, water year 1978. Vol. 1. Great Basin and Snake River Basin above King Hill. USGS Water-Data Rpt. ID-78-1. 457 pp. 1980. Water resources data for Idaho, water year 1979. Vol. 1. Great Basin and Snake River Basin above King Hill. USGS Water-Data Rpt. lD-79-1. 385 pp. 1981. Water resources data for Idaho, water year 1980. Vol. 1, Great Basin and Snake River Basin above King Hill. USGS Water-Data Rpt. ID-80-1. 375 pp. 1982. Water resources data for Idaho, water year 1981. Vol. 1. Great Basin and Snake River Basin above King Hill. USGS Water-Data Rpt. ID-81-1. 447 pp. POA L. IN NEW MEXICO, WITH A KEY TO MIDDLE AND SOUTHERN ROCKY MOUNTAIN SPECIES (POACEAE) Robert J. Sorenj;' Abstract — The 23 species and subspecies o( Poa that occur in New Mexico are described in detail. Collection locations of these species in New Mexico are given in dot distribution maps. A descriptive key to the species oiPoa that occur in Arizona, Colorado, Utah, Wyoming, and New Mexico, including general distribution for species not in New Mexico, is provided. Three new nomenclatural combinations are proposed: Poa arctica suiisp. aperia ; P. fendleriana subsp. lon^iligula ; P. f. subsp. albescens. Subgeneric affinities of species of the southern Rocky Mountain region are indicated. Of the 25 species that occur within the region, the geographic affinities are; circumpolar {P. alpina, P. arctica, P. glattca, P. interior, and P. leptocoma), Beringian (P. lettennanii, and P. pattersonii), Eurasian weeds (P. annua. P. bulbosa, P. compressa, P. palustris, P. pratensis (also circumpolar), and P. trivialis). Pacific Northwest (P. bolanderi, P. ciisickii, P. nervosa, and P. stenantha), northern Great Plains (P. arida), southern Great Plains and South American Pampas (P. arachnifera). Great Basin-Californian (P. fendleriana and P. secunda), and Middle and Southern Rock-y Mountain (P. bigelovii, P. curta, P. occidentalis, P. reflexa, and P. tracyi). The only native diploid Poa species known in the Southern Rocky Mountains, the contiguous United States, and southern Canada are P. lettermanii and P. occidentalis. Poa L. (Bluegrass) Low or moderately tall annuals or perenni- als, tufted, rhizomatous, or stoloniferous. Blades flat, folded, or rolled, tips usually curved and prowlike, glabrous on back, glabrous, scabrous, or pubescent above. Sheaths glabrous to pubescent, margins fused at least at base. Ligules membranous, some- times hairy on back, smooth or minutely fringed terminally. Inflorescence an open or contracted panicle. Spikelets with 2-several (rarely I) florets, disarticulating above glumes and between florets. Ghimes narrow to broad, sharply acute to rounded, usually keeled on back, I- to 3-nerved, shorter equaling or rarely longer than first lemma. Lemmas awnless, acute to rounded at apex, typically firm with scarious hyaline margin and tip, usually 5-nerved (3-11), prominently keeled on back or rounded (Secundae), glabrous scabrous or pubescent, with hairs confined to nerves or throughout, frequently with tuft of long cobwebby hairs from callus (pubescence characters are best seen on lower lemmas of spikelet). Paleas with chlorophyll (unlike Koe- leria), two keels glabrous, scabrous, ciliate, or villous. Flowers perfect or unisexual (dioe- cious, gynodioecious, gynomonecious). Sta- mens 3, anthers 0.2-4 mm long, or vestigial (early aborted). Pistil glabrous. Carijopses el- lipsoidal and usually somewhat compressed ventrally, hilum oval, less than 2/5 the cary- opses in length. Lodiciiles membranous, broadly lanceolate, usually lobed, 0.3-1.2 mm long. Basic chromosome number X — 7. (Reported 2n chromosome numbers are recorded in species descriptions, modal num- bers are in italics, and frequent numbers are boldfaced.) About 250 species in temperate and colder regions worldwide, extending into the sub- tropics and tropics as montane species and as cool-season grasses at low altitudes. Identifi- cation of Poa species is difficult because of the large number of species and because the lim- ited variety of characters distinguishing them mostly overlap. In addition, our traditional species criteria fail in certain groups (i.e., Poa, Stenopoa, and Secundae) wherein the high frequency of asexual reproduction by seed (agamospermy), high polyploidy, and hy- bridization combine to make the variation more or less continuous among many of the taxa. Biology Department, New Mexico State University, Las Cruces, New Mexico 88003. This study was Rinded in part by a Sigma Xi "Grant-in-Aid of Research, " The Scientific Research Society. 395 396 Great Basin Naturalist Vol. 45, No. 3 New Mexico county abbreviations: B = Bernalillo, DA = Dona Ana, CB = Cibola, CT = Catron, CV = Chaves, CF = Colfax, E = Eddy, GU = Guadalupe, GR = Grant, HD = Hidalgo, HR = Harding, LA = Las Alamos, LN = Lincoln, LU = Luna, MK = McKinley, MR = Mora, OT = Otero, RA = Rio Arriba, SD = Sandoval, SF = Santa Fe, SJ = San Juan, SM = San Miguel, SC = Soccoro, SR = Sierra, V = Valencia, TO = Taos, TR = Tor- rence, U = Union. Distribution records for all the species were checked at GH, NMC, NMCR, NY, UNM, US, UTP. Much ofthese data come from field- work, vouchers for which are housed at NMC. Subgeneric classification of the species of Poa occurring in the Southern Rocky Moun- tains is as follows: :lassified Oreinos A. & G. Abbreviatae Nannf. ex Tzvel Stenopoa Dumort. Tricopoa A. & G. "unclassified" "Secundae" "unclassified" Dioicopoa E. Desv. exC. Gay P. occidentalis. P. bolanderi, P. reflexa, P. traciji, P. ciirta (seiisu aiict) P. leptocoma P. lettennanii, P. pattersonii P. palustris, P. glaiica, P. interior, P. ncmoralis. P. camp res sa P. arida P. secunda sensit lato. P. stenantha . P. arachnifera [P. fendleriana, P. cusickii, P. nervosa?] Subgenus Section Poa Poa Bolbophorum A. &G. Ochlopoa(A. 6c P. annuo G.)Jiras Coenopoa P. trivialis Hylander Diversipoa Chrtek P. bigelovii & Jirasek Species TREATED here P. pratensis, P. arctica P. bulbosa, P. alpina Key TO Southern Rocky Mountain Species OF Poa This key includes all the species of Poa found in Arizona, Colorado, Utah, Wyoming, and New Mexico. Species definitely known from New Mexico are typed in bold face. Those species not known from New Mexico are preceeded by their general geographic distribution and followed by an author citation. 2(1). 3(2). 4(3). 5(4). Anthers, most of them, over 1 mm long, or aborted (vestigial); flowers perfect, unisexual, or forming bulblets; perennial 10 Anthers consistently 1 mm or less in length; flowers mostly perfect; annual or perennial 2 Callus of lemmas long webbed; lemmas glabrous to minutely scabrid on the keel; plants annual, of midmontane habitats, rare in n UT, ID, NV, and WA, mostly in CA, OR P. bolanderi Vasey Callus of lemmas webbed or not; lemmas glabrous to pubescent, if webbed, then prominently pubescent at least on nerves; plants annual or perennial, of various habitats 3 Callus glabrous; lemmas glabrous or with a few minute hairs on keel, 2.. 5-3 mm long, relatively broad, blunt at apex; spikelets 3-4 mm long, anthers averaging 0.6 mm long; alpine plants 2-10 cm tall; sw CAN, w US, to COL P. lettennanii Vasey Callus webbed or, if glabrous, lemmas prominently \'illous at least on keel; other characters various 4 Plant perennial, sometimes flowering first year, sometimes alpine; if flowering first year, then callus webbed and panicle branches eventually widely spreading or rettexed; palea keels various 6 Callus glabrous; panicles more or less open F. annua Callus webbed; panicles long and contracted P. bigelovii Plants annual, rarely biennial, rarely alpine; callus glabrous and lemmas \ illous on all 5 nerves, or callus webbed and panicle narrowly contracted; palea more or less villous on keels 5 July 1985 SoRENC; New Mexico PoA 397 6(4). Plants 3-15 cm tall, tufted from narrow base; panicles narrow, short l)ranched, little-exerted above basal leaves; callus sometimes webbed; lemmas pale green to purple, not bronze colored at tip, pubendous all over or pubescent only on nerves, rarely nearly glabrous; basal sheaths persistent, papery, upper ones closed \ the length or less; leaves green, soft, flat, lax; plants high alpine, on cold exposures, uncommon, sw CAN to UT and CO P. pattersonii Vasey — Plants mostly over 15 cm tall, loosely tufted; panicle relatively broad, long branched, well exerted above basal tuft of leaves; callus webbed; other charac- ters variable; plants of midelevations to low alpine 7 7(6). Sheaths closed | or less the length; ligules mostly fringed on margins; first glume mostly 3-nerved (short anthered plants of F. nemoralis and P. palustris) 19 — Sheaths closed \-l the length; ligules entire or jagged, smooth on margins; first glume mostly 1-nerved 8 8(7). Sheaths rarely glabrous, more or less densely retrorse-scabrous; panicle (8) 13-40 cm long, internodes (3) 4+ cm long; palea nerves glabrous or sparsely scabrous; lemmas mostly sparsely puberulent between 5 nerves, mostly pale green, margins and tips white-hyaline or sometimes purple tinged; plants of midmontane forested habitats P. occidentalis — Sheaths glabrous, roughened, or sparsely retrorse-scabrous; panicle mostly less than 12 cm long, internodes shorter than 3.5 cm; palea nerves mostly villous or ciliate-scabrous; lemmas rarely pubescent between the nerves, pale green to strongly purplish, tips bronze tinged or white-hyaline; plants usually of more moist subalpine and alpine habitats 9 9(8). Palea keels glabrous, scabrous, or with relatively regularly spaced, antrorsely curved, ciliate hairs, internerves with long cells only; glumes and lemmas narrowly lanceolate, unequal, first very narrow; lemmas sharply acute, inter- mediate nerves usually obscure, never villous; plants of wet habitats ... P. leptocoma — Palea keels rarely glabrous, mostly villous-pilose, internerves with both short and long cells; glumes and lemmas broadly lanceolate, glumes subequal; lemmas bluntly acute, intermediate nerves frequently distinct, frequently somewhat villous; plants of wet or dryish habitats P. reflexa 10(1). Florets normally forming leafy bulblets; plants tufted, culms somewhat bulbous at base P. bulbosa — Florets not normally forming bulblets; habits various, but culms not bulbous at base 11 11(10). Callus glabrous (infrequently with hairs similar to and continuous with those of lemma-keel), or (in P. stenantha and less often in P. secunda) with short, relatively straight hairs generally distributed around top of callus, these not restricted to dorsal side 22 — Callus webbed with tuft of short-kinky or long-sinuate hairs on dorsal surface, these isolated from those of lemma-keel 12 12(11). Plants dioecious, rhizomatous; panicle oblong, compact, terminal branches densely flowered from near base and, at least in female plants, densely scabrous; plants of se Creat Plains, rare and doubtfully native in our area P. arachnifera ~ Plants perfect flowered (rarely pistillate); if pistillate, panicles more open, branches sparsely flowered in distal | 13 13(12). Sheaths closed 1-^ length; flowers perfect with anthers averaging 2.2 mm long, and/or pistillate with vestiges of anthers; panicles mostly 13-29 cm long, lower internodes mostly over 3.5 cm long, branches widely spreading to sharply reflexed; tall subrhizomatous plants P- tracyt 398 Great Basin Naturalist Vol. 45, No. 3 — Sheaths closed ca I or less the length; flowers normally all perfect; panicles mostly less than 13 cm long, internodes rarely over 3.5 cm long (if panicles and internodes longer, then lemmas glabrous between nerves and anthers less than 1.9 mm long); habits various 14 14(13). Culms and nodes strongly flattened; plants strongly rhizomatous; sheaths closed I or less the length; panicle short, compact to loose, with short branches; spikelets compact; ligule ciliate margined; lemma with a weak web .... P. compressa — Culms and nodes not decidedly flattened; other characters variable 15 15(14). Plants densely tufted (or in damp habitats with only a few culms in a bunch and then decumbent at base and rooting at nodes, but not definitely rhizomatous); sheaths closed near base to ca | (5) the length (to ca | in P. trivialis); panicle branches distinctly scabrous, mostly angled 18 — Plants with strong rhizomes present; sheaths closed ca l-l the length; panicle branches glabrous to sparsely scabrous, terete 16 16(15). Glumes distinctly keeled, scabrous on nerves, second evidently shorter than first lemma; lemmas villous on keel and marginal nerves, glabrous between them (intermediate nerves rarely distinct, rarely sparsely pubescent); callus strongly webbed; ligules truncate, fringe margined; panicle often with 4 or more branches at lowest node; plants mostly lower than alpine P. pratensis — Glumes weakly keeled, nearly glabrous, second subequal to or longer than first lemma; lemmas villous on keel and marginal nerves and frequently on interme- diate nerves, often puberulent between nerves; callus hairs variable; ligules truncate to long-acute, entire; panicle usually with fewer than 4 branches at lowest node; plants subalpine and alpine 17 17(16). Callus web scant and short to long and copious; panicles loose, branches flexu- ous; culms not wiry, mostly single, with several sterile shoots; ligules truncate to acute P. arctica subsp. grayana {sensii lato) — Callus web scant, short; panicles narrow or wide open, branches strict; culms wiry, mostly closely tufted, with relatively few sterile shoots; ligules acute to acuminate P. arctica subsp. aperta 18(15). Lemmas pubescent on keel below, glabrous or rarely very sparsely puberulent on marginal nerves, glabrous between nerves; ligules 3-10 mm long, entire or lacerate; first glume very narrow, 1-nerved, curved inward P. trivialis — Lemmas pubescent on keel and marginal nerves and often between nerves; ligules less than 4 mm long, often minutely ciliate fringed on margins; first glume very narrow or quite broad, 3-nerved, not strongly arched inward; (long- anthered individuals of P. pattersonii may key out here, see lead 6) 19 19(7,18).Culms somewhat scabrous near nodes, stout, leafy to well above middle, fre- quently decumbant and rooting at nodes, sometimes branching above base, mostly 2.5-120 cm tall; ligules (1) 2-6 mm; panicles mostly 10-30 cm long, branches with abundant tertiary branching, many flowered; lemmas glabrous between keel and marginal nerves, abruptly incurved near tip; callus-web well developed; rachilla glabrous or scabrous P. palustris — Culms mostly glabrous near nodes, either, "lax, slender, leafy to above middle," or, "wiry, strict, with few leaves, these mostly in lower ^," never decumbent and rooting at nodes, never branching above base, height variable; ligules 0.2-3 mm long; panicles mostly shorter (except in P. nemoralis), main panicle-l)ranches unbranched or with some secondary branches, mostly fewer flowered; lemmas glabrous or pubescent between keel and marginal nerves, not abruptly incurved near tip; callus-web from a minute tuft to well developed; rachilla glabrous, scabrous, or puberulent 20 July 1985 SoRENG: New Mexico PoA 399 20(19). Glumes very narrow (subulate), margins nearly straight; anthers 0.8-1.2 mm long; ligules nearly absent to 0.2 (1) mm long, truncate, pubescent on back, fringe margined; panicles many flowered, 5-25 cm long, branches widely spreading; foliage green; introduced, rare in w US, not known from our region P. nemoralis L. — Glumes broader, margins of second frequently angled outward; anthers (1) 1.2-2 mm long; ligules various; panicles mostly fewer flowered, mostly less than 12 cm long, branches ascending; foliage green or glaucous; native species 21 21(20). Lemmas glabrous between nerves, rarely sparsely villous on intermediate nerves; callus web short or infrequently absent; plant green; culms erect, lax to slender-wiry, often geniculate at lower nodes, uppermost node in middle \; ligules ca 0..3-1.5 (2) mm long, scabrous on back, truncate, fringe margined; glumes sharply keeled, sharply acute and frequently curved in or out at tip; panicles 2-15 cm long, somewhat lax (or strict in small alpine forms), branches slender p. interior — Lemmas mostly pubescent between nerves or pubescent on intermediate nerves; callus web well developed to frequently absent; plant green or glaucous; culms erect or spreading, stout to wiry, strict, uppermost node in near base; ligules 1-3+ mm long, similar to P. interior or, more often, rounded to obtuse and lacerate and then less scabrous on back and not or weakly fringe margined; glumes similar to P. interior, or less sharply keeled, more obtuse at tip and tips not divergent; panicles mostly less than 6 cm long, strict, branches more stout P. glauca (sensu lato ) 22(11). Sheaths closed 5-^ the length, glabrous; panicles 10-20 cm long, open and diffusely flowered, lowest axial internodes mostly 2.. 5-6 cm long, branches mostly reflexed; plants loosely tufted, short-rhizomatous, perfect-flowered or anthers of some or most flowers vestigial; lemmas strongly keeled, glabrous to sparsely hirtelous or puberulent on keel base and between nerves near base; plants of midmontane habitats, uncommon, ne UT, w ID, w WY, possibly to sw CAN Poa curta (sensu. aiict non Rydb.) — Sheaths closed ^ to only near base, if closed over | length and plant rhizomatous, then panicles mostly less than 13 cm long, axial internodes less than 2.5 cm long, branches mostly ascending, and at least lower sheaths puberulous; panicles various, mostly more condensed, or axial internodes mostly less than 2 cm long, branches mostly ascending; plants rhizomatous or not, variously flowered, lemmas various; plants of various habitats, mostly more widespread 23 23(22). Flowers predominently perfect, stamens and stigmas more or less synchronously developed (remnants of a few anther sacs usually remain on even the most mature plants, filaments persistent); lemmas keeled or not; sheaths frequently closed I the length or less (unless rhizomatous) 30 — Flowers pistillate (anthers vestigial or partially developed but nonfunctional) or staminate (ovaries and stigmas very reduced or undeveloped) or infrequently perfect in part; lemmas keeled on the back; sheaths closed ^-^ the length 24 24. Plants strongly rhizomatous; culms few, loosely tufted; upper culm blade well developed; basal tuft of leaves weakly developed; ligules of lower culm leaves obtuse to truncate, finely scabrous to puberulent on back; spikelets green; lemmas villous to scabrous on keel and marginal nerves and mostly scabrous between; sheaths (at least lower ones) finely puberulent (rarely nearly glabrous); midmontane forest openings and thickets, to lower alpine, widespread P, nervosa var. wheeleri 400 Great Basin Natur\list Vol. 45, No. 3 — Plants tufted, culms several to many or, if stolonous or rhizomatous, then the upper culm blade mostly very reduced; basal tuft of leaves well developed (except in P. epilis); ligules various; spikelets green or pale green and shining; lemmas and sheaths various; plants of various habitats 25 25(24). Lemmas sparsely to prominently pubescent on keel and marginal nerves, some- times pubescent on intermediate nerves, and infrequently between nerves 28 — Lemmas scabrous to glabrous, less commonly finely and very sparsely puberu- lous on the lower keel and marginal nerves 26 26(25). Plants tufted, short-rhizomatous; culm blades strongly reduced upward, upper- most blade reduced or absent on most culms; basal blades relatively broad (1-3.5 mm wide) and firm; sheaths mostly smooth to sparsely scabrous, collar margins spiculate; ligules 0.25-1 (2) mm long, truncate to rounded, very scabrous on back, fringed on upper margin; lemmas mostly very smooth, sparsely scabrous to very sparsely puberulent on keel and marginal nerves, glabrous between nerves; plants of Sierra Madre Occidental P . fendleriana subsp. albescens — Plants densely tufted, never rhizomatous; culm blades not regularly and strongly reduced upward, uppermost blade mostly over 1 cm long; basal blades mostly 1 mm broad or less; ligules various; lemmas glabrous, or scabrous on nerves and frequently between them, infrequently very sparsely puberulous on keel; plants not known from south of s UT-CO state line in interior United States 27 27(26). Panicle axis and branches moderately to densely scabrous; blades mostly basal, mostly filiform, involute, culm blades few, occasionally broader above and to 1.5 mm wide; culm sheaths more or less scabrous, mostly covering nodes, upper sheath often rather loose; lemmas scabrous on nerves, and usually between them, never glabrous all over; spikelets pale green, shining; plants either pisti- late or staminate; n Great Plains grasslands, high sagebrush communities to alpine grasslands, widespread nw US, sw CAN, infrequent in n UT and CO (P. suharistata Rydb. form in Rocky Mts.) P.cusickii Vasey subsp. cusickii — Panicle axis and branches smooth to infrequently moderately scabrous; blades about equally basal and cauline, basal ones usually filiform, involute, culm blades generally broader and flatter and mostly over 1.5 mm wide; culm sheaths mostly glabrous, mostly exposing culm nodes; lemmas glabrous or scabrous on and between nerves, rarely very sparsely puberulous on keel; spikelets green, often purple tinged; flowers pistilate; subalpine to alpine, common w US, sw CAN, to s UT and CO P. cusickii subsp. epilis (Scribn.) W. A. Weber 28(25). Lemmas more or less evenly pubescent over basal portion; plants densely tufted, lacking any rhizomes; blades very narrow, basal ones less than 1 mm wide, uppermost blade usually filiform and over 5 mm long; flowers rarely with well developed anthers; dry hills, w ID, w MT, w WY, ne UT, nw CO, and across northern plains of CAN, in ALB, SKW, MAN. (Although the type of this is like P. cusickii, many of the plants that key-out here are more like P. fendleriana in having firm, broader, folded blades, with the upper ones very reduced. This taxon appears to have evolved from hybridization between the two species) P. cusickii subsp. puhens Keck — Lemmas strongly villous on keel and marginal nerves, infre(}uently pubescent between nerves; plant more loosely tufted, mostly with short, lateral, rhizoma- tous shoots; blades over 1.5 mm wide, uppermost blade, when present, firm, mostly very reduced; flowers pistilate or staminate, (staminate plants found occasionally in CO and more fre(|uently to south and west); widespread 29 July 1985 SoRENc;: New Mexico Poa 401 29(28). Ligules from merely a spiculate ridge to 1 (2) mm long, truncate to rounded and minutely fringed terminally, mostly quite scabrous on back; sheaths mostly scabrous (especially so on margins about collar) or finely puberulous; lemmas infrequently villous on intermediate nerves, glabrous between nerves P- fendleriana subsp. fendleriana — Ligules 1.8-11 mm long, obtuse, acute, or acuminate, mostly sparsely scabrous to glabrous on back, entire; sheaths glabrous to scabrous or minutely puberu- lous, but not distinctly so on margins near collar; lemmas occasionally villous on intermediate nerves, glabrous or occasionally sparsely puberulent between nerves P . fendleriana subsp. longiligula 30(23). Anthers not more than 1.2 mm long; lemmas distinctly keeled; foliage bright green; panicles not much exerted above basal tuft of leaves; blades less than 2 mm broad, soft, lax; plants narrowly-tufted, 2-15 cm tall; plants of high alpine situations, with both vegetative and fertile shoots; uncommon (see P. pattersonii, couplet 6) — Anthers usually more than 1.2 mm long; lemmas keeled or not; foliage green or glaucous; panicles well exerted above basal tuft of leaves; blades various; plants, if less than 15 cm, tall and alpine, then tufts dense, (in P. glauca, most shoots fertile, leaves relatively firm and strict) 31 31. Lemmas prominently pubescent; sheaths closed \-\ the length, or, open most of length 34 — Lemmas glabrous or scabrous, rarely very sparsely puberulent on keel; sheaths open most of length 32 32(31). Ligules 2.5-6 mm long, acute or acuminate, glabrous or sparsely scabrous on back P. secimda (P. nevadensis form) — Ligules 0.25-4 mm long, truncate to obtuse, sparsely to densely scabrous on back; culms in large loose tufts (plants rarely rhizomatous) 33 33. Foliage somewhat coarse; blades involute, mostly less than 1.5 mm wide; plants mostly of low mountains and desert plains in poorly drained soils. Great Basin, WA to MT, south to UT and CO P. secunda (P. juncifolia form) — Foliage more lax; blades flatish, mostly 1.5-3 mm wide; plants of high sagebrush slopes and higher, mostly of well-drained, rich soils P. secunda (P. ampla form) 34(31). Plants densely tufted; lemma pubescence variable 36 — Plants rhizomatous; lemma pubescence definitely stronger on nerves than between them, or glabrous between them 35 35(34). Sheaths closed (|) |-| the length; glumes weakly keeled; lemma tips acute; callus glabrous or with an inconspicuous web; panicle branches strict or flexuous; plants subalpine to alpine P. arctica subsp. grayana s. str. (see also couplet 17) — Sheaths closed I the length or less; glumes strongly keeled; lemma tips often blunt; callus glabrous or with a few hairs continuous with and like those of the lemma keel; panicles and branches strict; plants of arid, alkaline plains and piedmont valleys, e (rare w) of continental divide P. arida 36(34). Upper sheaths closed more than | the length; culm bases enclosed in persistent, thickened, closely overlapping sheaths; fohage green; spikelets broadly rounded, almost cordate at base; panicle branches glabrous to sparsely scabrous, terete, strongly divergent, intricately rebranched, and closely flowered; plants of moist alpine situations, circumboreal, to s UT and CO P. alpina 402 Great Basin Naturalist Vol. 45, No. 3 — Upper sheaths closed less than \ the length; culm bases not enclosed in persis- tent, thickened, closely overlapping sheaths; foliage green or glaucous; spikelets broadly rounded or more elongate, not at all cordate at base; panicle branches moderately to strongly scabrous, distinctly angled to terete, but the branches not strongly divergent, intricately rebranched, and closely flowered; plants of various situations 37 37(36). Lemmas distinctly keeled, 4—6 mm long, pubescence longer and stronger on nerves than between them, or glabrous between them; rachilla internodes mostly over 0.8 mm long; spikelets mostly over 7 mm long; panicles open, somewhat lax, mostly 7-17 cm long, lower panicle branches mostly more than 3.5 cm long, variously divergent, moderately scabrous on weak angels; plants 2.5-6 cm tall, with few flowering shoots and many vegetative shoots; blades mostly well over 3 cm long, soft, lax, bright green; ligules 2-4 mm long, obtuse to acute, entire to lacerate on margins; callus often with hairs across top around base of lemma (different from those of surface of lemma); plants of mountain slopes, AK, sw CAN, to n UT, and central CO, where rare [including P. macroclada Rydb. ] P. stenantha Trin. — Lemmas keeled or not; if keeled, lemmas to 4 mm long, rachilla internodes mostly less than 0.8 mm long, spikelets less than 5 mm long, panicles less than 7 cm long with branches less than 4 cm long, plants less than 3 dm tall 38 38(37). Lemmas keeled on back, pubescence mostly longer and stronger on keel and marginal nerves than between them, occasionally glabrous between them; plants mostly less than 30 cm tall, with few vegetative and many flowering shoots; callus glabrous; rachillas to 0.8 mm long; panicles strict, branches strictly ascending, strongly scabrous on prominent angles; leaf blades less than 3 cm long, strict, not thin and soft; foliage green or glaucous; ligules 1-2 mm long, truncate to obtuse, often minutely fringe margined; plants of high mountains on dry slopes and ridges P. glauca subsp. rupicola (see also couplet 19) — Lemmas more or less rounded across back, crisp-puberulent all across base, pubescence usually not or little longer and stronger on nerves (except in P. gracillema form), infrequently nearly glabrous between nerves; callus glabrous or with a few hairs no longer than and not separated from those of lemma keel, or with hairs across top around base of lemma (different from those of surface of lemma); plants of various heights, with many vegetative shoots and few flowering shoots; rachilla (0.6) 0.8-1.9 mm long; panicles various, branches moderately scabrous, weakly angled; leaf blades of various lengths, lax, very thin, soft; foliage green or infrequently glaucous; ligule 2-7 mm long, acute to acuminate, margins entire or lacerate; plants of various habitats (forms of P. secunda, continue to couplet 39) 39(38). Plants mostly less than 4 dm tall; basal tuft of leaves fine, mostly less than 3 cm long and panicles contracted except in anthesis; plants of dry open ground at moderate elevations, flowering in early spring, sw CAN, to UT, and n CO P. sandhergii form — Plants mostly over 4 dm tall; basal tuft of leaves mostly over 4 cm long, or panicles persistently open; plants of more mesic situations or higher elevations, flowering late spring to late summer 40 40. Panicles persistently open, lemmas evenly puberulent over base, or frequently with stronger pubescence on nerves than between them; plants of high moun- tains to alpine, sw CAN, south to n UT and CO P. gracillema form — Panicles open only in anthesis, lemma pubescence fairly evenly developed over base; plants of various situations, widespread P. canbyi form fuly 1985 SoRENG: New Mexico Poa 403 Taxa of Poa in New Mexico Poa annua L. Poa annua L. , Sp. Pi. 68. 1753. Annual Bluegrass. Annual, slender, tufted. Culms erect or as- cending, often geniculate at the nodes, 0.1-3+ dm tall. Leaves light green, soft, glabrous. Sheaths closed about I the length. Ligules glabrous, entire, about as long as blade is wide. Blades mostly flat, soon wither- ing, mostly 1-2 (4) mm wide, 1-8 cm long. Panicles 1-5 cm long, open, ovate or broader, branches smooth, strict, divergent, densely flowered in distal i Spikelets 2-6-flowered, 2.5-6 mm long. Glumes narrow, unequal, the first often curved inward and l-l as long as adjacent lemma, second glume about 5-5 as long as adjacent lemma. Lemmas broadly lanceolate, smooth, prominently 5-nerved, villous on nerves (rarely nearly glabrous) and glabrous between. Callus glabrous, rachilla internodes glabrous and short. Paleas mostly villous on nerves, glabrous between. Flowers of lower florets perfect, terminal one reduced to nob on rachilla, or developed and then pistillate (gynomonecious). Anthers 0.3-1 mm long. Chromosome numbers: 24-26, 28, 52. Habitat: A common lawn weed, potential in every habitat in New Mexico so long as there is shade, winter moisture, and distur- bance. Flowering continually in irrigated ar- eas, otherwise primarily late winter-early spring. J Distribution: Introduced from Europe. New Mexico: B, DA, OT, LN, RA, SD, and SF, but probably in every county. Comment: This species is perhaps one of the world's most widespread weeds. It was present in Sitka, Alaska, by at least 1829, where collected and mixed with the type of Poa leptocoma Trin. Probably the first collec- tion in New Mexico was made by S. M. Tracy in 1887 near Santa Fe. Poa arachnifera Torr. Poa arachnifera Torr., in Marcy, Expl. Red Riv. 301. 1853. Lectotype (A. Hitchc): Marcy "crop tim- bers, Arkansas," in 1852 (NY). Texas Bluegrass. Perennial from long, slender rhizomes. Culms tufted, erect, 2.5-8.5 dm taU. Leaves green, firm. Sheaths closed |-| the length, keeled. Ligules 1^ mm long, acute. Blades flat or folded, often inrolled on margins, 1.5-4.5 mm wide, elongate. Panicles oblong, contracted, somewhat lobed, 3-15 cm long, branches slender, strongly scabrous (female) to nearly glabrous (male), terminal ones densey flowered. Spikelets slightly dimor- phic, compact, compressed; male 2-10-flow- ered, 4-8 mm long; female 2-5-flowered, 4-9 mm long. Glumes 1-5-nei-ved, smooth to scabrous, narrow, subequal. Lemmas 5-7- nerved, smooth to sparsely papillose-rough- ened; male 3.5-5 mm long, acute, sparsely villous to scabrous on keel, callus with several slender, long, villous hairs; female 4.2-6.4 mm long, with long hyaline, acute tips, densely villous on keel and marginal nerves, sometimes sparsely so on intermediate nerves, glabrous between nerves, callus with a copious tuft of long, plicate hairs. Paleas glabrous to sparsely long-ciliate (male) to vil- lous (female). Rachillas glabrous, internodes less than 1 mm long. Flowers unisexual (dioe- cious, hermaphroditic flowers developed in- frequently and then resembling the female ones). Anthers 1.6-2.7 mm long. Chromo- some numbers: 42, ca 54, 56, ca 63, 84. Habitat: One record in New Mexico from Bosque del Apache (V), in salty flood plain, ca 1520 m, in 1957. Flowering May. Doubtfully native in New Mexico. One collection, C. Wright 2042 in 1851-1852 labeled as "New Mexico," belongs to this species, but where it was actually collected it is not known. Distribution: Apparently native to the south central Great Plains in Kansas, Okla- homa, Texas, Arkansas, but possibly intro- duced from South America in historical times. Introduced in all the southeastern states, but rare to the west. Seeded as a pasture and lawn grass in some areas. Comment: Poa arachnifera is more similar to such South American dioecious species as P. denudata Steud., P. bonarensis (Lam.) Kunth., and P. montevidensis Arech. than to any other North American species. Poa arctica R. Br. Poa arctica R. Br., in suppl. App. Parry's Voy. 288 ("188"). 1823. Type: Parry Mellville Island, Arctic America. 404 Great Basin Naturalist Vol. 45, No. 3 Poa arctica subsp. aperta (Scribn. & Merr.), comb. nov. Poa arctica subsp. aperta (Scribn. & Merr.) comb. not. ; Poa aperta Scribn. & Merr., USDA, Div. Agrost. Circ. 35:4. 1901. Type: Shear ir Shear 98, open mountainside, 2896 ni elev., Telluride, Colo., 1 Sept. 1900 (US 28611600!, NY). Perennial with rhizomatous. Culms wiry, decumbent, often densely tufted, smooth, 2-6 dm tall. Leaves greenish to glaucesent. Sheaths \-l open, rounded. Ligules 3-7 mm long, acute to attenuate, smooth. Blades as- cending, strict, flat or folded, to 2.5 mm wide. Panicles erect, narrow to ovate, 4-15 cm long, few-flowered, branches strict and steeply as- cending (or sharply divergent in anthesis). Axis internodes mostly 1-2 cm long, rarely over 3 cm long. Spikelets ovate, 2-3 (4) flow- ered, proportionally more green and less pur- ple than typical P. arctica. Glumes large, lanceolate, the second broadly so, with broad- hyaline-margins, smooth, weak keel glabrous or very sparsely scabrous near tip, second glume subequaling first lemma in length. Lemmas 3.5-6 mm long, lanceolate, acute, obscurely 5-nerved, densely villous on keel and marginal nerves, abundantly short villose between nerves, this occasionally nearly re- stricted to intermediate nerves, pubescence occasionally uniformly and abundantly short villose over nerves and whole base of lemma. Callus glabrous to sparsely webbed. Paleas glabrous, sparsely scabrous, or abundantly villous on keels, glabrous to puberulent be- tween them. Rachilla internodes glabrous, usually visible from side. Flowers perfect but anthers occasionally abortive. Anthers 1.4-2.5 mm long, yellow to purplish (aborted ones remaining yellow). Chromosome num- ber: 99. Habitat: Spruce-fir forest to alpine mead- ows and grasslands, mostly 2440-3800 m. Flowering July-August. Occurring from deep, rich soils to rocky places, in somewhat drier and warmer situations than the other subspecies. Distribution: Southern Rocky Mountains in s Utah, s and c Colorado. New Mexico B, RA, SD, SM,TO. Comment: This subspecies is a most dis- tinctive race of Poa arctica s. lat. It is markedly different on first sight in its more tufted and stricter habit, with fewer sterile shoots and proportionally more flowering shoots, more pale or glaucus foliage, and more southerly and often subalpine habitat. It may introgress with Poa sccunda, but it shows strong affinity to P. arctica and grades into that species. Poa arctica subsp. graijana (Vasey) Love, Love & Kapoor Poa arctica subsp. gratjana (Vasey) Love, Love & Ka- poor, Arctic & Alpine Res. .3:143. 1971; Poa graijana Vasey, Contr. U.S. Nat. Herb. 1:272, 1893. Type: Patterson 14, Grays Peak, Colorado, (US!). Poa alpicola Nash in Rydb., Mem. N.Y. Bot. Card. 1:272, 1893. Poa phoenicea Rydb., Bull. Torr. Bot. Club 32:60.5, 1905. Poa chiono- genes Gandog., Bull. Soc. Bot. France 66:302, 1920. Poa longipila Nash in Rydb., Mem. N.Y. Bot. Card. 1:46, 1900. Poa calhchroa Rydb., Bull. Torr. Bot. Club. 32:603, 1905. Poa tricolepis Rydb., ibid., 606, 1905. Arctic Bluegrass. Perennial, with well developed rhizomes. Culms usually decumbent, loosely tufted, smooth, 1-6 dm tall. Leaves green, firm. Sheaths closed \ to I the length. Ligules 2-3 (4) mm long, truncate to acute, mostly entire. Blades flat or folded, 1-3 mm wide, those of culms less than 8 cm long, often curved up- ward. Panicles erect to nodding, ovate, 3.. 5-15 cm long. Branches 2-3 (5) per node, slender, often drooping or contorted, smooth to sparsely scabrous, bearing few spikelets toward tip. Axis internodes mostly 1-2 cm long, rarely over 3 cm long. Spikelets ovate, loosely, 2-6-flowered, 4-8 mm long, strongly purplish. Glumes large, lanceolate, second broadly so, with broad, purplish, hyaline mar- gins, smooth, weak keel glabrous or very sparsely scabrous near tip, second glume subequaling first lemma in length. Lemmas (3) 4—6 mm long, 5-nerved, compressed- keeled, densely villous on keel and marginal nerves, mostly villous on intermediate nerves, mostly sparsely puberulent between nerves at base, mostly strongly piuplish, bronze colored near tip. Callus glabrous or scantly webbed [subsp. (graijana (s. str.)] to densely long-villous [P. longipila form], i^ileas glabrous, sparsely scabrous, or abun- dantly villous on keels, glabrous to puberu- lent between them. Rachilla internodes glabrous or villous. Flowers perfect but an- thers occasionally abortive. Anthers 1.4-2.5 mm long, yellow to purplish (aborted ones July 1985 SoRENC: New Mexico Poa 405 remaining yellow). Chromosome numbers 36-106, most frequently reported being 56 and 70. Habitat: Alpine. Occurring most fre- quently in cold, mesic sites, usually in peaty soils, 3100-3800 m in New Mexico. Flowering July-August. Distribution: The typical subspecies: Cir- cumboreal, high arctic. Subspecies gra/yonfl: Cascade Mountains, and Canadian Rocky Mountains south to n New Mexico. CF, MR, RA, TO. Poa longipila is the most common form of subsp. grayana in the n Rocky Moun- tains and is known in the U.S. from as far south as s Colorado, where it is less common than the subspecies grayana in the strict sense. Comment: Rocky Mountain phases of Poa arctica, based on the supposed larger stature of the plants, are often referred to P. grayana. However, population studies in Alaska and the Rocky Mountains do not support this dis- tinction. Subspecies grayana, as originally proposed, is a heterogenous group and in- cluded all U.S. material of the species. How- ever, many plants and populations from the Rocky Mountains match or lie between the more northern complex of the species. The complex requires a comprehensive study of population and herbarium material from its full geographic range. Poa arida Vasey Poa arida Vasey, Contr. U.S. Natl. Herb. 1:270. 1893. Type: Vasey, Socorro, New Mexico, in 1881 (US, GH!). Poa planifolia Scribn. & Will, in Scribn. USDA Div. Agrostol. Circ. 9:3. 1899, not Kuntze, 1898; Poa glaucifolia Scribn. & Will., ibid. 10:6. 1899. Type: Williams 2814, moist banks, Spring Creek, Big Horn Basin, Washakie Co., Wyo., 4 Aug. 1897 (US). Plains Bluegrass. Perennial with strong rhizomes. Culms round, 1.5-8 dm tall, not crowded, glabrous to scabrous, conspicuously striate. Leaves pale green, often glaucous, quite firm. Sheaths closed about I the length, mostly glabrous. Ligules 1-4 mm long, acute, entire or lacerate. Blades 1-5 (mostly 2) mm broad, flat to folded, uppermost leaf usually 1-6 cm long. Panicles 4-12 (18) cm long, narrow and compact or infrequently open, branches more or less scabrous on sharp angles. Axis in- ternodes 1-2 rarely over 3 cm long. Spikelets (2) 3-7-flowered, 4-7 mm long, compact, ovate or longer. Glumes glabrous or more often sparsely scabrous on upper portion of keel, ^-5 as long as subtended lemmas. Lem- mas 2.-5-4 mm long, obtuse to acute, promi- nently 5-nerved, often weakly keeled, densely pubescent on keel and marginal nerves and usually also on intermediate nerves, densely short-villous between nerves to glabrous. Callus glabrous. Rachilla usually pubescent. Paleas villous to long scabrous on keels. Flowers perfect. Anthers 1.3-1.7 (2 + ) mm long. Chromosome numbers: 63, 64, 76, 84, 90, 103, (P. glaucifolia form: 50, 56, 70, 81, 84, 86, ca 100). Habitat: In New Mexico, P. arida is the only native perennial bluegrass on the open plains and alkaline or saline flood plains, 1070-1980 m. Flowering May-July. Distribution: Western Great Plains Canada south to New Mexico; B, CV, G, MR, SC, SM, U, V. The several Utah records of this species 1 have seen are afl alpine and are referable to Poa arctica subsp. aperta. The Arizona records of this species have been based on rhizomatous specimens of Poa fendleriana. Comment: Poa glaucifolia appears to be a shade or mesic form having more lush foliage, smaller, more numerous, less pubescent spikelets, and larger, more open panicles. Poa arida shows some affinity to the P. secunda complex and appears to hybridize with it where ranges overlap. This species may have evolved from Pleistocene hybridization be- tween species of section Poa and the "Secundae" group. Poa bigelovii Vasey & Scribn. Poa bigelovii Vasey & Scribn., in Grasses U.S. Descr. Cat. 81. 1885, nomen nudum, Vasey & Scribn. in Vasey, Contr. U.S. Natl. Herb. 1:270. 1893. TYPE: Fendler 931, New Mexico [probably along Santa Fe Cr. above Santa Fe], in 1847. Poa annua var. stricta Vasey, Bull. Torrey Bot. Club 10:31. 1883. Bigelow Bluegrass. Erect, loosely tufted annuals (rarely bi- ennial). Culms sometimes geniculate at base, leafy, 1-6 dm tall. Leaves bright green, mostly cauline. Sheaths sharply keeled, closed ca^-l the length. Ligules 1-6 mm long, acute, entire, glabrous to scabrous on back. Blades 1.5-5 mm broad, flat or folded, promi- 406 Great Basin Naturalist Vol. 45, No. 3 nently keeled and prow tipped. Panicles very narrow, elongate, internodes frequently over 4 cm long. Branches appressed, often flow- ered from base, scabrous. Spikelets crowded on branches, compact, 4—7 mm long, ovate, strongly compressed. Glumes narrow, tend- ing to curve inward, first subequal to second, scabrous on 1-3 nerves, with narrow hyaline margins. Lemmas broadly lanceolate, acute to blunt, with whitish hyaline margins, 3—5 mm long, 5-nerved, intermediate usually distinct, villous on keel and marginal nerves and some- times on midnerves, finely papillose-rough- ened glabrous or puberulent between nerves on lower |. Callus with tuft of long hairs. Paleas mostly sparsely short villous on keels (at least near middle) and scabrous above, papillose between nerves. Rachilla in- ternodes glabrous, slender, short, and hidden from side view. Flowers perfect, mostly cleis- togamous. Anthers 0.2-1 mm long. Chromo- some numbers: 28, 28+1. Habitat: Frequent to locally abundant in warm deserts to ponderosa pine zone, spo- radic northward and upward. In shade of rocks and shrubs and in arroyo bottoms 1070-2900 m in New Mexico. Flowering March- May. Distribution: Southwestern U.S., Okla- homa, s Colorado, s Utah, s Nevada, to Cali- fornia, Arizona, south to Texas, Baja Califor- nia, and c Mexico. New Mexico: CB, DA, ED, CT, HD, LU, LN, MK, MR, OT, RA, SC, SD, SF, SM, TO, TR. Comment: Although considered by most authors to be close to Poa annua, P. bigelovii is more closely related to P. occidentalis. Poa bulbosa L. Poa bulbosa L., Sp. Pi. 70, 1753. Bulbous Bluegrass. Perennial, densely tufted. Culms 0..5-6 dm tall, somewhat bulbous at base. Leaves green. Sheaths round, usually smooth, closed only at base, lower ones often reddish, basal ones fibrous. Ligules 2-3.5 mm long, glabrous, ob- tuse, entire or lacerate. Blades near base of plant soft and filiform, those of culm 1-2 mm wide. Panicles lax, broadly lanceolate, up to 10 cm long. Spikelets usually with few "normal" florets beneath terminal, mostly bulbiferous ("viviparous") florets, com- pressed. Glumes mostly normal, sparsely scabrous on keel. Lemmas of normal florets strongly keeled, glabrous to sparsely villous on keel and marginal nerves, minutely papil- lose-roughened between nerves, with or without a sparse web on callus. Paleas sparsely scabrous on keels. Flowers at base of spikelet usually perfect but often incom- pletely developed, the upper producing bul- bous vegetative offsets. Anthers 1.2-1.5 mm long, questionablv functional. Chromosome numbers: 14, 21, 28, 39, 40, 42, 45, 56, 58. Habitat: The few collections to date (1936 and 1938) from New Mexico have been from grassland communities. It is likely to have spread into ponderosa pine zone of northern New Mexico as well, 18.30-1980 m. Flowering April-September. DISTRIBUTION: Introduced from Europe, widespread in w North America. New Mex- ico: MK, RA, SJ. Poa compressa L. Poa compressa L., Sp. Pi. 69. 1753. Canada Bluegrass. Perennial, strongly rhizomatous, fi-e- quently sod forming (reportedly cespitose in one strain, but this not known from New Mex- ico). Culms slender to very stout, flattened, often geniculate at the nodes, nodes strongly flattened, 20-60 cm tall, lower internodes usually shorter than sheaths. Leaves green, firm. Sheaths flattened, closed from only near base up to \ the length. Ligules 1-3 mm long, mostly obtuse, jagged, ciliate fringed on mar- gin, mostly glabrous on back. Blades of culm mostly regularly divergent, 1.5-^ mm broad, 2-10 cm long flat or folded, upper-one fixed well above middle of culm. Panicles mostly erect, contracted, 2-10 (15) cm long. Branches short, mostly less than 3.5 cm long, mostly steeply ascending (to widely spreading in anthesis), strongly scabrous on prominent angles. Axis internodes mostly less than 3 cm long. Spikelets strongly compressed, 2.5-8 mm long. Glumes mostly l-^ as long as adja- cent lemmas, mostly 3-nerved, scabrous on keel in upper part. Lemmas broadh-acute to blunt, hyaline-tipped, strongly 5-nerved, vil- lous on keel and marginal nerve, glabrous between nerves and distinctly finely papil- lose-roughened. Callus mostly with short web, sometimes glabrous. Paleas scabrous on keels, glabrous between. Rachillas short. July 1985 SoRENG: New Mi-Aico Po.^ 407 glabrous. Flowers perfect. Anthers 1.3-1.8 mm long. Chromosome numbers: 14 , 35, 39, 42,45,49-50,56,84. Habitat: Open ground, roadsides, dis- turbed meadows, often in riparian areas, in forested habitats, 1520-3050 m in New Mex- ico. Flowering June-August. Distribution: Introduced from Europe, widespread in North America. New Mexico c:f, ct, hr, ln, mk, ot, ra, sd, sf, sm, TO, TR. Comment: Frequently confused with Poa pratensis, but the latter has more closed, weakly keeled sheaths and terete culms. Poafendleriana (Steud.) Vasey Poafendleriana subsp. fendleriana Poa fendleriana (Steud.) Vasey subsp. fendleriana. USDA Div. Bot. Bull. 1.3(2):pl. 74. 1893; Era- grostis fendleriana Steud., Syn. Pi. Glum. 1:278. 18.54. LECTQ-nPE (Marsh): Fendlcr 932 "Mexico" [New Mexico, probably in Santa Fe Canyon above Santa Fe.], in 1847 (NY!, plant no. 1 of that sheet); Iso-nPES: NY!, GH!, GH!. Fendler Muttongrass. Perennial, tufted, usually producing short rhizomes (these infrequently collected). Culms mostly slender to stout, (15) 25-45 (60) cm tall (male), (15) 30-60 (80) cm tall (female). Leaves green, infrequently bluish, quite firm, mostly basal, basal tuft 2-40 cm tall. Sheaths closed ca \ the length, strongly striate, more or less coarsely spiculate about collar near sheath margins, mostly puberulent or finely scaberu- lous (if glabrous or sparsely scabrous, then lemmas quite villous on keel nerve). Ligules from merely spiculate ridge to 1 (2) mm long, truncate to rounded, spiculate fringed on up- per margin and abundantly scabrous on back. Blades (0.5) 0.8-2 (3) mm' wide, thick, firm, mostly scabrous to puberulent ventrally, flat or folded, occasionally with margins inrolled, within a plant upper blade of at least some culms 0-3 mm long, others occasionally ex- ceeding 1 (4.5) cm in length. Panicles nar- rowly oblong, compact (more open in flower), branches scabrous, often flowered from base, spikelets loosely arranged on short pedicels, sexually dimorphic, 2-6 (9) cm long in males, (3) 4-8 (12) cm long in females. Axis in- ternodes mostly less than 2.2 (3) cm long. Spikelets ovate, somewhat plump, 3-10 mm long, 2-7-flowered, mostly shiny, and pale green, sometimes turning purplish, ca 1.5-60 per panicle. Glumes lanceolate to broadly .so, 1-3-nerved, mostly glabrous and shining. Lemmas 3-6 mm long, smooth to papillose- roughened, 5-nerved, intermediate nerve of- ten obscure, prominently long-villous on keel and marginal nerves and glabrous between them (1% of New Mexico specimens sparsely villous on intermediate nerves as well). Callus glabrous or with hairs no longer than those on rachilla internodes. Paleas roughened, scabrous to villous on keels and between them. Rachillas less than 1.3 mm long, glabrous to scabrous, infreciuently pubescent. Flowers unisexual, (dioecious). Anthers 1.5-3 mm long. Chromosome number: 56. Habitat: Rocky slopes and meadows, up- per desert-grasslands, interior chaparral to subalpine grasslands, frequent in ponderosa pine-gambel oak associations, 1220-3350 m. Flowering February-June (October). One of most abundant early spring flowering species in New Mexico. Distribution: Southern Montana south to n Mexico, w Utah, Arizona, Wyoming, South Dakota, Colorado, Oklahoma, Texas, Coahuila, Chihuahua, New Mexico DA, B, CB, CF, CT, GR, HD, LA, LN, LU, MK, MR, OT, RA, SC, SD, SF, SM, SR, TO, TR, U. Comment: This subspecies appears to be restricted to areas under the influence of sum- mer monsoons. As averaged over the range, approximately 15% of the specimens are male plants, these originating predominently from New Mexico, Arizona, and Texas. Intermedi- ate sexual populations between this and subsp. longiligida occur where the sexual races of each subspecies are geographically in contact. Many of the plants from outside the se.xual zones exhibit intermediate characteris- tics as well, but most tend strongly toward one or the other sexual race in morphology and ecology. Poafendleriana subsp. albescens (Hitchc), comb. nov. Poa fendleriana subsp. albescens {Hitchc), comb, nov.; Poa albescens Hitchc, Contr. U.S. Nat. Herb. 17(3):375. 1913. HoLQ-nPE: Rose 11648, .Mexico, Chihuahua, Minaca, 1 Apr 1908, (US 454361!). Poa griffithsii Hitchc, Contr. U.S. Nat. Herb. 17(3):375. 1913. Holotype: Griffiths 4865, Mex- 408 Great Basin Naturalist Vol. 45, No. 3 ico, Sonora, Cananea, 7-8 Jul 1907, (US 691228! (an aberrant specimen, tending to be bulbifer- ous)). Mexican Muttongrass. Similar to Poa fendleriana subsp. fendleri- ana. Culms slender to stout, 20-60 cm tall. Leaves mostly basal, green. Sheaths glabrous to rarely strongly scabrous, collars spiculate near leaf margins. Ligules from nearly absent to 1.8 mm long, truncate to rounded, spicu- late-fringed on margin, scabrous on back. Blades flat or folded, sometimes inrolled on margins. 1-2.5 (3.5) mm wide, uppermost blade 0-6 (12) mm long. Panicles (3) 4-12 (20) cm long, branches scabrous. Axis internodes 1-3 cm long. Lemmas glabrous to sparsely puberulous on keel, sometimes sparsely scabrous on upper keel, marginal nerves usu- ally obscurely and regularly short ciliate, glabrous between nerves. Paleas glabrous be- tween keels. Rachillas glabrous. Flowers uni- sexual (dioecious; primarily sexual but asexual populations known). Anthers 1.5-3 mm long. Chromosome number: 28+1. Habitat: Pine-oak to spruce forests and subalpine grasslands, in rocky, organic soils and steep meadows, 1680-3350 m. Distribution: Sierra Madre Occidental Mexico in w and Chihuahua, ne Sonora, in the U.S. in the mountains of se Arizona and sw New Mexico: HD. Comment: Poa albescens and P. griffithsii, each described from single specimens, differ only in minor ways. Since their publication, an array of specimens have been collected that are intermediate and serve to bridge the main differences between them. The lemmas of plants of P. f. subsp. albescens are mostly glabrous or very sparsely pubescent, and the sheaths are mostly glabrous. Other than in the characters mentioned above, subspecies albescens stands only slightly moq3hologi- cally removed from the typical subspecies, yet it is cytologically distinct. The proposed sub- species is restricted to the Sierra Madre Occi- dental. This taxon intergrades to subsp. fend- leriana where the two occur together, but very few intermediate staminate specimens have been discovered. Poa fendleriana subsp. longiligula (Scribn. &Will.), comb. nov. Poa fendleriana subsp. longiligula (Scribn. & Will.), comb, nov.; Poa longiligula Scribn. & Will, USDA Div. Agrost. Circ. 9:3, 1899. HOLOTYPE: Jones 5149, Silver Reef, Washington Co., Utah, 3 May 1894 (US 28539100!. IscnPES: MO!, OSC!); Poa fendleriana var. longiligida (Scribn. & Will.) Gould, Madrono 10:94, 1949. Longtongue Muttongrass. Perennial, tufted, sometimes stooling, but less evidently rhizomatous, and generally more robust than typical subspecies. Leaves firmer, pale green to somewhat bluish. Sheaths glabrous to scabrous or infrequently puberulent, but hairs not noticeably coarser and congested around collar margins. Ligules obtuse to acuminate, entire or only very faintly scabrous on margin, glabrous to sparsely scabrous on back, (1.5) 2-18 mm long. Blades somewhat broader than in typical subsp., (0.8) 1..5-2.2 (4) mm wide, flat or folded, rarely inrolled on margins, glabrous, scabrous, or puberulent ventrally, upper culm blade of some culms reduced to 4 mm long or less, longest ones to 2 (7.8) cm long. Panicles slightly dimorphic, (2) 3-8 (10) cm (males), (2) .5-12 (30) cm long (females). Axis internodes to over 4 cm long. Spikelets 2-13- flowered, 4—12 mm long, pale green, shiny. Lemmas 5-nerved, intermediate nerves ob- scure to prominent, prominently long villous on keel and marginal nerves, glabrous or sparsely villous on intermediate nerves, glabrous or occasionally puberulent between nerves. Rachillas hirtellous to puberulous. Flowers unisexual, (dioecious, infrequently perfect in part). Anthers 1.5-3.8 mm long. Chromosome numbers: 56. Habitat: From interior chaparral and pinyon-juniper to subalpine meadows, often in somewhat dryer situations than typical subsp., 1830-2440 m in New Mexico, 910-3510 m elsewhere. Flowering March- July. Distribution: Mexico in Buja California, California, Oregon, Idaho, Nevada, Oregon (almost exclusive of subsp. fendleriana), sw Canada in British Columbia, Arizona, Colo- rado, Montana, South Dakota, Utah, Wyo- ming (where occasionally intermediate to suhsp. fendleriana). Northwest New Mexico: B, CB, MK, SC, SD, SJ, TO. C(JMMENT: This subspecies is predomi- nately distributed west of the summer mon- soon region. Agamospermy is the predomi- nant mode of reproduction in this subspecies, but staminate plants are frequent in Arizona, July 1985 SoRENG: New Mexico Poa 409 Utah, and s California and Nevada. Only one staminate plant has been collected in New Mexico. In contrast to subsp. fendleriana, about half the plants of" subsp. longiligula ex- hibit pubescent lemma intermediate nerves. None of the New Mexico material of the spe- cies has pubescence between the lemma nerves, but to the north and west some plants of subsp. longiligula are hairy in this region. Poa glaiica Vahl. Poa glauca subsp. glauca Poa glauca Vahl subsp. glauca, Fl. Dan. fasc. 17:3. Pi. 964. 1790. Type: Norway. Glaucous Bluegrass (Timberline Bluegrass). Perennial, cespitose, most shoots flower- ing. Culms mostly 1-3+ dm tall, strict, stout, wiry, mostly ascending, glabrous or rarely sparsely scabrous about the nodes. Leaves glaucous, few on culm. Sheaths rounded or slightly keeled, closed from near base up to \ the length. Ligules 0.5-3+ mm long, short- truncate to obtuse, upper edge entire to erose, often minutely fringed, sparsely to strongly scabrous on back. Blades 1-2.5 mm broad, firm, upper one located below middle of culm, 1-5 (6) cm long. Panicles 1.5-7 (12), open or closed, ovate to lanceolate, branches strict and steeply ascending, to 3+ cm long, M mostly strongly scabrous on angles, and with a few flowers near tips. Axis internodes rarely over 2 cm long (in Rocky Mountains). Spikelets compressed, ovate, glaucous, often somewhat purplish, 2-3 (5)-flowered, 2-5 (7) mm long. Glumes mostly 3-nerved, subequal, mostly sparsely scabrous on nerves and fre- quently between them, the second mostly with margin distinctly angled or rounded near middle, ^-l as long as first lemma. Lemmas 2-4 mm long, distinctly 5-nerved, papillose- roughened to nearly smooth, glabrous to sparsely puberulent between nerves, villous on keel and marginal nerves and frequently also on intermediate ones, hyaline margin narrow, obtuse, and frequently blunt at tip. Callus with distinct tuft of hairs up to | as long as lemma. Paleas scabrous or infrequently glabrous on keels, papillose, scabrous, or pu- berulent between keels, usually short spicu- late between keels. Rachillas glabrous, scabrous, or puberulent. Flowers all perfect but frequently anthers infertile. Anthers 1.4-2 mm long. Chromosome numbers: 42, 44, 56-58, 62-63, 70-72, 75, 78. Habitat: Subalpine to alpine, meadows, rocks, exposed ridges, wind balds, from open ground to among densely thatched cushion plants, 3350-3960 m in New Mexico. Flower- ing July-August. Distribution: Circumpolar boreal. Poa glauca subsp. glauca is found in most of tiaga- tundra, subalpine, and alpine regions of Canada but is infrequent southward in U.S. Rocky Mountains, occurring south to .south central New Mexico: LN, OT, RA, SF, TO. Comment: Poa glauca subsp. glauca is less frequent than subsp. rupicola in the U.S. Rocky Mountains. The latter may be distin- guished by its smaller, more slender, and erect culms, greener foliage, and the pres- ence of pubescence between the lemma nerves, and callus with a minute web or glabrous. The transition between the forms is highly complex. In population samples of "P. rupicola' from the Rocky Mountains, it is common to find the extremes of P. glauca and P. interior, and all intermediate combinations of habit, coloration, and spikelet pubescence. However, subsp. ritpicola occasionally forms monomorphic populations, and in some re- gions of the west, such as in the Sierra Ne- vada, this is the only form present. In trans- plant studies with Rocky Mountain alpine material, identified in the field as "P. rupi- cola," glaucousness or greenness, spikelet pubescence, shape of spikelet parts, and gen- eral leafiness were stable characters. Stature and panicle dimensions tended to be more plastic. Western U.S. Poa glauca (s. lat. ) is part of an extremely variable agamic com.plex. It of- ten occurs with, and shows forms intermedi- ate to, P. secunda, but in most cases may be distinguished as keyed. Poa glauca subsp. rupicola (Nash) W. A. Weber Poa g/fluca subsp. nipico/a (Nash) W. A. Weber, Phytolo- gia 51:375. 1982; Poa rupicola Nash, Bull. Torrey Bot. Club 14:94. 1887. Lectotype (A. Hitchc): Wolf 341, South Park, Colorado, in 1873 (NY!); Poa rupestris Vasey, not Roth 1817, not Bieb 1831, not With. 1796; Poa glauca var. rupicola B. Boivin, Naturahste Canad. 94:527. 1967. Timber- line Bluegrass. 410 Great Basin Naturalist Vol. 45, No. 3 Similar to subspecies glatica, but varying as follows: Culms more slender and erect, mostly 0.5-1.5 (2.5) dm tall. Leaves green to somewhat glaucous, few on culm. Blades 1-2 mm wide, firm, upper-one usually located be- low middle of culm and 1-3 (4) cm long. Pani- cles 1.5-5 cm long, open or closed, ovate to lanceolate, branches more slender. Axis in- ternodes to 2 cm long. Spikelets more com- pact, usually somewhat glaucous, often red- dish or purplish, 2-3 (4) flowered, 2-5 mm long. Glumes mostly sparsely scabrous on nerves and frequently between them, both broad, mostly 3-nerved, subequal, the second mostly with margin distinctly angled or rounded near middle. Lemmas 2-3.5 mm long, distinctly papillose-roughened, villous on keel and marginal nerves and often on intermediate nerves, usually puberulent be- tween nerves, acute to obtuse and sharp or blunt at tip. Callus with tiny tuft of crinkled hairs or glabrous. Anthers 1.1-1.6 mm long. Chromosome numbers: 48-50, 54-56. Habitat: Subalpine to alpine, meadows, rocks, exposed ridges, wind balds, from open ground to among densely thatched cushion- plants, 3350-3960 m in New Mexico. Flower- ing July-August. Distribution: Alpine regions of Western North America. Generally replacing typical subspecies throughout western U.S. Canada in Alberta and British Columbia, south to alpine of California, northern Arizona, and south central New Mexico: LN, OT, RA, SF, TO. Comment: This race occurs to the exclusion of subsp. glauca and P. interior in the Pacific Cordillera. However, see comments under those taxa. Poa interior Rydb. Poa interior Rydb., Bull. Torrey Bot. Club 32:604. 1905. Type: Tweedy 3706, Headwaters of Clear Cr. and the Crazy Woman Rv. Wyoming, in 1900 (NY!); Poa nenioralis var. interior (Rydb.) Butt. & Abbe; Poa nemoralis subsp. interior (Rydb.) W. A. We- ber, Phytologia51:,37.5. 1982. Inland Bluegrass. Perennial, cespitose, most shoots fertile. Culms 0.5-5 dm tall, densely tufted at base, slender, wiry, frequently geniculate at lower nodes, more or less glabrous about nodes. Leaves green, sometimes reddish, usually several on the culm. Sheaths as in P. glauca. Ligules truncate to rounded, to 2 mm long, mostly scabrous on back, upper margin fre- quently jagged, mostly minutely fringed. Blades 1-2.5 mm wide, strictly ascending to divergent at about 60°, sometimes lax, upper one (3) 6-15 cm long. Panicles strict or some- what lax, 2-15 cm long, the 1-3 (4) branches prominently scabrous on distinct angles, slen- der, ascending, to 6 cm long. Axis internodes to 4 cm long. Spikelets 2-3-flowered, 2.5-5 mm long, compressed, ovate, mostly bright green, sometimes purplish. Glumes narrow to broad, mostly 3-nerved, scabrous on keel, tips sharply acute, usually abruptly curved in or outward at apex, second glume mostly with distinct to rounded angle near middle of mar- gin. Lemmas 5-nerved (intermediate nerves frequently obscure), prominently villous on keel (to near tip) and marginal nerves, only occasionally villous on intermediate nerves (then sparsely so), glabrous and papillos- roughened between nerves. Callus web minute, or to about I lemma in length. Paleas scabrous to occasionally glabrous on keels. Rachillas as in P. glauca. Flowers perfect. Anthers 1.1-1.8 mm long. Chromosome numbers: 28, 42, 43, 56. Habitat: In Douglas-fir and spruce-fir forests in moist meadows, on mossy ledges, to alpine meadows and wind balds, 2740-3660 m in New Mexico. Flowering July-August. Distribution: Interior w North America extending to the Great Lakes region, north to Alaska, south to Arizona (reported from, but doubtfully in, Te.xas) and New Mexico: B, CB, CF, RA, SD, SM,TO. Comment: Frequently intergradient with and difficult to distinguish in forest and meadow habitats from Poa palustris, to thor- oughly intergradient with populations of Poa glauca Vahl at higher elevations. A close rela- tionship to the introduced P. nemoralis L. is evident, but the forms of P. interior in the western United States more closely approach P. glauca. A thorough, worldwide study of population samples from this facultatively ag- amospermous complex is sorely needed. Poa leptocoma Trin. Poa leptocojua Tr\u. , Mem. Acad. St. Petersb. VI, 1:374. 18-30. HoLQ-nPE: D. Mertens Sitka, Alaska (LE!). Bog Bluegrass. July 1985 SoRENG: New Mexico Poa 411 Perennial, loosely tufted or somewhat rhi- zomatous in mossy habitats. Culms 2-10 dm tall, smooth or scabrous. Leaves bright green. Sheaths closed \-l the length, smooth or scabrous. Ligules 1.5-4 mm long, truncate to obtuse, entire to jagged, glabrous on back. Blades flat or folded, mostly lax, 1-4 mm wide, upper (5) mostly 8-15 cm long (in U.S. Rocky Mountains). Panicles nodding, lax, open, the 1-3 (mostly 2) branches per node capillary, scabrous, spreading to occasionally strongly reflexed, few- to many-flowered in distal 5. Axis internodes mostly less than 4 cm long. Spikelets green to strongly purplish, strongly compressed, narrowly ovate to oblong, loosely 2-5-flowered, 4-8 mm long. Glumes sharply acute, mostly scabrous on nerves, unequal, the first shortest, very nar- row, 1-nerved, the second lanceolate, 1-3- nerved. Lemmas 5-nerved, intermediate nerves mostly obscure, very smooth and glabrous between nerves, densely to sparsely villous on lower | of keel and marginal nerves (to nearly glabrous), hyaline margin promi- nently bronze colored near tip (unlike P. occi- dentalis). Callus with a sparse, long web. Paleas keels with regularly spaced, slender, antrorsely curved scabers, or nearly glabrous, with only long cells between the nerves. Rachillas elongate, usually visible from side- view at maturity. Flowers perfect, mostly cleistogamous. Anthers 0.25-1 (1.1) mm long. Chromosome numbers: 42 . Habitat: Subalpine and alpine, in wet meadows, along stream banks, springs, and bogs, 2200-3350 m in New Mexico. Flower- ing June-August, generally slightly later than the very similar, and frequently sympatric, P. reflexa. Distribution: Northeast Asia, boreal w North America to the Canadian Rocky Moun- tains (P. paucispicula Scribn. & Merr.), se Alaska south and east through high montane and alpine regions of w North America to s New Mexico (P. leptocoma). New Mexico: CB, CF, LN, OT, SF, TO. Comment: This species and Poa reflexa of- ten occur together in mixed stands, and the species are sometimes considered as vari- eties. However, the species have different chromosome numbers, and intermediate plants are rare. The morphological distinc- tions, once recognized, make them easy to distinguish and indicate that they may be only distantly related (see subgeneric placement above). Poa nervosa (Hook.) Vasey Poa nervosa (Hook.) Vasey, Bull. U.S. Depr. Agr. Div. Hot. 132:pl. 81. 1893; Festuca nervcsa Hook. Fl. Bor. Am. 2:251, pi. 232. 1840. Tyfe: Scouler, Nootka Sound, Vancouver Lsl, Briti.sh Columbia (NY!, GH!, US!). Poa nervosa var. wheeleri (Vasey) C. L. Hitchc. Poa nervosa var. wheeleri (Vasey) C. L. Hitchc. Va.scl. Pi. Pac. Northwest. 1:671. 1969; Poa wheeleri Vasey inRothr. Cat. Pi. Surv. W. 100th Merid. 55. 1874. Type: Wolf 1131, South Park, Colorado (US!). Poa carta Rydb. (non auct. ) Bull. Torr. Bot. Club 36:534. 1901. Type: Tweedy 13, Spread Creek, Teton Co., Wyoming (NY!). Perennial, loosely tufted, with prominent short rhizomes. Culms erect, often decum- bent at base, terete, 20-85 cm tall. Leaves light green. Sheaths prominently striate, round to weakly keeled, closed (5) | to nearly entire length, at least those of lower culm usually finely retrorsely scabrous-puberulent, infrequently glabrous throughout. Ligules 1-3 (4) mm long, truncate to acute (above), usually densely spiculate on back (especially below). Blades flat to folded, weakly keeled, soft, steeply ascending, glabrous to sparsely spiculate above, glabrous below, those of culm 1.5-3 mm broad, and to 10 cm long, upper culm one 1-6 cm long, those of innova- tions to 30 cm long. Panicles erect, nodding at tip, (3.5) .5-13 (18) cm long, narrowly ovoid, sparsely flowered, long-peduncled. Branches spreading to ascending, scabrous on angles, (1) 2-4 per node, with (1) 3-8 (11) spikelets in distal i Axial internodes mostly 1-3.5 cm long. Spikelets compressed, 5-11 mm long, 2-8-flowered, light green to somewhat pur- phsh. Glumes about 5 the length of subtended lemma, acute, keels distinct, sparsely scabrous above. Lemmas, keeled, 4—6 mm long, broadly acute, papillose-roughened to sparsely spiculate between nerves, frequently scabrous above, 5-nerved, often scabrous on nerves, occasionally puberulent on keel and marginal nerves. Callus glabrous. Paleas moderately to sparsely scabrous along keels, sparsely spiculate between keels. Rachiila in- 412 Great Basin Naturalist Vol. 45, No. 3 ternodes usually spiculate. Flowers pistillate or infrequently appearing hermaphroditic (in var. wheeleri), (dioecious or gynodioecious and sexually reproducing in the typical vari- ety, reproduction by agamospermy in var. wheeleri). Anthers to 3 mm long, usually ves- tigial in var. wheeleri. Chromosome numbers: 28, 29, 56, 61-64, 70, 74, 91. Habitat: Open mountain slopes from up- per sagebrush and lower pine belts to lower alpine (New Mexico subalpine), in rich soils and duff. The one New Mexico collection from 3500 m. Flowering (May) June-August. Distribution: Southwestern Canada, Pacific Northwest, south to California, and Nevada. In Rocky Mountains south to north central New Mexico: TO. Comment: Poa nervosa (sensu stricto), ap- plies to the sexually reproducing race found west of the Cascade Mountains in the Pacific Northwest. It is distinguished by the pres- ence of long hairs on the collar margins. Vari- ety wheeleri is one of the most common native Poas of the northern Rocky Mountains and, although highly variable in form, is distin- guishable by its puberulent lower sheaths and hgules, relatively uniformly developed upper culm leaves, and pistillate florets. However, the one New Mexico collection (Gold Hill) has glabrous sheaths, only sparsely puberulent ligules, and pubescent lemmas. The complex needs further evaluation. Poa occidentalis Vasey Poa occidentalis Vasey, Contr. U.S. Natl. Herb. 1:274. 1893. HOLOTYPE: Vaseij, Las Vegas, New Mexico, in 1881 (US 28537500!); Poa platyphylla Rydb. New Mexico Bluegrass. Perennial, loosely tufted, probably short- lived. Culms stout, erect or decumbent, red- dish at base, erect, 2-11 dm tall. Leaves mostly cauline, commonly blue-green. Sheaths longer than internodes, closed \-l the length, strongly keeled, glabrous or mostly abundantly retrorsely scabrous, lower ones often reddish. Ligules acute to acuminate, entire, scabrous on back, those of upper culm leaves to 12 mm long, mostly longer than leaf is wide. Blades strongly keeled, tip distinctly prow shaped, upper 4-18 cm long, 1.2-5.5 (10) mm wide. Panicles open, pyramidal, elongate, (6) 12^0 cm long, with loosely as- cending branches when young, erect and with widely spreading branches at maturity, tips drooping. Branches 5-23 cm long, densely flowered in distal I, moderately to sparsely scabrous. Axis internodes mostly over 4 cm long. Spikelets oblong, strongly compressed, 3-8 mm long, usually longer than pedicel, 2-7-flowered. Glumes subequal, sharply acute, the first narrowly lanceolate, 1-nerved, the second broader, 1-3-nerved, nearly equaling first lemma in length. Lemmas usu- ally green with whitish hyline margin and tip, sometimes purplish but rarely bronze colored at tip (unlike P. leptocoma ), 2.6-4.2 mm long, usually distinctly 5-nerved, keel scarcely in- curved at tip, villous on lower | of keel and lower I of marginal nerves, glabrous to sparsely puberulent between them. Callus with sparse long web. Paleas glabrous to finely scabrous on keels, with long and short cells between keels. Rachilla internodes glabrous, usually hidden from side view. Flowers per- fect, mostly cleistogamous. Anthers 0.3-1.0 mm long. Chromosome numbers: 14, ca28. Habitat: Montane. Usually in mesic sit- uations, on cool exposures, in sparsely vegetated, disturbed, and natural forest open- ings, 2300-3500 m. Flowering mid-July- September. Distribution: Southwestern Colorado, White Mountains of e Arizona, to the Guadalupe Mountains of Texas. New Mexico: B, CB, CF, GR, LN, OT, RA, SD, SF, SM, SR, TO. Past, more northerly reports of this species (including Utah and Wyoming) have been in error. Poa occidentalis, sensu Har- rington, 1954, is mostly P. tracyi. Poa pahistris L. Poa pahistris L., Syst. Nat. lOth ed. 2:874. 1759. Fowl Bluegrass. Perennial, most shoots fertile. Culms loosely tufted, often decumbent at base and rooting at nodes, stoloniferous, stout, more or less scabrous below nodes, loosely tufted, 2-12 dm tall, often branching above base. Leaves green, often turning reddish, mostly cauline and numerous. Sheaths keeled, usu- ally closed less than \ the length, mostly glabrous. Ligules 2-6 mm long, rounded to acute, mosth- fringed on margin, entire or jagged, often scabrous on back. Blades 1-4 mm wide, flat or folded, those of culm steeply July 1985 SoRENG: New Mexico Poa 413 ascending, longer ones drooping, uppermost one fixed well above middle of culm, 5-20 cm long. Panicles open, (narrow and congested when young), elongate, broad, densely flow- ered, mostly 10-30+ cm long. Branches mostly 4 or more per node, usually 2 or 3 times rebranched, capillary, and spreading, scabrous on the angles. Axis internodes mostly over 3 cm, to 6 cm long. Spikelets compressed, smallish, 1-6 (mostly 2-4) flow- ered, 2-4 mm long. Glumes subequal, about ^ as long as the subtended lemma, narrow, broadest near middle, scabrous on keel, sec- ond glume margin mostly gradually rounded. Lemmas 2-2.7 mm long, 5-nerved (interme- diate nerves mostly obscure) villous on keel and marginal nerves, glabrous and finely pa- pillose-roughened between nerves, tip dis- tinctly incurved, broadly acute, obtuse, or blunt, prominently bronze colored, hyaline margin very narrow. Callus with sparse but prominent web. Paleas scabrous to minutely ciliate on keels. Rachillas glabrous or scabrous. Flowers perfect. Anthers 0.8-1.4 mm long. Chromosome numbers: 28, 30, 32, 42. Habitat: Montane. Mixed-conifer forests from riparian, where often in shallow water, to moist meadows and open ground, ca 2440-3050 m in New Mexico. Flowering June-August. Distribution: Introduced from Europe, common to the north. New Mexico CF, GR, RA, SM, TO. Comment: Poa palustris appears to inter- grade in form and habitat with P. interior, and frequently entire populations cannot be satis- factorily assigned to one or the other taxon. Poa pratensis L. Poa pratensis L., Sp. Pi. 67. 1753. Poa agassizenssis Boivin & D. Love, Bernard, Svensk Bot. Tidsker, 53:371. 1959, nomen nudum; Poa agassizenssis Boivin & D. Love, Naturaliste Canad. 87:176. fig. 1. 1960. Kentucky Bluegrass. Perennial with extensive creeping rhi- zomes, often forming dense sods. Culms 0.5-80 cm tall, only slightly compressed. Leaves green to glaucescent. Sheaths promi- nently nerved, closed |-| their length, glabrous or occasionally sparsely long villous. Ligules 0.5-3 mm long, mostly truncate, cili- ate on upper margin, glabrous to minutely scabrous on back. Blades flat or folded, soft or occasionally rather firm, strict, 1.5-2.5 (5) mm broad, prominently keeled and prow tipped, those of upper culm 1-10 cm long and only slightly divergent. Panicles (1.5) .5-10 (16) cm long, pyramidal to elliptic, mostly erect with (2) 4 or more branches at lower nodes, branches sparsely scabrous. Axial internodes mostly 1-2 cm long, rarely over 3 cm long. Spikelets compact-ovate, crowded on branches, strongly compressed, green or pur- plish, 2-7 mm long, 2-7 flowered. Glumes 1-3-nerved, strongly keeled, more or less scabrous on upper l of keel, the first narrow, often somewhat curved inward, both sharply acute and about f-j the length of subtended lemma. Lemmas 2-4 mm long, 5-nerved, in- termediate nerves usually faint, glabrous and minutely papillose-roughened between nerves, keel and marginal nerves densely vil- lous, rarely with few hairs on intermediate nerves. Callus with copious tuft of kinky hairs from I to as long as lemma. Paleas glabrous, or closely and finely scabrous on keels, glabrous between keels. Rachilla internodes glabrous, short. Flowers perfect. Anthers 1.2-2 mm long. Chromosome numbers: 25-127 (with nearly every intervening number recorded). Habitat: An aggressive and variable weedy species composed of numerous apomictic races worldwide, but which are poorly under- stood in our region. Widespread from pon- derosa pine zone to subalpine, where it often forms a turfalong water courses. Occasional in lower habitats along perennial water courses, in greasewood communities, and on north- eastern plains of New Mexico. Commonly planted in lawns and pasture mixes. Impor- tant as a soil-stabilizer and as a forage. Natu- ralized from ca 1700-3500 m. Flowering May-September. Distribution: Holarctic, now occurring worldwide except for low elevations in the tropics. Introduced or possibly native in tem- perate latitudes of North America. First col- lected in New Mexico in 1887 by S. M. Tracy, near Santa Fe. New Mexico: B, CB, CT, DA, LA, LN, MK, OT, RA, SD, SF, SJ, SM, TO, TR, and undoubtedly all others. Comment: Successful synthetic hybrids have been formed between Poa pratensis and members of most other sections of Poa. It appears that such hybridization is a continu- 414 Great Basin Naturalist Vol. 45, No. 3 ing process in nature. Specimens and partial populations identifiable as P. pratensis infre- quently display intermediate characteristics between it and some sympatric native spe- cies. A peculiar, possibly native form, which has prominent long spiculae on ventral blade surfaces and is sparsely hairy between lemma nerves, has been collected in MR and LN counties. Poa reflexa Vasey & Scribn. Poa reflexa Vasey & Scribn., Contr. U.S. Natl. Herb. 1:276. 1893. HOLOTYPE:Lettermon Kelso Mt. near Torrey Pk. in 1885 (US 28544900!). Nodding Blue- grass. Short-lived perennial. Culms slender, loosely tufted, 1-6 dm tall. Leaves bright green, mostly cauline. Sheaths closed \-\ the length, keeled, smooth. Ligules 1..5-3.5 mm long, truncate to obtuse, entire to jagged, glabrous. Blades of the culm flat or folded, 1.5-4 mm broad, upper 2-10 cm long, promi- nently keeled and prow tipped (broad and short relative to P. leptocoma). Panicles nod- ding, open, pyramidal, 4-15 cm long, branches capillary, glabrous, mostly 1-3 per node, widely divergent to steeply reflexed, with spikelets crowded in distal ^. Axis in- ternodes mostly shorter than 4 cm. Spikelets green to purplish, strongly compressed, mostly 3-4-flowered, ovate, compact, rachilla rarely visible in side view. Glumes acute, smooth or sparsely scabrous on keel near tip, subequal, the first lanceolate, 1-nerved, the second broadly lanceolate, 1-3-nerved. Lem- mas broadly lanceolate, acute, 5-nerved, densely villous-pilose on keel (to near the tip) on marginal nerves and often on intermediate nerves, mostly glabrous between nerves to sparsely pubescent between them on upper florets, hyaline margin narrow. Gallus long webbed. Paleas villous (sometimes obscurely so) on keels, with long and short cells between keels. Flowers perfect, frequently cleistoga- mous. Anthers 0.25-1 mm long. Chromo- some numbers: 28. Habitat: Subalpine, alpine, from dry open ground to meadows, streams, and bogs, 3050-3660 m in New Mexico. Flowering July to August. Distribution: Middle and Southern Rocky Mountains (very sporadic westward), occur- ring from s Montana to n Arizona and n New Mexico: CF, SF, TO. Comment: See comment under Poa lepto- coma, and subgeneric placement above. Poa reflexa appears to be closely related to P. occidentalis. Poa secunda Presl Poa secunda Presl, Rel. Haenk. 1:271. 1830. Type: Haenke, "Cordillera Chilensibus, " in 1790 (PR, MO, GH). (It has been remarked by Marsh and Keck, among other workers, that the type may have come from s California rather than Chile, but, according to historian S. D. McKelvey (1955), if the 1790 date is correct, this could not be be- cause Haenke had not reached North America until 1791.) Poa sandbergii Vasey, Contr. U.S. Natl. Herb. 1:276. 1893. Poa incurva Scribn. & Will. Poa scahrella (Thurb.) Benth. e.\ Vasey. Poa canhiji (Scribn.) Beal. Poa gracillema Vasey. [The following two taxa are considered to be conspecific with P. secunda (.s. lat. ). However, they are con- sidered to be distinctive ecotypes and are de- scribed separately: Poa ampla Merr., Rhodora 4:145. 1902. Type': Vasey 3009 Steptoe, Washing- ton (US 28610400). Poa nevadensis Vasev ex Scribn., Bull. Torrey Bot. Club 10:66. 1883. Type: Jones Austin, Nevada (US 28858200).] Sandberg Bluegrass. (Pine Bluegrass, Canby Bluegrass). Perennial, cespitose. Culms densely tufted, erect or divergent, wiry, 1.5-12 dm tall, frequently becoming reddish. Leaves mostly basal, green or occasionally glauces- cent. Sheaths nearly open or closed to \ the length, glabrous to scabrous, weakly keeled. Ligules 1.5-5 (7) mm long, acute to acumi- nate, entire to lacerate, glabrous to scabrous on back. Blades firm or mostly soft, mostly 1-5 cm long on culm, steeply ascending or some- times laxly so, 1-3 mm wide, flat or weakly folded, basal leaves often filiform. Panicles mostly narrowly contracted, 2-27 cm long, branches (2) 3-4 (8) per node, mostly scabrous. Spikelets 3.5-9 (12) mm long, gen- erally more than 3 times as long as wide (closed), 2-6-flowered, more or less terete in cross-section. Glumes lanceolate, 1-3- nerved, with broad hyaline margins, broadest about middle, mosth gradualK- rounded on margins, mostly smooth below and sparsely scabrous toward tip on keel and less scabrous on lateral nerves. Lemmas relatively long and narrow, tapered to tip from well above mid- dle, acute, obtuse, rounded, or retuse at tip, 5-nerved, intermediate nerves often obscure, weakly keeled or keel obscure, papillose- roughened to scabrous over surface, basal portion glabrous or usually uniformly puberu- July 1985 SoRENG: New Mexico Poa 415 lent, this infrequently confined to nerves (rarely much denser or longer on nerves). Cal- lus glabrous (sometimes with few short hairs). Paleas mostly scabrous on keels to puberulent on lower half, puberulent between keels be- low and scabrous in upper portion. Rachillas glabrous, scabrous, or puberulent, usually elongate and visible from side view. Flowers perfect, anthers occasionally abortive. An- thers 1-3.8 mm long, yellow or purple or both. Chromosome numbes 44, 56, 61-66, 68, 70-72, 78, 81-106. Habitat: Upper pinyon-juniper, pon- derosa pine, subalpine, and alpine, 2130-2440 m and 3350-3810 m in New Mex- ico. Flowering June-August. Distribution: South America in Chile and Argentina. North America, se Alaska across s Canada (sporadic east of Rocky Mountains), all w United States and n plains states to Great Lakes region, and Gaspe Peninsula. Infre- quent in n Arizona. New Mexico: SJ {canbyi form), SF & TO (alpine forms usually referred to P. canbyi). Comment: Poa secunda {s. lat. ) is a large, polymorphic, facultatively agamospermous species composed of many eco types. Of 40 different taxa that have been proposed, mod- ern taxonomic treatments of this group recog- nize as few as one to as many as 13 species and subspecies. Although 1 have not studied this group in depth, I am familiar with the varia- tion patterns in other agamous complexes in Poa and am inclined to follow the monogra- phers who recognize only one species here. Beyond this, it would be useful from an eco- logical perspective to recognize the more dis- tinctive races at a subspecific level. However, appropriate name combinations are not avail- able, and, until the most recent revision of the complex is published, I refrain from propos- ing any new ones. New Mexico is at the south- ern boundary of the complex where several of the more customarily recognized races are I apparent and allopatric. The above descrip- tion applies to Rocky Mountain forms in- cluded as species of the "scabrellae" group by A. S. Hitchcock. The following descriptions apply to two forms he included as species in the "nevadenses" group. It must be kept in mind, though, that these are intergradient ' over their greater ranges. The highly variable alpine forms of this complex occurring in the Rocky Mountains are usually referred to P. canbyi, but fre- (juently they are exceedingly difficult to dif- ferentiate from Poa glauca, and occasionally from P. arctica. This taxon can generally be distinguished from the latter alpine species by the presence of more lax foliage, generally reddish culm bases, long acute ligules, short, uniform lengthed puberulence of the lemmas, and elongate spikelets. However, it appears that there are continuous intergrades be- tween these facultatively agamous species. Poa a7npla phase of Poa secunda Big Bluegrass Robust perennials, infrequently rhizoma- tous. Culms several in large tufts, 6-18 dm tall. Leaves green to very glaucous. Ligules 0.25-4 mm long, sparsely to densely scabrous on back, obtuse to truncate, entire to jagged. Blades 2-5.5 mm wide, those of culm 9^20 cm long, flat to weakly folded, mostly rather firm. Panicles mostly 12-25 cm long, rather nar- row. Glumes 3-nerved, scabrous on keel. Lemmas 4-6 mm long, glabrous and papil- lose-roughened all over to very sparsely pu- berulent on lower keel and marginal nerve, scabrous or glabrous on upper keel. Callus glabrous. Paleas scabrous on keels. Rachillas glabrous. Flowers perfect. Anthers 1.5-3.5 mm long. Chromosome numbers: 62—65, 68, 70-71, 91, ca 100. Habitat: The only records of Poa a7npla in New Mexico have been from ponderosa pine zone, in open ground, 2130-2440 m. Flower- ing May-August. Distribution: Alaska, Alberta, British Co- lumbia, North Dakota, South Dakota, Ne- braska, Montana, Wyoming, Idaho, Washing- ton, Oregon, Nevada, California, and nw ^ Colorado. Introduced into New Mexico along roads and in campgrounds in CB, GR (SM?, not seen); all three collections made since 1980. Poa nevadensis phase of Poa secunda Nevada Bluegrass Similar to P. secunda 5-10 dm tall, very leafy throughout, sheaths and blades often scabrous, flat or folded, bright to pale green, 416 Great Basin Naturalist Vol. 45, No. 3 1-3.5 mm wide, upper ligules 3-6 mm long, acute to acuminate. Panicles narrow, elon- gate, to 25 cm long, somewhat loose. Spikelets 2-5-flowered, 4-8 mm long. Glumes more strongly keeled than typical of species. Lemmas 3.. 5-5 mm long, scabrous apically or throughout, usually somewhat ob- tuse. Chromosome numbers: 62, 63-66, 70. Habitat: In New Mexico: Moist ground around springs and moist meadows, pon- derosa pine zone, 2350-2450 m. Flowering June-July. Distribution: California, Oregon, Mon- tana, Idaho, Wyoming, Colorado, Utah, Ari- zona, and New Mexico. Not collected in New Mexico since 1906, CF (Bell), CT (Fitzgerald Cienaga, n of Reserve). A collection of Dr. Bigelow's 1853—4 #3 (NY) made somewhere between Fort Smith, Arkansas, and the Rio Grande, also appears to represent this form. Poa tracyi Vasey Poa tracyi Vasey, Bull. Torrey Bot. Club 15:49. 1888. Lectotype: Tracy, New Mexico, mesa sides near Raton, in 1887 (US 556764!). Poa flexuosa var. robusta. Poaf. var. occidentalis , Poa occidentalis (Vasey) Rydb. (not P. occidentalis Vasey). Tracy Bluegrass. Perennial, frequently subrhizomatous. Culms erect or decumbent at base, 25-125 cm tall, mostly glabrous. Leaves blue-green, mostly cauline. Sheaths mostly shorter than internodes, closed l-jo the length, strongly keeled, retrorsely puberulent, or scabrous, mostly glabrous. Ligules obtuse to acute, glabrous to abundantly hairy on back, those of the upper leaves to 4.5 cm long, mostly shorter than leaf is wide. Blades strongly keeled, flat, prominently prow shaped at tip, upper ones 6-18 cm long, 2-5.5 mm wide. Panicles narrowly to broadly pyramidal, (8) 13-29 cm long. Branches 1-5 per node, 2.5-18 cm long, sparsely flowered in distal I, widely spreading to reflexed at maturity, sparsely scabrous. Axis internodes mostly over 4 cm long. Spikelets strongly com- pressed, oblong, 3-8 mm long, mostly shorter than pedicels, 2— 8-flowered. Glumes nar- rowly lanceolate, acute, the first usually less than i the length of first lemma, 1-nerved, the second slightly shorter than first lemma, 1-3- nerved. Lemmas 2.6-5 mm long, lanceolate, keel scarcely incurved at tip, pale green with white hyaline margins, 5-nerved, intermedi- ate nerves often obscure, villous on lower |-f of keel, less so on marginal nerves, usually sparsely puberulent between nerves at base. Callus with long cobwebby hairs abundant (rarely nearly absent). Paleas mostly minutely scabrous on keels, sometimes also minutely puberulent near base, with long and short cells between keels. Rachillas usually visible from side view, glabrous. Flowers perfect or pistillate, two types usually mixed within pan- icle, sometimes all one or the other (partially gynodioecious). Anthers vestigial (aborted) or 1.25-3 mm long. Chromosome numbers: 28, 28+1 fragment. Habitat: Montane. In gambel oak thick- ets, mixed-conifer and spruce-fir forest open- ings, and subalpine meadows, mesic sites, in humus-rich soils, 2000-3500 m. Flowering May-mid-July and in south central New Mex- ico in August. Distribution: Southern rocky Mountains of Colorado and New Mexico; B, CF, LN, OT, RA, SM, U. All of the Wyoming records of P. tracyi that I have seen represent other spe- cies, especially P. leptocoma. Comment: Mature material of Poa tracyi is easily distinguished (by longer, or vestigial anthers) from P. occidentalis, P. leptocoma, and P. reflexa. Poa tracyi is morphologically very similar to the more western P. curta (s. auct), and both species are partially g\'nodioe- cious. However, they are geographically iso- lated from one another, and the moi-phologi- cal distinctions between them are not bridged. Poa trivial is L. Poa trivialis L., Sp. Pi. 67. 17.53. Meadow or Roiighstem Bluegrass. Short-lived perennial, frequently stolonif- erous. Culms loosely tufted, erect, or decum- bent and rooting at lower nodes, 4-11 dm tall, smooth to densely scabrous. Leaves bright green, mostly cauline. Sheaths closed for 4-3 the length, often keeled and prominently stri- ate. Ligules (1.5) 3-10 mm long, acute, entire or lacerate, glaiirous or sparsely scabrous on back. Blades of culm fiat, 2-8 mm wide, 10-25 cm long, liLx, tips scarcely prow shaped. Pani- cles (4) 6-22 cm long, open, pyramidal to oblong, slender branches spreading-ascend- July 1985 SoRENG: New Mexico Poa 417 ing in flower and fruit, many flowered from near base, lower branches often in 5s. Axis internodes 2-5.2 cm long. Spikelets small, 3-5 mm long, strongly compressed, 2-4- flowered, ovate to elliptic. Glumes unecjual, the first narrowly lanceolate, more or less in- curved and with narrow white hyline margin, the second somewhat broader and less ^ curved, the 1-3 nerves prominent, pale, and I mostly scabrous. Lemmas 2.. 5— 4 mm long, prominently 5-nerved, smooth or faintly, finely papiflose-roughened, sparsely villous on lower keel, otherwise glabrous. Callus with web of villous hairs 2 as long to longer than lemma. Paleas very finely and closely scaberulous to papilose-roughened on keels. Rachilla internodes short, slender, and glabrous. Flowers perfect. Anthers 1.4-2 mm long. Chromosome numbers: 14, 15, 28. Habitat: E.xpected to spread in montane New Mexico in shady, cool, wet to mesic soils, in disturbed sites, from 1980-3050 m. Flow- ering May-July. I Distribution: Introduced from Europe, common in nw and ne U.S. but infrequent in interior West. The one New Mexico record of this species (Soreng 6- Ward 1609, Eagle Cr., in 1982) was growing in wet soil, at water's edge, at about 8000 ft: LN. Comment: This species is superficially very similar to P. occidentalis, but has branches flowered from near the base; has longer an- thers and is partially self-incompatable; has a frequently stolonous habit, tuberculate nobs on the palea keels, no pubescence between the lemma nerves or on the marginal nerves, and narrow, scarcely prow-tipped blades. Acknowledgments Acknowledgments are due to V. L. Marsh, Elizabeth Kellogg, Lowis Arnow, and David D. Keck. I am grateful to Nancy Soreng, Rich Spellenberg, Kelly Allred, and all others who have reviewed this paper. Literature Cited Butters, F K , and E C. Abbe. 1947. The genus Poa in Cook County, Minnesota. Rhodora49:l-21. Chrtek, J , AND V JiRASEK. 1962. Contribution to the systematics of species of the Poa L. genus, section Ochlopoa(A. etGr.)V. Jiras. Presha .34:40-68. Edmondson, J R. 1978. Infragenaric taxa in European Poa L. Botanical J. Linn. Soc. 76:329-334. Hitchcock, C. L. 1969. Vascular plants of the Pacific Northwest. University of Washington Publica- tions in Biology 17(l):648-683. illust. Keck, D D. UnpubUshed typescript, untitled [a revision of Poa of the contiguous western United States] ca 65 pages, ca 1949. Kellogg, E. A. 1985. A biosystematic study of the Poa secunda complex. J. Arnold Arboretum 66: 201-242. Marsh, V. L. 1950. A ta.xonomic revision of the genus Poa of United States and southern Canada. Unpub- lished dissertation. University of Washington. 1952. A taxonomic revision of the genus Poa of the United States and southern Canada. Amer. Midi. Naturalist 47:202-250. Soreng, R. J , and S L. Hatch. 1983. A comparison of Poa traciji and Poa occidentalis. Sida 10:123-141. Tzvelev, N 1976. Grasses of the Soviet Union, Zlaki SSSR. [English translation, Smithsonian Inst.: Oxonian Press, Pvt. Ltd., New Delhi. 1983]. 418 Great Basin Naturalist Vol. 45, No. 3 POA ARCTICA subsp. APERTA (. ~r ~1 ^ ( o CtT , ^ n ( " / ^^v\J: o oo o ^^ 5 q. rQ\ o o 7 [ o -^ H ^ / "l ~n^^^ o o _- 0 o o S M \ A BIGELOVII PO Fig. 1. New Mexico distrihutions for Poa spp. , as labeled. July 1985 SoRENC: New Mexico Poa 419 POA FENDLERIANA subsp. ALBESCENS 3 '^\ -r 1 r r (1 9. r -1 /^ fh 3 r O o -d. / r J ir i \ fr \ 1 r IIANA sub£ 1 1 PC )A FENDLEI p. LONGILIGU POA GLAOCA subsp. GLADCA Fig. 2. New Mexico distributions for Poa spp. , as labeled. 420 Great Basin Naturalist Vol. 45, No. 3 POA OCCIDENTAL IS POA PALUSTRIS Fig. 3. New Mexico distriliutions for Poa spp. , as labeled. July 1985 SoRENG: New Mexico Poa 421 POA SECUNDA (nevadensis form) POA TRACY I Fig. 4. New Mexico distributions for Poa spp. , as labeled. 422 Great Basin Naturalist Vol. 45, No. 3 POA TRIVIALIS Fig. 5. New Mexico distribution for Poa trivialis. DWARF MISTLETOE-PANDORA MOTH INTERACTION AND ITS CONTRIBUTION TO PONDEROSA PINE MORTALITY IN ARIZONA Michael R. Wagner' and Robert L. Matliiaseii' Abstr.\CT. — The interaction between Southwestern dwarf' mistletoe, Arcciitlwhiuin lafiinatwn subspecies cryp- topodum, infestation and defoliation by the pandora moth, Coloradia pandora pandora, on the Kaibab Plateau in Arizona was evaluated. Heavy defoliation of ponderosa pine, Pimis ponderosa, in 1979 and 1981 resulted in mortality of individual trees in areas of heavy dwarf mistletoe infestation. Postmortem evaluation of ponderosa pines indicated that dead trees had a significantly higher dwarfmistletoe rating than did nearby paired live trees. Of 25 tree pairs evaluated, only two live trees had higher dwarfmistletoe ratings than the paired dead tree. Mean dwarf niistletoe ratings were: live trees 2.9, dead trees 4.6 (6 class dwarfmistletoe rating system). Implications for management of the pandora moth are discussed. An outbreak of the pandora moth, Color- adia pandora pandora Blake (Lepidoptera: Saturniidae), began in 1979 on the Kaibab Plateau in northern Arizona. Defoliation of ponderosa pine, Pinus ponderosa Dougl. ex Laws., occurred over 5,000 acres in 1979 and 19,000 acres in 1981 (Bennett and Ragenovich 1982). Pandora moth defoliation resulted in radial growth loss between 17% (Miller 1983) and 25% (Bennett and Andrews 1983). Tree mortality was not significantly greater in defo- liated plots than nondefoliated control plots (Bennett and Andrews 1983). Though there was no significant effect of defoliation on mor- tahty on a stand basis, there were clearly small pockets of mortality (Wagner pers. obs.). Field observations by the authors indicated that many of the trees that died were infected with Southwestern dwarf mistletoe, Arceii- thobium vaginattim subsp. cnjptopodum (En- gelm.) Hawksw. & Wiens. Bennett and An- drews (1983) found that radial growth loss in trees with mistletoe and pandora moth was greater than for those with only pandora moth in this area. Many insects and diseases do not cause di- rect serious impact on their host but rather function to predispose trees to other damag- ing agents. Numerous authors have reported that pandora moth defoliation increased the incidence of bark beetle mortality (Carolin and Knopf 1968, Keen 1952, Massey 1940, Patterson 1929, Wygant 1941). Dwarf misde- toes, which are serious damaging agents in western forests (Hawksworth 1961, Hawks- worth and Wiens 1972), are a common agent predisposing trees to other agents. The nu- merous interactions between dwarf mistle- toes and forest insects have been reviewed by Stevens and Hawksworth (1984). We report in this paper that ponderosa pine mortality, fol- lowing the pandora moth outbreak in north- ern Arizona, occurred primarily on trees heavilv infected with dwarfmistletoe. Materials and Methods The study site was approximately two miles north of Jacob Lake, Arizona, in an area heav- ily defoliated in 1979 and 1981 by the pandora moth. The sample area was approximately 100 acres in size and constituted the largest pocket of mortality that had occurred following defo- liation (Wagner pers. obs.). The study site was outside the area sprayed with Acephate® in 1981 (based on maps in Bennett and Ragen- ovich 1982). A systematic sample with a random start was used to survey the area of mortality. Sam- ple plots were two chains apart and one chain in radius. The nearest dead tree to the plot center was selected, and the nearest live tree of similar diameter (maximum acceptable dif- ference four inches dbh) identified (Fig. 1). The diameter breast height (dbh) and 6-class dwarf mistletoe rating (DMR) (Hawksworth School of Forestrv, Box 4098. Northern Arizona University, Flagstaff, Arizona 86011. 423 424 Great Basin Naturalist Vol. 45, No. 3 Fig. 1 . Typical pa dead tree. d sample of a living and dead pond.-n.sa pine. \..te the header dwarf nustletoe inieeticn on the July 1985 Wagner. Mathia.skn: Mistletoe-Moth Interaction 425 Table 1. Mean diameters and dwaif mistletoe ratings ;)f paired live and dead ponderosa pines. Live trees Dead trees dbh DMR dbh DMR Mean (X) SD Range 16.4 8.1 5.9-35.9 2.84 1.75 0-6 16.5 4.64 8.1 1.35 5.6-34.2 1-6 1977) were determined for each tree in the pair. If a suitable pair of trees could not be located within a plot, the plot was rejected. A total of 38 plots were examined, of which 25 met the criteria of having a suitable pair of trees. Data were analyzed using a nonparametric sign test (Conover 1980) (a = 0.05). The sign test indicates whether one random variable in a pair tends to be larger than the other random variable in the pair. The null hypothesis was that there was no difference in the dwarf mistletoe ratings between dead and live trees in the stand. Results The mean diameters and mean DMR's of the paired trees are given in Table 1. There was no statistically significant difference in mean dbh between live and dead trees, as should be expected because of the pairs cho- sen. However, there was a highly significant difference in the mean DMR of live and dead trees. Dead trees sampled generally had signs of heavy dwarf mistletoe infection (DMR 5 or 6). Dead trees had a higher mean DMR rating than their paired live trees in all but two sam- pled pairs, where the live tree had a higher DMR than the dead tree. We observed that the mortality occurred over a considerable range of tree diameters. None of the dead trees we examined were free of dwarf mistletoe. Discussion Our data indicate ponderosa pine mortality tended to occur on trees that were heavily infected with dwarf mistletoe. We can con- clude that the probability of mortality as a result of pandora moth defoliation is greater in stands heavily infested with dwarf mistletoe. Our experimental design does not permit the establishment of a cause-effect relationship. but it appears reasonable that dwarf mistletoe is predisposing trees to mortality following defoliation. We do not feel there is evidence to suggest that the pandora moth prefers dwarf-mistletoe-infested trees, since defolia- tion is often uniform over large areas. Rather, we feel trees weakened by dwarf mistletoe infection probably are less tolerant of defolia- tion than are healthier trees. We did not at- tempt to systematically determine the cause of mortality of each sampled tree in the study area. However, the few trees we did examine did not appear to be killed by bark beetles or other secondary agents. These findings have important implications for management of the pandora moth. Since growth loss is moderate and probably does not occm- for more than one or two years, control attempts directed at reducing growth loss are not justified. Mortality is a more serious im- pact and would justify control measures if ex- pected to occur over large areas. We would recommend controlling the agent predispos- ing trees to mortality (dwarf mistletoe) as the preferred option. Silvicultural control strate- gies for dwarf mistletoes are well established (Scharpf and Parmeter 1978). We would specifically recommend selectively removing trees with a DMR of 3.0 or greater in stands likely to be heavily defoliated by the pandora moth. Because the importance dwarf mistle- toes play in reducing growth is well known (Hawksworth and Wiens 1972), the added ef- fect of expected defoliation would certainly justify control efforts in managed forest stands. In forest areas not under intensive forest management, defoliation by the pandora moth may actually have a beneficial effect. Since mortality preferentially occurs on the more heavily dwarf-mistletoe-infected trees, pandora moth defoliation may have the affect of reducing stand mistletoe infestation levels. This may increase the desirability of these areas for future use as managed stands. Cer- tainly the mortality of some trees would provide considerable wildlife habitat for a va- riety of cavity-nesting birds. Acknowledgment We acknowledge E. A. Blake and J. M. DiMatteo for their assistance in data collec- tion. 426 Great Basin Naturalist Vol. 45, No. 3 Literature Cited Bennett, D. D., and M. A. Andrews. 1983. Biological evaluation of the pandora moth — Kaibab National Forest. USDA Forest Service. Southwestern Re- gion, Forest Pest Management Report R-3 83-7. 15 pp. Bennett, D. D . and 1 R Ragenovich 1982. A pilot control project to evaluate Acephate® for control of the pandora moth, Coloradia pandora Blake (Lep- idoptera: Saturniidae), Jacob Lake, Arizona 1981. USDA Forest Service. Southwestern Region, Forest Pest Management Report R-3 82-10. 36 pp. Carolin, V. M.. AND J. A. E. Knopf. 1968. The pandora moth. USDA Forest Service, Pacific Northwest Forest and Range E.xperiment Station. Forest Pest Leaflet 114. 7 pp. Conover. W. J. 1980. Pages 122-142 in Practical non- parametric statistics, 2d ed., John Wiley & Sons Inc., New York. Hawksworth, F. G 1961. Dwarf mistletoe of ponderosa pine in the Southwest. USDA Forest Service Technical Bulletin 1246, 112 pp. 1977. The 6-class dwarf mistletoe rating system. USDA Forest Service. General Technical Report RM-48. 7 pp. Hawksworth, F G., and D. Wiens 1972. Biology and classification of dwarf mistletoes (Arcetithobiutn). USDA Forest Service Agriculture Handbook 401. 234 pp. Keen. F. P. 1952. Pages 83-86 in Insect enemies of west- ern forests. LISDA Miscellaneous Publication 273. Massey, C L. 1940. The pandora moth {Coloradia pan- dora Blake), a defoliator of lodgepole pine in Colo- rado. Unpublished thesis. Duke L'niversity, Durham, North Carolina. 33 pp. Miller, K. K. 1983. The ecology and impact of the pan- dora moth, Coloradia pandora pandora Blake, in northern Arizona. Unpublished thesis. Northern Arizona University, Flagstaff. 62 pp. Patterson, J E 1929. The pandora moth, a periodic pest of western pine forests. USDA Technical Bulletin 137. 20 pp. SCHARPF, R. F., AND J. R. PARAMETER, EDS. 1978. Proceed- ings of the symposium on dwarf mistletoe control through forest management. L^SDA Forest Ser- vice General Technical Report PSW-31, 190 pp. Stevens, R E , and F G H,\wksworth. 1984. Insect- dwarf mistletoe associations: an update. Pages 94—100 in Proceedings biology of dwarf mistletoe. USDA Forest Service General Technical Report RMilll. Wygant, N. D. 1941. An infestation of the pandora moth, Coloradia pandora Blake, in lodgepole pine in Colorado. J. Econ. Ent. 34(5): 697-702. OCCURRENCE OF ANISAKID LARVAE (NEMATODA: ASCARDIDIA) IN FISHES FROM ALASKA AND IDAHO Richard Hocknuuiii' aiul Terry Otto' Abstract. — All 25 sablefish (Aiwplopoma fimbrki) exainined from two hays near Sitka, Alaska, were infected with anisakid larvae. There were 1 to 11 larvae per infected fish, with worms encysted in the mnscniatnre of the body wall of 5 fish and in the liver of 4 fish. For the other hosts the viscera was the site of infection. ( Ihinook salmon (Onchurinpulitis tshawytscha ) from Barrow, Alaska, and Obsidian, Idaho, were also infected with anisakid larvae. These data extend the known northern distribution of the anisakids along the Pacific Coast for sablefish and chinook salmon. The pathogenesis of the migratory pathway of anisakid larvae is described, and comments on human health implications are presented. Anisakine nematodes have been a major prob- lem in the fishing indnstry for years because their presence reduces the commercial value offish (Meyers 1979, Wootten and Waddell 1977, Wootten 1978). During the past 10 years it has been observed that two genera, Anisakis sp. and Phocanema sp., are danger- ous to humans in South America and North America who consume raw or poorly cooked infected fish (Jackson 1975, Meyers 1975, Cat- tan 1976, Kates 1973, Morbidity and Mortal- ity Weekly Report 1975). The primary species observed for this study was Anisakis simplex. The disease, anisakiasis, has been recognized in Europe and the Far East for several decades (Oshima 1972, Shiraki 1974, Smith and Wootten 1978). The survival of anisakid larvae in various fish-processing methods has been demonstrated by Hauch (1977). Precau- tions are necessary in preparing fish for hu- man consumption. During the last two decades there has been renewed interest in the importance of the anisakine nematodes (Myers 1979, Hadidjaja et al. 1978). Surveys have been conducted throughout the world to determine the occur- rence and distribution of the anisakids in fish, especially those hosts of commercial value. The larval worms have recently been reported in a bowhead whale {Balaena mysticetus) har- vested at Barrow, Alaska, (Migaki et al. 1982) and in fish from Chile (Carvafal 1981, Torres et al. 1983). It appears that these larvae have a worldwide distribution in fish. One study of fish hosts demonstrated the presence of an- isakid larvae in 138 species in marine fish and one species of squid (Ono 1975). One objective of this study was to extend the northern geographical distribution of an- isakias and to determine pathogenesis for host tissue. Because of the availability of the main host species, sablefish (Anoplopoma fimbria) were studied in Sitka, Alaska, to determine the geographical range of anisakine larvae. A limited number of chinook salmon were ob- tained from two other sites for study. The proposed life cycle of Anisakis begins with the release of eggs from the adult worm that usually is found in the large intestine of marine mammals (cetaceans and pinnipeds: Myers 1970, Vik 1964, Smith and Wootten 1978). The eggs develop into stage I and II larvae, which are preyed upon by krill and other crustaceans such as Thysanoessa sp. (Smith 1971). The larvae penetrate the intesti- nal tract and develop into stage III larvae in the crustacean host. Fish prey upon the in- fected crustaceans and become paratenic hosts for marine mammals that are the defini- tive hosts. In marine mammals the anisakids develop into adult worms and release eggs (Smith and Wootten 1978). Anisakid larvae found in fish are usually tightly coiled on the mesenteries, liver, and gonads and in the musculature of the body wall. Prusevich (1964) demonstrated that the capsule surrounding the larvae is of host origin and that the inflammatory reaction to the presence of the parasite in the liver of the shorthorn sculpin began during the first few Department of Zoology, Brigham Young University, Provo, Utah 84602. 427 428 Great Basin Naturalist Vol. 45, No. 3 Fig. 1. Anisakid larva (AL) encapsulated (K(',) in liosl muscle (HM). (400 x) hours of the invasion. The musculature of her- ring was a common sight of infection for the host taken from British marine waters (Davey 1972). The infection in herring was indepen- dent of host age or length. Freshwater fish such as Hemibarbus bar- bus (Ono 1975) and trout (Wootten and Smith 1975) have also been reported as hosts for anisakid larvae. Materials and Methods During the summer months of 1978, 25 sablefish {Anoplopoma fimbria) were taken by line from Starrigavan Bay and Thompson Harbor, Sitka, Alaska. Sixteen fish were har- vested from Starrigavan Bay and nine from Thompson Harbor. Each fish was examined immediately upon death or 24 hours later. The later fish were stored under refrigeration until examined. Infected muscle and liver tis- sue were fixed in 10% buffered formalin fol- lowed by histological preparation liy standard methods (Humason 1972). The tissue sections were stained with haemotoxylin and eosin (H + E) and Mallory's Triple, a trichrome stain. During 1982 and 1983 a total of 10 chinook salmon, from the spawning traps near Obsid- ian, Idaho, were examined for parasites in- cluding Anmi/cw. Samples of infectcxl chinook salmon were sent to the senior author from Barrow, Alaska. The occurrence of anisakid larvae was tabu- lated. Sections from the sablefish infected with anisakine larvae were studied in the labo- ratory using light microscopy. Vy^- 2. Aiiisaki-89, Navicula paramutica ■ 90-92, Navicula incotispicua ; 93-95, Nitzschia /lantoc/uana ; 96-98, Nitoc/iia pa/ea; 99-100, Nitzschia communis , 101-102, Pimmlaria borealis , 103-105. Rlwpalo- dia (iihiH-rula var. lanhcurckii- 106, Stcphanodisnts carconcnsis. All ligiiros art- X2{)0(). July 1985 RUSHFORTH, JOHANSEN: UTAH S()ILAl(;AE 441 10 |xm. This species was rare in our samples and has not been observed previously in the soils of our area. Nitzschia inconspicua Gnniow (Figs. 90-92). Valves 7-8 ^JLnl long by 2.7-3.5 |xm M/ide; fibulae 12-14 in 10 |xm; striae 28-30 in 10 ixm. This taxon has not been reported pre- viously from soils of the Great Basin and Colo- rado Plateau. Nitzschia palea (Kuetz.) W. Smith (Figs. 96-98). Valves 17-22 |xm long by 2.5-3 |xm wide; fibulae 12-16 in 10 |xm; striae unre- solved. This taxon has occasionally been found in soils throughout the West. It was rare in samples from the Uintah Basin. Pinnularia horealis Ehr. (Figs. 101-102). Valves 26-31 |Jim long by 7-8.5 |xm wide; striae 5-6 in 10 |xm. This taxon is a common constituent of soils worldwide. It was one of the most common diatoms observed in this study, ranking third in importance behind Hantzschia amphioxys and the Navicitla mu- tica complex. Many of our valves have the rectangular outline typical of F. borealis var. rectangidaris but did not have the coarser striae characteristic of that variety. Rhopalodia gihha (Ehr.) O. Mueller. Valve 55 (xm long by 16 (xm wide; costae 8 in 10 |xm; alveoli rows 14 in 10 |xm. Only a single speci- men of this taxon was observed in our sam- ples. Rhopalodia gibherida var. vanheurckii O. Mueller (Figs. 103-105). Valves 1.5-49 fjtm long by 5.5—11 |xm wide; costae 4—7 in 10 |xm; alveoli rows 16-20 in 10 jxm. We have ob- served this taxon in other soils of Utah where it is as rare a species as it is here. The samples from the Uintah Basin contained some very small specimens that were well below the minimum size recorded by other authors. The larger specimens (Fig. 105) are transitional between K. gibbenda var. vanheurckii and the nominate variety. Stephanodisciis carconensis (Eulens.) Grunow (Fig. 106). Valve 16 |xm in diameter; costae 3 in 10 |xm; aerolae forming 4 rows between costae, 12 in 10 |xm. This species is widely distributed in the soils of Utah and Arizona, though it is always present in low numbers (Johansen et al. 1981, 1984). It has occasionally been reported as S. astraea var. minutula (Anderson and Rushforth 1976). Literature Cited Ali, S . AND (; R. Sandu 1972. Blue-green algae of the saline soils of the Punjab. Oikos 22:268-272. Anantani, Y. S., and D. V Makatiik. 1974. Soil aggregat- ing effects of some algae occurring in the soils of Kutch and Rajasthan. J. Univ. Bombay 41(68): 94-100. Andkkson, D C , AND S. R. RusHKOHTn. 1976. The cryp- togamic flora ofdesert soil crusts in southern Utah. Nova Hedwigia 28:691-729. Ashlky, J . and S R Rushforth. 1984. Growth of .soil algae on top soil and processed oil shale from the Uintah Basin, Utah, USA. Reclam. Reveg. Res. ,3: 49-63. Bailey. D , A P Mazukak, and J R Rosowski. 1973. Aggregation of soil particles bv algae. J. Phvcol. 9:99-101. Breazeale, J. M. 1929. Algae and their effect upon the growth of citrus seedhngs. Unpublished thesis. University of Arizona, Tucson. Brotherson. J. D, and S. R Rushforth. 1983. Influence of cryptogamic crust on moisture relationships of soils in Navajo National Monument, Arizona. Great Basin Nat. 43(l):73-78. Desikachary, T V. 19.59. Cyanophyta. Indian Council of Agricultural Research, New Delhi. 686 pp. DuRRELL. L. W., andL. M. Shields. 1961. Characteristics of soil algae relating to crust formation. Trans. Amer. Microsc. Soc. 80:73-79. Fletcher. J E., andW P. Martin 1948. Some effects of algae and molds in the rain-crust of desert soils. Ecology 29(1):9.5-100. Hunt, C. B.. and L. W. Durrell. 1966. Distribution of fungi and algae. Pages .55-66 in D. B. Hunt, ed.. Plant ecology of Death Valley. California Geol. Sur. Prof. Paper 509, U.S. Government Printing Office, Washington, D.C. Johansen, J. R., S. R., Rushforth, and J. D Brotherson. 1981. Subaerial algae of Navajo National Monu- ment, Arizona. Great Basin Nat. 41:433-439. Johansen, J R , L L St Clair, B L Webb, and G T Nebeker. 1984. Recovery patterns of cryptogamic soil crusts in desert rangelands following fire dis- turbance. The Bryologist 87:238-243. Kleiner, E. F., and K. T. Harper. 1972. Environment and community organization in grasslands of Canyonlands National Park. Ecology 53:229-309. 1977. Soil properties in relation to cryptogamic ground cover in Canyonlands National Park. J. Range Manag. 30(3): 202-205. Klubek, B., and J. Skujins. 1980. Heterotrophic N fixa- tion in arid soil crusts. Ann. Biol. Biochem. 12:229-236. Lewin, R. a. 1977. The use of algae as soil conditioners. CIBASIO Trans. 3:,33-35. Mackenzie. H. J , and H. W Pearson 1979. Prehminary studies on the potential use of algae in the stabi- lization of sand wastes and wind blow situations. Abst. Br. Phycol. Soc. Winter Meeting. Br. Phy- col. J. 14:126. Mayland, H F , T H MclNTOSH, and W H Fuller 1966. Fixation of isotopic nitrogen in a semi-arid soil bv algal crust organisms. Soil Sci. Soc. Amer. Proc. 30:56-60. 442 Great Basin Naturalist Vol. 45, No. 3 Metting, B 1981. The systematics and ecology of soil algae. Botanical Review 47:19.5-312. Nebeker, G T. and L L St. Clair 1980. Enhancement of seed germination and seedling development by cryptogamic soil crusts. Bot. Soc. Amer. Misc. Ser. 158:81. RusHFORTH, S R. AND J D. Brotherson. 1982. Cryp- togamic soil crusts in the deserts of North Amer- ica. American Biology Teacher. 44(8):472^75. Rychert, R. C , and J Skujins 1974. Nitrogen fixation by blue-green algae-lichen crusts in the Great Basin desert. Soil Sci. Soc. Amer. Proc. 38: 768-771. Shields, L. M., and L. W. Durrell. 1964. Algae in rela- tion to soil fertility. Bot. Rev. 30:92-128. Shields, L. M.. C. Mitchell, and F. Drouet. 1957. Al- gae and lichen stabilized surface crusts as soil ni- trogen sources. Amer. J. Bot. 44:489-498. Shubert, L. E., andT. L. Starks. 1979. Algal succession on orphaned coal mine spoils. Pages 661-669 in M. K. Wali, ed. , Ecology and coal resource devel- opment. Pergamon Press, New York. 1980. Soil-algal relationships from surface mined soils. Br. Phycol. J. 15:417-128. Singh, R. N 19.50. Reclamation of "Usar" lands in India through blue-green alga. Nature 165:32.5-326. St. Clair, L. L., and S. R. Rushforth 1976. The diatoms of Timpanogos Cave National Monument, Utah. Amer. J. Bot. 63(l):49-59. St Clair, L L , B L Webb, J R. Johansen, and G T Nebeker. 1984. Cryptogamic soil crusts: enhance- ment of seedling establishment in disturbed and undisturbed areas. Reclam. Reveg. Res. 3: 129-136. Starks, T L , and L E. Shubert. 1978. Algal colonization and succession on a reclaimed strip-mine test plot. North Dakota Acad. Sci. Proc. 32:13. 1982. Colonization and succession of algae and soil-algal interactions associated with disturbed areas. J. Phycol. 18:199-207. Wilson, L, M E Olsen, T B Hutchings, A R. Southard, and A J Erikson. 1975. Soils of Utah. Agric. E.xpt. Sta. Bull. 492. Utah State University, Logan. 94 pp. IN MEMORIAM: WILLIAM WALLACE NEWBY (1902-1977) William H. Bi-hlf' Abstract. — W. W. Newby, professor of biology at the University of Utah from 1927 to 1971, was reknowncd as a great teacher and a clear and forceful writer and for his meticulous research in invertebrate embryology. Ancillary skills were counseling (especially of premedical students), illustrating, wood crafting, and paper preserving. Some of his writings pertained to the history of research in the biological sciences at the University of Utah. His specialty areas in teaching were embryology and genetics, and he served as chairman of the Department of Genetics and Cytology from the year of its creation in 1948 until 1962. The most outstanding example of his research pertained to the early embryology of the echiuroid worm Urechis caupo. Among his colleagues at the University of Utah he had the reputation of being one of the finest teachers that the university ever had. This appraisal of his teaching ability was shared by thousands of students whom he came in contact with during a teaching career that spanned 44 years (1927-1971). He was particularly appreciated by pre- medical students, since one of the many courses he taught was embryology, part of a sequence of required courses in biology in his time for this group. Early in his professional life he did pioneer research in the specialized field of developmental embryology. In later years he eschewed research in favor of com- mittee work and premedical counseling, in which activities he again made significant and prolonged contributions. He served as head of a newly created Department of Genetics and Cytology from 1948 to 1962. He liked to work with his hands and make things. One of his attributes was a remarkable skill in drawing and preparing illustrations for his own research reports and numerous labo- ratory manuals. He was also an expert wood craftsman. Upon his retirement in 1971, at which time he received a distinguished teaching award and the honorary rank of professor emeritus of biology. President James C. Fletcher charac- terized Dr. Newby as one who was "always willing to go the extra mile not only for stu- dents but for colleagues as well." Upon retire- ment he continued to work at the university as a volunteer in the library, serving as a special- ist in paper preservation and restoration. The following resume of Dr. Newby's life and work not only elucidates his career and ,lca^-fis>-*, J y Fig. 1. William Wallace Newby. pays tribute to the man, but also touches on the history of the University of Utah, with which he was affiliated for 50 years. He was one of the outstanding scholars who helped make it the great institution that it is. William Wallace Newby was born in Day- ton, Ohio, on 17 September 1902, the son of William Wallace and Emelia Vornholt Newby. His father was a photoengraver and moved several times through the midwest fol- Department of Biology, 201 Biology Building, University of Utah, Salt Lake City. Utah 84112. 443 444 Great Basin Naturalist Vol. 45, No. 3 lowing this trade. The family lived for several years in NoiAvood, Ohio, near Cincinnati, where young W. W. Newby received his early education in the elementary schools. (He re- ceived a certificate for good penmanship while in the third grade.) He recalled seeing the last passenger pigeon in the Cincinnati Zoo and was greatly impressed that a thou- sand dollars had been offered for a mate in the hope that if one were found the pair would propagate and save the species, but the effort came too late. The family next moved to Kansas City, Mis- souri, where Newby attended Westport High School from 1917 to 1920. He had a natural aptitude for working with his hands, so as a freshman (his ninth year) he took a course in carpentry shop and made a book shelf, table, and couch. This was prophetic of his working up to bigger and better things. The next year he took a course in pattern making that in- volved the preparation of sand castings of iron and brass. His good work in these classes led to a job in the summer of 1918 making wooden propellers for airplanes used in World War I. His third year he took machine shop, which gave him experience in forging and shaping metal, but he preferred woodworking over metalworking. The fourth year of high school he repeated carpentry shop just to have a place to work so he could continue to make various items of furniture for the family home. High school was followed by two years (1920-1922) attending Kansas City Junior College. During his last two years of high school and the two years in junior college, he held several jobs after school and on Saturdays in various shops, big and little, in Kansas City. He worked one summer on a surveying crew mapping a portion of the Missouri River. He was active in the Boy Scouts, and during the summer after junior college, as an a.ssistant Scout executive, he helped run a Scout camp at Noel, Missouri, in the Ozark Mountains. The next summer he assisted at a camp at Pleasant Hill, Missouri, for underprivileged boys from Kansas City. His supervisors urged him to continue in counseling work with them, but he was offered a position at the Haskell Institute, which was one of tiie federal boarding and training schools for Indian chil- dren. The buildings for the education com- plex were situated on the campus of the Uni- versity of Kansas at Lawrence. He had charge of the smaller children and received board and room plus a salary of $25 per month. Occasionally he would bring a carload of chil- dren to Kansas City to see the sights and attend a movie. The group slept on the floor of his mother s living room. In return, the chil- dren painted tribal symbols on a tanned deer skin which they presented to him. During the academic years 1924—1926, he was a student at the University of Kansas at Lawrence. During this time he became inter- ested in fencing and developed proficiency in the sport, winning a medal at one special event and being elected vice-president of the Fencing Club. He joined the DeMolay or- ganization, which led to his joining the Ma- sons. He was active while at Kansas and for his first year in Utah, but, as responsibilities of academic life increased, he gradually became inactive. At the University of Kansas he initially planned to go into engineering but instead switched to zoology, probably being influ- enced by Dr. H. H. Lane, who was the head of that department. Newby served during his senior year as president of the Zoology Club. He focused on mammals, especially rodents, doing some collecting and preparation in the field with follow-up museum curatorial work. He obtained the A. B. in 1926. By this time he had decided he wanted to teach for his liveli- hood and thinking that a master's degree would enhance his chances of obtaining a uni- versity position, he next went to Iowa State College (later Iowa State University) at Ames, Iowa, where he had been offered a teaching assistantship. His major professor there was George O. Hendrickson and his thesis prob- lem pertained to rodents. He completed the recjuired course work and did his research during the regular 1926-1927 academic year and wrote his thesis during the following sum- mer. The M.A. degree was awarded him in 1927. During the spring of 1927, being reason- ably sure that he could finish his work at Ames b\ the end of the summer term, he com- menced looking for a teaching position and received offers from three institutions: Lawrence College in Appleton, Wisconsin; Christian ('oUege, a girl's school in Columbia, Missouri; and the Uni\'ersitv of Utah. He July 1985 Behle; W. W. Newby (1902-1977) 445 chose the latter. For one thing, it was a uni- versity rather than a college. It was also the farthest west of the three institutions and hence closest to the Pacific Ocean. Appar- ently he harbored a latent interest in marine life and a desire to work sometime in a marine laboratory. Fiu'thermore, Utah was situated in the intermountain region, and he had never experienced mountains. Here his ever present curiosity about nature was manifest, but there was also a fortuitous element in- volved. The head of the Department of Zool- ogy at the time was Harold R. Hagan, who was a friend of Dr. Hendrickson at Iowa State. Dr. Hendrickson had taught in Utah at one time. In a letter to him Dr. Hagan appended a note saying "We have an opening. Do you have anybody?" The opening was occasioned by the resignation of David T. Jones. Newby was told of the position and immediately wrote to Dr. Hagan. Before anything was finalized, Hagan became seriously ill and was replaced as de- partment head by Dr. Ralph V. Chamberlin. It happened that Dr. Chamberlin was a friend of Dr. Lane at Kansas, whom Newby had suggested as a reference. Also Chamberlin was searching for a mammalogist, which Newby was at the time. So Newby was se- lected for the position. He arrived on the University of Utah campus about a week be- fore the autumn term started for the 1927-1928 academic year. He rented a room from Professor and Mrs. F. F. Hintze near the campus but went elsewhere for his meals. Dr. Hintze was in the Geology Department, which was then housed along with biology in the Museum Building. Newby's teaching assignment that first year was strenuous, especially for a person with little experience and virtually no time for prior preparation. It consisted of three classes per quarter, and they were large-sized classes. The first term he had two sections of genetics and one of invertebrate zoology. The genetics sections continued through the next two quarters, but the zoology changed. Win- ter quarter he taught comparative anatomy of vertebrates. This was followed spring quarter with a class in ecology. The second year the pressure eased a little, because the schedule was the same. The writer was a member of Dr. Newby's second winter quarter comparative anatomy class in 1928. I was impressed with his youthful appearance, friendly personality, enthusiasm for the subject matter, and his teaching effectiveness. A year or so later I took an advanced genetics class from him. When Dr. Newby arrived on the Univer- sity of Utah campus, the Biology Department occupied the second floor of the building on the lower campus then known as the Museum Building. Later it became the Biology Build- ing, then the North Biology Building (when a "South" Biology Building was built), and fi- nally the James E. Talmage Building of today. The Geology Department was on the ground floor, and the top floor was mostly a large, open, high-ceilinged hall that served as an auditorium for assemblies, lectures, and plays. There was a stage at the east end, and folding chairs were used. Some classrooms at the west end were used mostly by the Psy- chology Department. As the biology area was growing rapidly under Dr. Chamberlin's leadership, he antici- pated that before long more space would be needed. Soon after Newby's arrival. Dr. Chamberlin, knowing of Newby's manual dexterity and shop work experience, asked him one day if he knew how to make blueprints. When Newby replied yes, he was given the assignment of drawing up plans for the division of the open space on the top floor into classrooms, laboratories, and offices. Dr. Chamberlin wanted something tangible to show to President Thomas so as to "sell " the administration on expanding quarters for biol- ogy. This approach was successful and the remodeling was subsequently done according to the plans that Newby drew up, even to the extent of using his blueprints without the fur- ther aid of an architect. Biology got all the space on the top floor except that previously assigned to psychol- ogy. Eventually, biology crowded out both psychology and geology and took over the entire building, which then became known as the Biology Building. Every time after that, when alterations were made on the structure, Newby was consulted since he knew where the bearing walls and other architectural fea- tures such as lowered ceilings were located. Ironically, the top floor was partly opened up again in Newby's later years when certain par- titions were torn out and the two large rooms 446 Great Basin Naturalist Vol. 45, No. 3 at both the east and west ends were converted into large classrooms for teaching by televi- sion. Another project he did soon after his arrival was to make wooden models to illustrate all the changes in chromosomes during the stages of mitosis or cell division. Indeed he made duplicate sets, since there were often several sections of the genetics course. Later he prepared large drawings showing the dif- ferences between regular cell division and the reduction division that occurred in the pro- duction of gametes. The comparable stages were arranged side by side to show the con- trast. He also prepared charts comparing spermatogenesis with oogenesis. These aids were tremendously effective and came at a time when such items could not be easily purchased from biological supply houses. Newby's father died on 6 November 1927, not long after Newby had arrived in Utah. His mother and brother decided to join him out West. He rented an apartment for them, and they made the move just before Christmas. One of the people in the Biology Depart- ment when Newby arrived was Elizabeth Johnson, a student at the university whose home was in Midway, near Heber City. She was working at the time for board and room at the home of Professor and Mrs. Joseph Mer- rill of the Engineering Department. Dr. Dolly Lutjeharms, who was then assistant professor of botany, also hved with the Mer- rills. Through this connection Beth gained employment in the Biology Department do- ing some secretarial work but mostly serving as a reader for examination papers for Dr. Chamberlin's courses in zoology and evolu- tion. (She must have graded some of the ex- aminations of the writer, for I took all of Dr. Chamberlin's courses). Newby's almost daily contact with her led to a romance. She had been addressing him formally as Mr. Newby. One day he suggested that she call him by his middle name, Wal- lace— apparently he didn't care for his first name, William, or the nickname Bill. It seems she had an aversion to the name Wallace, however, so she started calling him Kim, which name stuck. They were married on 7 June 1928 at her home in Midway. They had one daughter, Navee, who specialized in nu- trition at the Universitv of Utah, later mar- ried, and moved east. She is employed as a nutritionist with a government food supplier. Kim retained his youthful appearance all through the years. People who knew him for a long time often remarked how little he changed. This and his always being impeccab- ly dressed made him seem eternally youthful. He especially liked to wear bow ties, and he had a penchant for many styles of shoes. Kim came from a deeply religious Baptist family; his great grandfather had built a hand- some Baptist church in Seymour, Indiana. So Kim affiliated with the First Baptist Church in Salt Lake City. An incidental point, going back to his University of Kansas days, is that in his early years on the University of Utah cam- pus he and Professor Joseph Smith used to fence together. They were the only two on the faculty who knew this sport. Later Kim fenced with a medical student Marcell Marquis, an experienced fencer and organizer of fencing clubs. Kim eventually gave up fencing, again because of his teaching responsibilities. Having arrived at the University of Utah with a master's degree, Newby next set a goal to obtain the doctor of philosophy degree. In keeping with his long-standing desire to see the ocean and study marine life, he decided to attend Stanford University's marine station at Pacific Grove near Monterey, California. In the meantime, Stephen D. Durrant had moved into the field of mammalogy, and Dr. Chamberlin wanted Newby to specialize in experimental biology. The Newbys first went to the marine station in the summer of 1930. Kim bought a Model A Ford and, with Beth and Navee, who was then only about 10 months old, journeyed across western Utah and Nevada on graveled roads, taking three days to reach Pacific Grove. He took three marine invertebrate courses that summer. His destiny turned from the planned exper- imental biolog)' specialization to developmen- tal morphology when he elected to study the early development of a marine worm Urechis caupo, an unpretentious denizen of the mud flats of the intertidal zone. His interest in Urechis was initially aroused by contact with Professor G. E. MacGinitie, who was one of the describcrs of the species, but it was Dr. Harold Heath who pointed out the need for studies on the embrxologv of invertebrates in July 1985 BehlE: W. W. Newby (1902-1977) 447 general and the suitability of the eggs and larval stages of Urechis in particular for such studies. Newby started his research under Professor Heath. His first publication (1932) dealt with the early embryology of this worm. When Heath retired in 1933, Newby contin- ued his work under Professors Douglas M. Whitaker and Tage Skogsberg. Whitaker be- came chairman of Kim's graduate committee. After three summers at Pacific Grove, Kim spent a full academic year, 1933-1934, on the main Stanford campus at Palo Alto continuing his research, taking further course work, pass- ing the qualifying examinations, and starting the writing of his dissertation. He obtained the Ph.D. in 1939. Dr. Whitaker was greatly impressed with Newby s meticulous morphological work and cell lineage studies and envisioned that his research would accentuate the relatively new specialty area in experimental embryology whereby particular cells could be marked at critical stages and the results of many genera- tions of cell divisions subsequently traced with great precision. He stopped off in Salt Lake City one time on his way east to urge Kim to expand his research. Kim did so, even though it delayed completion of the disserta- tion. Later Dr. Whitaker arranged for the results to be published in the memoirs of the prestigious American Philosophical Society (1940). The book was illustrated by 85 of Newby's superb original drawings. With pub- lication of the book, many co-workers in em- bryology wrote to Dr. Newby complimenting him on his fine, meticulous contribution. Among those who did so were such illustrious people as E. G. Conklin, J. Frank Daniels, and E. D. Goldsmith. Dr. Skogsberg was es- pecially lavish in his praise of Kim's work be- ing an original contribution. Kim was men- tioned in the Encyclopedia Britanica in conection with his work on Urechis. Dr. Newby s embryological study con- tributed to taxonomy in that it helped estab- hsh the echiuroid group of worms as a sepa- rate phylum according to some authorities. A sequel paper (1941) was concerned with the development and structure of the slime-net glands of Urechis. Several years later (1946) Dr. Newby made a similar study of the slime glands and thread cells of the hagfish Polistro- trema stouti, one of the Cyclostome or jawless fishes. The material was furnished by Rolf Bolin, an associate at Hopkins Marine Station and a former Utahn, whose father taught physical education at the University of Utah. (Incidentally, Newby's colleague Seville Flowers bought the Bolin residence near the campus). A carryover from the Stanford period was that Newby established in one of the laborato- ries in the Biology Building at the University of Utah an aquarium for marine invertebrates for teaching purposes. He brewed up salt wa- ter with the same constituents as the ocean and obtained sea anemones, starfish, and other marine organisms from Dr. Bolin, who collected them along the Pacific Coast. This effort was not long sustained because of the difficulties of continuously providing the proper environment for species whose habitat was the intertidal zone, especially during the long summers when no classes in invertebrate zoology were given. The momentum and stimulus of his re- search for the doctorate motivated Newby for several more years. For the 1941-1942 aca- demic year he took a sabbatical leave to be- come research associate professor at the Uni- versity of Texas. He worked with Dr. J. T. Patterson and his team studying development in the fruit fly Drosophila. The approach was to first study the development of normal Drosophila larvae as a basis for comparison with larvae of strains possessing structural or biochemical abnormalities. One paper by Newby (1942) soon appeared dealing with in- tersexes produced by a dominant mutation in Drosophila viridis. Some years later a second paper (1949) dealt with abnormal growths on the head oi Drosophila melanogaster. About the time the Newbys arrived in Texas, war clouds were gathering, and the United States soon became actively engaged in World War II. At Texas, Kim was ap- proached by U.S. Army recruiters about tak- ing part in a new high-altitude aviation physi- ology program of research and instruction of pilot trainees. It was hinted that if he signed up his initial rank would be that of captain. Being a patriotic and compulsive individual, Kim enlisted but strangely didn't discuss the matter with his wife. After only one full quarter of resumed teaching at the University of Utah, autumn quarter 1942, Beth was 448 Great Basin Naturalist Vol. 45, No. 3 shocked to receive a telephone message com- ing to the house that Second Lieutenant Newby was to report for duty on a certain date in February 1943, and that his assignment was in the Aviation Corps of the U.S. Army. He obtained a military leave from the Uni- versity and spent the next three years or thereabouts in a new phase of his career in the Army. He had short initial assignments at Randolph Field in Texas and at an air base near Philadelphia, followed by about two years at Santa Ana Army Air Base in Califor- nia. The last year he was stationed at Nellis Air Base at Las Vegas, Nevada, where he was attached to the hospital unit and taught night vision. Also he was an instructor in high-alti- tude physiology in the basic training program, utilizing chambers specially designed for that purpose. He received commendation for the high calibre of his work, his effectiveness as an instructor, and for his many other contribu- tions to the success of the ground training program — and he was promoted to the rank of first lieutenant. While Newby was in the army, the writer taught one of his courses at the University of Utah, namely, vertebrate embryology, which was taken mostly by pre- medical students in the Army Specialized Training Program. Upon Newby's release from the army in September 1945, he once again resumed his teaching duties at the University of Utah. While on military leave he had been advanced to full professor rank. Promotions and salary raises came slowly during the time that Dr. Chamberlin was head of the Biology Depart- ment. In Newby's case he was an instructor from 1927 to 1934 and assistant professor from 1934 to 1939. Having obtained the doctorate in 1939, he was promoted to associate profes- sor that year, which rank he held until 1945 when he became a full professor. It is not clear whether Dr. Chamberlin offered the promo- tion to get Kim to return to Utah or whether Dr. Newby made that a condition of his re- turn. In any event. Dr. Chamberlin visited Kim while he was stationed in Las Vegas. Unfortunately for the science of develop- mental embryology, the long period of mili- tary service interrupted Newby's research and seemingly lessened his desire to do more of it. Furthermore, when he returned to Utah, classes in the postwar period were very large, and numerous sections were held to accommodate the great influx of postwar stu- dents. Hence teaching loads for instructors in the department increased. I recall one quar- ter when Kim had three laboratory sections for his embr\ ology course going at the same time. Although he had teaching assistants for each, he was continuously rotating between them up and down stairs many times an after- noon. Still another factor was that he was now one of an intermediate group of professors on whom committee work rested heavily. Among the more important of his committee assignments was the involvement of serving many years on the Scholarship Standards Committee, including two years as chairman; being on the Committee on Academic Free- dom and Tenure; and serving on the Faculty Council. In addition, he served many years on the Academic Board of the U.S. Navy unit on campus. All were time-consuming and tension- producing assignments carried in addition to a full teaching load plus administrative duties. All this reduced time for research. Being a conscientious individual, he would not shirk other duties to do research so the latter had to suffer. He remarked once that some staff members could teach, serve casually on com- mittees, and still do taxonomic research, but that with his type of research he couldn't and wouldn't. He felt that it was a decision forced upon him that he had to give up research. During the summer of 1949, Dr. Newby taught a course in embryology at Stanford University. Several years after the strenuous postwar interval, another responsibility was thrust upon him in connection with a new advanced placement program designed to identify gifted students in high school and allow them to enter universities even before their gradua- tion from high school. This was a coordinated statewide movement, and Dr. Newby was se- lected to serve as director of advanced place- ment at the University of Utah. Working with Dean Sydney W. Angleman and the general education board, he was responsible for de- veloping a curriculum, counseling students, some as young as 15 years of age, and synchro- nizing the university's program with that of July 1985 BehlE: W. W. Newby (1902-1977) 449 other institutions in the state. He served in this capacity from 1965 to 1969. Another development in which he was vi- tally involved was a reorganization of the Biol- ogy Department. Although the catalog indi- cated for many years that there were two departments in the biology area, botany and zoology, in reality there was but one large Biology Department, with Ralph V. Cham- berlin as head. Most members of the professo- rial staff retired in those days at age 65, but Dr. Chamberlin evidently had an under- standing that he could continue to 68. As his retirement approached, there was consider- able discussion concerning a departmental re- organization. Dr. Newby led one faction, urging reten- tion of a single integrated biology department with a new chairman being brought in from outside the university. Supporting this posi- tion were the writer, Walter P. Cottam, El- den J. Gardner, and several others. Taking a different position were doctors Chamberlin, Rees, Woodbury, and their sup- porters. Their rationale was that the creation of several departments would lead to greater representation on the Faculty Council and to greater funding from the administration. The end result was that the Chamberlin group prevailed, and five departments were formed within a Division of Biology. The de- partments and their chairmen were as follows: Botany, Walter P. Cottam; Invertebrate Zool- ogy, Don M. Rees; Vertebrate Zoology, An- gus M. Woodbury; Genetics and Cytology, W. W. Newby; General Biology, William H. Behle. To establish harmony Dr. Horace Daven- port, chairman in the Department of Physiol- ogy in the Medical School, was persuaded to serve for a time as chairman of the division. The writer, in addition to heading the general biology, was selected to act as executive secre- tary of the division and be the liaison person with Dr. Davenport, and I moved into the former office of Dr. Chamberlin, which was centrally located. The division, with its multiple depart- ments, was an artificial arrangement that seemed to the writer to have been motivated largely to reward several full professors, but it worked for several years because we all wanted it to work and pulled together. In the many long meetings Kim Newby's sound thinking and composure constituted a steady- ing influence. After three years. Dr. Daven- port stepped out, and Dr. Rees became chair- man of the division in addition to being department head of Invertebrate Zoology. The writer could see no need for the position of executive secretary any longer, with every- thing now centrahzed, so that job was aban- doned and Dr. Rees moved into Dr. Cham- berlin s old office. Without going into all the details, eventu- ally a consolidation took place piecemeal over the years, and we were back to one big Biol- ogy Department, the situation that Dr. Newby et al. had argued for in the first place. Dr. Newby served as chairman of the Genet- ics and Cytology Department for 14 years, until 1962. Dr. Vickery then served as chair- man of the Department of Genetics for three years, 1962-1965, before another reorganiza- tion led to a merger of the Department of Genetics with the Department of Experimen- tal Biology, which had been established in the meantime. Another phase of Dr. Newby's career to note was his serving as premedical counselor, succeeding the writer in that position. One of his innovations was the preparation of a "Guide for Premedical Students" (1954a). This he personally published, but it had to be sold through the bookstore in accordance with university regulations. Many years later it was revised and reissued, this time in pamphlet form published by the university (1967). A second such aid was called "Becoming a Doc- tor. " It too started as a mimeographed product circa 1956 and was later published as a pam- phlet by the university (1965a). It covered a broader scope than just the premedical cur- riculum at the university, and the realistic advice contained therein was helpful to par- ents as well as students. For both of these aids he adopted a question and answer format, first posing the question in boldface type, which was then followed by the answer or explana- tion. Dr. Newby enthusiastically carried on this premedical counseling for 18 years until his retirement. As premedical counselor and instructor in the premedical embryology course. Dr. Newby had the burden, as did Durrant, of writing hundreds of letters of rec- ommendation, no small task. His appraisals of 450 Great Basin Naturalist Vol. 45, No. 3 applicants were highly regarded by medical admissions committees throughout the coun- try as well as at the University of Utah. Although Dr. Newby had largely forsaken research for these various new assignments, he continued his affiliation with several pro- fessional societies. He was a member of the American Association for the Advancement of Science, the American Institute of Biological Sciences, the Genetics Society of America, Sigma Xi, Phi Sigma, Phi Kappa Phi, and the Utah Academy of Sciences, Arts, and Letters. He didn't go to many meetings but did attend the Genetics Congress held in Montreal in 1958. Through the years he had one graduate student who obtained the Ph.D. under him and more than 10 others who received the M.S. or M.A. degrees. Even though research was essentially be- hind him and he was overburdened through the years with a heavy teaching load, commit- tee work, and special assignments. Dr. Newby continued to write and be creative. In his teaching, in addition to his carefully pre- pared lectures, he always stressed laboratory work as a practical learning experience. In the laboratory sections of his embryology course in the early years, he used a bulky carbon arc projector to show images on a screen of struc- tures appearing in prepared slides, mostly cross sections of embryos. His routine in the laboratory was to have an initial orientation session using this instrument, followed by the students studying slides with individual mi- croscopes and making drawings of the struc- tures seen. Newby enhanced his teaching by preparing teaching aids such as charts and models and especially laboratory manuals, which had elaborate drawings to depict the structures, organs, and organ systems. Not only did he prepare and publish laboratory manuals for his own classes, but the effort carried over to the general education classes as well. Since there were multiple sections of general biol- ogy being taught by many different instruc- tors, the use of his manuals helped the writer coordinate the coverage by all instructors of certain proscribed subject matter. The mechanism of reduction division as op- posed to normal cell division was particularK difficult for students to visualize. Trouble- some, too, were certain phenomena of hered- ity such as crossing over. Dr. Newby, with the collaboration of a colleague in his department. Dr. George Lefevre, Jr., prepared "An Illus- trated Introduction to Heredity and Develop- ment" (1954b). Portions of this were extracted and published for use in the general biology course. The laboratory manuals that Newby pre- pared undei-went various revisions over a pe- riod of many years under various titles (see 1956a, 1956b,' 1964, 1965b). One deserves additional comment, namely, his "Guide to the Study of Development " (1960). Although this was designed for his course in embiyol- ogy, it was actually a textbook in his special- ized field. Seemingly, an earlier informal mimeographed version of this had appeared as early as 1953. There are some tag ends concerning his writing and bibliography to note in passing. An early paper with student Perry Plummer, who later became a prominent field biologist with the U.S. Forest Service, described a technique for preparing microscopic sections of stems and roots. This was when Newby was teaching a course in microtechnique. A sec- ond item is an abstract (1950) of a paper he gave before the Utah Academy of Sciences discussing the recapitulation theory of devel- opment. Another was a joint paper (1965c) on the embryonic development of the California Gull. Dr. Newby's publications, listed chronologically at the end of this memorial, fall into four categories, namely, reports on his research, his laboratory manuals and sup- plements, the guides for premedical students, and articles dealing with historical aspects of the university, especially pertaining to re- search in the field of biology. Regarding those in the first two categories, his exquisite de- tailed drawings are especially noteworthy. There are some miscellaneous items wor- thy of comment in connection with Dr. Newby's long and distinguished career. He was invited by George Lefevre to teach em- bryology and genetics at Harvard University as a visiting professor during the siunmer of 1963, and he continued to do so during the summers of 1964 and 1965. Bob Vickerv' had preceded Kim in the summer of 1961, teach- ing genetics there. Kim was once asked how students at Har- vard compared with those at lUah. He didn't July 1985 BehlE: W. W. Newby (1902-1977) 451 think they were more highly endowed intel- lectually, but they were "better read" and hence more knowledgeable. He attributed this to a cultural factor wherein reading was more traditional for eastern students, corre- lated with their having more free time to do so. In contrast, many students at Utah had to work while attending the university. Never- theless, he thought that Utah students were better trained in mathematics and biology. Kim was impressed with the Harvard cus- tom of everyone being introduced as Mr. or Mrs. rather than by their academic title of Dr. or Professor. Another surprise was the stand- ing ovation given him at the end of the course in appreciation for his instruction. Newby influenced many students to be- come biologists — and not through proselyt- ing. Rather, their decisions were the subtle result of students desiring to emulate him. The best example is Stephen D. Durrant, who turned from studying languages to zoology and went on to become a reknowned mammalogist. Another Newby quality was his gregarious- ness. A round table in a corner of the Panorama Room Restaurant in the student union was reserved for faculty. A certain group of professors that included Newby reg- ularly met there for lunch. Many of these same people joined others for morning coffee in the faculty lounge, so Kim became conver- sant with university and faculty affairs. He was prominent in the affairs of the Aztec Club, an intellectual and social group that met monthly on campus. Membership consisted mostly of university people, but there were a few from downtown. Kim was vice-president in 196.5-1966 and president in 1966-1967. As a result of all these contacts, he became well known both on and off campus and was called upon many times by the administration for special assignments. For example, to cele- brate the fifteenth anniversary of the founding of the graduate school, a special series of pub- lic lectures by distinguished scholars was ar- ranged, and on 20 February 1961 a sympo- sium on "Graduate Education: The Basis of Our Technical Society" was held. Newby gave one of the papers for this (1961a). It was titled "The Spirit of Research at the University." To commemorate the same occasion, a booklet was published called "The Advancement of Learning: Fifteen Years of Graduate Instruc- tion, Research, and Service at the University of Utah 1946-1961." For this Dr. Newby (1961a) prepared the chapter on the biological sciences. When the old Biology Building on the lower campus was rechristened the James E. Talmage Building, after one of the early presi- dents of the institution. Dr. Newby was asked to tell the history of the building. Indeed, he had earlier been involved in selecting names among the university's presidents for all the buildings on the lower campus in the area that came to be known as the President's Circle. All through the years Dr. Newby was closely associated with Sydney W. Angleman in the general education program, one feature of which was an orientation course for fresh- men entering the university. A syllabus for this course was originally prepared by a com- mittee under the editorship of Dr. Virginia P. Frobes. The third edition of the syllabus was edited by Dr. Newby, assisted by an advisory committee. This entailed an extensive rewrit- ing. The first five chapters were written by Dr. Newby with subsequent modifications by others. The remainder of the syllabus was taken from earlier editions with minor changes. Dr. Newby was a charter member of Uni- versity of Utah chapters for two professional societies. The first was the Alpha Lambda Chapter of Phi Sigma Biological Society. The writer as a senior student happened to be president of the predecessor society when it "went national " and recalls how helpful Kim Newby was at the time of the installation cere- monies. Indeed, he and Beth were staunch supporters through all the years of the chap- ter's existence. The second was the Society of the Sigma Xi. Kim was one of the group of researchers who petitioned that a chapter be established at the University of Utah. He served on numerous committees through the years and one term as president of the local chapter for the 1963-1964 academic year. Two high honors were bestowed on Dr. Newby by the university. In 1966 he was made an honorary alumnus of the College of Medicine in recognition of his many years of teaching premedical students and serving as premedical counselor. Steve Durrant was similarly honored at the same time. In 1971, 452 Great Basin Naturalist Vol. 45, No. 3 the last year of Newby's teaching, he was one of four professors selected by senior students to receive the prestigious Distinguished Teaching Award, as mentioned earlier. Two of many favorable comments made by these stu- dents on nomination forms were that he "makes difficult material clear" and "he is out to teach, not to outguess students. " The award was presented to him at the annual com- mencement exercises, at which time the rank of professor emeritus in biology was conferred upon him. That same year, shortly after his retire- ment, the older group among the members of the staff of the biology department solicited letters of appreciation from his colleagues, the administration, and as many of his former stu- dents as could be contacted. The person largely responsible for this was Bob Vickery. The letters received were then bound and presented to Dr. Newby. It became one of his treasured possessions. A significant sidelight in this connection is that one of the last things that Sid Angleman did on campus was to per- sonally deliver to Bob Vickery his letter of appreciation for the book of letters. The next day Dean Angleman died following heart surgery. His death coming just prior to Kim's retirement affected Kim greatly. Subse- quently, Kim instigated a movement to place a rock bearing a bronze memorial plaque along the south side of the east-west mall on the upper campus near the education building. One of the inherent characteristics of Dr. Newby, as previously noted, was his ability to work with his hands. He shared this skill with his brother Gordon, who taught manual train- ing and art at Highland High School in Salt Lake City for many years and on the side made many artifacts of wood such as cabinets, fireplace fronts, and carved statues. (Inciden- tally, Kim and Gordon looked remarkably alike.) Kim made about 20 small ornamental mahogany tables, 15 of which he gave to friends. Each was carved with a special design appropriate to the scholar. Later he made canes with wooden handles fitted specially to the grip of each recipient. As sort of a sequel to this, dining his six years after retirement, he volunteered many hours to the Marriott Library working on the preservation and restoration of early manuscripts, published books, maps, and prints in the Special Collections Division of the Library. According to Dr. Everett Cooley, curator of this collection and univer- sity archivist, Dr. Newby contributed his own time to learn paper preservation. Indeed, Kim went to Harvard University one summer to learn the techniques at a special workshop. His subsequent work at Utah was invaluable in that he de-acidified and sealed in mylar envelopes hundreds of early maps and about 750 extremely valuable lithographs of Ameri- can Indians from the Edward Curtis collection that had been purchased by the university. This made it possible for material to be han- dled by researchers that before had been too fragile to touch. Kim's work on the preservation of rare and valuable items in the Marriott Library led to his being chosen in 1973 to serve on the board of directors of the Friends of the Library, an organization created to increase university and public awareness of the needs and achievements of the library. He served on the board two terms, during which time he played an important role in developing a significant Friends program. For his service to the li- brary and Beth's volunteer work in cataloging manuscripts, they were both made Honorary Friends of the University of Utah Library. Kim was also active in the Professors Emer- iti Club and served one term as president during the 1973-1974 academic year. Beth also worked for many >'ears as a volunteer in the Salt Lake City Public Library. Dr. Newby was not a robust individual, but he had no physical problems either. Yet the subtle effects of many years of stress had taken its toll. In the spring of 1968 the premedical honor society Alpha Epsilon Delta was invited to hold its 17th annual national meeting at the University of Utah. Dr. Newby and the offi- cers of the local chapter had the responsibility of making local arrangements and planning the program. Just prior to the meeting in the first week of April, Kim suffered a heart attack while at his home. The first action of the con- vention after it convened was to pass unani- mously a resolution thanking Dr. Newby for the excellent arrangements he had made and wishing him a speedy recovery. After a stay in the hosjiital he was brought home on the very day that his colleague Bill F^lowers died from a July 1985 BehlE: W. W. Newby (1902-1977) 453 heart attack. To recuperate, Kim deter- minedly followed a regimen involving walking at least three miles a day. This was often done at the university. He figured out how many times around the landing on the outside of the main floor of the new South Biology building (where he had his office) it took to constitute a mile, and then at noon aroimd and aroiuid he would go until the three miles were covered. He also took up golf and played several times a week. Perhaps as a result of this regimen, he lived for nearly ten more years. The end came suddenly on the evening of 24 March 1977 at the Hotel Utah while he was the guest of Everett Cooley at a dinner meeting of the Timpanogos Club, another intellectual soci- ety of essentially prominent people down- town. After the lecture, during the discussion period, he suddenly slumped foi'ward on the table without a sound. Although there were several doctors in the group, including some who had taken his classes as premedical stu- dents, he couldn't be revived. Thus ended his 74-year life span, his 50-year association with the University of Utah, his 44 years of teach- ing and research, and a long, productive, varied career. Memorial services were held at noon on 28 March 1977, with interment in the City Cemetery in Heber. Dr. Newby was one of a triumvirate of scholars in diverse fields who were recruited by the University of Utah from institutions in other states all about the same time in the late 1920s. The others were Sydney W. Angle- man, who initially taught English literature, and Jacob Geerlings, whose specialty area was Greek and Roman history. The three became fast friends, and their combined influence and academic leadership at the university for roughly four decades is incalculable. Dr. An- gleman became dean of the lower division and thus built up and guided the general educa- tion program for many years, assisted by Newby and others. Dr. Geerlings served as the first dean of the faculty. Dr. Newby be- came head of the Department of Genetics and Cytology and eventually the elder statesman of the biology area. In retrospect, William Wallace Newby was a model of perfection as a teacher. He made highly important contributions in his research in developmental embryology. He was a su- perb illustrator. He was an effective adminis- trator. As a premedical counselor, he gave advice that was timely and realistic. Although not an alumnus of the University of Utah, no one could have been more loyal to or support- ive of the institution. The university and his adopted state suffered a great loss with his passing. I am greatly indebted to Beth Johnson Newby, Everett Cooley, and Robert K. Vick- ery for their help in furnishing information for this memorial. They, along with Gordon L. Newby, reviewed the manuscript. Their sug- gestions are greatly appreciated. Bibliography OF W W. Newby 1932. The early embryology of the echiuroid Ure- chis. Biol. Bull. 6:387-399, 5 pis. 1936. Technique for preparing microscopic sections of woody stems and roots. Bot. Gaz. 98:198-199 (with Perry Plummer). 1940. The embryology of the echiuroid worm Urechis catipo. Amer. Philosoph. Soc. Mem. 16:xvi + 1-219, 85 figs. 1941. The development and structure of the slime-net glands of (7rec/ii.s. J. Morph. 69:303-316. 10 figs. 1942. A study of intersexes produced by a dominant mutation in Drosophila virilis, Blanco stock. In Studies in the genetics oi Drosophila. II. Gene variation and evolution. Univ. Texas Publ. 4228: 11.3-145, 28 figs. 1946. The slime glands and thread cells of the hagfish, Polistrotrema stouti. J. Morph. 78:397-409, 8 figs. 1948. Histological techniques applied to insects (Ab- stract). Utah Acad. Sci., Arts, Lett. Proc. 25:16.5-166. 1949. Abnormal growths on the head of Drosophila mehinogaster. J. Morph. 85:177-196, 33 figs. 1950. The present position of the recapitulation theory (Abstract). Utah Acad. Sci., Arts, Lett. Proc. 27:78. 19.54a. A guide for premedical students. 16 pp. Univer- sity of Utah Press. 1954b. An illustrated introduction to heredity and devel- opment (with George Lefevre, Jr.). University of Utah Press. 31 unnumbered pp., 12 pis. I9.56a. The biology of heredity: an illustrated outline (with George Lefevre, Jr.). Univ. Utah Press: 1-91, 14 pis. with numerous figs. 1956b. Genetics supplement for Biology 2 (with George Lefevre, Jr.). [An illustrated outline abridged from The biology of heredity]: 1-47 pp. 1960a. A guide to the study of development. W. B. Saun- ders, Philadelphia, xi + 1-217, many figs. 1960b. [Editor of| Syllabus for general education. Uni- versity of Utah, Salt Lake City, 93 pp. 454 Great Basin Naturalist Vol. 45, No. 3 1961a. The spirit of research at the University of Utah. In Graduate school: the basis of our technical cul- ture. [A lecture delivered on 20 February 1961 as part of the University of Utah and the Advance- ment of Learning series, in recognition of achievements in graduate instruction, research, and service at the University of Utah during the years 194^1961]: 10-13. 1961b. The biological sciences. In The advancement of learning; fifteen years of graduate instruction, re- search, and service at the University of Utah 1946-1961: 49-63. University of Utah Press. 1964. Genetics supplement for biology; an illustrated outline (with George Lefevre, Jr.) [revised edi tion] Universitv of Utah Printing Service: 1-5.5, many figs. 1965a. Becoming a doctor. University of Utah Printing Service. 16 pp. 196.5b. The biology of heredity; an illustrated outline (with George Lefevre, Jr.) [revised edition]. Uni- versity of Utah Printing Service; 1-20. 1965c. The embryonic development of the California Gull {Larits californicus) with B. A. Dawkins and W. H. Behle). Univ. Utah Biol. Ser. 13(2):l-26. 1967. A guide for premedical students. University of Utah Printing Service: 20 pp. SYMBOS CAVIFRONS (ARTIODACTYLA: BOVIDAE) FROM DELTA COUNTY, COLORADO Jerry N. McDonald' Abstract. — A partial cranium belonging to the extinct woodland musk-ox, Symbos cavifrons, is reported from the headwaters of Oak Creek, Delta County, Colorado. This is the first cranium of the genus Symbos to be described from the Colorado Plateau, and it helps to define the southwest boundary of the known range of the genus. The extinct musk-ox genus Symbos was dis- tributed widely across much of North America during the late Quaternary, ranging from Alaska to Louisiana and from California to the Atlantic continental shelf off New Jersey and Virginia (Kurten and Anderson 1980, J. N. McDonald, unpubl. data, C. E. Ray, unpubl. data). Records o( Symbos from the Basin and Range Province, however, are rare except from the pluvial Lake Bonneville region of Utah and its discharge area in southeastern Idaho. Indeed, all 16 cranial specimens from the Basin and Range identified as Symbos in the literature are from this area (Gazin 1935, Nelson and Madsen 1978). At least 10 other cranial specimens are known from Basin and Range localities, including: Minidoka County (4) and one unknown location (1) in Idaho; Modoc County (1), California; Delta (1) and Montezuma (1) counties, Colorado; Wasco County (1), Oregon, and Whitman County (1), Washington (S. W. Neusius, written comm., 12 January 1984; J. A. White, oral comm., 9 May 1984; J. N. McDonald, un- publ. data; C. E. Ray, unpubl. data). In addi- tion to specimens positively identified as Sym- bos, several other ovibovine specimens are also known from the Basin and Range, includ- ing cranial records assigned to the low-horned genera Bootheriiwi and Gidleya as well as numerous postcranial, facial, mandibular, and dental remains that cannot yet be identified with confidence to the genus level (Gidley 1906, Cossmann 1907, Allen 1913, Gazin 1935, Nelson and Madsen 1978, Kurten and Anderson 1980, J. N. McDonald, unpubl. data, C. E. Ray, unpubl. data). A partial skull treated here as Symbos cav- ifrons, from Delta County, Colorado, is of special interest because (1) it represents the first record of Symbos from the Colorado Plateau, and (2) it falls upon, and thus helps to define, the southwestern limits of the docu- mented range of the genus (Fig. 1). This speci- men was donated to the Field Museum of Natural History, Chicago, Illinois, in August 1946 by Alfred A. Look of Grand Junction, Colorado. The following notes made by Bryan Patterson upon receipt of the specimen at the museum provide some relevant historical details: In July, 1946, Mr. Alfred A. Look of Grand Junction, Colorado, to whom Paleontology is indebted for the dis- covery or preservation of a number of interesting fossil vertebrates, sent in for determination an incomplete cra- nium of an extinct musk-ox. This specimen had been turned over to him by the members of a fishing party who had found it on the south side of Grand Mesa, near the rim, at an elevation of some 9000 feet. Mr. Look later visited the spot, which he informs me is about 200 yards downstream on Oak Creek from the dam face of Davey Reservoir. This places it approximately in Section 17, Twp 13 S, R 96 W, Delta Co. The fossil was evidently derived from a pit south of the dam out of which sand and clay had been taken for construction purposes. He was unsuccessful in finding any additional remains. Thanks to the original finders and to Mr. Look, the specimen has been presented to this Museum and now bears number PM . The deposits from which the fossil appears to have come is almost surely post-Wisconsin. Henderson (1923) has described glacial tills on the top of Grand Mesa that he attributed to the Wisconsin stage. (See also Flint et al., 1945); it is likely that the sediments observed by Mr. Look in the pit just mentioned were derived from this source. (Notes attached to letter: B. Patterson to C. E. Ray, Nov. 4, 1968). Department of Geography, Radford University, Radford, Virginia 24142. 455 456 Great Basin Naturalist Vol. 45, No. 3 E .^^~^ Symbos records: •published o unpublished X Possible provenience of PM526 — -■ Basin and Range Province boundary y^^^/y^^ Colorado Plateau Fig. I. Records oi' Stptihos from the Basin and Range Province of the western United States, and the provenience of PM 526 near Grand Mesa, Colorado. The insert is from the Hells Kitchen, Colorado, Quadrangle, USGS 7.5' series, 1965 edition. July 1985 McDonald; Fossil Musk-Ox 457 Later, Patterson wrote in a foreword to Look s book on the Cok)rado Pkiteau that Look "has foHowed leads that have taken him hunting a fossil musk-ox in the pines and spruces at ten thousand feet on Grand Mesa" (Look, 1955: ix). The specimen is cataloged as PM 526. The museum records that put the place of discov- ery in Section 17, T 13 S, R 96 W are, how- ever, incorrect. Section 17 lies about 10,000 ft atop Grand Mesa, > l| miles west of the "Davey" (actually Davies, now Porter) reser- voirs, outside the Oak Creek watershed (Grand Mesa National Forest, Colorado, Map, USFS, 1959 ed.; Indian Point, Colo- rado, Quadrangle, USGS 7.5' series, 1965 ed.; L. Brooks, oral comm., 9 May 1984). Porter Reservoir 4 (formerly Little Davies Reservoir) is located in Section 15, just below 9400 ft elevation, at the base of the southeast- facing scarpment of Grand Mesa (Fig. 1). Porter Reservoir 1 (formerly Davies Reser- voir) lies at about 9100 ft elevation in Section 14. The segment of Oak Creek below Porter 4 flows southeasterly through forest, whereas the first half-mile of Oak Creek below Porter 1 flows northeasterly through a treeless region. Porter 1 is nearer the elevation of discovery as recorded in Patterson's notes. Available infor- mation does not permit an unequivocal deter- mination of the provenience of this specimen, yet Patterson's notes and the battered condi- tion of the specimen indicate that it came from below, not on. Grand Mesa. The provenience should be given, therefore, as along or near Oak Creek, Section 14 or 15, T 13 S, R 96 W, Hells Kitchen, Colorado, Quadrangle, USGS 7.5' series (Fig. 1). The geographic coordi- nates of this area are approximately 38°55'30" N, 108°6' W; it hes within the 1-km scjuare formed by UTM coordinates 750,500 E, 4,312,000 N, 12, N. Symbos cavifrons is considered to have been extant from the late Irvingtonian to the Wisconsin-Holocene boundary (i.e., from ca. > 500,000 y.a.-10,000 y.a.) (Kurten and An- derson 1980). Description of the Specimen This specimen consists of most of the cra- nium caudal to the orbits; it is badly abraded and weathered and is stained by iron. Most of the horn cores are missing, as are most of the left and part of the right occipital condyles, the jugular processes, the lateral part of the right half of the basioccipital, the rostral part of the presphenoid, the zygomatic processes, and the external occipital process, the nuchal line, and the temporal crest. The pattern of bone loss found in this specimen suggests that it has been battered by heavy objects and "rounded," such as would probably occur to skulls subjected to stream transport (Haring- ton 1968, 1975, Shipman 1981). The dorsal surface of the parietals and re- maining frontals is concave transversely. The depression occupying the intercornual region is oval, with the greatest breadth being near the caudal edge of the horn cores. Although the intercornual surface shows abrasive dam- age, it appears to have been completely cov- ered with exostoses associated with the spread of the keratinous bases of the horn sheaths into the intercornual region. There is no evi- dence that a longitudinal crest and trough of exostoses existed on the median plane. Some of the frontal sinuses are exposed, probably a result of abrasion of the dorsal surface (Fig. 2). The bases and proximal parts of both horn cores remain. The dorsal surface of the horn core bases are relatively flat (i.e., they are not concave) rostrocaudally, but this flatness could have been exaggerated by abrasion. The horn core bases are convex on the ventral surface where — especially on the right core — prominent longitudinal grooves are evident (Fig. 3). The transverse breadth of the cra- nium at the level of the frontal sinuses is greater than the breadth at the temporal fos- sae (Fig. 4). The frontoparietotemporal suture is roughly parallel to the dorsal surface of the cranium (Fig. 3). The caudal surface of the cranium is abraded and so badly weathered that most of the natural bone surface is missing (Fig. 5). Part of the natural surface of the right half of the occipital remains, and the deeper reaches of the right insertion for M. semispinalis capi- tis is evident. The breadth of the caudal end of the basioccipital, if complete, would be ap- proximately one-half that of the complete oc- cipital condyles. The lateral edges of the ba- sioccipital trend gradually, then abruptly, toward the median line, producing a V or shield shape with the apex directed rostrad. A 458 Great Basin Naturalist Vol. 45, No. 3 Fig. 2. Dorsal surface, PM 526. Rostral direction toward top. Line represents 10 ci Fig. 3. Lateral surface, right side, PM .526. Rostral direction toward right. Line represents 10 ci July 1985 McDonald: Fossil Musk-Ox 459 Fig. 4. Rostral view, PM 526. Line represents 10 ( Fig. 5. Caudal surface, PM 526. Line represents 10 cm. 460 Great Basin Naturalist Vol. 45, No. 3 Fig. 6. Ventral surface, PM 526. Ro.stral direction toward top. Line represents 10 cm. median groove divides the caudal half of the basioccipital into right and left halve.s (Fig. 6). Measurements of selected characters are given in Table 1. Identity of the Specimen PM 526 is referred to Symhos (Osgood 1905a, 1905b) on the basis of the concave, exostosis-covered dorsal surface of the pari- etals and the remaining frontals, the deep frontal sinuses, and the morphology of the ventral surface of the basioccipital, including (a) the presence of a median groove, (b) the V or shield shape resulting from the conver- gence of its lateral edges rostrally toward the median plane, and (c) the relatively great transverse breadth of the caudal part, being approximately one-half the transverse breadth of the occipital condyles (Leidy 1852, Osgood 1905a, Harington 1968, 1975, J. N. McDonald, unpubl. notes). Table L Cranial measurements (mm). Greatest breadth, occipital condyles Greatest breadth, occipital condyles with auxiliary articular surface Greatest breadth of basioccipital Least breadth of parietals Least breadth, dorsal surface of frontals Least breadth of Irontals at the ventral base of horn cori's CIreatest (rostrocaudal) length, horn core base Greatest dorsoventral diameter, horn core base Greatest circumference of horn core 137.2 {156.3r (76. 2r 1.32.4 1,56.0 160.9 H-llI.y L- 105.4 R L- 66.2 R L-268.0 * estimates. July 1985 McDonald: Fossil Musk Ox 461 Symhos is considered informally to he a monospecific genus containing only Sipnbos cavifrons (Kurten and Anderson 1980), to which species this specimen is referred on the hasis of its horn core, dorsal cranial, and ha- sioccipital morphology and fronto- parietotemporal suture pattern. The differ- ence in size between the type specimen of Symbos cavifrons (Leidy 1852, Osgood 1905a) and PM 526 is slight. The characters of PM 526 differ from those of the type specimen of Symbos cavifrons (Leidy 1852, Osgood 1905a) and other referred specimens only in minor details, none of which seem to be greater than should be expected of individual variation within a representative population sample. Acknowledgments I thank William Turnbull (Field Museum of Natural History) for permission to study PM 526 and Len Brooks (U.S. Forest Service), Clayton E. Ray (National Museum of Natural History), and John A. White (Idaho Museum of Natural History) for providing unpublished information for use in this paper. I also thank Susan L. Woodward for preparing Figure 1. Literature Cited Allen, J. A. 1913. Ontogenetic and other variations in musko.\en with a systematic review of the muskox group, recent and extinct. Mem. Amer. Mus. Nat. Hist.,n.s., 1(4): 10.3-226. CossMANN, M 1907. (Proposal to replace Liops with Gtd- lei/a.) Rev. (>riti(jue Palcozoologie 11th year (1):64. Flint, R F., etal. 1945. Glacial map of North America. Geol. Soc. Amer. Spec. Paper 60. Map and 37 pp. Gazin, G. L 1935. Annotated list of Pleistocene mam- malia from American Falls, Idaho. J. Washington Acad. Sci. 25:297-302. GiDLEY, J. W. 1906. A new ruminant from the Pleistocene of New Mexico. Proc. U.S. Natl. Mus. 30(1447): 16,5-167. Harington, C. R. 1968. A Pleistocene muskox (Symhos) from Dease Lake, British Columbia. Canadian J. Earth Sci. 5:1161-1165. 1975. Pleistocene muskoxen (Symbos) from Al- berta and British Columbia. Canadian J. Earth Sci. 12:90,3-919. Henderson, J. 1923. The glacial geology of Grand Mesa, Colorado. J. Geol. 31:676-678. Kurten, B., and E. Anderson 1980. Pleistocene mam- mals of North America. Columbia Univ. Press, New York. 442 pp. Leidy, J 1852. Memoir on the extinct species of American ox. Smithsonian Contrib. Knowl. 5, art. 3. 20 pp. Look, A. A. 19,55. 1,000 million years on the Colorado Plateau, land of uranium. Bell Publications, Den- ver, Co. 344 pp. Nelson, M E., and J H Madsen, Jr. 1978. Late Pleis- tocene musk oxen from Utah. Trans. Kans. Acad. Sci. 81:277-295. Osgood, W H. 1905a. Scaphoceros tyrelli, an extinct ruminant from the Klondike gravels. Smithsonian Misc. Coll. 48(1,589): 173-185. 1905b. Symbos, a substitute for Scaphoceros. Proc. Biol. Soc. Washington 18:223-224. Shipman, p. 1981. Life history of a fossil, an introduction to taphonomy and paleoecology. Harvard Univer- sity Press, Cambridge. 222 pp. COMPARISONS OF PRESCRIBED BURNING AND CUTTING OF UTAH MARSH PLANTS Loren M. Smith' " and John A. Kadlec' Abstract. — The efficacy of cutting (haying) versus burning was compared for control of marsh vegetation in Utah. Cutting reduced production of hardstem bulrush (Scirpus lacustris), alkali bulrush (S. maritimus), and cattail {Typha spp.) compared to levels found on burned plots, but differences were significant (P < 0.05) only within the alkali bulrush vegetation type. Clipping saltgrass (DisticJilis spicata) plots greatly reduced production upon reflooding, which produced results similar to prescribed burning and reflooding. Heat penetration into the sediments during the fire was not sufficient to cause substantial belowground mortality. Without Ijelowground mortality, prescribed burning alone did not change aboveground production or species composition. Flooding after fire did eliminate saltgrass, but a single prescribed burning or cutting was not an effective management tool for reducing production of cattail, hardstem bulrush, and alkali bulrush. Dense emergent marsh plants often restrict nesting and loafing by waterfowl. Wetland managers attempt to reduce the density of emergent marsh plants with prescribed burn- ing or cutting (haying) (Nelson and Dietz 1966, Linde 1969). Smith and Kadlec (1985a) detailed the effects of fire on cattail (Typha spp.), saltgrass (Distichlis spicata), hardstem bulrush (Scirpus lacustris), and alkali bulrush (S. maritimus). In this study the objective was to compare the effects of cutting relative to fire on primary production of these four vege- tation types and to assess the impact of heat penetration into the soil on aboveground plant response. Although some data exist for cattail (Nelson and Dietz 1966), little has been published regarding heat penetration and the cutting of the other vegetation types. Study Area The study was conducted at Ogden Bay Waterfowl Management Area, Weber County, Utah. The area was established in the 1930s and 1940s when dikes were constructed along the shore of the Great Salt Lake that entrapped freshwater from the Weber River (Nelson 1954). Four species dominated the emergent vegetation: saltgrass, alkali bul- rush, hardstem bulrush, and cattail (Smith and Kadlec 1983). Botanical nomenclature fol- lows Cronquist et al. (1977) and Scoggan (1978). Unit 1 of Ogden Bay (see Nelson 1954 for detailed description) was drained in April 1981. Portions of the area were burned 2 Sep- tember 1981. The fire removed all above- ground plant material. There was a wind speed of 16.6 km with a mean dew point of 5 C and a maximum temperature of 28.5 C on the day of the fire (Smith et al. 1984). Methods Annual production was determined in burned and unburned sites for cattail, alkali bulrush, hardstem bulrush, and saltgrass communities. Five quadrats in each vegeta- tion type within the unburned area were clipped at the substrate level and raked when the other plots were burned. Quadrat size was 1.0 X 0.5 m in cattail, alkali bulrush, and hardstem bulrush sites and only 0.25 X 0.25 m within saltgrass sites due to its smaller size and high shoot density (see below). Five quadrats in each vegetation type were also established within the burned area. Standing crops were estimated six times (May-August) during 1982 using length-mass regressions on tagged shoots (Smith and Kadlec 1985b) of alkali bulrush, hardstem bul- rush, and cattail. Because water depths influ- ence plant recovery, depths were also recorded throughout the season in burned and mowed plots. Total annual production Department oi Fisheries and U'ildlil -Present address: Department of Ran lltah State and Uildlil Utah 84322. ■xasTeehUnr 462 July 1985 Smith, Kadlec: Marsfi Ecology 463 Bulrush 15200- 1 24''C - Soil surface -MT'C 204°C 48''C- centimeters Fig. 1 . Schematic of asbestos board illustrating spacing of depressions that contained waxes with a specific melt- ing point. Boards were used to indicate temperature gradient within the soil during prescribed burning at Ogden Bay Waterfowl Management Area, Utah, 1981. and mowed plots. Total annual production was calculated by incorporating mortality of tagged shoots. Because of the dense nature of saltgrass (2000 shoots/m"), seasonal estimates of biomass were not made and total produc- tion was estimated using clipped plots at peak standing biomass. Biomass was also estimated for the proposed burned and unburned areas for all species prior to treatment. If pretreat- ment estimates of biomass were different among proposed treatment sites, pretreat- ment biomass was used as a covariate in subse- quent analyses of variance (Steel and Torrie 1980). Data were square root transformed ac- cording to Zar (1974) to conform with normal variate assumptions. Heat penetration into the soil during the prescribed burn was estimated by using 7 as- bestos strips (Fig. 1) within each vegetation type. Each strip had waxes with 8 different melting points. Tops of the strips were placed flush with the soil surface prior to the fire (Shearer 1975). Temperature profile in the soil during the fire was reconstructed by lin- early measuring the depth to which the vari- ous waxes had melted. Mortality of roots and rhizomes was estimated by taking 20 ran- domly located soil cores (5.0 X 15.2 cm) within each vegetation type immediately prior to and after the prescribed burn. The number of hving roots and rhizomes was Ma y July August Fig. 2. Comparisons of standing crops among burning and clipping treatments within cattail, hardstem bulrush, and alkali bulrush vegetation types (P = 0.05; LSD = 14.2, 15.1, 6.0, respectively) at Ogden Bay Waterfowl Management Area, Utah. number of living roots and rhizomes was counted at 4 depths: 0-3.8, 3.8-7.6, 7.6-11.4, and 11.4-15.2 cm. A solution of orthotohdine (C6H3-4NH2-3-CH3)2 in 95% methanol and a 3% hydrogen peroxide (H^O,) solution (Hare 1965, Shearer 1975) were sprayed on the sur- face of the flattened core. Living material re- acted by producing a blue color, but dead material did not change in color. Results Smith and Kadlec (1985a) demonstrated that burning did not affect the subsequent annual production of hardstem bulrush, alkali bulrush, or cattail, but saltgrass was virtually eliminated as a result of flooding following fire. In this study, standing crop biomass on clipped plots was consistently less than on burned plots (Fig. 2), but the difference was significant (F < 0.05) only within the alkaU 464 Great Basin Naturalist Vol. 45, No. 3 bulrush vegetation type. Total annual produc- tion results also indicated a significant (F < 0.05) reduction in clipped versus burned plots (Table 1) only within the alkali bulrush sites. Within saltgrass sites cutting was similar (P > 0.05) to prescribed burning in that little vege- tative biomass was produced following clip- ping and reflooding. Water depths following the prescribed burn were not different (F > 0.15) among burned and clipped plots, re- spectively, for saltgrass (8.0, 10.2 cm), cattail (22.2, 16.8 cm), hardstem bulrush (14.2, 15.0 cm), and alkali bulrush (16.8, 16.4 cm). There- fore, water levels in the two areas were not a factor in production differences. Heat penetration into the soil (Table 2) suf- ficient to kill (Shearer 1975) plant tissue (60 C) was minimal but greatest within saltgrass and alkali bulrush vegetation types. Tempera- tures within the 48-69 C range were never deeper than 3.8 cm. In cattail, the number of living roots and rhizomes decreased after the prescribed burn at all 4 depths (Table 3; F < 0.05). There was no decrease (F > 0.20) in the upper 3.8 cm of sediment in hardstem bulrush sites; however, differences were evident (F < 0.05) in the three deepest zones. Within alkali bulrush sites differences existed (F < 0.05) in the num- ber of living roots and rhizomes for the three most shallow categories but not (F > 0. 10) in Table 1. Mean annual production (g m~" yr ') of 4 vegetation types under clipping and burning treatments at Ogden Bay Waterfowl Management Area, Utah. Vegetation type Burned Clipped Distichlis spicata " 89(122r 268(179) Scirpus lacustris 1559(811) 920(612) Scirpus maritimus 849(229) 319(129) Typha spp. 1173(1161) 979(616) ■"Peak standing biomass ''Not adjusted for cosariate 'Standard deviation in parentheses the 11.4-15.2 cm range. Saltgrass sites also did not show a difference (F > 0.20) in the number of roots and rhizomes in the upper 3.8 cm, but differences were evident (F < 0.05) in the three deepest categories. Discussion Data on rhizome mortality and heat pene- tration provided insight into the effects of pre- scribed burning on aboveground vegetative- reproduction. Cattail and alkali bulrush sites were the only vegetation types where differ- ences in the number of living roots and rhi- zomes occurred in the upper 3.8 cm of soil (Table 3). This zone is where temperature during the fire was greatest (Table 2) and sug- gests that the prescribed burn may have caused belowground plant mortality in these two cases. However, most vegetation types Table 2. Mean heat penetration (cm) into the soil, within 4 vegetation types, during a prescribed burn at Ogden Bay Waterfowl Management Area, Utah, 1981. Temperature (C) Vegetation type 104 177 204 Distichlis spicata 2.21(0.63)' 1.56(0.51) 1.01(0.24) 0.77(0.37) 0.60(0.45) 0.29(0.23) 0.13(0.11) 0.07(0.10) Scirpus lacustris 1.91(0.83) 1.33(0.72) 0.89(0.66) 0.63(0.57) 0.49(0.55) 0.29(0.36) 0.13(0.19) 0.03(0.08) Scirpus maritimus 2.34(0.70) 1.84(0.61) 0.93(0.30) 0.69(0.29) 0.37(0.21) 0.13(0.11) 0.06(0.05) 0.03(0.05) Typha spp. 1.50(0.85) 1.01(0.65) 0.64(0.52) 0.43(0.46) 0.28(0.29) 0.13(0.18) 0.08(0.12) 0.04(0.07) "Standard deviation in parentheses Table 3. Mean number of living roots and rhizomes at four depths in foiu" vegetation t\ pes prior to and after a prescribed burn at Ogden Bay Waterfowl Management Area, Utah, 1981. Depth range (cm) Prefire Postfire Vegetation type 0.0-3.8 3.8-7.( 7.6-11.4 11.4-15.2 0.0-3. 3.8-7. 7.&-11.4 11.4-15.2 1.7(1.1) 1.1(1.0) 0.7(0.7) 0.5(0.8) 0.8(0.8) 0.2(0.5) 0.9(0.5) 0.4(0.5) Distichlis spicata Scirpus lacustris Scirpus maritimus Typha spp. 2.8(1.4)^ 0.8(0.6) 2.4(1.1) 1.5(1.0) 2.9(1.4) 1.3(0.9) 2.0(0.7) 1.6(0.7) 3.0(1.6) 1.7(0.9) 1.3(0.8) 1.3(0.9) 2.2(1.6) 1.4(0.7) 0.6(0.8) 1.1(0.8) 2.3(1.3) 0.6(0.9) 1.0(0.9) 0.8(0.7) 2.0(1.1) 0.7(0.7) 1.4(1.0) 0.8(0.6) "Standard deviation in parentheses July 1985 Smith, Kadlec: Marsh Ecology 465 had decreases in the number of hving roots and rhizomes in the three deepest sediment categories, where lethal heat penetration into the soil was not common. Mortality that oc- curred at these depths therefore should be regarded as non-fire-induced, and perhaps that was also true in more shallow layers. Therefore, decreases in the number of roots and rhizomes in the three deepest categories (3.8-15.2 cm) were natural or resulted from the drawdown preceding the fire, which caused decreased soil moistures and in- creased salinities. (Smith and Kadlec 1983). Valiela et al. (1976) noted that biomass of cord- grass (Spartina spp.) roots peaked in midsum- mer and declined through autumn. At Farm- ington Bay Waterfowl Management Area, Utah, Anderson (1977) found that, under the stress of increased salinity, the growth of a fungus {Chaetophoma confliiens) was encour- aged on the surface of cattail rhizomes. Ander- son noted the fungus was consistently associ- ated with rhizomes of declining plants. Because the aboveground production of cat- tail and alkali bulrush was not reduced follow- ing fire (Smith and Kadlec 1985a), below- ground decreases were apparently un- important in terms of potential control of these vegetation types. In contrast, the de- crease of saltgrass biomass was not due to fire per se because there was no decrease in living roots and rhizomes. Saltgrass decreases there- fore were likely due to fire and reflooding acting together. Faulkner and de la Cruz (1982) found that both increased soil temperature and slight in- creases in sediment nutrients (from fire residue) may have aided vegetation recovery and resulted in increased nutrient levels found in plant tissues after fire. At Ogden Bay, biomass of clipped plots were consistently lower than burned plots (Fig. 2 ), suggesting that nutrients released by the fire may have speeded vegetation recovery. De la Cruz and Hackney (1980) compared clipped and burned plots within a cordgrass marsh and stated, "Presumably, nutrients left by the ash enhanced the growth of plants in the burned area." Nitrogen levels of the postburn vegeta- tion increased in some species at Ogden Bay (Smith et al. 1984), which suggests a possible nutrient effect. An alternative hypothesis is that fire caused enough rhizome mortality to reduce intraspecific competition and allow in- creased vigor in individual shoots, such as cattail at Ogden Bay (Smith and Kadlec 1985a), which resulted in vegetative recovery. Conclusions Controlled burning has been used with lim- ited success to control dense and undesirable plants such as cattail, Phragmites, and cord- grass (Lay 1945, Nelson and Dietz 1966, Ward 1968). Reasons cited for poor control include lack of proper water levels and little below- ground mortality. Nelson and Dietz (1966) reported that cattail could be controlled if the area was flooded to greater than 45 cm imme- diately after fire. Water levels of that depth are often difficult to attain. In this study sedi- ments were dry (Smith and Kadlec 1983) and heat penetration into the sediments was slight, resulting in little belowground plant mortality. Sediments of higher organic con- tent might allow greater belowground mortal- ity and better overall control. Control of salt- grass through burning or cutting and subsequent reflooding was effective. How- ever, saltgrass is usually considered as good nesting cover for waterfowl and seldom is con- trol warranted. Mowing (haying) may be more effective, albeit probably temporary, than a single fire for control of emergent macrophytes, which may be due to the availability of ash (nutri- ents) on burn plots. Clipped plots did not receive this pulse of nutrients. Also, rather than using vegetative removal methods to reduce cattail, salinity manipula- tions may promote more desirable species. Dropping water levels may lead to increased salinity (Smith and Kadlec 1983) favoring al- kali bulrush. Nelson and Dietz (1966) found that cattail plots dried for two years (thereby increasing salinity) and mowed often showed increases in alkali bulrush. Acknowledgments M. Wolfe, F. Lindzey, M. Barkworth, and D. Sisson provided comments on project de- sign and drafts of the manuscript. We thank J. Smith, K. Reese, M. Hadfield, B. Weber, V. Bachman, D. Irving, and N. Nelson for field 466 Great Basin Naturalist Vol. 45, No. 3 assistance. The Utah Division of Wildhfe Re- sources, Utah Cooperative Wildhfe Research Unit, Utah State University Ecology Center, Texas Tech University (Paper T-9-403, Col- lege of Agricultural Sciences), and the Fed- eral Aid to Wildlife Restoration Project pro- vided financial and logistic support. Literature Cited Anderson, C M 1977. Cattail decline at Farmington Bay Waterfowl Management Area. Great Basin Nat. 37:24-34. Cronouist. A , A H Holmgren, N H. Holmgren, J. L Reveal, and P K. Holmgren. 1977. Intermoun- tain Flora. Vol. 6. Columbia University Press, New York. DE LA Cruz, A. A., andC. T. Hackney. 1980. The effects of winter fire and harvest on the vegetational struc- ture and primary productivity of two tidal marsh communities in Mississippi. Mississippi-Alabama Sea Grant Consort. M-AS6P-80-013. Ocean Springs, Mississippi. Faulkner, S. P., and A. A. de la Cruz. 1982. Nutrient mobilization following winter fires in an irregu- larly flooded marsh. J. Environ. Qual. 11; 120-133. Hare, R C. 1965. Chemical test for fire damage. J. For. 63:939. Lay, D. W. 1945. Muskrat investigations in Te.xas. J. Wildl. Manage. 9:56-76. LiNDE, A. F. 1969. Techniques for wetland management. Wisconsin Dept. Nat. Resour. Res. Rep. 45. Nelson. N. F 1954. Factors in the development and restoration of waterfowl habitat at Ogden Bay Refuge, Weber County, Utah. Utah Dept. Fish Game Publ. 6. Nelson, N. F., and R. A. Dietz 1966. Cattail control methods in Utah. Utah Dept. Fish Game Publ. 66-2. ScogganH. J. 1978. The flora of Canada. Nat. Mus. Nat. Sci. Publ. Bot. 7. Winnipeg, Manitoba. Shearer, R. C. 1975. Seedbed characteristics in western larch forests after prescribed burning. Intermoun- tain Forest Range Ex. Sta. Res. Pap. 167. Ogden, Utah. Smith, L. M., and J. A Kadlec. 1983. Seed banks and their role during drawdown of a North American marsh. J. Appl. Ecol. 20:67.3-684. 1985a. Fire and herbivory in a Great Salt Lake marsh. Ecology. 66:259-265. 1985b. Comparisons of marsh plant loss estimates in production techniques. Amer. Midi. Nat. In press. Smith, L M , J A Kadlec, and P. V Fonnesbeck 1984. Effects of prescribed burning on the nutritive quality of marsh plants in LHah. J. Wildl. Manage. 48:285-288. Steel, R G D, and J H Torrie 1980. Principles and procedures of statistics: a biometrical approach. McGraw-Hill, New York. Valiela, 1., J. M. Teal, and N. Y. Persson. 1976. Produc- tion and dynamics of experimentally enriched salt- marsh vegetation: belowground biomass. Limnol. Oceangr. 21:245-252. Ward, P 1968. Fire in relation to waterfowl habitat of Delta marshes. Tall Timbers Fire Ecol. Conf. 8:255-267, Zar, J. H. 1974. Biostatistical analysis. Prentice-Hall, New Jersey. NEW SPECIES AND RECORDS OF NORTH AMERICAN PITYOPHTHORUS (COLEOPTERA: SCOLYTIDAE), PART IV: THE SCRIPTOR GROUP D. E. Bright' Abstract — Six new species ofPityoplithorus from Mexico are described. The new species, all in the Scriptor group, are: atkinsoni, diminutivus, equihtiai, thamnus, trunculus and zexmenivora. An additional locality record for P. coronarius Blackman is included. A new key to the 22 species included in the Scriptor group is presented. Recently, Dr. T. H. Atkinson (Colegio de Postgraduados, Centro de Entomologia Y Acarologia, Chapingo, Mexico) sent a large collection of Pityophthonis collected by him- self and his students to me for identification. A number of samples were collected from plants not previously known to harbor scolytids and are thus a rich source of previously unknown and unsuspected host relationships. A prelim- inary count indicated that almost 23 un- described species were included in the collection. Rather than simply describe and name the new species, I have segregated them into species groups as used in my recent mono- graph of the genus (Bright, 1981, Mem. Ent. Soc. Canada 118). The species in each group will be described separately, and a new key will be presented where appropriate. The present contribution is the first and includes those species in the Scriptor group. Consult my recent monograph for group diagnoses, for keys to species groups, and for other information. I thank Dr. T. H. Atkinson for sending the specimens and my colleagues Dr. Y. Bous- quet and Dr. J. R. Vockeroth for reading the initial draft of this manuscript. Pityophthorus atkinsoni, n. sp. Length 1.9-2.2 mm, 3.0 times longer than wide. Female. — Frons flattened on a semicircu- lar area extending from epistoma to well above upper level of eyes, flattened area occu- Biosystematics Research Institute, Agriculture Canada, Ottawa, On pying about 66% of distance between eyes, surface of flattened area densely and finely punctured with dense yellowish setae, these longer and incurved on periphery and along epistoma, surface above and lateral to flat- tened area shining with large, sparse punc- tures. Antennal club 1.25 times longer than wide, widest through segment 2; first two sutures straight; first two segments occupy almost I of total club length. Pronotum 1.2 times longer than wide, widest at level of summit; sides evenly, moderately arcuate; as- perities on anterior slope arranged into two to four evenly concentric rows, if two then two additional rows irregularly arcuate, usually one or two additional broken rows at summit; summit distinctly elevated; posterior area of disc with deep, moderate punctures, these separated by distance equal to or less than their diameters, decreasing in size and more widely separated laterally; surface between punctures shining, with numerous fine im- pressed points; median line evident, flat, im- punctate, with fine impressed points. Elytra 1.9 times longer than wide; apex slightly acuminate; discal striae punctured in even rows, punctures definitely larger than those on posterior portion of pronotal disc, deeply impressed, close, decreasing slightly in size posteriorly, each puncture bearing a very short seta; discal interstriae narrower than striae, surface shining, minutely sculptured with impressed points and lines or reticulate and glabrous. Declivity steep, weakly bisul- cate; interstriae 1 moderately elevated, very slightly higher than interstriae 3, with a me- CanadaKlAOCe. 467 468 Great Basin Naturalist Vol. 45, No. 3 dian row of four or five small rounded gran- ules; interstriae 2 very slightly impressed, surface moderately shining, minutely reticu- late, slightly wider than interstriae 1 or 3; interstriae 3 weakly elevated, lower than 1, with a median row of five or si.x small rounded granules; punctures in striae 1 weakly visible, those in striae 2 obsolete. Male. — Frons deeply, transversely im- pressed from epistoma to upper eye level, margins of impressed area sharply defined, especially medianly, surface of impressed area with large, close punctures, a very weakly elevated, longitudinal, impunctate line may be visible, surface above and lateral to impressed area deeply and closely pimc- tured. Pronotum and elytra essentially as in female. Declivity as in female except granules may be slightly larger and punctures in striae 2 may be weakly visible. Type material. — The holotype ( ? ) bears the labels: "Cardonal (cerca Ixmiquilpan), Edo. Hgo. 27. III. 81, 2250 msnm, S-209, col. T. H. Atkinson'T'Hosp.: Compositae'7 "HOLOTYPE Pityophthorus atkinsoni D. E. Bright, 1985." The allotype and four paratypes bear the same labels plus appropri- ate type labels. Twenty-eight paratypes bear the same labels plus the host label "Hosp. Flourencia resinosa. " The holotype, allotype, and 10 paratypes are in the Canadian National Collection of Insects, Ottawa (CNC No. 18402). Twenty paratypes were returned to Dr. Atkinson, and 2 paratypes were sent to S. L. Wood, Brigham Young University, Provo, Utah. Comments. — This species does not appear to be related or similar to any species presently placed in the Scriptor group. Adults differ from those of all other species in the group by the combination of the slightly acuminate elytral apex, by the deeply, trans- versely impressed male frons, by the weakly bisulcate elytral declivity, by the large strial punctures, and by the characters of the female frons. This species is named after its collector. Dr. T. H. Atkinson, who has collected many new species of Scolytidae in Mexico. Pitijophtliorus diminuth us , n. sp. Length 0.9-1.0 mm, 3.0 times longer than wide. Female. — Frons deeply concave from epistoma to well above upper level of eyes, concavity more strongly impressed just above epistoma; surface of concave area brightly shining, with very fine points, or less shining, with fine microreticulation, periphery of con- cavity with closely placed, long, yellowish se- tae, those on vertex of concavity reaching about midway to epistoma. Antennal club oval, about 1.3-1.4 times longer than wide, widest near middle; only one suture visible, straight, located just below middle; anterior face glabrous, shining. Pronotum about 1.2 times longer than wide, widest at middle; side straight, parallel; asperities on anterior slope arranged into three definite concentric rows, a vague fourth row detectable at summit; pos- terior area of disc shining and densely, minutely reticulate, punctures widely sepa- rated, very obscure, shallow, barely apparent; median line very vague, faint, impunctate, shallowly impressed. Elytra 2.0 times longer than wide; apex distinctly acuminate; striae punctured in regular rows, punctures large, close, each with an extremely short seta; in- terstriae about half as wide as striae, impunc- tate. Declivity steep, generally flattened, bisulcate; interstriae 1 distinctly elevated above 2, impressed on upper portions slightly below level of interstriae 3, bearing a median row of very small acute granules; interstriae 2 impressed, widened toward apex, surface glabrous, shining, smooth; interstriae 3 slightly elevated, about as high as 1 on lower half, armed with median row of distinct but small acute granules, these larger than those on interstriae 1; punctures in striae 1 and 2 generally distinct, only slightly smaller than those on disc; vestiture of \'er\' fine, hairlike setae. Male. — Frons very slightly flattened or shallowly, transversely impressed from epis- toma to upper level of eyes, upper margin marked by a weak transverse carina; surface with scattered fine setae. Pronotum as in fe- male except asperities in more even, clearly defined concentric rows. Elytra as in female except vestiture on base and face of declivity distinctK' broader and scalelike. Declivity es- sentialK as in female except interstriae 1 slightly more deeply impressed below inter- striae 3, granules slightly larger; granules on interstriae 3 distinctly larger, arranged near July 1985 BRIGHT: New American Bark Beetles 469 middle of declivital face; apical and lateral margins more definitely acute, subtubercu- late, extending from interstriae 3 around apex to opposite interstriae 3; pimctures in striae 1 and 2 obsciue. Type material. — The holotype (?) is la- beled "Estacion de Biologia, Chamela, Edo. Jalisco, 19-VIII-82, S-761, 110 msnm. Col. Armando Equihua'V'Host.: Leguminosae"/ "HOLOTYPE Pityophthorus diminutivus D. E. Bright, 1985." The allotype and nine paratypes bear the same labels plus the appro- priate type labels. The holotype, allotype, and four paratypes are in the Canadian National Collection of Insects, Ottawa (CNC No. 18403). Five paratypes were returned to Dr. Atkinson. Comments. — This very small species seems to be most closely related to P. hy- lociiroides and related species but differs by the less deeply impressed declivital inter- striae 1, by the very different female frons, and bv the much smaller size. Pityo])]itlu)riis eqiiilutai , n. sp. Length 1.4—1.8 mm, 2.6 times longer than wide. Female. — Frons broadly flattened from epistoma to well above upper level of eyes and laterally nearly from eye to eye; surface shin- ing and densely pubescent over entire area, setae on periphery much longer, very densely placed, setae arising on upper margin reach- ing to or beyond epistoma. Antennal club nar- rowly oval, 1.3 times longer than wide, widest through segment 2; sutures 1 and 2 straight, transverse; first two segments occupy more than half of total club length. Pronotum about 1.2-1.3 times longer than wide, widest at level of summit; asperities on anterior slope arranged into four somewhat irregidar con- centric rows, these more even and regular on lateral areas, with one or two irregular rows around summit; summit weakly elevated; pos- terior area of disc with shallow, small, ill- defined punctures, these widely separated by a distance greater than their diameter; surface between punctures weakly shining, very finely reticulate; median line distinct, narrow, shallowly impressed, impunctate. Elytra 1.6 times longer than wide; apex distinctly acumi- nate; discal striae punctured in uneven regu- lar rows, punctures larger and deeper than those on posterior portion of pronotum, very close, more irregularly placed on base, each with a very short seta; discal interstriae nar- rower than striae, surface more brightly shin- ing than surface of pronotum, smooth, with scattered minute points, glabrous. Declivity moderately deeply bisulcate; interstriae 1 dis- tinctly elevated, slightly lower than inter- striae 3 on upper portions, with a distinct row of five to seven small rounded granules; inter- striae 2 distinctly, moderately impressed, slightly widened on apical half, surface smooth, shining, or moderately dull, with nu- merous very fine points or very fine reticulate; interstriae 3 convex, moderately elevated, slightly higher than interstriae 1 on upper portions, bearing a median row of three to five distinct rounded granules, these slightly larger than those on interstriae 1; several very small granules present lateral to interstriae 3 on apex of interstriae 4—8; punctures in striae 1 and 2 small, weakly impressed but readily visible, especially in striae 2. Male. — Frons shallowly, transversely im- pressed from epistoma to well above upper level of eyes, surface of impressed area shin- ing, densely, finely punctured with numer- ous, erect, short setae, surface above and lat- eral to impressed area more deeply, sparsely punctured, glabrous, shining. Pronotum as in female except asperities on anterior slope larger, in more even concentric rows. Elytra as in female except strial punctures larger, more deeply impressed, declivity more deeply bisulcate, interstriae 1 distinctly lower than interstriae 3 and granides on interstriae 1 and 3 much larger, acute. Type material. — The holotype (?) is la- beled: "Est. de Biologia Chamela, Edo. de Jalisco, S-832, 12-XI-82, 100 m, col. Armando Equihua M'T'HOLOTYPE Pityophthorus equihuai D. E. Bright, 1985." The allotype and 18 paratypes bear the same label plus an appropriate type label. The holotype, allotype, and eight paratypes are in the Canadian National Collection of Insects, Ottawa (CNC No. 18404). Eight paratypes were returned to Dr. Atkinson and two paratypes were sent to S. L. Wood, Brigham Young University, Provo, Utah. Comments. — Adults of this species are similar to those of P. mexicanus, coronarius. 470 Great Basin Naturalist Vol. 45, No. 3 and vesculus but differ in a number of signifi- cant characteristics as mentioned in the key. This species is named after its collector, Armando Equihua, who has collected many undescribed species of Scolytidae in recent years. Pitijophthorus thamnus, n. sp. Length 1.9-2.1 mm, 3.0 times longer than wide. Female. — Frons flattened on large area extending from epistomal margin to upper level of eyes, laterally occupying about 86% of distance between eyes; surface densely, finely punctate, bearing dense, short, erect setae, all of about equal length. Antennal club oval, 1.3 times longer than wide, widest at middle through segment 2; sutures 1 and 2 trans- verse; segments 1 and 2 together occupy more than half of total club length. Pronotum very slightly longer than wide, widest at middle; sides weakly arcuate, very weakly constricted before anterior margin; asperities on anterior slope arranged into four concentric rows, with one or two small, vague additional rows at summit; summit weakly elevated; posterior portion of disc with moderate size punctures, these widely separated by a distance greater than their diameter; surface between punc- tures moderately shining, densely, finely, minutely reticulate-punctate; median line weakly elevated behind summit, broad, im- punctate toward base. Elytra 2.0 times longer than wide; apex moderately acuminate; discal striae punctured in regular rows, punctures very slightly larger than those on posterior portion of pronotal disc, moderately im- pressed; discal interstriae about as wide as striae, smooth, moderately shining, minutely reticulate. Declivity sloping, moderately deeply bisulcate; interstriae 1 distinctly ele- vated, impressed below level of 3 on upper portion, with a few fine granules on basal area and toward apex, central portion devoid of granules; interstriae 2 moderately deeply im- pressed, widened toward apex, surface smooth, minutely reticulate; interstriae 3 dis- tinctly, moderately elevated, higher than 1 on upper portions, bearing a median row of fine large granules, each bearing an erect seta; puncture in striae 1 and 2 obsolete; if visible, then much smaller, shallower than those on disc. Male. — Frons flattened on an area equal to female, weakly, transversely impressed above epistoma; surface with coarse rather large punctures, lower median area just above epistoma weakly, longitudinally elevated, im- punctate; vestiture sparse. Pronotum and ely- tra essentially as in female except punctures, asperities etc. coarser. Declivity as in female except interstriae 3 more strongly elevated above interstriae 1, granules larger; inter- striae 1 more deeply impressed and strial punctures slightly larger. Type material. — The holotype (?) is la- beled: "Pachuca Edo. de Hidalgo, S-462, 21.V.82, 2400 m., col. A. Equihua M.7 "Zaluzaniaangusta(Lag.) Sch. Bip. (Composi- tae)'7"HOLOTYPE Pityophthorus thamnus D. E. Bright, 1985." The allotype and two paratypes bear the same data and appropriate type labels. The holotype, allotype, and two paratypes are in the Canadian National Collection of Insects, Ottawa (CNC No. 18405). Four paratypes were returned to Dr. Atkinson. Comments. — Adults of this species are similar to those of P. coronahus and related species but differ by the more gradually slop- ing elytral declivity, by the smaller granules on declivital interstriae 3, by the less strongly acuminate elytral apex, and by the female frons that bears setae all equal in length. Pityoplithonis tninculus, n. sp. Length 1.0 mm, 3.0 times longer than wide. Female. — Frons broadly flattened on a semicircular area extending from epistoma to slightly above upper eye level and laterally occupying about 65% of distance between eyes; surface of flattened area densely, finely punctured, a minute tubercule evident just above epistomal margin in some specimens; vestiture short, moderately abundant, gener- ally of ecjual length, slightly longer on periph- ery. Antennal club oval, about 1.2 times longer than wide, widest through segment 2; first two sutures straight, first two segments occupy slightK more than half of total club length. Pronotum less than 1.1 times longer than wide, widc\st on posterior half; sides moderately arcuate; asperities on anterior slope arranged into three concentric rows, a J July 1985 BRIGHT: New American Bark Beetles 471 vague fourth row detectable at summit; poste- rior area of disc distinctly to weakly shining, punctures entirely obscure, only very weakly indicated; median line not obvious. Elytra 2.0 times longer than wide; apex distinctly acumi- nate; discal surface completely, minutely reticulate, strial punctines not visible, or at most very weakly indicated. Declivity con- vex, sloping; interstriae 1 weakly elevated, devoid of granules; interstriae 2 weakly im- pressed, not sulcate; interstriae 3 not elevated or only weakly so, equal or very slightly higher than 1, with a median row of extremely fine granules; punctures in striae 1 and 2 obso- lete. Vestiture sparse, consisting of fine, hair- hke setae on declivital interstriae 1, 3, 5, etc., those on interstriae 1 much shorter. Male. — Frons weakly transversely im- pressed from epistoma to above upper level of eyes, upper margin of impression slightly more evidently elevated into a short, trans- verse carina; surface finely, densely punc- tured. Pronotum and elytra as on female ex- cept granules on declivital interstriae 3 very slightly larger. Type material. — The holotype ( ? ) bears the labels; "Est. de Biologia Chamela, Edo. de Jalisco, S-831, 12-XI-81, 100 m, col. Ar- mando Equihua M. "/"HOLOTYPE Pityoph- thorus trunculus D. E. Bright, 1985." The allotype and seven paratypes bear the same data plus appropriate type labels. The holotype, allotype, and two paratypes are in the Canadian National Collection of Insects, Ottawa (CNC No. 18406). Five paratypes were returned to Dr. Atkinson. Comments. — This species will key to P. dimidiatus Blackman or to P. minutidis in my 1982 key. Adults of F. trunculus differ from those of both species by the entirely impunc- tate, reticulate elytra (faint strial punctures visible in some specimens), by the nearly im- punctate posterior area of the pronotum, and by the smaller size. Pityophthorus zexmenivora, n. sp. Length 1.8 mm, 2.7 times longer than wide. Male. — Frons broadly flattened from epistoma to slightly above upper eye level, very weakly transversely impressed above epistoma; surface with coarse, dense. moderate-size punctures, these distinctly im- pressed, each bearing a short, yellowish, semirecumbent seta. Antennal club oval, about 1.6 times longer than wide, widest through segment 2; sutures 1 and 2 trans- verse; segments 1 and 2 occupy more than half of total club length. Pronotum as long as wide, widest on posterior half; sides subparallel on basal half, weakly but distinctly constricted before anterior margin; asperities on anterior slope arranged into four concentric rows with one or two indistinct rows at summit; summit very weakly elevated; posterior area of disc with coarse moderate size, distinctly im- pressed punctures, these separated by a dis- tance equal to their diameter; surface be- tween punctures moderately shining, with minute lines and points. Elytra 1.8 times longer than wide; apex strongly acuminate; discal striae punctured in regular rows, punc- tures equal to or slightly larger than those on posterior portion of pronotum, moderately deep; discal interstriae about as wide or slightly wider than striae, surface smooth, moderately shining, very finely reticulate, impunctate, except near declivity. Declivity sloping, strongly sulcate; interstriae 1 slightly elevated, bearing a few acute granules on base and 1 or 2 large granules at apex; interstriae 2 deeply impressed, broadened toward apex, surface smooth, very minutely reticulate; in- terstriae 3 strongly elevated, much higher than 1 on upper two-thirds, bearing a median row of large acute granules and long, hairlike setae; punctures of striae 1 and 2 obsolete. Type material. — The holotype { S ) is la- beled "Pachuca, Edo. de Hidalgo, S-461, 21. VI. 82, 2400 m, Armando Equihua"/ "Zexmenia sp. (Compositae)"/"HOLOTYPE Pityophthorus zexmenivora D. E. Bright, 1985." One paratype bears the same data with an appropriate type label. The holotype is in the Canadian National Collection of Insects, Ottawa (CNC No. 18407). The paratype was returned to Dr. Atkinson. Comments. — Adults of P. zexmenivora are similar to those of P. mexicanus but differ by the lack of a fine, elevated median line on the posterior portion of the pronotum, by the finer, shallower punctures on the pronotum, and by the impunctate elytral interstriae. Adults also key out near P. equihuai but diffier 472 Great Basin Naturalist Vol. 45, No. 3 by the size, by the characters of the declivity. This species was previously known from and by other characters given in the key. Jalisco, Mexico, from Samhucus sp. A series of Pityophthorus coronarius Blackman eight specimens are labeled: "Rancho Tetela, Pityophthorns coronarius Blackman, 1942, Proc. U.S. Cuernavaca, Mor. [elos], Compositae, 10 En- Nat. Mus. 92; 220; Bright, 1981, Mem. Ent. Soc. loco T7c;o r^ l tjttca ca/^it Canada 118: 33; Wood, 1982, Great Basin Nat. ^^^ 1^82, 1750 m, Col. BUSA-SACE- Mem. 6:1115. MAFE. Key to species in Scriptor group (Modified from Bright 1981) 1. Interstriae 1 moderately to strongly impressed below level of interstriae 3 on upper declivity (especially in males); granules on declivital interstriae 3 usually large and prominent 2 — Interstriae 1 not impressed on declivity or only weakly impressed, equal in height to interstriae 3, or at most very slightly lower; granules on interstriae 3 usually very small 15 2(1). Declivital sulcus very wide, flattened to interstriae 5, interstriae 1 only slightly lower than 3 on upper half; male frons strongly, transversely impressed from epistoma to upper level of eyes; posterior portion of pronotum obscurely punctured, reticulate; body slender, about 3. 1 times longer than wide; Guatemala elegans Schedl — Declivital sulcus more convex, more deeply bisulcate; male frons variable; posterior portion of pronotum distinctly punctured 3 3(2). On conifers in SW United States; declivital interstriae 2 very broad, moderately sulcate; pronotal asperities arranged in even or irregular rows; female frons pubescent over entire surface, setae on periphery longer and incurved arcanus Bright — On shrubs in SE and W United States or in deciduous trees or shrubs in Central America; declivital interstriae 2 not broadly sulcate; pronotal asperities scattered or in even concentric rows; female frons variable but not as above 4 4(3). Pronotal asperities numerous, confused, in no apparent order; declivital striae 1 and 2 with distinct punctures, these almost equal in size to those on disc; declivital setae of male narrowly spatulate; Panama vesculus Wood — Pronotal asperities arranged in even concentric rows; punctures in declivital striae obscure; declivital setae hairlike in males 5 5(4). Larger species; male frons convex, without a carina; declivity more gradual, apex strongly acuminate, interstriae 3 armed by median row of distinct granules, forming lateral margin of declivity 6 — Small species; male frons transversely impressed, with a distinct, transverse carina at upper level of eyes; declivity steeper, apex moderately to weakly acuminate; interstriae 3 granulate to middle of declivital face, not forming lateral margins of declivity 11 6(5). Declivity sloping, beginning on posterior one-fourth of elytra, moderately im- pressed, granules on interstriae 3 small; elytral apex moderately acuminate; female frons with short setae, all equal in length; Mexico tluunnus Bright — Declivity steep, beginning on posterior third of elytra, usually strongly im- pressed, granules on interstriae 3 large; elytral apex moderately to strongly acuminate; female frons with long setae, especially on periphery 7 July 1985 BRIGHT: New American Bark Beetles 473 7(6). Posterior portion of pronotum hearing rounded, elevated granules on lateral or posterior edges of punetures; surface between punctures on posterior portion of pronotum strongly and densely reticulate; female frons densely pubescent, with longer, downward-pointing setae on upper and lateral margin 8 — Posterior portion of pronotum not hearing rounded, elevated granules on edges of punctures; surface between punctures on posterior portion of pronotum not strongly reticulate, or punctures obsolete; females frons variable 9 8(7). Body slender, about 2.9-3.0 times longer than wide; female frons broadly flattened, densely pubescent with long setae on upper and lateral margins; male frons flattened; interstriae 1 on declivity with small granules along entire length, granules on interstriae 3 slightly larger; Mexico coronarius Blackman — Body stouter, 2.7-2.8 times longer than wide; female frons broadly concave, with long setae on complete periphery; male frons moderately, transversely impressed; interstriae 1 on declivity devoid of granules except at base and at apex, granules on interstriae 3 acute, prominent; Mexico concinnus Wood 9(7). Median line on posterior portion of pronotum sharply, narrowly elevated; punc- tures on posterior portion of pronotum large, deep, almost touching; discal interstriae 1, 3, 5, 7 usually with a median row of sparse setae extending nearly to base; Mexico mexicanus Blackman — Median line on pronotum not elevated; punctures on posterior portion of prono- tum finer, shallower, and more widely separated; discal interstriae usually impunctate 10 10(9). Body length 1.8 mm; declivity deeply sulcate, interstriae 3 strongly elevated, inner slope abrupt; median portion declivital interstriae 1 devoid of granules; Mexico zexmenivora Bright — Body length 1.4-1.8 mm; declivity weakly sulcate, interstriae 3 only slightly higher than 1, granules on interstriae 1 and 3 very large; setae on female frons very long, extending from vertex to epistoma; Mexico equihuai Bright 11(5). Length 0.9-1.0 mm; female frons concavely impressed, shining, glabrous in center, periphery with long, yellowish setae; punctures on posterior portion of pronotum obsolete, surface between punctures reticulate; Mexico diminutivus Bright — Length greater than 1.2 mm; female frons flattened, densely pubescent over entire surface; punctures on posterior portion of pronotum distinct, surface between punctures smooth 12 12(11). Declivity deeply sulcate in male, moderately sulcate in female, interstriae 3 much higher than 1 on declivital base, with large coarse granules; interstriae 2 distinctly broadened toward apex in male, punctures in striae 2 obscure in male, distinct in female 13 — Declivity shallowly sulcate, interstriae 3 moderately or only slightly higher than 1, with moderately large or small granules; interstriae 2 variable, but not broadened toward apex in male, punctures in striae 2 distinct in both sexes 14 13(12). Declivity of male very steep, flattened, with an acute, tuberculate margin extending from top of interstriae 3 around apex to opposite interstriae 3, granules on interstriae 1 and 3 large, acute; declivity of female convex, less deeply sulcate, with prominent granules on interstriae 1 and 3; central Mexico. . . hylocuroides Wood — Declivity of male less steep, obscurely flattened on lower half, lateral margin lateral to interstriae 3 rounded, granules on interstriae 1 small to absent, on interstriae 3 slightly larger, acute; declivity of female more convex, less deeply sulcate, with small granules on interstriae 1 and 3; Idaho to northern Mexico and west Texas virilis Blackman 474 Great Basin Naturalist Vol. 45, No. 3 14(12). Granules on declivital interstriae 1 and 3 small; female frons evenly pubescent on a large semicircular area, setae slightly longer on periphery; male frons moderately deeply, transversely impressed; southeastern USA . . scriptor Blackman — Granules on declivital interstriae 1 and 3 moderately large; female frons densely pubescent, the setae on the periphery very closely placed and very long, extending downward almost to epistoma and masking surface of frons; male frons shallowly, transversely impressed; Honduras hermosus Wood 15(1). Declivity distinctly flattened, very weakly sulcate, interstriae 1 and 3 very weakly elevated with fine granules; male frons distinctly, transversely im- pressed, impression deeply punctured; female frons densely, evenly pubescent; Mexico atkinsoni Bright — Declivity variable but distinctly convex and bisulcate; male frons flattened or weakly, transversely impressed; female frons variable 16 16(15). Basal half of pronotum dull, minutely reticulate, punctures barely visible, if at all; Mexico dimidiatus Blackman — Basal half of pronotum dull to shining, smooth, punctures usually distinctly visible 17 17(16). Punctures in declivital striae 2 obscure, difficult if not impossible to see 18 — Punctures in declivital striae 2 distinct, obvious 23 18(17). Declivital interstriae 1 and 3 devoid of granules or granules extremely minute .... 19 — Declivital interstriae 1 and 3 with generally distinct granules 20 19(18). Elytral strial punctures obsolete, entire surface minutely reticulate; punctures on posterior surface of pronotum obsolete; female frons flattened, with abundant setae of equal length over surface; length 1.0 mm; Mexico trunculus Bright — Elytral strial punctures distinct, in even rows, entire surface smooth; punctures on posterior surface of pronotum distinct; female frons flattened, pubescence long on periphery, somewhat sparser on a median longitudinal space; length 1.0-1.2 mm; Mexico minutalis Wood 20(18). Male frons convex, with a weak longitudinal carina; female frons concave, uniformly covered with dense pubescence, setae on periphery longer and in- curved; southwestern USA torridus Wood — Male frons very weakly, transversely impressed, with an extremely fine, longitu- dinal carina or a narrow, longitudinal, smooth space; female frons flattened, with either three tufts of extremely long, fine setae extending downward to apex of mandibles or with flattened area uniformly pubescent 21 21(20). Female frons with three tufts of extremely long, downward-pointing setae that extend almost to tips of mandibles; male frons generally conxex, narrowly, transversely impressed above epistoma, with a weak longitudinal carina; Guatemala uufialis Wood — Female frons uniformly pubescent on a subcircular area occupying 70%-75% of distance between eyes, setae short, usually all of ecjual length; male frons weakly, transversely impressed from epistoma to upper level of eye, usualh' with a longitudinal smooth space; Mexico 22 22(21). Size smaller, 0.8-1.2 mm; declivital interstriae 2 ver\ weakK impressed, interstriae 3 weakly elevated, granules on 1 and 3 minute, obscure . . . atumus Wood — Size larger, 1.2-1.5 mm; declivital interstriae 2 more deeply impressed, interstriae 3 distinctly elevated, granules on 1 and 3 distinct . . . attcnuatus Blackman July 1985 BRIGHT: New American Bark Beetles 475 23(17). On Piniis spp.; surface between punctures on posterior portion of pronotum sm ooth, with only a few minute points and/or lines . . 24 On vines- surface between punctures on posterior portion of pronotum densely micropunctate; Costa Rica sobrinusWood 24(23). Elytral apex distinctly, strongly acuminate, apices projecting downward on female; southern Mexico and Guatemala (see Confertus group). . . . suhstmilis Schedl _ Elytral apex moderately acuminate, apices not projecting downward on female; thern Mexico (see Confertus group) subimpressus Bright sou NEW SPECIES AND NEW RECORDS OF NORTH AMERICAN PITYOPHTHORUS (COLEOPTERA: SCOLYTIDAE), PARTY: THE JUGLANDIS GROUP Donald E. Bright' Abstract — Five new species of Mexican Pityophthonis in the Juglandis group are described: P. abhisus (Hidalgo), P. costifera (Guerrero), P. cracentis (Morelos), P. desultoritis (Puebla), and P. insuetus (Morelos). New locality or host records are given for P. costahilis Wood, P. costatulus Wood, and P. diligens Wood. Previously published keys are modified to accommodate these species. This is the second contribution describing the unnamed species of Pity aphtha rus col- lected by Dr. T. H. Atkinson and his col- leagues (Centro de Entomologia y Acarologia, Colegio de Postgraduados, Chapingo, Mex- ico). As in the previous paper (Great Basin Nat. 45:467-475), all the species in a species group are described together, and the key in my I98I monograph (Mem. Ent. Soc. Canada 118:72) is modified to accommodate the newly named taxa. I wish to thank Dr. T. H. Atkinson for send- ing the specimens to me and also thank him and his students for their diligence in search- ing for Scolytidae in previously unrecognized host plants. I also thank my colleagues Dr. Y. Bousquet and Dr. J. M. Campbell for review- ing the manuscript of this paper. PityaphtJianis ablusiis , n. sp. Length 1.4-1.7 mm, 3.2 times longer than wide. Female. — Frons convex, slightly to al- most indeterminably flattened in median area extending to or above upper eye level; surface moderately shining with numerous, shallow, indefinite punctures and short, fine, scattered setae; epistoma with a very weak, blunt, me- dian callus, this sometimes extended into a weakly elevated, longitudinal carina extend- ing across flattened area. Antennal club oval, 1.5 times longer than wide, widest through segment 2; sutures 1 and 2 transverse, 1 heav- ily septate for almost entire length, 2 septate at lateral margins only; segments 1 and 2 to- gether occupy two-thirds of total club length. Pronotum 1.1 times longer than wide, widest at middle; sides parallel behind middle, slightly constricted and broadly rounded be- fore middle; asperities on anterior slope ar- ranged into three distinct concentric rows and 1 additional short, slightly irregular row around summit, asperities in each row dis- tinct; summit distinctly elevated; posterior area weakly impressed behind summit, with large, distinct, moderately deep punctures, these separated by a distance nearly equal to diameter of puncture; surface between punc- tures moderately shining, marked with nu- merous fine lines and surface irregularities; median line not evident. Elytra almost 2.0 times longer than wide; apex narrowly rounded; discal striae punctined in regular rows, punctures large, moderately im- pressed, about equal in size and depth to those on posterior portion of pronotum; inter- striae about 2.0 times wider than striae, sur- face impunctate, smooth, shining. Declivity sloping, weakly bisulcate; interstriae 1 mod- erately elevated, equal in height or slightly lower than 3, devoid of granules, with a few very fine punctures and setae; interstriae 2 weakly impressed, flat, broader than discal width, surface moderately shining with nu- merous fine lines and points; interstriae 3 weakly elevated, devoid of granules; punc- tures in striae 1 distinct but only very weakly impressed, those in 2 almost obsolete, much smaller than those on disc; interstriae 3, 5, 7 with a median row of several short setae. Male. — Frons moderately deepK', trans- versely concave on each side of median line; smi'acc with sliglitK larger and deeper punc- Biosvslfniatics Rt-scarcl 47fi July 1985 Bricht: New American Bark Beetles 477 tures than on female, setae sparser. Pronotuni as in female except punctures on posterior portion more distinct. Elytra and declivity similar to that of female. Type material. — The holotvpe (?) is la- beled: "Pachuca, Edo. Hgo., 21-V-S2, S-461, 2400m, Col. T. H. AtkinsonTZe.xmeniasp. 7 "HOLOTYPE Pitvophthorus ablusus Bright, CNC No. 18434." The allotype and 14 paratypes bear the same labels plus appropri- ate type labels; 7 additional paratypes bear the same data except that the collector is A. Equihua M. The holotype, allotype, and 6 paratypes are in the Canadian National Collection, Ottawa (CNC), 2 paratypes were sent to S. L. Wood, Brigham Young University, Provo, Utah, and the remaining paratypes were returned to T. H. Atkinson. Comments. — Adults of this species are dis- tinguished from those of P. fransehae Wood by the absence of granules in declivital inter- striae 1 and 3, by the very weakly flattened female frons, by the male frons, which is con- cave on each side of a weakly elevated median line, and by the distributions. Pityophthorus costifera, n. sp. Length 1.2-1.4 mm, 2.4 times longer than wide. Female. — Frons evenly, weakly convex; surface smooth, with small, shallow, evenly spaced punctures, these separated by dis- tances equal to their diameters; pubescence sparse, nearly absent. Antennal club about 1.2 times longer than wide, widest through segment 2; sutures 1 and 2 transverse, straight, 1 distinctly septate, 2 septate only at lateral margins; segments 1 and 2 together occupy more than half of total club length. Pronotum as long as wide, widest at level of summit; anterior margin broadly rounded with a subserrate elevated costa; anterior slope bearing four elevated, evenly concen- tric, subserrate costae, these not divided into individual asperities, with one or two indis- tinct costae around summit; summit weakly elevated, distinct; posterior portion of disc distinctly, evenly punctured, punctures shal- low, of moderate size, separated by distances about equal to their diameters; surface be- tween punctures moderately shining, with numerous minute points; median line broad. impunctate, not elevated. Elytra 1.5 times longer than wide; apex broadly rounded; dis- cal striae punctured in regular rows, punc- tures small, shallowly impressed; discal inter- striae about 2.0 times wider than striae, surface moderately shining, densely micro- punctate. Declivity very weakly impressed; interstriae 1 narrowly, weakly elevated, weakly impressed below elytral surface, with a median row of about four extremely small granules, each with a short, fine, erect seta at base; interstriae 2 very weakly flattened, es- sentially as on disc; interstriae 3 very weakly elevated, devoid of granules, with median row of short, fine erect setae; punctures in striae 1 and 2 obsolete, striae 1 narrowly impressed. Male. — Very similar to female except frons less strongly convex, very weakly im- pressed above epistoma, with a low subtuber- culate elevation at midpoint on epistoma, de- clivital setae very slightly stouter, declivital interstriae 3 with a median row of very fine granules, and striae 1 and 2 more conspicu- ously punctured on declivity. Type material. — The holotype (?) is labeled: "Taxco, Guerrero, 22.11.82, S326, 1900 msnm, col. Atkinson y EquihuaV "Apocynaceae'7 ? /"HOLOTYPE Pityoph- thorus costifera, D. E. Bright, 1985, CNC No. 18420." The allotype and six paratypes bear the same data plus appropriate type la- bels. One specimen, thought to be this spe- cies but not designated as a paratype, is la- beled: "Chilapa, Guerrero, 23.n.83, S-344, 1640 msnm, col. Atkinson y Equihua." Eight additional paratypes are labeled: "Cuautla, Morelos, 1 Oct-82, 1230 m, SM-119, E. Saucedo — A. Burgos"/ "(Apocynaceae) Theve- tia peruviana " and a paratype label. The holotype, allotype, and 6 paratypes are in the CNC, 6 paratypes were returned to T. H. Atkinson, and 2 paratypes were sent to S. L. Wood. Comments. — This species is closely re- lated to P. costatiihis Wood but differs by its larger size, by the very different female frons as indicated below, and by the very weakly impressed elytral declivity. Pityophthorus cracentis, n. sp. Length 1.1-1.2 mm, 3.3 times longer than wide. 478 Great Basin Naturalist Vol. 45, No. 3 Female. — Frons flattened on a smafl area extending from epistomal margin to wefl above eyes, flattened area laterafly occupying about 60% of distance between eyes; surface of flattened area densely, minutely punc- tured, with moderately long, erect, yellowish setae, these longer on upper portion, becom- ing progressively shorter toward epistoma; surface above and lateral to flattened area smooth, shining, glabrous, with a few widely separated, small punctures. Antennal club small, about 1.2 times longer than wide, widest through segment 2; sutures 1 and 2 transverse, lightly septate at lateral margins, 1 more strongly septate than 2; segments 1 and 2 together occupy about two-thirds of total club length. Pronotum less than 1.1 times longer than wide, widest at middle; sides very slightly converging behind middle, broadly rounded before middle; asperities on anterior slope arranged into four definite, regular, concentric rows, fourth row small, placed around summit, asperities in each row dis- tinct; summit weakly elevated; posterior area with widely separated, small, very shallow punctures; surface between punctures smooth, shining to very finely reticulate; me- dian line evident as a narrow, weakly elevated line extending from summit to posterior mar- gin. Elytra about 1.5 times longer than wide; apex broadly rounded; discal striae punctured in regular rows, punctures large, distinctly but shallowly impressed, these much larger, deeper than those on posterior portion of pronotum, each puncture with an extremely short seta equal to or slightly longer than di- ameter of puncture; discal interstriae shining, about 1.5 times wider than striae, surface im- punctate, marked with extremely fine lines. Declivity steep, weakly bisulcate; interstriae 1 weakly elevated, slightly impressed below level of 3 on upper half, each interstriae with 1 small, acute granule near apex; interstriae 2 weakly bisulcate, shining, glabrous; inter- striae 3 slightly elevated above 1, with a me- dian row of 3 small but distinct, acute gran- ules, middle granule larger than other two; remaining interstriae unmodified with a few, long, erect setae. Type material. — The holotype (?) is la- beled: "Carr. Xochicalco — C^uentepec, Km. 6, Temixco, Mor., 14-Julio-84, 1220 m, SM- 347, E. Saucedo — E. Martinez'V'Composi- tae'T'HOLOTYPE Pitvophthorus cracentis D. E. Bright, 1985, CNC No. 18430." Four female parat\'pes bear the same labels plus appropriate paratype labels. The holotype and 1 paratype are in the CNC, the remaining 3 paratypes have been returned to T. H. Atkinson. Comments. — This species seems to be unique among the Pityophfhorus known to me. Adults differ from the other species in the Juglandis group by the distinctly but slightly impressed elytral declivity. It cannot be placed in the Scriptor group since the elytral apex is not acuminate, and it doesn't fit at all in the other species groups with concentric rows of pronotal asperities. However, to avoid creating monotypic species groups unless ab- solutely necessary, I have broadened the con- cept of the Juglandis group to include this species. PifyophtJioriis desultuhus, n. sp. Length 1.2-1.4 mm, 2.8 times longer than wide. Female. — Frons flattened on a small area extending from epistomal margin to upper level of eyes, with a weakly elevated, impunc- tate, longitudinal carina extending from epis- toma to upper margin of flattened area, flat- tened area laterally occupying about 60% of the distance between eyes; surface on each side of carina shining, with distinct small, shallowly impressed punctures and with short, scattered, inconspicuous setae. Anten- nal club circular, as long as wide, widest through segment 2; sutures 1 and 2 trans- verse, straight, septate at lateral margins; seg- ments 1 and 2 together occupy about two- thirds of total club length. Pronotum 1.1 times longer than wide, widest at level between summit and posterior margin; asperities ar- ranged into three or four irregular rows, these especially irregular on median portion, rows more regular on lateral areas; summit weakly elevated; posterior portion of disc moderately shining, with large, deep punctures, these separated by distances much less than their own diameters; surface between punctures with extremely fine lines and points; median line flat, broad, impunctate. Elytra 1.2 times longer than wide; apex broadK rounded; dis- cal striae punctured in distinct, regular rows. July 1985 BRIGHT: New American Bark Beetles 479 punctures about same size and depth as those on posterior portion of pronotuni, each punc- ture bearing a very short seta equal to or slightly longer than diameter of puncture; dis- cal interstriae moderately shining, equal in w idth or slightly narrower than striae, surface impunctate, moderately shining with very tine points and lines. Declivity steep, almost evenly convex, very weakly bisulcate; inter- striae 1 weakly elevated, devoid of granules; interstriae 2 very weakly impressed, moder- ately shining; interstriae 3 convex, not ele- vated, equal in height to 1, devoid of granules; punctures in striae 1 and 2 not visible; inter- striae 1, 3, 5, 7, each with a median row of narrowly spatulate setae. Male. — Almost identical to female, differs by frons more distinctly convex, setae shorter and sparser. Type material. — The holotvpe (?) is la- beled: "Oriental, Pue., 4.V.8l', 2370 m, col. T. H. Atkinson, A. Equihua, S-213'7"Hosp.: Compositae'T'HOLOTYPE Pitvophthorus desultorius, D. E. Bright, 1985, CNC No. 18421." The allotype and 10 paratypes bear the same labels plus appropriate type labels. The holotype, allotype, and 4 paratypes are in the CNC, four paratypes were returned to T. H. Atkinson, and 2 paratypes were sent to S. L. Wood. Comments. — Adults of this species are similar to those of P. insuetus (described be- low) but differ by the larger, deeper punc- tures on the posterior portion of the prono- tum, by the steeper, more convex elytral declivity, and by the slightly smaller size. Pityophthorus insuetus, n. sp. Length 1.5-1.8 mm, 3.1 times longer than wide. Female. — Frons convex, very weakly flat- tened, with a distinct, weakly elevated longi- tudinal carina extending from epistoma to above upper level of eyes; surface on each side of carina shining, with densely placed, weakly impressed punctures; pubescence sparse, erect, consisting of short, scattered, erect se- tae, these longer along epistomal margin. An- tennal club oval, about 1.5 times longer than wide, widest through segments 1 and 2; su- tures 1 and 2 transverse, 1 more distinctly septate; segments 1 and 2 together occupy more than half of total club length. Pronotum about 1.1 times longer than wide, widest slightly behind level of summit; asperities ar- ranged into four irregular concentric rows, these especially irregular in median portion of first two rows, these first two rows more even laterally, rows around summit more even, regular; summit weakly elevated; posterior portion of disc shining, punctures moderate, deep, very close, these separated by less than diameter of puncture; surface between punc- tures with very finely impressed, minute points; median line weakly elevated, smooth, impunctate. Elytra 2.0 times longer than wide; apex narrowly rounded; discal striae punctured in distinct regular rows, punctures distinctly, moderately impressed, each with a seta slightly longer than diameter of punc- ture, punctures slightly smaller than those on posterior portion of pronotum; discal inter- striae smooth, equal in width or slightly nar- rower than striae, impunctate. Declivity con- vex, weakly bisulcate; interstriae 1 weakly elevated, weakly impressed below level of in- terstriae 3, devoid of granules; interstriae 2 very weakly impressed, flattened, shining, equal in width to discal width; interstriae 3 very weakly elevated above interstriae 1 and 2, devoid of granules; punctures in striae 1 and 2 moderately distinct, much smaller than those on disc, striae 1 narrowly impressed; interstriae 1, 3, 5, 7 each with median row of sparse, erect setae. Male. — Very similar to female except frons more deeply, densely punctured, me- dian carina slightly more strongly elevated, declivity slightly more deeply bisulcate, setae on posterior portion of elytral interstriae nar- rowly flattened or narrowly spatulate. Type material. — The holotype ( ? ) bears the labels: "Huitzilac, MOR., 25-IX-81, 2700 m, S252, Col. Atkinson — Equihua'V'Hosp.: Compositae'T'HOLOTYPE Pityophthorus insuetus D. E. Bright 1985, CNC No. 18422." The allotype and 15 paratypes bear the same data plus appropriate type labels. The holotype, allotype, and 6 paratypes are in the CNC. 7 paratypes were returned to T. H. Atkinson, and 2 paratypes were sent to S. L. Wood. Comments. — This species does not appear to be very closely related to any species in the 480 Great Basin Naturalist Vol. 45, No. 3 Juglandis group but is similar to P. desulto- punctures in striae 1 and 2 on the declivity, rius. It will key out in my 1981 key to P. and by the irregular concentric rows of asperi- detentus Wood but differs by the sparsely ties on the pronotum (very evenly concentric pubescent female frons (densely pubescent in in P. detentiis). Adults differ from those of P. P. detentus), by the more deeply impressed desultorius by the characters mentioned elytral declivity, by the more distinct strial below. Revised KEY TO SPECIES IN THE Juglandis GROUP 1. Anterior slope of pronotum bearing 4 or more concentric, continuous costae, the summit of each costa subserrate, individual asperities not detectable; female frons flattened to weakly concave, pubescence short; male frons convex, dis- tinctly punctured, devoid of carina; posterior portion of pronotum with numerous impressed points on surface between punctures 2 — Anterior slope of pronotum with asperities arranged in several concentric rows, each row strongly serrate, divided to or near their bases, individual asperities usually detectable; female and male frons variable but not as above; surface between punctures on posterior portion of pronotum smooth and shining or reticulate 4 2(1). Sutures of antennal club straight to moderately procurved; declivital striae 1 and 2 finely punctured in both sexes; interstriae 2 weakly impressed on declivity in female; male frons subglabrous 3 — Sutures of antennal club strongly procurved; declivital striae 1 and 2 rather coarsely punctured in female, interstriae 2 more strongly impressed on declivity in female; male frons pubescent near epistoma; body length 1.6-1.8 mm; Jalisco, Guerrero costabilis Wood 3(2). Female frons flattened with moderately abundant, long, erect, equal length setae evenly scattered over surface; declivity not impressed, declivital inter- striae 1 and 3 with median row of fine granules in both sexes; body length 0.9-1.2 mm; Jalisco, Oaxaca costatulus Wood — Female frons convex, setae absent or very sparse, inconspicuous; declivity very weakly impressed, declivital interstriae 3 devoid of granules in female, granules distinct but very small in declivital interstriae 1 and 3 in male; body length 1.2-1.4 mm; Guerrero, Morelos costifera Bright 4(1). All declivital interstriae (except 2 and rarely 6) with a median row of short setae .... 5 — Declivital interstriae 1, 3, 5, 7 with a median row of short setae 8 5(4). Pronotum evenly arched from base to anterior margin, pronotal summit not elevated and transverse impression not present; pronotal asperities arranged in up to 8 broken concentric rows; interstriae 1 on declivity impressed below level of interstriae 3; Guatemala tenax Wood — Pronotum with a definitely elevated (sometimes weak) pronotal summit and with a transverse impression behind summit; pronotal asperities arranged in 3-5 even, concentric rows; declivity variable 6 6(5). Declivital interstriae 3 with small but distinct granules; declivital setae scalelike in male, hairlike in female; Gosta Rica <:,alerius Wood — Declivital interstriae 3 devoid oi granules; if present, granules extremely minute; declivital setae hairlike in both sexes 7 July 1985 BRIGHT: NewAmerican Bark Beetles 481 7(6). Antennal club elongate-oval, more than 1.4 times longer than wide; female frons flattened on a very large semicircular area extending far above upper level of eyes, pubescence dense and long on periphery, sparser in central area, male frons flattened on a smaller area, pubescent but pubescence much less dense than on female; Mexico burserae Wood — Antennal club oval, 1.4 times or less longer than wide; female frons convex, pubescence short, male frons weakly, transversely impressed, punctured, Costa Rica strictus Wood 8(4). Declivital interstriae 3 higher than 1 9 — Declivital interstriae 3 equal in height to 1 or lower 11 9(8). Interstriae 1 on elytral declivity distinctly impressed below level of interstriae 3, with one, small, acute granule near apex; interstriae 3 distinctly elevated, with three larger, acute granules; female frons flattened, pubescent on small median area; body very slender, about 3.3 times longer than wide; Morelos . cracentis Bright — Interstriae 1 on elytral declivity only very slightly impressed below level of interstriae 3, with or without a median row of numerous small granules; inter- striae 3 weakly elevated, with or without a median row of numerous small granules; body stouter, 3.0 or less times longer than wide 10 10(9). Pronotal asperities arranged into three concentric rows (vague 4th row infre- quently seen); declivital interstriae 3 devoid of granules in both sexes or granules extremely fine; female frons pubescent over entire surface, setae short, of equal length; declivity steep, interstriae 3 only slightly higher than 1; Mexico diligens Wood — Pronotal asperities arranged in four or more definite concentric rows; declivital interstriae 3 of male with distinct granules, female devoid of granules; female frons densely pubescent only on lower portion below upper level of eyes, setae long, dense, extending at least to midpoint of mandibles; declivity sloping, interstriae 3 slightly but definitely higher than 1; Mexico nanus Wood 11(8). Antennal club with only one suture septate; declivity evenly convex; male frons strongly convex, surface dull, minutely reticulate, impunctate; Mexico indigens Wood — Antennal club with two sutures septate; declivity evenly convex to weakly impressed; male frons weakly convex, transversely impressed or broadly flattened, surface punctured 12 12(11). Declivital interstriae 2 weakly impressed, below level of 3 13 — Declivital interstriae 2 not impressed, equal in height to 3 14 13(12). Declivital interstriae 1 and 3 devoid of granules; declivity sloping; female frons very weakly flattened, male frons concave on each side of weakly elevated median line; Hidalgo ahlusus Bright — Declivital interstriae 1 and 3 with a median row of fine granules; declivity more steeply convex; female frons distinctly flattened, male frons transversely impressed, with transverse carina at upper level of eyes; New Mexico franseriae Wood 14(12). Elytral declivity strongly, evenly convex, punctures in striae 1 and 2 distinct; serrations on anterior margin of pronotum located in median area only; pronotal asperities in broken concentric rows; female frons flattened to well above eyes, setae on upper margin very long, masking surface; Mexico pudicus Blackman 482 Great Basin Naturalist Vol. 45, No. 3 — Elytral declivity less strongly convex to flattened, punctures in striae 1 and 2 indistinct to obsolete; serrations on anterior margin of pronotum more generally located, extending to lateral margin; pronotal asperities in even to slightly broken concentric rows; female frons variable but not as above 15 15(14). Body length 1.7-2.0 mm; declivital interstriae 1 and 3 each with a row of small granules in male; concentric rows of pronotal asperities usually somewhat irregular; southwestern United States juglandis Blackman — Body length 1.2-1.8 mm; declivital interstriae 1 and 3 devoid of granules in both sexes; pronotal asperities arranged in regular to irregular concentric rows; Mexico 16 16(15). Pronotal asperities in regular, even rows; punctures in striae 1 and 2 obsolete on declivity; declivity evenly convex; female frons flattened, with long setae detentus Wood — Pronotal asperities irregular in median portion, more even laterally; punctures in striae 1 and 2 visible, smaller than those on disc; declivity weakly bisulcate to almost evenly convex; female frons convex, with short, sparse setae 17 17(16). Declivity sloping, weakly bisulcate, more deeply impressed in male; punctures on posterior portion of pronotal disc moderate in size, deeply impressed; body length 1.5-1.8 mm instietus Bright — Declivity steeper, almost evenly convex in both sexes; punctures on posterior portion of pronotal disc larger, more deeply impressed; body length 1.2-1.4 mm desidtorius Bright New Records Pityophthorus costabdis Wood Pityophthorus costabilis Wood, 1976, Great Basin Nat. .36:352; Bright, 1981, Mem. Ent. Soc. Canada 188, p. 74; Wood, 1982, Great Basin Nat. Mem. 6:1121. This species was previously known from eight specimens from Jalisco. A series of 19 specimens that are probably this species were seen with the labels: "Chilapa, Guerrero, 23.11.82, S-344, 1640 msnm, Col. Atkinson y Equihua." When this species was named, the frons of the females was undescribed because it was largely concealed on the available specimens. The discovery of additional specimens enable the following comments to be made. The female frons is broadly flattened, weakly plano-concave on a large area extend- ing well above eyes, with very dense, minute punctures and dense, erect, moderately long setae that are shorter in the center and longer and incurved on the periphery. The size of specimens is given in the original description as 1.6-1.8 mm. The additional specimens at hand range in size from 1.4 to 1.8 nun. PityopJithonis costatidus Wood Pityophthorus costatidus Wood, 1976, Great Basin Nat. 26, p. .351; Bright, 1981, Mem. Ent. Soc. Canada 118:73; Wood, 1982, Great Basin Nat. Mem. 6:1120. This species was previously known from Jalisco and Oaxaca from Thevetia sp. A series of seven specimens bearing the labels: "Chilapa, Guerrero, 23.11.82, S-344, 1640 msnm, col. Atkinson y Equihua" and a series of seven specimens labeled: "Cuauhtenango, GRO., 23.11.82, S-344, Atkinson-Equihua" were seen. Pdyophtliorus ddigens Wood Pityophthorus dili^ens Wood, 1976, Great Basin Nat. 36:363; Bright, 1981, Mem. Ent. Soc. Canada 188:77; Wood, 1982, Great Basin Nat. Mem. 6:1132. This species was previously known only from Hidalgo from a desert shrub with bluish leaves. Mixed in a series of 25 specimens of F. ablusus was one specimen that is probably this species. It bears the labels: "Pachuca, Edo. de Hidalgo, S-461, 21. V.82, 2400 m, col. A. Ecjuihua M'TZexiuenia sp. (Compositae)." SECOND NESTING RECORD AND NORTHWARD ADVANCE OF THE GREAT-TAILED CRACKLE {QUISCALUS MEXIC ANUS) IN NEVADA Jennifer A. Holmes', David S. Dohkin' ", and Bruce A. Wilcox' Abstract. — The second nesting record for the Great-tailed Crackle (Quiscalu.s mexicanus) in Nevada is reported from the central part of the state approximately 240 km north of the previous record. Since 1912 this species has undergone a dramatic northward extension of its previous range in the United States, presumably as a result of increased agricultural irrigation in areas that were previously desert or short-grass prairie. Range expansion of the Great-tailed Crackle {Quiscalus mexicanus) into Nevada is of relatively recent occurrence. Linsdale's surveys of the Nevada avifauna made no men- tion of the species (Linsdale 1936, 1951). The first published record for Nevada is from April 1973 in southeast Nevada near Las Vegas, Clark Co. (Oberholser and Kincaid 1974). Since then the species has been recorded from several localities in Clark, Lincoln, and Nye counties in the southeastern part of the state (Kingery 1980, 1984) and due north of there in the Ruby Valley area, Elko Co. (Kingery 1978, 1981, 1982, 1984). The first record outside these two areas was recently reported for central Nevada between Tonopah and Austin, Nye Co., during the summer of 1983 (Kingery 1984). Despite this plentitude of observations, there is only one record of nesting by this species in Nevada: a single active nest found in 1980 at Beatty, Nye Co., in southern Nevada (Kingery 1980). In this report we document the second Nevada nesting record for this species approximately 240 km north of the previous nesting record for the state. During the first week of June 1983, a female Great-tailed Crackle was seen carrying nest material to a small stand of narrowleaf cotton- woods (Populus angustifolia) surrounding a shallow pond at Carver's ranch in Carvers, Nye Co., elevation 1715 m. This locality is in Big Smokey Valley approximately midway be- tween Tonopah and Austin and is probably the basis for the observation record reported by Kingery (1984). In mid-July we observed an adult male, adult female, and a fledgling in the same stand of trees. The area used by the birds is typical of nesting areas for this species: close to water, human habitations, and agri- cultural land (Skutch 1958). Between 22 May and 15 July 1984, we observed adult male and female grackles at this same location, although we did not ascertain whether they nested there. Our observations were made incidental to a broad ecological analysis of riparian avi- faunas of the Toiyabe Range (Dobkin and Wilcox, in press); we have never seen more than a single adult of each sex at the Carver's location at any time. Prior to 1912 the Great-tailed Crackle ranged only as far north as southern Texas, New Mexico, and Arizona (Oberholser and Kincaid 1974). In recent years this species has expanded its range northward into Colorado (Stepney 1975), Nebraska (Faanes and Nor- ling 1981), Utah (White et al. 1983), California (Small 1974), and Oregon (Littlefield 1983), as well as Nevada. We agree with Littlefield's assessment that the increased introduction of mechanized sprinkler irrigation systems into newly created agricultural areas within for- mer deserts is the most likely factor enabling the Great-tailed Crackle to expand its range northwestward from southern Arizona. The north/south axes of the isolated mountain ranges in the Great Basin provide valley corri- dors with scattered agricultural "oases" con- taining suitable nesting habitat for this spe- cies. In the coming decade we should expect Center for Conservation Biology, Department of Biological Sciences, Stanford University, Stanford, California 94305. ^o whom reprint requests should be addressed. 483 484 Great Basin Naturalist Vol. 45, No. 3 to see increased nesting by Great-tailed Grackles in the area extending from southern Nevada northward to southeastern Oregon. Acknowledgments Our fieldwork in Nevada was made possible by grants from the National Science Founda- tion (DAR 8022413), the Koret Foundation, and the World Wildlife Fund. We thank the Tonapah Ranger District of the U.S. Forest Service for logistic support. Literature Cited DOBKIN, D. S., AND B A Wilcox. 1986. Avian communi- ties in natural riparian forest fragments. In J. Verner, M. L. Morrison and C. J. Ralph, eds., Wildlife 2000: modeling habitat relationships of terrestrial vertebrates. Univ. Wisconsin Press, Madison. In press. Faanes, C a., and W. Norling. 1981. Nesting of the Great-tailed Crackle in Nebraska. Amer. Birds .35:14^-149. KiNGERY. H. E 1978. The spring migration. Mountain West region. Amer. Birds 32:1036-1040. 1980. The spring migration. Mountain West re- gion. Amer. Birds 34:800-803. 1981. The spring migration, Mountain West re- gion. Amer. Birds 35:846-849. 1982. The spring migration. Mountain West re- gion. Amer. Birds 36:877-880. 1984. The autumn migration. Mountain West re- gion. Amer. Birds 38:227-230. LiNSDALE, J M 19,36, The birds of Nevada. Pacific Coast Avifauna 23:1-145. 1951. A list of the birds of Nevada. Condor .53:228-249. LiTTLEFiELD. C. D. 1983. Oregon's first records of the Creat-tailed Crackle. Western Birds 14:201-202. Oberholser, H C, AND E. B. KiNCAiD, Jr. 1974. The bird life of Texas. Vol. 2. University of Texas Press, Austin. Skutch, a. F 1958. Boat-tailed Crackle. In A. C. Bent, ed., Life histories of North American blackbirds, orioles, tanagers, and allies. U.S. Natl. Mus. Bull. 211. Smithsonian Inst., Washington, D.C. Small, A. 1974. The birds of California, Winchester Press, New York, Stepney. P. H. R. 1975. First recorded breeding of the Creat-tailed Crackle in Colorado. Condor 77: 208-210. White, C M , H H Frost, D L Shirley, C M Webb, AND R D. Porter 1983, Bird ^distributional and breeding records for Southeastern Idaho, Utah, and adjacent regions, Creat Basin Nat. 43: 717-727. NEW SPECIES OF TALINUM (PORTULACEAE) FROM UTAH N. Duane Atwood' and Stanley L. Welsh" Abstract — Named and described is Talimim thompsonii Atwood & Welsh. The species is evidently most closely allied to T. validuhim Greene from northern Arizona. The new species is known from the Cedar Mountain region of Emerv County, Utah. In the late summer of 1970 the authors, accompanied by Dr. Glen Moore, visited the summit of Cedar Mountain in Emery County, Utah. The mountain summit is protected fiom excessive erosion by the Buckhorn Conglom- erate Formation of Jurassic age. Rounded sili- cious pebbles mark the surface, which is clothed by a pinyon-juniper woodland, inter- spersed here and there with ponderosa pine. Crevices in the conglomerate provide habitats where water accumulates from the impervi- ous surface. The crevices are vegetated by those plants capable of survival through long drought periods, and it is there that we dis- covered a species of Talinum. The plants con- sisted of fleshy-leaved rosettes 2-4 cm broad, projecting only a few centimeters above the surface. Bright pink flowers were helpful in the discovery. The season was very dry, and few plants were found. Subsequent collec- tions demonstrated that larger material was not exceptional when moisture was more abundant. The plants flower very late in the season, when most taxonomists have returned to other pursuits. This accounts, in part, for the long interval between initial discovery and this publication. Attempts at identification were thwarted by lack of similar material in Utah and Arizona herbaria and by the real lack of information in contemporary keys to the Portulacaceae. Ten- tatively we settled on an identification as T. validuhim Greene? Ultimate disposition of the plants as a new taxon awaited location of the type of that species at US. The type was taken in the Tusayan (now Kaibab) Forest Re- serve, Coconino Co., Arizona, at 2013 m, 11 August 1912 by R. R. Hill. The type was bor- rowed, through the kindness of the curator at the Smithsonian Institution, and compared with our material. The plants are strikingly similar but differ in stamen number, leaves that average longer, and larger flov/ers. The type specimens of T. validuhim consist of three specimens and a slide containing a mounted, dissected flower. A note on the sheet indicates that there are 12 stamens, not 10, as in our specimens. The Cedar Mountain talinum is described as follows: Talimim thompsonii Atwood & Welsh, sp. nov. Planta similis T. validulo Greene in radices, caudices, et staturas sed in folius et floribus majoribus et staminibus (10 nee 12) differt. Type.— USA: Utah: Emery Co., Cedar Mountain, east of Castle Dale, T19S, R12E, S18, pinyon-juniper-ponderosa pine commu- nity, on conglomeritic rock, at 2288 m elev., 19 July 1981, N. D. Atwood & R. Thompson 8056 (Holotype BRY; isotypes NY, POM, US). Perennial glabrous herbs from a fusiform or cylindric, reddish, tuberous root and a short perennating rootcrown bearing branches of the season; stems spreading, rosettelike, forming caespitose clumps to 10 cm wide; leaves 0.8-3.2 cm long, fleshy, cylindroid, to 3 mm wide when pressed, with auriculate, clasping base; flowers (1) 3-6 in cymes, ca 1 cm wide; petals pink; sepals 4.3-4.8 mm long, ovate, reticulately veined, greenish or brown- ish, with scarious margins, abruptly acumi- nate apically, tardily deciduous; stamens 10; capsules 6-6.5 mm long, 3.2-3.8 mm wide, Forest Service, U.S. Department of Agricultui Life Science Museum and Department of Bota Uinta National Forest, Provo, Utah 84603. and Range Science, Brigham Young University, Provo, Utah 84602. 485 Great Basin Naturalist Vol. 45, No. 3 486 Fi^. 1. Talinumthompsonii M^^ood^ Welsh,A.IlahU.H..VUaotl.avesana>ntWscence. July 1985 Atwood. VVelsh: Utah Talinum 487 keeled along the sutures apically; seeds gray- ish black, 1.2-1.3 mm long. Additional COLLECTIONS. — Utah. Emery Co., summit of Cedar Mt., ca 50 km SSW of Price, T19S, RUE, S13, Pinyon-juniper woods on conglomerate, 31 August 1970, S. L. Welsh, N. D. Atwood, & G. Moore 10781 (BRY); Ca 35 km due SE of Huntington, T19S, R12E, S7, 2166 m, S. L. Welsh & S. Clark 16134, 16166 (BRY). This low, clump-forming fleshy plant with beautiful pink flowers occurs on fused silicious conglomeratic gravel of the Buckhorn Con- glomerate Formation. It occurs with another rarity, Hymenoxys depressa (T. & G.) Welsh & Reveal, which is known from other sites in Emery County. However, the physical fea- tures of the summit of Cedar Mountain are hardly matched by any other in the vicinity. The substrate is present over a large region, but it does not occur in the same context or at the same elevation in any other area. The extent of the formation on Cedar Mountain is relatively large, standing above the north rim of the San Rafael Swell proper. It is difTicult to predict where the plant might be found else- where. The plant is named for Robert (Bob) Thompson, long-time collector and botanical enthusiast, who works for the U.S. F'orest Service in Price, Utah. References Greene, E. L. 1912. Miscellaneous specific types — VI. Leafl. Bot. Obs. & Grit. 2:270-272. Kearney, T. H., and R. H, Peebles. 1979. Arizona flora. University of Galifornia Press. Berkeley. 108.5 pp. TYPES OF NEVADA BUCKWHEATS (ERIOGONUM: POLYGONACEAE) James L. Reveal' Abstract. — The types and type specimens of Nevada species and varieties of Eriogonw alphabetically listed. Appropriate lectotypes are selected as necessary. (Polygonaceae) are The growing need for type information of Nevada plants for various projects, and espe- cially that by A. Tiehm, makes necessary the publication of critical type data o^Eriogonum (Polygonaceae), one of the state's largest gen- era. The format followed here is similar to that used by Welsh (1982). The entries are ar- ranged alphabetically by basionym. A number of Gandoger names proposed for Nevada buckwheats in 1906 and reported by Heller 1907 have been shown to be invalid (Reveal 1980). Readers are referred to Reveal (1985) for a taxonomic treatment of the genus of the state. Eriogonum anemophilum Greene, Pittonia 3:199. 1897. Pershing Co.: Windswept sum- mits at the north end of the West Humboldt Mts., probably on Star Peak, Jul 1894, Greene s.n. Holotype, NDG! Isotype, NY! Eriogonum angulosuni Benth. var. jlabella- tum Gand., Bull. Soc. Roy. Bot. Belgique 42:187. 1906. Washoe Co.: near the Central Pacific Railroad entrance into the Virginia Mts., 16 Jun 1894, Hillman s.n. Holotype, LY! Isotype, NESH! = E. maculatum A. A. Heller. Eriogonum angulosum Benth. var. patens Gand., Bull. Soc. Roy. Bot. Belgique 42:187. 1906. Washoe Co. : Wadsworth, 7 Aug 1899, Hillman s.n. Holotype, LY! Isotype, NESH [as 9 Jul 1899]! = E. maculatum i. A. Heller. Eriogonum angulosum Benth. var. pauci- florum Gand., Bull. Soc. Roy. Bot. Belgique 42:187. 1906. Washoe Co.: Reno, 25 Jun 1895, Hillman s.n. Holotvpe, LY! Isotypes, DS! NESH! UC! = E. maculatum A. A. Heller. Eriogonum argophyUum Reveal, Phytolo- gia 23:168. 1972. Elko Co.: In sandy washes on crusty mineralized sand below Sulphur Hot Springs, 7 Jul 1969, Holmgren ir Kern 3661. Holotvpe, US! Isotvpes, ID! IDS! KANS! MIN! MSC! NY! OKL! RM! RSA! UBC! UC! UTC! Eriogonum aridu7n Greene, Pittonia 3:200. 1897. ElkoCo.: Holborn, 16 Jul 1896, Greene s.n. Lectotype selected here, NDG! Dupli- cate of the lectotype, NDG! = E. umhellatum Torr. var. dichrocephalum Gand. Eriogonum azaleastrum Greene, Pittonia 5:67. 1906. Pershing Co.: Black Canyon, West Humboldt Mts., 29 Jul 1895, Greene s.n. Holotype, NDG! Isotype, NY! = E. um- hellatum Torr. var. aureum (Gand.) Reveal. Eriogonum haileiji S. Wats. var. por- phyreticum Stokes ex Jones, Contr. W. Bot. 11:17. 1903. Eureka Co.: Palisades, 6 Aug 1881, M. E.Jones s.n. Lectotype, POM! Du- plicate of the lectotype, DS! = E. baileyi S. Wats. var. baileyi. Eriogonum heatleyae Reveal, Aliso 7:415. 1972. NyeCo.: About0.9miNofU.S. Hwv. 6, 5.3 mi W of Salisbury Wash Rd., about iS mi E of Tonopah, 26 Jun 1971, Reveal et al. 2498. Holotvpe, US! Isotypes, ARIZ! ASC! ASU! BRY! CAS! G! GH! K! MICH! MO! NY! OSC! OKL! RENO! RM! RSA! SD! SMU! TEX! UC! UTC! WTU! Eriogonum bifurcatum Reveal, Aliso 7:357. 1971. Nye Co. : Pahrump Valley, 13 Jun 1970, Reveal 2283. Holotvpe, US! Isotvpes, ARIZ! ASC! ASU! BRY! CAS! COLO! G! GH! K! KANU! KSC! MICH! MO! NCU! NTS! NY! OKL! OSC! P! RM! RSA! SD! SMU! UC! UTC! WIS! WT! WTU! Department of Botany, University of Maryland, College Park. Maryland 20742, and National Museum of Natural History, Smithsonian Institution, Washington, DC. 20560. Research supported by National Science Foundation Grant BMS75-13063. This is Scientific Article A3838, Contribution No. 6818 of the Maryland Agricultural E.\periment Station. 488 July 1985 REVEAL: Nevada Eriogonum 489 Eriogonum ceniuum Nutt. var. tmilti- pedunculatum S. Stokes, Leafl. W. Bot. 2:48. 1937. Lander Co.: 40 mi W of Austin, 25 Aug 1931,;. T. Howell 7988. Holotype, CAS! Iso- types, GH! NY! US! = E. watsonii Torr. & Gray. Eriogonum cernuum Nutt. var. purpuras- cens Torr. & Gray, Kept. Explor. Surv. As- cert. Pract. Econ. Route Railroad Miss. River to Pacific Ocean 2:124. 1855. Washoe Co.: Mud Lake Valley, 16 Jun 1854, Snyder s.n. Holotype, NY! Isotypes, GH! MO! = E. nu- tans Torr. & Gray var. nutans. Eriogonum cernuum Nutt. var. tenueTorr. & Gray, Proc. Amer. Acad. Arts 8:182. 1870. Elko Co.: East Humboldt [now Ruby] Mts., 1869, Watson 1036. Lectotype selected here, GH! Duplicates of the lectotype, K! PH! = £. cernuum Nutt. var. cernuum. Eriogonum cernuum Nutt. subsp. viminale S. Stokes, Gen. Eriog. 41. 1936. Elko Co.: 44 mi SWofWendover, 24 Aug 1931,/. T. How- ell 7952. Holotype, CAS! Isotype, GH! = E. cernuum Nutt. var. viminale (S. Stokes) Re- veal in Munz. Eriogonum chrijsocephalum A. Gray subsp. desertorum Maguire, Leafl. W. Bot. 3:11. 1941. Elko Co. : Foothills of dry gravelly lake bar, 8 mi W of Wendover, 5 jun 1939, Holmgren 6- Lund 163. Holotvpe, UTC! Iso- types, NY! OKL! UC! US! = E. desertorum (Maguire) R. J. Davis. Eriogonum collinum Stokes ex Jones, Contr. W. Bot. 11:15. 1903. Washoe Co.: Reno, 19 Jun 1900, Stokes s.n. Lectotype, UC!, vide Madrono 18:169. 1966. Duplicates ■ of the lectotype, DS! MIN [as 20 Jun 1900]! y NY! SD! US! n Eriogonum commixtum Greene ex Tide- strom, Proc. Biol. Soc. Wash. 36:181. 1923. Carson City Co.: Eagle Valley, 31 Jul 1902, Baker 1402. Holotype, US! Isotypes, B! GH! MO! MSC! NY! POM! UC! = E. baileiji S. 1 1 Wats. var. praebens (Gand.) Reveal I Eriogonum comosum (M. E. Jones) M. E. Jones var. plaijanum M. E. Jones, Contr. W. Bot. 11:16. 1903. Clark Co., Mica Springs, 14 Apr 1894, M. E. Joties 5064bc. Lectotype selected here, POM! = E. pusillum Torr. & Gray. Eriogonum concinnum Reveal, Bull. Tor- rey Bot. Club 96:476. 1969. Nye Co. : Buck- board Mesa Rd. , near Cat Canyon and Timber Mtn., 5 Jul 1968, Reveal 1501. Holotype, UTC! Isotvpes, ARIZ! BRY! CAS! CS! DS! GH! IDS! isC! KSC! MIN! MO! MSC! NTS! NY! OKLA! RENO! RM! RSA! TEX! UC! UT! UTC! WIS! WTU! Eriogonum cusickii Gand. var. califor- nicum Gand., Bull. Soc. Roy. Bot. Belgique 42:193. 1906. Elko Co.: Little Lakes Canyon, near Stampede, 1 Jul 1902, Kennedy 543. Holotype, LY, not found at LY. Isotypes, NESH! RM! UC! = E. strictum Benth. sub.sp. proliferum (Torr. & Gray) S. Stokes var. pro- liferum (Torr. & Gray) Reveal. Eriogonum deflexum Torr. in Sitgr. var. nevadense Reveal, Phytologia 25:206. 1973. Nye Co.: Near Lunar Crater, 18 Jul 1972, Reveal ix Reveal 2785. Holotype, US! Iso- types, ARIZ! ASC! ASU! BRY! CAS! COLO! G! GH! ISC! K! MO! NY! OKL! OSC! P! RENO! RM! RSA! SD! SMU! TEX! UC! UTC! WIS! WT! WTU! Eriogonum elatum Dougl. ex Benth. var. erianthum Gand., Bull. Soc. Roy. Bot. Bel- gique 42:188. 1906. Elko Co.: Little Lakes Canyon near Stampede, 14 Jul 1902, Kennedy 563. Holotype, LY! Isotypes, NESH! RM! UC! = E. elatum Dougl. ex Benth. var. elatum. Eriogonum esmeraldense S. Wats., Proc. Amer. Acad. Arts 24:85. 1889. Esmeralda Co.: Miller Mtn., Jul 1888, Shockley 581. Lectotvpe selected here, GH! Duplicates of the lectotvpe, DS! JEPS! K! POM [as 381]\ UC! Eriogonum esmeraldense S. Wats. var. toiyabense]. T. Howell, Leafl. W. Bot. 6:178. 1952. Lander Co.: Majogany Canyon, Toiyabe Mts. , Linsdale ix Linsdale 550. Holo- type, CAS! Eriogonum exaltatum M. E. Jones, Contr. W. Bot. 15:61. 1929. Clark Co.: Riverside, 2 Jul 1927, M. E. Jones s.n. Holotype, POM! Isotypes, GH! NDG! US! = E. insigne S. Wats. Eriogonum exiiniumTidestrom, Proc. Biol. Soc. Wash. 36:181. 1923. Washoe Co.: near Franktown, 16 Aug 1912, Heller 10649. Holo- type, US! Isotypes, CAS! Cl! DS! DUKE! E! F! G! GH! LA! MO! NESH! POM! = E. ovali- folium Nutt. var. eximium (Tidestrom) J. T. Howell. Eriogonum gracile Benth. var. effusum Torr. & Grav, Proc. Amer. Acad. Arts 8:178. 490 Great Basin Naturalist Vol. 45, No. 3 1870. Carson City Co.: Near Empire City, 1865, Torrey 439. Lectotype selected here, GH! Duplicate of the lectotype, NY! = E. baileyi S. Wats. var. baileyi. Eriogonum heermannii Dur. & Hilg. var. clokeyi Reveal, Phytologia 34:437. 1976. Clark Co.: Lee Canyon, Spring Mts., 10 Aug 1966, Holmgren 6^ Reveal 2990. Holotvpe, US! Isotvpes, ARIZ! BRY! CAS! CS! DAO! DS! IDS! ISC! KSC! MIN! MO! MSC! NY! OKL! OSC! RM! RSA! UC! US! UT! UTC! Eriogonum heer7nannii Dur. & Hilg. subsp. humilius S. Stokes, Gen. Eriog. 90. 1936. Humboldt Co. : E of Golconda, Hot Springs Range, 22 Jul 1930, Keck 937. Holo- type, CAS! Isotypes, DS! MO! RSA! UC! = E. heermamiii Dur. & Hilg. var. humilius (S. Stokes) Reveal. Eriogonum heracleoides Nutt. var. virde Gand., Bull. Soc. Roy. Rot. Belgique 42:190. 1906. Washoe Co. : Peavine foothills, 15 Jun 1894, Hillman .s.n. Holotype, LY! Isotype, NESH! = E. umhellatum Torr. var. nevad- ense Gand. Eriogonum holmgrenii Reveal, Leafl. W. Bot. 10:184. 1965. White Pine Co.: W base of Pyramid Peak N of Johnson Lake, Snake Range, 10 Aug 1964, Holmgren ir Reveal 1576. Holotvpe, UTC! Isotvpes, ARIZ! BRY! CAS! DS! GH! IDS! K! KSC! MARY! MIN! MO! MSC! NY! OKL! OSC! PH! RENO! RM! RSA! UC! US! UT! UTC! WIS! Eriogonum howelliiS. Stokes, Gen. Eriog. 91. 1936. Eureka Co.: 34 mi W of Eureka, 24 Aug 1931,;. T. Hoivell 7963. Holotype, CAS! = E. heermannii Dur. &: Hilg. var. argense (M. E. Jones) Munz. Eriogonum inflatum Torr. &; Frem. var. contiguum Reveal, Aliso 7:221. 1970. Nye Co.: E of Ash Meadows, 3 Jun 1969, Reveal ir Matthews 2157. Holotype, US! Isotypes, BRY! CAS! MICH! MO! NTS! NY! OKL! RENO! RM! RSA! SD! UC! UTC! WIS! = E. contiguurn (Reveal) Reveal. Eriogonum kingii Torr. & Gray, Proc. Amer. Acad. Arts 8:165. 1870. Elko Co.: East Humboldt [now Ruby] Mts., Jul 1868, Wat- son 1020. Lectotvpe selected here: GH! Du- plicates of the lectotype: BM! K! NY! UC! US! Eriogontim laetum S. Stokes, Gen. Eriog. 23. 1936. Pershing Co. : Near Humboldt, 24 May 1901, George 9. Holotype, CAS! = E. ruhricaule Tk\e^hom. Eriogonum lemmonii S. Wats., Proc. Amer. Acad. Arts 12:266. 1877. Washoe Co.: Sand hills between Wadsworth and Reno, 1875, Lemmon 861. Holotvpe, GH! Isotvpes, ARIZ! ISC! NY! ?UC [as "Lava ridge, 1874"]! US! Eriogonum leucocladum Gand., Bull. Soc. Roy. Bot. Belgique 42:189. 1906. Washoe Co.: Reno, Jun 1898, Hillman s.n. Holotype, LY! = E. baileyi S. Wats. var. praebens (Gand.) Reveal. Eriogonum lewisii Reveal, Great Basin Nat.' 45. 1985. Elko Co.: White Elephant Butte, S of Elk Mtn., .30 Jul 1976, Reveal ir Reveal 4596. Holotvpe, US! Isotypes, BRY! CAS! F! MARY! MEXU! MICH! MO! OKL! RSA! TEX! UTC! Eriogonum microthecum Nutt. subsp. in- termedium S. Stokes, Gen. Eriog. 75. 1936. White Pine Co.: near Elv, 24 Aug 1931, /. T. Hoivell 7956. Holotype, CAS! Isotypes, GH! US! = E. microthecum var. .simpsonii (Benth in DC.) Reveal. Eriogonum microtJiecum Nutt. var. lapidi- cola Reveal, Brigham Young Univ. Sci. Bull., Biol. Ser. I3(l):28. 1971. Nye Co.: N end of Rainier Mesa at the S end of the Belted Range at the head of The Aqueduct, 25 Aug 1968, Reveal 6 Holmgren 1926. Holotvpe^ UTC! Isotvpes, ARIZ! BRY! CAS! CS! DS! GH! MO! MSC! NTS! NY! OKLA! RENO! RM! TEX! UT! WIS! Eriogonum monticola S. Stokes, Gen. Eriog. 123. 1936. Esmeralda Co.: S of Queen Mine, White Mts., 4 Sep 1926, Ferris 6753. Holotype, DS! Isotype, POM! = E. latens Jepson. Eriogonum nevadense Gand., Bull. Soc. Roy. Bot. Belgique 42:188. 1906. Washoe Co.: Near Reno, 1878, Hillman s.n. Holo- type, LY! = E. ochrocephalum S. Wats. var. ochrocephalum. Eriogonum nutatis Torr. & Gray var. bre- vipedicellatum S. Stokes, Gen. Eriog. 43. 1936. Eureka Co. : 30 mi W of Eureka, 25 Aug 1931,;. T. Howell 7974. Holotype, CAS! Iso- type, GH! E. ntitans Torr. & Gray var. nutans. Eriogonum nutans Torr. & Gray var. glabratum Reveal, Madrono 18:172. 1966. Elko Co.: 1 mi W of the Deeth turnofl', 20 Jun -All .iWir.-viations follow those rccoini.HMuU-d In B-l'-H Botanico- pciiodii imi-luintianiMii fxct-pt that for C:rcat Basin Naturalist. July 1985 REVEAL: Nevada Eriogonum 491 and 14 Jul 1964, Holmgren 6 Reveal 1037. Holotype, UTC! Isotypes, ARIZ! BRY! CAS! DS! GH! KSC! MO! MSC! NY! OKL! RENO! RM! RSA! SMU! UC! US! UTC! WIS! Eriogonum ochrocephahim S. Wats, in Brewer & Wats., Bot. Calif. 2:480. 1880. Washoe Co.: "Valleys of Northwestern Ne- vada," likely from near Pyramid Lake, sin date, Lemmon s.n. (perhaps 91). Holotype, GH! Isotype, NY! Eriogonum ochrocephahim S. Wats. var. alexanderae Reveal, Great Basin Nat. 45:000. 1985. Lyon Co.: Wilson Canyon between Smith and Mason, 12.8 mi NE of Smith, 21 Jun 1978, Reveal et al. 4737. Holotype, US! Isotypes, BRY! CAS! DUKE! F! MARY! MEXU! MICH! MO! OKL! RSA! TEX! UTC! Eriogonum ovalifolium Nutt. var. cae- lestinum Reveal, Great Basin Nat. 32:115. 1972. Nye Co. : South Fork of Pine Creek, Toquima Range, 23 Jul 1964, Reveal 629. Holotvpe, US! Isotypes, ARIZ! BRY! DS! GH! MIN! MO! NY! OKL! OKLA! RENO! RM! RSA! UC! UT! UTC! WTU! Eriogonum ovalifolium Nutt. var. nevad- ense Gand., Bull. Soc. Roy. Bot. Belgique 42:193. 1906. Washoe Co.: Hills near the Truckee River, 4 May 1901, Kennedy s.n. Holotype, LY! Isotypes, MIN! NESH! RENO!RM!UC! Eriogonum ovalifolium Nutt. var. williamsae Reveal, Brittonia 33:446. 1981. Washoe Co. : Steamboat Springs, 7 Jun 1979, Williams 6- Tiehm 79-71. Holotype, US! Iso- types, MARY and to be distributed. Eriogonum praebens Gand. var. divarica- tum Gand., Bull. Soc. Roy. Bot. Belgique 42:196. 1906. Washoe Co.: Peavine foothills, 27 Sep 1893, Hillmans.n. Holotype, LY! = E. baileyi S. Wats. var. praebens (Gand.) Re- veal. Eriogonum puberulum S. Wats. var. veno- sum S. Stokes, Gen. Eriog. 35. 1936. Eureka Co.: 34 mi E of Eureka, 24 Aug 1931, /. T. Howell 7962. Holotype, CAS! = E. puberu- lum S. Wats. Eriogonum pusillum Torr. & Gray, Proc. Amer. Acad. Arts 8:184. 1870. Churchill Co. or Pershing Co. : Foothills of the Trinity Mts. , May 1868, Watson 1037. Holotype, GH! Iso- types, K! NY! US! Eriogonum reliquimi S. Stokes, Leafl. W. Bot. 2:52. 1937. Clark Co. : Clarleston Park, Spring Mts., 9 Aug 1937, Clokey 7491. Holo- tvpe, CAS! Isotypes, AHFH! ARIZ! B! BALT! BM! BRY! CAN! CAS! CI! DAO! DS! DUKE! F! G! GH! lA! IDS! IND! ISC! K! LA! ECU! MARY! MICH! MIN! MO! MSC! NEB! NESH! NY! OKL! OKLA! ORE! OSC! P! PAG! PENN! PH! POM! RENO! SD! TEX! UC! US! UT! UTC! WILLU! WIS! WS! WTU! = E. panamintense Morton var. panamintense . Eriogonum reniforme Torr. & Frem. var. asarifolium Gand., Bull. Soc. Roy. Bot. Bel- gique 42:196. 1906. Washoe Co. : Reno, 9 Aug 1894, Hilhnan s.n. Holotype, LY! Isotype, NESH! = E. pusillumTorr. & Gray. Eriogonum reniforme Torr. & Frem. var. comosum M. E. Jones, Proc. Calif. Acad. Sci. II, 5:719. 1895. Clark Co.: Near Hole in the Rock, 10 mi above Stone's Ferry, 12 Apr 1894, M. E. Jones 5036ao. Lectotype, POM! = E. reniforme Torr. & Frem. Eriogonum restioides Gand., Bull. Soc. Roy. Bot. Belgique 42:199. 1906. Washoe Co.: Reno, 9 Aug 1894, Hilhnan s.n. Lecto- type, LY! Duplicate of the lectotype, NESH! = E. baileyi S. Wats. var. baileyi. Eriogonum revolutum Goodding, Bot. Gaz. 37:54. 1904. Clark Co. : In low mountains near the Virgin River near Bunkerville, 9 May 1902, Goodding 753. Holotype, RM! Iso- types, F! G! GH! ISC! LY! MO! NEB! NESH! NY! P! POM! RM! UC! US! UT! UTC! = Eriogonum fasciculatum Benth. var. poli- folium (Benth. in DC.) Torr. & Gray. Eriogonum rhodanthum Nels. & Kenn., Proc. Biol. Soc. Wash. 19:35. 1906. Washoe Co.: Mt. Rose, 17 Aug 1905, Kennedy 1184. Holotype, RM! Isotypes, NESH! NY! UC! = E. ovalifolium Nutt. var. nivale (Canby in Gov.) M. E. Jones. Eriogonum robustium Greene, Bull. Calif. Acad. Sci. 1:126. 1885. Storey Co.: Geiger Grade, NW of Virginia City, Jul 1884, Curran s.n. Holotype, CAS! Isotypes, BM! F! GH! = E. lobbii Torr. & Gray var. robustium (Greene) M. E. Jones. Eriogonum rosen.se Nels. & Kenn., Proc. Biol. Soc. Wash. 19:36. 1906. Washoe Co.: Summit of Mt. Rose, 17 Aug 1905, Kennedy 1180. Holotype, RM! Isotypes, NDG! NESH! UC! Eriogonum rubricaule Tidestrom, Proc. Biol. Soc. Wash. 36:181. 1923. Churchill Co.: 492 Great Basin Naturalist Vol. 45, No. 3 Near Lahontan, 21 May 1916, Headley 4. Holotype, US! Isotype, RENO! Eriogonum rupinum Reveal, Aliso 7:226. 1970. Nye Co.: below Rose Spring, Cedar Pass, southern Kawich Range, 9 Aug 1969, Beatley9458. Holotype, US! Isotypes, ARIZ! ASC! ASU! BRY! COLO! DUKE! G! GH! ISC! K! KANU! KSC! MICH! MO! NCU! NTS! NY! OKL! OSC! P! RM! RSA! SMU! SD! TEX! UC! UT! UTC! WTU! Eriogonum saxatile S. Wats, subsp. miilti- caule S. Stokes, Leafl. W. Bot. 3:201. 1943. Esmeralda Co.: Gold Mtn., 5 Jul 1941, Alex- ander ir Kellogg 2478A. Holotype, UC! Iso- type, CAS! = E. saxatile S. Wats. Eriogonum sericoleiiciim Greene ex Tide- strom, Proc. Biol. Soc. Wash. 36:182. 1923. Carson City Co. : King's Canvon, 4 Jun 1902, Baker984. Holotvpe, US! Isotvpes, B! F! GH! MO! MSC! NY! POM! RM! VT! = E. caespito- sum Nutt. Eriogonum shockleifi S. Wats., Proc. Amer. Acad. Arts 18:194. 1883. Mineral Co.: Candelaria, Jun 1882, Shockleij 248. Holo- tvpe, GH! Isotvpes, DS! ISC! JEPS! MICH! NY!P!POM!UC!US! Eriogonum tenellum Torr. ex Benth. in DC. var. sessiliflorum Gand., Bull. Soc. Roy. Bot. Belgique 42:198. 1906. Washoe Co.: Reno, Sep 1894, Hillman s.n. Holotype, LY! Isotype, NESH! = £. microtheciim Nutt. var. laxiflorum Hook. Eriogonum thurberi Torr. var. acutangu- lum Gand., Bull. Soc. Roy. Bot. Belgique 42:198. 1906. Washoe Co.: Virginia Mts., 27 Aug 1894, Hillman .s.n. Holotype, LY! = E. maculatum A. A. Heller. Eriogonum tiehmii Reveal, Great Basin Nat. 45:493-519. 1985. Esmeralda Co.: Silver Peak Range N of the road fiom Silverpeak to Fish Lake Valley, 1.2 air mi NNW of Cave Springs, sec. 27, TIS, R37E, 1830 m eleva- tion, 18 May 1983, Tiehm et al. 8534. Holo- type, US! Isotvpes, BRY! CAS! MARY! NY! RENO! RSA! UTC! and to be distributed. Eriogonum umhellatum Torr. var. califor- nicum Gand., Bull. Soc. Roy. Bot. Belgique 42:199. 1906. Washoe Co.: Peavine foothills, 11 Jun 1894, Hillman s.n. Holotype, LY! Iso- type, NESH! = E. umhellatum Torr. var. nevaden.se Gand. Eriogonum umhellatum Torr. vsiv. jtinipor- inum Reveal, Great Basin Nat. 45. 1985. White Pine Co. : Sacramento Pass, N of the Snake Range, 13 Aug 1975, Reveal ir Reveal 3925. Holotvpe, US! Isotvpes, BRY! CAS! MARY! OKL! TEX! Eriogonum umhellatum Torr. var. nevaden.se Gand., Bull. Soc. Roy. Bot. Bel- gique 42:198. 1906. Washoe Co.: HuflFakers, near Reno, 27 Mav 1893, Hillman s.n. Holo- type, LY! IsotypcNESH! Eriogonum umhellatum Torr. var. suh- aridumS. Stokes, Leafl. W. Bot. 2:53. 1937. Clark Co.: Kyle Canyon, Spring Mts., 17 Jul 1937. Clokei/ [6 Clokey] 7492. Holotvpe, CAS! Isotvpes, B! BALT! BR! BRY! CAN! CAS! CI! DS! DUKE! GH! lA! ILL! IND! K! KANU! LA! LAM! LL! MICH! MIN! MO! MSC! NESH! NY! OKL! OKLA! P! PAG! PENN! PH! POM! RM! TEX! UC! US! UTC! VT! WIS! WVA! Eriogonum umhellatum Torr. var. vernum Reveal, Great Basin Nat. 28:157. 1968. Nye Co.: N end of Shoshone Mtn. along Buck- board Mesa Rd. , 1 mi W of Tippipah Spring, 4 Jun 1968, Reveal 1139. Holotvpe, UTC! Iso- tvpes, ARIZ! BRY! CAS! CS! DS! GH! IDS! KSC! MIN! MO! MSC! NTS! NY! OKLA! RENO! RSA! TEX! US! UT! UTC! WIS! Eriogonum viscidulum J. T. Howell, Leafl. W. Bot. 3:138. 1942. Clark Co.: Virgin River Bridge, Riverside, 8 Mav 1941, Eastwood ir Howell 9031. Holotype, CAS! Isotype, CAS! Eriogonum watsonii Torr. & Grav, Proc. Amer. Acad. Arts 8:182. 1870. Elko Co.: East Humboldt [now Ruby] Mts., 1865, Torrey 450. Lectotype, NY! Duplicates of the lecto- type, NY! US! Literature Cited Reveal, J. L. 1980. The genus Eriogomnn Michx. (Polyg- onaceae ) and Michel Gandoger. Great Basin Nat. 40:143-148. 1985. Annotated key to Eriogoiititn (Polygo- naceae) of Nevada. Great Basin Nat. 45:493-519. Welsh. S L. 1982. Utah plant types — historical perspec- tive 1840 to 1981— annotated list, and hihliogra phv. Great Basin Nat. 42:129-195. ANNOTATED KEY TO ERIOGONUM (POLYGONACEAE) OF NEVADA Janu's L. Reveal' Abstract. — Seventy-three species ofEriogonum (Polygonaceae) are reported for Nevada. A key is provided for the identification of these species along with brief notes on their distribution in the state and elsewhere. The genus Eriogonum is a prominent mem- The key is followed by a comment section, ber of the Nevada flora, and, except for the The arrangement of the species in the key is treatment of the genus by Tidestrom (1925), ^^^.g^jy ^^^jfi^.^l ^^^^ ^j^^^^jj ^^^^ ^^ ^^j^^^ ^^ no key tor the native species is available. The ^ ^. r ^ i rr- •.• t i \ , ^ .11 1 r representative ot natural afimities. In the present treatment has been prepared tor a , doctoral dissertation by John Kartesz, a grad- comment section species are arranged in a uate student at the University of Nevada, who sequence that may be regarded as more natu- is writing a manual on the flora of Nevada, ral. The distribution data given for Nevada is New entities discussed in this treatment may detailed, but that for adjacent states or regions be found in a companion paper (Reveal 1985). is more generalized. 1. Plants perennial, not annual, but see E. inflatum with its inflated stems and yellow, hirsute flowers 2 — Plants annual, or if perennial then stems inflated and flowers yellow, hirsute, and in pedunculated involucres 41 2(1). Flowers not stipelike at the base 3 — Flowers stipelike at the attenuated base, sometimes weakly so 33 3(2). Plants distinctly shrubby or subshrubby, woody above the basal caudex and not dying back completely to the ground after each year 4 — Plants herbaceous, cespitose or pulvinate perennials, not at all shrubby or subshrubby 9 4(3). Flowers pubescent without, 2.5-3 mm long, white to pink; low shrubs; Esmeralda, Nye, Lincoln, and Clark cos 6. E. fasciculatum — Flowers glabrous without 5 5(4). Stems and branches smooth, glabrous to tomentose, not angled or scabrellous; inflorescences with involucres arranged in loose to compact terminal cymes or racemosely along the straight branches, the branches not zigzag 6 n — Stems and branches angled or ribbed, or if smooth then obviously scabrous, or, if smooth and tomentose, then inflorescences of zigzag branches 8 6(5). Inflorescences cymose throughout with involucres dichotomously arranged even at the tips of the branches; leaves less than 8 mm wide, or, if broader, then flowers yellow 7 Department of Botany, University of Maryland, College Park, Maryland 20742, and National Museum of Natural History, Smithsonian Institution, Washington, DC. 20560. Research supported by National Science Foundation Grant BMS75-13063. This is Scientific Article A3839, Contribution No. 6819 of the Maryland Agricultural Experiment Station. 493 494 Great Basin Naturalist Vol. 45, No. 3 — Inflorescences large with numerous branches and branchlets bearing racemosely arranged involucres at their tips; flowers white; White Pine and Lincoln cos. westward across central Nevada to Esmeralda, Mineral, and Ormsby cos 2. E. nummulare 7(6). Leaf-apices acute, the blade less than 8 mm wide; flowers white, or, if yellow, then plants of western Nevada; widespread and common throughout the state L E. microthecum — Leaf-apices rounded, the blades .5-L5 mm wide; flowers yellow; local, near Las Vegas, Clark Co 3. £. corymbosum 8(5). Stems and branches angled, ribbed or smooth and scabrous; involucres 0.7-2 mm long; throughout the lower two-thirds of the state 4. E. heermannii — Stems and branches tomentose; involucres 2-2.5 mm long; rare, southernmost Clark Co 5. E. plumatella 9(3). Plants cespitose or pulvinate perennials with a terminal, capitate cluster of involucres 10 — Plants upright herbs with open, often elongated or much-branched in- florescences 27 10(9). Tepals monomorphic, glabrous or pubescent 11 — Tepals strongly dimorphic, the outer tepals broader than inner tepals, or, if only slightly dimorphic, then the plants with oval to rotund leaf-blades 2-8 mm long and wide and restricted to the high mountains; common nearly throughout the state 28. E. ovalifolium 11(10). Flowers glabrous or glandular without, not pubescent with long hairs 12 — Flowers hairy pubescent without, not glabrous or glandular 26 12(11). Involucres rigid and distinctly tubular 13 — Involucres membranaceous and indistinctly forming a tube, mostly 2-3.5 mm long, glandular or glandular-hairy without 23 13(12). Scapes glabrous or tomentose, not glandular or glandular-hairy 14 — Scapes glandular-hairy or glandular, not glabrous or tomentose 22 14(13). Scapes glabrous 15 — Scapes tomentose to floccose 16 15(14). Leaves ovate to obovate, 1-2 cm long, .5-10 mm wide; scapes 1-2 dm long; involucres floccose without; clay hill or flats, Storey, Lyon and Washoe cos. 8. £. ochrocephalum — Leaves spathulate to oblanceolate, 0.4-0.8 (1) cm long, 2-3 (4) mm wide; scapes 0.2-0.8 dm long; involucres glabrous or with a few hairs on the teeth; volcanic slopes and outcrops in northern Washoe Co 9. £. prociduum 16(14). Flowers bright yellow 17 — Flowers pale yellow or more often whitish with a reddish midrib, 2-3 mm long; involucre 2-5 mm long 20 17(16). Involucres 3.5—5 mm long, floccose over the upper half of the tube; leaf-blades oblong to lanceolate, (1) 1.5-2.5 cm long, the petioles 1..5-3 cm long; scapes (0.6) 1-J.5 cm long; Lyon Co. northward to southern Washoe and Pershing cos. 8. E. ochrocephalum — Involucres 2-3 mm long 18 July 1985 REVEAL: Nevada Eriogonum 495 18(17). Leaf-blades 1.5-2.5 (3) cm long, 5-9 mm wide; scapes densely tomentose; involucres tomentose without; flowers 2.. 5-3 mm long; Elko Co. south to north- ern White Pine Co. and westward to adjacent northern Eureka Co., 170(>- 2630 m elev 7. £. desertorum — Leaf-hlades less than 1.5 cm long and 7 mm wide; scapes floccose; involucres floccose without and then mainly on the upper half of the tube; flowers mostly less than 2.5 mm long 19 19(18). Flowers (2) 2.5-3 mm long; involucres 2.5-3 mm long; leaf-blades 4-7 mm wide; high elevation rocky slopes in Elko Co 10. £. lewisii — Flowers 1.5-2 mm long; involucres 2-3 mm long; leaf-blades 1..5-4 (5) mm wide; low elevation clay outcrops in Humboldt Co 11. E. crosbyae 20(16). Involucres 2-2.5 mm long, floccose without; leaf-blades broadly elliptic to obovate or suborbicular, (7) 9-13 (15) mm long, .5-9 (11) mm wide; flowers glabrous; Lander, Pershing, and southwestern Humboldt cos. . 14. E. anemophilwn — Involucres 4—5 mm long, sparsely floccose without; leaf-blades elliptic to oblong, (8) 12-25 mm long, .5-8 (10) mm wide; flowers sparsely glandular without; Esmeralda Co 15. E. tiehmii 21(13). Involucres tomentose without, 3-4 mm long; leaf-blades 1-2 cm long; flowers 2-2.5 mm long; scapes 0.7-2 dm long; clay hills mainly in Washoe Co., mostly below 2000 m elev 8. E. ochrocephalum — Involucres glandular-hairy or glandular, not tomentose; plants above 2000 m elev. or, if at lower elevations, then not of northwestern Nevada 22 22(21). Leaves oblanceolate, 4—15 mm long, 2.5-5 (6) mm wide; involucres more or less turbinate, 2.. 5-3 mm wide with (5) 6-8 teeth; achenes 1.5-2 mm long; montane forests and alpine areas in the Sierra Nevada and adjacent ranges of Esmeralda Co. northward to Washoe Co. mostly 2440-3300 m elev 12. E. rosense — Leaves elliptic, 10-25 mm long, 5—16 mm wide; involucres campanulate, 3-4 mm wide with 5 teeth; achenes (2.5) 3-3.5 mm long; clay hills and slopes in Lander, Eureka, Mineral, Nye, and Churchill cos., 1700-2750 m elev 13. E. beatleyae 23(12). Flowers greenish yeflow or pale yellow, not white or rose 24 — Flowers white to rose or red, not greenish yellow or pale yellow 25 24(23). Involucres (2.8) 3-3.5 mm long; petioles 4-12 (15) mm long, thinly tomentose; flowers (2.. 5) 3-3.5 mm long, glabrous; rocky outcrops at high elevations, 2250-3250 m, East Humboldt and Ruby mts., Elko Co. and Cherry Creek Range, northern White Pine Co 16. E. kingii — Involucres 2-2.5 mm long; petioles 0.5-1 (1.5) mm long, densely tomentose; flowers 2.. 5-3 mm long, sparsely glandular; clay flat near Sulphur Hot Springs, 1850 m, Ruby Valley, Elko Co 17. £. argophylhim 25(23). Leaves 1-2 cm long, densely white-tomentose below, less so and white above; scapes 3-8 cm long, glandular, not floccose; inflorescences of 5-7 involucres; pedicels glandular at the tips; high elevation sandy to granitic outcrops, White Mts. , Esmeralda Co 18. £. gracilipes — Leaves 0.3-1 cm long, densely white-tomentose and greenish below, less so and greenish above; scapes up to 3 cm long, floccose and glandular; inflorescences of 2-4 involucres; pedicels glabrous except for a few scattered glands at the base; high elevation limestone and quartzite outcrops, Snake Range, White Pine Co. 19. E. holmgrenii 496 Great Basin Naturalist Vol. 45, No. 3 26(11). Ovaries and achenes glabrous; plants loosely cespitose with 10-20 rosettes; branches prostrate to weakly erect, 2-8 cm long; inflorescences cymose- umbellate to more or less capitate; local and infrequent on limestone gravel. Eureka, Nye, White Pine, and Lincoln cos 20. E. villiflorum — Ovaries and achenes pubescent; plants densely cespitose with 15-50 or more rosettes; branches erect, up to 3 cm long; inflorescences capitate; common on clay to gravel flats and slopes from Esmeralda to Nye and Clark cos. eastward to Elko and Lincoln cos 21. E. shockleyi 27(9). Tepals monomorphic; plants not with elliptic leaves .5-15 mm long and tomentose stems 28 — Tepals dimorphic; plants 1-3 dm tall, the branches and stems tomentose; leaves mostly elliptic, .5-25 mm long, .5-15 mm wide, tomentose on both surfaces; flowers white or yellow; northern and northwestern Nevada 29. E. strictiim 28(27). Inflorescences with involucres racemosely arranged along the branches, involucres solitary 29 — Involucres with involucres in cymes with usually clustered involucres dichotomously arranged throughout the inflorescence 32 29(28). Plants suffrutescent and branched at the base; leaves many, oblanceolate to elliptic, 0..5-1 cm long, 2-5 mm wide; involucres turbinate, 2-2.5 mm long; flowers white; rare and local in southern Clark Co. or common from Mineral Co. northward 22. E. wrightii — Plants not suffrutescent at the base; basal leaves few, more than 1 cm long and 5 mm wide 30 ,30(29). Basal leaves roundish to broadly ovate, the leaf-blades 1..5-4 cm long, on petioles 1-5 cm long; plants with branched and woody spreading caudices; involucres and flowers .3-5 mm long; dry rocky slopes of Esmeralda Co. southeastwardly to the Spring Mts. and Sheep Range of Clark Co 23. E. panamintense — Basal leaves oblong, cordate, ovate or elliptic, the leaf-blades (2) .3-10 cm long, on petioles .3-10 cm long; plants arising from a single, woody, mostly unbranched caudex; involucres 2-5 mm long; flowers 2-4 mm long; Mineral and Nye cos. eastward to White Pine, Lincoln, and northern Clark cos 31 31(.30). Inflorescences cymose with 3-5 racemostly arranged involucres at the ends of the branches; plants 3-5 dm high with 3-7 stems arising from the caudex, rarely with 7-20 stems on somewhat more spreading caudices in some; involucres 2-3.5 mm long; widespread and common from Mineral and Esmeralda cos. across the central portion of the state to Lander and Nye cos 24. E. rupinum — Inflorescences cymose with .5-20 or more racemosely arranged involucres at the end of the branches; plants 3-10 dm high with 1-3 (5) stems arising from the compact caudex; widespread and locally common in Lincoln, White Pine, Nye, and northern Clark cos 25. E. racemosum 32(28). Leaves tomentose below, somewhat less so above, the leaf-blades spreading with apices mostly obtuse; flowers glabrous; along the western edge of the state from Douglas and Carson City cos. northward 26. £. nudum — Leaves villous and green on both surfaces, the leaf-blades erect, with acute apices; flowers thinly pubescent; rather common throughout the northern half of the state 27. E. datum .33(2). Flowers with a long, tubular, more or less winged stipe at the base of white flower, together (3) .5-7 mm long, glabrous, the tepals dimprohic; leaves broadly obovate to roundish, 1-2 cm long, 1-1.5 cm wide, lanate to tomentose on both surfaces; flowering branches .5-10 cm long; inflorescences 0..5-1.5 dm long; rare and local in Clark, Nve, and Esmeralda cos 30. E. saxatile July 1985 REVEAL: Nevada Erioconum 497 — Flowers abruptly stipitate, the stipe sometimes obscure and never winged; flowers generally yellow or cream 34 34(33). Involucres with lobes at least half as long as the tube, the lobes usually reflexed or spreading, never erect and toothlike 35 — Involucres with lobes much shorter than the tube, toothlike and erect or nearly so 39 35(34). Flowers glabrous without 36 — Flowers pubescent without 37 36(35). Flowering stems not bracteated near the middle; leaf-blades oblong to elliptic; inflorescences umbellate to once or twice compoundly umbellate; common throughout the state 31. E. umbellatum — Flowering stems bracteated near the middle; leaf-blades linear to oblanceolate; inflorescences compoundly umbellate three to several times; common across the northern half of the state 32. E. heracleoides 37(35). Flowering branches with a whorl of subtending bracts at the base of the umbel or near the middle of the ray 38 — Flowering branches without subtending bracts and with a solitary, terminal involucre; widespread and common throughout the Great Basin portion of the state 35. E. caespitosum 38(37). Involucres more than 1, umbellate, subtended by (2) 3-several leafy bracts below the umbel, or, if seemingly in the middle of the flowering branch, then the leaves glabrate above; leaves 1-3 (4) cm long; flowers 5-9 mm long; Carson City north and eastward to western Elko Co 33. E. sphaerocephalum — Involucres sohtary, terminal, not immediately subtended by leafy bracts, the flowering branches with a whorl of bracts near the middle; leaves densely tomentose on both surfaces; flowers 5-8 mm long; Peavine Mtn. area of southern Washoe Co 34. E. douglasii 39(34). Flowers pubescent without, cream to pale yellow, 5-6 mm long; leaves short- pilose to subglabrous, 1-3 cm long; White Mountains, Esmeralda Co. . . 37. E. latens — Flowers glabrous without; leaves tomentose at least on the lower surface 40 40(39). Flowering branches erect, thinly tomentose; involucres 2-3 mm long; flowers unisexual, the male flowers yellow, 1.5-3 mm long, the female flowers yellow to lemon yellow, 4-7 mm long; infrequent, Carson City, Washoe, and Humboldt cos 36. E. inarifolium — Flowering branches prostrate to weakly erect, usually densely tomentose; in- volucres 5-15 mm long; flowers perfect, white to rose or red, 5-9 mm long; locaUy frequent in Washoe, Storey, Lyon, and Carson City cos 38. E. lohhii 41(1). Involucres smooth, not ribbed or angled, usually distinctly peduncled, or if sessile then involucres not vertically appressed to the stems; annuals except for E. inflatwn 42 — Involucres angled to strongly ribbed, usually tightly appressed to the stem and always sessile; strictly annuals 68 42(41). Leaves basal, not cauline, occasionally the leaves sheathing up the base 43 — Leaves basal and cauline at the lower nodes 66 43(42). Leaves glabrous, pilose, hispid or villous on one or both surfaces, not densely tomentose at least on the lower surface; flowers mostly yellow 44 — Leaves densely tomentose to floccose-tomentose on one or both surfaces; flowers mostly white, glabrous or glandular-puberulent 52 498 Great Basin Naturalist Vol. 45, No. 3 44(43). Flowers pubescent without 45 — Flowers glabrous without 49 45(44). Plants glabrous, or, if glandular, the glands infrequent and restricted to the base and lower nodes of the stems and branches 46 — Plants glandular, the glands dense and frequent throughout the plant 48 46(45). Involucres 5-toothed 47 — Involucres 4-toothed; plants strictly annual; flowering stems glabrous or hirsute at the base, green to yellowish; widespread and common in southern Nye and Lincoln cos. and throughout Clark Co 41. E. trichopes 47(46). Flowering stems glabrous and glaucous or with a few hirsute hairs at the base, grayish or green; plants first-year flowering perennials; widespread and common in the southern third of the state 39. E. inflatiun — Flowering stems glandular at the base and occasionally at the lower nodes, otherwise glabrous above, reddish; plants strictly annual; low desert valleys and foothills in extreme southern Nye and northern Clark cos 40. E. contigiium 48(45). Flowers yellow, 1-1.5 (2) mm long; involucres turbinate-campanulate, 1.3—2 mm long on erect peduncles 3—10 mm long; achenes 1.-5-1.8 mm long; Nye Co. eastward to Lincoln, White Pine and Elko cos. mostly on volcanic or limestone ranges 42. E. howellianum — Flowers white, 1-1.8 mm long; involucres narrowly turbinate, 0.8—1.2 (1.5) mm long on deflexed peduncles 2-5 mm long; achenes 1-1.3 mm long; southwestern Nye and northern Clark cos. on low limestone desert ranges .... 43. E. glandidosiim 49(44). Involucres long peduncled at least at the lower nodes 50 — Involucres sessile or the lowermost short-pedunculate, the tube campanulate, 2-2.5 mm long, 2.. 5-3 mm wide; flowers pink to rose, 1..5-2 mm long; clayey foothills and flats in northwestern Nevada 47. E. lemmonii 50(49). Flowers white to greenish white; involucres turbinate; peduncles descending .... 51 — Flowers pale yellow to yellow; involucres campanulate, 2. 5-3 mm long and wide; peduncles 2—3 cm long, erect; volcanic ranges in northwestern Nevada 46. E. ruhhcaule 51(50). Peduncles slender to filiform, 2-15 mm long, spreading to deflexed; leaves obovate to round-obovate with tapering leaf-bases; involucres narrowly tur- binate, 0.8-1.8 mm long, 0.. 5-1. 2 mm wide; flowers white to pink or red; gravelly slopes from Humboldt and Mineral Co. south to Nye Co. eastward to Lander and Eureka cos 44. £. esmcraldense — Peduncles slender, (0.5) 1-2.5 mm long, sharply deflexed; leaves subcordate with cordate leaf-bases; involucres turbinate, 1.2-1.6 mm long, 1-1.4 mm wide; flowers white to greenish white; clayey or sandy-clay soils in southwestern Nye Co 45. E. concinnum 52(43). Outer tepals cordate or subcordate at the base, mostK oblong to orbicular 53 — Outer tepals truncate to obtuse at the base 58 53(52). Involucres deflexed, sessile or on peduncles up to 25 mm long 54 — Involucres erect on peduncles less than 5 mm long 57 54(53). Stems and branches glabrous 55 — Stems and i)ranches glandular, stoutish and usually short, the crowns flat- topped; peduncles up to 15 mm long; in\{)lucrcs turbinate to campanulate, 1-2.5 mm long; common throughout the southern half of the state . . . 51. E. hrachijpodum July 1985 REVEAL: Nevada Eriogonum 499 55(54). Involucres narrowly turbinate to tiuhinate-campanulate; peduncles up to 25 mm long; tepals as long as to longer than wide, white to pink 56 — Involucres campanulate or hemispheric, 1-2 mm long; peduncles lacking; flow- ers yellow to reddish yellow; scattered throughout much of central Nevada 50. £. hookeri 56(55). Involucres 1..5-3 mm long; plants variously branched; flowers not gibbous at the base; common throughout the southern two-thirds of the state 48. E. deflexum — Involucres 1-1.5 mm long; plants branched in a series of layers one above the other so as to be pagodalike; flowers gibbous at the base when mature; infrequent in extreme southern Nye Co 49. E. rixfordii 57(53). Flowering branches short, less than 3 cm long; plants 1-4 dm high, 3-15 dm across, the crown spreading and flat-topped; extreme southern Nye Co. and Clark Co 52. E. bifurcatum — Flowering branches long, 2-20 cm long; plants 2-12 dm high, 1-5 dm across, the crowns erect and strict; southern Nye and Clark cos 53. E. insigne 58(52). Flowers smooth or saccate, glabrous or glandular, not strongly pustulose 59 — Flowers strongly pustulose without; northwestern Nevada 65 59(58). Tepals monomorphic, mostly oblong to ovate 60 — Tepals dimorphic, or, if similar, then glandular-pubescent, pandurate to flabellate or ovate 61 60(59). Plants glabrous; involucres 2-3 mm long, 5-toothed; peduncles deflexed, slen- der; flowers white, 2-2.5 mm long; local in northwestern and northcentral Nevada 54. E. watsonii — Plants minutely viscid; involucres 1-1.2 mm long, 4-toothed; peduncles erect, filiform; flowers yellow, 1.3-2 mm long; rare and local in Clark Co ". 60. £. visciduhim 61(59). Flowers glabrous without, not glandular 62 — Flowers glandular without 63 62(61). Peduncles glabrous, cernuous to ascending, straight or nearly so, lacking in var. viminale; involucres turbinate, 1-1.5 mm wide; flowers white, the outer tepals pandurate, crisped along the margin; common throughout nearly all of the state 55. £. cernuum — Peduncles glandular, or, if glabrous, then curving downwardly; involucres cam- panulate, 1.5-2.5 mm wide; flowers white to rose, the outer tepals oblong to oval, not crisped along the margin; infrequent and local across the northern half of the state 56. £. nutans 63(61). Outer tepals saccate-dilated at the base, usually white when mature; involucres 0.6-1.2 mm long, glabrous; southern Nye, Lincoln, and Clark cos. . . 57. £. thomasii — Outer tepals smooth; involucres 1-2 mm long, glabrous or glandular without; flowers yellow 64 64(63). Tepals and outer involucral surface glandular-puberulent; bracts glandular on the outer surface; western and southern Nevada 58. £. pusillum — Tepals glandular-puberulent without; involucres glabrous on the outer surface; bracts villous on the outer surface; Churchill Co. southward through the southern half of the state 59. £. reniforme 65(58). Upper involucres peduncled, the peduncles curving upwardly, 1-5 cm long, the involucral tube (1.5) 2-3 mm long; flowers white to yellow, 1-2.5 mm long; achenes 2-2.5 mm long; Washoe, Douglas, and Lyon cos. northward to Hum- boldt Co 61. £. collinum 500 Great Basin Natur.\list Vol. 45, No. 3 — Upper involucres sessile, erect, the peduncles at the lower nodes 1-5 mm long, the involucral tube 1..5-2 mm long; flowers white to rose, 1.2-1.8 mm long; achenes 1.6-2 mm long; rare, Humboldt Co 62. E. salicornioides 66(42). Involucres glabrous to hispid or villous without 67 — Involucres glandular-puberulent without; tepals dimorphic, the outer tepals inflated at the base and middle, white to rose, ofl:en with a large purplish spot on the outer tepals; common throughout most of the state 65. £. macidatum 67(66). Tepals distinctly dimorphic, the outer whorl oblong-ovate and bisaccate, yellow; infrequent, Lincoln Co 63. E. pharnacoides — Tepals essentially monomorphic, not saccate, white; northwestern Nevada. . . . 64. E. spergulinum 68(41). Leaves tomentose on one or both surfaces; stems and branches glabrous to tomentose 69 — Leaves puberulent to villous or sericeous; stems puberulent to villous, with spreading hairs 74 69(68). Involucres 2-4 mm long; stems and lower branches think' floccose or rarely glabrous; flowers 1..5-2 mm long, white to pink; local and often common from Carson City Co. north and eastward to Washoe, Humboldt, and Elko co. 67. £. vimineum — Involucres 1-2 mm long; stems and branches glabrous to densely tomentose, or, if sparsely tomentose, then the flowers yellow to yellowish red 70 70(69). Stems glabrous, or, if tomentose, then the tepals glandular and the outer whorl not fan shaped 71 — Stems tomentose to floccose; tepals fan shaped, white or yellow 73 71(70). Involucres at the tips of slender branchlets and at the node of dichotomous branchlets or branches, not appressed to the stems, the tube turbinate-campan- ulate, smooth, 1.5—2 mm long; flowers 1—1.5 mm long with large, roundish, greenish, or, more commonly, reddish bases and white tepal tips; rare and local in Mineral Co 66. E. ampullaceum — Involucres scattered along and appressed to the stems, the tube narrowly turbinate and slightly angled, 1-1.5 (2) mm long; flowers 0.6-1.5 mm long, yellow, or, if white, then glandular, the base slender and never roundish; common 72 72(71). Flowers white, 1.5-2 mm long; stems and branches glabrous or tomentose; common throughout the Great Basin portion of western Nevada 68. E. badeiji — Flowers yellow, 0.6-1 mm long; stems and branches glabrous; infrequent to locally common in western half of the state south to Nye Co. . . 69. E. hrachyanthum 73(70). Flowers yellow to red; plants rather densely branched; involucres 1 mm long; widely scattered throughout much of the state 70. E. nidulahum — Flowers white or rarely pale yellow; plants open with few branches; involucres 1.5-2 mm long; widely scattered throughout most of the southern two-thirds of the state 71. E. pcdmeriaiiitm 74(69). Outer tepals oblong to narrowly ovate, not hooded, white to red, 1-1.5 mm long, glabrous to hispidulous; involucres 4-toothcd, 1-1.5 mm long; local but often common on volcanic soils in east central and southern Nevada ... 72. E. pidiendum — Outer tepals fan shaped and hooded, pale yellow to pink, 1.5-2 mm long, hirtellous; involucres 5-toothed, 2-2.5 mm long; rare and local on clayey soils in Nve and White Pine cos 73. £. darrovii [ulv 1985 REVEAL: Nevada Erioconum 501 1. Eriogonutn microthecum Nutt. A highly variable shrubby species common throughout the state mainly on gravelly, clayey, or sandy soils, occasionally on rock outcrops and ledges, 1160-3200 m, mostly in sagebrush communities. Flowering from June through October. Widespread in the western United 2.5 States. Great Basin buckwheat. Low subshrubs and shrubs with white or yellow flowers; rep- resented in Nevada by the following varieties: 1. Flowers white, not yellow or yellowish 2 — Flowers yellow; shrubs and subshrubs to 5 dm high; western Nevada from Esmeralda Co. northward to Washoe and Humboldt cos var. ambiguum 2(1). Tomentum whitish; plants shrubs or well-formed subshrubs; common 3 — Tomentum brownish to reddish; plants low, compact subshrubs less than 1.5 dm high; southern Great Basin from Esmeralda and Nye cos. eastward to southern White Pine Co. and Lincoln Co var. lapidicola 3(2). Leaves plane; stems and inflorescences floccose to glabrous; shrubs or subshrubs mostly 2-4 dm high; mainly in the Great Basin portion of the state . . . var. laxiflorum — Leaves revolute; stems and inflorescences densely lanate to tomentose; shrubs mostly 4-15 dm high; mainly in the Mojave Desert portion of the state var. simpsonii The most common expression in the state is var. laxiflorum Hook. [E. confertiflonwi Benth. in DC; E. microthecum vox. conferti- floriim (Benth. in DC.) Torr. & Gray; E. tenelhim Torr. var. sessiliflorum Gand.; E. microthecum subsp. laxiflorum (Hook.) S. Stokes; E. tnicrothecum subsp. confertiflo- rum (Benth. in DC.) S. Stokes; E. microthe- cum var. spathulare S. Stokes] occurs mainly in the Great Basin section of Nevada. It is found mostly above 1500 m on slopes and ridges. The var. simpsonii (Benth. in DC.) Reveal [E. simpsonii Benth. in DC; E. ef- fusum var. foliosum Torr. & Gray. E. mi- crothecum var. rigidum Eastw. ; £. friscanum M. E. Jones; E. nelsonii L. O. Williams; E. effusum subsp. simpsonii (Benth. in DC.) S. Stokes; E. effusum subsp. nelsonii (L. O. Williams) S. Stokes; E. microthecum subsp. rigidum (Eastw.) S. Stokes; E. microthecum subsp. intermedium S. Stokes; E. microthe- cum var. friscanum (M. E. Jones) S. Stokes; E. microthecum var. foliosum (Torr. & Gray) Reveal] is the common expression in the Mo- jave Desert region of the state but extends northward into the Great Basin portions of Esmeralda, Lander, Eureka, northern Nye, and White Pine cos. It occurs mostly below 2150 m on flats and slopes. The name var. simpsonii replaces var. foliosum used previ- ously for this entity (Reveal 1971, 1983). The var. lapidicola Reveal is a dwarfed polygamo- dioecious subshrub mainly of rocky outcrops and ledges. It is found in Esmeralda, Nye, White Pine, and Lincoln cos. A Jaeger collec- tion (POM) from Potosi Mtn., Clark Co., probably is representative of this variety. A closely related form from Eureka Co. may represent an undescribed expression (see Ripley ir Barnebij 9330 and 9333— CAS). The yellow-flowered expression in Nevada is now restricted to the var. ambiguum (M. E. Jones) Reveal in Munz [£. tennelum var. erianthuin Gand.; E. microthecum var. expansum S. Stokes] that occurs in western Nevada from Esmeralda Co. northward into Washoe and Humboldt cos. It is a plant of slopes and ridges in the mountain in the southern part of its range, but of flats and foothills in the northern part of the state. The reference to var. mi- crothecum (Reveal 1971) in Humboldt Co. is now considered an error, and these plants should be referred to var. ambiguum. It is not unusual to find var. laxiflorum and var. am- biguum growing together. 2. Eriogonum nummulare M. E. Jones [E. kearneyi Tidestrom; E. nodosum var. kear- neiji (Tidestrom) S. Stokes; E. dudleyanum S. Stokes; E. nodosum subsp. monoense S. Stokes; E. kearneyi var. monoense (S. Stokes) Reveal; E. kearneyi subsp. monoense (S. Stokes) Munz ex Reveal]. Kearney's buck- 502 Great Basin Naturalist Vol. 45, No. 3 wheat. A large shrub of sandy places, 1130-1850 m elevation, in saltbush and sage- brush communities, eastern California east- ward across the Great Basin portion of Nevada to western Utah, then southward into north- western Arizona. In Nevada the plant occurs in Esmeralda, Mineral, and Ormsby cos., then eastward to Humdoldt, Nye, and Lin- coln cos. As now defined, E. kearneiji and its var. monoense are reduced to synonymy un- der E. nummulare. Plants similar to the var. monoense are found in scattered locations in west central Nevada. 3. Eriogonum corymbosum Benth. in DC. Corymb-flowered buckwheat. A widespread and variable shrub from Nevada and Arizona to Colorado and New Mexico represented in Nevada by var. aureum (M. E. Jones) Reveal [E. aiireiiniM. E. Jones; E. aureumvar. gluti- nosum M. E. Jones; E. fruticosum A. Nels.; E. crispum L. O. Williams; E. microthecum subsp. aureum (M. E. Jones) S. Stokes; E. microthecutn war . crispum (L. O. Williams) S. Stokes; E. corymbosum var. ghitinosum (M. E. Jones) Reveal] that is a large yellow-flow- ered shrub of sandy places of southern Utah and northern Arizona, with a disjunct popula- tion near Las Vegas, Clark Co., Nevada. It is in flower from July to mid-October. The name var. aureum must now be used over the more familiar var. ghitinosum because of a recent change in the International Code (Reveal 1983). A specimen of E. jonesii S. Wats, gathered by Wheeler (US) is labeled "Nevada" and "1872," no doubt an error be- cause this plant is restricted to northern Ari- zona, where it is infrequent. 4. Eriogonum heermannii Dur. & Hilg. A variable shrub found nearly throughout Ne- vada except for the extreme northwestern part, on sandy, clayey, or rocky soils often of a limestone origin, 940-2200 m. Flowering from April to October. A desert shrub from southern California eastward through Nevada to Utah and northern Arizona. Heermann's buckwheat. A large shrub to densely branched subshrub with green, glabrous or sometimes floccose, scabrellous or smooth branches that can be smooth, angled, or ribbed; represented in Nevada by the fol- lowing varieties: 1. Stems smooth, not scabrous or angled, glabrous or floccose 1 — Stems scabrous or angled, not smooth or glabrous 4 2(1). Involucres at the tips of the branches not racemosly arranged or only the last two or three so disposed, the inflorescence diffuse, glabrous; northern two-thirds of the state from Nye and Lincoln cos. northward to Humboldt and Elko cos. var. humilius — Involucres at the tips of the branches racemosely arranged, the inflorescence open, glabrous or floccose 3 3(2). Branches glabrous; common in Clark Co. and adjacent southwestern Nye Co. var. clokeyi — Branches floccose; rare, McCuUough Mts., Clark Co \ar . floccosum 4(1). Stems scabrellous but not sharply and deeply angled; mainly in the Great Basin portion of Nevada from Esmeralda and Nye cos. eastward to Eureka, White Pine, and Lincoln cos var. argense — Stems scabrellous to scabrous, sharply and deepK' angled; mainly in the Mojave Desert portion of Nevada from southern Nye Co. and Clark Co. eastward to southern Lincoln Co \'ar. sulcatum The most common expression in Nevada is var. humilius (S. Stokes) Reveal [E. heerman- nii subsp. humilius S. Stokes] that occurs in the Great Basin portion of the state from Nye and Esmeralda cos. north and east to Washoe, Humboldt, and southern Elko cos. It occurs on a variet)' of soils, but mainK those of \'ol- canic origins. In the Mojaxt' Desert portion of the state is var. clokeyi Reveal. It is found in southern Nye Co. and (>lark Co. and occurs mainly on limestone foothills and slopes. In the McCuIlough Mts. of southern Clark Co. is luly 1985 REVEAL: Nevada Erioconum 503 var. floccostim Munz [E. hccnnannii siibsp. floccosiim (Munz) Munz]. All of these vari- eties are well-defined shrubs and usually oc- cur on gravelly soils. The var. argense (M. E. Jones) Munz [E. howeUii S. Stokes; E. heer- monnii subsp. argense (M. E. Jones) Munz] occurs mainly in the Great Basin region, oc- curring from Esmeralda and Nye cos. east- ward to Eureka, White Pine, and Lincoln cos. The var, sulcatum (S. Wats.) Munz & Reveal [E. sulcatum S. Wats.; E. heennannii subsp. sulcatum (S. Wats.) S. Stokes] is found mainly in the Mojave Desert region of southern Nye and Lincoln cos. and in Clark Co. The var. argense is a small shrub or subshrub, and var. sulcatum is a densely branched subshrub. Both occur mainly on limestone cliffs and rocky outcrops. 5. Eriogonum plumatella Dur.& Hilg. [£. palmeri S. Wats.; E. nodosum var. jaegeri Munz & Johnst.; E. plumatella var. jaegeri (Munz & Johnst.) Stokes ex Munz]. Flat- topped buckwheat. A small shrub with glabrous or tomentose branches. Infrequent in extreme southern Clark Co. , 1000-1220 m. Flowering from June to November. The spe- cies ranges from southern California eastward to extreme southwestern Utah and western Nevada. 6. Eriogonum fasciculatum Benth. Califor- nia buckwheat. Common in the arid South- west and northwestern Mexico with only var. polifolium (Benth. in DC.) Torr. & Gray [E. polifolium Benth. in DC; E. revolutum Goodding; E. fasciculatum subsp. polifolium (Benth. in DC.) S. Stokes; £. fasciculatum var. revolutum (Goodding) S. Stokes] found in Nevada. A low shrub mainly of the Mojave Desert of California and northern Baja Cali- fornia, Mexico, eastward across southern Ne- vada to southwestern Utah and western Ari- zona. Flowering throughout the year. The variety occurs from southern Esmeralda Co. eastward across southern Nye Co. to southern Lincoln Co. on the edge of the Great Basin southward to Clark Co., 520-1800 m elev. Flowering from late March to September. 7. Eriogonum desertorum (Maguire) R. J. Davis [£. chnjsocephalum suhsp. desertorum Maguire; E. brevicaide Nutt. var. desertorum (Maguire) Welsh] Cold desert buck-wheat. A compact cespitose perennial with tomentose scapes and yellow flowers of desert ranges and flats in northeastern Nevada and adjacent northwestern Utah, 1700-2330 m elev. Flow- ering from late May to mid-July. In Nevada the species is found in Eureka, White Pine, and Elko cos. where it often occurs on clay slopes and flats. 8. Eriogonum ochrocephalum S. Wats. Ocher-flowered buck-wheat. A compact cespi- tose perennial with glabrous, glandular, or tomentose scapes and yellow flowers of north- western Nevada and adjacent northeastern California northward into southeastern Ore- gon and southern Idaho. Flowering from May to late June at lower elevations and to early September at higher elevations. The var. ochrocephalum [E. nevadense Gand.] is lo- cally common on clayey outcrops and flats of Lyon and Storey cos. northward through Washoe Co. to the Oregon line. This variety occurs 1310-2470 m elevation, reaching its highest elevations on Peavine Mtn. The scapes of this variety are mainly glabrous, but occasional specimens in Washoe Co. have glandular scapes. The newly proposed var. alexanderae Reveal, characterized by its floc- cose scapes, occurs mainly east of var. ochro- cephalum, being found from Mineral Co. northward to Washoe and Pershing cos. It occurs on clay outcrops 1430-2070 m eleva- tion and flowers from late May to early July. 9. Eriogonum prociduum Reveal Austin's buckwheat. A compact cespitose perennial with glabrous scapes and yellow flowers known only from northern Washoe Co., Ne- vada, and adjacent Lassen Co., California, northward to Lake Co. , Oregon, 1400-2450 m elevation. Flowering from May to early July. The Nevada site is vouchered by Tiehm 8056 (MARY). It is mentioned in the latest install- ment of the rare and endangered species re- port for Nevada (Pinzl 1983). 10. Eriogonum lewisii Reveal Lewis' buck- wheat. A compact cespitose perennial with floccose scapes and yellow flowers known only from northeastern Nevada and adjacent northwestern Utah. In Nevada the species is known from the high mountains of Elko Co. where it occurs above 2400 m elevation. It flowers from late June to early September. IL Eriogonum crosbyae Reveal Crosby's buckwheat. A compact cespitose perennial with floccose scapes and yellow flowers known only from northwestern Nevada and adjacent 504 Great Basin Naturalist Vol. 45, No. 3 southeastern Oregon. In Nevada the species occurs on clay outcrops in northern Washoe and extreme southern Humboldt cos., where it occurs 1600-1700 m elevation. It flowers from late May through July. This species, pre- viously known only from Oregon (Reveal 1981), was first discovered by Tiehm ir Bird- sey 5013 (MARY) in Humboldt Co. and more recently in Washoe Co. by Tiehm alone (8040, 8043— MARY). 12. Eriogonum rosense Nelson & Kennedy [E. ochrocephalwn var. agnellum Jeps.; E. ochrocephalurn subsp. agnellwn (Jeps.) S. Stokes]. Mt. Rose buckwheat. A compact ce- spitose perennial with glandular scapes and yellow flowers common in the Sierra Nevada, and Sweetwater and White mountains of Cali- fornia northward into Nevada from Esmeralda Co. northward to Washoe Co., where it oc- curs (2000) 2440-3300 m elevation. It flowers from July to early September. 13. Eriogonum beatleyae Reveal Beatley s buckwheat. A compact cespitose perennial with glandular and eglandular villous scapes and yellow to cream-colored flowers of ex- treme east central California and Nevada. Lo- cal and scattered on clayey soils in west cen- tral and central Nevada from northern Nye Co. northward to Eureka and Lander cos., then westward to Churchill and Mineral cos. 1700-2320 (2750) m elevation. It flowers from May to August. Since this species was de- scribed (Reveal 1972), E. beatleyae has been found in a variety of sites in Nevada. For the most part the species is found at elevations lower than that of E. rosense, but a recent collection {Ertter ir Strachan 2804— MARY) from the north end of the Monitor Range at 2750 m elevation is probably best referred to E. beatleyae rather than E. rosense as I origi- nally annotated the collection. Cream -colored specimens of E. beatleyae occur in Nye, Churchill, Lander, and Eureka cos. 14. Eriogonum anemophilum Greene Wind-loving buck-wheat. A compact cespitose perennial with floccose scapes and white flow- ers endemic to Nevada; restricted to the West Humboldt Mts. of Pershing Co. and the Jack- son Mountains and Sonoma Range of Hum- boldt Co. 2500-2800 m elevation. Flowering from June to August. 15. Eriogonum tiehmii Reveal Tichm's buckwheat. A compact cespitose pereimial with floccose scapes and cream-white glandu- lar flowers endemic to Nevada; restricted to white clay hills near Cave Springs, Esmeralda Co., 1830 m elevation. Flowering from early May to late July. This new species may be quickly recognized by its glandular tepals and its longer, acutely toothed involucres. The floccose scapes distinguishes Tiehm's buck- wheat from the cream-colored specimens of E. beatleyae. 16. Eriogonum kingii Torr. & Gray King's buckwheat. A compact cespitose perennial with floccose to glabrous scapes and greenish yellow to pale yellow flowers endemic to the Ruby Mts. and East Humboldt Range of Elko Co., and in the Cherry Creek Range of north- ern White Pine Co., Nevada, 2400-3170 m elevation. It flowers from June to August. 17. Eriogonum argophyllum Reveal Sul- phur Hot Springs buckwheat. A compact perennial with floccose scapes and yellow flowers endemic to mineralized soil at Sul- phur Hot Springs, Elko Co., Nevada. It oc- curs at 1850 m elevation and flowers from June to July. 18. Eriogonum gracilipes S. Wats. [E. kenneclyi subsp. gracilipes (S. Wats.) S. Stokes; E. ochrocephahim var. gracilipes (S. Wats.) J. T. Howell]. White Mountain buck- wheat. A compact perennial with glandular- hairy scapes and white to reddish flowers of the White Mountains and adjacent portions of the Sierra Nevada of east central California and adjacent Esmeralda Co., Nevada. It is found 3000-4000 m elevation, but in Nevada it is at 3200 m. The plant flowers from July to September. 19. Eriogonum hohngrenii Reveal Holm- gren's buckwheat. A compact perennial with floccose and stipitate-glandular scapes and white to reddish flowers endemic to the Snake Range of White Pine Co., Nevada, where it occurs 2870-3700 m elevation. It flowers from July to September. 20. Eriogonum villiflorum A. Gray Shaggy-haired buck-wheat. A compact peren- nial of 10-20 rosettes, prostrate \ illous flower- ing stems, and densely pilose white flowers with glabrous achenes. Local and infre(}uent in western Utah and eastern Nevada on grav- elly flats and slopes 1900-2200 m elevation. In Nevada the species is known onK horn White July 1985 REVEAL: Nevada Erioconum 505 Pine, Lincoln, Nye, and Eureka cos. It flow- ers from May to early June. 21. Eriogonum shockleyi S. Wats. A vari- able species rather common throughout the Intermountain Region and along its immedi- ate borders in the western United States, mostly on gravelly, clayey or sandy soils, or on rocky outcrops and ledges, 730-2750 m eleva- tion. Flowering from May through August. Shockley's buck-wheat. Cespitose to pul- vinate perennials with up to a hundred or more rosettes with erect, floccose to tomen- tose stems, and white, reddish or yellow densely pilose flowers and pubescent ach- enes. The var. shockleyi [E. viUiflorum var. condidum M. E. Jones; E. acaule var. shock- leyi (S. Wats.) M. E. Jones; E. pidvinatum Small; E. shockleyi subsp. candidum (M. E. Jones) S. Stokes] is common in Nevada and ranges from Mineral and Esmeralda cos. east- ward across Nye Co. to Lincoln Co., and northeastwardly to Elko Co. This plant occurs mainly on gravelly to clayey soils throughout this range 1500-2100 m elevation. Both white- and yellow-flowered specimens belong to this variety. A population at Baking Powder Flat in Spring Valley, White Pine Co., is on deep moving sand, and the resulting plants are large, open, loose mats {Emmel 199 — CAS; Reveal 4845— MARY, US and else- where). This may prove to represent a new and as yet undescribed variety. A M. E. Jones collection (POM) gathered at Pioche, Lincoln Co., 31 August 1912, is similar to E. soredium Reveal which is now known only from Beaver Co., Utah. Attempts to rediscover specimens similar to the Jones collection have failed, and the extant specimen is not suitable to ade- (juately determine its identity. 22. Eriogonum wrightii Torr. ex Benth. in DC. A highly variable cespitose to sub- shrubby or shrubby perennial of western and southern Nevada on sandy to gravelly soil, 1280-2450 m. Flowering from late June to early November. Widespread in North Amer- ica from northern California and western Ne- vada southward to Baja California, Mexico, eastward through Arizona and New Mexico to western Texas southward to central Mexico. Wright's buck-wheat. Low shrubs or sub- shrubs with small elliptic leaves and white flowers. The var. wrightii is locally common in the McCullough Mts. and infrequent to rare in the Spring Mts., Clark Co. It is a distinct shrub and occurs from extreme south- eastern California eastward across southern Nevada and southwestern Utah to Texas southward to central Mexico. The var. sub- scaposum S. Wats. [E. wrightii subsp. sub- scaposiivi (S. Wats.) S. Stokes] is a low sub- shrub of the mountains of western Nevada from the Sweetwater Mts. of Mineral Co. (and to be expected in the Nevada portion of the White Mts.) northward to southern Washoe Co. Other varieties are found in California and adjacent western Mexico. 23. Eriogonum panamintense Morton An erect herbaceous perennial with large leaves and white flowers on loam to gravelly soil in the desert ranges of southwestern Nevada and adjacent southeastern California, 1830-2600 m. Flowering from mid-June through September. Panamint buck-wheat. The var. pana- mintense [E. reliqiiiim S. Stokes; E. racemo- sum var. desertorum S. Stokes] with numer- ous solitary involucres and large elliptic to ovate or obovate leaves is rather common in the mountains of southwestern Nevada from Mineral and extreme western Nye Co. south- ward to the Spring Mts. and Sheep Range of Clark Co. This variety is similar to E. nipinum. The var. mensicola (S. Stokes) Re- veal in Munz [£. panamintense subsp. mensi- cola (S. Stokes) Munz] with few solitary in- volucres and generally rotund leaves is restricted to the Sheep Range of Clark Co., Nevada, and the Death Valley region of Inyo Co., California. 24. Eriogonum rupinum Reveal Canyon buck-wheat. A stout erect perennial herb with large, oblong to elliptic leaves and 3-5 solitary involucres racemosely arranged at the tips of the branches, on gravelly to sandy (rarely clayey) soil on the foothills and canyon bot- toms 1830-2600 (3489) m, in the mountains of central Nevada to eastern California. Flower- ing mainly from July to early October. In Ne- vada, the species occurs mostly below 3100 m from Lander and Nye cos. westward to Min- eral and Esmeralda cos. It is a stouter plant than E. panamintense. 25. Eriogonum racemosum Nutt. [E. or- thocladon Torr. in Sitgr. ; E. obtusum Benth. in DC.]. Redroot buck-wheat. An erect, grace- ful perennial herb with large, variably shaped 506 Great Basin Naturalist Vol. 45, No. 3 leaves on sandy to gravelly soil from north- western New Mexico and southwestern Colo- rado westward across northern Arizona and most of Utah to central Nevada. Mostly 1220-2500 m, and flowering from June to Oc- tober. In Nevada the species is rather com- mon in the desert ranges of Nye, White Pine, and Lincoln cos., but infrequent and local in the Sheep Range, Clark Co. 26. Eriogonum nudum Dougl. ex Benth. A highly variable perennial herb with numerous varieties. Widespread and common in the Pacific coast states from Washington to Baja California Norte eastward to western Nevada from sea level to 3400 m. Flowering from April through September. Naked-stemmed buck-wheat. Erect peren- nial herbs with small, variably shaped leaves and glabrous to tomentose stems; currently represented in Nevada by four poorly defined varieties. The montane expression is the var. deductiim (Greene) Jeps. [£. dediictum Greene] characterized by glabrous stems and solitary involucres. It occurs from Mineral Co. northward in the Sierra Nevada and adja- cent ranges to southern Washoe Co. The var. nudum [E. latifolium subsp. nudum (Dougl. ex Benth.) S. Stokes] is rare in Nevada. It is known only from two old Nevada collections (Kennedy 962 and Peterson 279— NESH) gathered near Verdi, Washoe Co. This variety occurs from Washington and Oregon south- ward into northern California. It is character- ized by clustered involucres and glabrous stems. The distinction between the two vari- ants in low elevation populations is generally blurred in the Sierra Nevada, and the diffi- culties are repeated in Nevada. The var. oblongifolium S. Wats. [E. harfordii Small; E. sulphureum Greene; E. capitatum A. A. Heller; E. latifolium subsp. sulphureum (Greene) S. Stokes] occurs from Douglas Co. northward to southern Washoe Co. This vari- ant ranges from southern Oregon to northern California and adjacent Nevada. It is charac- terized by a tomentose stem and pubescent flowers. The poorly differentiated var. pubi- florum Benth. in DC. occurs in northern Washoe Co. and southwestern Humboldt Co. It differs from var. oblongifolium in having glabrous stems along with its pubescent flow- ers. At present only the yellow-flowered ex- pression of var. pubiflorum has been discov ered in Nevada. The var. gramineum (S. Stokes) Reveal is to be sought in the warm desert mountain ranges of extreme south- western Nye Co. that border Death Valley. It has inflated, glabrous stems and yellow, pubescent flowers, and it is found on lime- stone outcrops. All occur 1500-2500 m eleva- tion in Nevada and flower from June through September. 27. Eriogonum elatum Dougl. ex Benth. A tall, erect, perennial herb with large lance- olate to lance-ovate leaves and slightly pubescent flowers of Washington and Idaho south to central California and central Nevada 1200-2900 m. Flowering mainly from May through August. Tall buckwheat. The var. elatum [E. elatum var. erianthum Gaud.] occurs in the moun- tains and foothills of Nevada from Nye Co. northward. It is often locally common, al- though individual populations can be rather scattered. The var. villosum Jeps., character- ized by its pubescent stems, is restricted in Nevada to the foothills of the Sierra Nevada from Douglas Co. northward to southern Washoe Co. It is rather infrequent. 28. Eriogonum ovalifolium Nutt. A wide- spread and highh \ariable, compact peren- nial herb of many ecological niches found in the Nevada and throughout the western United States and southwestern Canada, 920-3700 m. Flowering from April to August. Cushion buckwheat. A compact to cespi- tose herb with pubescent scapes and white to yellow or reddish, strongly dimorphic tepals (except in some high elevation expressions) in capitate terminal inflorescences mostly in the sagebrush communities throughout the state; consisting of the following \arieties: 1. Leaves mostly more than 1 cm long; scapes 5-30 (or more) cm long; involucres (3.5) 4-7 mm long, turbinate; flowers (3) 4-1 mm long; plants mostly below 2450 m ' 2 — Leaves mostly less than I cm long; scapes up to 6 (10) cm long; involucres 2.5—4.5 mm long, turbinate-campanulate; flowers 2.5-4 mm long; plants mostly above 2450 m except in the Mt. Rose area of west central Ne\ada 3 July 1985 REVEAL: Nevada Erioconum 507 2(1). Flowers white var. ovalifolium — Flowers yellow vur. nevadense 3(1). Flowers white 4 — Flowers yellow; endemic to the To(]iiima and Toiyabe ranges of Nye Co. var. caelestinwn 4(3). Leaves without brown-edged margins; plants mainly of high elevation sites 5 — Leaves with distinct brown-edged margins; plants mainly of low elevation sites .... 6 5(3). Leaves densely white-tomentose on both surfaces; widespread and common var. nivale — Leaves greenish white pubescent at least on the upper surface; infrequent, Elko Co var. de})ressum 6(4). Leaves densely lanate tomentose on both surfaces, the tomentum brownish, the brown edge pronounced; plants of the Mt. Rose area in Carson City and southern Washoe cos var. eximium — Leaves white-tomentose on both surfaces, the tomentum whitish, the brown edge thin; plants endemic to the Steamboat Springs area of southern Washoe Co. var. williamsiae The var. ovalifolium [E. purpureitm (Nutt.) Renth. in DC; E. davisiamim S. Stokes] is the low elevation expression with white flow- ers, and var. nevadense Gand. [E. or- thocaidon Small; E. ovalifolium var. celsum A. Nels.; E. ovalifolium var. orthocaulon (Small) C. L. Hitchc. ] has yellow flowers. The var. ovalifolium is widespread throughout the Intermountain Region section of Nevada, and var. nevadense occupies essentially the same area although it is more common in the north than in the south. Two other low elevation variants are var. eximium (Tidestrom) J. T. Howell [E. eximium Tidestrom; E. ovali- folium subsp. eximium (Tidestrom) S. Stokes] and var. williamsae Reveal. Roth are found in the Mt. Rose area of southern Washoe Co. and adjacent Carson City Co., with var. williamsae restricted to the Steamboat Springs area, Washoe Co. The former has larger leaves with distinctly brown-edged leaf-blades, and the latter has small leaves arranged in densely compact mats. Three high-elevation variants occur in Nevada. The most common phase is var. nivale (Canby in Gov.) M. E. Jones [E. nivale Canby in Gov.; E. rhodanthum Nels. & Kenn.] which has densely white-tomentose leaves and white flowers. It is rather common throughout the state. The var. depressum Blank., with its greenish white tomentose leaves and white flowers, is infrequent in Elko Co. The high elevation yellow-flowered variety is the var. caelestinum Reveal, that is restricted to the Toquima and Toivabe ranges of northern Nye Co. 29. Eriogonum striatum Benth. A branched, erect perennial herb with white or yellow dimorphic tepals ranging from north- ern Washington south to northern California, and eastward to western Montana south to northern and northwestern Nevada mainly below 2600 m. Flowering from May through August. Rlue Mountain buckwheat. Erect peren- nial herbs with branched inflorescences bear- ing white or yellow dimorphic tepals; repre- sented in Nevada by two varieties of subsp. proliferum (Torr. & Gray) S. Stokes differing chiefly on flower color. The var. anserinum (Greene) R. J. Davis [E. anserinum Greene; E. strictum subsp. anserinum (Greene) S. Stokes; E. ovalifolium subsp. flavissimum S. Stokes; E. strictum var. flavissimum (S. Stokes) C. L. Hitchc; £. proliferum subsp. anserinum (Greene) Munz] has yellow flowers and is the common expression. It is found from western Elko Co. across Humboldt Co. to Storey and Washoe cos. The var. pro- liferum (Torr. & Gray) Reveal [E. proliferum Torr. & Gray; E. cusickii Gand. var. califor- nicum Gand.] is white flowered and occurs in Elko Co., although a Kennedy collection (i38i —NESH, UTC) presently referred to this variety has been found at Broncho Greek in Washoe Co. This species differs from E. 508 Great Basin Naturalist Vol. 45, No. 3 ovalifolium in having branched rather than m. Flowering from May to late July. In Ne- capitate inflorescences. vada the species is rather local in Nye and 30. Eriogoniim saxatile S. Wats. [E. southern Esmeralda cos. bloomeri Parish; E. stokesae M. E. Jones; E. 31. Eriogonuyn umbellatum Torr. A large saxatileyar. stokesae (M. E.Jones) S.Stokes ^^^^ exceedingly complex species found ex Jones; E. saxatile subsp. multicaiile S. throughout most of western North America f,^ , T „ , , , , ^ . 1 , , , from southern Canada to near the Mexican StokesJ. Rock buclcwheat. A low, looselv to , , . -, . . n i •. . i <■ , , , .111 . I 1 ' 11 border on a wide variety ot habitat and trom densely matted perennial herb with broadlv i i , i ' .^aa x?^ , ^ ^, , ' near sea level to nearly 4000 m. Flowering obovate to rotund leaves and white to cream ^^^^ ^^^ ^^ October. The species is corn- flowers on elongated, winged stipes of sandy, po^gj ^f ,-,-,01-e than 30 varieties, several of decomposing soil in desert ranges of southern which have as yet to be described formaUy . In California and adjacent Nevada, 1220-3400 Nevada the following variants are known: 1. Primary branches of the inflorescences simple, not branched 1 — Primary branches of the inflorescences branched 9 2(1). Flowers bright yellow 3 — Flowers mostly whitish to red, occasionally pale yellow; plants subshrubs (see also the shrubby var. venium ) 7 3(2). Leaves pubescent at least below on plants in full anthesis 4 — Leaves glabrous on both surfaces on plants in full anthesis 6 4(3). Leaves densely tomentose on the lower surface even in fruiting plants; plants low, spreading subshrubs; Elko Co var. umbellatum — Leaves sparsely pubescent on both surfaces or only glabrate on the upper surface on plants in full anthesis; plants upright to spreading subshrubs or shrubs 5 5(4). Flowers less than 7 mm long, always sulphur yellow; involucral tubes 2-3.5 mm long; plants late spring to summer flowering subshrubs or small shrubs, common throughout nearly all of Nevada except for Elko, Clark, and perhaps White Pine cos var. nevadense — Flowers 6-10 mm long, sulphur yellow or infretiuently cream colored; involucral tubes 1.5-2.5 mm long; plants spring flowering shrubs endemic to Nye Co. var. venium 6(3). Inflorescences umbellate or merely subcapitate; widespread at lower elexations from eastern Humboldt Co. south to Nye Co., then eastward to Elko and White Pine cos var. aureum — Inflorescences capitate or nearly so; rare in subalpine and alpine habitats in the Ruby and East Humboldt mts var. poi-teri 7(2). Leaves sparsely tomentose to glabrous on both suriaces or with the tomentum slightly more below than above; flowers ;3-8 mm long, whitish, cream-colored, or pale yellow to reddish brown to rose or pink with large, colored midribs 8 — Leaves densely tomentose below, bright green to olive-green and floccosc to glabrous above; flowers cream colored, .3—7 mm long; to be expected in Elko Co. but currently unknown for Nevada var. majus 8(7). Flowers cream colored to pale yellow with tannish midribs; mostlx" northern Nevada from Nye Co. northward \ ar. dichrocephalum — Flowers reddish brown to pink with reddish to purplish midribs; Spring Mts., Clark Co., and northward to extreme southern Nye Co. and adjacent south- western Lincoln Co var. versicolor 9(1). Flowers yellow or strongK yellowish 10 July 1985 REVEAL: Nevada Erioconum 509 — Flowers cream colored or reddish brown to rose or pink, not bright yellow 11 10(9). Leaves densely tomentose below, thinly floccose or more commonly glabrous and green above; flowers bright yellow, (5) 6-8 mm long; Sierra Nevada from Lake Tahoe area northward to southern Washoe Co var. fitrcosum — Leaves evenly thinly pubescent to glabrous on both surfaces; flowers bright yellow, 6-7 mm long; desert ranges from Esmeralda Co. south to Clark Co. eastward to Lincoln and White Pine cos var. subaridum 11(9). Flowers cream colored or whitish; plants forming low shrubs or subshrubs up to 8 dm tall; Lincoln and White Pine cos var. juniporiniim — Flowers reddish brown to rose or pink with large reddish or purplish midribs; plants forming low, spreading, matted subshrubs mostly less than 2 dm tall; Spring Mts. , Clark Co var. versicolor Sulphur-flower. The Rocky Mountain ex- pression, var. umbellatum, occurs in Nevada only in the high mountains of Elko Co., it being found typically in the Ruby and East Humboldt mts., but only infrequently in the mountains to the north. More widespread in the state is the glabrous-leaved expression var. aureum (Gand.) Reveal [£. neglectum Greene; E. azaleastrum Greene] that differs from the typical variant only in this single feature; it is generally found at a lower eleva- tion than var. umbellatum. The var. aureum occurs from eastern Humboldt Co. south to northern Nye Co. eastward. The high eleva- tion expression related to var. aureum is var. porteri (Small) S. Stokes [£. porteri Small] that occurs in the East Humboldt and Ruby mts. generally above 2750 m elevation. All three variants are more common in the Rocky Mountains to the east and reach their most westward point of distribution in Nevada. The most common phase oiEriogonum um- bellatum in Nevada is the western var. nevadense Gand. [E. reclinatum Greene; E. heracleoides yar. virde Gnnd.;E. umbellatum var. californicum Gand.]. It is found through- out the Great Basin portion of Nevada from Nye Co. northward. It flowers mainly in the summer, forming well-defined subshrubs in most populations and is typically found at 1400-2900 m elevation. The variety is com- mon in the mountains of western Nevada, but it becomes less frequent in the central and eastern portions of the state. At times, in Elko Co. for example, the differences between var. U7nbellatum and var. nevaden.se are obscured. The early spring-flowering expression related to var. nevadense is the northern Mojave Desert-southern Great Basin expression var. vernum Reveal. This plant, typically a large and well-defined shrub, is in full anthesis in May and continues to blossom into June. Bright yellow-flowered plants are common, but on the Nevada Test Site scattered popula- tions are dominated by pale yellow-flowered individuals. At higher elevations, and mainly in the northern half of the state, is the cream- colored expression close to var. nevadense, the var. dichrocephalum Gand. [E. aridum Greene; E. umbellatum subsp. aridum (Greene) S. Stokes; E. umbellatum var. aridum (Greene) C. L. Hitchc.]. Unlike the subshrubby var. nevadense, the var. dichro- cephalum is often more similar in habit to the spreading and mat-forming var. umbellatum. The var. nevadense occurs from the Sierra Nevada of California northward to southern Oregon and eastward across Nevada, and the var. vernum is endemic to Nye Co., Nevada. Both are found mostly below 2440 m eleva- tion. The var. dichrocephalum, on the other hand, ranges from southeastern Oregon southward in the Sierra Nevada to Inyo Co., California, then eastward across Nevada and Utah to extreme western Colorado. This vari- ety can occur above 3100 m but is most com- mon 2200-2800 m elevation. Yet to be discov- ered in Nevada is the widespread, northern expression, var. mo/us Hook. [E. suhalpinum Greene; E. umbellatum subsp. majus Piper; E. umbellatum subsp. subalpinum (Greene) S. Stokes; E. umbellatum var. subalpinum (Greene) R. J. Davis]. This is a mat-forming perennial with pale, cream-colored tepals, and olive-green leaves that are glabrous (or nearly so) on the upper surface. The var. ma- jus occurs from British Columbia, Canada, southward to southern Oregon, then eastward 510 Great Basin Naturalist Vol. 45, No. 3 across Idaho to Alberta and southward in the Rocky Mts. to Colorado. The remaining variants of Eriogonum um- bellatiitn differ from those mentioned above in having branched flowering inflorescences with each node subtended by a whorl of bracts. The majority of these expressions are found in the southern part of the state. The one exception is the var. furcosum Reveal that is restricted to the Sierra Nevada portion of Nevada in the Lake Tahoe region. This expression was previously referred in the lit- erature to the more northern and less shrubby var. ellipticum (Nutt.) Reveal [E. stellatiim Benth.; E. ellipticum Nutt.; E. umhellatiim var. stellatum (Benth.) M. E. Jones; E. umbcl- latwn subsp. stellatum (Benth.) S. Stokes] (see Reveal 1983). The most common expres- sion in southern Nevada is the var. subaridum S. Stokes [E. biiimbcllatum Rydb.; E. ferrissii A. Nels.; E. umbellatum subsp. ferrissii (A. Nels.) S. Stokes; E. umbellatum subsp. sub- aridum (S. Stokes) Munz]. It differs from the var. furcosum in the degree of pubescence on the leaves. The var. subaridum ranges from Esmeralda Co. eastward across the state to Eureka and White Pine cos., then southward to Lincoln and Clark cos. In Clark Co., the var. subaridum is restricted to the Sheep and Spring mts. The variety occurs at 1830-2800 m elevation. A cream-colored expression re- lated to the var. subaridum is the newly pro- posed var. juniporinum Reveal, which occurs in Lincoln and White Pine cos. and reappears again in the desert ranges of San Bernardino Co. in southeastern California. In Nevada var. juniporinum occurs at 1830-2500 m eleva- tion. Also in the southern desert ranges is the rose to pink- or reddish-tepaled var. versi- color S. Stokes, a montane phase that may or may not have a compound inflorescence. It ranges from northern Clark and southern Nye cos., Nevada, westward to eastern Inyo and southern Mono cos., California. It occurs at 1980-2750 m elevation. 32. Eriogonum heracleoides Nutt. Wy- eth's buckwheat. A matted perennial herb with branched flowering inflorescences, cream or whitish yellow flowers, and a whorl of bracts about midlength along the pubescent, erect, flowering stem. The species occurs from southern British Columbia, Canada, southward to northern California, then east to western Montana, Wyoming, and Colorado. Common in the mountains of northern Nevada from northern Washoe Co. eastward to Elko Co. and southward into Lan- der and northern White Pine Co. at 1740-3100 m elevation. Flowering from June through August. A Stokes collection (OKL) from Verdi, Washoe Co., is likeK' mislabeled. 33. Eriogonum sphaerocephalum Dougl. ex Benth. A variable species of the northwest- ern United States from Washington and Idaho southward to northern California and north- ern Nevada. Round-headed buckwheat. A low, woody subshrub with yellow or ochroleucous, vil- lous-tomentose flowers and achenes with a slightly pubescent beak; mostly on volcanic soils 920-2140 m elevation. Flowering from May to mid-July. The yellow-flowered ex- pression is the var. sphaerocephalum. It oc- curs from Washoe Co. southward to the Car- son City area (M. E. Jones s.n. — POM) eastward across Humboldt and Eureka cos. to Elko Co., mostly at 1430-1950 m elevation. The pale-flowered expression, var. hal- imoides S. Stokes, is restricted to northern Washoe Co., northwestern Humboldt Co., and northern Elko Co. It occurs mostly at 1740-2130 m elevation. 34. Eriogonum douglasii Benth. in DC. Douglas' buck-wheat. Low, spreading, mat- ted, and cespitose perennial herbs with yel- low to ochroleucous, densely to sparsely vil- lous-tomentose flowers, a pubescent, 3-angled beak, and a whorl of leaves about midlength along the flowering stems. The species ranges mostly in the sagebrush and woodland communities of central Washington south to east central California and western Nevada; it occurs at 610-2450 m elevation. The Nevada expression is the yellow-flowered var. douglasii [E. caespitosum var. douglasii (Benth. in DC.) M. E. Jones; E. caespitosum subsp. douglasii {Benih. in DC.) S. Stokes]. It is known only from the Peavine area of Washoe Co., where it occurs at 1600-2140 m elevation. Flowering maiuK' from April to late .b'ly- 35. Eriogonum caespitosum Nutt. [E. and- inum Nutt., E,. sericolcucum Greene exTide- strom; E. sphaerocephalum var. sericolcucum (Greene ex Tidestrom) S. Stokes]. Cespitose buckwheat. Low, compact, cespitose peren- July 1985 REVEAL: Nevada Erioc;onum 511 nial herbs with a single involucre atop the scapose flowering stem hearing yellow, densely pilose to villous-puhescent flowers, and glabrous to sparsely pubescent achenes. Widespread and common horn eastern Cali- fornia across Nevada to northern Utah and northwestern Colorado, and north to south- eastern Oregon, southern Idaho, western I Montana, and western Wyoming. In Nevada j the species is common throughout the Inter- mountain Region portion of the state. It oc- curs at 1400-3400 m elevation. Flowering from May through August. 36. Eriogonum marifolium Torr. & Gray Marum-leaved buckwheat. Low, spreading, loosely matted, dioecious perennial herb with green to olive-green leaves and numerous nonrooting caudices, the flowers yellow, glabrous, with those of the male plants smaller than the female plants. Common in sandy or pumice soils in the Sierra Nevada of California northward to central Oregon, and eastward to Humboldt Co., Nevada, mostly 1070-3400 m elevation. Flowering from June to August. In Nevada the species occurs in the Sierra Ne- vada portion of the state and is isolated on the Pine Forest Range of western Humboldt Co., mostly 2000-2930 m elevation. The closely related Gray-leaved buckwheat, E. incanum Torr. & Gray, which is characterized with densely pubescent leaves, is not known to occur in Nevada but is to be expected in the Sierra Nevada portion of the state. 37. Eriogonum latens Jeps. [E. monticola S. Stokes]. Onion-flowered buckwheat. Low, compact perennial herbs with short-pilose leaves and an erect, slender, essentially glabrous flowering scape bearing a single clus- ter of involucres with numerous cream to pale yellow flowers. Local and infrequent in the White Mts. of Esmeralda Co. , Nevada, south- ward to the Inyo Mts. of Inyo Co., California, and along the eastern Sierra Nevada in north- ern Inyo Co., mostly 2000-3400 m elevation. Flowering from late June to late August. 38. Eriogonum lohhi Torr. & Gray A vari- able species of the numerous soil types in the mountains of northern and eastern California, and in western Nevada, that occurs mostly at 1310-3700 m elevation. Flowering from June through August. Lobb's buckwheat. Low, spreading, small to robust, compact to densely matted peren- nial herbs with stoutish caudex branches bear- ing prostrate to decvunbent or weakly erect flowering stems with large (5-15 mm long), solitary involucres bearing white to rose, glabrous flowers. The high elevation phase in Nevada is var. lohbii [E. lohbii var. minus Torr. & Gray] that occurs in the Sierra Ne- vada portion of the state in Washoe Co. The plant may be found at approximately 2150-2870 m elevation on granitic soils. The more common expression in Nevada is the endemic var. robustum {Greene) M. E. Jones [E. robustum Greene]. This phase is much more robust, with larger leaves, involucres, and flowers and erect or nearly so flowering stems. It occurs on altered andersite soils in Storey and Washoe cos. at 1310-2440 m elevation. 39. Eriogonum inflatum Torr. & Frem. Desert trumpet. A highly variable plant of arid regions of western North America from California eastward to Colorado southward to northwestern Mexico mostly on sandy, vol- canic soils below 2100 m elevation. Flowering nearly throughout the year but mostly from March to October. Our expression, var. infla- tum, is an annual or perennial herb with in- flated stems and branches bearing numerous, erect, filiform to capillary peduncles of 5- toothed involucres and yellow, pubescent flowers. The variety is the common expres- sion in Nevada; it may be found nearly throughout the state except the northernmost tier of counties. In southern Arizona south- ward is the var. deflatum I. M. Johnston. In northeastern Utah and adjacent states is the var. fusiforme (Small) Reveal. The first two varieties are first-year flowering perennials, but the latter is strictly an annual. 40. Eriogonum contiguum (Reveal) Reveal [E. inflatum var. contiguum Reveal]. Ash Meadow buckwheat. Erect annual herbs with slender, basally glandular stems and branches bearing numerous erect to spreading capillary peduncles bearing 5-toothed involucres and yellow, pubescent flowers. The species is re- stricted to the Mojave Desert region from Ash Meadows southward to the Pahrump Valley area of southern Nye Co., Nevada, then west- ward to the Death Valley region south to Tecopa, Inyo Co., California. It is found from near sea level to 762 m elevation. Flowering from April to late June. 512 Great Basin Naturalist Vol. 45, No. 3 41. Eriogonum trichopes Torr. Little trumpet flower. Erect annual herbs with slen- der or inflated, glabrous or occasionafly basally glandular stems and branches bearing numerous, erect capillary peduncles of 4- toothed involucres and yellow, pubescent flowers. The species ranges from southern California, Nevada, and Utah southward to northwestern Mexico and eastward to south- ern New Mexico. It occurs from below sea level to 2000 m elevation and flowers through- out the year. In Nevada the var. trichopes [?E. cordatum Torr. & Frem.; E. tri- chopodium Torr. ex Benth.; E. trichopodiiim var. minor Benth. in DC.;£. fnc/jope.s subsp. cordatum (Torr. & Frem.) S. Stokes] occurs mainly in the Mojave Desert portion of the state from southern Nye and Lincoln cos. southward to Clark Co., 300-1770 m eleva- tion. Flowering mainly from late Maich to early July. The Fremont collection of £. cor- datum, described prior to that of £. trichopes, has been lost and the description is not clear enough to determine its actual identity. It is possible this is an early name for E. trichopes or E. contiguum. It is not likely an earher name for E. glandulosum. A\\ three species occur in the eastern Mojave Desert area of California where the type of £. cordatum was gathered. 42. Eriogonum howellianum Reveal How- ell's buckwheat. Low, spreading, pilose-hir- sutulous and glandular annual herbs with ascending peduncles bearing turbinate- campanulate involucres with dense pilose, yellow flowers. Local and rare on dry sandy soil in the desert ranges of the Great Basin from western Utah eastward across Nevada from Elko, White Pine, and Lincoln cos. to Nye and north central Clark cos., mostly 1500-1900 m elevation. Flowering from late Jime through August. 43. Eriogonum glandulosum (Nutt.) Nutt. ex Benth. in DC.[ £. trichopes subsp. glandu- losum (Nutt.) S. Stokes; £. glandulosum var. carneum]. T. Howell; £. carneumi]. T. How- ell) Reveal in Munz]. Gambel's buckwheat. Low, spreading, pilose-hirsutulous and glan- dular annual herbs with deflexed peduncles bearing narrowly turbinate involucres with dense pilose, white to pinkish flowers. Local and rare on sandy soil in the desert ranges of the northern Mojave Desert from southwest- ern Nye and northwestern Clark cos., Ne- vada, westward into eastern Inyo Co. and northeastern San Bernardino cos., California. The species occurs at 880-1600 m elevation. Flowering from June through August. 44. Eriogonum estneraldense S. Wats. A species of arid mountain ranges and foothills in the western portion of Nevada and adjacent eastern California, mostly at 1770-3170 m ele- vation. Flowering from June to late August. Esmeralda buclcwheat. Diffusely branched, erect, annual herbs with slender to filiform peduncles bearing small, narrowly turbinate, 5-toothed involucres with white, glabrous flowers. The var. esmeraldense is the wide- spread and common phase being found in Ne- vada in Humboldt, Esmeralda, Mineral, and Nye cos. The endemic var. toiyabense J. T. Howell is restricted to the Toquima and Toiyabe ranges in Lander and northwestern Nye COS. It differs from var. esmeraldense in having scattered glands at the very base of the stem. 45. Eriogonum concinnum Reveal Darin's buckwheat. Erect annual herb with slender to fistulose, glabrous stems bearing erect or spreading peduncles with narrow involucres and white, glabrous flowers. Endemic to the Nevada Test Site regions of southern Nye Co., occurring at 1480-2050 m elevation. Flowering from late May through August. 46. Eriogonum rubricaule Tidestrom [£. laetum S. Stokes; £. trichopes \dr. rubricaule (Tidestrom) S. Stokes]. Lahontan Basin buck- wheat. Erect annual herbs with slender to fistulose, glabrous stems bearing erect pedun- cles with broad involucres and pale yellow to yellow, glabrous flowers. Endemic to the La- hontan portion of the western Great Basin of Nevada from Mineral and Nye cos. northward to Humboldt Co., mostly 1290-1830 m eleva- tion. Flowering from late May to early Au- gust. 47. Eriogonum lemmonii S. Wats. Lem- mon's buck-wheat. Erect annual herbs with slender to slightly fistulose, glabrous stems bearing sessile or shortly pedunclcd, campan- ulate involucres with pinkish to dark red, glabrous flowers. Endemic to west central Nevada from Lyon to southern Washoe cos. eastward to Churchill and extreme southern Pershing COS., mostly 1280-1460 m elevation. Flowering from Ma\' to earl\- July. July 1985 REVEAL: Nevada Eriogonum 513 48. Eriogonum deflexum Torr. in Ives A widespread and highly variable species rang- ing from northern Nevada and Utah south- ward through southern California and Arizona to northwestern Mexico and southwestern New Mexico, mostlv below 2300 ni elevation. Flowering throughout the year but mainly from May through October. Skeleton weed. Annual, glabrous herbs with erect to spreading branches bearing de- flexed peduncles with narrowly turbinate to turbinate involucres and white flowers. The following varieties are found in Nevada. 1. Involucres turbinate, 1..5-2 mm long; peduncles up to 5 mm long; stems not inflated 2 — Involucres narrowly turbinate, (2) 2.5-3 mm long; peduncles 3-15 mm long; stems often inflated; desert ranges on the southern margin of the Intermountain Region and the adjacent edge of the Mojave Desert var. baratum 2(1). Tepals obtuse basally; plants of the Intermountain Region portion of the state var. nevadense — Tepals cordate basally; plants of the Mojave Desert portion of the state var. deflexum The Intermountain Region phase of the species is the var. nevadense Reveal which occurs throughout that portion of the state except for Eureka and Elko cos. The plant is often found on volcanic soils. It is generally restricted to the valley bottoms and foothills, where it occurs nearly always above 1300 m elevation. In the Mojave Desert portion of the state the var. deflexum is the common phase. It differs from the more northern expression in having distinctly cordate tepal bases as op- posed to the obtuse tepal bases of the flowers in var. nevadense. The var. deflexum is re- stricted to southern Nye and Lincoln cos. and Clark Co., where it generally occurs below 1100 m elevation (some populations may oc- cur as high as 1900 m in the Spring Mts.). The taller, more erect, and inflated stem phase of the species, var. baratum (Elmer) Reveal [E. baratum Elmer; E. deflexum subsp. baratum (Elmer) Munz], occurs in the mountains along the interface between the Intermountain Re- gion and the Mojave Desert. In general, var. baratum occurs from Esmeralda Co. eastward across southern Nye Co. to west central Lin- coln Co., where it is found at 1350-2050 m elevation. 49. Eriogonum rixfordii S. Stokes [£. de- flexuyn subsp. rixfordii { S. Stokes) Munz]. Pagoda buckwheat. Erect glabrous annual herbs with inflorescence branches arranged in numerous tiers of horizontally arranged branches forming a pagodalike crown, with sessile, deflexed involucres bearing small. white flowers. Local and occasionally weedy in the Death Valley region of Inyo Co., Cali- fornia, and just entering Nevada in the Beatty area northward onto the eastern foothills of the Grapevine Mts. in extreme southern Nye Co. at 990-1600 m elevation. Flowering from late June to October. 50. Eriogonum hookeri S. Wats. [S. de- flexum subsp. hookeri (S. Wats.) S. Stokes]. Hooker's buckwheat. Spreading glabrous an- nual herbs with sessile, deflexed involucres bearing yellow flowers. Widespread from east central California eastward across Nevada to eastern Idaho, Utah, western Colorado, northern Arizona and extreme northwestern New Mexico. Rather common in the Inter- mountain Region portion of Nevada except in Washoe and Humboldt cos., at 1250-2000 m elevation. Flowering from July to October. 5L Eriogonum brachypodum Torr. & Gray [E. parnji A. Gray; E. deflexuin var. brachypodum (Torr. & Gray) Munz; E. de- flexum subsp. brachypodum (Torr. & Gray) S. Stokes; E. deflexum subsp. parry i (Torr. & Gray) S. Stokes]. Parry's buckwheat. Low, spreading, glandular annual herbs forming a flat-topped inflorescence with sessile or pe- duncled, deflexed, involucres bearing white flowers. Widespread and locally common mainly on the Mojave Desert from southeast- ern California across southern Nevada to southwestern Utah and western Arizona. In Nevada the species occurs in the Lahontan Trough (Reveal 1980) as far north as Churchill 514 Great Basin Naturalist Vol. 45, No. 3 and Pershing cos. and along the southern edge of the Intermountain Region from Min- eral and Esmeralda cos. eastward across southern Nye Co. to southern Lincoln Co. southward throughout Clark Co. , at 180-1900 m elevation. Flowering from March to October. 52. Eriogonum bifurcatum Reveal Pah- rump Valley buckwheat. Low spreading glabrous annual herbs with erect peduncles bearing narrow involucres with white flowers. Local and restricted in the Pahrump and Stewart Valley area of Inyo Co., California, and adjacent Nye Co., Nevada, southward to Mesquite Valley in California and the Las Ve- gas area of Clark Co., Nevada, at 300-800 m elevation. Flowering from late April to late June. 53. Eriogonum insigne S. Wats. [£. exal- tatum M. E. Jones; E. deflexum var. insigne (S. Wats.) M. E. Jones; E. deflexum subsp. insigne (S. Wats.). S. Stokes; E. deflexum subsp. exaltatum (M. E. Jones) S. Stokes]. Exalted buckwheat. Tall, erect, glabrous an- nual herbs with elongated, whiplike inflores- cences bearing erect, sessile, or short- peduncled involucres with white flowers. Local and infrequent from southwestern Utah westward across southern Nevada to southern California. In Nevada the species is most com- mon in the Bunkerville area southwestwardly to the Hoover Dam area in Clark Co.; infre- quent elsewhere as on and near the Nevada Test Site in southern Nye Co. and in the Panaca area in Lincoln Co., mostly 300-1480 m elevation. Flowering from late June through October. 54. Eriogonum watsonii Torr. & Gray [£. deflexum subsp. watsonii (Torr. 6c Gray) S. Stokes; E. cernuum var. multipedunculatum S. Stokes; E. deflexum var. watsonii (Torr. & Gray) R. J. Davis; £. deflexum var. multi- pedunculatum (S. Stokes) C. L. Hitchc.]. Watson's buck-wheat. Low, spreading, glabrous annual herbs with slender, deflexed peduncles bearing long, narrowly turbinate involucres with white flowers. Widespread in the Intermountain Region from southeastern Oregon and southern Idaho southward to northwestern and northern Nevada. In Ne- vada the species ranges from northern Nye Co. northward to Pershing, Churchill, and Lander COS. and northeastwardK through Eu- reka Co. to Elko Co., at 1400-2200 m eleva- tion. Flowering from late May to early Sep- tember. 55. Eriogonum cernuum Nutt. Wide- spread and common throughout much of tem- perate western North America from south- western Canada southward on the western edge of the Great Plains to northern New Mexico and westward to southeastern Wash- ington, eastern Oregon, and the mountains of eastern California; found up to 3300 m eleva- tion. Flowering from June through October. Nodding buckwheat. The var. cernuum [£. cernuum var. tenuelorv. &Gray;£. cernuum subsp. tenue (Torr. & Gray) S. Stokes]. An- nual herbs with straight, cernuous peduncles bearing turbinate involucres and white flow- ers with crisped or wavy margins. Wide- spread and common throughout most of the Intermountain Region portion of Nevada and the higher mountain ranges in the Mojave Desert portion of the state; mostly 1200-3100 m elevation and flowering from June through September. The var. viminale (S. Stokes) Re- veal in Munz [£. cernuum subsp. viminale S. Stokes] is similar to var. cernuum except the involucres are sessile. The var. viminale is rather frequent on the desert floors of the Intermountain Region ranging from Elko and Eureka cos. southward to Lincoln Co., then westward to Lander and Nye cos. The variety occurs at 1500-2400 m elevation and flowers from late July to early September. 56. Eriogonum nutans Torr. & Gray Drooping buck-wheat. A low, often slightly spreading annual herb with curved, cernuous peduncles bearing campanulate involucres and white, rose, or red flowers ranging from western Utah eastward across Nevada to ex- treme eastern California and southeastern Or- egon, mostly 1350-2000 m elevation. Flower- ing from May to September. The var. nutans [£. cernuum \ dr. purpurascens Torr. &: Grdy, E. ruhriflorum M. E. Jones; £. nutans var. hrevipedicellatum S. Stokes] is the most com- mon expression of this relatively rare species and in Nevada is found from Elko and White Pine cos. westward to Mineral, Esmeralda, and Washoe cos. It is characterized by glandu- lar peduncles. The var. glahratum Reveal is restricted in Nevada to Elko Co., where it occurs along Interstate Highway 80 from near Elko to Wells. It diflcrs from \ar. nutans in July 1985 Rfa'eal: Nevada Erioconum 515 having glabrous peduncles. The var. glabra- tiim has been found near Hirschdale, Nevada Co., California, also on Interstate Highway 80, where it was likely introduced as a result of" highway trafiPic. 57. Eriogonum thomasii Torr. Thomas buckwheat. Low, spreading, often diffusely branched, annual herbs, glabrous throughout except for a few scattered glands near the base of the flowering stems, the spreading, capil- lary peduncles bearing small involucres of yel- low to white or rose, short-hispidulous flowers with the cordate base of the outer tepals dis- tinctly saccate-dialated at full anthesis. Wide- spread and often common on the warm deserts of northwestern Mexico northward through southern California and southern Ne- vada to southwestern Utah and western Ari- zona, where it occurs from below sea level to 1400 m elevation. Flowering from March through June. In Nevada the species is re- stricted to the Mojave Desert portion of the state in southern Nye and Lincoln cos. and Clark Co. 58. Eriogonum pusillum Torr. & Gray [E. comosum var. plaijanum M. E. Jones; £. reni- fonne var. asarifoliiim Cand.; E. reniforme subsp. pusillum (Torr. & Gray) S. Stokes; E. reniforme var. plaijanum (M. E. Jones) S. Stokes]. Puny buckwheat. Spreading annual herbs, glabrous throughout except for glands near the base of the flowering stems, the spreading, slender peduncles bearing broad, glandular involucres with yellow to reddish yellow, glandular flowers. Widespread and often common on the warm deserts of south- ern California northward through the Lahon- tan Trough of western Nevada to southeastern Oregon and southwestern Idaho and eastward across southern Nevada to southwestern Utah and western Arizona, where it occurs from near sea level to 1800 (rarely 2600) m eleva- tion. Flowering from March through June. In Nevada, the species ranges from the Mojave Desert portion of the state in Clark, southern Lincoln, and Nye cos. northward through the Lahontan Trough (Reveal 1980) to Washoe Co., occurring at 680-1900 m elevation. 59. Eriogonum reniforme Torr. & Frem. [£. reniforme var. comosum M. E. Jones; E. comosum {M. E. Jones) M. E. Jones]. Kidney- leaved buckwheat. Spreading annual herbs, glabrous throughout except for a few scattered hairs near the base of the flowering stems, the spreading, slender to capillary peduncles bearing broad, glabrous involucres with yel- low to yellowish red, glandular flowers. Wide- spread and often infrecjiient and local from northern Baja California, Mexico, and south- ern California eastward to western Arizona and north to southern and southwestern Ne- vada from near sea level to 1600 m elevation. Flowering from March through June. In Ne- vada the species occurs throughout the Mo- jave Desert portion of the state and extends up the Lahontan Trough to Churchill Co. It differs from the related E. pusillum (see no. 58) in the nature of its leaf pubescence and its glabrous, often glaucous involucres. 60. Eriogonum viscidulum J. T. Howell Clammy buckwheat. Tall, erect, diffusely branched, minutely viscid annual herbs with erect, filiform peduncles bearing small, nar- row involucres with small yellow flowers. En- demic and rare along the Virgin River, Clark Co., Nevada, mostly 300-475 m elevation. Flowering from April to late June. 61. Eriogonum collinum Stokes ex Jones Hilly buck"wheat. Low to tall and erect, essen- tially glabrous annual herbs with open inflo- rescences of upwardly curved peduncles bearing glabrous involucres of white to yel- lowish flowers with pustulose tepals. Local and widely scattered from northwestern Ne- vada and adjacent northeastern California northward to southeastern Oregon, occurring at 1300-2000 m elevation. Flowering from June to mid-September. In Nevada the spe- cies ranges from southern Wahoe Co. and Lyon Co. northward to west central Hum- boldt Co. 62. Eriogonum salicornioides Gand. [E. demissum S. Stokes; E. demissum var. ro- manum S. Stokes; E. viinineum var. sal- icornioides (Gand.) S. Stokes]. Glasswort buckwheat. Low, spreading, glabrous annual herbs with sessile or short-peduncled involu- cres bearing small white flowers with pustu- lose tepals. Rare and infrequent in heavy clay soil of southeastern Oregon and southwestern Idaho, just bearly entering Nevada in north- ern Humboldt Co. (Train s.n. — PAG) at 1000-1400 m elevation. Flowering from late April to mid-August. 63. Eriogonum pharnaceoides Torr. in Sitgr. Ginseng buckwheat. Erect villous 516 Great Basin Naturalist Vol. 45, No. 3 herbaceous annual herbs with green, basal, and sheathing leaves, the slender and erect peduncles with campanulate involucres bear- ing 5 lanceolate lobes 1-3 mm long and white or yellow flowers with saccate-dialated tepals. Widespread and infrequent from southeast- ern Nevada and adjacent southern Utah south to northern Arizona and western New Mex- ico, mostly at 1350-2500 m elevation. Flower- ing from August through October. The Ne- vada expression is the yellow-flowered var. cervinum Reveal that is known in Nevada only from the Deer Lodge area of Lincoln Co. It also occurs in southwestern Utah (Iron and Washington cos.) and northern Mohave Co., Arizona. The var. pharnaceoides is restricted to Arizona and New Mexico. 64. Eriogonum spergulinum A. Gray Spurry buckwheat. Prostrate to spreading or erect hispid and often glandular annual herbs with basal and sheathing leaves, the filiform peduncles with turbinate involucres bearing 4 erect teeth and white, glabrous or sparsely pubescent flowers. Widespread and occasion- ally weedy from southern and eastern Califor- nia northward to southeastern Oregon and southwestern Idaho, mostly 1450-3450 m ele- vation. Flowering from June to late Septem- ber. The Nevada expression is the widespread and common var. reddingianum (M. E. Jones) J. T. Howell [E. spcrgulinuin subsp. reddingianum (M. E. Jones) Munz ex Reveal] that occurs throughout the range of the spe- cies. In Nevada the variety ranges from Min- eral Co. northward to Washoe and Humboldt COS., where it occurs at 1500-2800 m eleva- tion. Other variants of the species are re- stricted to the Sierra Nevada of California. 65. Eriogonum maculatum A. A. Heller [E. angulosumya.r. rectipesGand.,E. angulo- sum var. pauciflorwn Gand.; E. angulosum var. flabellatum Gand.; E. angulosum var. patens Gand.; E. thurheri var. acutangulum Gand.; E. angulosum subsp. maculatum (A. A. Heller) S. Stokes; E. cernuum subsp. acu- tangulum (Gand.) S. Stokes]. Spotted-flow- ered buckwheat. Low, spreading, tomentose annual herbs with sheathing leaves, the fili- form, spreading peduncles bearing campanu- late involucres with white to yellow or pink to red, glandular-puberulent flowers, the outer tepals often with a single large purplish spot. Widespread and common in the desert re- gions of extreme northern Baja California, Mexico, northward through southern and eastern California to southern Washington, and eastward to southern Idaho, western Utah, and western Arizona, from just above sea level to 2450 m elevation. Flowering from April to November. In Nevada the species occurs essentially throughout the state at 510-2100 m elevation and flowers from late April through September. 66. Eriogonum ampullaceum J. T. Howell [E. mohavense var. ampullaceum (J. T. How- ell) S. Stokes]. Bottle-shaped buck-wheat. An erect slender annual with strictly basal leaves, sessile involucres, and minute flowers with enlarged, roundish, reddish bases and white tepal tips. Rare and infrequent in Mono Co., California, and in adjacent Mineral Co., Ne- vada, at 1980-2105 m elevation. Flowering from late June through August. In Nevada this species is known only from Alkali Valley on the north side of Alkali Lake (Tiehm 6- Lavin 8143 — MARY). This represents a new record for the state. 67. Eriogonum vimineum Dougl. ex Benth. [E. shoshonense A. Nels.; E. vimineum var. shoshonense (A. Nels.) S. Stokes]. Wicker buckwheat. Erect glabrous to floccose annual herbs with basal leaves and occasional smaller cauline leaves, the involu- cres sessile and strongly angled, bearing white to rose or yellow glabrous flowers. Widespread and common in the Pacific Northwest from Washington and Idaho south- ward to central California and northwestern Nevada, 30-2400 m elevation. Flowering from June through September. In Nevada the species occurs from the Carson City area northward to Washoe, Humboldt, and Elko COS. at 1500-2400 m elexation. 68. Eriogonum baileyi S. Wats. Erect glabrous or tomentose annual herbs with basal leaves, sessile and small involucres, and white, glabrous or, more commonly, glandu- lar-puberulent flowers. Widespread and often rather common from eastern California north- ward to Washington and eastward across Ne- vada and Idaho to western Utah, mostly 460-2250 m elevation. Flowering from May through September. Bailey's buckwheat. In Nevada the com- mon expression is var. haileyi [E. gracile var. effusum Torr. & Gra\'; E. haileyi \'ar. })or- July 1985 REVEAL: Nevada Erkx.onum 517 phyreticinn Stokes ex Jones; E. restioides Gaud.; E. vimineum subsp. baileyi (S. Wats.) S. Stokes; £. vimineum viw. restioides (Gdnd.) S. Stokes; E. vimineum var. porphijreticum (Stokes ex Jones) S. Stokes; £. vimineum var. baileyi (S. Wats.). R. J. Davis] that is found nearly throughout the state (exeept Clark Co.). It oecurs at 1300-2250 m elevation. The var. praebens (Gand.) Reveal [£. leucu- cladum Gand.; £. praebens Gand.; £. praebens war. divaricatum Gand.; £. commix- tum Greene ex Tidestrom; £. vimineum var. commixtum (Greene ex Tidestrom) S. Stokes; £. baileyi var. divaricatum (Gand.) Reveal in Munz] differs from var. baileyi in having to- mentose rather than glabrous branches and stems. The var. praebens ranges from eastern California northward into western Nevada. In Nevada the variety is common from Douglas Co. northward to central Washoe Co., then eastward to Humboldt and Eureka cos. (where rare). 69. Eriogonum brachijanthum Coville [£. baileyi var. brachyanthum (Coville) Jepson; £. baileyi subsp. brachyanthum (Coville) S. Stokes]. Short-flowered buckwheat. Low, rounded, glabrous annual herbs with basal leaves, sessile involucres, and minute yellow flowers. Common on the desert valley floors and foothills from southern California north- ward to Nevada at 610-1900 m elevation. Flowering from May through August. In Ne- i vada the species ranges from Nye Co. north- ward to Washoe Co., then eastward to west- ern Humboldt and Lander cos., 1250-1900 m elevation. 70. Eriogonum nidularium Coville [£. vimineum subsp. nidularium (Coville) S. Stokes]. Bird-nest buckwheat. Low, spread- ing, tomentose to floccose annual herbs with numerous incurved branches and basal leaves, the involucres minute and sessile, bearing yellow to reddish flowers. Common on the desert valley floors and foothills of southern California eastward to Arizona and northward to Oregon and Idaho, from near sea level to 2150 m elevation. Flowering April through October. In Nevada the species is found throughout the state, mainly at 1300-2150 m elevation. 71. Eriogonum palmerianum Reveal in Munz [£. phanatella var. pabneri Torr. & Gray; £. baileyi var. tomentosum S. Wats.]. Palmer s buckwheat. Low, spreading, tomen- tose to floccose annual herbs with few to many spreading branches and basal leaves, the in- volucres minute and sessile, bearing white flowers. Common on the desert valley floors and foothills of southern California eastward to extreme southwestern New Mexico, and northward to Nevada, Utah, and southwest- ern Colorado, mostly 670-2680 m elevation. Flowering from June through October. In Ne- vada the species is common nearly throughout all the state except the northwestern corner, 1300-2300 m elevation. 72. Eriogonum puberulum S. Wats. [£. puberulumvAr. venosumS. Stokes]. Puberu- lent buckwheat. Low, spreading, silky-pu- berulent, reddish annual herbs with basal leaves, the involucres divided to near the base into 4 lobes, with white to bright red, glabrous or hispidulous flowers. Local and infrequent mainly on volcanic soils of Inyo Co. , Califor- nia, eastward across Nevada to southwestern Utah, mainly 1300-2850 m elevation. Flower- ing from May through August. In Nevada the species is found in Nye, Eureka, White Pine, Lincoln, and extreme eastern Clark cos. 73. Eriogonum darrovii Kearney Dar- row's buckwheat. Low, spreading, sericeous annual herbs with several to many compact branches bearing basal and cauline leaves, the involucres minute and 5-lobed with pale yel- low to pink and white-hirtellous flowers. Rare and infrequent in northwestern Arizona and southeastern Nevada, mostly 1650-1880 m el- evation. Flowering from late June to early September. In Nevada the species is known only from near Sunnyside, Nye Co., and south of Major's Place, White Pine Co. The latter locality was reported by Barneby (1947) as £. divaricatum Hook. , a species not known to occur in Nevada. Index TO Names The following index is to entities of Eri- ogonum mentioned in the above treatment. Those marked by an asterisk (*) indicate ba- sionyms established on type material gath- ered in Nevada (see Reveal 1985b for more details). Synonyms are printed in italics. acaule shockletji, 21 andinui7J, 35 ampullaceum, 66 anemophilum, 14* angulosuni 518 Great Basin Naturalist Vol. 45, No. 3 flahellatum, 6.5* maculatum, 65 patens, 65* pauciflorum, 65* rectipes, 65 anserinum, 29 argophylluni, 17* aridtim, 31* aureum, 3 glutinosum, 3 azaleastrum, 31* baileyi, 68 brachyanthum, 69 divaricatum, 68 porphyreticum, 68* praebens, 68 tomentosum, 71 baratiim, 48 beatleyae, 13* bifurcatum, .52* biumbellatum, 31 bloomeri, 30 brachypodum, 51 caespitosum, 35 douglasii, 34 capitatum, 26 corneum, 43 cernuum, 55 actitanguhim, 65 muUipedunculatum. .54* purpurascens, 56* tenue, .55* viminale, .55* chrysucephahim desertoniin, 7* collinum, 61* cornmixium, 68* coinosiim, 59 playanum, .58* concinnum, 45* confertiflorum, 1 contiguum, 40 cordatum, 41 corymbosum aureum, 3 glutinosum, 3 crispum, 3 Crosby ae, 11 custcfctt californicum, 29* darrovii, 73 davisianum, 28 deductum, 26 deflexum, 48 baratum, 48 brachypodum, 51 deflexum, 48 exaltatum, 51 hookeri, 50 insigne, 51 multipedunculatum, .54 nevadense, 48* parry i, 51 rixfordii, 49 watsonii, .54 demissum, 62 romanum, 62 desertorum, 7 divaricatum, 72 douglasii, .34 dudleyanum, 2 effusum foliosum, 1 nelsonii, 1 simpsonii, 1 elatum, 27 erianthum, 27* villosum, 27 eUipticu7n, 31 esmeraldense, 44* toiyabense, 44* exaltatum, 51* exiinium, 28* fasciculatum polifolium, 6 revohitum, 6 ferrissii, 31 friscanum, 1 fruticosum, 3 glandulosum, 43 carneum, 43 gracile effusum, 68* gracilipes, 18 harfordii, 26 heermannii, 4 argense, 4 clokeyi, 4* floccosum, 4 humilius, 4* sulcatum, 4 heracleoides, 32 tiinV/c, 31* holmgrenii, 19* hookeri, 50 howellianum, 42 howellii, 4* incanum, 36 inflatum, 39 contiguum, 40* deflatum, 39 fusiforme, .39 insigne, .53 jonesii, 3 kearneyi, 2 monoense, 2 kingii, 16* latifolium niidutn. 26 sulphureum, 26 laetum, 46* latens, 37 lemmonii, 47* leucocladum, 68* lewisii, 10* lobbii, ,38 minus, 38 robustum, 38 macnlatum, 65 mariiblium, 36 microthecum, 1 ambiguum, 1 aureum, 3 confertiflorum, 1 crispum, 3 cxpansum, 1 foliosum. 1 friscanum, 1 intermedium, 1* lapidicola, 1* laxiflorum, 1 rigidum, 1 simpsonii, 1 spathulare, 1 mohavense ampullaceum, 66 monticola, 37* neglectum, 31 nelsonii, 1 nevadense, 8* nidularium, 70 nivale, 28 jaegeri, 5 kearneyi, 2 monoense, 2 nodum, 26 deductum, 26 gramineum, 26 oblongilolium, 26 publiflorum, 26 nummulare, 2 nutans, 56 brciipedicellatum, .56* glabratuni, .56* ochrocephalum, 8* agnellum, 12 alexanderae, 8* obtusum. 25 orthocaulon, 28 orthocladon, 25 ovalifoliuni, 28 caelestinum, 28* celsum, 28 depressum, 28 eximium, 28 flavissimum, 29 nevadense, 28* nivale, 28 orthocaidon, 28 williamsae, 28* palmeri, 5 panamintense, 2.3 mensicola, 23 parriji, 51 pharnaceoides, 63 cervinum, 63 plumatclla, 5 jaegeri, 5 palnwri. 71 polifoHum. 6 portcri, 31 praebens, 68 divaricatum. 68* procidnum, 9 proliferum, 29 anserinum, 29 puberulum, 72 venosum, 72* pulvinatum, 21 purpureum, 28 pusillum, 58* racemosum, 25 desertorum, 25 reclinatum, 31 reliquum, 23* reniforme, 59 asarifolium, .59* como.suffi, 59* playanum, 58 pusillum, 58 restioides, 68* rcvolutum, 6* rhodanthum, 28* rixfordii, 49 robustum, 38* rosense, 12* rubidum, 28 rubiflorum, 56 rubricaule, 46* rupinum, 24* salicornioides, 62 saxatile, 30 multicaule, 30* stokesae, 30 sericolcucum. .35* shockleyi, 21* candidum, 21 shoshonense, 67 simpsonii. 1 soredium, 21 spathulare, 1 spergulinum reddingianum, 64 sphaerocephalum, 33 halimioides, 33 sericolcucum. 35 stcllatum. 31 stokesae, 30 strictum anserinum, 29 flavissinuim. 29 proliferum, 29 subal))i>uii)i. 31 sulcatum. 4 sulphureum. 26 tenellum erianthum. 1 scssilijlorum. 1* thomasii, 57 thurberi acutangulum, 65* tiehmii, 15* tiichopes, n corddtum, 41 glaiuhdosum, 43 ruhi-icaide, 44 tricho])odium. 41 July 1985 REVEAL: Nevada Erioconum 519 minus, 41 umbt'llatiini, 31 aridum, 31 aureum, 31 californicum, 31* dichrocephaliiin, 31 ellipticuni, 31 ferrissii, 31 furcosuni, 31 juniporinuin, 31* majus, 31 nevadense, 31* porteri, 31 stelhitiiiu. 31 siihalpiituni, 31 suharidiiin, 31* vernum, 31* versicolor, 31 illiflonim, 20 (•(iiulidinn, 21 iniiiu'iiin, 67 Ixiilciji, 68 coniniixtiiin, 6S ni(ltil(irimii. 70 porphyreticum, 68 restioide.s, 68 s(dicornioides , 62 shoshonense, 67 visciduliini, 60* watsonii, 54* wrightii, 22 subscaposum, 22 Literature Cited Barneby, R. C. 1947. Distributional notes and minor nov- elties. Le;ifl. W. Bot. 5:61-66. PlNZL, A 1983. 1983 revised maps. Unpublished, Nevada State Museum, Capitol Complex. Carson City, Nevada. Reve.al, J. L 1971. Notes on Erio^onum — VI. A revision of the Eriof!,oi}um inicrothecian complex (Polygo- naceae). Brigham Young Univ. Sci. Bull., Biol. Ser. 13(l);l-45. 1972. Two new species of Eriogonurn (Polygo- naceae) from California and adjacent states. Aliso 7:415-119. 1980. Intermountain biogeography — a specula- tive appraisal. Mentzelia 4:1-92. 1981. Notes on endangered buckwheats {Eri- ogonurn: Polygonaceae) with three newly described from the western United States. Brittonia 33:441-448. 1983. The Demoulin Rule and newly mandated combinations in Eriogonurn (Polygonaceae). Taxon 32:292-295. 1985a. New Nevada entities and combinations in Eriogonurn (Polvgonaceae). Great Basin Nat. 45: 27^280. 1985b. Types of Nevada buckwheats. {Eriogonurn: Polygonaceae). Great Basin Nat. 45:488-492. TiDESTROM. 1 1925. Flora of Utah and Nevada. Contr. U.S. Natl. Herbarium 25:1-665. HIGH RATES OF PHOTOSYNTHESIS IN THE DESERT SHRUB CHRYSOTHAMNUS NAUSEOSUS SSP. ALBICAULIS Tim D. Davis', N. Sankhla^ W, R. Andersen', D. J. Weber', and B. N. Smith' Abstract. — Basic aspects of photosynthesis were investigated in white rubber rabbitbrush (Chrysothamnus nau- seosus (Pallas) Britt. ssp. albicaulis), a common C3 deciduous shrub native to arid regions of the western U.S. Under favorable field conditions, net photosynthesis (P„) ranged from 36 to 73 mgCOj dm ^ hr ', which is relatively high for a woody species. The leaves from the actively growing flowering shoots exhibited higher P^ than those on the vegetative shoots. Pn also varied according to the age of the leaves and the location of the plants. P„ did not light saturate even at quantum flux densities (QFD) equivalent to full sunlight. The light compensation point was relatively high (ca 100 [Jtmol ■ m'" • S~'), perhaps due to the presence of a tomentose vestiture on the leaf surface. At high QFD's, the stomatal conductance was high (ca 520 mmol m~ s') for a woody species. RUBP-carboxylase content of the leaves ranged from 20 to 22 mg per gram F.W., which is similar to that found in most C3 crop species. These results suggest that rabbitbrush is able to maintain high rates of P^, at least under nonstressed conditions. The genus Chrysothamnus (rabbitbrush) consists of deciduous subshrubs or shrubs en- demic to western North America in open plains, valleys, foothills, and mountains (McArthur et al. 1979). Like other species of the genus, Chrysothamnus nauseosus (Pallas) Britt. (rubber rabbitbrush) is an excellent plant for soil stabilization because of its deep roots, heavy litter, and ability to establish on severe sites. It can grow in the cold deserts of the Colorado plateau, the Great Basin plateau, and the warm deserts of the south- western U.S. In fact, Chrysothamnus is able to survive and grow vigorously from Mexico to Canada, an area that represents a very wide range of environmental conditions. Recently there has been a renewed surge of interest in rabbitbrush as a nontraditional source of rubber. Acquisitions containing up to 6% rubber per unit dry weight have been reported (Ostler 1980). As a part of an on-go- ing project on the potential oi Chrysothamnus as a rubber source, we became interested in evaluating factors controlling rubber produc- tion. Very little information is available on the photosynthetic potential and physiological characterization of rabbitbrush. This paper describes some basic aspects of the photosyn- thetic characteristics of this potentially impor- tant plant. Materials and Methods For initial experiments, rates of net photo- synthesis (P„) were measured during Septem- ber and October 1984 on vigorous, healthy, white rubber rabbitbrush plants growing in the Range Plant Garden at Brigham Young University (elevation approximately 1500 m) using the in situ CO, depletion technique (Ehleringer and Cook 1980). The cuvette was clamped onto the shoots for 45 sec. The CO2 analysis system consisted of a Beckman 865 Infrared Analyzer through which No gas was flowing. Samples were injected into the gas stream, and sample peaks were printed out by a Hewlett-Packard Model 3390-A Reporting Integrator. The rate of CO2 exchange was cal- culated from the CO2 depletion rate based on the chamber volume, using the ideal gas equation. Plants were about three years old and had been irrigated periodically through- out the summer. For comparative purposes, P„ was also measured on healthy, vigorous plants of several additional woody species growing under similar enxironmental condi- tions. P„ was also measured on rabbitbrush plants growing on three native sites in Utah County, Utah. These plants had received con- siderable rainfall about fi\ e da\ s prior to mea- surement. Rates of P„ were expressed on leaf area, dry weight, and chlorophyll bases. Leaf College of Biological and Agricultural Sciences, Bri^l Permanent address: Department of Botany, I'niscrM tyotjo.lhpur, Ituli. 520 July 1985 Davis etal.: Shrub Photosynthksis Rates 521 area was determined using Li-Cor model LI- 3000 and LI-3100 area meters. Because of the narrowness of rabbitbrush leaves, it was found that the LI-3000 area meter underestimated leaf area by 45%. Hence leaf area data ob- tained by the LI-3000 area was multiplied by 1.82 to correct for this error. Chlorophyll con- tent was determined using the method of Lichtenthaler et al. (1982), and the dry weights were determined by oven drying at 70 C for at least 16 hours. To determine the response of rabbitbrush P„ to light, shoots from plants growing in the Range Plant Garden were excised under wa- ter and placed in an open gas exchange system as described in detail by Ehleringer (1983). Shoots were first exposed to a quantum flux density (QFD) of 2650 |xmol m"' S \ The QFD was then lowered in steps down to 80 |jLmol m"^ ■ S~\ Shoots remained at a given QFD until a stable photosynthetic rate was achieved (usually about 45 min.). Leaf tem- perature was held constant at 25 C, and COo concentration was about 350 fxl 1' during the measurements. Stomatal conductance and in- tercellular COo concentrations were calcu- lated as described previously (Ehleringer 1983). For enzyme assays, fully expanded leaves were collected from young vigorous shoots. One-gram samples were each ground in 5 ml buffer (0.1 M Tris-Cl, pH 8.2 (25 C), 20 mM MgClo, 4 mM ethylendiaminetetraaceticacid, 4 mM dithiothreitol, and 5% acid and deion- ized HoO-washed polyvinylpolypyrrolidone (Sigma). Assay and activation of ribulose bis- phosphate carboxylase (RuBPCase) followed the method of Lorimer et al. (1977). Concen- tration of NaH'^C03 was 10 mM in the activa- tion and assay media. Concentration of ribu- lose-1,5-bisphosphate was a 0.4 mM with MgCL at 20 mM in the activation and assay media. Determination of amount of enzyme followed the Beckman Model-E ultracen- trifuge method of Andersen et al. (1970) using the Schlieren optical system at a bar angle of 50°. Amount of enzyme as mg ml' extract was determined by dividing the 15 x magni- fied area of the Schlieren peak by a factor of 4.107. For scanning electron microscopy (SEM), leaf tissue was fixed in glutaraldehyde- acrolein (Hess 1966). After dehydration to Net photosynthesis ( mg CO • dm ■ -hr" ) o S Stomatal conductance (mmolm'-i) or Intercellular CO, (yllit^r) Fig. 1. Response of net photosynthesis (P„), stomatal conductance (gHoO), and intercellular C02(C,) of rabbit- brush to incident quantum flux. acetone, tissue was critical-point dried and sputter coated with gold. Results White rubber rabbitbrush consistently ex- hibited high P„ rates for a woody species (Fig. 1; Tables 1-4). Rates obtained with intact shoots in the field using the CO, depletion technique compared favorably with those ob- tained with excised shoots in the open gas exchange system. On a leaf area basis, the P^ in this plant was considerably higher than that of the other woody C3 species (Table 1). Even on a dry weight basis, the P in rabbitbrush equaled that of Atriplex canescens, a C4 spe- cies, and was nearly twofold or more than that of the other C3 species. When calculated on chlorophyll basis, only the C4 A. canescens exhibited a higher P„ rate than that of rabbit- brush. The P„ rates in plants growing on the nonirrigated native sites were found to be somewhat lower than those recorded for the plants maintained at the Range Plant Garden (Table 2). In rabbitbrush the stem is also photosyn- thetic, and leaf senescence progresses acro- petally. To assess the contribution of stem photosynthesis in overall P„ as well as to eval- uate the role of leaf senescence, additional 522 Great Basin Natur.\list Vol. 45, No. 3 Table 1. Net photosynthetic rates (P„) of Chrysothamnus nauseosiis ssp. albicaulis and six other woody species under similar environmental conditions. All plants were growing outdoors under irrigated conditions except for A. tndentata, which was measured on a native site after a heavy rainfall. Plus/minus values indicate standard of error of the mean (n = 4). Environmental Pn conditions QFD Temp. Species mgCO, dm ^ hr ' mgCO, g 'D.W. hr mgCOa mg 'Chl hr (ixmol.m " s ' ) rc) Chrtjsothamnus nauseosus 45.9±1.2 31.3±1.8 5. 9 ±0.3 1050 20 Acer saccharimim 16.8±1.9 21.0±2.4 6.4±0.7 1700 24 Artemisia cana 26.8±2.9 17.4±2.3 3.1±().4 1150 20 Artemisia tndentata 30.0±3.8 13.9±1.0 5. 3 ±0.4 1600 26 Atriplex canescens 5.5.8±2.7 30.2±2.0 8.3±0.6 1050 20 Ephedra viridis — * 5.2±0.5 3.7±0.4 1600 20 Maltis domestica "Red Delicious' 23.1±2.1 18.9±1.7 4.0±0.4 1800 26 ♦This species has photosynthetic stems and hears no leaves. Hence ?„ on a leaf area basis was not calculated Table 2. Net photosynthesis by Chrysothamnus nauseosus at three native locations in Utah County, Utah. Plus/minus values indicate standard error of the mean (n = 4). All measurements made at QFD of 1800 fjimol m ' s ' and on nonflowering shoots. Location mgCOidm'hr"' mgCO, g ' D.W. hr' 1. Provo Canyon— partially shaded, 39.5±4.3 42.0±4.6 southern slope, near Provo River (temp. = 20 C) 2. Provo Canyon— open field, flat, full 36.1±2.1 28.7±1.6 sun (temp 20 C) 3. Mouth of Rock Canyon— slight 40.9±2.2 28.6±1.6 western slope, full sun (temp. = 26 C) Table 3. Net photosynthesis rates (P„) of nonflowering and flowering shoots oi Chrysothamnus nauseosus with and without leaves. Plus/minus values indicate standard error of the mean (n = 5). Shoot type mgCOa dm ^ hr ' mgCOj g 'D.W. hr ' mgCO. gChl^ hr ' Nonflowering, leaves intact 54.3±3.7 45.2±3.1 10.5±0.7 Flowering, leaves intact 73.3±6.7 61.5±5.6 16.8±1.5 Nonflowering, leaves removed — 3.1±1.2 2.7±1.0 Flowering, leaves removed — 4.7±1.6 4.3±1.5 Table 4. Net photosynthesis of the ti'nnina! 10 cm and the adjacent 10 cm i)cl()\\ Chrysothamnus nauseosus shoots. Plus/minus values indicate standard error of the mean (n 4). Environmental conditions during measurement: QFD - 1700 nmolm'^s ', Temp. =20C. Shoo! 1()( ation mgCOadm^hr' mgCO, g 'D.W. hr ' rngmgChl' hr Upper 10cm 48.2±8.2 28.1±4.6 9.2±1.3 Section between 10 and 20 cm below apex 28.4±7.2 13.9±4.8 6.8±2.3 mea.siireineiits of P„ were undertaken. The leaves on both flowering and nonflowering results iudicate that stem P,„ on a dry weight shoots (Table 3). Leaves on flowering shoots at basis, w as only about 7% of that observed for anthesis exhibited about a 35% higher P„ than fulv 1985 Davis etal.: Shrub Photosynthesis Rates 523 Fig. 2. Scanning electron micrograph of rahbitbrush leaf surface (X300) showing pubescence. Arrows point to stomata(s). the leaves on nonflowering shoots on the same plant. Results relating to the effect of leaf senes- cence on Pn are presented in Table 4. The leaves on the terminal 10 cm of rahbitbrush shoots exhibited a P„ rate that was about two- fold higher that observed on the same shoots just 10 cm below the terminal section that had senescing leaves. In this plant P„ did not light saturate at QFD's near full sunlight (Fig. 1). The light compensation point was also found to be rela- tively high (ca 100 ixmol ■ m ' • S"'). A SEM of the leaf surface revealed that rahbitbrush leaves are covered with a dense pubescence (Fig. 2). High P„ rates in rahbitbrush were accompanied by high stomatal conductance values (Fig. 1). The intercellular CO^ concen- trations in rahbitbrush leaves were similar to those found in other C3 drought-deciduous shrubs that exhibit high P„ and stomatal con- ductance (Ehleringer and Bjorkman 1978, Comstock and Ehleringer 1984). RuBPCase content and catalytic activities at V„,_,^ were measured for separate extractions of six samples of fully expanded leaves from activelv growing shoots. Amount of RuBP- Case was 21 ± 1.25 mg g^' F.W., 12.93 ± 0.74 mg • mg ' chlorophyll, and 1.07 ± 0.06 mg cm " leaf area. The catalytic activities of RuBPCase at V,,,,,, levels of substrate were 712 ± 96 nmol '^COo • mg"' enzyme min"\ 9.66 ± 0.87 fxmol '^Cb., • mg"' chlorophyll ■ min~\ and 761 ± 102 nmol "CO. cm"" min~\ Discussion Desert ecosystems are inhabited by a vari- ety of plant life forms including ephemerals, drought-deciduous and desiccation-tolerant evergreen shrubs and herbaceous perennials (Walter and Stadelmann 1974, Ehleringer 524 Great Basin Naturalist Vol. 45, No. 3 and Mooney 1983). Moisture stress is a con- tinual limiting factor for the photosynthetic process in most desert plants. Therefore, it is not surprising to note that desert plants adapt the photosynthetic apparatus to desiccation tolerance and/or drought avoidance. It ap- pears that white rubber rabbitbrush has re- sorted to drought avoidance in its photosyn- thetic adaptation. Like ephemerals (Mooney et al. 1976), this deciduous shrub is able to maintain high rates of P„ during nonstress periods. In fact, the rates of P„ exhibited by rabbitbrush were found to be similar to those observed in well-watered, drought-deciduous shrubs of warm deserts that typically exhibit very high P^ rates for woody species (Eh- leringer and Mooney 1983). The rates of P„ in rabbitbrush even compared well with those found in many herbaceous C3 crop species (Leopold and Kriedemann 1975). Rates of P„ were particularly high in flowering shoots at anthesis. The reason for this is not clear, but the presence of reproductive sinks is known to stimulate P„ in at least some plants (Milthorpe and Moorby 1974). The higher amounts of RuBPCase per unit leaf area in rabbitbrush could also be a signifi- cant factor contributing to high rates of P„ in this species. In seedling leaves of Pisitm sa- tivum and in fully expanded leaves of Medicago sativa, the estimated amounts of RuBPCase were 11 and 21 mg g ' F.W., respectively (Gordon et al. 1978, Meyers et al. 1982). Thus, on a fresh weight basis, the amount of enzyme in rabbitbrush leaves is similar to that foimd in C3 crop species. Based upon mass of enzyme per unit leaf area, how- ever, RuBPCase in rabbitbrush is relatively high (1.07 mg cm "). Some of the highest amounts of this enzyme on a leaf area basis (0.5 mg • cm^) for C, species have been mea- sured in desert winter annuals (Seemann et al. 1980). Depending upon leaf age, RuBPCase amounts in soybean were 0. 1 to 0.4 mg • cm ' (Wittenbach et al. 1980). Similarly, in fully expanded leaves of spinach the amount of this enzyme was 0.3 mg cm " (Seemann and Berry 1981). The amount of RuBPCase on a leaf area basis measured in our study of rabbit- brush exceeds the extremes of these values by more than twofold. However, on the bases of chlorophyll or fresh weight, rabbitbrush val- ues are similar to those in other (>, species. The carboxylation activities for rabbitbrush per mg RuBPCase or per mg chlorophyll are also similar to that found in other C3 species (Koivuniemi et al. 1980, Seemann et al. 1980, Seemann and Berry 1981). However, car- boxylation activities in rabbitbrush at V^,^^ were high on a leaf area basis compared to other C3 species showing high RuBPCase ac- tivities per unit leaf area (Mooney et al. 1976). Among the different desert life forms, the highest photosynthetic rates and leaf conduc- tances have been recorded for ephemerals (Mooney et al. 1976, Mooney and Ehleringer 1978, Ehleringer et al. 1979). Some of the drought-deciduous shrubs and herbaceous perennials, which are active for somewhat longer periods than ephemerals, also have high P„ rates (Ehleringer and Bjorkman 1978). High P„ rates in rabbitbrush were also accom- panied by high stomatal conductance. How- ever, it should be pointed out that these high conductance values were measured on well- watered plants. Under more dry native condi- tions, such high values may not be observed. Even the intercellular COo concentrations in rabbitbrush leaves were found to be similar to C3 drought-deciduous shrubs, which exhibit high P„ and high stomatal conductance (Eh- leringer and Mooney 1983). Leaves of many desert plants are pubescent. The presence of pubescence not only modulates leaf spectral characteristics and leaf boundary layer resistance, but it also reduces leaf absorptance resulting in reduced heat load, lower leaf temperatines, and lower transpiration rates (Ehleringer and Bjorkman 1978, Ehleringer and Mooney 1978) and may have adaptive significance. In rabbitbrush the shoot is covered with a green, yellow-green, gray-green to white, feltlike tomentum, and the leaves are clothed with a tomentose vesti- ture. A characteristic feature of P„ in rabbit- brush is that it is not light saturated at QFDs near full sunlight. In addition, the light com- pensation point is relatively high. Similar re- sponses of P„ to light ha\ e also been recorded in some C3 species native to the Sonoran desert (Ehleringer and Bjorkman 1978). In Enci'lia farinosa, the high light saturation point has been attributed to the pubescent nature of the leaf surflice. It is likely that the nonsaturation of P„ at near full sunlight and July 1985 Davis etal.: Shrub Photosynthesis Rates 525 the high hght compensation point in rabhit- brush is due to its tomentose vestiture. In conchision, this study indicates that rab- bitbrush is capable of maintaining high photo- synthetic rates during nonstress periods. Thus, at least under favorable environmental conditions, a potential exists for high rates of dry matter accumulation per unit of biomass. The extent to which the dry matter produc- tion can be partitioned into rubber as well as an elucidation of the factors promoting rubber production should be a worthwhile subject for future investigation. It will also be of interest to determine how P„ responds on a seasonal basis to water stress and other environmental parameters. Acknowledgments Gratitude is expressed to J. Ehleringer and J. Comstock (University of Utah) for their helpful comments and the use of their open gas exchange system. We also express appre- ciation to W. M. Hess (Brigham Young Uni- versity) for the SEM of the rabbitbrush leaf surface. This research was supported in part by a grant from Brigham Young University. Literature Cited Andersen, W R., G F. Wildner. and R. S. Griddle. 1970. Ribulose diphosphate carboxylase III. Al- tered forms of ribulose diphosphate carbo.xylase from mutant tomato plants. Arch. Biochem. and Biophys. 137:84-90. Comstock, J, and J Ehleringer. 1984. Photosynthetic responses to slowly decreasing leaf water poten- tials in Encelia frtiitescens. Oecologia 61:241-248. Ehleringer, J 198.3. Ecophysiology of Amaranthus palmeri, a sonoran desert summer annual. Oe- cologia 57:107-112. Ehleringer,]., AND O. BjORKMAN. 1978. A comparison of photosynthetic characteristics of Encelia species possessing glabrous and pubescent leaves. Plant Physiol. 62:185-190. Ehleringer, J, AND G S Cook 1980. Measurements of photosynthesis in the field: utility of the GOo de- pletion technique. Plant, Cell, and Environment 3:479-482. Ehleringer, J., and H. A. Mooney. 1978. Leaf hairs: effects on physiological activity and adaptive value to a desert shrub. Oecologia .37:183-200. 1983. Productivity of desert and mediterranean- chmate plants. Pages 20,5-231 inO. L. Lange, P. S Nobel, G. B. Osmond, and H. Ziegler, eds.. En- cyclopedia of plant physiology. Vol. 120. Springer- Verlag, Berlin. Ehleringer, J., H A Mooney, and J A. Berry. 1979. Photosynthesis and microclimate of Camissonia clavifonnis, a desert winter annual. Ecology 60:280-286. Gordon, K H J , M B People.s, and D. R Murray. 1978. Aging-linked changes in photosynthetic ca- pacity and in fraction 1 protein content of the first leaf of pea Pistivi sativum L. New Phytol. 81:,3,5-t2. Hess, W. M. 1966. Fixation and staining of fungus hyphae and host plant root tissues for electron mi- croscopy. Stain Technology 41:27-35. Koivuniemi, P M., N. E. Tolbert, and P. S. Carlson. 1980. Ribulose- 1,5-bisphosphate carboxylase/oxy- genase and polyphenol oxidase in the tobacco mu- tant su/SU and three green revertant plants. Plant Physiol. 65:828-833. Leopold, A. C, and P. E Kriedemann. 1975. Plant growth and development. McGraw-Hill, New York. Lichtenthaler, H. K., T. J Back, and A. R. Wellburn. 1982. Cytoplasmic and plastidic isoprenoid com- pounds of oat seedlings and their distinct labelling from '^C- nievalonate. Pages 489-500 in J. F. Win- ternans, G. M. Kuiper, and P. J. G. Kuiper, eds.. Biochemistry and metabolism of plant lipids. El- sevier Biomedical Press, Amsterdam. Lorimer, G. H , M R Badger. .andT J Andrews 1977. D-ribulose-L5-bisphosphate carbo.xylase-oxyge- nase: improved method for the activation and as- say of catalvtic activities. Anal. Biochem. 78:66-75. McArthur, E. D., a. C. Blauer, A. P Plummer, and R. Stevens. 1979. Characteristics and hybridization of important intermountain shrubs. III. Sunflower family. USDA Forest Service Research Paper lNT-220, U.S Dept. of Agriculture. Meyers, S P , S L Nichold, G R Baer, and W. T Molin 1982. Ploidy effect in isogenic populations of alfalfa. I. Ribulose- 1,5-bisphosphate carboxy- lase, soluble protein, chlorophyll and DNA in leaves. Plant Physiol. 70:1704-1709. Milthorpe, F. L., and J. Moorby. 1974. An introduction to crop physiology. Cambridge Press. Mooney, H. A., and J. R. Ehleringer. 1978. The carbon gain benefits of solar tracking in a desert annual. Plant Cell Environ. 1:307-311. Mooney, H A.J E Ehleringer, and J. A Berry. 1976. High photosynthetic capacity of a winter annual in Death Valley. Science 194:322-324. Ostler, W. K. 1980. Selection of acquisitions of Chnjsothamnus for rubber production. Native Plants Inc., Salt Lake City. NSF Report. Seeman, J R, and J A Berry 1981. Interspecific differ- ences in the kinetic properties of RuBP carboxy- lase protein. Carn. Inst. Yr. Bk. 81:78-83. Seeman, J R , J M Tepperman, and J. A Berry. 1980. The relationship between photosynthetic perfor- mance and levels and kinetic properties of RuBP carboxylase oxygenase from desert winter annu- als. Carn. Inst. Yr. Bk. 80:67-72. 526 Great Basin Naturalist Vol. 45, No. 3 Walter H and E Stadelmann. 1974. A new approach Wittenback, V. A , L C. Ackerson, R. T. Giaquinta^ and to the water relations of desert plants. In G. W R R. Hebert. 1980. Changes in photosynthesis, ed Desert biology. Vol. 2. Academic ribulose bisphosphate carboxylase, proteolytic ac- tivity, and ultrastructure of soyabean leaves dur- crown Press, New York ing senescence. Crop. Sci. 20:225-231. FOOD HABITS OF THE WESTERN WHIPTAIL LIZARD {CNEMIDOPHORUS TIGRIS) IN SOUTHEASTERN NEW MEXICO Troy L. Best' and A. L. Gennaro" Abstract. — This study presents the first food habit assessment for the western whiptail hzard {Cnemklophoriis ti^ris) in the shinnery oak-mesquite habitat {Quercus havardii-Prosnpis glandulosa) of southeastern New Mexico. Short-horned grasshoppers, termites, anthons, beetles, and spiders formed the major portion of the diet during the four-year stud\'. Discriminant analyses were used to evaluate annual, seasonal (monthly), and sexual variation. Incidental food categories were responsible for most of the annual and seasonal variation. Dominant foods varied little between months and years. Sexual variation was more evident; it may act to reduce intraspecific competition for food resources and may be associated with secondary sexual size dimorphism. Food habits of the western whiptail Hzard {Cnemidophorus tigris) have been studied in several areas of the western United States (e.g., Pack 1923, Milstead 1957a, 1958, 1961, 1965, Johnson 1966, Echternacht 1967, Medica 1967, Milstead and Tinkle 1969, Pi- anka 1970, Bickham and MacMahon 1972, Scuddav and Dixon 1973, Vitt and Ohmart 1977, Mitchell 1979, Best and Polechla 1983). Some of these investigators have examined intraspecific variation: e.g., Johnson (1966) found that the diet of immature whiptails was similar to that of adults, and Johnson (1966) and Pianka (1970) found little difference in diet between sexes. Conversely, there is con- siderable geographic (Milstead and Tinkle 1969, Pianka 1970), seasonal (Johnson 1966, Milstead and Tinkle 1969, Pianka 1970, Vitt and Ohmart 1977, Mitchell 1979), and annual variation (Milstead 1965, Medica 1967, Mil- stead and Tinkle 1969, Mitchell 1979). In southeastern New Mexico, Best and Polechla (1983) reported diet data for C. tigris in their study of C. gularis, but their sample of C. tigris was small and the habitat where they collected specimens was quite different from where those examined herein were obtained. Subsequently, Best and Gennaro (1984) stud- ied Uta stanshuriana from the shinnery oak-mesquite habitat of southeastern New Mexico using specimens collected in sympa- try with the C. tigris reported herein. In view of the previous studies of food habit variation and because no extensive studies of C. tigris have been conducted in the shinnery oak-mesquite association of southeastern New Mexico, the present study was initiated. Our objectives were to assemble a listing of food items consumed in that area and to exam- ine annual, seasonal (monthly), and sexual variation. Materials and Methods From 1976 through 1979, 174 C. tigris were collected approximately 40 km E of Carlsbad in Eddy and Lea counties, New Mexico (within an 8-km radius of drill hole ERDA 9, SE corner. Sec. 20, T22S, R31E). Specimens were fixed in 10% formalin and stored in 40% isopropyl alcohol. Stomach contents were later removed, placed into individual vials, and identified. Arthropod taxonomy follows Borroretal. (1981). Two separate data sets were used in the analyses. One contained the number of indi- viduals in each arthropod order. The second included the number of individuals identified to family except where identification was im- possible (e.g., unidentified Coleoptera were entered as Coleoptera, Buprestidae was an- other character, Cleridae another, etc.). Dis- criminant analyses (Nie et al. 1975) were used to test for annual, seasonal (monthly), and sexual variation in food habits. Best and Gen- naro (1984) presented a summary of this tech- 'Ceneral College. Department of Biology, and Museum of Southwestern Biology, University of New Mexico, Albuquerque, New Mexico 87131. -Llano Estacado Center for Advanced Professional Studies and Research, Natural History Museum, Eastern New Mexico University, Portales, New Mexico S8130. 527 528 Great Basin Naturalist Vol. 45, No. 3 Table 1. Food items found in 174 western whiptail lizards {Cnemidophorus tigris). Sample sizes are given in parentheses and occurrence data are presented as: number of lizards containing a food category; total items observed. Food category 1976(10) 1977(42) 1978(69) 1979(53) Combined(174) Arthropoda 1;1 1;1 Arachnida 1;1 1;1 Scorpionida 1;1 1;1 7;7 5;5 14;14 Araneae 5;8 13; 18 13; 17 31;43 Solifugae 8;9 3;3 11;12 Chilopoda 1;1 1;1 He.xapoda 4;5 4;5 Orthoptera (9)' 7;9 32;61 61;146 46;82 146;298 Isoptera (3) 14;299 19;509 23;487 56; 1295 Psocoptera (1) 2;2 2;2 Hemiptera (5) 1;1 13;18 2;2 5;5 21;26 Homoptera (5) 2;3 6;14 5; 10 1;1 14;28 Neuroptera (2) 1;2 , 15;22 19;29 4;5 39;58 Coleoptera (9) 4;4 23;62 38;85 13;20 78;171 Lepidoptera (4) 1:1 13;29 20;48 15;22 49;100 Diptera (2) 2;4 7;8 2;2 11;14 Hymenoptera (4) 1;1 7; 156 16;24 16;22 40;203 Miscellaneous Insect eggs 1;18 1;18 Lizards 2;2 2;2 Sand 2;2 2;2 Empty stomachs 1;1 1;1 'Minimum number of families represented. nique in relation to lizard feeding ecology studies. Analyses were conducted using the IBM computer system at University of New Mexico. Specimens and their stomach con- tents were deposited in the Eastern New Mexico University Natural History Museum in Portales. Lizards were collected in a shinnery oak-mesquite association (Quercus havardii- Prosopis glamhdosa). Although shinnery oak and mesquite did not have the greatest plant density on our study area, they were among the most obvious plant taxa. Extensive vege- tation analyses of this noncultivated region are given in Best and Jackson (1982). Results Food items found in the stomachs of the 174 C. tU^ris are presented in Table 1. In addition, one platyhelminth was found in July 1978, and 12 nematodes were found in seven 1978 speci- mens. Arthropods represented by the highest frequencies of occurrence (number of speci- mens containing a food category/total number of specimens x 100) were: Orthoptera, 84%; (loleoptera, 45%; Isoptera, 32%; and Lepi- doptera, 28%. Except for the lack of isopter- ans in 1976, these categories were repre- sented in each of the four annual samples. Other consistently occiuring arthropods were Scorpionida, Hemiptera, Homoptera, Neu- roptera, and Hymenoptera. Although the Or- thoptera represented at least nine families, there were about 10 times as many Acrididae as any of the others. Their frequency of occur- rence was 78%, with 242 acridids being counted. The Coleoptera were more evenly distributed among the nine families identi- fied, but Scarabaeidae, Tenebrionidae, Elate- ridae, and Curculionidae had 11%-17% fre- quencies of occurrence, respectively. Other coleopteran families occurred at frequencies less than three percent. In addition, there was an 8% frequency of unidentified coleopterans (18 beetles). There were at least three families of Isoptera; Termitidae was the most common and occurred at a frecjuencN' of 4%. Half the lepidopterans were unideritilied, and most of the remainder were Geometridae (14% frequency). The results of the discriminant anaKsis be- tween years (sexes combined), using the num- ber of arthropods in each order as characters, are shown in Table 2. Onl\' 47% of the C. tigris were classified correctly, indicating little dif- ference between years. The analyses using all arthropod taxa (orders, suborders, super families, families) classified 69% of the speci- mens corr(>ctl\-. In decreasing order of impor- July 1985 Best, Gennaro: New Mexico Lizard 529 Table 2. Discriminant analyses between years and months for Cnemidophorus tigris. Actual group n Predicted group membership Years (1976-1979)'^ 1976 1977 1978 1979 1976 10 7(70.0%) 2(20.0%) 1(10.0%) 0 1977 42 11(26.2%) 21(50.0%) 7(16.7%) 3( 7.1%) 1978 69 19(27.5%) 7(10.1%) 32(46.4%) 11(15.9%) 1979 53 19(35.8%) 3( 5.7%) 9(17.0%) 22(41.5%) Months (1977-1979f May June July August May 31 13(41.9%) 8(25.8%) 10(32.3%) 0 June 60 15(25.0%) 16(26.7%) 25(41.7%) 4( 6.7%) July 59 6(10.2%) 3( 5.1%) 48(81.4%) 2( 3.4%) August 14 0 1( 7.1%) 6(42.9%) 7(50.0%) Months— 1977-' May June July May 9 5(55.6%) 2(22.2%) 2(22.2%) June 14 1( 7.1%) 9(64.3%) 4(28.6%) July 19 2(10.5%) 0 17(89.5%) Months— 1978' May June July August May 14 4(28.6%) 4(28.6%) 6(42,9%) 0 June 26 4(15.4%) 14(53.8%) 7(26.9%) 1( 3.8%) July 20 1( 5.0%) 1( 5.0%) 17(85.0%) 1( 5.0%) August 9 0 0 4(44.4%) 5(55.6%) Months— 1979** May June July August May 8 3(37.5%) 0 5(62.3%) 0 June 20 2(10.0%) 5(25.0%) 13(65.0%) 0 July 20 0 0 20(100%) 0 August 5 0 0 3(60.0%) 2(40.0%) 'The data in subsequent footnotes are given as: percent of the specimens that were correctly classified; in decreasing order of importance, the variables accounting for most or all of the differences. "47.1%; Hemiptera, Coleoptera, Orthoptera, Homoptera, Neuroptera, Araneae, Solifugae, Chilopoda, He,\apoda, and lizards. ^.51.2%; lizards, Lepidoptera, Isoptera, Scorpionida, Coleoptera, Psocoptera, Diptera, Orthoptera, Hemiptera, and Araneae. ^73.8%; Lepidoptera, Coleoptera, Hemiptera, Homoptera, and Scorpionida. ^58.0%; Diptera, Hexapoda, Coleoptera, Psocoptera, Araneae, Lepidoptera, Scorpionida, and Isoptera. ^56.6%; lizards, Orthoptera, Scorpionida, and Solifugae. tance, the variables separating years were Gryllacrididae, Acanalonidae, Coleoptera, Coreidae, Hemiptera, Cydnidae, Curculion- idae, Orthoptera, and Lepidoptera. This anal- ysis indicated there was some variation be- tween years. However, most of the variability was in food categories that were incidental (occurred at very low frequencies) or that were abundant during only one or two of the four years. The major food categories oc- curred every year, but fluctuated in fre- quency and total number of items observed. Discriminant analysis was performed be- tween months (May through August) combining sexes and data for 1977 through 1979 (Table 2); 51% of the specimens were classified correctly. Using all arthropod taxa, 64% were classified correctly. Variables con- tributing the most to the classification were lizards (juvenile Phrynosoma), Termitidae, Scorpionida, Coleoptera, Asilidae, Pso- coptera, and Elateridae. These analyses showed some variation between months, but most of the variability was associated with incidentally occurring food categories. When months were considered for individ- ual years, discriminant analyses using arthropod orders as characters showed greater differences between months than when data were combined (Table 2). Analyses using all arthropod taxa correctly classified 88, 75, and 72% of specimens to month for 1977, 1978, and 1979, respectively. For 1977 vari- ables accounting for the most differences were Hymenoptera, Sphingidae, Cicadellidae, Lepidoptera, Acrididae, Coleoptera, and Ela- teridae. For 1978 Mantidae, Curculionidae, Isoptera, Gryllacrididae, Elateridae, Pso- coptera, and Gryllidae accounted for the most differences. For 1979 variables accounting for the most differences between months were lizards, Acrididae, Pentatomidae, Geometri- dae, Blattoidea, Lepidoptera, and Tenebrion- idae. For each of the three years, the separa- 530 Great Basin Naturalist Vol. 45, No. 3 Table 3. Discriminant analyses between sexes for Cnemidophorus tigri.s. Actual group n Predicted group membership Male Female 1978-19791-' Male 64 51(79.7%) 13(20.3%) Female 58 22(.37.9%) 36(62.1%) 1978^ Male 37 34(91.9%) 3( 8.1%) Female 32 11(34.4%) 21(65.6%) 1979' Male 27 20(74. 1%) 7(25.9%) Female 26 6(23.1%) 20(76.9%) May 1978' Male 9 9(100%) 0 Female 5 1(20,0%) 4(80.0%) June 1978*^ Male 13 12(92.3%) 1( 7.7%) Female 13 2(15.4%) 11(84.6%) July 1978' Male 11 10(90.9%) 1( 9.1%) Female 9 2(22.2%) 7(77.8%) August 1978'* Male 4 4(100%) 0 Female 5 0 5(100%) May 1979*^ Male 5 5(100%) 0 Female 3 0 3(100%) June 1979'° Male 12 12(100%) 0 Female 8 0 8(100%) July 1979" Male 6 4(66.7%) 2(33.3%) Female 14 2(14.3%) 12(85.7%) 'The data in subsequent footnotes are given as: percent of the specimens that were correctly classified, in decreasing order of importance, the variables accounting for most or all of the differences. ^1.3%; Neuroptera, Hemiptera. Orthoptera, Coleoptera, Isoptera, Solifugae, Psocoptera. Araneae, Scorpionida. and lizards. 79.7%; Neuroptera, Solifugae, Orthoptera, Psocoptera, Isoptera, and Hexapoda. ^75.5%; Hemiptera, Lepidoptera, Orthoptera, sand, Solifugae, Coleoptera, Araneae, and Isoptera. ^92.9%; Neuroptera, Solifugae, Homoptera. and Coleoptera. 88.5%; Neuroptera, Isoptera, Coleoptera, Solifugae, Orthoptera, Hemiptera, Diptera, Hexapoda, and Scorpionida. ^85.0%; Neuroptera, Hemiptera. Solifugae, Homoptera, and Araneae. 100%; Coleoptera, Isoptera, Homoptera, Orthoptera, and Neuroptera. 100%; Scorpionida, Araneae, and Coleoptera. '"100%; Hymenoptera, Isoptera, Neuroptera, Hemiptera, Lepidoptera, Coleoptera, insect eggs, Scorpionida, Diptera, and sand. 80.0%, Araneae, Hymenoptera, and ."Vrthropoda. tion of months was primarily based upon inci- dental occurrences. Thus, monthly-seasonal variation was evident but was mostly reflected by incidentally occurring food categories. This was the same type of variation observed between years. The sex of each lizard was determined in 1978 and 1979, and discriminant analyses were used to assess sexual variation for these years. When 1978 and 1979 males were com- bined and compared to females, there were differences between sexes (Table 3). Except for Hemiptera, Solifugae, and Psocoptera, food categories accounting for the most differ- ences were represented in iioth sexes (Table 4). Analysis using all arthropod ta.xa provided 79% correct classifications; Myrmeleontidae, Isoptera, Noctuidae, Hymenojitera, Solifu- gae, Psocoptera, Araneae, Scorpionida, and Rhopalidae contributed the most to the differ- ences. Of these, M\ rmelconlidae, Isoptera, and Araneae were the most consistently oc- curring food categories. Each year was then examined separately. For 1978, 80% of the lizards were classified correctly to sex (Table 3). Psocoptera and Hexapoda were the only variables accounting for differences that could be considered as incidental (Table 4). Ct)nsidering all arthropod taxa, 90% were classified correctly, and Myrmeleontidae, Solifugae, Tettigoni- idae, Hymenoptera, Cicadellidae, Pso- coptera, and Elateridae accounted for the dif- ferences. Of these, Myrmeleontidae, Solifugae, and Tettigoniidae were the only consistently occurring food categories. For 1979, 76% of the specimens were classified correctly (Table 3). Except for Hemiptera, sand, and Solifugae, the variables accounting for the differences represented some of the most consistently occurring food categories (Table 4). Eight\ -thixn^ percent of the lizards July 1985 Best, Gennaro: New Mexico Lizard 531 Table 4. Food items in Cnemidophorus tigris collected during 1978 and 1979, See Table 1 for data presentation format. Food category Year and month of collection 1978 May (14) Total (691 June 26) July (20) August (9) S 5(9) ? ?(5) <5 • '1 V- - «hS'** Figs. 2-7. View,s of the beach between Niget Island ;mcl Hlaek Point. Mono Lake, CaUiornia: 2-3. Annnal zone; 4-5, Transition zone showing long rhizomes of" salt grass; (>-7, Salt grass zone. July 1985 Brotherson, Rushforth: Beach Stabilization 545 suggesting that vegetation differences are due to invasional phenomena rather than soil factors. Literature Cited Daubenmire, R. 1959. A canopy coverage method of veg- etational analysis. Northwest Sci. 33:43—66. Hansen. D. S, P. Dayanandan, P B. Kaufman, and J D Brotherson. 1976. Ecological adaptations of salt marsh grass. Disticliilif spicata (Gramineae), and environmental factors affecting its growth and dis- tribution. Amer. Jour. Bot. 63(5): 63.5-650. MuNZ, P A 1959. A California flora. University of Califor- nia Press, Berkeley. 1681 pp. Young, G. 1981. The troubled waters of Mono Lake. Nat. Geog. 160(4):504-519. GRASS SPIDER MICROHABITAT USE IN ORGAN PIPE CACTUS NATIONAL MONUMENT, ARIZONA Mark Roliert Deutschinan' Abstract. — The grass spider {Ag,elena naevia), commonly found in Organ Pipe Cactus National Monument, Arizona, uses rodent burrows located under a shrub canopy more frequently for web construction than burrows located in the open. The average number of prey available in canopy microhabitat was greater than in open microhabitat, and unequal prey abundance may explain spider microhabitat use. Fixed-web foragers must assess prey abun- dance when selecting a web site (Riechart 1979). Locomotion, silk production, and res- piration while waiting for prey all require energy (Ford 1977). When selecting a web site, spiders might choose a microhabitat that maximizes prey availability. TurnbuU (1964) reported that Achaearanea tepidariorum (Koch), a web-building spider, used prey availability as an index to determine web loca- tion. Webs were placed where wind currents maximized prey availability and minimized web damage. Horton and Wise (1983) found web location in two species of orb-web-build- ing spiders to be affected by the degree of environmental stress. Turnbull assumed (1964) that solar radiation and wind velocity influence web location. The grass spider (Agelena naevia), common to the Sonoran Desert, constructs webs in the openings of rodent burrows. Therefore, mi- crohabitat use may be a consequence of bur- row location. Because prey capture should be maximized, microhabitat preference may also be determined by prey abundance. In this paper, I seek to determine whether or not the distribution oiAgelena naevia is independent of Imrrow location and whether food availabil- ity may be a possible explanation for the pref- erential use of canopy microhabitat. Material AND Mkthods This study was conducted on the desert flats of Organ Pipe Cactus National Moinunent, Arizona, in late March 1982. Hie number of rodent burrows, with and without webs built in the burrow opening, were counted in a 3- by 50-m transect in two microhabitats. Bur- rows were located in canopy microhabitat if below the downward projection of a bush canopy (normally Larrea tridentata or Am- brosia deltoidea), otherwise, burrows were in open microhabitat. All animal burrows were considered available for spider occupancy, and I made no attempt to distinguish if rodent burrows were currently being used. Twelve plastic boards (10 cm") covered with Tanglefoot were used to assess insect availability. Twelve boards were placed in each microhabitat on each of two successive days. Boards in canopy microhabitat were randomly placed either north, south, east, or west of the bush under the edge of the canopy. Boards in open microhabitat were arbitrarily placed at least 2 m from a bush canopy. Spider body length (front of head to tip of abdomen) was also measured in each micro- habitat using a vernier caliper while randomly searching for webs. Results Spider distribution was related to burrow location (X" ^ 5.37, p = .02). Spiders occu- pied 33.4% of the burrows in the canopy mi- crohabitat and 4.8% of the burrows in the open microhabitat. The number of pre\- were also different be- tween microhabitats (ANO\'A, F 8.79, p < .01); an average of 1. 16 ± 0.9 insects/day were caught in the open microhabitat, and 2.42 ± 'Department of Zoology, Uiiivcrsilv of Moiil Hazardous Waste Management and Special Stinl 5980:! c. Roc 546 July 1985 Deutschman: Arizona Grass Spidkr 547 1. 1 insects/day were caught in the canopy nii- crohabitat. Significantly larger spiders occurred in the canopy microhabitat (F = 4. 14, p < .05). The mean spider body length was 0.66 ± 0. 19 cm (n = 22) in the open microhabitat and 0.76 ± 0. 19 cm (n ^ 29) in the canopy microhabitat. Discussion Greater food abundance may explain the preferential use of canopy microhabitat by Agelena naevio. However, other hypotheses include: (1) lower environmental stress in the canopy microhabitat, (2) more suitable strata for web construction in the canopy microhabi- tat, (3) rodent burrows may not be equally available as sites for web construction in canopy and open microhabitats. These hy- potheses are discussed below. Although Castillo and Eberhard (1983) re- ported that artificial webs were inaccurate in assessing the exact species composition of prey captured by webs, they do conclude that artificial webs are effective in comparing dif- ferent properties of the environment (e.g., relative insect abundance). Trapping with sticky plastic boards indicated a greater num- ber of potential prey in the canopy microhabi- tat. If microhabitat use was based solely on prey abundance, burrows located in canopy microhabitat would be used more often. In- creased prey consumption may allow greater growth and reproductive success (Calow 1981). Less severe environmental conditions may characterize canopy microhabitat. A dimin- ishing of the intense solar radiation of summer should be beneficial in maintaining body tem- perature at an optimal level. Shrub branches and litter may also provide better physical strata for web construction, resulting in less web destruction and energy for web repair. Eisner and Nowicki (1983) suggested that web destruction resulted not only in the loss of time spent in web repair, but in the loss of valuable proteinaceous silk. Spiders may choose to establish webs only in inactive (or active?) rodent burrows. If true, then spider residency in a microhabitat is a consequence of the distribution of rodents and the location of inactive rodent burrows. I assumed all burrows were available for spider use and made no distinction with respect to the degree of rodent activity. At the time of spring hatching, spiders may be seeking burrows. If burrows were limited and canopy microhabitat preferred, competi- tion for web sites might occur. The difference in average spider body length may be evi- dence of intraspecific competition (Schoener 1974). Acknowledgments This research was conducted under the guidance of R. HuttoandJ. McAuliffie while I was a student at the University of Montana. I thank them for their assistance and support during the project. I also thank R. Hutto, G. Allen, R. Nelson, and D. Carter for reviewing the manuscript. Literature Cited Calow, P 1981. Invertebrate biology: a functional ap- proach. Groom Helm London. 183 pp. Castillo. J. A., and W. G. Eberhard. 1983. Use of artifi- cial webs to determine prey available to orb weav- ing spiders. Ecology 64:1655-1658. Eisner. T.. and S. Nowicki 1983. Spider web protection through visual advertisement: role of stabilimen. Science 219:185-187. Ford, M. J. 1977. Energy costs of the predation strategy of the web spinning spider Lepthyphantes zim- mennanni. Oecologia 28:.341-349. HORTON, C C , and D H. Wise 1983. The experimental analysis of competition between two syntopic spe- cies of orb-web spiders (Araneae: Araneidae). Ecology 64:929-944. Reichert, S. E 1979. Games spiders play II. Resource assessment strategies. Behav. Ecol. and Soc. Biol. 6:121-128. Schoener, T. W 1974. Resource partitioning in ecologi- cal communities. Science 185:27-39. Turnbull, a. L. 1964. The search for prey by a web building slider Achaearanea tepidariorum (C. L. Koch) (Aranea, Theridiidae). Canadian Ent. 96:568-579. NEW SPECIES OF PRIMULA (PRIMULACEAE) FROM UTAH Ronald J. Kass' and Stanley L. Welsh' Abstract. — Named as a new species from the House Range of Millard County, Utah, as Primula domcnsis Kass & Welsh. The House Range in west central Millard County, Utah, is famous for its massive lime- stone and dolomite sequence, which exposes cliffs with a relief of more than 1280 m on its western flank. Drainage plunges steeply on the western side, more gently so on the east- ern side of the range. The foot slopes are clothed sparsely by mi.xed desert shrub vege- tation, with pinyon-juniper woodland becom- ing important at about the 2135 m contour. The range is arid, despite its high elevation, and has only a few truly mesic sites. Collections of plants have been taken from the House Range for almost a century, but no intensive study had been undertaken until the present time. Beginning in 1981 a study of the flora of the House Range was undertaken by the senior author (Kass 1983) as partial fulfill- ment of the requirements for completion of a master of science degree at Brigham Young University. Those collections, consisting of some 366 species, were routine for the most part, yielding few surprises. An exception among the specimens taken is a dwarf prim- rose species, whose characteristics, in combi- nation, are unique among the primroses of the west. Except for the rather widely distributed Primula parryi Gray and P. specuicola Rydb. , the only other primulas known from Utah are rare and restricted (Welsh 1985). Primula in- cami Jones is known from Daggett County and, until 1982, only from its historic type locality in western Garfield County, and P. maguirei L. O. Williams is known only from Logan Canyon in Cache County. During 1982 P. incana was rediscovered in Garfield County by E. S. Nixon and in 1984 by Sherel Goodrich, presumably neither locality far re- moved from where it was initially discovered by Marcus E. Jones in 1894. Thus, the discov- ery of a primrose in the House Range was unexpected. However, P. nevadensis N. Holmgren is known from the Mt. Washington area of the Snake Range in White Pine County, Nevada, only 80 km distant from the newly discovered population of primrose growing in the House Range. Relationships of the House Range primrose apparently lie most closely with P. maguirei and P. nevaden- sis and more remotely with P. cusickiana Gray, a plant of eastern Oregon and Idaho. Primula domensis Kass & Welsh, sp. nov. Species habitu cum Primula maguirei L. O. Williams sed in coroUae tubis calyce sub 1.5 longioribus (nee 1.5-2) et corollae lobis la- tioribus (4-12 nee 4-5 mm) et foliis plus den- tatibus diifert; ab Primula nevadensis N. Holmgren in inflorescentia foliis superans dif- fert; similis Primula cusickiana Gray in habitu calyce et corollae tubis ad calyce statu ra sed in foliis plus dentibus et majoribus et plantis majoribus differt. Type — USA: Utah: Millard Co., Sawtooth Canyon, House Range, T19S, R14W, S23, 2590 m, limestone cliff faces, in Cercocarpus intricatus-Ephcdra viridis community, 4 June 1982, R. and J. Kass 884 (Holotype BRY; isotypesNY, RM, POM). Plants 7-15 cm tall, from a short, rhizoma- tous caudex, this clothed with persistent, brown leaf bases; leaves 2-8 (11) cm long, 5-22 mm wide, oblanceolate to spatulate, dentate to subentire, tapering to a broad peti- ole, green and more or less glandular on both sides; bracts usually 3, 1.5-10 mm long, lance- olate, not swollen at the base, glabrous or mealy; peduncle apex glabrous or somewhat nd Department of Bo 548 July 1985 Kass, WELSH: A Utah Primula 549 Fig. 1. Primula domensis Kass & Welsh. A, Habit. B, Detail of flower. 550 Great Basin Natur.'VLIst Vol. 45, No. 3 mealy; umbels 1- to 5-flowered, the pedicels 5-22 mm long; calyx 8-12 mm long, mealy or glabrous, the teeth shorter than the tube, more or less accrescent in fruit; corolla rose to lavender, the tube surpassing the calyx, but less than twice its length, the limb 12-25 mm wide, the lobes shorter than the tube, 4—12 mm wide; capsule to 8 mm long, not surpass- ing the calyx. Additional specimens. — Utah: Millard Co. , House Range, Notch Peak, T19S, R14W, S23, moist limestone cliffs at 2623 m, 22 May 1981, R. Kass 289 (BRY); Ibid., T19S, R14W, S22, limestone cliffs at 2745 m, 27 June 1981, R. Kass & S. White 473 (BRY; UT, UTC). The House Range primrose is most closely similar to P. nevadensis and P. rnaguirei, be- tween whose ranges it occurs. It differs horn the former in the inflorescence to leaf propor- tions (leaves are overtopped by the inflores- cence) and leaf shape (leaves spatulate to oblanceolate, not cuneate) and from the latter in the corolla tube to calyx proportions, mostly broader corolla lobes, and more con- sistently toothed leaves. It seems probable that each of these geographically and geneti- cally isolated microspecies would be treated best as portions of an expanded P. cusickiana, the first named of the group. However, no such combination is intended or implied herein. The specific epithet is taken from the word "house" {do77ius in Latin) from the House Range. References Kass, R J 1983. A floristic study of the House Range, Millard County, Utah. Unpublished thesis. Brigham Young University, Provo, Utah. 75 pp. Welsh, S. L 1985. Utah flora: Primulaceae. Unpublished manuscript. Brigham Young University, Provo, Utah. 8 pp. NEW SPECIES OF ASTRAGALUS (LEGUMINOSAE) FROM SOUTHEASTERN UTAH Rupert C. Harnel)\' and Stanley L. Welsir Abstract. — Named and deseribed is Astr(i^,alu.s piscator Barnehy & Welsh, a species of sect. Argophylli subsect. Missourienses. The species occurs in Grand and San Juan counties, Utah. For several years the authors have been aware of an undescribed species oi Astragalus masquerading within the specimens desig- nated as A. amphioxijs 'And A. cymboides from Grand and San Juan counties in southeastern Utah. The plants begin flowering in late March and are in fiuit by mid-May. Few col- lections represent flowering material, possi- bly due to the earliness of anthesis. Relation- ship of flowering to fruiting materials has only recently become apparent. The species is de- scribed as follows: Astragalus (sect. ArgophijUi subsect. Missourienses) ptsca a species occurred on divided by the total number of bums. Table 3. Occurrence (percent of relev^s a species occurs on) and cover (percent of the relev^s a species occurs on that has ^ 5% cover) of species' on five successional stages. Occurrence Cover^ Successional stages Successional stages Early Late Early Late Early mid Mid mid Late Early mid Mid mid Late Tree species Juniperus osteosperma (f (f 2' 27" 56" 0" 0" 0" 0" 100" Finns monophylla C .5" 5'- 80" 100" 0" 0" 0" 5" 100" Shrub species Artemisia tridentata 9' 74b 79" 98" 88"" 0'' 15' 36" 88" 30" Chrysotliamnus nauseosus 2'' 66" 63" 65" 8" 0" 10" 49" 32" 0" Chrysothamnus viscidiflorus 37'- M"" 75" 44bc 22'' 6' 21" 33" O' 0* Ephedra viridis It 52" 68" 60" 51" — — — — — Prunus andersonii I' 35'' 67" 31" 13' — — — — — Purshia tridentata 28' 60'' 61" 83" 73" C 3' 15' 45" 24" Ribes velutinum 25" 72" 48" 59"" 56" Sambucus cerulea 45^ 28'' .30" ft3" 4' — Symplwricarpos oreophilus 39" 27"" 22"" 38" 20" gb. 47" 18"" 0^ 29" Tetradymia canescens S' 37" 10" 4" 13" 0" 22" 0" 0" 0" Grasses Bromus tectorum 42* 86" 61" 60" 49"' 13' 76" 71" 48" 7' Festuca idahoensis^ 44" 19" 0" 33" Oryzopsis hymenoides 12'- 22'' 56" 16"* 8'' — — — Foa fendleriana 4' r ^ 31" 17"" — — — — — Foa secunda 75" 19' 39" 49" 63" — — — — — Sitanion hystrix 43' 58" 49"- 90^ 44' 3" 15" 28" 0" 0" Stipa thurberiana 33*> 12* 37"" 57" 34" 0'" 9" 23" 21" 3"^ Perennial forbs Arabis holboeUii 19" 9^ 26" 26" 50* — — — — — 560 Great Basin Naturalist Vol. 45, No. 3 Table 3. continued. Occurrence Cover" Successional stages Successional stages Early Late Early Late Early mid Mid mid Late Early mid Mid mid Late Argemone munita 44'*' 56' 26'^ 3l'' 0^ 2*' S-" O'' O'' O'' Astragalus purshii 14' 6' 33' 14' S' — — — — — Balsamorhiza sagittata 21' 29'' 20''' ll'' 19''' _____ Chaenactis douglasii S"* 7'' 16'' 13' 7' — — — — — Crepis acuminata 40" 24'' 29'' 36''' 13' 5"'' 6"'' 16'' O'' O'' Enogunum umheUatum 5"'' 4'' U' 5''' 10' — — — — — Lupinus caudatus 46'' 20' 32''' 25''' 19'" 9''' 7''' ()'' 13'' O'' Lygodcsmia spinosa 31' 32' 19''' 27' 13'' — — — — — Machaeranthei'a canescens 18' 15' 19'' 28' 6 ' — — — — — Phacelia hastata 4T 36' 48" 15'' 'J _____ Phlox hoodii 66" 6^ 45'' 0' 32'' _____ Phlox spp. 26" 13^ 23"'' 31" 12'' _____ Annual forbs Collinsia parvijlora 56" I'' 0'' 6' 27'' _____ Cryptantha spp. 56" 12'' O'' O'' U'' 11" O'' O'' O'' O'' Dcscurainia spp. 37" 7' 3" 0' 23" — — — — — Eriogonum spp. V 33'' 8"'' 0' 6*' — — — — — E radium cicutarium O'' 25" 25" 20" O'' O'' 3'' O'' 60" O'' Gayophytum spp. 57" 10' 1 1 " 29 6' — — — — — Gilia spp. 65" 4' 2' 46"'' 34'' 58" O'' 0^ O'' O'' Lactuca serriola 25" 7^' lO'' O'' O'' — — — — — Mentzelia albicaulis 32" 35" 0" 8'' 2'" _____ Nicotiana attenuata 38" 4*' O'' O'' O'' — — — — — Sisymbrium alti.ssimu7n t 46" 35" 12'' 4'' O'' 35" 21'' O'' O'' Species that occurred on S 20% of the total releves or ^ 20% of the releves in any one of the categories examined. ^Only species that occurred on s= 5% of the actual releves the species occurs on with s= .5% cover were examined. ■ Chi square test including all successional stages was not valid; therefore. onl\ earlv and late stages were compared. year-old burns (Table 3). Tree species were essentially eliminated by burning. Shrubs such as big sagebrush and bitterbrush, which regenerate by seed, were easily killed by burning but began to reoccupy some burned sites within one year. Reestablishment of big sagebrush appeared to occur through migra- tion and germination of seed from adjacent unburned areas or germination of on-site soil seed reserves. Bitterbrush reestablished pri- marily by germination of rodent caches (pers. obs.). Ecotypes of bitterbrush that can re- sprout were not found on the study sites. Both species had negligible cover on one-year-old burns. Although some individual plants were killed, all root-sprouting species studied re- mained part of the postburn plant commu- nity. This is consistent with other studies (Wright et al. 1979). Occurrence of rubb(>r rabbitbrush and horsebrush (Tetradijniia canescens) was not significantly different be- tween mature woodland stands and one-year- old l)urns. One-year-old wildfire sites had sig- nificanth higher occurrences of blueberr\ el- der (Sanibucus cerida ), mountain snowberry {Symplioricarpos orepliihts), and low rabbit- brush than did mature woodlands and signifi- cantly lower occurrences of Mormon tea (Ephedra liridis), desert gooseberry, and An- derson peachbriish [Prunus andersonii). Only mountain snowberry and low rabbitbrush oc- cinred on early postburn sites with co\er val- ues ^ 5%. Even though occurrence of moun- tain snowberry was significantly greater on the one-year-old burns than on the woodland stands, the percent of these releves with cover ^ 5% was significantly less on one-year-old burns. The differences were not significant between the two successional stages lor low rabbitbrush. Hoot crowns oi most perennial grasses can survive wildfire, continuing growth when conditions are favorable (Wright et al. 1979, White and Currie 1983). Comparison of one- year-old burns and mature woodlands showed that onK' the occurrence of one species, mut- ton bhiegrass (Poa frudleriana). was signifi- July 1985 KoNiAK: Ecological Succession 561 cantly reduced. Occurrence of other peren- nial grasses remained the same or increased slightly. For most perennial grasses, high cover values (^ 5%) were not found on either one-year-old burns or mature woodlands. Idaho fescue {Festuca idahoensis) occurred on the highest number of releves, all on mature woodlands, with cover ^ 5%. Cheatgrass, an annual grass, was killed by wildfire but re- gained postburn levels of occurrence and cover within one year from germination ot either soil seed reserves or seed from adjacent areas (Young and others 1976, Merrill and others 1980). Most perennial forbs occurred more fre- quently on one-year-old burns than on mature woodlands. Only Holboell rockcress {Arabis holhoeUii) had significantly lower occurrence on the early successional stage than on the late stage. Few releves had cover values ^ 5% for any perennial forb species. No significant dif- ferences occurred between cover values of the two stages, although cover tended to be higher on one-year-old burns. Annual forbs are killed by burning but rapidly reoccupy most sites by germination of seed from adjacent unburned areas or on site soil seed reserves (Koniak and Everett 1982). Occurrence of most annual forbs was signifi- cantly greater on one-year-old burns than on mature woodlands. Only gilia (Gilia spp.) and cryptantha (Cnjptantha spp.) occurred with cover ^ 5% and only on the one-year-old burns. Occurrence and cover of understory spe- cies in pinyon-juniper woodlands generally remained the same or increased after burn- ing. High postfire occurrence or cover of perennial species may be attributed to a com- bination of high tolerances to burning, in- creased visibility of plants from either in- creased size or elimination of litter or duff, and increased germination, survival, and growth following release from competition from late successional species. High postfire occurrence or cover of annual species may result from increased germination and growth of on-site seed or seeds from adjacent un- burned sites. Several shrub species and Holboell rock- cress decreased substantially after fire. The decrease may be attributed to a high suscepti- bility to fire-related injury and death, to the inability to compete with early successional species or, for woody species, to the slow accumulation of bioniass. Wildlife grazing may also contribute to the decline in a number of browse species. Species Variation over Five Successional Stages Annual forbs displayed the greatest varia- tion in distribution patterns among succes- sional stages (Table 3). Some species occurred most frequently on early successional sites, others occurred most frequently on midsuc- cessional sites, and still others had higher oc- currence on early and late sites than on mid- successional sites. Most perennial forbs were found on a disproportionately higher number of early and midsuccessional sites than on late sites. Hood's phlox (Phlox hoodii) had signifi- cantly greater occurrence on the early succes- sional stage than on later stages, but occur- rence was also high during midsuccession and late succession. Holboell rockcress was the only nontree species that occurred more fre- quently on late successional sites than at other stages. Perennial and annual grasses were consis- tently found on more mid-successional sites than on early or late sites. A notable exception was Sandberg bluegrass, whose high occur- rence on early and late successional sites par- alleled the distribution patterns of several an- nual forbs. All shrubs, except blueberry elder and mountain snowberry, had relatively low oc- currence on early successional sites compared to midsuccessional sites. Occurrence of all root sprouting shrubs, except desert goose- berry and Mormon tea, decreased between midsuccession and late succession. Big sage- brush and antelope bitterbrush, shrub species that regenerate by seed, maintained high oc- currence throughout midsuccession and late succession. Unlike occurrence, cover of an- telope bitterbrush and big sagebrush de- creased significantly from midsuccession to late succession. Apparently, as tree species dominated a site, the resources available to understory species rapidly diminished and cover was greatly reduced. Tree species began to reestablish 20 to 30 years after fire, but cover was minimal even 60 562 Great Basin Naturalist Table 4. Aspects on which species' occurred most frequently. Species North Aspect East South West Shrubs * Artemisia tridentata X Chrysothamnus nauseosus * Chrysothamnus viscidiflorus X Ephedra viridis * Prunus andersonii * Purshia tridentata X * Ribes velutinum X * Symphoricarpos oreophilus X * Tetradymia canescens X Grasses * Bromus tectorum * Festuca idahoensis X Oryzopsis hymenoides * Poa fendleriana Poa secunda Sitanion hystrix * Stipa thurberiana Perennial forbs Argemone munita X * Arabis holboellii X Astragalus purshii X * Balsamorhiza sagittata X * Crepis acuminata X Eriogonum elatum X * Eriogonum umbellatum X * Lupinus caudatus X Lygodesmia spinosa X * Machaeranthera canescens X * Phacelia hastata Phlox hoodii X Phlox spp. X Annual forbs Collinsia parviflora Cryptantha spp. X * Descurainia spp. Eriogonum spp. X * Erodium cicutarium Gayophytum spp. * Gilia spp. Lactuca serriola Mentzilia albicaulus Nicotiana attenuata X * Sisymbrium altissimum XX Species that occurred on * 20% of the relev^s in any one of the categories investigated (i.e., north, south, east, west) Occurrence of species with X's in all categories was not significantly different among aspects. •Differences are significant at P < .05. years after burning. Recent work (Tausch and West, in prep.) indicates that at this point trees rapidly increase in density and cover, dominating a site 100 to 150 years after burn- ing. Species with high occurrence in early suc- cession apparently survived wildfire as seeds on the burned site or on adjacent unburned sites, or as buds at the root crowns, or both. They were able to rapidly take advantage of the increased availability of water, nutrients, and light in the postfire environment. Lim- ited occurrence of species in midsuccessional or late stages may be related to several factors (Barbour et al. 1980). Seed may germinate most effectively after fire scarification. As July 1985 KoNiAK: Ecological Succession 563 shrubs aud later tree species overtop the low- growing herbaceous forbs, reducing the light intensity at the soil level, photosynthesis is reduced and shade-intolerant plants will not survive. Competition with midsuccessional or late species for water and nutrients may limit occurrence of some species. Plants prominent in later stages of succession may produce al- lelopathic chemicals that may inhibit growth of other plants. Increase in occurrence of perennial species in midsuccession may be correlated to the attainment of the critical biomass necessary for seed production or to the increase in ground cover providing a fa- vorable microsite for seed germination and plant survival. Reentry of tree species into the plant com- munity depends upon perennial nurse plants associated with midsuccessional and late stages (Everett and Ward 1984). Nontree spe- cies with high occurrence on late stages can successfully compete with other late species and may actually benefit from the combina- tion of shade and relative lack of ground cover. However, many of these species exhibit a sub- stantial reduction in cover with increasing tree dominance. Common factors may explain similar spe- cies occurrence on two seemingly dissimilar sites, one-year-old burns (early succession) and mature woodlands (late succession): an abundance of bare ground and a lack of com- petition for resources in the upper soil hori- zons. Species in this group appear to be toler- ant of late successional conditions (shade, tree competition, alleopathy) but cannot compete with species that have high cover in midsuc- cessional stages. A number of species, although often occur- ring more frequently in specific successional stages, exhibited high occurrence (5= 20%) in all stages. These species could be a major component of a plant community at any point in the successional cycle and include cheat- grass, bottlebrush squirreltail, Sandberg bluegrass, tailcup lupine, mountain snow- berry, low rabbitbrush, bitterbrush, and desert gooseberry. Species Variation with Aspect Species can be grouped according to differ- ences in occurrence on the various aspects. Many root-sprouting shrubs and perennial forbs typically had higher occurrence on north and east aspects (Table 4). Other species associated with east slopes included Indian ricegrass (Oryzopsis hymenoides), mutton bluegrass, Thurber needlegrass (Stipa thurberiana), and Anderson peachbrush. Holboell rockcress occurred most frequently on north aspects. Many annual forbs were more prominent on south and west slopes than on north and east slopes. Rubber rabbit- brush and cheatgrass exhibited significantly higher occurrence on west slopes than on north or east slopes. South slopes had medium occurrence of these species. Shrubs that regenerate by seed (i.e., big sagebrush and bitterbrush). Hood's phlox {Phlox hoodii), and horsebrush {Tetradymia canescens) tended to occur least frequently on south slopes and with equal occurrence on other slopes. Most other species, including the two most frequently occurring perennial grasses, Sandberg bluegrass and bottlebrush squir- reltail, had no significant differences among aspects. North and east slope conditions appear to favor the establishment of perennial species. These slopes generally have better moisture relations, less variation in temperature, and generally less harsh conditions than south and west slopes. Conditions limiting the occur- rence of annual species may range from reduc- tion of their soil seed reserves from long-term preburn competition by perennial species to the prolonging of the winter dormancy of many winter annuals (Baskin and Baskin 1981), by cooler temperatures and longer snow cover on north and east slopes. Other site factors appear to determine the establishment patterns on south and west as- pects. On these sites, large daily temperature and moisture differentials, especially in spring and early summer, can easily damage susceptible seedlings. Annual species appear to survive these fluctuations better than most perennial species (Evans and Young 1982, Young and Evans 1982). Species that showed no marked preference for aspect appear to germinate and survive equally well under a variety of environmental conditions. Variation of the Successional Cycle among Aspects Canopy cover by growth forms reflected and summarized individual species cover 564 Great Basin Naturalist Vol. 45, No. 3 Table 5. Percent of releves with cover values ^ 5% for five growth forms, two aspect groups, and five successional stages. North and east aspects South and west aspects successional stages Earlv Late Earlv Late Growth forms Early mid Mid mid Late Early mid Mid mid Late Shrubs .lis"' ,71'^ ,79'' ,97" ,49' , 0'- .53' ,76" ,90"'' ,47'' Annual forbs ,38" , 2'' , 3'' , ()'' .4'' ,6.5" ,35'' ,11' ..25'" . 2^* Perennial forbs ,.,35-' ,20'' „16'' , 0' V (?' .22"'' ,20"'' ,13'" , 5'"- X 3' Annual grasses V 0' ,,.56'' „24"' ,16''" •7' , 9'' ,80" .55'' .45^- .3^ Native perennial grasses ..15"^ ,19'' „26-'" ,,42" .25"'' , C' , 9"' ,32" ,1,5'"- .14"^ 'Percents in a column preceded by the same letters V , w. X, V, or 7. do not differ sign ficantlv at F <.05. ^Percents in a row followed by the same letters a, b c, d. e. or f do not differ significantly at P < 0.5. among successional stages and aspects (Table 5). Early succession on all slopes was domi- nated by perennial and annual forb cover, with higher perennial forb cover on north and east slopes and higher annual forb cover on south and west slopes. In addition, north and east slopes had high cover of shrubs and native perennial grasses. Early midsuccession brought a sharp increase on all aspects in the number of releves with shrub and annual grass cover ^ 5% and a sharp decrease in the same measure for annual forbs. Annual forbs remained an important component on south and west slopes throughout midsuccession, but not on north and east aspects. During all stages of midsuccession, annual grass cover was significantly greater on south and west slopes than on north and east aspects, and perennial grasses tended to be higher on north and east slopes. By late succession, the number of releves with perennial and annual forb and annual grass cover ^ 5% decreased on all aspects to less than 6%. Perennial grass and shrub cover also de- creased by late succession but still maintained relatively high cover on all aspects. North and east slopes tended to support higher cover of perennial grasses than did the south and west slopes, init the difference was not significant. Differences between aspect groups were even less for other growth forms. Different mois- ture, temperature, and light regimes on the different aspects appear to determine the spe- cies and subsecjuently the growth forms that can establish and survive, as well as the speed with which a species or vegetational group becomes prevalent or diminishes. This in turn controls the successional pattern. Affect of Seeding on Other Species Eight of the 21 burns were seeded. Domi- nant seeded species included crested wheat- grass (Agropyron desertomm), intermediate wheatgrass {Agropyron intermedium), and smooth brome {Bromiis inennis). Seven of the eight seeded burns were identified as early midsuccessional and midsuccessional stages. The eighth seeded burn, which was only one year old, will not be discussed because seeded species generally appeared only in trace amounts. No consistent pattern emerged when comparing occmrence of shrubs on seed and nonseeded burns (Table 6). Three species had higher occurrence on nonseeded burns, one had higher occurrence on seeded burns, and three exhibited no significant difference between the two groups. All nonseeded grass species occurred sig- nificantly less on seeded burns. This would be expected because most seeded species were grasses selected for their competitive ability. Cheatgrass, however, is a highly competitive grass whose decrease on seeded burns ap- pears to be limited to north and east slopes and high elevation sites, where seeded spe- cies have a competitive edge (Koniak 1983). Occmrence of perennial forbs was generally not affected by seeding. OnK two species, sulfiu- buckwheat [Eriogoniim umhcllatum) and Hoods phlox, occurred less frecjuently on seeded burns. These species are relatively woody and often classified as half-shrubs. Only five annual species occurred on enough midsuccessional sites to examine. Tumble mustard and alfileria occurred less frequently on seeded burns than on nonseeded. The other species showed no significant difference July 1985 KoNiAK: Ecx)L()GicAL Sucx:ession 565 Table 6. Effect of seeding on naturally occurring species. ' Species with significantly higher occurrence on non- seeded sites than on seeded sites Species with no significant differences in occurrence between seeded and nonseeded sites or with greater occurrence (*) on seeded sites Shrubs Artemisia tridentata Chnjsothamntis natiscosus Chnjsothamnus viscidiflorus Grasses Bromiis tectortim Onjzopsis hijmenoides Poafendh' liana Poa sandhergii Sitanion hystrix Stipa thurberiana Agropijron spicatum Forbs Eriogonum umlwHatum Phlox hoodii Sisymbrium altissimum Erodium cicutarium Shrubs Ephedra viridis Pursliia tridentata (*) Hihes ichilinnm Symphcnicarpos oreophiltts Grasses None Forbs Arabis holboellii Argemone munita (*) C repis acuminata Lupintis candatus Lygodcsniia spinosa (*) Machaeranthera canescens Phacelia hastata Astragalus ptirshii Chaenactis douglasii Phlox spp. (perennial species) Penstemon speciosus Gilia spp. (annual species) Mentzelia albicaulis Eriogonum spp. (annual species) Species that occurred on at least two seeded and two nonseeded bur between occurrence on seeded and non- seeded sites. Seeding appears to be detrimental to a number of naturally occurring species. This should be taken into account when examining the results of this study. However, if percent occurrence in early midsuccession and mid- succession from Table 3 were adjusted to com- pensate for apparent losses due to seeding, the changes would generally be less than 10%. In addition, other factors may be confounding the apparent decrease in species occurrence correlated to seeding because the burns are of different years and locations. Because of the difficulty in isolating the cause of decreased occurrence on seeded sites and the relatively small decrease involved, it was not considered worthwhile to analyze seeded burns sepa- rately from nonseeded. Conclusions Plant succession following wildfire within the pinyon-juniper type has previously been described as relay floristics (Arnold and others 1964, Erdman 1970, Barney and Frisch- knecht 1974), a sequential migration of later successional species into a site (Egler 1954). Everett and Ward (1984), however, have indi- cated that initial floristics, the sequential dominance of a site by species on the burn immediately after wildfire (Egler 1954), may be of equal or even greater importance. This study supports the latter hypothesis. Species that became dominant in midsuccession and late succession were present in the early stage. Slow growth or establishment rates precluded early dominance. Tree species, which rely on nurse plants for establishment and survival, were the only exceptions. Within the pinyon-juniper woodlands, postburn succession may follow multiple pathways (Everett and Ward 1984). Succes- sional patterns varied with aspect and associ- ated preburn species composition. To in- crease the predictability of postburn plant and community response, effects of elevation, soil type, seeding, postburn climate, and severity and timing of disturbance should also be con- sidered and studied at greater length. 566 Great Basin Naturalist Vol. 45, No. 3 Literature Cited Arnold. J H . D A Jameson, and E H Reid 1964. The pinyon-juniper type of Arizona: effect of grazing, fire, and tree control. USDAFor. Sevr. Prod. Res. Rept. 84. 28 pp. Barbour, M. G.. J H Burk, and W. D Pitts 1980. Ter- restrial plant ecology. Benjamin/Cunimings Publ. , Menlo Park, California. Barney, M. A., and N C. Frischknecht 1974. Vegeta- tive changes following lire in the pinyon-juniper woodland in Nevada. J. Range Manage. 27:91-96. Baskin, J, M., and C. C Baskin. 1981. Ecology of germi- nation and flowering in the weedy winter annual grass Bromus japonicus. J. Range Manage. 34:369-372. Egler. F E 1954. Vegetative science concepts. I. Initial floristic composition: a factor in old-field vegeta- tion development. Vegetatio 4:412—117. Erdman, J. A 1970. Pinyon-juniper succession after natu- ral fires on residual soils of Mesa Verde, Colorado. Brigham Young Univ. Sci. Bull., Biol. Series 11(2). 24 pp. Evans, R, A., and J. A Young. 1982. Russian thistle and barbwire Russian thistle seed and seedbed ecol- ogy. USDA Agric. Res. Serv. Agric. Res. Results W-25. 39 pp. Everett, R L , and K Ward. 1984. Early plant succes- sion on pinyon-juniper control bums. Northwest Science 58:57-58. Holmgren, A. H., and J L. Reveal 1966. Checklist of the vascular plants of the Intermountain Region. USDAFor. Serv. Res. Pap. INT-32. 60pp. Klebenow, D., R. Beall, A. Bruner, R Mason, B, RouNDY, W, Stager, and K. Ward. 1977. Con- trolled fire as a management tool in the pinyon-ju- niper woodland. Nevada. USDA For. Serv. and Nev. Agric. Exper. Sta. Summary' Progress Rept. 155 pp. KoNiAK, S . and R L Everett. 1982. Seed reserves in soils of successional stages of pinvon woodlands. Amer. Midi. Nat. 108:295-303. KONIAK. S. 1983. Broadcast seeding success in eight pinyon-juniper stands after wildfire. USDA For. Serv. Res. Note lNT-334. 4 pp. Merrill. E H . H F Mayland, and J. M Peek. 1980. Efifects of a fall wildfire on herbaceous vegetation on .xeric sites in the Selway-Bitterroot Wilderness, Idaho. J. Range Manage. 33:363-367. MuELLER-DoMBOis, D , AND H. Ellenberg. 1974. Aims and methods of vegetative ecology. John Wiley and Sons, New York. 448 pp. Stager. D W 1977. Mule deer response to successional changes in the pinyon-juniper vegetation type af- ter wildfire. Unpublished thesis. University of Nevada, Reno. 41 pp. White. R S . and P. O Curie 1983. Prescribed burning in the Northern Great Plains: yield and cover responses of three forage species in the mLxed grass prairie. J. Range Manage. 36(2): 179-183. Wright, H A., L, F Neunschwander, and C, M. Brit- ton 1979. The role and use of fire in sagebrush- grass and pinyon-juniper plant communities: state-of-the-art review. USDA For. Serv. Gen, Tech. Rept. INT-58. 48 pp. Young, J A , R A Evans, and R. A Weaver. 1976. Esti- mating potential downy brome competition after wildfires. J. Range Manage. 29:322-325. Young, J A , and R A Evans. 1982. Temperature profiles for germination of cool season range grasses. USDA Agric. Res. Ser. Agric. Res. Results W-27. 92 pp. USE OF RADIO TRANSMITTER IMPLANTS IN WILD CANIDS Jeffrey S, Green', Richard T. Golightly, Jr.^, Susan Lyndaker Lindsey\ and Brad R. LeaMaster' Abstract. — Twelve adult and five juvenile coyotes and 20 adult kit foxes were implanted with radio transmitters using relatively simple surgical procedures. Four foxes were successfully implanted in the field. None of the animals implanted exhibited noticeable behavioral effects, and no deaths were confirmed to result from implantation. Implants were attached to the peritoneum in adult coyotes and kit foxes and were left free-floating within the abdominal cavity of the coyote pups. Both procedures produced satisfactory results. Radio signals transmitted from implants had less range than those from traditional neck collar transmitters. Implants offered benefits unavailable with traditional collar transmitters: no external packaging to influence behavior, ability to radio monitor small or juvenile animals, and ability to acquire various physiological data on free-ranging individuals. , . Implantable radio transmitters have beenused to monitor physiological parame- ters and movements of numerous mammals (primates, Stone et al. 1972; canids, Golightly and Ohmart 1983; ursids, Jessup and Koch 1984; mustelids, Melquist and Hornocker 1979, Melquist et al. 1981, Garshelis and SiniflP 1983, Eagle et al. 1984; sciurids. Eagle et al. 1984, Golightly and Ohmart 1978; cas- torids, Davis et al. 1984; and others. Folk et al. 1971). Implants may offer advantages over externally attached transmitters such as a de- creased effect on animal behavior, ability to instrument small or juvenile animals, and ability to acquire physiological data. One dis- advantage of implantation is surgical risk to the animal. Techniques of implantation have been described for several species, yet there is a lack of detail concerning the implantation procedure and effectiveness of implants for monitoring physiological parameters and movements of canids. In this paper we report the results of two independent studies using implants in wild canids. In one study (authors JSG, SLL, and BRL) implants were used to monitor the movements of juvenile coyotes (Canis la- trans), and in the other (author RTG) im- plants were used in adult coyotes and adult kit foxes {Vulpes macrotis) to monitor body tem- perature (Golightly and Ohmart 1983) and lo- cation (Golightly 1981). Study Areas and Methods Coyote Pups Five pups (two male, three female) were obtained from two captive litters in two 65-ha enclosures at the U.S. Sheep Experiment Sta- tion, Clark County, Idaho. Pups were 100-136 days old and weighed 3.7-5.8 kg. Each pup was pretranquilized with 2.2 mg/kg xylazine (Rompun) intramuscularly and 0.05 mg/kg atropine sulfate subcutaneously. In- duction and surgical anesthesia were obtained by administering 16 mg/kg thiamylal sodium intravenously until a desired plane of anesthe- sia was reached. Each pup was placed in dor- sal recumbency and its abdomen shaved from the xyphoid process to the pubis. This area was washed, and a 4-cm incision through the skin was made caudal to the umbillicus. The peritoneal cavity was entered through the linea alba by a short incision made with a scalpel and extended with blunt-tipped scis- sors. Each transmitter (22 g cylinder, 6 cm long, 2 cm diameter; Telonics, 932 E. Impala, Mesa, AZ 85204) was presoaked for approxi- mately 15 minutes in a disinfectant solution (Nolvasan) and inserted into the peritoneal cavity. The peritoneum and internal and external rectus sheaths were sutured with a simple interrupted pattern using size 00 chromic gut. U.S. Department of Agriculture, Agricultural Research Service, U.S. Sheep Experiment Station, Dubois, Idaho 83423. 'Department of Wildlife, Humboldt State University, Areata, California 95521. Department of Zoology and Entomology, Colorado State University, Fort Collins, Colorado 80523. 567 568 Great Basin Naturalist Vol. 45, No. 3 The superficial fascia and skin were reposi- tioned with a simple interrupted pattern of 0 chromic gut. Each pup was given a prophylac- tic injection of 150,000 units of procaine peni- cillin G and 150,000 units of benzathine peni- cillin G (1 cc Benza-Pen) intramuscularly. Two days later the pups were given a second 1-ml dose of the antibiotic and released into the enclosures. From August 1982 through May 1983 per- formance of the implanted transmitters was compared with that of 10 collar-mounted transmitters (weight 290 g; Telonics) on adult coyotes concurrently within the enclosures. Coyotes were observed from an elevated (9 m) observation booth located at a boundary com- mon to both enclosures. Signals were re- ceived with a phase-combined stacked array of two 5-element beams, 15.5 m above the ground. Maximum distance from transmitter to receiver was 1.14 km. Signals were classi- fied as good (signal strong, direction easily determined), fair (signal of medium strength, direction diflPicult to determine), poor (signal audible, unable to determine direction), or no signal. The implants were recovered when the coyotes were recaptured nine months follow- ing implantation. The surgical procedure for removal of the implants was similar to the implanting procedure. Coyote Adults In 1977, 12 adult coyotes (10 male, 2 fe- male) weighing 9.1-12.7 kg were captured near Phoenix, Maricopa County, Arizona, and immobilized with a mixture of 1.25 mg/kg ketamine hydrochloride (Ketaset), 0.5 mg/kg Rompun, 0. 12 mg/kg Acepromazine, and 0.006 mg/kg atropine sulfate administered in- tramuscularly. A surgical plane of anesthesia resulted in 5-7 minutes. The peritoneum was opened in a manner similar to that described for pups, and a temperature-sensitive trans- mitter (13 g, 5.7 by 2.4 by 1.7 cm; J. Stuart Enterprises, Grass Valley, CA 95965) was in- serted. A length of size 0 Vetafil ligature was run through a loop of Vetafil attached to the transmitter, and both ends of the ligature were held by clamps outside the incision. The ligature was then tied to the peritoneum 1 cm lateral to the superior end of the incision, thus securing the transmitter to the peritoneal wall. The peritoneum and the linea alba were closed with a running suture of size 0 chromic gut. Powdered Furacin was applied to the sutured incision. The skin was then closed with mattress sutures of size 0 Vetafil, and Betadine ointment was applied to the suture line. Following surgery, 75,000 units of Bi- cillin were administered intramuscularly. Coyotes were confined to cages for 12 hours before being released into a 0. 13 ha enclosure at Arizona State University. To remove an implant, an incision was made similar to the one described previously. The ligature was located, clamped, and cut, the implant was removed, and closure was accomplished as described previously. Kit Foxes In 1978 and 1979, 12 kit foxes (five male, seven female) were captured near Apache Junction, Pinal County, Arizona, implanted with transmitters, and released in a 0. 13-ha enclosure at Arizona State University. Eight additional kit foxes (three male, five female) were captured and released into the wild fol- lowing placement of an implant and a conven- tional neck-collar radio transmitter (110 g, Telonics). Body weights of kit foxes ranged from 1.4 to 2.4 kg. Kit foxes were immobilized with a mixture of 2 mg/kg Ketaset, 0.25 mg/kg Rompun, 0.25 mg/kg Acepromazine, and 0.012 mg/kg at- ropine sulfate administered intramuscularly. Anesthesia resulted in 3-5 minutes. The im- plant and implantation procedure as pre- viously described for adult coyotes was used. Following surgery, 25,000 units of Bicillin were administered to each animal intramus- cularly. Kit foxes were released into their dens or into the enclosure 6-8 hours following surgery. Four of the foxes released into the wild were implanted in the field; othei-wise, all surgeries were performed in a laboratory. Results and Discussion Coyote Pups We did not obserxe any behavioral changes or mortality resulting from implanting coyote pups. Implants provided an advantage over July 1985 Green etal.:CanidTransmitter implants 569 collars in that we were able to e(]uip each pup with a transmitter at a young age. This made capture more efficient and less stressful to the animal since young pups usually retreated to a den fi-om which they were dug, and older animals usually ran until cornered or ex- hausted. Also, data on sex, weight, physical measurements, and movement were obtained earlier than if coyotes had been captured after they reached adult size when conventional radio collars could be used. (Expandable neck collar transmitters may also be used on juveniles.) Three disadvantages of implants were rec- ognized. The transmitted signals from the im- plants (58% rated good, 21% fair, 7% poor, 14% no signal, n = 254) were not as strong as those of the collars (93% good, 5% fair, 1% poor, 1% no signal, n = 630). However, the implants were usually adequate for identify- ing location or direction of the pups. Second, use of a motion-sensing option with the im- plant would have necessitated attachment of the implant to the abdominal wall (probably more extensive than the attachment de- scribed previously for adult coyotes and kit foxes). We were unsure of the risk (e.g., inter- nal complications) involved with such a proce- dure. Third, the implants had an approxi- mately two-thirds shorter operational life than collar transmitter packages (based on data from Telonics). When the time came to remove the im- plants from the coyote pups, three of the four implants were floating freely within the ab- dominal cavity and the fourth was encased within a thin membrane. There was no evi- dence of trauma or irritation of tissue adjacent to the site of implantation. Although coyote pups were implanted at 3.5 and 4.5 months of age, implanting com- parably sized transmitters would likely be possible in coyote pups as young as 2-2.5 months. Coyote Adults From 1977 to 1980, 21 transmitter implan- tations were performed on 12 adult coyotes. Each implantation procedure lasted approxi- mately 30-35 minutes. Postsurgical infection requiring medical attention developed in only one instance. All coyotes appeared healthy and fed normally during the 2-4 month exper- imental periods following implantation. One female successfully produced a litter of two pups while carrying an implant. Serial implantations (replacement of an im- plant in each of six coyotes) resulted in moder- ate scar tissue and made final retrieval surgeries slightly more difficult. The ligature attachment did not cause ap- parent damage to internal organs or to the peritoneum, and it facilitated retrieval of the implant through a relatively small incision. Some experimental conditions (e.g., where public visibility is a concern or in heat ex- change studies) necessitate that a small surgi- cal site be prepared, thus potentially making retrieval of a free-floating implant difficult. Attachment of the implant to the peritoneum reduced time in surgery and trauma during recovery or replacement of implants. This might be an important consideration with rare species or animals that are difficult to obtain. The attachment also provided a consistent location for measurement of body tempera- ture. Some experiments require accurate measurement of relatively small differences in body temperature with implanted transmit- ters (Golightly 1981). Body temperature varied at locations within the abdominal cav- ity of mammals during arousal from hiberna- tion (Lyman et al. 1982), and body tempera- ture of normothermic mammals may vary at different locations within the body. It would be particularly important to attach implanted transducers for measuring heart or respiration rate. The transmitters were effective in provid- ing information on body temperature to re- mote receiving locations outside the enclo- sure. Signals were routinely received with a hand-held Yagi antenna at a distance of 70-100 m from the animal despite the pres- ence of physical barriers (i.e., brick walls, rock piles) between the transmitter and the receiver. Kit Foxes Sixteen transmitters were implanted into foxes in the enclosure, including four serial exchanges (two implants in each of four foxes). The surgery lasted approximately 20-35 min- utes. The kit foxes were observed every 2 or 3 570 Great Basin Naturalist Vol. 45, No. 3 days, and infections were not evident. The animals maintained weight and appeared healthy. One female died of undetermined causes while carrying the implant. No pathol- ogy was evident upon necropsy, and the tissue around the implant appeared normal. Two males developed hernias at the site of incision several months following implanta- tion. The transmitter may have physically stressed the site of incision and caused or contributed to the hernia. Both foxes had re- ceived replacement implants. One fox had continued problems with the hernia following repair of the surgical site. Because kit foxes have a thin peritoneum compared with that of coyotes, we suspect that the serial implanta- tions precipitated the hernias. Four of the eight free-ranging kit foxes died during the study but none from transmitter-related causes. All foxes appeared to feed normally and maintain weight, and some moved substantial distances at night (maximum distance of ap- proximately 21 km in a single night, Golightly 1981). No infections were observed. One free- ranging female was implanted early in preg- nancy and carried the implant for four months while successfully whelping and rearing a lit- ter of three pups. A signal transmitted from an implant nor- mally could be received 75-100 m away from foxes in dens (1-2 m underground), whereas transmitters affixed by neck collars could be received 500-700 m. Outside dens, signals from implants and collars were received at 200+ m and 1,000-2,000 m, respectively. Implants were useful for obtaining body temperature data and for determining move- ment within a den after the fox was located with the signal from the neck collar. Implants were not useful as a sole means of locating kit foxes. However, kit foxes should be able to accommodate the larger implant described previously for coyote pups, which would per- haps increase their range of transmission. Acknowledgments The authors thank R. Woodruff, W. Bow- ers, L. Pritchett, and J. Hamilton for help with collection of data and G. Bjotuedt and H. Hood for consultation during the develop- ment of procedure. D. Garshelis, J. Litvaitis, D. Guynn, Jr., J. McGrew, W. Melquist, S. Tomkiewicz, and R. Woodruff commented on the manuscript. Literature Cited Davis, J. R , A F Von Recum. D D Smith, and D. C. Guynn. 1984. Implantable telemetry in beaver. Wildl. Soc. Bull. 12;322-324. Eagle, T C , J C. Norris, and V. B. Kuechle, 1984. Implanting radio transmitters in mink and Franklin's ground squirrels. Wildl. Soc. Bull. 12:180-184. Folk, G E, Jr . W O Essler, and M. A. Folk. 1971. The abdominal cavitv for transport of instruments. Fed. Proc. 30:700. Garshelis, D. L., and D. B. Siniff. 1983. Evaluation of radio-transmitter attachments for sea otters. Wildl. Soc. Bull, 11:378-383, Golightly, R. T. 1981, The comparative energetics of two desert canids: the coyote (Canis latrans) and the kit fox {Vulpes macrotis). Unpublished disserta- tion. Arizona State University, Tempe, 175 pp. Golightly, R T., and R D Ohmart 1978, Het- erothermy in free-ranging Abert's squirrels (Sctu- rusabertii). Ecology 59:897-909, 1983. Metabolism and body temperature of two desert canids: coyotes and kit foxes, J, Mammal, 64:624-635, Jessup, D. a , and D S. Koch 1984, Surgical implanta- tion of a radiotelemetry device in wild black bears, Ursus americanus. California Fish and Game 70:163-166, Lyman, C P , J S Willis, A Malan, and L C H Wang. 1982, Hibernation and torper in mammals and birds. Academic Press, New York, 317 pp, Melquist, W. E, and M. G. Hornocker. 1979, Develop- ment and use of a telemetry technique for study- ing river otter. Pages 104-114 in F, M, Long, ed,, Proc, Second Int. Conf Wildl, Biotelemetry. Laramie, Wyoming. 259 pp, Melquist, W E , J S Whitman, and M G Hornocker 1981, Resource partitioning and coexistence of sympatric mink and river otter populations. Pages 187-220 in J, A, Chapman and D. Pursley, eds,, Worldwide Furbearer Conf Proc, Vol, I, Frost- burg State College, Frostburg, Maryland. 2056 pp. Stone, H L., H. H. Sandler, and I. B Fryer. 1972, Im- plantable telemetry in the chimp. Proc. Int. Telem, Conf 8:464, NOTICE TO CONTRIBUTORS Manuscripts intended for publication in the Great Basin Naturalist or Great Basin Natural- ist Memoirs must meet the criteria outhned in paragraph one on the inside front cover. They should be directed to Brigham Young University, Stephen L. Wood, Editor, Creat Basin Naturalist, 290 Life Science Museum, Provo, Utah 84602. Three copies of the manuscript are required. They should be typewritten, double spaced throughout on one side of the paper, with margins of at least one inch on all sides. Use a recent issue of either journal as a format, and the Council of Biology Editors Style Manual, Fourth Edition (AIBS 1978) in preparing the manuscript. An abstract, about 3 percent as long as the text, but not exceeding 200 words, written in accordance with Biological Abstracts guidelines, should precede the introductory paragraph of each article. All manuscripts receive a critical peer review by specialists in the subject area of the manuscript under consideration. Authors may recommend one or two reviewers for their article. Accepted manuscripts less than 100 printed pages in length will automatically be assigned to the Great Basin Naturalist. Those larger than 100 printed pages will be considered for the Memoirs series. Illustrations and Tables. All illustrations and tables should be made with a view toward having them appear within the limits of the printed page. The original illustrations or glossy photoprints of them should accompany the manuscript. Illustrations should be prepared for reduction by the printer to either single-column (2| inches) or double-column (5| inches) width, with the length not exceeding 7| inches. Costs Borne by Contributor. Contributors to the Great Basin Naturalist should be prepared to donate from $10 to $40 per printed page toward publication of their article (in addition to reprint costs). Excessive or complex tables requiring typesetting may be charged to the author at cost. Authors publishing in the Great Basin Naturalist Memoirs may be expected to contribute $40 per printed page in addition to the cost of the printed copies they purchase. No reprints are furnished free of charge. Reprint Schedule for the Great Basin Naturalist 100 copies, minimum cost for 2 pages, $26. Each additional 2 pages, $6. Each additional 100 copies, $4 for each 2 pages. Examples: 300 copies of 10 pages = $82; 200 copies of 13 pages = $86. Great Basin Naturalist Memoirs No. 1 The birds of Utah. ByC. L. Hayward, C. Cottam, A. M. Woodbury, H. H. Frost. $10. No. 2 Intermountain biogeography: a symposium. By K. T. Harper, J. L. Reveal et al. $15. No. 3 The endangered species: a symposium. $6. No. 4 Soil-plant-animal relationships bearing on revegetation and land reclamation in Nevada deserts. $6. No. 5 Utah Lake monograph. $8. No. 6 The bark and ambrosia beetles of North and Central America (Coleptera: Scolytidae), a taxonomic monograph. $60. No. 7 Biology of desert rodents. $8. TABLE OF CONTENTS Quaternary paleontology and paleocology of Crystal Ball Cave, Millard County, Utah: with emphasis on mammals and description of a new species of fossil skunk. Timothy H. Heaton 337 First record of Cliinacia californica (Neuroptera: Sisyridae) and its host sponge, Ephydatia mulleri (Porifera: spongillidae), from Idaho with water quality rela- tionships. William H. Clark 391 Poa L. in New Mexico, with a key to middle and Southern Rocky Mountain species (Poaceae). Robert J. Soreng 395 Dwarf mistletoe-pandora moth interaction and its contribution to ponderosa pine mortality in Arizona. Michael R. Wagner and Robert L. Mathiasen 423 Occurrence of anisakid larvae (Nematoda: Ascardidia) in fishes from Alaska and Idaho. Richard Heckmann and Terry Otto 427 Soil algae of cryptogamic crusts from the Uintah Basin, Utah, U.S.A. John Ashley, Samuel R. Rushforth, and Jeffrey R. Johansen 432 In memoriam; William Wallace Newby (1902-1977). William H. Behle 443 Symbos cavifrons (Artiodactyla; Bovidae) from Delta County, Colorado. Jerry N. McDonald 455 Comparisons ofprescribed burning and cutting of Utah Marsh plants. Loren M. Smith and John A. Kadlec 462 New species and records of North American Pityophthonis (Coleoptera: Scolytidae), Part IV: the Scriptor group. D. E. Bright 467 New species and new records of North American Pityophthonis (Coleoptera; Scolyti- dae), Part V: the Juglandis group. D. E. Bright 476 Second nesting record and northward advance of the Great-tailed Crackle (Quiscalus mexicanus) in Nevada. Jennifer A. Holmes, David S. Dobkin, and Bruce A. Wilcox 483 New species oiTalinum (Portulaceae) from Utah. N. Duane Atwood and Stanley L. Welsh 485 Types of Nevada buckwheats {Eriogonum : Polygonaceae). James L. Reveal 488 Annotated key to Eriogonum (Polygonaceae) of Nevada. James L. Reveal 493 High rates of photosynthesis in the desert shrub Chrysothamnus nauseosus ssp. albicaulis. Tim D. Davis, N. Sankhla, W. R. Andersen, D. J. Weber, and B. N. Smith 520 Food habits of the western whiptail lizard {Cnemidophorus tigris) in southeastern New Mexico. Troy L. Best and A. L. Gennaro 527 Vegetation patterns in relation to slope position in the Castle Cliffs area of southern Utah. Jack D. Brotherson and William J. Masslich 535 Invasion and stabilization of recent beaches by salt grass {Distichilis spicata) at Mono Lake, Mono County, California. Jack D. Brotherson and Samuel R. Rushforth 542 Grass spider microhabitat use in Organ Pipe Cactus National Monument, Arizona. Mark Robert Deutschman 546 New species o{ Primula (Primulaceae) from Utah. Ronald J. Kas.s and Stanley L. Welsh 548 New species ofAsfraga/us(Legumino.sae) from southeastern Utah. Rupert C. Barneby and Stanley L. Welsh 551 New Sclerocactus (Cactaceae) from Nevada. Stanley L. Welsh and Kaye Hugie Thorne 553 Succession in pinyon-juniper woodlands following wildfire in the Great Basin. Susan Koniak 556 Use of radio transmitter implants in wild canids. Jeffrey S. Green, Richard T. Golightly, Jr., Susan Lyndaker Lindsey, and Brad R. LeaMaster 567 IE GREAT BASIN NAI UKALIb I je45No.4 MCZ LIBRARY 31 October 1985 Brigham Young University AUG 2 0 1986 HARVARD UNIVERSITY GREAT BASIN NATURALIST Editor. Stephen L. Wood, Department of Zoology, 290 Life Science Museum, Brigham Young University, Provo, Utah 84602. Editorial Board. Kimball T. Harper, Chairman, Botany; James R. Barnes, Zoology; Hal L. Black, Zoology; Stanley L. Welsh, Botany; Clayton M. White, Zoology. All are at Brigham Young University, Provo, Utah 84602. Ex Officio Editorial Board Members. Bruce N. Smith, Dean, College of Biological and Agricultural Sciences; Norman A. Darais, University Editor, University Publications. Subject Area Associate Editors. Dr. Noel H. Holmgren, New York Botanical Garden, Bronx, New York 10458 (Plant Taxonomy). Dr. James A. MacMahon, Utah State University, Department of Biology, UMC 53, Logan, Utah 84322 (Vertebrate Zoology). Dr. G. Wayne Minshall, Department of Biology, Idaho State University, Pocatello, Idaho 83201 (Aquatic Biology). Dr. Ned K. Johnson, Museum of Vertebrate Zoology and Department of Zoolog) , University of California, Berkeley, California 94720 (Ornithology). Dr. E. Philip Pister, Associate Fishery Biologist, California Department of Fish and Game, 407 West Line Street, Bishop, California 93514 (Fish Biology). Dr. Wayne N. Mathis, Chairman, Department of Entomology, National Museum of Natural History, Smithsonian Institution, Washington, D.C. 20560 (Entomology). Dr. Theodore W. Weaver III, Department of Botany, Montana State University, Boze- man, Montana 59715 (Plant Ecology). The Great Basin Naturalist was founded in 1939 and has been published from one to four times a year since then by Brigham Young University. Previously unpublished manuscripts in English of fewer than 100 printed pages in length and pertaining to the biological natinal history of western North America are accepted. Western North America is considered to be west of the Mississippi River from Alaska to Panama. The Great Basin Naturalist Memoirs was established in 1976 for scholarly works in biological natural history longer than can be accom- modated in the parent publication. The Memoirs appears irregularly and bears no geographical restriction in subject matter. Manuscripts are subject to the approval of the editor. Subscriptions. The annual subscription to the Great Basin Naturali.st for private individuals is $16; for institutions, $24 (outside the United States, $18 and $26); and for student subscrip- tions, $10. The price of single issues is $6 each. All back issues are in print and are available for sale. All matters pertaining to subscriptions, back issues, or other business should be directed to Brigham Young University, Great Basin Naturalist, 290 Life Science Museum, Provo, Utah 84602. The Great Basin Naturalist Memoirs may be purchased from the same office at the rate indicated on the inside of the back cover of either journal. Scholarly Exchanges. Libraries or other organizations interested in obtaining either journal through a continuing exchange of scholarly publications should contact the Brigham Young University Exchange Librarian, Harold B. Lee Library, Provo, Utah 84602. Manuscripts. See Notice to Contributors on the inside back cover. 4-86 650 21168 ISSN 017-3614 The Great Basin Naturalist Published AT Provo, Utah, by Bricham Young University ISSN 0017-3614 Volume 45 31 October 1985 No. 4 LIFE HISTORY OF THE CUI-UI, CHASMISTES CUJUS COPE, IN PYRAMID LAKE, NEVADA: A REVIEW William F". Sigler', Steven Vigg", and Minii Bres' Abstract— The cui-ui, Chasmistcs ciijus Cope, a member of the .sucker family and endemic to Pyramid Lake, Nevada, is listed as endangered by the U.S. Fish and Wildlife Service. Cui-ui was once a major source of sustenance for native Americans, who have inhabited the Lahontan region for at least 11,000 years. The Northern Paiutes developed sophisticated fishing technology to harvest this resource. The original distribution of cui-ui was the ancient Lake Lahontan comple.x, but as a result of climatic changes it was restricted to the Pyramid-Winnemucca-Truckee system by the turn of the 20th century. Transbasin water diversions (190.5 to present) have resulted in further restrictions of habitat. The species is now limited to Pyramid Lake and the lower Truckee River. Reproduction is from hatcheries as well as limited natural reproduction. Females produce more than 40,000 2-mm eggs per year. The normal develop- ment is described from the unfertilized egg through 912 hours post-hatching, when the fry are actively feeding and approaching adult body form. The unusual feature of adult cui-ui morphology is the relatively large ventro-terminal i mouth, with thin and obscurely papillose lips. Cui-ui grow slowly and may live 18 years or possibly much longer; females generally live longer and attain a greater size than males. The highest adult mortality probably occurs during spawning runs. At this time they are vulnerable to predation, stress, and sometimes environmental degradation. The highest larval mortality probably occurs from predation when they are planted or migrate into the lake. The trophic ecology of the species is poorly understood, but they are known to ingest algae and zooplankton. Spawning behavior is documented. At present, natural reproduction is probably still the limiting factor for the cui-ui population. Cui-ui ccMHposed less than one percent of the total fish in Pyramid Lake dining 1975-1977. During 1982 the largest cui-ui spawning run (13,000) in recent years occurred. The acti\it\' of cui-ui in the lake closeK' resembles that of the Tahoe ^^ucker being most active during the spawning season each spring. C'ui-ui inhabit the inshore-benthic zone and the ^ pelagic waters of Pyramid Lake (<46 m). The cui-ui, Chasmistes cujiis Cope, a mem- reached its maximum size of about 22,300 km' ber of the sucker family (Catostomidae), is some 13,000 years before present (BP) and present only in Pyramid Lake and the affluent inundated a large portion of northwestern Ne- lower Truckee River, Nevada (Fig. 1). Be- vada. The cui-ui was present in Winnemucca cause of its limited range and depleted num- Lake until the late 1920s or early 1930s (Fig. bers, it is listed as endangered (Federal Regis- 2). ter. Vol. 32/48, 11 March, 1967). Cui-ui until There is general agreement that the ecologi- recently was an important food source for cal devastation of the cui-ui's lake and river Northern Paiute, the native Americans who environment was caused in part by the New- have inhabited the region for at least 11,000 lands Reclamation Irrigation Project (NRIP), y years. Prehistorically the habitat of cui-ui con- which was authorized by the U.S. Congress in n sisted of the Lake Lahontan system, which 1903. In 1905 Derby Dam was dedicated, and 'W. F. Sigler& Associates Inc., 309 East 200 South. Logan, Utah 84321 "Biological Sciences Center, Desert Research Institute, University of Nevada Svstein, Reno. Nevada ■ Department ofBiological Sciences, George Washington University, W'asliington. D.C. 571 572 Great Basin Naturalist Vol. 45, No. 4 Fox Valley Needles Cormorant Rock True North Hells Kitchen Anderson Bay Dago Bay Truckee River Popcorn Fi«. 1. liatlivinrtricniapolT'Mann.l Lake, N.'N ada; cU'iUli o.ntou ■Icisatrlrvation 1154.9 in. October 1985 SiCLER ET AL.: Cui-Ul IN PVHAMII) LAKE 573 WINNEMUCALAKE ; Mud Lake Slough To Lahonton Reservoir Fig. 2. The Truckee River- Pyramid Lake Ecosystem. 574 Gkeat Basin Naturalist Vol. 45, No. 4 transbasin water diversion from the Truckee River to the Carson River system began. La- hontan Dam on the Carson River was com- pleted in 1915, creating Lahontan Reserv- ior — the water storage imponndment for the NRIP. From 1915 to 1970 as much as half or more of the total flow of the lower Truckee River was diverted to the NRIP. Because of droughts and diversions, the level of Pyramid Lake declined more than 24.4 m from 1909 to 1968, Pvramid Lake increased in total dis- solved solids (TDS) from about 3500 to 5500 mg/1, and Winnemucca Lake disappeared in 1938. A delta developed at the mouth of the Truckee River in the early 1930s that was virtually impassable to spawning migrations of cui-ui. Natural reproduction in the Truckee River was very limited for about 50 years (un- til the new Marble Blufldam and the fishway became functional for cui-ui in 1982). How- ever, during years of exceptionally high flow, natural reproduction may have been possible. Upstream, Siphon dam (washed out in 1958), and about 1 mile below it the original Marble Bluff" dam (washed out in 1950) were also bar- riers to migrating cui-ui. Neither of these ob- structions had functional fish ladders. The original Numana Dam also barred cui-ui mi- gration. It is our objective to synthesize information collected during the Pyramid Lake Ecological Study, conducted by W. F. Sigler & Associ- ates Inc. during 1975-1978, with available data from agency reports and research publi- cations to present an overview. It is hoped this paper will contribute to the knowledge of the cui-ui, and that its deficiencies will point out areas where additional research is needed. Historical Over\ iew Cope (1883) first diagnosed and revised the genus Chasmistcs and named a new species C. ciijus from Pyramid Lake. In 1918, Snyder published the first life history information on C. cujus and other fishes of the Lahontan System; even at this early date Snyder consid- ered the fate of the cui-ui to be uncertain. Sumner (1940) collected environmental and fishery data from Pyramid Lake and the Truc- kee River, compiled a chronology of the fish- ery, and stated that the major cause of the decline of the fishery was the transbasin dixer- sion of Truckee River water. T. J. Trelease, the first fishery biologist for the Nevada Department of Fish and Game (NFG), did preliminary work on the diet and reproduction of cui-ui (La Rivers 1962). Jonez (1955) and Johnson (1958) (both NFG biolo- gists) worked with cui-ui during the 1950s conducting evaluations of cui-ui behavior and habitat. La Rivers made many observations over the years and developed a life history for cui-ui, incorporating information from previ- ous workers. Koch (1972, 1973) supplied information on life history, reproductive characteristics, and spawning behavior of cui-ui, Koch and Contreras (1973) advanced artificial hatching techniques, and Koch (1976) summarized available life history information. The U.S. Fish and Wildlife Service operated a cui-ui hatchery in 1974-75. Pyramid Lake Fisheries (PLF) has operated the David L. Koch Cui-ui Hatchery since 1977 and has further refined hatching and rearing technicjues. In 1971 the U.S. Department of the Interior (DI) reported the classification status of the cui-ui. Federal restoration of the species be- gan in 1973 by the U.S. Fish and Wildlife Service (FWS) cui-ui recovery team. This team completed a Draft Cui-ui Recoverv Plan in 1977 (Pyle et al. 1977). The 1982 revision of the original Cui-ui Recovery Plan was ap- proved bv the FWS and reviewed bv DI (U. S. Fish and Wildlife Service 1983). In 1975 the U.S. Bureau of Indian Affliirs (BIA) funded studies on the fisheries of the Truckee River and Pyramid Lake. The results of the Pyramid Lake Ecology Studies, includ- ing data on cui-ui ecology, are presented in Sigler and Kennedy (1978). The results of the Truckee River studies are in preliminary FWS reports. McConnell, Galat, and Hamil- ton-Galat (1978) and Galat and McConnell (1981) discuss Pyramid Lake fish production in relation to potential changes in total dis- solved solids (TDS). In the early 196()s the NFG developed plans for a fishwa\ that would enable upstream mi- grating fish to b\ pass the delta and enter the lower Truckee River. The plans were submit- ted to the Fleischmann Foundation, Reno, Ne\ada, but the facility was not funded be- cause the Foundation could be given no assur- ance of a water right. The NFG, along with the FWS and the U.S. Bureau of Reclamation ( )ctober 1985 SiGLER ET AL.: Cl'I-UI IN PYRAMID LaKE 575 (BOR), then developed plans for a larger and more elaborate facility. The NFG also lobbied with state and national agencies for the Washoe Project Act, which made funding pos- sible (T. J. Trelease personal communication 1984). The Washoe Project Act was made much more salable by the earlier develop- ment, largely by NFG, of highly successful Lahontan cutthroat trout, Sahno clarki Jicn- sluiivi, fishery. In 1975 BOR completed the Marble Bluff" Fishway. The FWS operates the Marble Bluff facility and monitors spawning migra- tions of cui-ui and Lahontan cutthroat trout. Data collected by FWS on cui-ui spawning populations in the lake and fishway are pre- sented by U.S. Fish and Wildlife Service, Nevada Department of Fish and Game, ( California Department of Fish and Game (1976), Ringo and Sonnevil (1977), and Son- nevil (1977a, 1977b, 1978, 1981). The age structure of cui-ui in 1978 was determined bv Robertson (1979). Scoppettone et al. (1981, 1983, and G. Scoppettone personal communication 198.3) studied the spawning behavior and habitat re(|uirements of cui-ui in a natural side channel of the lower Truc- kee River. Research on the habitat and ecology of fish species in Pyramid Lake was conducted by Vigg (1978a). Vertical distribution patterns and relative abundance are reported (Vigg 1978b, 1980, 1981). Research on the effects of increasing levels of TDS on cui-ui was initiated bv Earl Pvle of FWS during 1975-1978. Chatto (1979) pre- sented preliminary data on hatching success of cui-ui eggs in various proportions of Pyra- mid Lake water. Lockheed Ocean Sciences Laboratories (LOSL) (1982) studied the ef- fects of various levels of TDS on the embryos, larvae, and juveniles of cui-ui. T. J. Trelease first reared larvae in 1947, and Kay Johnson and Ivan Young (all NFG person- nel) raised them to adult size — about 31 cm. Koch et al. (1979) estimated 91.6% hatching success in controls during nitrogen-species bioassays. However, they were unable to ob- tain definitive results on toxicity because of high mortality in all treatments and controls. Koch (1981) conducted preliminary tempera- ture tolerance studies of cui-ui embryos and larvae. Various morphological studies have been conducted on catostomid fishes, including cui-ui. Nelson (1948, 1949, 1961) studied the comparative morphology of the Weberian ap- paratus, the opercular series, and the swim bladder, respectively. Miller and Evans (1965) studied the external morphology of the catostomid brain and lips. Snyder (1981a, 1981b, 1983) studied larval development of cui-ui, mountain sucker [Catostomus })hityrliynclius), and Tahoe sucker {Catostomus tahocnsis ) and prepared a key for their identification. Miller and Smith (1967, 1981) discuss the paleohistory, systematics, distribution, evolution, and status of each spe- cies oi'Chasmistes . Donald R. Tuohy, Nevada State Museum, Carson Gity, has conducted extensive archae- ological studies within the Pyramid Lake re- gion; however, the data are largely unpub- lished. Archaeokjgical finds at Pyramid Lake are reported bv Ting (1967) and Tuohy and Clark (1979). Hattori (1982) studied the ar- chaeology of the Winnemucca Lake area and relates the importance of aquatic resources, including cui-ui, to human prehistoric habita- tion. The importance of the fishery, especially cui-ui, to the native Americans is discussed by Bath (1978). The ethnographic record of Pyra- mid Lake Northern Paiute fishing is pre- sented bv Fowler and Bath (1981). Follett (1963, 1974, 1977, 1980, 1982) has studied cui-ui remains in aboriginal deposits. Stewart (1941) discusses the culture element distribu- tions of the Northern Paiute. Procedures Cui-ui were captured with variable mesh bottom-set gill nets in Pyramid Lake and at the Marble Bluff facility on the Truckee River. Vigg (1981) presents a description offish sam- pling design and methodology. For age and growth data, fish were weighed to the nearest gram, measured (nearest mm), and sexed in- ternally, except at spawning time. Scales, op- ercula, otoliths, and fin rays were taken to compare accuracy of aging using different bony parts. The length-weight relationship is expressed by the formula W = aL*' (Sigler 1951), where W = weight (g), L = fork length (cm), and a and b are constants. The value of the constants (a and b) are calculated by the 576 Great Basin Naturalist Vol. 45, No. 4 method of least squares using log transforma- tions of weight and length (log W = log a + log b length). Validity of the aging method was determined by criteria suggested by Van Oosten (1923, 1929, 1944) and Hile (1941). To avoid possible bias, scales and other bony parts were first read without knowledge of the size of the fish. They were read at least three times. The length of body-bony part relation- ship was calculated according to Tesch (1971). The condition factor K =^ Wxl07L^ was calcu- lated according to Carlander (1969), where W = weight (g) and L = fork length (mm). Age and growth calculations were accomplished using a computer program (Nelson 1976). Cui-ui eggs and embryos used in this study were obtained from the David L. Koch Fish Hatchery. They were collected at regular in- tervals postfertilization and preserved in both Bouin's solution and Puckett's fixative. Serial sections of the entire embryo were cut at 8- 10 micrometers and stained with hematoxylin and eosin and Mallory's Triple Stain. Em- bryos to be sectioned were chosen from among the best preserved of 12- 15 specimens from each sample. In addition to sectioned material, whole mounts were also used, rang- ing in age from 9 to 912 hours post-hatching (Bres 1978). There were 19 water sampling stations lo- cated along 4 transects designed to represent the horizontal areas of the lake and to facilitate measuring the influence of the river upon the system. Stations were sampled on a monthly basis from November 1975 through October 1977. Conductivity, oxygen, pH, tempera- ture, and turbidity in relation to time, depth, and location were measured in the field with an InterOcean probe (Model 513D). Mea- surements were taken at 2-m intervals from the surface to 22 m and at 5-m intervals from 25 m to the bottom. Conductivity measiue- ments were standardized to 25 C. Water sam- ples were collected for analysis of major chem- icals and trace elements the third week of every month from January through Decem- ber 1976 and again in April and September 1977. Samples to be tested for nutrients were collected at least once a month from January 1976 through December 1977. Water sami)l('s were collected at the surface (Im), middeplh, and bottom levels at the midpoint of each ol three transects (Lider 1978). Analyses were done by the Desert Research Institute Water Chemistrv Laboratorv. R\NGE AND Distribution Four recent species of CJiasmistes are known: C. ciijits Cope, C. liurus Jordan, C. hrevirosths Cope, and C. miiriei Miller and Smith; the latter species, known from a single collection, is now extinct. Two additional ex- tinct species, C. bcitrachops Cope and C. spatulifer Miller and Smith, are known only from the fossil record. Miller and Smith (1981) discuss the distribution and evolution of the various forms oi'CJwsmistcs (Table 1). CJiasmistes is a lacustrine sucker; all living species and most extinct forms are associated with lake systems. However, the oldest known form, CJiasmistes sp. from the fluvial beds of the Miocene Deer Butte Formation in Oregon, is an exception (Miller and Smith 198 i). The Pyramid Lake cui-ui population is the last remaining pure species of the genus; the other species have considerable hybridization and introgression with Catostomus spp. (Miller and Smith 1981). Cui-ui inhabited Lake Lahontan during the late Pleistocene period (Fig. 3). At its maximum extent, ap- proximately 12,000 years BP, Lake Lahontan covered about 22,300 km" and received drainage from about 117,000 km" (Russell 1885). Fossil cui-ui have been discovered in the Carson Desert, which was once contained in the largest basin of Lake Lahontan; addi- tional CJiasmistes sp. fossils have been found in the Honey Lake basin to the northwest (Miller and Smith 1981). As Lake Lahontan desiccated during the last 10,000 years, its contiguous basin became nine remnant lakes. Cui-ui persisted for variable lengths of time in these remnant waters until desiccation caused extinction of most populations. Cui-ui was not present in Walker Lake during historical times. This idea is confirmed by the work of Spencer (1977) and Benson (1978a), which in- dicated Walker Lake was dry sometime dur- ing the period 9050 to 6400 years BP. During historic times cui-ui li\ed in both P\ lamid and Winnenuicca lakes and spawned in the Truckce Ri\ er as far upstream as just below Reno (Snyder 1918). When Derby Dam was completed in 1905, spawning cui-ui October 1985 SiCLER ET AL.: CUI-UI IN PYRAMID LAKE 577 Table 1. The geographic distribution of recent and Fossil species of C7(rt.s»ii.sfr.v (Miller and Smith 1981). Recent SPECIES Common name C'lii-ui June sucker Shortnose sucker Snake River sucker Scientific name C. ciijus Cope C. /lor^.v Jordan^ C.I. C. I. mictiis C. brevirostris Cope C. miiriei Miller and Smith Drainage basin Lahontan Bonneville Klamath River Snake River Present range Pyramid Lake, Nevada Utah Lake, Utah Upper Klamath Lake, Oregon Extinct" Fossil SPECIES Scientific name Chasmistes sp. C. spatulifer Miller & Smith Chasmistes sp. Chasmistes sp. Chasmistes sp. Chasmistes sp. Chasmistes sp. C. batrachops Cope Chasmistes cf C. batrachops C. batrachops Chasmistes cf C. liorus C. cujus C. brevirostris Geologic epoch Miocene Pliocene and Pleistocene-Recent Pliocene Pliocene Pliocene Pliocene Pliocene Pleistocene - Recent Pleistocene - Recent Pleistocene - Recent Pleistocene - Recent Pleistocene - Recent Pleistocene - Recent Geologic fo nna tion Deer Butte, OR Glenns Ferry, ID to Adrian, OR Glenns Ferry, ID Secret Valley, CA Honey Lake sediments Calcareous sands Teevimom, WY Fort Rock Basin, OR White Hills, CA Duck \'alley, NV Black Rock Can von, UT Pleistocene gravels, Fallon, NV Indian middens, Klamath Lake, OR Paleohabitat Fluvial Lake beds Lake beds Lake beds Lake Lahontan Mono Lake Fossil lake China Lake Pleistocene Lake Lake Bonneville Lake Lahontan Klamath Lake A. Catostomus fecundus Chasmistes liorus > Catostomiis aniens B. Based on a single collection from the Snake River below Jackson Lake Dam were restricted to the river below that point. As water was diverted to the NRIP via the Truckee Canal, the water level in Win- nemucca and Pyramid lakes dropped. VVin- nemucca Lake dried in 1938. Pyramid Lake and the affluent lower river is the only remain- ing habitat for cui-ui. Embryology Koch (1972, 1976) did limited work on the larval development of cui-ui, finding many similarities to the development of the white sucker, Catostomus commersoni, as de- scribed by Stewart (1926). Long and Ballard (1976) document the stages of embryonic de- velopment of the white sucker and cite diag- nostic structural characteristics for each stage. They also review previous work on embryol- ogy of other fishes within the order Cyprini- formes. Snyder (1983) found that sequences of developmental events are nearly equal for cui-ui, Tahoe sucker, and mountain sucker and typical at least for the tribe Catostomini. However, the latter two species, at any given size, are slightly more developed than cui-ui. The following is a detailed discussion of the embryological development of the cui-ui in a 13 C environment (Bres 1978). Egg-Embryos The unfertilized egg of the cui-ui is about 2 mm in diameter and is surrounded by a noncellular chorion. It has one micropyle at the animal pole. After fertilization, during a process known as water hardening, the eggs 578 Great Basin Naturalist Vol. 45, No. 4 Lake Lahontan > 10,000 years B. P. Remnant Lakes > 6400 years B.P. California Utah Arizona B. Pyramid Lake Winnemucca Lake Lower Truckee River WINNEMUCCA LAKE PYRAMID LAKE Sutcllffe Before 1938 Pyramid Lake to Marble Bluff Dam D. Mud Lake Slough Nixon PYRAMID LAKE Derby Dam RENO/SPARKS 1975- Present Fishway Marble Bluff Dam Fi^. 3. Dc'CR-ase in the niiiiie ol'llu' cvii-ui from Lake I.alioiiton times to the pr October 1985 SiCLER ET AL.; CUI-UI IN PYKAMID LaKE 579 imbibe water and swell to 3 mm. Koch (1976) recorded an 83% increase in egg volnme dur- ing water hardening, which took 39 minutes. Trelease (personal communication 1984) recorded 75% increase dining water harden- ing and a time of 60 to 75 minutes. The blastodisc appears at 6 hours postfertilization, 0.5 mm in diameter, and is elevated above the surface of the egg at the animal pole. By 18 hours postfertilization, 8 blastomeres are present, with an exponential increase in num- ber thereafter. After 19 hours, "giant" nuclei are seen associated with the syncytial cells of the yolk sac. The marginal periblast is at the periphery of the blastoderm. At 48 hours postfertilization, the first dis- tinction between the three germ layers is ap- parent. The neural plate has formed, along with a thickened precursor to the neural tube. The notochord and somites are present. At 96 to 120 hours, the neural tube and notochord are well developed. Myotomes have differentiated from somites, and the dor- sal fin fold has begun to develop. The pronephric ducts are formed anteriorally but are undifferentiated posteriorally. The gut has no lumen and is incomplete posteriorally, and the cloaca has not yet formed. Anterior neural crest migration occurs at 144 to 168 hours. The diencephalon exhibits cruciform shape. The optic vesicles have de- veloping lenses, and the opticoel joins the diocoel. Auditory vesicles are also present. Myoblasts the length of one somite can be seen. The coelomic cavity is developing be- tween the somatic and splanchnic mesoderm. At 192 hours cranial ganglia V, VII, and X are visible. Presumptive medulla is develop- ing, and the lateral ventricles are present. The pronephric duct has increased in length, and tubule development is beginning. The liver diverticulum and developing gut are visible. Vitelline circulation is well developed, and the dorsal aorta and postcardinal veins are visible. Precursors of the pigmented retina (a single layer of cells) and the neural retina are forming in the eye. At 13 C hatching occurs at 216 hours. Larvae At hatching cui-ui are white and threadlike in appearance, 6 to 7 mm in length, without Tabi.k 2. Time .sequence of eui-iii development at 13 A(;k Dkvelopmkntal stage (Hours postfertilization) 0 Unfertilized ovum 6 Formation of blastodisc 18 19 Eight-cell stage Earlv hlastula 48 Early neurulation, somites 96 to 120 192 present Neurulation complete, organ development begins Well-developed circulation. appearance of retinal pig- inent 216 Hatching (Hours post-hatching) 26 to 31 51 to 56 84 120 312 384 672 912 S-shaped heart 4 pairs of gill arches, optic chiasma forms, secondary reopening of gut begins 6 pairs of gill arches, recan- alization progresses to foregut Extensive nerve develop- ment, spinal cord differenti- ated, internal melanophore development Development of lateral line system and external melan- ophores Begin directional swimming Functional mesonephros, 5 functional gill arches Mouth open, eyes functional, first development of swim bladder Yolk completely absorbed, functional gut Fry actively feeding, ap- proaching adult body form functional vision, and have only limited pow- ers of locomotion (Table 2 and Fig. 4). Central Nervous System. — -The anterior curvature of the brain is noted 26 to 36 hours (post-hatch). Considerable nerve develop- ment has occurred by 72 hours. The cerebel- lum is still relatively small compared to the large medulla. The neural tube has differenti- ated into a spinal cord, and both gray and white matter are present. At 84 hours the potential neurohypophysis of the pituitary is developing in the brain. The III and IV ventri- cles are present, with the Aqueduct of Sylvius connecting them; the region of the epiphysis is also beginning to develop. Spinal ganglia 580 Great Basin Naturalist Vol. 45, No. 4 yolk blastomere Fig. 4a. i hours postfertilization otic vesicle telencephalon Fig. 4b. 9 hours post-hatching Fig. 4c. 4.5 days post-hatching October 1985 SiCLEK ET AL.: CUI-UI IN PVKAMID LaKE 581 r-^' ■Ti.\'.'''j^ ^yi^.'':. Fig. 4d. 21 days post-hatching Pig. 4e. 38 days post-hatching Fig. 4. Embryonic stages of development of the cui-ui are visible along the spinal cord. At 384 hours the epiphysis continues to develop. The pitu- itary and hypothalamus are visible, although no differentiation has occurred in the pitu- itary. Motor neurons are well developed in the mesencephalon. At 504 hours the devel- oping chondrocranium is visible. Eye . — The optic cup and retina continue to develop after hatching occurs. The optic chi- asma is first observed at 26 to 31 hours (post- hatch), with the optic nerve connected to the retina. The horseshoe-shaped retina, derived from the optic cup, is apparent at 51 to 56 hours. At this time the oculomotor nerve is visible, extending from the brain to the eye region. By 72 hours the lens is present and the pigmented retina is represented by a thin layer; however, no differentiation has oc- curred in the sensory portion of the retina. Presumptive cornea has formed by 84 hours, and differentiation in layers of sensory retina has occurred. The optic nerve is attached to the retina. Extrinsic ocular muscles are well developed. By 120 hours, heavy pigmentation has been laid down on the retina. After 384 hours the pigmented iris, cornea, lens, and many sensory layers of the retina are visible. The eyes are functional and capable of move- ment. Ear and Lateral Line . — Seventy-two hours post-hatching, the otic vesicle, the rudiment of the inner ear, begins to develop. The first complete distinction between the dorsal sac- culus and the central utriculus takes place in the otic vesicle at 84 hours post-hatching. At this time the first indication of lateral line system development occurs. By 168 hours the otic vesicle is well developed. After 384 hours of larval development, the otic capsule has divided into 3 parts, the latter part being com- pletely closed off. Otoliths are visible in the inner ear, and the cranial nerves that supply the ear are visible. The vestibular ganglia has developed outside the otic capsule from the stato-acoustic nerve (VIII). Olfactory Sense and Taste Buds. — By 20 hours the olfactory placodes are well devel- oped in the anterior portion of the head. The 582 Great Basin Naturalist Vol. 45, No. 4 neural connection of the nasal placode to the brain (olfactory nerve) is visible by 26 to 31 hours. By 168 hours the olfactory organ has developed from the nasal placode. By 384 hours indentations are forming at the site of the future external nares. The mouth is open, and developing taste buds are visible in the mouth and gills by 384 hours. These are very abundant on the head, mouth, and gills of adults and compensate for incomplete devel- opment of the internal nares. Gills. — Four pairs of gill arches are visible at 26 to 31 hours. The aortic arches leave the center of the gill arch to fuse together and open into the conus arterious. At 51 to 56 hours six pairs of gill arches are present and the gill cleft is developing. By 60 hours each of the six pairs of well-developed gill arches has a central core, the aortic arch. By 72 hours the aortic arch has increased substantially in size. At 84 hours the first gill cleft has opened. By 312 hours the 6 primitive gill arches have been reduced to 5 functional gill arches, the definitive adult condition. Each arch has at least 3 filaments composed of loops of capil- laries. After 384 hours of development, gill filaments are evident, as are gill cartilages associated with muscles for moving the gills. Heart. — The S-shaped heart is visible 9 hours post-hatching. After 20 hours the endo- cardial cushion, which is the precursor to valve development, is forming in the atrio- ventricular canal. Separation between endo- cardium and myocardium is pronounced by 56 hours post-hatching. The heart and associ- ated vessels are well developed by 72 hours. Cardiac jelly is visible after 82 hours. After 120 hours all 4 chambers of the heart and the atrio-ventricular canal are visible. After 312 hours the muscular wall of the heart is well developed and the ventricle has become tra- beculated. By 384 hours all blood vessels con- tain eosinophilic plasma. Muscle . — Myotomes and myocommata are well developed by 9 hours. By 26 to 31 hours connective tissue is present in the myocom- mata. At 72 hours myofibrils appear as ribbons around the periphery of the muscle cells; this conforms to the standard configuration of the adult fish. Skeleton. — After 20 hours the sites of the future chondrification of the ribs are visible as individual swellings along the dorso-lateral in- tersegmental myosepta. By 51 to 56 hours condensation is beginning to form the initial skeletal elements. The trabeculae of the chon- docranium are visible, although they are not true cartilage but simply condensations of the mesenchyme. After 312 hours a large number of caudal rays are present. At 384 hours carti- lage is present in the gill arches, opercula, and the roof of the mouth (precursor to palate). Liver and Pancreas. — The liver pri- mordium is well developed by 20 hours. At 26 to 31 hours the sinus venosus has been dis- placed to a crescent shape at the side of the liver. The liver primordium is well developed by 84 hours; a pancreatic rudiment is visible next to the intestinal swelling. The liver has an adult pattern of organization and is functional by 384 hours. The pancreas is forming lobules that will later spread out forming the adult diffuse pancreas. The gall bladder is visible; bile and pancreatic ducts are separate and fuse together at the entrance to the gut. Kidney. — At 9 hours the pronephric ducts join with the intestine posteriorally to form the cloaca. By 26 to 31 hours, ciliated nephrostomes, the opening of the kidney tubule to the coelom, have developed in the pronephros, and coelomic fluid is pumped into the tubule. After 72 hours of larval devel- opment, kidney tubules are well developed in the pronephros. For the first time, the mesonephros and mesonephric tubules are visible. At 84 hours the mesonephric duct is visible, opening into the mesonephros and contacting the cloaca. By 312 hours the mesonephros has greatly enlarged, is very well developed, and has reached a functional state. At 672 hours the mesonephric duct and anus empty together into the cloacal aperture. Alimentary Canal. — The pronephric ducts join with the intestine posteriorly to form the cloaca 9 hours post-hatching. The tiny, solid gut begins to form the loop of the intestinal swelling at about 20 hours. The larval cui-ui, like the adults, do not have a true stomach since it contains no glands. At 26 to 31 hours the secondarx reopening of the gut begins, small in the liver mass but enlarging in the midgut region posterior to the liver. Mesen- teries supporting the gut are visible. Absorp- ti\ e cells are apparent in the yolk sac, and the mouth cleft is present. Further recanalization of the Ibregut is occurring at 51 to 56 hours. At October 1985 SiCLER ET AL.: CUI-UI IN PYRAMID LaKE 583 72 hours there are many secondary openings in the foregut. Also the hnnen of the gut has greatly increased fiom 1 to 2 to 10 to 15 micrometers in diameter. At 84 hours the loops of the gut are beginning to form; early differentiation of the intestinal swelling and visceral cavity occurs. The pharyngeal cavity is open at 120 hours. After 384 hours the mouth is open, and many mucous-secreting cells are visible in the oral cavity. Material present in the pharyn.x suggests feeding, al- though some parts of the pharynx are still undifferentiated. The gut is broadly open and has developing longitudinal folds. From 384 to 504 hours the yolk sac is greatly diminishing in size. After 672 hours of larval development, the yolk is absent and the gut is functional, with food present in the intestine. By 840 hours the larvae are 20 to 25 mm long (Koch 1976). After 912 hours fry are actively feeding and the digestive tract is filled with food. Integument and Pigmentation. — By 20 hours lateral fin folds are well developed, and many mucous secreting cells are visible in the ectoderm. Connective tissue is present in the dermis of the skin at 26 to 31 hours. After 72 hours the epithelium is still simple, and many secretory cells are present. Melanophore de- velopment is beginning internally. At 84 hours goblet cells are observed in the epithe- lium. Granular cells, filled with eosinophilic granules, are present, characteristic of the adult condition. Both small and large external melanophores are visible by 120 hours post- hatching. At 384 hours mucous-secreting gob- let cells are present in the skin. Swimming . — After 18 hours the larvae are 8 to 9 mm long, and sudden bursts of energy constitute their initial swimming attempts; at 192 to 240 hours the larvae are 12 to 14 mm in length and continually swim at the surface (Koch 1976). Between 240 to 360 hours they swim to keep their position in the water column (Koch 1976). After 384 hours the pneumatic duct enters the gut from the devel- oping swim bladder, and at 504 hours the swim bladder is clearly visible. The swim bladder has increased in size during 672 hours. Identification. — Larval and juvenile cui-ui are sometimes difficult to identify in Pyramid Lake; they are easily confused with another resident catostomid, the Tahoe sucker. This may, in part, account for the fact that rela- tively few cui-ui less than 300 mm in length have been identified. Ramsey (letter to E. A. Pyle, 16 September, 1974) offers the following points of contrast between the two larvae: Ventral-Pi^incntation: A consistent character for dis- tinguishine; lar\ al stages of Tahoe sucker from larval cui-ui is the presence of a superficial row of melanophores on the midventral skin posterior to the pectoral basis. This ab- dominal pigmentation is generally absent in cui-ui, al- though a row of melanophores sometimes is present but confined to the breast anterior to the pectoral bases. The row of midventral melanophores in larval Tahoe suckers is still present at age 66 days (17 to 19 mm total length). Intestinal Coiling: At age 66 days the intestine of the Tahoe sucker loops far anterior in contrast to the cui-ui, where it is either straight or has a left twist. Mouth : The lips of the Tahoe sucker are thicker and the mouth is placed further ventrally than in the cui-ui. Other: A character sometimes useful at ages earlier than 66 days is the presence in cui-ui of a depigmented "one to one" on top of the head, just posterior to the eyes. There is considerable occluding of this pigmentation by age 66 days. Snyder (1981a, 1981b, 1983) studying larval development of cui-ui in comparison to the other catostomids that spawn in the Truckee River system, i.e., Tahoe sucker and moun- tain sucker, developed a taxonomic key that separates the larvae and early juveniles of the three species. Snyder concludes the larvae can be separated on the basis of midventral pigmentation, peritoneal pigmentation, gut- loop formation, and mouth characters. The following differential characteristics are included to complement previous descrip- tions of larval development and morphology (Snyder 1983). At a total length (TL) of 11 to 21 mm, cui-ui are characterized by absence of midventral melanophores on the head or ab- domen anterior to the bases of pelvic fin or their precursors and anterior to the vent. If midventral melanophores exist, they are present as a short line only in the branchial and heart regions between and anterior to pectoral fin bases. Mesolarvae have a straight gut until about 19 mm TL; metalarvae to 21 mm may develop a primary loop extending foi-ward less than two-thirds of the length of the stomach and not crossing over the stom- ach. Metalarvae have peritoneal pigmenta- tion largely restricted to the dorsal and dor- salateral visceral cavity. The following characteristics apply to meta- larvae > 21 mm and juveniles < 50 mm. The pigmentation of the peritoneum is mostly lim- 584 Great Basin Naturalist Vol. 45, No. 4 Fig. 5. Adult female cui-ui. Photo by Thomas J. Trelease. ited to the dorsal and dorsalateral visceral cav- ity. The primary loop of the gut is relatively straight along the left side of the stomach until about 30 mm TL, at which size secondary loops cross the stomach in an S -shape, persist- ing through 50 mm TL. The mouth is termi- nal— usually slightly oblique but sometimes very low and almost horizontal, approaching a subterminal condition. Adult Morphology Description The cui-ui is a large, big-mouthed sucker. The head is wide and somewhat roimd in cross-section. Its interorbital space is greater than half the length of the head. The mouth is unsuckerlike with a ventro-terminal position. The lips are thin and obscurely papillose. The lower lip is somewhat pendant and divided by a wide median notch. The cui-ui is coarsely scaled, with counts of 13 to 14 above the lat- eral line, 59 to 66 ak)ng the lateral series and 22 to 26 around the caudal peduncle. The total body length is 9 times that of the dorsal fin base. The length of the anal fin, from the insertion to the tip, is about one sixth the total body length. Fin ray counts are: dorsal, 10 to 12; anal, 7; and caudal, 8 or less. The caudal is weak to moderately forked. The caudal pe- duncle is thick, with the smallest depth going 12 times into standard body length (SL). In triangular section, the pharyngeal teeth are delicate. The last pharyngeal arch bears a row of more than 10 comblike teeth confined to a single row. The swim bladder is 2-celled; the peritoneum is nearly black. Each gill raker is branched like broccoli (Fig. 5). Sexual Dimorphism Breeding males display a brilliant red to brassy color on the sides; in general they are black or brown above, fading into flat white below. Females have a bluish gray cast year- round. Female cui-ui attain greater length and heavier weight than males. During the spawning season the vent of females becomes swollen and extended, whereas males de- velop nuptial tubercles on their fins. Appar- ent sexual dimorphism exists in the meristics associated with fin size (Table 3). The length of the base of the dorsal and anal fins, the height of the dorsal and anal fins, and the length of the pectoral, peKic, and caudal fins are all proportionally greater for males. Snyder (1918) refers to difFerences between the sexes: Tlu' females are more stocky than the males, and with their huge heads, large rounded bodies, and relati\el\ October 1985 SiGLER ET AL.: CUI-UI IN PYRAMID LaKE 585 Table 3. Meristics ofChasmistes cujtis from near the mouth of the Truckee River (Snyder 1918). Mean me asurement Morphological Males Females characteristic n=ll n ^^ 7 Standard length (mm) *** 427.1 487.3 range (410-444) (445-538) Percent of body length Length head 28.0 27.8 Depth body * 21.1 22.4 Depth caudal peduncle 8.5 8.2 Length caudal peduncle 15.8 15.2 Length snout 12.9 13.2 Diameter eye 3.1 2.9 Interorbital width 12.4 12.5 Depth head 18.7 18.6 Snout to occiput 22.3 22.0 Snout to dorsal 51.3 50.5 Snout to ventral 58.1 58.4 Length base of dorsal *** 15.1 13.3 Length base of anal *** 9.4 8.1 Height dorsal 12.9 12.5 Height anal*** 19.9 15.6 Length pectoral ** 18.9 17.5 Length pelvic *** 13.8 11.6 Length caudal *** 20.4 18.4 Dorsal rays 11.1 10.7 Anal rays 7.2 7.0 Scales lateral line 62.1 61.6 Scales above lateral line 13.6 13.9 Scales below lateral line 10.4 10.1 Scales before dorsal 31.6 30.3 * Significant differences between sexes, P<0.05. ** P<.01 *•* P<.001 short fins are very ungainly looking fish. The scales and fins are without tubercles. Snyder (1918) describes the differential col- oration patterns between the sexes. He also reports that Indians could differentiate cui-ui from Pyramid and Winnemucca lakes by the grayer color of the Winnemucca cui-ui, al- though he was unable to detect any differ- ence. Comparative Morphology The ventro-terminal position of the mouth is a diagnostic characteristic of Chasmistes spp. It is so exceptional among the usually ventral- mouthed sucker family that it has been re- garded as an extreme specialization; however, certain primitive suckers (e.g., Amyzon and Ictiobus cyprinellus) and presumed sucker ancestors are also characterized by relatively terminal mouths (Miller and Smith 1981). Cui-ui is the largest living species of Chas- mistes. Snyder (1918) collected specimens ranging from 410 to 670 mm in SL. Of various adult meristic data summarized from the literature, Snyder (1983) determined that lateral series scale counts prove to be diagnostic in separating cui-ui (59 to 66) from mountain sucker (75 to 100) and Tahoe sucker (79 to 95). For juveniles, Snyder found this character useful only when squamation is complete, usually by 35 to 50 mm TL. The comparative morphology of Catostomi- dae has been studied with reference to the swim bladder (Nelson 1961), the opercular series (Nelson 1949), the Weberian apparatus (Nelson 1948), and the brain and lips (Miller and Evans 1965). Chasmistes spp. have a two- chambered swim bladder that is characteristic of all catostomids except Moxostoma , which has a three-chambered structure. It is the posterior chamber in catostomids that regu- lates buovancy. The usual catostomid swim bladder is 35 to 45% of the SL of the fish (7% by volume); however, the cui-ui swim bladder is only 32. 1% of SL (Nelson 1961). Nelson (1949) presents a generalized com- posite of the catostomid opercular series, which consists of a large operculum, relatively small suboperculum and interoperculum, and invariably three branchiostegal rays. On the basis of the opercular series, the genera of Catostomidae can be arranged into three well- defined groups; Chasmistes belongs to the group including Catostomus and Xyratichen (Nelson 1949). The Weberian apparatus of catostomids in- cludes the first four vertebrae and associated structures that form two separate functional units. Chasmistes has the same general mor- phological pattern as Catostomus and Xyrauchen; however, it differs in having en- larged esophageal supports and obliterated second to third intervertebral space (Nelson 1948). Based on the comparative morphology of the Weberian apparatus. Nelson concludes Chasmistes is an early divergent of the catostomid stock. Miller and Evans (1965), studying the mor- phology of the brain and lips in catostomids, conclude: Their principal value probably lies in providing a basis for making inferences about the life history, and espe- cially the habitat preferences and feeding behavior of little-known species. Thus, morphological evidence may shed light on aspects of the ecology of cui-ui about which there has been much speculation. The 586 Great Basin Naturalist Vol. 45, No. 4 facial lobe of the brain is associated with taste buds on the lips and skin, whereas the vagal lobes receive fibers from taste buds in the mouth and pharynx. The brain morphology of cui-ui is unique in several ways: the optic lobes are small and separated, the postcere- bellar medulla is elongated, and the vagal lobes are well developed but located more posteriorly than is usual in catostomids. The overall pattern suggests a well-developed "mouth tasting" apparatus (Miller and Evans 1965). Suckers that have large vagal lobes are characteristic of lotic habitats, and mouth tasters probably sort food within the oral cav- ity. Thus the cui-ui is probably not a sight- feeder in surface waters but may use the oral cavity to sort out food (e.g., algae and inverte- brates). Other genera with well-developed vagal lobes include Xyrauchen, Ictiohus, and Carpiodes. Age AND Growth The cui-ui is a slow-growing, long-lived fish, living 18 or more years (Robertson 1979). Scoppettone (report to Desert Fishes Council 1983) stated it may live much longer (> 40). Growth in length is rapid for the first 4 to 5 years and slower thereafter. Annuli in older fish are formed between June and August; in younger fish it may occur the first week of June. Back-calculated fork length (FL) at scale for- mation is 46.0 mm for known age fish (1 to 111), from the NFG Washoe Rearing Station, Reno. The calculated FL was skewed substan- tially higher when advanced age groups (IV and VI) from Pyramid Lake were added. In aquarium-reared fish, E. Pyle (personal com- munication 1977) found they started forming scales at 49.0 mm FL, and fish 50 mm had from 3 to 7 scales at the base of the caudal peduncle. Scales are judged not to be reliable for aging cui-ui older than age VI. Other bony parts, otolith, opercula, and fin ray, are more nearly reliable. No technique is reliable when there is no, or almost no, growth and no discernible annulus. This is a definite possibility in older cui-ui. There is reasonably good agreement between fin rays and otoliths and excellent agreement l)etwcen otoliths and operculum through age XIII (Table 4). There was gener- £ 580 E T 560 14 15 16 17 18 Annulus Fig. 6. Absolute growth rates using four methods of age assignment (scales, fin rays, operculum, and otoliths) for cui-ui collected from Pyramid Lake, Nevada, 1978 (Robertson 1979). ally good agreement for otolith, opercula, and scale in age I to IV. Data from opercula were chosen because it is reliable and easy to collect and process. Since growth differences were not significant, sexes are combined (Table 5). The absolute growth is in good agreement with calculated growth (Fig. 6). The body fork length-opercula (X) relationship, sexes com- bined, is: FL - 229.2 + 7.0 x (r = 0.92). The body length-bony part radius regres- sions are highly correlated: fin ray (r^ = 0.93), opercula (r" = 0.92), otohth (r" = 0.80), and scales (r' - 0.63) (Robertson 1979). The drop in niunbers of fish older than age XV may be attributed largely to natural mor- tality or no growth, but the low numbers of fish in some ot the younger age groups are, in part, a result of moderate to weak year classes (Table 6). Sonnevil (1978) suggests reduced spawning populations and consecjuent weak year classes can be attributed to reduced river flows at the time cui-ui spawn. There appears to be good correlation be- tween strength of year classes and flow levels of the river for 5 of the 12 years and poor correlation for 3 of the 12 years (Table 7). October 1985 SiCLER ET AL.: CUI-UI IN PYRAMID LaKE 587 Table 4. Comparison of assigned age by various aging methods (or 28 c 1978 from Pyramid Lake, Nevada. sexes combined. Collected April to July Methods of age assignment Fork length (mm) at Scale Fin ray Operculum Otolith Centrum ID# capture age age age age age 32A 599 8 10 11 10 33A 587 8 11 11 11 36A 612 8 12 13 13 17A 638 8 13 13 14 MB 584 8 — 11 11 KiB 615 8 11 11 11 YY-3 565 8 11 — 12 — 24A 604 8 11 11 11 17 578 8 12 12 12 24 633 8 15 14 15 31 607 8 11 12 12 33 618 8 11 12 12 9B 588 8 — 11 11 2()A 618 8 — 14 13 — 21A 615 8 — 12 12 28A 591 8 12 12 12 — 8B 588 8 7 12 12 4B 620 8 — 13 13 27A 648 9 14 14 14 — 38A 632 9 12 13 13 19A 609 9 — 11 12 — AA 610 9 12 12 F 573 9 — 10 11 10 E 638 9 — 16 16 — 2B 598 — — 10 10 ZZ-B 575 — — 11 — 3B 601 12 12 23 632 — 13 13 13 — Table 5. Calculated average fork length and annual growth increments using opercula of 79 cui-ui, sexes combined. Collected April 1978 to July 1978 from Pyramid Lake, Nevada (Robertson 1979). Age Number of fish Mean calculat( ed fish 1 length ( FL-mm) at each annu ilus group 5 6 7 8 9 10 11 12 13 14 15 16 17 V 13 466 VI 12 465 490 \TI 3 453 481 499 .\ 6 451 470 485 511 527 468 .\1 10 455 477 499 518 539 559 576 .\II 17 453 471 494 512 530 547 563 583 .XIll 10 447 468 492 511 523 546 562 579 596 XIV 4 455 478 493 515 531 547 566 580 594 611 XV 1 453 474 495 508 522 529 536 553 564 578 599 XVI 1 418 440 460 484 508 522 543 564 578 585 599 613 XVII 2 440 457 471 488 507 524 540 550 564 583 596 610 624 Total number 79 66 54 51 51 51 45 35 18 8 4 3 2 Grand , average (mm) 459 475 493 512 530 548 564 578 589 597 598 611 624 calcii dated fork length Range 1 [mm) in 418- 439- 467- 484- ■ 484- 495- 515- 522- 536- 553- 564- 578- 592- calcu dated length 488 529 536 533 571 592 606 634 634 620 627 641 655 Length 1 increments 459 16 18 19 18 18 16 14 11 8 1 13 13 (mm ) 588 Great Basin Naturalist Vol. 45, No. 4 Table 6. Age and year class composition of 665 cui-ui sampled in Pyramid Lake, Nevada, 1978 (Robertson 1979). Age class Year class % Composition Number offish IV 1974 6.0 37 V 1973 32.0 211 VI 1972 22.0 149 VII 1971 7.0 49 VIII 1970 5.0 34 IX 1969 3.0 18 X 1968 2.0 16 XI 1967 2.0 14 XII 1966 4.0 27 XIII 1965 4.0 28 XIV 1964 3.0 21 XV 1963 6.0 41 XVI 1962 2.0 13 XVII 1961 0.9 6 XVIII 1960 0.2 1 There appears to be good correlation between strength of year classes and flow levels of the river for 5 of the 12 years and poor correlation for 3 of the 12 years (Table 7). Robertson (1979) determined the length- weight relationship for 139 females, ranging from 453 to 653 mm FL, and for 147 males, ranging from 448 to 577 mm FL. Only spawned fish were included in these data. The relationships for males and females are: males logio W = 3.4725 + 2.4639 (log,,, L) (r' = 0.77); females log,,, W = 4.5046 + 2.8485 (log,oL)(r' = 0.93). Males weighed less than females of equivalent length and age. This is in agree- ment with Johnson (1958) and Koch (1972). The length-weight relationship in the 1975 to 1977 Pyramid Lake studv (Robertson and Koch 1978) was: log,,, W =' -1.240 + 2.5738 (log,,, L); this is in agreement with work done by Robertson (1979) in 1978-1979. The condition factor, or general robustness of the fish, K(FL) for spent (spawned out) fish ranged between 1.08 and 1.64, with a mean of Table 7. Discharge in cubic feet per second (cfs), of the Truckee River near Nixon, Nevada, for the calendar years 1962-1973 (USGS 1962-1973) and year class-levels of flow relationship. Strength Level of Age group Mean flows (cfs) by month of year class* river Year April May June J'lly flows* 1962 XVI 315.0 229.0 31.3 18.8 1 3 1963 XV 183.0 1391.0 926.0 53.7 1 2 1964 XIV 53.4 93.2 48.3 26.2 3 3 1965 XIII 580.0 1325.0 515.0 62.2 2 1 1966 XII 64.6 61.4 47.0 33.2 2 1 1967 XI 5.9 7.5 11.4 31.7 3 3 1968 X 321.0 67.9 52.4 38.8 3 3 1969 IX 3392.0 3454.0 3469.0 430.0 3 1 1970 VIII 530.0 212.0 291.0 445.0 2 2 1971 VII 770.0 1234.0 1744.0 451.0 2 1 1972 VI 236.0 249.0 110.0 43.3 1 3 1973 V 8.54.0 991.0 4.53.0 .321.0 1 1 ♦Rated on a 1 to 3 seal Although 1969 was a high water year, a weak year class resulted. According to Robertson (1979), this appears to contradict the hypothe- sis of high flows and successful cui-ui spawn- ing; however, it is pointed out this was the year of exceptionally high suspended sedi- ment discharges, which may have been lethal to fertilized eggs. Koch (1972) reports the mean age of spawn- ing cui-ui in 1971 and 1972 as 7.5 and 7 years, respectively. These represent the strong year classes produced in high water years of 1963 and 1965. Koch also found a low number of fish representing ages IV, V, and VI in 1971 and 1972. 1.21 for females and 0.81 to 1.61 and a mean of 1.20 for males (Robertson 1979). The condi- tion factor decreases moderately with an in- crease in length, and the decrease is higher for males than females. Condition factors for the 1976 to 1977 studv showed similar trends (Table 8). Food AND Feeding The diet of cui-ui is not well known; how- ever, we made some observations under arti- ficial conditions. Koch (1972, 1976) reports larval cui-ui, older than 20 days, readily con- sume zooplankton introduced into an aquar- October 1985 SiCLER ET AL.: CUI-UI IN PYRAMID LaKE 589 Tables. Length, weight, and concUtion iiictors K 10' FL for cui-ui, sexes combined, Pyramid Lake, Ne- vada, 1976 to 1977 (Robertson and Koch 1978). Fork length (mm) Weight (g) K factor 378 566 1.04 401 659 1.02 424 761 0.99 447 872 0.97 472 1002 0.95 495 113.3 0.93 518 1273 0.91 544 1444 0.89 589 1771 0.S6 ium of lake water. He reports the zooplankter Moina hutchinsoni is most preferred by lar- vae, presumably because it has limited mobil- ity, whereas Diaptomus sicilis is the least pre- ferred zooplankter (Fig. 7). We have observed aquaria-reared larvae and juvenile cui-ui graz- ing on periphyton growing on rocks. Since zooplankton is not abundant in the river habi- tat of larval cui-ui, periphyton is probably im- portant in their diet. When young cui-ui first enter the lake, they may feed both on peri- phyton and zooplankton. In the David L. Koch Hatchery larval and juvenile cui-ui feed on algae on the sides of the tanks, as well as on commercial fish feed (A. Ruger personal com- munication 1983). Snyder (1918) found spawning adults do not feed; he states: "The stomachs of all specimens examined were devoid of food." Koch (1972) reports cui-ui examined during spawning mi- grations of 1971 and 1972 had not recently fed. Examination of fish during the spawning mi- gration at the Marble Bluff facility also con- firm these observations. Johnson (1958) reports that, of 46 adult cui- ui examined, 43 had eaten zooplankton (93.5% occurrence), 4 sand and mud (8.7% occurrence), 2 unidentified material (4.3% oc- currence), and 1 insects (2.1% occurrence). La Rivers (1962) reports T. J. Trelease exam- ined specimens taken in commercial net hauls and found a mixture of algal filaments with zooplankton fragments. From this informa- tion La Rivers (1962) concludes, "It seems probable that most of the feeding is done about rocks where thick algae coatings are heavily populated with micro-crustacea." Based on the cui-ui's fine and numerous gill rakers. La Rivers hypothesized, "The strong Fig. 7. An adult copepod, Diaptomus sicilis, a com- mon food of cui-iii. possibility exists that they can extract useable quantities of micro-crustacea from the open lake waters." T. J. Trelease (personal commu- nication 1984) observed cui-ui in large dough- nut-shaped schools near the surface over deep water and far from shore first in 1954. He assumed they were feeding since tui chub form similarly shaped schools when they are feeding. He saw these schools somewhat fre- quently as late as 1968. Vigg (1978a, 1980) documented that adult cui-ui primarily in- habit the shallow benthic areas and not the limnetic water column. It may be that when they inhabit the benthic zone they generally feed further off the bottom than species of suckers with ventral mouths. Reproduction Migration Snyder (1918) and Scoppettone et al. (1981, 1983) made detailed observations on cui-ui spawning migrations. Snyder observed the annual cui-ui spawning run begins about April 15, depending on the condition of the river. La Rivers (1962) states it is about a month later than in Snyder's time as a result of river condi- tions: it extends from mid-May to early June. However, Trelease (1971) reports cui-ui may spawn as early as April and as late as July, when a surge of fresh water often triggers the spawning run. The cui-ui apparently homes fresh water, including springs. Scoppettone et al. (1981) also found a sudden heavy surge of very turbulent water often triggered spawn- ing activities, even in the daytime. The cui-ui prefers depths of water for spawning that 590 Great Basin Naturalist Vol. 45, No. 4 range from 9 to 43 cm, velocities that range from 23 to 87 cm/sec, and substrate with about 60% gravel. Historically, cui-ui spawning runs up the Truckee River only occasionally reached downtown Reno, a distance of over 100 km (Snyder 1918). Today they generally run no farther upstream than 15 to 20 km, although they can go further. Koch and Contreras (1972) report spawn-laden cui-ui reach ex- haustion in 18, 10, 2, and 0.5 hours at veloc- ities of 1.2, 1.8, 4.6 and 5.2 m/s, respectively. Spawning Behavior Spawning cui-ui often choose the head of gravel bars, where the flow is rapid and the substrate relatively free of silt (McGarvey 1974). At times the dorsal fins of the cui-ui project above the water, and in very shallow places, where there is much crowding, the entire backs of the fish are exposed (La Rivers 1962). Trelease (1971) notes the numbers of cui-ui at the mouth of the Truckee River in past years were so immense at spawning time that fish near the surface were literally forced out of the water, and during periods of peak activity schools offish covering 0.4 ha or more would form a mass of writhing fish on the surface of the water. Some runs of cui-ui were so extensive that, as fish worked their way upstream in dense schools, their numbers ac- tually blocked the flow of water and diverted it around them. As a result, a new channel was sometimes cut through the sandy delta, leav- ing large numbers offish stranded. Migrating and spawning cui-ui are more ac- tive at night than in daytime (Snyder 1918). Scoppettone et al. (1983) found that peak spawning occurs between the hours 2000 and 0600 over a .3-day period and postulate that nocturnal spawning lessens egg predation. Adhesive eggs are broadcast over a large area (Koch 1973). One spawning act, lasting from 3 to 6 seconds, is participated in by 1, or occa- sionally 2, females and from 2 to 4 males; although a typical spawning act has 1 female and 2 males. Scoppettone et al. (1983) found the most active male spawned 294 times, the most active female 114 times. The length of the spawning run for individual males was 3 to 5 days, for females 2.5 to 4 days. Just prior to spawning, two males position themselves on either side of a female, the heads of the males just aft of the female's head. With bodies touching and quivering, the fe- male deposits eggs, followed by the males expelling sperm. The cupping and vibration of the male's caudal, along with the female's cau- dal, creates an eddy preventing the eggs from drifting away before they are fertilized (Scop- pettone et al. 1983). Although the cui-ui does not build a nest, the fanning of the caudal fins serves to clear the area of silt. Optimum Hatching Temperature In an 8-day period when temperatures ranged from 13.8 to 20.8 C, with a mean of 16.7 C, mean viability of the embrvos was 47% (Scoppettone et al. 1981). Koch (1981) found 13.9 C optimum for cui-ui egg incuba- tion; embryos incubated at 17.8 C had a 60% survival to hatching, whereas embryos incu- bated at 21.7 C had a 30% survival. High temperatures cause preemergence of larva, and a lower rate of survival (Lockheed Ocean Sciences Laboratories 1982). Larval Migration Larval peak downstream migration is 14 or more days after hatching (Scoppettone et al. 1983). Hatchery-reared larval cui-ui, 15 to 18 days old, released in 3 areas of the Truckee River, began migrating downstream immedi- ately. The peak migration occurred the night of release followed by several days' lull. All three groups showed a tendency for immedi- ate outmigration (Scoppettone et al. 1981). It should be noted that our embryological stud- ies show that larvae are not developed well enough to feed or swim actively before 21 to 18 days at 13.6 C. It may be that early migra- tions (< 28 days) greatly reduce chances of survival. Lake Spawning There are several reports of cui-ui spawning in the freshwater-lake saline interface. Snyder (1918) reports, "On May 1, 1913 large numbers of cui-ui were found depositing eggs along the shallows near some springs on the southwest shore. " Johnson (1958) observed ready-to-spawn cui-ui aroimd the periphery of the lake. Koch (1973) documented the spawning behavior of cui-ui near the inflow of freshwater springs (0.014 cms) in 17.3 C lake October 1985 SiGLER ET AL.: CUI-UI IN PYRAMID LAKE 591 Table 9. Number of cui-ui eggs taken at the Marble Bluil' facility, 1978-1983 (Source; Alan Rnger, Pyramid Lake Fisheries director). Number ofspawners Male Female 188 226 112 92 333 320 166 158 422 436 184 244 Number of eggs Eggs per female 1978 1979 1980 1981 1982 1983 4,838,660 21,410 2,706,308 29,416 12,140,480 37,939 5,437,886 34,417 17,707,268 40,613 13,706,700 56,175 Totals and weighted average 1,476 56,537,197 38,304 water in the Hell's Kitchen area, which is approximately 29.5 km north of the Truckee River. Most biologists agree that cui-ui spawn in the lake (> 5000 mg/1 TDS) as well as in the river (< 600 mg/1 TDS), but the success of the lake spawning is not known. Observed lake spawning has been in the vicinity of fresh water, e.g., springs or stream-lake interfaces (Koch 1973). T. J. Trelease (personal commu- nication 1984) lists seven places around the lake where he observed cui-ui spawning over the years. Experiments by LOSL (1982) and Chatto (1979) indicate that eggs must be water hardened in fresh water (< 600 mg/1 TDS), or they will either not hatch or the larvae will not survive in lake water. The issue then revolves around the question. Is there enough fresh water in these lake microhabitats for cui-ui eggs to produce healthy larvae? This is a diflfi- cult question. According to Chatto (1979), from 2 to 3 days are required if hatching is to be successful.' LOSL (1982) declare freshly fertihzed eggs are intolerant of 5,897 mg/1 of TDS, but within a day the embryos have ac- quired considerable resistance. Fecundity Koch (1972) found cui-ui become sexually mature in their fifth or sixth year and produce 20,000 to 30,000 eggs per year. Frazier and Ferjancic (1977) estimated the average-sized female produced 35,700 eggs. Mean number of eggs per female taken at the Marble Bluff facility from 1978 to 1983 was 36,662 (Table 9). This is a nearly linear increase in the number of eggs/female. This may be due to increased efficiency in egg-taking caused by such factors as riper females, better water conditions, bet- ter fish-holding facilities, increased use of hor- mone injections, and/or increased experience of workers (A. Ruger personal communication 1983). The 1983 value of over 55,000 eggs/fe- male may be more indicative of the actual mean fecundity of the species than lower esti- mates. The 1983 run was quite different from previous years in timing and size of females. Possibly more large females increased the av- erage number of eggs taken. It is understood that, in the wild, a female would have to spawn several times to reach this number, and realistically this may not happen. Morbidity' and Mortality Large, long-lived cypriniform fishes (such as cui-ui) with relatively small eggs and high fe- cundity usually experience extremely high mortality rates in their early life stages. At present recruitment is derived from both arti- ficially and naturally reared cui-ui. The hatch- ing success is moderate in the PLF operation, e.g., 75.2% in 1983 (A. Ruger personal com- munication 1984). Cui-ui mortality can be divided into five stages: spawning adults, eggs (embryos), lar- vae, juveniles, and maturing adults. Each stage has differing levels of vulnerability and causes of death. Spawning adults are ad- versely affected by low flows, high tempera- tures, and predation. Egg mortality is affected by condition ofspawners, high temperatures, and silt. Larvae survival is determined by a complex of factors during their early life in the river, including temperature, flow, food availability, parasites, disease, and predation. In the lake the mortality of juvenile cui-ui is determined by food availability, salinity change, competition, and predation. Non- spawning adults are subject only to mortality factors in the lake environment. Predation 592 Great Basin Naturalist Vol. 45, No. 4 there is minimal; therefore, if food supplies are adequate, then parasites, disease, and se- nility are probably the most significant ad- verse factors. Egg and Larvae Mortality If the Truckee River spawning habitat were optimal, one would expect high hatching suc- cess from the river-spawning cui-ui. How- ever, using the fecundity of 35,700 eggs per female estimated by Frazier and Ferjancic (1977), Scoppettone et al. (1981) projected that if 21 females deposited 750,000 eggs only 20,000 larvae would be produced. Their esti- mated survival rate to emergence is 2.7%, this was attributed to high temperatures, poor egg viability, and predation by Lahontan red- sides, Richardsonius egregius . Adult Mortality The highest adult mortality probably occurs during the spawning season, when cui-ui are most vulnerable to predation. Historically fishing mortality may or may not have been significant; it continued at low levels, as a snag fishery of spawners on and near the Truckee River Delta, until recent years. Since 1979 all fishing for cui-ui, even by tribal members, is prohibited. Death of adults as a result of spawning, as well as handling mortality dur- ing and following egg taking at the Marble Bluff and the PLF facilities occurs at unknown levels. Snyder (1918) reports a few dead indi- viduals along the Truckee River after each spawning season, and high mortality regularly occurred at the mouth of Winnemucca Lake. Fish-eating birds, primarily white pelicans, Pelecanus erythrorhynchos , double-crested cormorants, Phalacrocorax auritus , and Cali- fornia gulls, Larus californicus , can wound or kill adult cui-ui. Although large numbers of white pelicans and cormorants were observed on the Truckee River Delta during the 1976 and 1977 cui-ui spawning migration, Knopf and Kennedy (1980) found no evidence that these birds fed on cui-ui. Common carp, Cyprinus carpio, and tui chub, Gila hicolor, composed over 97% of the diet of the pelican. T. J. Trelease (personal communication 1984) states he has observed pelicans catch and swallow adult cui-ui. The pelicans then had great difficulty taking off' with so heavy a load. He also states he has seen several, but not a great many, cui-ui remains on Anahoe Island. He believes the major damage done by birds is pecking out eyes and gills. Pelicans preyed on adult cui-ui during the large run of 1982 (M. LaFever personal communication 1983). This phenomenon was also observed by D. L. Galat in recent years (personal communica- tion 1984). Snyder (1918) reports that when cui-ui migrate in dense schools considerable numbers are crowded into shallow water and even stranded out of water on sand bars: Cormorants, gulls, and pelicans in great numbers were attacking them, and many still wriggling fishes had lost their eyes and strips of flesh had been torn from their sides. Disease Pathological studies of the wild cui-ui popu- lations have not been conducted; therefore, the effect of internal and external parasites, fungal infestation, and viral and bacterial dis- ease is unknown. Effects ofTDS on Eggs, Larvae, and Juveniles Bioassay tests conducted by LOSE (1982) demonstrate the intolerance of fertilized and/ or water-hardened cui-ui eggs to TDS concen- trations above 525 mg/1. Embryos placed in 525 mg/1 water (i.e., Truckee River water) for 24-96 hours survived when transferred to Pyramid Lake water (5897 mg/1), although some abnormalities were found. Embryos placed in 5897 mg/1 water immediately after fertilization in 525 mg/1 water, were atypical within 24 hours. An average 90% mortality occurred in the 5897 mg/1 TDS concentration by the third day, and an average of only 8.3% of the embryos in this concentration produced apparently normal fry. One-day-old cui-ui larvae placed in test con- centrations of either 5781 or 3503 mg/1 showed differential mortality; 20% and 13.3% of the test fish died in the respective concen- trations within 72 hours. Three day old cui-ui larvae placed in test concentrations of 350 and 5781 mg/1 had 100% survival in the first 96 hours. After 192 hours there was no mortality in the 350 mg/1 level, but the 5781 mg/1 level had 7% mortality and an additional 8% abnor- malities. Chronic 180-day tests indicate that reduced sinvival of juvenile cui-ui, across a broad range of TDS levels extending from 3620 to October 1985 SiCLER ET AL.: Cui-UI IN PYRAMID LaKE 593 5225 mg/1, represents only 33% to 48% of the 96-hour median tolerance Hmit (LC50). This indicates that, although LC50 tests may show acute toxicity resulting in death only at high TDS levels, lower TDS levels may cause death or abnormalities when fish are exposed for extended periods of time (LOSL 1982). Habitat AND Ecology Physical At an elevation of 1154.9 m above mean sea level. Pyramid Lake is approximately 40.8 km long and from 5.8 to 17.3 km wide, with a north-south axis (Fig. 1). At this elevation it has a surface area of 437 km", a volume of 25.3 km^, a mean depth of 57.9 m, and a maximum depth of 100.6 m (Harris 1970). The only sig- nificant inflow into the lake most years is the Truckee River, which originates 193 km up- stream at Lake Tahoe in the Sierra Nevada. During 1976 and 1977 mean surface tempera- tures ranged from 6.1 to 23.1 C; the lake is monomictic, thermally stratifying in summer and mixing physically during winter. The most characteristic feature of Pyramid Lake is its high TDS— about 5,350 ' mg/1 during 1976-1977. Although sodium chloride is the dominant salt in the TDS (over 70%), the lake is high in bicarbonate alkalinity that is proba- bly important to the ecosystem. Since the baseload of TDS is relatively constant, the TDS of the lake varies with its volume (Ben- son 1978b). Temperature The maximum surface (0 to 1 m) water tem- perature in Pyramid Lake was 21.4 and 23. 1 C in July 1976 and August 1977, respectively (Lider 1978). The lake is thermally stratified from June through December; wind-gener- ated mixing occurs from January through May. A metalimnion forms at depths ranging from 16 to 22 m. The euphotic depth averaged 11 m for 1976 and 1977, which resulted in a trophogenic zone of about 4.67 km^ (Galat et al. 1981). Dissolved oxygen (DO) at the sur- face is always near saturation, about 8 mg/1. Metalimnetic and hypolimnetic DO deple- tion occurs beginning in July, following strati- fication and algal decomposition. Maximum DO deficits occur in the profundal zone just prior to fall mixing (Sigler et al. 1983). Plankton Diatoms Cyclotella sp. and Stephanodiscus spp. dominate the phytoplankton community during winter; the most abundant chloro- phyte, Crucigenia sp., attains its maximum abundance in spring. Blue-green algae are by far the dominate phytoplankton in Pyramid Lake (> 74%). Nodiilaria spumigena is the most abundant blue-green algae. Its bloom begins as early as July and last as late as Octo- ber. Following spring increases of algal growth, orthophosphate and nitrate are de- pleted and remain at low levels throughout the summer. Silica, in addition to nitrate, probably limits diatom production in Pyramid Lake (Galat et al. 1981). Chironomids are the lake's most abundant macroinvertebrates, fol- lowed by oligochaetes, which are especially abundant in the profundal zone (Robertson 1978). Two euryhaline amphipods, Garn- mariis lacustris and Hijallela azteca , are asso- ciated with tufa and rocks. La Rivers (1962) reports the Mormon creeper, Ambrysus mor- mon, common among the rocks around the periphery of the lake. The zooplankton community is composed of five cladocerans, three copepods, and four rotifers (Lider and Langdon 1978). The clado- ceran, Diaptomus sicilis, is a perennial spe- cies and the most abundant zooplankton throughout the year. Factors Affecting Fish Activity The cui-ui is the least abundant of the four major fish species native to Pyramid Lake. The other three species in increasing order of abundance are Lahontan cutthroat trout, Tahoe sucker, and tui chub. Vigg (1981) esti- mates cui-ui compose 0.03% by numbers and 0.47% by weight of the fish population. The mean cui-ui catch/gill net set slightly declined from 1976 to 1977 (1.29 to 0.95). This is not a statistically significant decrease (P = .21). During 1982 the largest spawning run in five years ascended the Marble Bluff fish way — 13,807 cui-ui (Scoppettone personal commu- nication 1983). Although it is not known what proportion this spawning migration repre- sents of the total adult population, now that the fishway is operational at a near constant efficiency, the magnitude of future spawning 594 Great Basin Naturalist Vol. 45, No. 4 N D J F 1976 1976 DATE (Month/y«ar) Fig. 8. Proportion of the catch from 15 bottom-set gill nets per month taken in the north (wavy lines), middle (clear), and south (dotted) sections of Pyramid Lake, Nevada, from November 1975 through December 1977. migrations during years of similar flow regimes will give an indication of cui-ui popu- lation trends. During 1976 and 1977 elevated net catch rates of cui-ui occurred during the spring. There was a concentration in the southern third of the lake during February to May 1976 and March to July 1977 (Fig. 8). Periods of increased proportional catches in the south- ern section corresponded to a decrease in the relative contribution of the middle third of the lake, with relatively little effect on the catch in the north end. The percent of total was 30, 10, and 60 for the north, middle, and south sec- tions, respectively. The Truckee River delta produced the highest catch rates: 26/net (38. 1 m) in May 1976 and 8/net in June 1977. These maxima correspond to the historical spawning period (April to June) and und()u!)tedly reflect spawning- related activity. It is a complex of environmental parame- ters, not just a single variable, that triggers year-round cui-ui activity patterns in Pyramid Lake. We would also expect a multivariate factor to trigger cui-ui spawning runs. Tahoe suckers exhibit a very similar response to the environmental complex in terms of temporal activity (Fig. 9); about 30% of the monthly cui-ui catches can be explained statistically by comparable Tahoe sucker catches (N = 26, P < .01). This relationship is even more con- vincing when the spatial effect is included; i.e., 373 individual net samples of the two species in benthic habitats throughout the lake were significantly correlated (r = .404, p < .001). Thus these two native catostomids are associated in terms of seasonality and habi- tat. Environmental variables that can be hy- pothesized to affect the activity of cui-ui in- October 1985 SiGLER ET AL.: Cui-Ul IN PYRAMID LaKE 595 c 500 S 400 3 300 CO UJ 200|- § 100 < H 30 S 25 ♦ 20 a. 15 O O 10 N 5 I 2000 i 1500 O 1000 i 500 a. I I I t I I I I I I I f Cuh-ui v« Tahoa sucker I ' ' — •— r vsTotal Phytoplankton r -.281 ft: 18 ,P>.05 va Truckaa RIvar inflow r .187 ,n: 28 , P>.05 NOJFMAMJJASONDJ FMAMJJASONO 60 30- 20? 10 o 1978 1978 1977 MONTH/YEAR Fig. 9. Comparison of monthly cui-ui catches (shaded) with Tahoe sucker catches, total zooplankton density, lake surface temperature, and monthly Truckee River inflow from November 1975 through December 1977. elude zooplankton, phytoplankton, lake tem- perature, and river inflow (Fig. 9). The gen- eral pattern of total zooplankton abundance was quite similar to cui-ui activity — unimodal in the spring of 1976 and bimodal in the spring and fall of 1977. The maxima did not corre- spond exactly, however, and the overall cor- relation (r = .262) was not statistically signifi- cant (P > .05). Peak phytoplankton con- centrations, primarily Noduloria spumigena , occurred in June 1976 and August 1977; dur- ing these months low numbers of both zooplankton and cui-ui occurred in the sam- ples. The overall correlation between cui-ui and phytoplankton is negative (r = —.281) but again not significant. Limited data suggest that cui-ui feed on benthic zooplankton, and Nodularia blooms may depress zooplankton populations; therefore these two trophic-re- lated variables may have a cause-effect rela- tionship with cui-ui activity. It is reasonable to hypothesize that Truckee River inflows affect river spawning-related cui-ui behavior and thus their lakeside activ- ity. The relationship between these variables, however, is very weak (r = . 167, P > .05). The flow regime of 1976 was relatively normal compared to the constant and extremely low flows of 1977. This situation provides an illu- minating comparison: in 1976 peak cui-ui ac- 596 Great Basin Naturalist Vol. 45, No. 4 1 1 1 1 1 1 O 1 1 10 12 14 16 18 20 22 21 Temperature (C) Fig. 10. Direct linear relationships (P<.05) between cui-ui catches and surface lake temperature during the winter-spring (W-S) periods of 1976 (asterisks) and 1977 (solid circles) compared to the inverse linear relationships (p<.05) during the summer-fall (S-F) periods of 1976 and 1977 combined (open circles). tivity in the lake occurred in May a.s flows began to subside after four months of high flows. In 1977 cui-ui activity again peaked in the spring in spite of the fact that river flows had been negligible for over seven months. This limited observation may illustrate that cui-ui have an innate response to spring envi- ronmental conditions that is not totally depen- dent on high river flows. It is notable, how- ever, that the magnitude of the 1977 cui-ui activity peak was much less than 1976 and had a different pattern (bimodal). Water temperature is another variable that is generally associated with the activity of spring-spawning fishes. There was no linear correlation, however, with lakewide cui-ui ac- tivity and the surface temperature of tlie lake (r = -.010, P > .05). The explanation for this apparent anomaly is that the relationship is (juadratic, not linear. During spring, as water temperature increases from winter minima, cui-ui activity increases in a direct relation- ship. As temperature continues to increase during the summer past a threshold value, cui-ui catches decline. This relationship is il- lustrated by Fig. 10; the temperature threshold was 11.0 C in 1976 and 17.6 in 1977. Thus the temperature range of maximum cui- ui activity during the winter-spring period of increase and summer-fall maximum tempera- ture decrease regime was 11.0 to 17.6 during 1976-1977. Photoperiod may also have an (unmeasured) effect on the prespawning mi- gration of cui-ui. Cui-ui catches varied significantlv bv season and depth for both 1976 and 1977 (P < .001). Maximum densities of cui-ui occurred in the inshore benthic zone from 0 to 15 m in depth, i.e., 2.0 fish/net (Table 10). Catch/effort pro- gressively decreased in benthic areas at depths of 23 and 46 m. No cui-ui was captured at depths 46 m, nor at the surface inshore, nor in the deep water colimin offshore. These dis- tribution patterns are similar to the Tahoe October 1985 SiCLER ET AL.: CUI-UI IN PyIUMID LaKE 597 Table 10. Distribution of 421 adult cui-ui captured in six depth-stratifit-d habitats of Pyramid Lake with experimen- tal gill nets from November 1975 through December 1977 (Vigg 1980). Inshore benthic: Offshore benthic: Surface inshore: Water column offshore: Number Depth of samples Cui-ui (m) Number Catch/effort 0-15 199 400 2.01 23 35 10 0.29 46 152 11 0.07 46-100 72 0 0 0-2 35 0 0 0-46 18 0 0 sucker, but according to net catches the cui-ui is apparently more inshore oriented (Vigg 1978a, 1978b, 1980). This does not overlook the fact that in the past large schools of sur- face-feeding cui-ui were observed over deep water and away from shore. Importance to Native Americans Native Americans inhabited the Lake La- hontan ecosystem from at least 11,250 ± 250 radiocarbon years BP, as indicated from arti- facts in the Winnemucca Lake basin (Hattori 1982). When the Indians arrived. Lake La- hontan occupied much of western Nevada (Russell 1885, Benson 1978a). Follett (1982) identified remains of four fish species at the Falcon Hill archaeological site, northwest of Winnemucca Lake: cui-ui, Tahoe sucker, La- hontan cutthroat trout, and tui chub; the earli- est materials associated with fish remains were radiocarbon dated at 9540 ± 120 radio- carbon years BP. D. R. Tuohy, curator of archaeology, Nevada State Museum, con- ducted extensive studies of human habitation in the Pyramid Lake basin; his data indicate that between 9500 and 500 BP at least 3 differ- ent prehistoric human cultures inhabited the region. The latest culture, the Northern Paiute, live in the vicinity of Pyramid Lake. The cui-ui at one time constituted the prin- cipal food of the Northern Paiute around Pyra- mid and Winnemucca lakes (Powers 1877). The fact that the Pyramid Lake Paiutes were called kuyuidikadi or kuyuitakuda (eaters of cui-ui) indicates the importance of this fish to the tribe's culture and sustenance. Snyder (1918) reports Pyramid Lake Indians pre- ferred cui-ui to trout. Bath (1978) reports cui- ui were preferred because, unlike trout, they could be dried in the sun and thus preserved for later use. Information collected by Stephen Powers (cited by Fowler and Bath 1981) indicate Tahoe suckers were also eaten regularly, but they were not as favored as cui-ui or trout. For example, a collection from Thea Heye Cave near the southern end of Pyramid Lake contained the desiccated re- mains of nine or more cui-ui but only one Tahoe sucker (Follett 1977). Pyramid Lake Indians made elaborate adaptations to various components of their wetlands, but they con- centrated on the capture of Lahontan cut- throat trout and cui-ui (Harner 1974). The Pyramid Lake Paiutes resisted all attempts by the federal government to turn them into irri- gation agriculturalists and instead actively pursued subsistence fishing (Knack 1982). Elaborate techniques were utilized by Indi- ans to capture fish (Fowler and Bath 1981). Fishing was a year-round subsistence activity at Pyramid Lake and could be separated into river and lake fishing (Bath 1978). River fish- ing could be further subdivided into (1) ex- ploitation of spawning runs (high water) and (2) low- water fishing. Lake fishing was an indi- vidual enterprise practiced during the sum- mer and early fall and accomplished with set lines (for trout), gill nets, harpoons, and spears (Fowler and Bath 1981). However, baited hooks were probably not used for cui- ui. Large treble hooks were utilized at one time to snag cui-ui congregated at the delta. Cui-ui were caught in large quantities and played an important role in the historic econ- omy of the Pyramid Lake Tribe as a trade item. Follett (1980) reports cui-ui remains at the Karlo Site, about 24 km north of Honey Lake, California. Archaeological sites in Ne- vada where cui-ui remains have been found include: Humboldt Cave and Humboldt Sink (Hubbs and Miller 1948, Heizer and Krieger 598 Great Basin Naturalist Vol. 45, No. 4 1956), Fishbone Cave and Winnemiicca Lake (Orr 1956), Lovelock Cave and Humboldt Sink (Follett 1967, 1970), the Nicolarsen Site at Winnemucca Lake (Follett 1974), Thea Heye Cave at Marble Bluff, Pyramid Lake (Follett 1977), and Falcon Hill (Follett 1982). The Pyramid Lake Tribe was the most widely known band of Northern Paiute. The Paiute name was familiar to Indians from Burns, Ore- gon, to Owens Lake, California, a distance of more than 805 km (Stewart 1939). T. J. Trelease (personal communication 1984), who talked to many of the older Paiutes and other local people (some whose observa- tions date back to 1906), believes the cui-ui and trout were taken by Indians in large num- bers only during spawning runs. These har- vests were so plentiful that they lasted for many months. The tui chub, however, was captured year-round, except during the more severe winter weather, and was a staple in the diet. It was taken from the lake in sagebrush bark nets and by hook and line. Some of the informants Trelease mentioned were Phil Orr, Margaret (Peggy) Wheat and L. W. Mor- gan. Spawning runs of cui-ui during high water were fished using platforms with lifting nets, with or without weirs (Fowler and Bath 1981). Sturdy winter platforms were built by several men who shared trout fishing privileges. Spring and summer platforms operated by individuals were less substantial. During summer and early fall, as well as winter, when flow was low and the water clear, harpoons and spears were used (Fowler and Bath 1981). Trelease (personal communication 1984), quoting L. W. Morgan, describes an Indian family fishing expedition sometime between 1906 and 1910 as follows: the father, using a gaff-hook fastened to a long pole, stood in waist-deep water and snagged cui-ui, which were tossed up on the bank. Mother and chil- dren built drying racks, cleaned the fish, and put them out to dry. Sometimes platforms were used in conjunction with weirs that di- rected fish over an area of river bottom paved with white rocks to improve visibility; the lighter bottom also facilitated night fishing. The fishing technology used by the Walker Lake Northern Paiutes, and at least to some degree by the Carson Lake and Humboldt Basin groups was similar to that of the Pyra- mid Lake Indians (Speth 1969). Fowler and Bath (1981) conclude native Americans in the western Great Basin have been involved in fishing complexes of various orders and vary- ing degrees for several millenia. Knack (1982) reports that efficient methods utilized by the Pyramid Lake Paiutes to cap- ture cui-ui and Lahontan cutthroat trout dur- ing their spawning runs were unacceptable to the Nevada legislature, which "imposed a definition of appropriate sporting technique, which was derived from the Anglo-European cultural past." Knack (1982) summarized the fishing laws the state of Nevada passed affect- ing the Indians: For over one hundred years, the state of Nevada at- tempted to impose its laws on the Northern Paiutes of Pyramid Lake. It declared which fish could be caught and where, as well as the techniques to be used. At first, the state tried simply to assume jurisdiction over Indians living on reservations, and then it employed a series of circumventions. Indians were cut off from sales markets and arrested as soon as they left federal trust land. Indian agents were encouraged to enforce state law on the reser- vation itself The opportunity to commercialize the one productive resource of the reservation was denied Paiutes by the imposition of state law; economic development was thereby blunted, prosperity stopped, and the local econ- omy allowed to stagnate. Meanwhile, Anglo economic developments, dependent on water diversions to agricul- ture, mining, and urban areas, produced drastic changes in the fishery population. The state defined fish as a luxury suitable only for sport, and subsequent Anglo actions assured that this would be so. Trelease (personal communication 1984) strongly disagrees with Knack. He believes the state had only the welfare of the resource in mind, and the federal government, whose responsibility it was, did nothing. Townley (1980) documents the historical devastation of theTruckee River, the Pyramid Lake trout and cui-ui fishery, and the atti- tudes of the various sides of the controversy. Snyder (1918) enunciated the attitude of those who believed that the fishery, so important to the livelihood of the Pyramid Lake Indians, could not stand in the way of white man's progress: A discussion of the economic \alue of the fishes of this region and any consideration of methods of propagation and prt)tection must begin and end with the assumption that agricultural and manufacturing interests are of paramount importance. A considerable and constantly increasing amount of the ilowing water must be used first for power and then for irrigation, and when any measure intendi'd for the protection of fishes is found to seriously JTiterfere with the working of power plants or the de- mands of agriculture it will have to be abandoned. October 1985 SiGLER ET AL.: CUI-UI IN PYRAMID LaKE 599 Fortunately for fishery resources in general and the cui-ui in particular, society is evolving a philosophy more compatible with the maintenance of renewable natural resources. Through federal laws, especially the Endan- gered Species Act, the cui-ui is deemed to be important to society as a whole. Management The primary management objective for the cui-ui is the restoration of a stable, naturally reproducing population, thereby allowing its removal from the endangered and threatened species list. This can best be done by increas- ing numbers substantially and by improving habitat. Ongoing programs designed to re- duce man-induced threats to the cui-ui popu- lation in Pyramid Lake include: (1) mainte- nance of water temperatures < 13.9 C during spawning, made possible by maintaining ade- quate flows and shading of the river; (2) reno- vation of the lower Truckee River so that it has a stable meandering channel and riparian habitat of trees and shrubs; (3) artificial propa- gation; (4) use of the Marble Bluff" Dam and Fishway for monitoring the spawning popula- tion, collecting eggs, and providing spawner access to the Truckee River; (5) maintenance of the fishway at Numana Dam to provide spawning access further upriver; (6) continua- tion of life history studies. The lower Truckee River temperatures fluc- tuate with flow, time of day, season, and year. The optimum temperature for cui-ui spawn- ing and egg hatching is 13.9 C. The lower river has a scouring, braided, exposed chan- nel; the need is for a meandering, stable chan- nel and banks that stand firm, with trees and shrubs for shading (Gregory 1982). The im- poundment above Marble Bluff" Dam has a population of predatory fish including sun- fishes and one or more species of catfish. This poses a problem for larval cui-ui that migrate downstream primarily at night. Removal or depletion of predators is a possible answer. In very low water years these larvae may also become disoriented in the impoundment. The tribe's Pyramid Lake Fisheries organi- zation is rearing millions of cui-ui fry annually, some stocked in Pyramid Lake and in the Truckee River. In 1982, a high water year, more than 11,000 adult cui-ui went up the fishway to spawning sites in the river (13,807 reached the trapping facility). Life history studies have and are exploring stages in the life of the fish and their current and optimum habitat. Artificial propagation should be con- tinued until the number of adult cui-ui in Pyramid Lake is at or near their optimum, if not historic, numbers. Barring disaster, the natural rims should then be able to maintain adequate numbers. The base load of TDS is essentially static in Pyramid Lake (Benson 1978b). This means the concentration varies inversely with the volume of the lake. The concentration of TDS in Pyramid Lake should not be allowed to increase appreciably; current levels are at or above optimum for cui-ui. The lake levels should not fluctuate beyond a range of plus or minus 3 m except in high water years. Nutri- ent loading should not be increased from mu- nicipal, industrial or agricultural sources. Summary The cui-ui, once so abundant that it was a staple in the diet of the Pyramid Lake Paiute Indians and an item of trade, is today endan- gered. It is a slow-growing, long-lived fish, reaching a length of > 70 cm. Cui-ui eggs must be water-hardened in relatively fresh water. It may or may not be able to spawn successfully in the Truckee River-Pyramid Lake interface, in temporary streams of high water years, or in springs in Pyramid Lake. Biologists are not in firm agreement on these points. Upriver spawning migrations are of- ten, but apparently not always, triggered by surges of fresh water. Spawning starts from mid-April to May and extends through June or, rarely, July. Modification of the Marble Bluff" Fishway provides upstream passage for cui-ui, espe- cially during low water years. Eggs are taken from part of the spawning population; others are allowed to move upstream. Spawning suc- cess depends largely on acceptable tempera- tures and ffows. Mortality, primarily from predation, is presumably high on both em- bryos and larvae in the stream. Once in the lake, young cui-ui undoubtedly face heavy predation. In addition to natural reproduc- tion, millions of larvae are released each year from the David L. Koch Fish Hatchery, Sut- cliff"e, Nevada. 600 Great Basin Naturalist Vol. 45, No. 4 Cui-ui feed on zooplankton, benthic inver- tebrates, and algae. They inhabit shallow to medium depth water (< 46 m) in the lake, where they are most active in spring and fall. Adults move into fresh water only to spawn; young cui-ui generally remain in the river for a few weeks after they are hatched. Conclusions The cui-ui is endangered today because of a progressive population decline resulting from transbasin water diversion, failure on the part of the federal government to originally protect the Indians' resources, upstream water use, and early adverse legal and political decisions. Percent of total river flow diversions that be- gan in 1905 reached a climax in the early 1930s, when the combination of low river flows and dropping lake levels caused a delta to form at the mouth of the river. The cui-ui could no longer migrate upstream to spawn; thereafter numbers of adults dropped sharply. To date the population has not stabi- lized or recovered. Artificial propagation and restoration of river spawning are providing an interim solution. The long-term answer is ac- ceptable spawning habitat: an adequate flow of < 13.9 C water from early to mid- April through June, a stabilized, nonbraided river bed with spawning gravels, and reestablish- ment of shaded raparian habitat. Acknowledgments This work was performed under Bureau of Indian Affldrs contract H50C 14209487. Assis- tance and cooperation was provided by em- ployees of W. F. Sigler & Associates Inc. (WFSAI), members of the Pyramid Lake Paiute Indian Tribe, and the U.S. Fish and Wildlife Service, Fisheries Assistance Office, Reno, Nevada. Denise Robertson, formerly of WFSAI, was responsible for part of the sec- tion on age and growth. The manuscript was reviewed by Dr. David H. Bennett, associate professor, Fish and Wildlife Resources De- partment, University of Idaho, Moscow; Dr. Peter B. Moyle, professor and chairman. Wildlife and Fisheries Biology Department, University of California, Davis; Dr. Sharon Ohlhorst, research assistant professor. Fish- eries and Wildlife Department, Utah State University, Logan; Mr. Alan Ruger, fisheries director. Pyramid Lake Fisheries, Sutcliffe, Nevada; Dr. John W. Sigler, biologist and projects director, W. F. Sigler & Associates Inc., Logan, Utah; Mr. Thomas J. Trelease, chief of fisheries (retired), Nevada Depart- ment of Wildlife, Reno. We also wish to ac- knowledge the long-term commitment of the late Dr. David L. Koch to the research and restoration of the cui-ui. Literature Cited Bath. J E 1978. Lake-iiiargin adaptations in Great Basin prehistory. Unpuhlislied Masters of Science the- sis. University of Ne\ada, Reno. Benson, L V. 1978a. Fluctuation in the level of pluvial Lake Lahontan during the last 40,000 years. Quar- ternary Res. 9:300-318. 1978b. The chemical composition of Pyramid Lake: past, present, and future. Pages 356-413 in W. F. Sigler and J. L. Kennedy, eds.. Pyramid Lake Ecological Study, W. F. Sigler & Associates Inc., Logan, Utah. Brks. M 1978. The embryonic development of the cui-ui, Chasmistes ctijiis (Teleostei, Catostomidae). Un- pubhshed Masters of Science thesis. University of Nevada, Reno. 76 pp. Carlander, K. D. 1969. Handbook of freshwater fisheries biology. Iowa State University Press, Ames. 752 pp. Cn.\TT(), D. A. 1979. Effects of .sahnity on liatching suc- cess of the cui-ui. Prog. Fish Cult. 41(2):82-85. Cope. E. D 1883. On the fishes of the recent and pliocene lakes of the western part of the Great Basin, and of the Idaho Pliocene Lake. Proc. Acad. Nat. Sci. Philadelphia. 35:134-166. FOLLETT. W. I 1963. Preliminary report offish remains from the Falcon Hill sites, Washoe County, Ne- vada. Pages 33-34 in The 1962 Great Basin An- thropological Conference, Nevada State Museum Anthropological Papers, No. 9. Carson City, Ne- vada. 1967. Fish remains from coprolites and midden deposits at Lovelock Cave, Churchill County, Ne- vada. Reports of the Univ. of California Archeol. Surv. 70:93-116. 1970. Fish remains from hiunan coprolites and midden deposits obtained during 1968 and 1969 at Lovelock Cave, Churchill County, Nevada. Con- trib. Univ. of California Archeol. Res. Fac. 10:163-175. 1974. Fish remains from site NV-Wa-197, Win- nemucca Lake, Nevada. Contrib. Univ. of Califor- nia Archeol. Res. Fac. 21:37-43. 1977. Fish remains from Thea Heye Cave, N\- Wa-385, Washoe Countv, Nevada. Contrib. Univ. of California Archeol. Res. Fac. 35:59-80. 1980. Fish remains from the Karlo Site (CA-Las- 7), Lassen Countv, California. J. (California Great BasinAnthro. 2(i):114-122. October 1985 SiGLER ET AL.: CUI-UI IN PYRAMID LaKE 601 1982. An analysis offish remains from ten archaeo- logical sites at Falcon Hill, Washoe County, Ne- vada, with notes on fishing practices of the ethno- graphic Kuyuidikadi Northern Paiute. Appendix A in E. M. Hattori, The archaeology of Falcon Hill, Winnemucca Lake, Washoe County, Ne- vada. Nevada State Museum Anthropological Pa- pers, No. 18. Carson City, Nevada. Fowler, C. S , and J. E. Bath 1981. Pyramid Lake North- ern Paiute fishing: the ethnographic record. J. California Great Basin Anthro. 3(2): 176-186. Frazier.J. L.andK. P Ferjancic 1977. Progress report: cui-ui hatchery operations. Pyramid Lake Indian Tribal Enterprises. Galat. D. L. andW J McCoNNELL, eds. 1981. Effects of increasing total dissolved solids on the dynamics of Pyramid Lake microcosm communities. Colorado Coop. Fish. Res. Unit, Colorado State University, Fort Collins. 131 pp. Galat, D L , E L Lider, S Vice, and S R Robertson. 1981. Limnology of a large, deep North American terminal lake. Pyramid Lake, Nevada, USA. Hy- drobiologia. 82:281-317. Gregory. D I 1982. Geomorphic study of the lower Truckee River, Washoe County, Nevada. Earth Resources Dept. , Colorado State University, Fort Collins. 125 pp. Harner, N. S. 1974. Indians of Coo-Yu-ee-Pah, the his- tory of the Pyramid Lake Indians and early tribal history, 1843-1959. Dave's Printing and Publish- ing, Sparks, Nevada. Harris, E. E. 1970. Reconnaissance bathymetry of Pyra- mid Lake, Washoe County, Nevada. Hydrologic Investigation Atlas HA-379, U.S. Geol. Surv. Ip. Hattori, E M 1982. The archaeology of Falcon Hill, Winnemucca Lake, Washoe County, Nevada. Ne- vada State Museum Anthrological Papers, No. 18. Carson City, Nevada. Heizer, R F . AND L K Krieger. 1956. The archaeology of Humboldt Cave, Churchill County, Nevada. Univ. California Pub. Amer. Archaeology and Ethnology 47(1). HiLE, R. 1941. Age and growth of rock bass Amhloplitcs rupestris (Rafinesque) in Nebesh Lake, Wiscon- sin. Trans. Wis. Acad. Sci., Arts, Lett. 33:189-337. HUBBS, C. L., AND R R Miller 1948. The Great Basin with emphasis on glacial and postglacial times. 11. The zoological evidence. Univ. of Utah Bull. 38(20): 17-166. Johnson, V. K. 1958. Fisheries management report. Lakes — Pyramid, Walker, and Tahoe. Dingell- Johnson Completion Rep. Nevada Fish and Game Department. FAF-4— R. 47 pp. JONEZ, A. 1955. Pyramid Lake life history of cui-ui. An- nual completion report, June 1955. Nevada Fish and Game Commission. FAF-4-R, Job 1-E. 2 pp. Knack, M. C. 1982. The effect of Nevada state fishing laws on the Northern Paiutes of Pyramid Lake. Nevada Hist. Soc. Quarterly. 25:251-265. Knopf, F. L . and J L Kennedy 1980. Foraging sites of White pelicans nesting at Pyramid Lake, Nevada. Western Birds. 11:175-180. K(K :n, D L. 1972. Life history information on the cui-ui lakesucker (Chasmistes cujus Cope, 1883) en- demic to Pyramid Lake, Washoe County, Nevada. Unpublished Ph.D. dissertation. University of Nevada, Reno. 343 pp. 1973. Reproductive characteristics of the cui-ui lakesucker (Chasmistes cujus Cope) and its spawn- ing behavior in Pyramid Lake, Nevada. Trans. Amer. Fish. Soc. 102(1) : 14.5-149. 1976. Life history information on cui-ui lakesucker (Chasmistes cujus Cope 1883) in Pyramid Lake, Nevada. Biol. Soc. Nev. Occas. Pap. 40:1-12. 1981. Temperature tolerance evaluations of cui-ui (Chasmistes cujus) fertilized eggs and larvae to swimup stage. Report to PLITE. Koch and Associ- ates, Reno, Nevada. Koch. D. L , andG P Contreras. 1972. Swimming abil- ity and effects of stress on the cui-ui lakesucker (Chasmistes cujus Cope). Biol. Soc. of Nevada, Occ. Pap. 31:1-8. 1973. Hatchingtechniqueof the cui-ui lakesucker (Chasmistes cujus Cope 1883). Prog. Fish Cult. 34(l):61-63. Koch, D. L.. J. J. Cooper. E. L Lider. R. L Jacobson. and R. J. Spencer. 1979. Investigations of Walker Lake, Nevada: Dynamic ecological relationships. Desert Res. Inst. Bioresources Center Publica- tions, Reno, Nevada. 191 pp. La Rivers, 1. 1962. Fishes and fisheries of Nevada. Ne- vada Fish and Game Commission, Reno. 782 pp. Lider, E L. 1978. Physical and chemical limnology of Pyramid Lake, Nevada. Pages 157-190 in W. F. Sigler and J. L. Kennedy, eds.. Pyramid Lake ecological study. W. F. Sigler & Associates Inc., Logan, Utah. Lider. E L., and R. W Langdon. 1978. Plankton ecol- ogy. Pages 241-293 in W. F. Sigler and J. L. Kennedy, eds.. Pyramid Lake ecological study. W. F. Sigler & Associates Inc., Logan, Utah. Lockheed Ocean Sciences Labor.\tories. 1982. Inves- tigation on the effects of total dissolved solids on the principal components of the Pyramid Lake food chain. USDI Bureau of Indian Affairs. Final report. 84 pp. Long. W. L . and W W. Ballard. 1976. Normal embry- onic stages of the white sucker, Catostomus com- mersoni. Copeia 1976 (2): 342-351. McConnell, W J . D L Galat. and K Hamilton- Galat 1978. Potential fish production based on organic matter contributions. Colorado Coop. Fish. Res. Unit, Colorado State University, Fort Collins. 84 pp. McGarvey. P. J. 1974. Survival ofunique and endangered cui-ui fish enhanced. U.S. FishWildl. Serv. News release. 4 pp. Miller, R. J., and H E Evans. 1965. External morphol- ogy of the brain and lips in catostomid fishes. Copeia 1965 (4):467-187. Miller, R R . and G. R. Smith. 1967. New fossil fishes from Plio-Pleistocene Lake Idaho. Occ. Pap. Mus. Zool. Univ. of Michigan 654:1-24. 1981. Distribution and evolution of Chasmistes (Pisces: catostomidae) in western North America. Occ. Pap. Mus. Zool. Univ. of Michigan 696:1-46. 602 Great Basin Naturalist Vol. 45, No. 4 Nelson, E M 1948. The comparative morphology of the Weberian apparatus of the Catostomidae and its significance in systematics. J. Morphol. 83:22.5-251. 1949. The opercular series of the Catostomidae. J. Morphol. 85:.559-567. 1961. The comparative morphology of the defini- tive swim bladder in the Catostomidae. Amer. Midland Nat. 65(1);101-110. Nelson, L 1976. SHAD II. A model for analysis of fish- eries age and growth data. Wildl. Sci. Dept., Utah State Univ., Logan. 15 pp. Orr, P C. 1956. Pleistocene man in fishbone cave, Persh- ing County, Nevada. Department of Archaeology, Nevada State Museum Bull. 2:1-20. Powers, S. 1877. Centennial mission to the Indians of western Nevada and California. Pages 449-^60 in Annual report of the Board of Regents of the Smithsonian Institution for 1876. U.S. Printing Offices, Washington, D.C. Pyle, E. a , J Frazier, D. King, K Johnson, D, Lockard, AND T J. Trelease. 1977. Cui-ui recovery plan- draft. U.S. Fish Wildl. Serv., Endangered Spe- cies Program. Region I. RiNGO, R. D., and G. M Sonnevil. 1977. Evaluation of Pyramid Lake fishway: operation and fish passage 1976-1977. U.S. Fish and Wildlife Service, Fish. Asst. Office, Reno, Nevada. Special Report. 21 pp. Robertson, D. A 1979. Cui-ui investigations: age and growth. Pyramid Lake, 1978. U.S. Fish Wildl. Serv., Fish. Asst. Office, Reno, Nevada. Special Report. 22 pp. Robertson, D. A., and D. L. Kogh. 1978. Life history of the cui-ui Chasmistes ciijtis Cope. Pages 51-68 in Pyramid Lake ecological study, W. F. SiglerandJ. L. Kennedy, eds. W. F. Sigler & Associates Inc., Logan, Utah. Robertson, S. R. 1978. The distribution and relative abundance of benthic macroinvertebrates in Pyra- mid Lake, Nevada. Unpublished thesis. Univer- sity of Nevada, Reno. 68 pp. Russell, I. C. 1885. Geological history of Lake Lahontan. U.S. Geological Survey Monograph 11. 288 pp. Sgoppettone, G. G.. M. Coleman. H. Burge, and G WedemeYER. 1981. Cui-ui life history: river phase. U.S. Fish Wildl. Serv. Fish. Asst. Office, Reno, Nevada. Annual report. 6.3 pp. Scoppettone, G. G., G. A. Wedemeyer. M. Coleman, and H Burge 1983. Reproduction by the endan- gered cui-ui in the lower Truckee River. Trans. Amer. Fish. Soc. 112(6):788-793. Sigler, W. F. 1951. The life history and management of the mountain whitefish {Prosopiiim williamsoni (Girard)) in Logan River, Utah. I'tah State Agric. Col. Bull. 347. ,36 pp. Sk;ler, W. F , W T Helm, P A Kucera, S Vigc;, and G W Workman 1983. Life history of the Lahontan cutthroat trout, Salmo clarki henshawi, in Pyra- mid Lake, Nevada. Great Basin Nat. 43(I):l-29. Sigler, W. F., and J. L. Kennedy, eds. 1978. Pyramid Lake ecological study. W. F. Sigler & Associates Inc., Logan, Utah. 545pp. Snyder, D E 1981a. Identification of larval and early juvenile fishes collected from Pyramid Lake and the Truckee River, Nevada, 1973, 1974, 1975, and 1980. Dept. Fish, and Wildl. Bio., Larval Fish Lab. Colorado State University, Fort Collins. 1981b. Developmental study and identification of the catostomid fish larvae of the Truckee River System, final report and key. U.S. Fish and Wildl. Serv. Fish. Asst. Office, Reno, Nevada. 36 pp. 1983. Identification of catostomid larvae in Pyra- mid Lake and the Truckee River, Nevada. Trans. Amer. Fish. Soc. 112(213);33;3-348. Snyder, J. O 1918, The fishes of the Lahontan system of Nevada and northeastern California. U.S. Bur. Fish. Bui. 191.5-16 (35):31-88. Sonnevil, G. M. 1977a. Cui-ui {Chasmistes cttjus) popu- lation monitoring: Pyramid Lake, Nevada 1977. U.S. Fish Wildl. Serv., Fish. Asst. Office, Reno, Nevada. Special report. 6 pp. 1977b. Analysis of catch effort data to reflect the population status of the cui-ui (Chasmistes cujus) in Pyramid Lake, Nevada. U.S. Fish Wildl. Serv., Fish. Asst. Office, Reno, Nevada. Special report. 6 pp. 1978. Cui-ui investigations: Pyramid Lake, 1978. U.S. Fish Wildl. Serv. Fish. Asst. Office, Reno, Nevada. Special report. 26 pp. 1981. Evaluation of the cui-ui restoration pro- gram: 1977-1980. U.S. Fish Wildl. Ser\'. Fish. Asst. Office, Reno, Nevada. Special report. 25 pp. Spencer, R.J 1977. Silicate and carbonate sediment-wa- ter relationships in Walker Lake, Nevada. Unpub- lished Masters of Science thesis. University of Nevada, Reno. 98 pp. Speth, L K 1969. Possible fishing cliques among the Northein Paiutes of the Walker River Reserva- tion, Nevada. Ethnohistory. 16(3):22.5-244. Stewart. N. H 1926. Development, growth, and food habits of the white sucker, Catostomus commer- so»i( Le Sueur. Bull. Bur. Fi.sh. 42:147-184. Stewart. O. C. 1939. The Northern Paiute bands. Univ. of California Anthropological Records. 2 (3). 1941. Culture element distributions: XIV, North- ern Paiute. I'niv. of California Anthropological Records. 4 (3). Sumner, F. H. 1940. The decline of the Pvramid Lake fisherv. Trans. Amer. Fish. Soc. 69(1939):216- 224. Tesch. F. W. 1971. Age and growth. Pages 98-131 in W. E. Ricker, ed.. Methods of assessment offish pro- duction in freshwaters. Blackwell Sci. Publ., Ox- ford. Ting, P 1967. \ Pyramid Lake surface artifact as.semblage located at or near the 3,8()()-foot ele\'ation. Ne- vada Archaeological Survev Reporter, Reno. 1(8):4-I1. TowNLEY. J M 1980. The Truckee River Basin Fishery, 1844-1944. Nevada Historical Society in coopera- tion with Desert Res. Inst., University of Nevada, Reno. 88 pp. Trelease. T J 1971. Pvramid Lake's cui-ui — relic of the past. Nev. Outdoors Wildl. Rev. 5(l):26-27. October 1985 SiGLER ET AL.: Cui-UI IN PYRAMID LaKE 603 TUOHY. D R , AND D T Clark 1979. Excavations at Mar- ble Bluff Dam and Pyramid Lake Fishwav, Ne- vada, Parts I, II, III, IV, V, and VI. U.S.Dept. Int., Heritage Conservation and Recreation Ser- vice. Interagency Archaeological Services. San Francisco, California. 492 pp. United States Fish and Wildlife Service (USDI). 1983. Cui-ui recovery plan (revision). Washing- ton, D.C. 28pp. United States Fish and Wildlife Service, Nevada De- partment of Fish and Game, and the California Department of Fish and Game. 1976. Life history and habitat evaluation of the Lahontan cutthroat trout and cui-ui in the Truckee River, Octo- ber-December 1976. Quarterly Progress Report. Fish. Asst. Office, Reno, Nevada. 14 pp. Van Oosten, J. 1923. The whitefishes [Coregonus clu- peafonnis). A study of the scales of whitefishes of known ages. Zoologica. 11(17):380-412. 1929. Life history of the lake herring {Coregonus arteilii LeSucur) of Lake Huron as revealed by its scales, with a criti(jue of the scale method. U.S. Bur. Fish. 44(1 928): 26.5-428. 1944. Factors affecting the growth offish. Trans. Ninth N. Amer. Wildl. Conf 9:177-183. ViG(;, S. 1978a. Fish ecology. Pages 191-240 in W. F. Sigler and J. L. Kennedy, eds.. Pyramid Lake Ecological Study. W. F. Sigler & Associates Inc. , Logan, Utah. 1978b. Vertical distribution of adult fish in Pyra- mid Lake, Nevada. Great Basin Nat. 38(4):417-428. 1980. Seasonal benthic distribution of adult fish in Pyramid Lake, Nevada. California Fish and Game 66(l):49-58. 1981. Species composition and relative abundance of adult fish in Pyramid Lake, Nevada. Great Basin Nat. 41(4):395-408. HELMINTH PARASITES OF THE WHITE-TAILED JACKRABBIT, LEPUS TOWNSENDI, FROM NORTHWESTERN COLORADO AND SOUTHERN WYOMING Larry M. Shults' and Lora G. Rickard" Abstract— Helminth parasites of white-tailed jackrabbits, Lepus townsendi , were surveyed from southern Wyo- ming and northwestern Colorado. A total of eight helminth species were identified, including two species of adull cestodes, Mosgovoyia pectinata and M. varahilis. three species of larval cestodes, Multiceps serialis. Taenia pisi- fonnis, and Taenis sp., and three species of nematodes, Dermatoxys veliger, Passalurus ambiguus, and a filariid. MicipseUa hrevicauda . In addition, eggs of an unidentified species o^Neniatodiriis were found in pooled fecal samples. The cysticercus larva ofTaenia sp . is a species new to science and will be reported elsewhere. Mosgovoyia varahilis anc MicipseUa hrevicauda are new records for the white-tailed jackrabbit. The helminth parasites of the white-tailed jackrabbit are not well known throughout most of its range. Only in North Dakota has any attempt at a survey been made (Voth and James 1965). Additional reports such as those of Honess and Winter (1956) and Thomas and Honess (1962) indicate that helminths occur occasionally in this host but give no informa- tion on number of hosts examined or percent of infection. Materials and Methods The hosts for this study were collected from three locations, i.e., near Meeker, Colorado; 20 miles north of Baggs, Wyoming; and 30 miles north of Medicine Bow, Wyoming. All were collected using firearms. Standard para- sitological techniques were used for the re- covery of helminths. Selected examples of all adult helminths were deposited in the United States National Museum Helminthological Collection (USNM#). Results A total of eight helminth species were foimd infecting white-tailed jackrabbits examined in this study. They consisted of five species of cestodes and three species of nematodes. A comparison of hosts from the three study areas is shown in Table 1. For the sake of clarity each species will be considered separately. Mosgovoijia pectinata (Goeze, 1782). — This is the only cestode found in jackrabbit* from all areas of collection. It has been re- ported previously from Albany County, Wyo- ming (Honess 1982), and from southwesterr North Dakata as Cittotaenia by Voth anc James (1965). It is found in the small intestine of the definitive host (USNM# 77145). Mosgovoyia varahilis (Stiles, 1895) Bever- ridge, 1978. — This species was found in onl\ one host from northwestern Colorado. It haj been reported previously from the cottontai rabbit, Sylvilagiis nuttalli, examined fron' southern Wyoming (Honess and Winter 1956 as Cittotaenia varahilis . This cestode, like A/ , pectinata , is a double-pored species that oc- curs in the small intestine. It may be distin- guished from the former by the arrangemeni of the testes, which are enclosed between the ovaries insteael of extending to the longitudi- nal excretory canals (USNM# 77144). Multiceps serialis Gervalis, 1847. — One host collected north of Baggs, Wyoming, con- tained a 5-cm coenurus of this species. It waj located in the posterior abdominal cavity ir association with the psoas muscle. This spe- cies has been reported pre\ iously in the cot- tontail rabbit from Carbon Coimt\\ W\"omin§ (lioness 1982), and in v\'hite-tailed jackrabbit* in North Dakota (Voth and James 1965). Taenia pisifonnis Bloch, 1780. — Cysticerci of this cestode were found encysted in the intestinal mesenteries of two hosts collected 'Department of Zoology and Physiology, University of Wyoming, Laramie, Wyoming 82071. College of Veterinary Medicine, Oregon State University, Corvallis, Oregon 97331. 604 ( )ctober 1985 Shults, Rickari): Jackrabbit Parasites 605 Table 1. A comparison of tlir lu-liiiintlis ol' wliitf-tailcd jackial)hits from three study areas. Percentages indicate percent infected in that locahty. Species 30 miles north 20 miles north Meeker, Colorado. Med icine Bow, Wyoming (n = 10) Baggs, Wyoming (n = ll) (n = 8) Mosiiovot/ia pcctinata 4 (40%) 1 (9%) 1 (12%) M. larahilia — — 1 (12%) Miilticcps si'iialis — 1 (9%) — Taenia pisiformis — 2(18%) — Taenia sp. — 1 (9%) — Passalitrusaiiihiauus — — 1 (12%) Dennatoxt/s r<7(gera 6 (60%) 2(18%) — MicipacUa brcvicauda — 5(45%) — north of Baggs, Wyoming. This larval cestode, commonly found encysted in the viscera of cottontail rabbits from the same area (Shults, unpublished data) is a parasite of coyotes, CUtnis latrans , and bobcats. Lynx nifiis . It has been reported from white-tailed jackrabbits in North Dakota (Voth and James 1965) and cottontail rabbits from Carbon and Fremont counties, Wvoming (Honess and Winter 1956). Taenia sp. — One host from the Baggs, Wy- oming, site was found to be infected with visceral cysts of an undescribed species of this genus. Similar cysts have been found in cot- tontail rabbits. Descriptions of this new spe- cies will be published elsewhere. Passalurus amhiguus (Rudolphi, 1819) Du- jardin, 1845. — This species was found in the caecum of one host collected in northwestern Colorado. Thomas and Honess (1962) indi- cated that this species has been found in cot- tontail rabbits in Wyoming (USNM# 77146). Dennatoxys veligera (Rudolphi, 1819) Schneider, 1866. — These caecal nematodes were found in hosts collected from Wyoming sites. They have also been reported from the white-tailed jackrabbit in Albany County, Wyoming, by Honess and Winter (1956) (USNM# 77147). Micipsella hrevicauda Lyons & Hansen, 1961. — Only hosts collected from near Baggs, Wyoming, were infected with this filariid ne- matode. The adults were found free in the abdominal cavity, and microfilaria were re- covered from the circulating blood. Morpho- logical comparisons with other species of this genus indicate that our specimens most closely resemble those of M. hrevicauda de- scribed from black-tailed jackrabbits, Lepiis californicus , in Kansas by Lvons and Hansen (1961). Voth and James (1965) found microfi- laria in blood smears from white-tailed jackrabbits collected in North Dakota but did not assign them to any genus, although they suggested that they might be M. hrevicauda (USNM# 77148). In addition to the above species, pooled fecal samples from each of the study areas revealed ova oi Nematodirus sp. No adults of this genus were found. It is possible that the specimens were N. neomexicanus , which has been reported from black-tailed jackrabbits in Colorado and cottontails in Wyoming (Thomas and Honess 1962). Discussion Helminth parasites found in the present study differ somewhat from those found in a similar study by Voth and James (1965). They found only the microfilaria of a filariid ne- matode present in their survey, whereas two cecal nematode species and the filariid Micipsella hrevicauda were found in our study. In addition, o\aoiNematodirus sp. was found by fecal flotation. In our study only the adult cestodes Mos- govoyia pectinata and M. varahilis were found. This is in contrast to data from both Voth and James (1965) and Honess (1982), who found Raillietina sp. in the white-tailed jackrabbit. Honess (1982) stated that this spe- cies occurs more often in hosts from an arid or semiarid area, and Mosgovoyia is most com- monly found in hosts living along streams or in foothills and forests. This was not the case in the present study; Raillietina sp. was not found in anv area, arid or otherwise. 606 Great Basin Naturalist Vol. 45, No. Mosgovoyia variabilis has not previously been reported from white-tailed jackrabbits, although Honess (1982) stated that this ces- tode is "probably a parasite of all wild rabbits and hares ' in Wyoming. Literature Cited Beveridge, I. 1978. A taxonomic revision of the genera Cittotaenia Riehm, 1881, Ctcnotaenia Railliet, 1893, Mosgovoyia Spasskii, 1951, and Pseiidocit- totaenia Tenora, 1976. (Cestoda: Anoplocephali- dae). Memoires du Museum National D'Histoire Naturelle. Series A. Zoologie 107: 1-64. Honess, R F 1982. Pages 164-169 in E.T. Thome, N. Kingston, W. R. Jolley, and R. C. Bergstrom eds.. Diseases of wildhfe in Wyoming (2d ed. Wyoming Game and Fish Department. Honess, R. F., and K. B. Winter. 1956. Diseases o wildlife in Wyoming. Wyoming Game and Fisl Comm. Bull. 9. 279 pp. Lyons, E. T. andM. F. Hansen. 1961. Observations oi Micipsella brevicauda n.sp. (Nematoda: Filari oidea) from the black-tailed jackrabbit, Lepus call fornicus melanotis, in southwestern Kansas Trans. Amer. Microsc. Soc. 80: 204-210. Thom.\s, G. M., and R. F. Honess. 1962. Intestina roundworms of rabbits and hares of the Unitec States: host, locality, and the authority. Univ. o Wyoming Publ. 27:17-25. VoTH, D. R. andT.R James. 1965. Parasites of the white tailed jackrabbit in southwestern North Dakota Proc. North Dakota Acad. Sci. 19: 15-18. THERMAL ECOLOGY AND ACTIVITY PATTERNS OF THE SHORT-HORNED LIZARD (PHRYNOSOMA DOUGLASSl) AND THE SAGEBRUSH LIZARD (SCELOPORUS GRACIOSUS)lN SOUTHEASTERN IDAHO Craig Cuyer' " and Allan D. Linder ' Abstract — A mark-recaptnre study of the short-horned lizard {Phnjno.soina doufilassi ) and the sagebrush lizard {Sceloporus ^raciosus) was performed from 1976 to 1977 in southeastern Idaho. Both species had mean cloacal temperatures of approximately 33 C. However, P. douglassi had more variable cloacal temperatures, particularly during morning and evening periods. This was caused by differences in sleeping sites chosen by the two species. Adults of both species were active from mid-April through late August, with peak activity in June. Juvenile P. dotiglassi displayed a seasonal activity pattern similar to that of adults. Juvenile S. graciosus were most active later in the year (August), when adults were disappearing. In both species, young-of-the-year appeared in early to mid-August. Adult and juvenile P. doiiglassi were active during all daylight hours and displayed no activity peaks, whereas young-of-the- year displayed a bimodal activity pattern. Adult and juvenile S. graciosus were active over all daylight hours but had peak activity between 1200 and 1500 h. Ants (Pogonomyrmex) were the lizard's principle prey. However, only young-of-the-year P. douglassi had activity patterns that paralleled that of ants on their mounds. This study was conducted to determine as- pects of thermal ecology and seasonal, daily, and reproductive activity patterns for the short-horned lizard {Phrynosoma douglassi ) and the sagebrush lizard (Sceloporus gracio- sus) near the northern edge of their distribu- tions. Thermal activity is a commonly studied aspect of lizard ecology (Brattstrom 1965, Heath 1^65, Pianka and Parker 1975, Prieto and Whitford 1971 for P. douglassi, and Brattstrom 1965, Burkholder and Tanner 1974, Mueller 1969 for S. graciosus), whereas activity patterns are much less commonly re- ported (Pianka and Parker 1975, for P. dou- glassi and Burkholder and Tanner 1974, and Stebbins 1944 for S. graciosus). Our results are compared to studies of P. douglassi from Utah (Pianka and Parker 1975) and S. gracio- sus from Utah (Burkholder and Tanner 1974, Woodbury and Woodbury 1945, Tinkle 1973), Wyoming (Mueller and Moore 1969), and California (Goldberg 1975, Stebbins 1944). Methods The study was conducted on the Idaho Na- tional Engineering Laboratory (IN EL) site in southeastern Idaho during June to October 1976 and April to October 1977. This area is ca 1500 m in elevation, is characterized vegeta- tively as a sagebrush-desert community (Mc- Bride et al., 1978), and is composed geologi- cally of Recent lava flows covered by wind and waterborne deposits. The climate is charac- terized by short, hot summers (average maxi- mum and minimum temperatures 30.5 and 10.0 C, respectively) and long, cold winter (average maximum and minimum —2.7 and — 16.1 C, respectively). The average yearly precipitation is 21.6 cm, mostly in the form of spring rains. Much of the data were collected on a 1-ha grid system with stakes placed 10 m apart. This grid system and the surrounding area were dominated vegetatively by big sage- brush (Artemisia tridentata), rabbitbrush (Chrysothamnus nauseosus), halogeton (Halogeton glomeratus), and squirreltail grass (Sitanion hystrix). Lizards were captured by hand or noose and were marked permanently by toe-clip and for field identification by color marks on their legs. Individuals were sexed and measured snout-to-vent (nearest mm), from which age and sex groups were deter- mined. Five age and sex groups were recog- nized; young-of-the-year (YOY, 23-30 mm SVL), juvenile males (30-50 mm SVL), juve- nile females (30-60 mm SVL), adult males (>50 mm SVL), and adult females (>60 mm SVL). Cloacal temperatures were measured Department of Biolog\', Idaho State University, Pocatello. Idaho 83201. -Present address: Department of Biology, University of Miami, Coral Gables, Florida 3.3124. 607 608 Great Basin Naturalist Vol. 45, No. 4 45-1 40- ^35- bJ £ 30- S 25- 20- 15- / ./>• I I I 09 10 11 I I I I 12 13 14 15 -I— 16 -r- 17 I I 18 19 I I 20 21 22 TIME OF DAY Fig. 1. The relationship between mean environmental temperatures on cloudless days and time of year. Solid squares are soil temperatures in sun, open diamonds are air temperatures in sun, solid triangles are air temperatures in shade, and open circles are soil temperatures in shade. to the neare.st 0.2 C with a Schiilteis quick- reading thermometer, e.xcept for YOY, which could not accept the bulb without injury. Handling time for individuals was generally less than 5 minutes, and temperatures were taken within the first minute of capture. Time and location where the animals were first sighted were recorded as were the following environmental temperatures: air temperature in sun (ATS), soil temperature in sun (STS), air temperature in shade (ATSh), and soil tem- perature in shade (STSh). All soil tempera- tures were taken with the bulb barely covered with loose soil, and all air temperatures were taken with the bulb 1 cm from the soil surface. Each day s observations averaged 5 h and starting times were rotated so that ol)ser\a- tions occurred during all daylight hours. Ants (princii)ally Po(!,on()niynnex) were the most abundant insects on the grid. Since ants are important prey of both lizards (Knowlton and Baldwin 1953, Knowlton et al. 1946, Burkholdcr and Tanner 1974), ant activity was monitored in 1977 to determine if lizard activ- ity paralleled ant activity. All Po^onomynncx mounds on the grid were mapped. During each grid sampling, each mound was visited. To inde.x ant activity, we recorded active mounds (>20 active ants) and inactive mounds (<20 active ants.) All data used to analyze activity patterns were from lizards seen on the grid. Tempera- ture data were collected from lizards on the grid as well as lizards marked in surrounding areas (Guyer 1978). Results Environmental temperatures increased parabolically from 0800 to 1500 h followed by a parabolic decline from 1500 to 2100 h (Fig. 1). The hottest microenxironment during all hours except 2000 to 2100 h was STS, and STSh was the coolest microenvironment until late afternoon (1400-1900 h), when ATSh was coolest. The peak for STSh was shifted to the right relatixe to curves of the other three mi- croen\ironments indicating a lag period in- volved witli heating and cooling the soil. Mean cloacal temperatures (± 1 SD) were 33.4 ±4.4 G for P. don^lassi (N = 84) and 33.9 ± 2.4 G for S. <^r(iciosiis (N = 61). Since October 1985 45 40 35H 30 25 20 ui 15- GUYER, LiNDER: IDAHO LlZARDS 609 ••.^« -.• • • • .. •:• ••• P. douglassi < U I 45. UJ K- _i 40- < O 35- 30H 25- 20- 15" I II I I I I I I I I I I I 09 10 11 12 13 14 15 16 17 18 19 20 21 22 • •••• • • • S. graciosus 09 10 11 ll2 13 14 15 I I 16 17 18 19 20 21 22 TIME OF DAY Fig. 2. The relationship of cloacal temperatures oi Phnjnosoma douglassi and Sceloporus graciosus and time of day. Circles are single observations, and triangles are double observations. animals captured on overcast days were much cooler than those captured on sunny days, only lizards caught on cloudless days were used to test for trends among age and sex groups. No differences were found in the dis- tribution of cloacal temperatures among age and sex groups for either species (Kruskal- Wallis test; p < 0.05 for both species), so data were pooled within each species. Cloacal temperatures of P. douglassi during daylight hours could be separated into three thermoregulatory periods (Fig. 2): a morning period (0800-1100 h), a midday period (1100-1800 h), and an evening period (1800— 2200 h). The morning period was character- ized by rising temperatures at each successive hour and wide variability. The midday period was characterized by relatively stable temper- 610 Great Basin Naturalist Vol. 45, No. 4 atures and reduced variability, and the evening period was characterized by falling and widely variable temperatures. During midday P. douglassi were found more fre- quently in shade than in direct sun (68 of 158 in direct sun), whereas during morning (58 of 70 in sun) and evening (23 of 36 in sun) periods lizards were found more often in direct sun. There was a significant association between thermoregulatory period and location of P. douglassi (Chi-square test of association; p < 0.05). However, there was no difference in location of individuals between morning and evening periods (Chi-square test; p > 0.05). One thermoregulatory period was found in S. graciosiis, corresponding to the midday pe- riod of P. douglassi (Fig. 2). No major shifts in cloacal temperature occurred during daylight hours. Because S. graciosus escaped at dis- tances prohibiting the determination of origi- nal microhabitat, locations with respect to sun or shade could not be analyzed. Adult P. douglassi were first sighted 12 April 1977, and juveniles were first sighted 23 April 1977. Fieldwork was begun too late in the season to determine time of emergence in 1976. Seasonal patterns of activity were simi- lar between years so these data were pooled. Adults and juveniles had similar seasonal ac- tivity patterns, with greatest activity occur- ring from June to July and declining activity from July to September (Fig. 3). Adults were last seen 29 August 1976 and 3 September 1977, and juveniles were last seen 18 Septem- ber 1976 and 11 September 1977. One copulation of P. douglassi was ob- served 10 May 1977. Young-of-the-year of this species were first seen 5 August 1976 and 10 August 1977. This group was encountered in- creasingly more often until their sudden dis- appearance in mid- to late September (Fig. 3). Individuals from this age class were last sighted 25 September 1976 and 12 September 1977. The earliest emergence of S. graciosus adults was 16 April 1977, whereas juveniles were first seen 31 May 1977. Again, seasonal patterns of abundance did not ditler, so data for the two years were pooled. Monthly acti\ - ity of adults and juveniles differed in that adult activity peaked during May and declined from June through September, whereas juveniles peaked in August and declined through Sep- tember (Fig. 3). Adults were last seen 29 Au- gust 1976 and 28 August 1977, whereas juve- niles were last seen 29 August 1976 and 17 September 1977. We estimated time of mating for S. gracio- sus using the intensity of orange, the nuptial coloration in females that indicates ovulation (Burkholder and Tanner 1974). In 1977 or- ange color was most intense 3-8 June 1977 with some color persisting through 21 June 1977, implying an early June mating season. Young-of-the-year were first seen 28 August 1976 and 14 August 1977. This group was active throughout August and September (Fig. 3) and was last seen 25 September 1976 and 1 October 1977. Daily activity patterns of adult and juvenile P. douglassi were uniform throughout all day- light hours (Fig. 4). This pattern did not shift during the study period so data for all months and both years were pooled. Daily activity of young of the year followed a bimodal pattern with decreased activity during the hottest part of the day (Fig. 4). Because insufficient cap- tures were made of S. graciosus juveniles and YOY, their daily activity patterns were not analyzed. Adults showed uniform daily activ- itv patterns, with peak activitv occurring from lioO to 1500 h (Fig. 4). The' daily activity of Pogonomynnex on their mounds was bimodal (Fig. 4), with activity depressed during the hottest part of the day. This pattern was simi- lar for all months sampled, so data were pooled. Discussion Mean cloacal temperatures of both species agree with those reported from other popula- tions (Brattstrom 1965, Pianka and Parker 1975 for P. douglassi, and Brattstrom 1965. Burkholder and Tanner 1974 for S. gracio- sus). Sccloporus graciosus had similar cloacal temperatures throughout the daylight hours. However, P. douglassi temperatures differed during the day, with reduced temperatures during early and late hours of the day. This difference appears to be related to daily activ- ity patterns and sleeping sites utilized. Plnjrnosoum douglassi were never observed using rodent burrows or burrowing under loose soil. Se\eral individuals were followed October 1985 GUYKR, LiNDER: IDAHO LiZAHDS 611 ADULTS JUVENILES YOY 2.0 1.5 1.0- 0.5 P. douglassi A« A M J J A S S. graciosus ■ ■ A M J J A S A M J J A S TIME OF YEAR / A S A / A S Fig. 3. The relationship between mean number of lizards seen per sample period and time of year. Circles are males, and triangles are females. 612 Great Basin Naturalist Vol. 45, No. 4 p. douglassi Adults and Juveniles S. graciosus ■I Ant mounds 08 09 10 U 12 13 14 15 16 17 18 19 20 21 TIME OF DAY Fig. 4. Mean number of lizards and active ant mounds seen per sample period and time of day. until they became inactive during evening hours and were relocated early the next morn- ing before becoming active. All were ob- served to spend the night above ground. On several occasions sluggish individuals were captured during cool, early-morning hours. In contrast, S. graciosus were often observed using rodent burrows for escape and were found at burrow entrances during early-morn- ing hours. These sites were apparently used overnight. Sluggish individuals were never observed. The fact that cloacal temperatures of S. graciosus were uniform throughout all daylight hours over which lizards were ob- served indicates that this species does not emerge until body temperatures reach levels at which activity takes place and that this level is maintained throughout the hours over which activity occurs. The morning and even- ing periods of low cloacal temperatures seen in P. douglassi apparently result from the lo- cation of these animals above ground during all hours of the day. Phrynosoma douglassi expedite their morning rise in cloacal temper- ature by locating in the warmest and fastest heating microhabitat (STS), a conclusion sup- ported by Heath (1965). Maintenance of a relatively constant cloacal temperature throughout midday hours, when environmen- tal temperatures continue to fluctuate, is probably accomplished by shuttling between sun and shade and may indicate the level pre- ferred for daily activity. Cloacal temperatures were maintained until STS dropped below this preferred level. Cloacal temperatures then decreased at a rate similar to STS. This cooling during evening hours may be delayed l)y relocating in sunny areas. Pianka and Parker (1975) found variabihty of cloacal temperature in Phrynosoma to be greater than that of any other North American iguanid. They attribute this to relaxed ther- moregulation. In this study both species had approximately equal mean cloacal tempera- tures, but the variation about the mean was approximately two times greater in P. doug- lassi than in S. graciosus. However, the in- creased variability of P. douglassi was due to lizards captured during morning and evening periods, when few S. graciosus were ob- served. If S. graciosus could have been cap- tured at these times, it is likely that similar variability would have been seen. During midday hours both species maintained rela- tively constant and similar cloacal tempera- tures. Thus, during comparable periods of ac- tivity P. douglassi probabK- does not exhibit wider thermal variabilit\ than S. graciosus. Seasonal activity of ju\ eniles differed be- tween the two species. Activity of P. douglassi juveniles was similar to that of adults, whereas activity of juvenile S. graciosus was inversely related to that of adults. The reduced activity of juvenile S. graciosus early in summer was due primarily to the low numbers of juveniles encountered, but this was confounded by poor capture success of smaller lizards. These results differ from those of Burkholder and October 1985 GUYER, LiNDKH; IdAIIO LIZAHDS 613 Tanner (1974), who fonnd aetivity patterns to be similar for S. graciosiis adults and juveniles in Utah. These authors did show that the level of activity maintained by juveniles was below that of adults until August, when adults began hibernating. Whitfbrd and Creusere (1977) showed an inverse relationship between adult and juvenile activity for P. cornutiiin and S. ma^ister in New Mexico. Seasonal activity patterns for adults of both species were similar, with emergence occur- ring in mid-April and hibernation beginning in late August to early September. Peak activ- ity of adults occurred approximately one month earlier in S. graciosus (May) than in P. dougjassi (June). The duration of adult sea- sonal activity was shorter than that reported for southern populations of the same two spe- cies (Goldberg 1971, 1975, Burkholder and Tanner 1974) due to later emergence and ear- lier hibernation. The active season reported for S. graciosus adults in this study is similar to that reported for northern populations (Stebbins 1944, Mueller and Moore 1969). A similar pattern of adult seasonal activity has been found for P. douglassi in Alberta, Ganada(G. Larry Powell, personal communi- cation). Mating occurred at different times of the year in the two species. This is probably due to the difference in reproductive method of the two species. Phnjnosoma douglassi is ovo- viviparous and gestation is approximately three months (Goldberg 1971), whereas S. graciosus is oviparous and gestation is approx- imately two weeks (Goldberg 1975). The time of mating predicted or observed in this study compares favorably with that found by Gold- berg (1971) for P. douglassi and by Burkholder and Tanner (1974) for S. graciosus . Young-of-the-year appeared at approxi- mately the same time of year (early August) in both species. The time of hatching is similar to that reported by Goldberg (1971) and Pianka and Parker (1975) for P. douglassi and by Woodbury and Woodbury (1945), Mueller and Moore (1969), Tinkle (1973), Burkholder and Tanner (1974), and Goldberg (1975) for S. graciosus. Because these studies were done in a variety of geographic locations, there seems to be little geographic variation in tim- ing ol the first clutch in either species. Bimodal activity patterns, particularly in the hot months of July and August, have been reported commonly for southern populations of PJinpiosoma (Parker 1971, Tanner and Krogh 1973, 1974, Baharav 1975) and S. gra- ciosus (Burkholder and Tanner 1974). This pattern was observed only in P. douglassi YOY during this study. Bimodality seen in southern populations appears to provide an escape from high midday temperatures. In this study at least one microhabitat was avail- able at all times that was lower than and one that was higher than or at least equal to the cloacal temperature maintained by the two species. This may explain the absence of bi- modality seen in this study. Because of this absence of midday decline in activity, north- ern popidations may have a longer daily activ- ity pattern than southern populations. The bimodal pattern of P. douglassi YOY may be a result of more rapid heating and cooling of small ectotherms (Brattstrom 1965). Because the duration of seasonal activity is much longer in southern than in northern popula- tions, it has been suggested that southern populations are exposed to predation for longer periods of time. This is thought to be an important factor causing the short life span of southern populations (Tinkle 1973). How- ever, because southern populations may have reduced daily activity compared to northern populations, the effects of seasonal and daily activity may cancel each other between north- ern and southern populations. Thus, specula- tion about the relative role of predation in northern versus southern lizard populations must account for not only the effect of differ- ences in seasonal activity patterns (Nussbaum and Diller 1976) but the effect of daily activity patterns as well. Ant activity on the grid was bimodal. How- ever, activity patterns of both lizard species were not directly related to that of ants. This differs from the observations of Baharav (1975), who found that P. solare in Arizona tracked ant activitv. The fact that P. douglassi 614 Great Basin Naturalist Vol. 45, No. 4 activity did not track that of ants may be re- lated to the broader diet of this species com- pared to P. solare (Pianka and Parker 1975), or it may indicate that the relationship between Phrynosoma activity and ant activity in Ari- zona is not a causal one. Instead both lizards and ants may require escape from high mid- day temperatures in Arizona, whereas lizards in Idaho require no such escape. Acknowledgments We are indebted to O. D. Markham, direc- tor of Environmental Studies on the INEL, for his assistance and encouragement. We thank T. D. Reynolds and R. W. Sehman for assistance in fieldwork and M. A. Donnelly, M. P. Hayes, H. W. Greene, G. L. Powell, J. M. Savage, and S. D. Werman for suggesting improvements in the manuscript. This re- search was supported by a DOE grant (#EY- 7&-S-07-1529) to A. D. Linder. Literature Cited Baharw. D 1975. Movement of the horned hzard Phrynosoma solare. Copeia 1975: 649-657. Brattstrom, B H 1965. Body temperatures of reptiles. Amer. Midi. Nat. 7.3: 37.3-422. BURKHOLDER, G L , AND W W. Tanner. 1974. Life his- tory ecology of the Great Basin sagebrush swift, Sceloporus ^raciosiis ^raciosus Baird and Girard, 1853. Brigham Young Univ. Sci. Bull., Biol. Ser. 14(5): 1^2. Goldberg, S R. 1971. Reproduction in the short-horned lizard Phrynosoma chmglassi in Arizona. Herpeto- logica 27: 311-314. Reproduction in the sagebrush lizard, Scch)ponis graciosns. Amer. Midi. Nat. 93: 177-187. Guyer, C. 1978. Comparative ecology of the short-horned lizard (Phrynosoma douglassi) and the sagebrush lizard (Sceloporus graciostis). Unpublished the- sis, Idaho State University, Pocatello. Heath, J. E. 1965. Temperature regulation and diurnal activity in horned lizards. Univ. (California Fubl. Zool. 64: 97-136. Knovvlton, G. F., and B A. Baldwin, 1953. Ants in horned toad. Herpetologica 9:70. Knowlton, G. F., D. R. Maddock, and S. L Wood 1946. Insect food of the sagebrush swift. J. Econ. Ent. 39: 382-383. McBride R . N R French, A H Dahl, .\nd J E Det- MER 1978. Vegetation types and surface soils of the Idaho National Engineering Laboratory Site. IDO-12084. NTIS, Springfield, Virginia. MlELLER, F C 1969. Temperature and energy charac- teristics of the sagebrush lizard (Sceloporus gra- ciosus) in Yellowstone National Park. Copeia 1969: 153-160. Mueller, F. C, and R E Moore 1969. Growth of the sagebrush lizard, Sceloporus graciosus , in Yellow- stone National Park. Herpetologica 25: 3.5-38. Nussbaum, R. a., and L. V. Diller. 1976. The life history of the side-blotched lizard, Uta stanshuriana Baird and Girard, in north-central Oregon. North- west Sci. 50: 24.3-260. Parker, VV S 1971, Ecological observations on the regal horned lizard (Phrynosoma solare) in Arizona. Herpetologica 27: 333-338. Pl^nka, E R , and W S Parker. 1975, Ecology of homed lizards: a review with special reference to Phryno- soma platy rhinos. Copeia 1975: 141-162. Prieto, a. a , AND W G Whitford 1971. Physiological responses to temperature in the horned lizards, Phrynosoma cornutum and Phrynosoma dou- glassi . Copeia 197 1 : 498-504 . Stebbins, R C 1944, Field notes on a lizard, the moun- tain swift, with special reference to territorial be- havior. Ecology 25: 233-245. Tanner, W. W , and J. E Krogh. 1969. Ecology oi Phrynosoma platyrhinos on the Nevada Test Site, Nye Co. Nevada. Herpetologica 29: 327-342. 1974. Variations in activity patterns as seen in foui sympatric lizard species in southern Nevada. Her- petologica 30: 303-308. Tinkle, D VV 1973. A population anah sis of the sage- brush lizard, Sceloporus graciosus. in southern Utah. Copeia 1973: 284-295. Whitford, W G , and F M Crelsere 1977, Seasonal and >earl\ fluctuations in Chihuahuan deserl lizard communities. Herpetologica 33: 5-4—65. Woodbury, M., and A. M. Woodbury. 1945. Life historv of the sagebrush lizard Sceloporus g. graciosus with special reference to cycles in reproduction. Herpetologica 2: 172-195. SNAKES OF WESTERN CHIHUAHUA Wilnier W. Tanner' Abstract — This is a report on the snakes of western Chihuahua that were taken at intervals from 1956 to 1972. At no time did we attempt to colleet east of Highway 45, rather expending our time in the foothills, valleys, and desert ranges east of the mountains and in the highlands of the Sierra Madre Occidental. Reference is made to reports on the geological and ecological aspects of the area as a whole, but without a major attempt to duplicate previous studies. A brief gazetteer and a map are included as guides. To provide orientation to the area traversed by the John Cross expeditions, a map of the lower Rio Urique and Rio San Miguel is also included. From the area studied, 28 genera and 51 species are listed, with three new subspecies described: two worm snakes (Leptotyphlops huinilis chihuahuaensis and Leptotyphlops diilcis sitpraocularus) and a garter snake (Thatnnophis rufipunctatus ttnilabialis). For a number of species it became necessary to expand the study into populations from adjoining states in Mexico and the United States. Such species as T. rufipunctatus and R. hesperia are examples. Where data were available systematic relationships were implied, as well as ecological and biological data. This study is an outgrowth of a series of herpetological investigations and a number of conversations with individuals who have spent many years in various parts of Mexico. I was constantly enthused with the idea of spending time in Mexico, and, upon my ar- rival at the University of Kansas in 1946, where I became associated with Dr. Edward H. Taylor, I was even more motivated by his constant references to the fabulous herpeto- logical fauna of this neighboring country to the south. My association with the plethodontid salamanders, which through the courtesy of Dr. Taylor served as the subject of my disser- tation, demonstrated there were new species and subspecies to be discovered. The thrill of discovering the new genus Lineatriton left httle doubt in my mind that someday I must visit the area and become more involved in understanding the herpetological faunas of this fascinating republic. I remember distinctly a conversation with Dr. Taylor during my visit to the University of Kansas in 1952, at which time I was trying to prepare for my first adventures into Mexico. After we discussed some of the problems that seemed pertinent to my understanding of the faunas extending from Mexico into the south- western U.S., he advised me emphatically to not worry about getting into the central and southern parts of Mexico, but to start collect- ing as soon as I had crossed the Mexican bor- der. He suggested that perhaps one of the areas most neglected was the state of Chi- huahua. Preliminary studies from the small collec- tion that then existed at Brigham Young Uni- versity (BYU) indicated there were a number of relationships, particularly as I understood them in the serpents, that must yet be under- stood not only from the standpoint of taxon- omy, but also from the standpoint of geo- graphical distribution. Also, a small collection made by Dr. D Eldon Beck in 1931 included a juvenile skink from the vicinity of Colonia Garcia. It was not possible to key this speci- men to any of the materials described and discussed by Taylor (1936) and others, and it was eventually found to represent a new spe- cies (Tanner 1957). My visits with some of the students from the Mormon colonies in north- ern Chihuahua convinced me they were fa- miliar with a fauna that was unique to that area, and certainly not one that was commonly understood by most of the herpetologists with whom I had visited. Thus, in May 1956, in company with Mr. Verl Allman and my oldest son, Lynn, we spent a month in northern Chihuahua, spending most of our time in the area of Nuevo Casas Grandes and Colonia Juarez, with one trip into the mountains of western Chihuahua at Three Rivers (Tres Rios) on the Rio Bavispe. Each year, from then until 1972, trips were Life Science Museum, Brigham Young University, Provo, Utah 84602. 615 616 Great Basin Natufl\list Vol. 45, No. 4 made into Chihuahua at various times during the summer from May into October. My com- panion in 1957 and on through the next three years was Dr. Gerald W. Robison. Perhaps the three most noteworthy trips were made in 1957, 1958, and 1960, during which times much of the mountain area west of Colonia Juarez and west of Ciudad Chihuahua was visited. The trip in 1958 reached its climax at the mining town of Urique on the Rio Urique in the Barranca del Cobre. In 1960 an extended trip was made into the barranca country, where we visited the min- ing town of Maguarichic, and thus were in the vicinity of the area collected by Dr. Irving W. Knobloch (Taylor and Knobloch 1940). Other trips will be mentioned as it becomes impor- tant to do so, but it should be noted that the emphasis on our biological surveys was to cover as best we could in the time available to us the area west of Highway 45 that extends from Ciudad Juarez south to Parral. The only important area that we were not able to visit was the area southwest of Parral across the headwaters of the Rio Conchos and into the headwaters of the Rio San Miguel, and the tableland area near Guadalupe y Calvo. We were, however, fortunate in having received a collection from Mr. John Cross as he tra- versed the area southeast of Guachochic and then boated down the Rio Verde and San Miguel rivers to their junction with the Rio Urique. Mr. Cross also provided some speci- mens from the Urique River north of the town of Urique. Figure 8 indicates the extent of his travels. The state of Chihuahua is 245,612 square km and is the largest state in the Mexican republic. Its territory is approximately 12.5% of the total area of Mexico. Within the state are many diverse ecological habitats ranging from deserts in the eastern and central areas to subtropical areas in the southwest barran- cas. As should be expected, the vegetation is very diverse, from desert and dime floras in the eastern deserts to the pine-oak forests in the western mountains. Between these ex- tremes are the multitude of ecological niches that change with the seasons from dry, rock\-, scrub, brush foothills to helds of waving grasses and other flowering plants after the summer rains have renewed the area. For a more complete study and review of the Chi- huahuan Desert see Morafka (1977). Ecologists have established various life zones in northern Mexico that have included the state of Chihuahua. Perhaps the report compiled by Leopold (1950) and summarized by Knobloch and Correll (1962) gives a gen- eral overview useful to an understanding of the diverse biotic zones encountered in this state. Knobloch and Correll (1962) summarized these zones as follows: 1. Boreal forest: rare, found only at 3000-3200 m (Cerro Mohinara). 2. Pine-oak forests: ranging from pure pine forests to mixed pine/oak and scrub oak. 3. Chaparral: mostly shrubs and small trees other than oak. 4. Mesquite-grassland: the short grass plains, mesquite shrubs and grasslands east of the Sierras. 5. Desert: cactus-euphorbia-yucca and creosote bush. 6. Tropical deciduous forests: short thorn forests of the western barrancas. Chihuahua may best be considered a steppe desert, in which there is no certainty as to when or how much precipitation may occur in any given year or at any locality. At times during July and August, heavy rains occur, producing heavy runoff in local streams, but this is not always widespread nor constant. Thus, Chihuahua has much dry terrain and there are many small or intermittent streams during most of the year. During the rainy season streams may be in flood for a short time, requiring traffic in local areas to wait for the streams to recede. We encountered these conditions several times, but usually for only a few hours or, at most, a day. The heavy runoff" in the desert \ alleys and mountains rutted the roads, most of which were dirt, exposing rock and making tra\'el off the main highways difficult and hazardous. Because of the rugged Sierra Madre Occiden- tal, which extends through most of western Cvhihuahua,, few roads are available, and these often in poor condition. Roads in the mountains were passable, particularly where mining or lumber companies were operating; however, if these companies closed or moved, the roads soon deteriorated. Furthermore, most traffic was from the mountains eastward October 1985 TANNER: Snakes of Western CiiHiUAHUA 617 since the abrupt escarpment and the deep canyons on the west Hmited travel to means other than a car. It was not until the 1960s that the railroad from Ciudad Chihuahua to Creel and on to Los Mochis, Sinaloa, was completed, provid- ing for the first time a route across the moun- tains. Prior to this, travelers going westward from eastern Chihuahua had to first go north, usually into the U.S., and then into Sonora, or go south to Durango. Our trip to Urique took advantage of the newly built railroad grade (then used as a road) from Creel to Cuiteco, but we were three hours going seven miles from Cuiteco to Cerocouhui and from there to Urique, a nine-hour trip by mule. Geology and Physiography of Chihuahua The state of Chihuahua has been part of an area of extensive mountain building that has occurred throughout mucli of western Mex- ico, extending northward through most of western Chihuahua. The Sierra Madre Occi- dental uplift reaches its highest elevation in the south central part of Chihuahua, where the Cerro Mohinara perhaps serves as the highest elevation, reaching to at least 3200 m (10,500 ft). There is a gradual reduction in elevation from south to north, with the moun- tains terminating in low passes of about 1500 m (5000 ft) near the border of Chihuahua and the states of Arizona and New Mexico. The uplift provided for a steep escarpment on the west side, resulting in heavy erosion cutting deeply into the mountains and developing the deep canyons and barrancas. To the east the escarpment is less severe, owing to the ero- sion that filled the adjoining valleys with sedi- ments in the form of many large and extensive alluvial fans. According to geological reports of scientists who have explored the Sierra Madre (Roth- well, Raymond, and Hobart 1901, Knobloch and Correll 1962, Goldman 1951, Hovey 1907, West Texas Geological Soc. 1964, and 1974), Chihuahua is underlain primarily with Cretaceous limestone that has been capped in many localities by rhyolite and basaltic flows in numerous areas along the major fault lines both to the east and west of the uplift. Accord- ing to Forrest Shreve (1939), the state can logically be divided into five physiographical 1. The bajados, gentle slopes in the eastern section of the country, including the lower portion of the Rio Conchos and the large area to the east and northeast. 2. The enclosed basins, particularly of the northwest, wliich are referred to as Bolsons and have no outlets except into the stagnant lakes such as the Guzman and Santa Maria. These lake beds are filled with sandy silt and usually possess shallow water with a high salt content. 3. The elevated plains, the central portion of the state, extending, for the most part, through the central part of Chihuahua, and including the lower desert ranges and valleys immediately east of the Sierra Madre uplift. 4. The Sierra Madre region, the mountain- ous western part of the state. Through this area winds the Continental Divide, separat- ing such streams as the Rio Conchos and the Rio Fuerte, whose headwaters immediately east of Guachochic are divided by a relatively narrow ridge. It is, however, in the western part of the Sierra Madre region that the deep barrancas have been formed — in contrast to the more gentle canyons and streambeds east of the Continental Divide. 5. From the standpoint of a herpetologist, a fifth region, which may be referred to as the barranca area, provides a distinct biological zone. The zone fingers into the mountainous areas from west to east and provides in the deep canyons (such as the Barranca del Cobre) a series of subtropical habitats not found in any other part of Chihuahua. The floor of these deep canyons are of reduced elevation, such as at Urique, to about 600 m (2000 ft), and thus the thorn forests of northern Sinaloa have extended far into these narrow canyons. It is estimated that the height of the barranca rim above and west of Urique is at least 1800 m (5800 ft). The sediments, occurring in much of west- ern Chihuahua, consist of limestones capped in many areas by a large outflow of tuft and other loose, volcanic materials. Such forma- tions have been subject to rapid erosion, re- sulting in the deep canyons along the western escarpment of the Sierra Madre. Only below the headwaters of such streams as the Rio Urique and the Rio Conchos occur waterfalls and severe rapids where the streams have encountered more resistant sediments. It 618 Great Basin Naturalist Vol. 45, No. 4 should be noted that during the rainy season the streams may become torrents and deepen rapidly these steep, deep canyons (Shreve 1944). Because of the rapid erosion through the various stratifications of the Sierra Madre uplift, deposits of minerals have been ex- posed, primarily silver and associated lead. However, copper is also exposed in various areas and has been responsible for the name "Barranca del Cobre." Because of the numer- ous ore deposits, areas in Chihuahua that would not otherwise be explored biologically may be reached by roads built to the mines. The mountainous area varies in width from about 130 to 160 km (80 to 100 miles) in the south, that is, west of Parral and from 65 to 80 km (40 to 50 miles) west of Colonia Juarez. Within these areas there may be many com- paratively flat park and meadow areas as well as gentle slopes with considerable timber. Most of the headwaters of such streams as the Bavispe, the Papagochic, the Conchos, and some of the tributaries of the Urique and Oteros drain these highland meadows and parks, and it is not until the streams reach near the west or east escarpments that the terrain is cut into deep canyons or barrancas. Perhaps the most noted of the barrancas is that of the Barranca del Cobre, in which the Rio Urique flows; it is not a swiftly flowing stream in its canyon toward the headwaters, but downstream it soon becomes a series of waterfalls and rapids, in some areas deeply undercutting the lateral canyon walls, in turn resulting in much shearing of the steep walls in the riverbed below — even to the point of permitting the river to flow among and under great masses of boulders. It was this type of terrain that made our river running trip in 1963 a failure (Fig. 9). Considerable publicity about this trip oc- curred in Chihuahua and the southwestern United States. Actually, we descended the river for about 16 km (10 miles); during this time, several rapids and waterfalls were en- countered. These did not stop us, but later, when the river partly or totally disappeared under large, granite lioulders for great dis- tances, it became impossible to continue. The next year John Cross and his associates did enter the Barranca del Cobre near the Di- visidero, and they ran the Rio Urique to the Rio El Fuerte. They also entered the Rio San Miguel south of Guachochic and ran it to the El Fuerte. During these river running trips, Mr. Cross made collections that included sev- eral genera and species not previously known to occur in Chihuahua. The Cross collections (three trips) are deposited in the BYU her- petological collection. Previous Herpetological Surveys Perhaps the first significant herpetological survey of Chihuahua was that of Edward A. Goldman (made under the direction of Dr. E. W. Nelson). Although his work was done from the fall of 1898 into October of 1899, his study was not published until July 1951. Much of his report is concerned with the geography, geology, and flora of various local- ities, but herpetological specimens were ob- tained and deposited in the U.S. National Museum. It should also be noted that the Goldman-Nelson travels in Chihuahua did not cover much of the central part of the state but were confined to the northern area, Nuevo Casas Grandes and areas to the north and west, and the southern area from Parral westward across the mountains to Batopilas and the Barranca del Cobre. During the 1930s and early 1940s, Dr. Irv- ing W. Knobloch investigated the fauna and flora of the west central mountains of Chi- huahua in the vicinity of Majorachic. This col- lection was reported by Taylor (1940) and by Taylor and Knobloch (1940). Several other biological surveys have also been made into Chihuahua; the one by Dr. James D. Ander- son was confined largely to the Sierra del Nido area, and his material is primarily deposited at the University of California at Berkeley. In 1942 Dr. Hobart M. Smith briefly re- viewed Mexican and Central American Thamnophis and described as new the sub- species T. ordinoides errans. He also spent time studying the herpetological fauna along the Rio Santa Maria and its environs. Other field trips originating at the Univer- sity of Kansas, the Universit>' of Illinois, and the University of Texas at El Paso have added to the specimens available for study. A collec- tion from Yepomera was made by members of the University of Arizona, and a special study of Tlumuiophis and Natrix was conducted by Dr. Roger Conant. A trip by Kenneth L. ( )ctober 1985 Tanner: Snakes of Western Chihuahua 619 Williams, Edward O. Moll, Francois Vuilleu- niier, and John E. Williams (Smith ct al. 1963) down the Conchos River in Angust 1962 was one of few that has attempted to collect in the desert areas east of the main highway. Reynolds and Scott (1982) reported on a series of 20 species taken along Highway 16 between Villa Aldama and El Pastor. This was a study of food and habitat selection in northeastern Cvhihuahua from 1975 to 1977. There un- doubtedly have been other collections made, but those indicated above apparently repre- sent the most important collections. Gazetteer Some of the localities from which many of our specimens were collected are listed below with various comments concerning their loca- tion and general habitat (Fig. 1). Colonia Juarez This is a Mormon colony established before 1900 and serving since then as headquarters for the colonists. It is basically a farming com- munity with a considerable emphasis in re- cent years on orchards, with apples and some pears and peaches the primary crops. It is located along the Rio Piedras Verdes which flows directly eastward from the mountains west of the colony. From its eastern border one looks toward the east at the old settlement listed on maps as San Diego. Rolling hills surround the town, with the escarpment of the Sierra Madre Oc- cidental showing sharply to the southwest. To the north and northeast are rolling hills and a relatively broad, open canyon known as the Tinaja extending westward into the moun- tains. To the northeast of the mouth of the Tinaja are some rolling hills that have been productive in our collecting program. The en- virons of Colonia Juarez support a herpetolog- ical fauna in which not only the more desert species occur along the base of the escarp- ment, but also species more commonly found in the mountains descend into the mouth of the canyons and along the streams. The climate in this part of Chihuahua is relatively dry from September through the winter months and into June. The summer rains usually begin in late June and continue intermittently through July, August, and early September. This part of Chihuahua is a steppe desert, receiving much of the moisture from the southeast as the tradewinds circle into the area from the Gulf of Mexico, but with some storms being initiated from the south- west. The winters are mild, but with some precipitation in the form of snow and an ade- quate period of cold to favor temperate zone vegetation. Some of the simimer precipitation is heavy, producing considerable runofiP and, at times, closing roads and flooding the lowlands. When the first heavy rains come in July, the desert flats east of Colonia Juarez and N. Casas Grandes are alive with amphibians; in fact, in some areas they become so numerous that it is impossible to drive the roads without a continual popping sound as the inflated indi- viduals are mashed on the road. In this area and throughout some of the valleys to the east, the dry season provides an opportunity for collecting mainly lizards, with only a small population of snakes extant. During the rainy season, there is a greater percentage and a greater variety of snakes to be found along the roads and under rocks and other debris throughout the area. Ciudad Chihuahua The basin in which Ciudad Chihuahua is located drains from the northwest toward the east. Our collecting was done primarily north- west of the city; this area was actually a lower, southern end of the Sierra del Nido. We were concerned with the foothill areas and did not move into the higher elevations of this range. The eastern sloping foothills and their alluvial fans provided a habitat in which the scrub brush and cacti were separated so as to provide no difficulty in walking through the terrain. Also, there were a number of rocky outcroppings and boulder-strewn flats provid- ing an opportunity for collecting, particularly during the rainy season that brought the more secretive burrowing forms to the surface. Ex- tending westward from Ciudad Chihuahua for many miles, almost to Cuauhtemoc, is a ter- rain similar to these foothills. 620 Great Basin Naturalist Vol. 45, No. 4 MAP OF CHIHUAHUA CASAS COLONIA PACHEjCO.l gavilaJn* •; GA CHUHUICHUI Fig. 1. MapofChihualuu] Maguarichic Maguarichic is located on a bluff oxerlook- ing the deep canyon of the Hio Oteros. It is a mining town and was at one time a thriving connnunity. We were there for three days during the rainy season of 1958 and were suc- cessful in securing a number of specimens, some of species not previously reported by Tavlor and Knobloch (1940) from Majorachic. We attempted to visit Majorachic, but the roads were so nuidcK that we were unable to ascend some of the hills leading into the town. October 1985 TANNER: Snakes of Western Ciiiiiuahua 621 We did, however, remain in the general area a few miles southeast of it. The entire area here was covered with vegetation, a long leaf pine, oak, madrone, and other trees and shrubs. This area is on the high, mountainous plateau and drains southwestward into the smaller tributaries of the Rio Oteros. The area between San Juanito and Maguarichic is the high, mountainous area lying mostly to the west of the Continental Divide that lies a few miles west of San Juanito; thus, there are in this generally mountainous area not only pines, but also Douglas-fir and Chihuahua spruce, particularly on the northern sloping ridges. Cuiteco-Cerocouhi Area We left Creel and traveled along the rail- road grade which, in 1958, was serving as a highway while the Mexican government built the railroad on through the mountains to Los Mochis in northern Sinaloa. The grade had been finished to Cuiteco, and we were privi- leged to stay in some government-owned cab- ins while we organized for the short run to Cerocouhi. The latter was only seven miles from Cuiteco by road but required over two hours to traverse. At Cerocouhi we were housed in some government buildings that had been used as a headquarters during a mosquito abatement survey. It adjoined the Catholic church and a nunnery. We spent two days there while Dr. Knobloch arranged for a mule train to take us to Urique. The area in and around Cuiteco and Cerocouhi was on small tributaries of the Rio Oteros, and during much of the time that we were there the streams were at flood stage. The area to the east of these towns was forested with oak, madrone, and scattered (particularly on the higher ridges) pine. Much of the area in and around Cerocouhi had been overgrazed, but in the small canyons to the east across the main stream were boulder-strewn alluvial fans mixed with some oak forests that pro- vided suitable habitat for a number of amphib- ians and reptiles. Urique Urique is a mining town that has been largely abandoned as far as mining is con- cerned. We discovered here the same situa- tion that occurred in other almost-abandoned mining towns where the inhabitants were left stranded after the mines closed; although Uricjue had been a very prosperous mining settlement, only a few people were working the mining area at the time we were there, and any ores mined were hauled out by burro packtrains. The town is located on the Urique River at the bottom of the Barranca del Cobre. The vegetation there is subtropical. We stayed in a large building next to the dwelling of the "Presidente," and to the side of the building were mangos, wild figs, and other subtropical trees and shrubs. A short distance away from the river on the slopes were thorn forests with the leguminous cat claw predominating and serving as a real deterrent to one wishing to hurry through. Although we were forced to select areas where we could collect without vegetational hindrances, we found a few areas of rocky outcroppings along the river. We were led by some of the Mexican children to the old stone wall around the cemetery, where we secured a number of iguanids and other rock-dwelling species. It may be of interest to note that in some of the old graves the deceased were not fully buried but placed in crypts and could now be seen through the cracks as skeletons. Many of these tombs had been elaborately constructed during the height of the mining boom of Urique. Three Rivers Area The Tres Rios area of western Chihuahua and eastern Sonora derives its name from the junction of three streams that form the Rio Bavispe. These streams (Chuhuichupa Creek, Trout Creek, and Black Canyon Creek) flow northwest as does the Bavispe. We collected along the stream, in the small side canyons, and on the sides of the main canyon. Although the canyon has steep slopes, there are ledges and talus to provide for a variety of habitats. Oak was dominant on the lower slopes, with some long leaf pine on the higher ridges. Along the streams were sycamore, cotton- woods, and willows. This area was visited in May 1956 and in June 1958. 622 Great Basin Naturalist Vol. 45, No. 4 Chuhuichupa This Indian name refers to the mist that forms in the valley on the cool mornings fol- lowing afternoon or evening thunderstorms. The few American inhabitants used the term "valley of the mists" to explain the term. The valley lies in a high mountain basin, draining to the north and then west to join with other streams (Black Canyon and Trout Creek) at Three Rivers to form the Rio Bavispe. Above the town of Chuhuichupa is a series of large springs that are the major sources of Chuhuichupa Creek. To the west through a low pass is Black Canyon Creek, which flows northward in a parallel canyon to join Chuhuichupa Creek at Tres Rios. The head- waters of these streams are in open basins with gentle slopes and meadow pastures. The steeper slopes and ridges surrounding these basins are forested; thus, these areas provide a variety of habitats for a large number of spe- cies. Creel The city of Creel is located in a high basin just north of the Barranca del Cobre. During our work in this area, this was the terminus of the railroad that has since been extended across the mountains to Los Mochis, Sinaloa. The area around Creel is a series of rolling hills and basins with slow-flowing streams. As one moves south, the canyons deepen as streams join the deep canyon of the Rio Urique. Besides being a railroad junction. Creel serves as a road junction for travel to the south and west. The La Bufa Road connected Creel to the headwaters of the Rio Uri(|ue and to Batopilas and ore mines south of the west- flowing Rio Urique. The area of Creel and north to San Juanito is in the higher mountain valleys, with the Con- tinental Divide lying between them. Thus, the roads into these valleys were difficult and, during the rainy season, often impassable. La Bufa Road This area about 25 miles southeast of Creel on a small, south-flowing stream with low, boulder-strewn hills was a very productive area. We were there in July during the rainy season. Low-growing trees (oak and madrona) and shrubs mixed with grass provided the general habitat. This area is on the plateau north of the Barranca del Cobre. List of Genera and Species The serpent fauna of the state of Chihuahua is rich, primarily because of the diverse habi- tats that occur in various areas of the state. Of the various faunas that have invaded the state of Chihuahua from practically all sides, we can now recognize from the Smith-Taylor check- list (1945) a total of 18 genera and 39 species and a total species-subspecies of 43. Their list has been modified by the deletion of some species, such as Crotahts semiconiutus, and by altering the status of others, such as placing Lampropeltis knoblochi as a subspecies of L. pyromclana. A number of genera and species have been added so that there are now at least 28 genera and 51 species plus an additional 6 subspecies. To understand the ser^Dent fauna of Chi- huahua, it became necessary to review speci- mens from those states (Mexican and United States) adjoining Chihuahua (Sonora, Sinaloa, and Durango and Texas, New Mexico, and Arizona). The strategic location of Chihuahua lends itself well to the reception of numerous species into diverse habitats within the state, which necessitated this review. Throughout Chihuahua there are only a few species and subspecies that may be considered endemic to the state. Several of these inhabit the mountainous area of western Chihuahua and represent subspecifically distinct linear ex- tensions of species into the narrow mountain- ous corridor. In an attempt to understand such species as Thamnophis nifipunctatus, which extends throughout Chihuahua and into the adjoining states to the north and south, I have taken the liberty to investigate again the basic charac- teristics of this wide-ranging species and to interpret, from the data axailable, the varia- tions in the populations as I have encountered and understood them. It should, therefore, be understood that where deemed necessary consideration is given to the geographical overlap of adjacent populations of \arious spe- cies and subspecies. October 1985 TANNER: Snakes of Western Chihuahua 623 Family Leptotyphlopidae Two species of the genus Lcptotyphlops have been found in Chihuahua. The first one treated befow was reported by Cope (1879) from a collection made by Edward Wilkinson, Jr., at Batopilas, a small mining town located on a tributary of the Rio San Miguel, itself a tributary of the Rio El Fuerte (see Smith and Mittleman 1943, 1944, Cope 1896, 1900). Ex- cept for the summer rains, this area is dry and slopes to the southwest (Goldman 1951). The second record was recently reported by Murphy (1975) from 5 km (3 mi) NW of Chilmahma, a suburb of Chihuahua City. We have collections from 9 to 11 km (6 to 7 mi) NW of Chihuahua City and from Colonia Juarez. The area northwest of Chihuahua City is an eastward-sloping alluvial plain, strewn with rocks and with rocky hills extending onto the plain from the higher hills to the west. Specimens were taken in late July and early September from under rocks while the soil was moist. Those taken at Colonia Juarez were collected by children in the school yard. Pre- sumably all specimens were on the surface and all were collected during April and May, which is usually the dry season. The species of Leptotyphlops apparently in- habit the foothills on the eastern edge of the Sierra Madre Occidental. The records now available suggest that a comparatively narrow area, extending from north (Colonia Juarez- Casas Grandes) to south (Chihuahua City- Cuauhtemoc) and undoubtedly on south- ward, supports worm snake populations. In these foothills, we have worked numerous sidehills and driven numerous hours night collecting, with only limited success. What is seemingly true for the eastern foothills is also apparently the case for the western foothills. In the west, however, the foothills are mostly in the states of Sonora and Sinaloa, so the only suitable habitat in western Chihuahua is on the edge of the low river valleys of the Rio Fuerte and its tributaries. Leptotyphlops humilis dugesii (Bocourt) Catodon dugesii Bocourt, 1881, Bull. Soc. Philoni. 7(6):81. Colima, Mexico. Leptotyphlops humilis dugesii Klauber, 1940a. Trans. San Diego Soc. Nat. Hist. 9:129. No specimen from Chihuahua is available. The hsting of this subspecies is based on the report of Cope (1879), reiterated by Klauber (1940a), of a specimen from Batopilas in south- western C'hihuahua. Bogert and Oliver (1945) reported a specimen from Alamos, Sonora, and Hardy and McDiarmid (1969) listed sev- eral records for Sinaloa. It is thus suspected that if specimens are taken along the Rio Fuerte and its tributaries they would be L. h. dugesii (Hahn, 1979a). In 1966, we received three specimens (BYU 23913-15) of this subspecies from 15.9 km (9.4 mi) W of Autlan, Jalisco. The follow- ing scale characters are significant: 14 rows around body, 12 rows around tail, 212-236 in dorsal row, prefrontal a little longer than the frontal but not wider, 5th dorsal enlarged and wider than other dorsals. There are 7 rows of dorsals with heavy pigmentation. In contrast to the data presented by Klauber (1940a), there are fewer dorsals, 212- 236 (Klauber gives 235-257), and the en- larged prefrontal may be unique to the west- ern humilis subspecies. These specimens ex- tend the lower extreme in the dorsals and indicate that total variation in the cline for the dorsals in humilis may be as much as 100 scales, with the known range 212-309. A specimen from west of Autlan (BYU 23914) was preserved with its mouth fully opened. There are only 3 infralabials on each side of the mental (Fig. 2A), and the 2nd and 3rd infralabials are pigmented nearly as heav- ily as the lateral and dorsal head scales. The other two specimens have the mouth closed but do show some pigmentation on the same scales. I have noted this pigmented character- istic in no other specimens of humilis. The color pattern in the Jalisco specimens shows 9 rows of dorsals with dark brown pig- mentation, 7 rows completely pigmented, and the 2 adjoining lateral rows with half the scales pigmented. Leptotyphlops humilis chihuahuaensis , n. subsp. Holotype.— BYU 17000, adult male from 10.7 km (6.7 mi) NW of Ciudad Chihuahua (west of Highway 45), collected by W. W. Tanner and W. G. Robison, 21 July 1960. Paratypes.— BYU 15211 and 16999, topo- types; MVZ 57331, 5 km (3 mi) NW of 624 Great Basin Natur\list Vol. 45, No. 4 Fig. 2. A, Lower lip oi' Lepfotyphlops hiimilis du^csii in spt-tinien BYU 2.3914; M = mental, 1-2-3 lower labials. B, Dorsal head scales in BYU 23929, same suhsp. C;, Dorsal head scales L. /). chiliualiuacnsis , BYU 17000. Chilmahma, a northwestern suburb of Chi- huahua City, Mexico. Diagnosis. — A subspecies of L. humilis having only 10 scale rows around the tail, whereas all others, except h. segregus and probably h. tenuiciilus, have 12. From L. h. segregus it is distinguished by having a low number of dorsal scales (253-257), and the first 4 dorsals of approximately equal size, the 5th greatly enlarged (compare Figs. 2B and C). Description.— Body cylindrical, head only slightly if at all distinct from body, tail slightly reduced and with the characteristic terminal spine. Snout rounded and extending beyond the lower jaw. From the ocular cau- dad the head merges into the body with no apparent deviation, so that from a dorsal view the head is not distinct from the body. The longest specimen (the type) is 158 mm in total length, with a tail of 19 mm from tip of spine to posterior edge of the vent. The paratypes are 133 and 112 mm, respectively, in total length, with tail lengths of 8 and 6 mm. The body has 14 rows of smooth, imbricate scales, uniform in size except for those on or near the head. There is a reduction to 12 rows a few scales anterior to the vent, and a final reduction to 10 rows at or just posterior to the vent. The anal scale is single and triangular in shape. The middorsal scale coimt is 253, not counting the rostral and tail spine. There arc 17 subcaudals in the type. The ])arat>pes, both females, have 256 and 257 dorsals and 16 and 17 subcaudals. The rostral curves over the snout between the 2 large nasals to contact the prefrontal, which is in contact laterallv with the nasals and oculars and posteriorly with the frontal. Nasal is divided through the nostril, with the upper nasal largest and the lower nasal form- ing the lateral edge of the snout between the rostral and first supralabial. Ocular large, ex- tending from lip to contact the prefrontal and frontal dorsally, eye spot near middle oi scale and above the level of the nostril. Posterior supralabial large, narrowing dorsally to con- tact the parietal below the level of the eye. Parietals and occipitals are elongate scales ex- tending from the dorsals laterally to be par- tially separated by the temporal, wedging be- tween them at their lateral ends. The first 4 dorsals are of approximately the same size, with the lateral edges forming nearly straight lines to the 5th dorsal, which is much enlarged laterally and is the largest scale in the dorsal series. The mental is broad and narrow, joined laterally by 3 infralabials on each side, with the posterior scale the largest; a single chin shield contacts the mental and divides the first infralabials. The middorsal scale row and the three ad- joining lateral body and tail rows on each side are finely pigmented, but the next lateral rows show a reduction of pigmentation near the middle of each scale and the ventral 1/4 to 1/2 of the scale row without pigmentation. This color pattern extends from the head posteri- orly to the tail spine. Below the dorsal pig- mentation the sides and ventral scales are light cream or a light bufl. Hkmarks — The relationship of clii- huahiiactisis is with segregus primarily be- cause of the 10 rows of scales on the tail and its October 1985 TANNER: SNAKES OF WESTERN (^HIHl'AHUA 625 geographical nearness. The occurrence of sc^- regus in south Texas and adjoining eastern Coahuila, and the fact that the Chihuahua basin (type locaHty) lies in the drainage of the Rio Conchos, which drains to the Rio Grande in southwest Texas, supports this assumption. These facts suggest that these two subspecies had a common ancestry. Whether the desert areas of eastern Chihuahua and western Coahuila have served as an isolation barrier is as yet unknown, since no specimens are avail- able and little collecting has been done in these areas. Hahn (1979) cites two localities (on map) for Chihuahua. The one from central Chihuahua must be from 5 km (3 mi) NW Chilmahma (Murphy 1975). The one from south central Chihuahua may also belong to this subspecies. A specimen of segregens from Coahuila, Mexico (USNM 93593), has 286 ventrals, 12 caudals, prefrontal larger than frontal and in- terparietal, and the 5th dorsal enlarged. Klauber (1940a) lists the dorsals for segregus as 261(271)275; a larger series from Coahuila would probably increase the known average for dorsals and add credence to the unique- ness of the Chihuahua population. In preserving the type of L. h. chihiia- huaensis, the mouth was opened so that the infralabials were exposed. In most preserved specimens the mouth is closed and the lower lip scales are partially, if not entirely, cov- ered. This is particularly the case for the most posterior infralabial scale. In most of the liter- ature (Klauber 1940a, Taylor 1939c) I have found a listing of 4 infralabial scales. Klauber believes there are 4 in L. h. segregus and other subspecies of humilis. Murphy (1975) also lists 4 for h. Vmdsaiji and h. levitoni, stat- ing that the first lower labial is very small. I have not seen Murphy's specimens, but those humilis available to me do not have a small scale by the mental. In fact, there appears to be but 3 infralabials in humilis, with the 3rd, the largest, wedged back to the corner of the mouth and with only the lower edge exposed when the mouth is closed. One or two scales may appear to be infralabials, but they contact the larger 3rd infralabial beneath the large, overlapping posterior upper labial and do not reach the inner margin of the lip. A humilis specimen from Jalisco (BYU 23913) was in pre-ecdisis condition, and the scales from around the lower lip were removed intact. After staining in eosin to show more clearly the sutures between the scales, it was appar- ent that there were only 3 scales on each side of the mental. Other humilis specimens were examined with the same results. The mental scale is grooved on each side, giving the ap- pearance of a small lateral scale, but I could not see any suture to indicate an additional scale. During preservation, the hemipenes of the type were everted. They are elongate tubular structures without spines, but with numerous grooves and irregularities. Leptotyphlops dulcis supraocularis, n. subsp. HOLOTYPE— BYU 30426, an adult taken at Colonia Juarez, Chihuahua, Mexico, by Vir- ginia and Herman Hatch during April 1965. Paratites.— BYU 1421, 19131, 30427-28, and 32417, topotypes. Diagnosis. — A subspecies of L. dulcis, with the anterior supralabial divided as in dul- cis dissectus but differing from dissectus in having the supraoculars elongate and wedging between the prefrontal and frontal to enclose or nearly enclose the prefrontal. The inter- parietal (3rd scale in dorsal row) is much larger than either the frontal (2nd) or the interoccipi- tal (4th) and is approximately equal in size to the enlarged 5th scale. Prefrontal noticeably larger than frontal. The occipital is not di- vided. Description. — The body is cylindrical from head to tail. Head only slightly distinct from body, with the snout slanting forward and downwards, beginning at about the pre- frontal. Tail short and terminating in a sharp spine. Total length 219 mm, snout-vent length 208 and tail length 11 mm. The topo- types range in total length from 105-257 mm. The ratio of the body length (S-V) to the mid- body diameter in four specimens averages .02. The tail length is approximately 5.0% of the total length. There are 14 rows of smooth, imbricate scales on the body from just posterior to the occipitals to about 5-7 scales anterior to the vent, where the rows are reduced to 12; just posterior to the vent the rows are reduced to 10 on the tail. The anal is single and triangu- 626 Great Basin Naturalist Vol. 45, No. 4 lar, followed by 14 subcaudals that range in the paratypes from 13 to 15. There are 237 middorsal scales, beginning with the prefron- tal and counting to, but not including, the tail spine. Type and paratvpe series range from 231(238) to 246. The rostral is the largest head scale and curves from the underside of the lip dorsally and posteriorly to contact the prefrontal at about the level of the eye. The rostral is only slightly narrowed from the snout between the nasals and has a rather broad, rounded contact with the prefrontal. On the upper lip are 5 scales extending posteriorly from the rostral: nasal, divided through the nostril to form an upper and a lower scale; first supralabial, di- vided into 2 scales by a vertical suture; ocular, with the eye spot just above the level of the nostril; and a large posterior labial that over- laps the posterior infralabials. Of this series, only the rostral, nasal, and ocular contact the dorsal head scales. The supraoculars are elongate (Fig. 3), ex- tending from their contact between the nasal and ocular posteriorly and medially to contact the parietal and to wedge between the pre- frontal and frontal. In two paratypes, the 2 supraoculars are in contact, thus separating the frontal and prefrontal. The supraoculars are about the same size as the prefrontal, which is larger than the frontal; the interpari- etal is larger than either the frontal or interoc- cipital, and approximately equal in size to the 5th dorsal. In none of the series are the first 4 dorsals of about equal size. From smallest to largest they are: frontal-interoccipital-pre- frontal-interparietal. The 5th dorsal is en- larged and is equal in size to the interparietal or slightly larger. The mental is broad and short with 3 infrala- bials on each side. The posterior infralabial is large and extends under the overlapping pos- terior supralabial to the corner of the mouth. A scale just posterior and lateral to the poste- rior infralabial appears to be a 4th infralabial. It is overlapped by the supralabial so that its true relationship to the 3rd infralabial and the lip cannot be seen as it joins the posterior infralabial, but it does not reach the lip. Color PATiERN.^Thc 5 dorsal rows of scales are finely pigmented, and the dorsal edges of the adjoining rows show some pig- mentation. The snout, including the most Fig. 3. Dorsal head scales of Leptotyplilops diilcis suprauciilaris (BYU 30427). anterior part of the rostral and extending lat- erally to include the area surrounding the nos- tril, is not pigmented. Those parts of the ros- tral, nasal, and ocular that are pigmented have small papillae, but these do not appear on the nonpigmented areas. Those scales on the lat- eral and ventral parts of the head and body are not pigmented. Remarks. — An attempt to key the Chi- huahua specimens (Klauber 1940a, Taylor 1939c) did not satisfy the key characters. The supraoculars were shaped more as in L. alb- ifrons than in the figures by Klauber (1940a) and dissect us specimens available. In fact, the first specimen collected in Colonia Juarez (1959) had the supraoculars completely sepa- rating the prefrontal from the frontal, and the occipitals are not divided. As additional speci- mens became available, it was obvious that this supraocular character was apparently unique and not represented in cither d. dulcis or (/. dissectus. A comparison of Figure 2 with those presented by Klauber (1940a) illustrates this basic difference in the dorsal head scales. I have not examined as many specimens from Texas, New Mexico, and Arizona as did Klauber; however, those seen from southern Texas are dissectus as described bv Klauber. October 1985 Tanner: Snakes of Western Chihuahua 627 A specimen from 4 km (2.5 mi) NW of Glen- wood, Catron County, New Mexico, has ap- proximately the same head characters as the Chihuahua series except that the interparietal is not greatly enlarged. Although this locality is northwest of the type locality of L. dulcis dissectus (near Lake Valley, Sierra County, New Mexico), it is west of the Continental Divide and in the Gila River drainage. A specimen from Thatcher, Graham County, Arizona, has characteristics of dissec- tus and supraocularis- that is, the supraocu- lars are elongate, but the occipitals are di- vided and the interparietal is not greatly enlarged. The terrain favors intergradation of the populations in southwestern New Mexico, southeastern Arizona, and the area directly south in northwestern Chihuahua. Specimens in Arizona and New Mexico west of the Conti- nental Divide are thus expected to show inter- grading characters between d. dissectus and the Chihuahua subspecies d. supraocularis, particularly those representing populations from the Gila River drainage to the west, and from the desert ranges to the south and into extreme northwestern Chihuahua. A specimen (KU 44264) from Rancho San Francisco in the extreme northwest corner of Chihuahua has all the characteristics of dulcis dissectus. The prefrontal is noticeably smaller than the frontal, and the supraoculars are equal in size to the prefrontal and not elongate and thus permit a wide contact between the frontal and prefrontal. This may be the speci- men cited (map) by Hahn (1979b). A specimen (BYU 41893) from Monterrey, Nuevo Leon, is heavily pigmented, with the dorsal rows dark brown and the more lateral and ventral scales a light slate color. The supraoculars are not curved to enclose the prefrontal, and the following scales (frontal, supraoculars, interparietal, and interoccipi- tal) are of approximately equal size. There are 230 dorsal scales, and the prefrontal is the smallest of the series. Small series of d. dulcis (19), d. dissectus (15), and d. supraocularis (7) provide the fol- lowing data for the dorsals: d. dulcis 212(224)239; d. dissectus 206(229.35)239; d. supraocularis 231(237.7)246. There is over- lapping in all subspecies; however, a cline is evident from east to southwest. Family C'olubridae The colubrid fauna of Chihuahua is large and diverse primarily because of the many different habitats existing from the deserts to the foothills and into the mountains. This di- versity increases the number of available eco- logical niches and supports a home for 24 gen- era and at least 40 species, with 5 additional subspecies in the state. Although the greater part of the state has been traversed by collec- tors and a number of collections made other than those of which 1 have been a participant, there are yet additions to be made to this large and interesting fauna. In view of the limited collecting in the desert areas east of Highway 45 and in the rugged barranca terrain in which the west coastal thorn forest has invaded the deep canyons, these areas undoubtedly con- tain additional genera and species for the state of Chihuahua. Salvadora g. bairdi, among others now occurring in northeastern Sinaloa but not yet reported for Chihuahua, may oc- cur there. The colubrid fauna of Chihuahua is com- posed of genera and species with strong affini- ties to the desert habitats that surround the state to the east, north, and northwest, per- mitting an invasion to occur from the deserts into eastern and northern Chihuahua and a major movement of the central Mexican spe- cies through the mountains from the south into the western highlands. It is important to note the species and subspecies that inhabit the central plains and foothills, extending from the area near Nuevo Casas Grandes south along the eastern foothills to the state of Durango. In this area some species and sub- species appear to have been isolated since the recent Pleistocene. This isolation has been intensified by the persistence of the desert areas to the north and east, and in turn these desert areas, extending as they do through northern Chihuahua into southern Arizona and New Mexico, have limited the northern movement of the montane species such as Thamnophis melanogaster and Storeria store- rioides. Therefore, one investigating the colu- brid fauna of Chihuahua should take into ac- count more than the great diversification of the habitats resulting primarily from changes in elevation from the northeast to the south- west. The geological past has contributed to 628 Great Basin Naturalist Vol. 45, No. 4 the changing habitat conditions and has re- sulted in some isolation as well as permitting certain species to expand their ranges into the various ecological areas in western and south- western Chihuahua. Arizona elegans expolita Klauber Arizona elegans expolita Klauber, 1946, San Diego Soc. Nat. Hist. 10(17):340-343. 2.2 mi SE N. Casas Grandes, 1 (BYU 15254). Tinaja Vallev, 2 mi S of highway, 2 (BYU 13899, 143.34). 1 mi S Sueco, 1 (BYU 17105). 3 mi W Galeana, 1 (BYU 17106). 17 mi N Chihuahua City, 4 (BYU 1.5293, 15299, 1.5342, 1.5346). 10 mi S Chihuahua Citv on Route 45, 1 (BYU 32041). 12 mi S Los Nieves, Durango, 1 (BYU 14073). 7.4 mi N Moctezuma, 1 (UTEP 4099). 12 mi from Chihuahua Citv, 1 (MVZ 436.56). 1 mi W Ojo de Laguna, 1 (MVZ 73030). 20 mi SE Ciudad Camargo, 1 (MVZ 80001). The original description cited a type and two paratypes (not examined). With the present series and including data for the types, 17 specimens are available for study. In all, scale and color patterns, range, and variation in scalation have been increased, although the averages are not appreciably changed. The scale rows at midbody are 27 in all except one specimen fi^om 27 km (17 mi) N Chihuahua City with 26 rows. The variation occurs anteriorly with 25—29 rows and before the vent with 17-20 rows. Two rows adjacent to the ventrals have larger (wider) scales than those near the dorsum. The head scales show little variation: supralabials 8-8, infralabials 12 or 13, loreals 1-1 except for one with 1-2, preoculars 1-1, postoculars 2-2, and temporals 2-3 or 2-4. Ventrals range from 192 to 207. The males are 192-203, and females 198-207. Caudals are 39-51, with males 50-51, and females 39-49. When ventrals and caudals are combined for males (246.8) and females (247.3), the averages are nearly equal. The total length/tail length ratios were: males 14.2, 14.6, 14.7, 14.8, and 15.1; females 12.7, 13.8, and 13.9. The male from Durango is 15.0. The color pattern consists of a series of ir- regular dorsal brown spots ranging from 44 to 54 (47.74) on the body, and from 15 to 23 (18.25) on the tail. Between the darker dorsal spots, the scales are flecked with brown on a cream background. Between the dorsal spots and the ventrals is a row of small spots alter- nating with the dorsal spots. The ventrals and caudals are without any spots or flecks. This subspecies, based on available rec- ords, occurs in the valleys and foothills east of the Sierra Madre Occidental into the desert valleys at least a short distance east of the main highway (45) and south from north central Chihuahua into Durango. Banta and Leviton (1961) report a DOR specimen from 1.6 km (1 mi) S Juan Batista, Aguascalientes, 12 Sep- tember 1957, and Dixon, Sabbath, and Wor- thington (1962) report 4 male specimens from 9 to 32 km (6 to 20 mi) SE of Nombre de Dios, Durango. These have ratios, tail to total length, of 12.4% to 13.6%. If there is justifica- tion for recognizing the subspecies australis, it is not supported by the northern Durango specimen. The specimen from 19 km (12 mi) S of Los Nieves, Durango, is near the mean for the Chihuahua series in all characters, and is thus considered to be an extension of the Chi- huahua population into Durango. Dixon and Fleet (1976) have mapped the approximate distribution of this species in Chihuahua; also, they show A. e. elegans in northeastern Chihuahua and A. e. philipi oc- curring in extreme north central Chihuahua. Two specimens (BYU 15431-2) o( philipi 5 km (3.4 mi) N of Columbus, New Mexico, are similar to the Chihuahua series in scalation but differ in color pattern with 61 and 64 dorsal body spots. Conopsis nasus lahialis Tanner Conopsis nasus Giinther, 1858, Catalogue of the snakes in the British Mus., p. 6. Conopsis nasus Tavlor & Smith, 1942, Lhiiv. Kansas Sci. Bull. 28(2)15:.329-,3,33. Conopsis nasus lahialis Tanner, 1961, Herpetologica 17:13-18. 2 mi SE Creel, 4 (BYU 14295-8). I mi W La Laja (6 mi E of Majoraehic), 2 (BYU 16854-5). 22.5 mi SE Creel. 1 (BYU 16952). 25 mi SE Creel (hv La Bufa Road), 7 (BYU 16856-62). Since the report of Tanner (1961), no addi- tional data on scale and color pattern varia- tions have been obtained. A field note refer- ence to their habitat is as follows: south of October 1985 Tanner: Snakes of Western (>hihuaiiua 629 Creel these snakes were found under rocks in moist to wet soil and in the crumbled volcanic gravel. Some were taken under loose rocks at the base of a small ledge from which water was issu- ing, keeping the base materials wet. In this area, three small individuals were taken, each of which showed large umbilical scars, suggesting that they were recent hatchlings (total length 101-124 mm), recorded 18 July 1960. Diadophis punctatns recalls Baird and Girard Diadophis regalis Baird and Girard, 1853, Cat. N. Amer. Rept. Pt. I, p. 115. Diadophis punctatus regalis GehXh&iih, 1965, Proc. U.S. Nat. Mus. 116:300-307. 10 mi W San Francisco del Oro, 1 (BYU 14251). Appro.ximately 2 mi E Cerocouhiii, 1 (BYU 14243). Yepomera, 2 (UAZ 34398 and 34792). Only two specimens were taken in the years we worked Chihuahua, and these seemed to have come from very different populations. The one from San Francisco del Oro lacks a light neck ring and has a very high ventral count (243); the specimen is otherwise very similar to the regalis specimens I have seen from New Mexico, Ari- zona, Utah, and Nevada. This female specimen was taken alive as it crossed a road, along the margins of which were numerous boulders. Gehlbach (1965) placed it in with the subspecies regalis. A second specimen from Cerocouhui is very different, with a wide neck ring and only 195 ventral scutes. The two specimens are from ap- proximately the same latitude but on opposite sides of the Sierra Madre Occidental. The latter specimen is a male, and this may account for some of the differences in the number of ventral scutes; however, a difference of 48 seems a rather wide margin to retain it in the same sub- species. Although I have not attempted to examine a large series (27 specimens) of the subspecies re- galis, the Cerocouhui specimen does not fit the basic color and scale patterns I have come to associate with the regalis populations in Idaho, Utah, and Nevada. Furthermore, in southwest- ern Chihuahua it is associated with the high mountainous habitats extending southward through Durango and into central Mexico. These areas do not provide the same foothill or xeric conditions I have associated with the sub- species regalis; for this reason, it seems that the dugesii subspecies may well extend as far north as the high moimtain habitats of southern Chi- huahua. This geographic pattern would be in keeping with the distribution of several other species that have extended their range north- ward throughout the high mountain habitats from Durango and into south and south central Chihuahua. It is more logical to include it in the subspecies dugesii. Cehlbach (1965) considered it a regalis-dugesii intergrade. The Cerocouhui specimen was taken after heavy rains in a boulder-strewn alluvial fan at the mouth of a small canyon and was taken in com- pany with Eumeces callicephalus, Crotalus lep- idus klauberi, and the Hylactophrine frog Eleutherodactylus taraluimaraensis. If it is in- deed a representative of D. p. dugesii, it not only represents an extension of its range but also a new record for Chihuahua. The two female specimens taken at Yepomera have wide distinct nuchal collars but were simi- lar to the specimen from San Francisco del Oro in scalation (ventrals more than 240). Since re- galis may or may not have a nuchal collar, these specimens are, on the basis of scalation, retained in the subspecies regalis. Eleven male speci- mens (at BYU) from southwestern United States (states listed above) and Chihuahua have the following ventral and caudal counts: ventrals 211(215.2)222, caudals 69(74.5)81; 15 females, ventrals 219(231.1)243, caudals 57(66.4)72. There is an indicated clinal increase from north to south if the Cerocouhui specimen is excluded. Drymarchon corais rubidus Smith Dn/marchun corais rtibidtis Smith, 1941a, Jour. Wash. Acad. Sci. 31(ll):474-476. Hardy and McDiarmid, 1969, Univ. Kansas Publ. Mus. Nat. Hist. 18(3): 159- 160. Appro.ximately 2 mi S Urique, near river, 1 (BYU 14245). Along San Miguel River, just below Arroyo Cienega, 1 (BYU 2,3708). Smith (1941a) reported the type series as hav- ing the ventrals ranging from 190 to 203, and the caudals as 69-78. Hardy and McDiarmid (1969) reported the range in ventrals as 187-197 for 17 specimens from Sinaloa. The two Chihuahua specimens (both males) have 199 and 196, re- spectively, in ventrals, and one has 72 caudals. In both specimens there are 8-8 supralabials; however, one (23708) has 9-9 infralabials, and in both the 6th supralabial contacts the 1st lower temporal but is widely separated from the lower 630 Great Basin Naturalist Vol. 45, No. 4 postocular. Both are adult males measuring 1457 and 1987 mm in total length, with the latter having lost approximately half of its tail. The color pattern is more nearly as described by Smith (1941a) for the type series; that is, the dorsum is black and the head markings are es- sentially the same color. One difference is noted: the ventrals anterior to those that are black have the posterior edge in black. This black edging becomes thin anteriorly until the median di- vides, leaving thin, dark margins extending lat- erally along the edges of the ventrals to within 15-20 ventrals before the gulars. The ventral color of the specimen from the San Miguel River is a deep, almost ruby, red. The Urique speci- men was a salmon pink (Tanner and Robison 1960). This is another extension of the coastal thorn forest fauna extending its range into the deep valleys of southwestern Chihuahua. McCranie (1980) lists a record for southwestern Chi- huahua. Elaphe guttata emoryi (Baird & Girard) Scotophis emoryi Baird & Girard, 1853, Smithsonian In- stitution, part I, p. 157. Elaphe guttata emoryi Dowling, 1952, Occ. Papers Mus. Zool. Univ. Michigan 540:2. 12 mi SE Nuevo Casas Grandes, 1 (BYU 13918). 11.7 mi W Ricardo Flores Magon, 1 (BYU 15.347). The scale and color patterns are normal for the subspecies, and the distribution is within the limits established by previous reports. Both specimens are males with 204 and 208 ventrals; one (14547) has 74 caudals, and the dorsal body spots are 43 and 37, respectively. In spite of extensive collecting in the area, only two specimens were found. This is surpris- ing, since none were seen as DOR specimens on roads. We attempted not only daytime collect- ing along roads in the valleys or on the hillsides, but we also did considerable road running dur- ing the evenings and early mornings. We have concluded that this is either a rare or very secre- tive species of Chihuahua. Elaphe stibocularis (Brown) Coluber subocularis Brown, 1901, Proc. Acad. Nat. Sci. Philadelphia .53:492. Elaphe subocularis Stejneger iind Barbour, 1917, C>'heck- list, p. 84. Elaphe subocularis Worthington, 1980, Gat. Amer. Amph. and Kept., p. 268. 12 to 39 mi NE Aldama on Chihuahua Road 16, 9 (Carnegie Mus. Nat. Hist. 59917-23, 49926 and 61792). 18 mi SE Giudad Chihuahua, 1 (NMMZ 9307). 17.7 mi E Goyame, 1 (TGWG 44005). The distribution of this species in Chihuahua is far from being fully understood. The present records place it primarily in the eastern part of the state; however, some records from central Chihuahua indicate that it could occur in much of the lower foothills and ranges lying east of the Sierra Madre Occidental. We did not collect east of Highway 45, and the records listed above (kindly provided by Dr. Richard D. Worthington) are all from localities east of the highway. Elaphe triaspis intermedia (Boettger) Coluber triaspis Cope, 1879, Proc. Amer. Philos. Soc. 18:261-277. Pityophis intermedius Boettger, 1883, Ber. Offenbach. Ver. Naturk 22/23:148. Elaphe triaspis intermedia Mertens & Dowling, 1952, Senckenbergiana 33:201. I am aware of only one specimen from Chi- huahua, previously reported by Taylor and Knobloch (1940), and now number 17681 in the University of Illinois Natural History Museum. Based on the records reported for Sonora (Bogert and Oliver 1945) and Sinaloa (Hardy and McDiarmid 1969), this species is obviously more common in the western lowlands than in any part of western Chihuahua. Its occurrence at or near Majorachic is undoubtedly the extreme eastern extension of its range and represents another species that has ascended the valleys of the Rio Fuerte, this time via its western tributary, the Oteros. Geophis aquilonaris Legler Geophis aquilomiris Legler, 1959, Univ. Kansas Publ. Mus. of Nat, Hist. ll(4):.327-334. Maguarichic, 2(BYU 16912, 16913). Both specimens (females) were taken 13 July 1960. One was underneath a rock, and the other was taken as it moved in leaves under low grow- ing shrubs. When the speciman was collected, the ground was wet and the temperatine hot and humid. The scalation and color pattern are generally within those of other CI. aquilonaris, but a few coinits extend the known variation (Legler 1959). One specimen (BYU 16912) has only 170 ventrals but 66 caudals. Thus, the ventral range for females is now 170-183 and the caudals 55-66. The color pattern is within the limits previously described for the species. Gyalopion canum Cope Gyalopion canum Cope, 1860, Proc. Acad. Nat. Sci. Philadelphia 12:241, 243. Ficimia cana Carman, 1883, Mem. Mus. Comp. Zool. 8(3):82. Gyalopion caniis Leviton & Banta, 1960, Occ. Pap. Cali- fornia Acad. Sci. 26:1-4. 5.5 mi NE of Colonia Juarez, 1 (BYU 15257). Scalation and color pattern as reported by Cope (1900). Ventrals 138, caudals 30, scale rows 17. The specimen was taken on a warm and humid night about 10:00, just after a light rain, 3 September 1959 on the road between Colonia Juarez and Casas Grandes. When the specimen was picked up, it produced several rather loud, sharp, popping noises. In spite of many hours of night driving dur- ing all types of weather, this is the only speci- men seen DOR or otherwise. The range is now definitely within Chihuahua and may be expected throughout most of the foothill area on the eastern front of the Sierra Madre Occi- dental. Hardy (1976) cites (map) a record for central Chihuahua. Heterodon nasicus kennerlyi Kennicott Hcterodon kennerlyi Kennicott, 1860, Proc. Acad. Nat. Sci. Philadelphia 12:336. Heterodon nasicus kennerlyi Cope, 1900, Ann. Kept. U.S. Nat. Mus. 1898:773. Smith 1943. Proc. U.S. Nat. Mus. 93(3169):432-433. Tinaja Valley, 1 mi S of highway to Colonia Juarez, 7 (BYU 13900, 14315, 15337, 15816, and 16106-8). 3 to 5 mi SE N Casas Crandes (along highway), 4 (BYU 15250, 17101, 17103-4). 54 mi S Juarez City, DOR on Highway 45, 1 (UTEP4696). Smith (1943) lists three specimens from the following localities: Progreso, 27 km (17 mi) W Carmen, and Corralitos. The scalation and color pattern are not dif- ferent from those provided by previous au- thors. In all, there are 23-23-19 scale rows; 131-146 ventrals, males 131-136, females 139-146; caudals 29-40, males 37-40, fe- males 29-33. Smith (1943) lists a specimen for Tlajualilo, Durango. This indicates that the range in Chi- huahua undoubtedly extends south through- out the desert valleys of central Chihuahua. We were fortunate to have half our collection given to us by Mr. Herman Hatch, who found them in his cultivated field a mile east of Colo- nia Juarez. The only specimens we found were on or along the highways. Once arriving in the Colonia Juarez/Casas Grandes area and letting it be known that we were interested in reptiles, we were intro- duced to a number of snake stories. One con- cerned the species at hand. It was referred to as a deadly adder that feigned death only to get you to pick it up or get close enough for it to strike. In 1959 we were fortunate to find a live adult a few miles SE of N. Casas Grandes, and it was possible to demonstrate that it was completely harmless. We soon found that many were afraid of reptiles, and before we engaged any help from the locals, we had to educate them that the only good snake is not a dead one. Genus Hypsiglena In 1860 Cope described two species of the genus Hypsiglena: ochrorhyncha from Cape San Lucas and chlorophaea from Fort Buchanan, Arizona. Most authors have placed the latter species as a synonym under ochrorhyncha, even though there are at least two subspecies geographically between Cape San Lucas in Baja California and the popula- tions in Arizona and northern Sonora. Now that adequate material is seemingly available, it does seem appropriate, on the basis of the ventral counts and combining the ventral-cau- dal counts and color pattern, that we recog- nize chlorophaea as a distinct subspecies and indicate its description and distribution in central and northern Sonora and in most, if not all, of Arizona. The material from Chi- huahua is, in several respects, more nearly similar to the Cape San Lucas population than to the Arizona-Sonora populations. An analysis of the available material from southern Sonora, northern Sinaloa, and southwestern Chihuahua indicates that in- tergradation of the torquata type (having a light nape band followed by a large, dark band) with the ochrorhyncha type (or chlorophaea) to the north has produced sev- eral head and nape color pattern combina- tions. It does appear that the light cream band on the nape of the torquata material is gradu- ally replaced in the intergrading specimens to the north by forward movement of the broad, dark band, which in turn is responsible in part a32 Great Basin Naturalist Vol. 45, No. 4 Table 1. Ventral and caudal variatioi in Hijps ^lena . Region No. Sex Ventrals Caudals Ventral-Caudals Cape San Lucas 13 M 166(170.8)176 42(49.5)53 212(220.5)224 and environs 8 F 17.3(178.4)188 44(45.8)48 218(225.7)235 Sonora 11 M 161(173.57)189 49(61.6)65 232(238.7)250 10 F 178(183.8)188 52(55)58 231(238.. 3)244 Arizona 19 M 175(179.6)193 51(55.7)66 222(230.9)248 29 F 175(185.2)190 44(47. 8 ).53 223(233.5)241 Chihuahua 4 M 164(168.3)171 48(52.3)56 216(220.7)227 3 F 170(172.0)175 41(44.2)49 211(216.2)221 for eliminating at least the posterior portion of the cream-colored band. The anterior part of the white band becomes pigmented and spot- ted like the dorsum of the head. To further reduce the prominence of the light band, a medial dark stripe two or three scales wide extends from the dark band to or within a scale of the parietals, and the orbital stripes are extended caudad to contact laterally the dark nuchal band. The elimination of the cream band is followed by a reduction in the size of the dark nape band, which becomes incised dorsolaterally, and, as the reduction in the length of the dark nape band continues, the dorsolateral indentations separate the nape band into a median spot posterior to the pari- etals and two lateral spots extending dorsad to the orbit. This intergradation of color pattern is more apparent than perhaps in any other subspecies thus far examined, and is, of course, the criterion that induced Dunn (1936) to suggest only one species existed in the genus. An examination of 32 specimens from the cape region of Baja California (that is, the area south of La Paz and in the cape area) suggests that the material from the type locality of ochrorhyncha and the general area north from the Cape to near the La Paz area repre- sents a population quite distinct from any other in the Baja California peninsula. Speci- mens taken a short distance north of La Paz show a substantial increase in the number ol ventrals and caudals, and a substantial in- crease in the number of dorsal spots, provid- ing the i)asic characteristics of the midpenin- sular subspecies venusta. A comparison of the material in the Loreto/Comondu area to that of the Cape shows an increase of approxi- mately 10-15 scales in the ventrals and an increase of approximately 10 in the caudals, so that a combination of the ventral-caudals re- sults in an average increase in the venusta population of 1.5-20 ventral-caudal scutes. Immediately north of the venusta subspecies, in the area of San Felipe and on into the Great Basin area of southern California, Nevada, and Utah, and as far north as British Colum- bia, the subspecies deserticola provides an even higher number of ventrals and caudals, with a combination of the ventral-caudal scutes exceeding 240 scales. Thus, the popu- lations at the cape and those populations in Arizona-Sonora have, until now, been re- ferred to as disjunct populations of ochrorhyncha. On the basis of statistical anal- ysis, the Arizona-Sonora populations should be separated and placed in a distinct subspe- cies. An examination of the Arizona-Sonora populations (85 specimens examined) indi- cates that the average ventral-caudal scales in these populations range from 232 to 239, in contrast to an average of approximately 220 in the typical ochrorhyncha popidation of the Baja cape. There is also an increase of approxi- mately 10 dorsal body spots in the Arizona- Sonora populations, averaging 57-60 spots in contrast to the 50-52 in the cape population (Table 1). It is thus proposed that the Arizona- Sonora populations be recognized as distinct and placed in the subspecies Hypsighnm torquata chk)rophaea Cope. Hypsi:lrn(i tcxaiui Sti-jiu'Hcr, 1893. N. .\incr. Fauna 7:205. Hypsi^lcita orhrorhymlia tcxatui Stcjuiiicr 6c Barbour, 1917. Checklist N. Amer. Amph. Hept. , p. 93. Colonia Juarez, 4 (BYl' 14300-3). 9iniNFColoniaJuarez, 1 (BVl' 1.5.37.3). Ca-sasCrandes (in ruins). 1 iBYl' 16988). 24 mi F Buenaventura. 1 (BVU 1.5260). 6.5 mi N ChiliuahuaCitv, 1 (BYU 16989). 5 mi N Cerro Clampana. 2 (MVZ 70995-6). October 1985 TANNER; Snakes oe Western Chiiuiaiilia 633 Ojo de Laguna (25 mi S Clallego), 2 (MVZ 73012-3). 16 mi N DurangoCity, Durango, 1 (MVZ 59299). 0.5 mi S Matachic on highway 16, 1 (UAZ 34420). The Chihuahua Hijpsiglcna that have, in recent studies, been referred to as part of the H. t. ochrorhyncha complex are more com- parable in scale patterns to the Baja cape pop- ulation than to any other subspecies referred to above. For example, the mean ventrals in the males (7) is 168.3 in contrast to the cape males (22) at 170.5. The females are equally similar in that the total of ventral caudal scutes is almost identical in the two populations, as is also the number of dorsal spots in both sexes. However, a comparison of the Chihuahua population samples to a series of specimens from southeastern New Mexico and Te.xas suggests that the central and eastern Chi- huahua material should be associated with the subspecies t. texana rather than retained in the subspecies t. ochrorhyncha or t. chlorophaea. Thus, the Hypsiglena of Chi- huahua can best be assigned to two subspe- cies: those east of the Sierra Madre Occidental in the area of Colonia Juarez and south into Durango to texana, and those on the west, at least in the Rio El Fuerte basin, on the basis of the material now available, to the subspecies chlorophaea. A female specimen (16 mi N Cd. Durango, MVZ 59299) has 174 ventrals and 51 caudals (225 V-C) and should perhaps be included in this subspecies. Hypsiglena torquata chlorophaea Cope Hypsiglena chlorophaea Cope, 1860, Proc. Acad. Nat. Sci. Philadelphia p. 246. Urique, 1 (BYU 14313). There is a real possibility that the popula- tion of the El Fuerte basin in extreme south- western Chihuahua may also have representa- tives of t. torquata, since adjacent southern Sonora and northern Sinaloa represent the area where intergradation seemingly occurs. I would, therefore, expect to find specimens with the t. torquata pattern entering from Sinaloa into southwestern Chihuahua. The specimen from Urique has a broad, dark, nape band 5 scales long with a median extension of 5 scales to the parietals. At the anterior edges of the dark band (on each side of the median) are light brown areas. This nape pattern is similar to that of a specimen from Colima (BYU 23962) except that, in the Urique specimens, the areas anterior to the dark band and on each side of the median nape stripe are cream colored. This pattern seems to be a further indication of changes in pig- mentation pattern that have occurred because of intergradation between H. t. torquata and H. t. chlorophaea. Summary for Hypsip,lena The taxonomic arrangement of Hypsiglena, indicated above, seems the most logical inter- pretation, based on geographical distribution and statistics. For a long time the Arizona- Sonora-Chihuahua material represented a taxonomic problem. With the material now available, there seems to be little justification for not recognizing Cope's subspecies de- scribed from Fort Buchanan, Arizona, in 1860. Averages provide ample key characters to serve the purpose of separating the two widely separated populations. The climatic changes that have occurred during the last 15,000-20,000 years may have subjected this widely dispersed genus to sub- stantially changing environments. The cape area of Baja California and the Arizona-Sonora area may not have changed as radically as did the area between (that is, central and north- ern Baja California and the lower regions of the Great Basin). The area thus retains ances- tral characteristics in these populations while necessitating a more radical change in the external color pattern and scalation character- istics of the populations in the intervening desert areas. Therefore, the primary differ- ence that has developed in the Arizona- Sonora-southwestern Chihuahua popula- tions has been an increase in the ventral-cau- dal scutes, with no major alteration in the basic color pattern except for an increase in the number of dorsal spots. The parameters of intergrading populations are, as yet, not well defined, but general areas as given above can now be indicated. They are as follows: a. H. t. torquata -t. chlorophaea: northern Sinaloa, southern Sonora, and perhaps southwestern Chi- huahua. b. H. t. ochrorhijncha-t. venusta: areas near La Paz and immediately north of Bahia de La Paz but not extending far beyond the Arroyo Salado. 634 Great Basin Naturalist Vol. 45, No. 4 c. H. t. venusta -t. deserticola -t. klaubeh : veniista in the area south of San Fehpe and along the gulf coastal areas, with more typical deserticola ranging northward into southeastern California, and klauberi extending westward to the coast and north into southwestern California. d. H. t. deserticola -t. chlorophaea : not as clearly de- fined, but occurring to a limited degree along the edges of the Colorado River to Glen Canyon Dam. e. H. t. chlorophaea -t. texana: apparently occurring in extreme eastern Arizona, southwestern New- Mexico, and northwestern Chihuahua. Lampropeltis getulus splendida (Baird & Girard) Ophibolus splendidus Baird & Girard, 1853, Cat. of N. Amer. Reptiles, p. 83. Lampropeltis splendida Cope, 1860, Proc. Acad. Nat. Sci. Philadelphia, p. 25.5. Lampropeltis gettdus splendidus Wright & Bishop, 1915, Proc. Acad. Nat. Sci. Philadelphia 67:168. 2.3 mi N Chihuahua Citv, 2 (BYU 15182-1.5283). 28 mi W Chihuahua City, 1 (BYU 14138). Rio Santa Maria at bridge W of Galeana, 1 (BYU 13515). 14.5 mi E Buenaventura, 1 (BYU 1,5252). 1 mi SW Casas Grandes, 1 (BYU 17691). 7 mi N El Sueco, 1 (UTEP 4018). Literature citations are for San Diego (Blan- chard 1921, AMNH 3752) and Rio Santa Maria and San Diego (Smith and Taylor 1945). Based on the available records, it appears that this species ranges in the area west of Highway 45, extending west through the val- leys and low ranges to the east base of the Sierra Madre Occidental. Records are avail- able from the desert areas of eastern Chi- huahua (Reynolds and Scott 1977). The scale counts of examined specimens are slightly higher than those listed by Blanchard (1921) and are as follows: scale rows 23-23-19 or 23-25-19, with 4 of the 7 having 25 rows at midbodv; ventrals 209-217, males 204-213 (210.3), females 210-217 (213.3); caudals 51-58; other variation as previously reported. Three of the specimens are juveniles and show a decided series of large dorsal spots, clearly divided by narrow, light lines. In the adults, an increase of dorsal dark pigment ob- scures this spotted pattern. The dorsum of the head and the nape are black, with this pattern extending for 15 scales posterior from the parietals. The ventrals are mostly black, but with light spots on the edges. There appears to be a strong inllucnce of the subspecies niat. ol Amer. .\niph. and Reptiles). In the northern mountains (north of the Rio Papigochic) of Chihuahua is one of several isolated populations of L. p. pyromelana. Be- cause of the many mountain islands in the total distribution of this species, we find ver> little e\'idence of intergradation between the subspecies. This ma\' also be the result of an incomplete understanding of its distribution, since few specimens are axailable. Both Blanchard (1921) and Smith and Tay- lor (1945) list a specimen from San Diego October 1985 TANNER: Snakes of Western Chihuahua 635 (AMNH 3716). In the early days San Diego was an important Rancho. It is located east of Colonia Juarez and south of Casas Grandes, and this is not a montane locality. Those who have seen pyromelana report them to be only in the mountains much to the west of Colonia Juarez. I suspect that the San Diego specimen was also taken in the mountains but reported from the then-recognized locality. The scale counts listed by Blanchard for this San Diego specimen are within a few scales of those ex- amined from the mountains to the west. Specimens reported from Guerrero (Smith and Taylor 1945) and from Yepomera (Van Devender and Lowe 1977) in west central Chihuahua include its range in the Rio Pa- pigochic basin. There is a strong indication that pyromelana has entered Chihuahua by ascending the tributaries of the Rio Yaqui from northwestern Sonora, since both the Bavispe and Papigochic drain the northern and central regions of the Sierra Madre Occi- dental. Lampropeltis pyromelana knoblochi Taylor Lampropehis knoblochiJ a.y\or, Copeia 1940:253. Lampropeltis pyromelana knoblochi Tanner, 1953, Great Basin Nat. 13:47-66. 25 mi S Creel (La Bufa Road), 2 (BYU 16864-5). The only other specimens known are the types from Majorachic (FMNH 23016-17), two specimens from Yecora, Sonora (UAZ 25131 and 28177), and a specimen in the British Museum from Yoquiro, Chihuahua. This subspecies shows the greatest depar- ture from the basic characteristics of L. py- romelana. The color pattern is unique in that the red is not constricted dorsally by the black bands, nor does the red reach the ventrals. Thus, the red is combined to large, dorsal spots surrounded by narrow, black bands dor- sally and laterally and by an irregular, light stripe ventrally between the black-edged red blotches and the ventral scales. In the other subspecies the red reaches the ventrals in some or all triads. The length of the triads is shorter than in other subspecies, and thus there is an increase in the number of white bars and/or triads (transverse white bars 63- (74)-85). The white bars terminate laterally by becoming a part of the lateral, irregular, light line; such bars equal or exceed the total Inumber of caudals. This subspecies is not ringed or banded but has a series ol red spots extending across the dorsum from the 3rd or 5th scale rows on each side. There is an increase in ventrals 225-238 and in caudals 64(67)74. The two specimens taken south of Creel were on a rocky hill above a small stream. Heavy rains the day before left the habitat wet; both were found under rocks. The distribution of knoblochi is not fully known. Those records indicated above are all from the high mountains of southwestern Chi- huahua and extreme southeastern Sonora. Since most of the area to the south has not been studied, it is suspected that this subspe- cies may range south into the Guadalupe y Calvo area and even into northern Durango. Leptodeira splendida ephippiata Smith & Tanner Leptodeira ephippiata Smith & Tanner, 1944. Copeia 3:131. Type locality, 13.3 km (8.3 mi) WNVV of Alamos, Sonora. Leptodeira splendida ephippiata Duellman, 1958. Bull. Amer. Mus. Nat. Hist. 114:82. Approximately 15 mi upstream from Divisadero Trail, Urique River, 1 (BYU 22658). A juvenile or subadult male 277 mm in total length and with a tail/body ratio of 0.267, less than that reported for juveniles by Duellman (1958) at 0.365. The scale rows are 21-21-17, ventrals 179, caudals 92, supralabials 7-8, in- fralabials 10-10, preoculars 3-3, postoculars 1-2, loreals 1-1 and temporals 0-2 on right side (with first temporal fused to the parietal) and 1-2 on the left side. The range of the caudals is increased from 78-85 (Duellman 1958) to 78- 92 in males. There are 26 spots on the body, and 20 on the tail. The first body blotch is connected to the nuchal stripe, which extends as a narrow, uniform streak to the posterior tip of the pari- etals. The area between the parietals and the first body spot is a light cream color, providing a noticeable contrast between the mottled head and the first body spot. The postorbital stripe fades on the posterior temporals and appears only as a stippling from the posterior supralabial on one row of scales to the first body spot. Otherwise, the color and scale pat- terns are as has been described by Duellman (1958), Hardy and McDiarmid (1969), and Taylor (1939a). 636 Great Basin Naturalist Vol. 45, No. 4 The importance of this specimen is not that it varies appreciably from specimens taken in Sinaloa, but primarily in that it was found at such a distance up the Urique River, suggest- ing that there may well be a sizeable popula- tion inhabiting the Rio El Fuerte basin (at least as far as the coastal thorn woodland habi- tat extends into the barrancas). The specimen had recently eaten an adult Hijla and several arthropods, the latter not easily recognized since they were fragmented. The reduced ratio between body and tail and the increase in caudals suggest that this species, as with others that inhabit the deep, narrow canyons, is sufficiently isolated there to have evolved distinctive variations. This specimen is also a new record for Chihuahua. Leptophis diplotropis diplotropis (Giinther) Ahaetulla diplotropis Giinther, 1872, Ann. Mag. Nat. Hist. 4(9):25-26. Leptophis diplotropis Giinther, 1894, Biologia CentraH- Americana, Kept., p. 130. Arroyo Cienaga Prieta, approximately 35 mi below Guachochic, 1 (BYU 22484). A single female specimen represents the first to be taken in the Rio El Fuerte basin of southwestern Chihuahua. It has a higher ven- tral count (188) than specimens examined from nearby Sinaloa as reported by Oliver (1948) and Hardy and McDiarmid (1969). Oliver (1948) also lists four specimens (MCZ 43268-43271) from Guasaremos, Chihuahua, a locale on the Rio Mayo. In these, the ventral counts range from 181 to 184. The average of the five available specimens from Chihuahua is 183.4. The ventral counts of Chihuahua specimens (181-188) are similar to those oc- curring in specimens from the Tres Marias Islands (185-186) and may, therefore, justify the retention of the island population within the nominal subspecies diplotropis. Other- wise, the scale counts are approximately the same as those observed by previous authors cited above. The color pattern is as described by Oliver (1948) and as figured by Bogert and Oliver (1945), except that the dorsolateral stripe does not involve the 4th-5th rows and the anterior half of the 7th supralabial. The scales in the 1st and 8th (middorsal) scale rows are noticeably larger than other scales in the intermediate scale rows. Faint keels occur only on the paravertebral rows of the body, but not on the tail. The differences in the size and shape of the body scales is very noticeable, with rows 3 to 6 on each side elon- gate, and with a diagonal position in contrast to the other more uniformly positioned rows. An increase in ventrals and differences in color pattern from south to north in western Mexico is apparent, but whether the northern populations can be considered to be sub- specifically distinct from those in southern Sinaloa southward must await a much larger series of specimens. Masticophis flagellum lineatulus Smith Coluber fla(:,ellum Shaw, 1802, General Geol. or System- atic Nat. Hist., p. 615. Masticophis flcif^ellum Ortenburger, 1928, Occ. Pap. Mus. Zool. Univ. Michigan 139:2. Masticophis flagellum lineatulus Smith, 1941, J. Wash. Acad. Sci. 31(9):.394. 13 mi S Acension, 1 (BYU 17102). 22 mi S Gallego, 1 (BYU 13975). 1 mi W Sueco, 1 (BYU 15360). 9 mi W Sueco, 1 (BYU 1.5339). 2 mi S Sueco, 1 (BYU 42244). 38 mi S Ahumada, 1 (BYU 15.340). 3.5 mi E Buenaventura, 1 (BYU 15358). Colonia Juarez, 1 (BYU 15468). 8 mi NW Colonia Juarez, 1 (BYU 15461). Lower Tinaja near Colonia Juarez, 1 (BYU 15341). 4 mi NE Colonia Juarez, 1 (BYU 17697). 25 mi N Cd. Chihuahua, 1 (BYU 30.381). Aside from the type (11 mi S of Buenaven- tura), Smith lists four additional paratype specimens from Chihuahua (USNM 14279, 14283 Chihuahua, and USNM 104675-6, Rio Santa Maria, near Progreso). In the UTEP collection are the following: 2519, Sierra del Nido; 3582 0.8 mi NW Zavalza, Durango (near Chihuahua border); 4097, 72 mi N Cd. Chihuahua and 4228, 6 mi NE Janos. The scale counts are as follows: ventrals, males 194(198.6)201, females 191(193.5)195; caudals, males 98(104.6)113, females 98(98.8)100; scale rows 17 reducing to 12 or 13 before the vent; supralabials 8, and infrala- bials 9 or 10. The color pattern does not vary from Smith's (1941) original description. Adult specimens preserved 25 years ago still show the brilliant, deep, salmon-red color. This alone is a remarkable distinction for this sub- species. A few specimens from northern Durango (12 mi S Los Nieves, BYU 14071 and 1 mi N Zarca, BYU 14072) have also retained this unique color and color pattern. October 1985 TANNER: SNAKES OF WESTERN ChIHUAHUA 637 Wilson (1973) indicates by map two and possibly three subspecies in Chihuahua. The subspecies testaceus enters northeastern Chi- huahua from Coahuila, and piceus intergrades with lineatiilus in the northwestern corner. There is reason to suspect that cingiihim may occur in southwestern Chihuahua by entering through the El Fuerte Basin. None of the specimens we have taken in central Chi- huahua, west of Highway 45, show any inter- grading of characters. Masticophis mentovariiis sfriolatus (Mertens) Coluber striolatiis Mertens, 1934, Zoologica 32:190. Masticophis s. striolatus Zweifel & Norris, 19.55, Amer. Midi. Nat. 54:242. Masticophis mentovariiis striolatus Johnson, 1982, Cat. Amer. Amph. and Rept., p. 295. Cueva Creek near Tres Rios, 1 (BYU 17112). This locality is just east of the Sonora border and is a small creek draining into the Bavispe River. This female has the following scale counts: scale rows 17-17-13, ventrals 177, caudals 109, supralabials 8-8, infralabials 9-10, preoc- ulars 2-2, postoculars 2-2, loreals 1-1 and tem- porals 2-2-3. An examination of specimens from Sonora and those reported from Sonora by Bogert and Oliver (1945) show, from the limited material available, lower ventral counts than those reported from farther south. Hardy and McDiarmid (1969) list the total range for Sinaloa specimens as 178-189. Specimens available to me, and those re- ported in the literature, give the range in ventrals as 177-195. Scale rows at midbody are 17 and at vent 13. The color pattern is unique in adults, with the dark spots on the tips of the scales forming fine, broken, diago- nal stripes across the dorsal scale rows. Johnson (1977 and 1982) reviewed the ta.\- onomy and distribution of the whipsnake Masticophis mentovarious (Dumeril, Bibron, & Dumeril), recognized five subspecies, and placed striolatus as the northern subspecies. The Chihuahua specimen is well within both the scale counts and color patterns provided in the studies by Johnson. Masticophis taeniatus girardi Stejneger & Barbour Masticophis ornatus Baird Si GirArd, 1853, Cat. N. Amer. Reptiles, pp. 102-103. Masticophis taeniatus ornatus Schmidt & Smith, 1944. Publ. P^ield Mus. Nat. Hist. Zool. Ser. 29:90. Coluber taeniatus girardi Stejneger & Barbour, 1917, Checklist, p. 89. Masticophis taeniatus girardi Parker, 1982, Cat. Amer. Amph. and Rept., p. 304. 48 mi W Ciudad Chihuahua, 1 (BYU 14128). 7.5 mi E Buenaventura, 1 (BYU 152.56). 6 mi E Ricardo Flores Magon, 1 (BYU 19134). Bridge at Rio Urique on La Bufa Road, 1 (BYU 22700). Sierra del Nido, 1 (UTEP 2519). 72 mi N Ciudad Chihuahua, 1 (UTEP 4097). 6 mi NE Janos, 1 (UTEP 4228). Smith and Taylor (1945) list the range of this species in Mexico as extending north from central Zacatecas through extreme eastern Chihuahua to western Texas. The records listed above indicate that most of Chihuahua may be a part of its range. While on the Rio Urique in September 1963, a large specimen was seen approximately four miles down- stream from the bridge. Unfortunately it es- caped, but it does establish its occurrence in western Chihuahua. Ortenburger (1928) recognized the subspe- cies girardi and extended the range south from the south central United States through adjoining central and eastern Chihuahua to Guanajuato. The scale counts are normal for the subspe- cies. Ventrals range from 199 to 205, the cau- dals 125 to 168. The single male has 168 cau- dals, and the females vary from 125 to 147. All have 15 dorsal rows at midbody and 11 or 12 before the vent. In adults the lateral light stripe fades into the general body color a short distance before the vent. However, in young and juveniles the light stripe is present, extending onto the tail and involving at the vent only the 3rd and 4th rows. The fading of the light stripes in adults is a character distinguishing girardi from the subspecies taeniatus. Opheodrys vernalis hhinchardi Grobman Colubes rcnia/i.s (Harlan) 1827. J. Acad. Nat. Sci., Phila., 5:361. Opheodrys vernalis Schmidt and Necker 1936. Herpeto- logica, 1(2):63. Opheodrys vernalis blanchardi Grobman 1941. Misc. publ., Mus. Zool. Univ. Michigan 50:11-37. 1.6 mi N Pederhales (24 mi SE Guerrero, on Highway 16), 1(UAZ 34416). The occurrence of the green snake is added evidence that the recent past did indeed have 638 Great Basin Naturalist Vol. 45, No. 4 a climate and thus ecological conditions per- mitting a wide dispersal of many species. The passed distribution of this species must have extended throughout much of the intermoun- tain area from Wyoming, Colorado, Utah, New Mexico, and northwestern Chihuahua. There is reason to believe that its distribution was halted by the deserts of the Great Basin extending from the Snake River in southern Idaho and south through central Utah. The slow desiccation during the last 10,000 to 20,000 years has driven this species as well as others (a good example is Larnpropeltis py- romelana) into more suitable habitats in mountains (6000 ft) well above the desert val- leys. The numerous disjunct areas suggest that there may be other mountainous areas not yet discovered in what appears to have been its past area of distribution. Oxybelis aeneus auratus (Bell) Dryinus aeneus Wagler, 1824, Serpentuni Brasiliensum Species Novas, p. 12. Dryinus auratus Bell, 1825, Zool. Jour. 2:324-326. Oxybelis aeneus auratus Bogert & Oliver, 1945, Bull. Amer. Mus. Nat. Hist. 83:381. Arroyo Santa Anita, 1 (BYU 22485). Rio San Miguel, 1 (BYU 38338). These female specimens were taken in the drainage basin of the Rio San Miguel in south- western Chihuahua; it is another species that has entered the state by way of the El Fuerte River basin. The scale counts are within the limits set forth by Bogert and Oliver (1945) for the spe- cies in Mexico; however, the ventrals at 201 and 202 and caudals up to 185 are more than for the Sinaloa female specimens (183-195 and 175-176) reported by Hardy and McDi- armid (1969) but are within the range re- ported for Sonora (Bogert and Oliver 145:387). On the right side of specimen 22485 there are 2 preoculars, one being formed from a division of the upper posterior (orbital) part of the normal preocular. These are new records for the state of Chi- huahua. It is obvious that the 1-2 preocular pattern is an anomaly in this specimen. An examina- tion of a few specimens in our collection from Venezuela, Costa Rica, and western Mexico indicates that throughout the greater range of this species the normal preocular formula is 1-1; however, the occurrence of 2 preoculars, as reported by Taylor (1941) in his description oi^ Oxybelis potosiensis, may not have been an anomaly. Bogert and Oliver (1945) did not see the type specimen or others from or near the type locality. Keiser (1970:227) provides a key to the spe- cies of the genus Oxybelis. His catalogue re- port (1982) provides a complete synonomy for Oxybelis aeneus (Wagler). The map for the species includes the edge of southwestern Chihuahua, but it is not clear as to whether a collecting locality is within the state. Pituophis melanoleucus affinis (Hallowell) Pityophis affinis Hallowell, 1852, Proc. Acad. Nat. Sci. Philadelphia 6:181. Pituophis melanoleucus affinis Smith & Kennedy, 1951, Herpetologica 7:93-96. 2 mi E Colonia Dublan, 2 (BYU 13878-9). 6 mi SE N Casas Grandes, 1 (BYU 15324). 16.4 mi SE N Casas Grandes, 1 (BYU 15359). 23.3 mi SE N Casas Grandes, 1 (BYU 15369). Eastern limits of Casas Grandes, 1 (BYU 15370). 8 mi NE Colonia Juarez, 1 (BYU 15374). Galeana, 1 (BYU 15255). 1.5 mi S Galeana, 1 (BYU 15258). 12.5 mi S Galeana, 1 (BYU 15353). 21 mi E Buenaventura, 1 (BYU 15354). 23.9 mi SE N Casas Grandes, 1 (BYU 15355). 2 mi W Sueco, 1 (BYU 42245). 18 mi W Sueco, 1 (BYU 15352). 25 mi W Sueco, 1 (BYU 15336). Colonia Juarez, 2 (BYU 15430 and 18249). 3 mi S Palomas, 1 (BYU 14651). 4 mi W San Francisco del Oro, 1 (BYU 15378). 1 mi W Chuhuichupa, 1 (BYU 13877). 21.7 N Ciudad Chihuahua, 1 (BYU 15323). 46.7 mi W Ciudad Chihuahua, 1 (BYU 22686). 23.5 mi N Ciudad Chihuahua, 1 (BYU 32042). 48 mi VV Ciudad Chihuahua, 1 (BYU 13881). 4 mi E Cuauhtemoc, 1 (BYU 15385). 25 mi E Cuauhtemoc, 1 (BYU 15428). 5 mi W Minaca, 1 (BYU 17083). 60 mi S Ciudad Chihuahua on Highwav 45, 2 (BYU 1.5772 and 41.339). Aside from the locality records listed above, DOR snakes were seen along Highways 45 and 10 and from Ciudad Chihuahua west to La Junta. Most were badly mashed and were not kept. The records do, however, place this species throughout central and western Chi- huahua. None of the Chihuahua specimens reach the upper limits of the ventrals set by Klauber (1947:220-251). Ventrals range in males from 213 to 232 (225. 1); females range from 223 to 238 (228.7). Caudals in males range from 62 to October 1985 TANNER: Snakes of Western Chihuahua 639 69 (64.66), and females range from 52 to 61 (55.28). In both sexes the caudal ranges varied less than 10 scales. The range in the ventrals was greater but showed a lower mean than was projected by Klauber for specimens stud- ied from the entire geographical area occu- pied by affinis. In the preoculars, 16 had 1-1, 7 had 2-2, and 3 had 1-2. Postoculars were 3 or 4 in nearly equal numbers, and the scale rows were 29 to 33 at midbody, with 31 rows being the usual number. With few exceptions, the rows at the vent were 23. Sixteen specimens had 8-8 supralabials, with others having 8-9 or 9-9. Infralabials were usually 12-12, but with 12-13 or 13-13 occurring frequently. The scale counts are generally lower than those obtained from specimens seen from Sonora and areas from the western part of the affinis range. Dorsal spots on the body range from 26 to 56, average 42.7. Klauber lists the range as 34-63 with an average of 48. The male speci- men with 26 spots came from the western mountains near Chuhuichupa; its spots are large, round, and without the usual convexity so common in most specimens. The next low- est number is the specimen from Mifiaca with 32 spots. Otherwise, the specimens are within the recorded range of variation in scale and color patterns. Klauber (1947) gives the ratio of the length of the tail to the total length as about 0. 136 in males and 0.125 in females. He obtained these ratios from western specimens and states that the more eastern population would have shorter tails. This is not borne out from the smaller series from Chihuahua in which the males range from 0.129 to 0.157 (.143), and the females from 0.120 to 0.144 (.132). Perhaps the present series (29) from Chi- huahua is more representative of the eastern populations than material seen by Klauber. His intrasubspecific trends are, except for ra- tios, seemingly valid; and, yet, an in-depth study of the affinis complex, including a more balanced distribution of specimens, would provide a better understanding of the varia- tion in this widely dispersed subspecies. Rhadinaea hesperia hesperioides Smith Rhadinaea hesperia Bailey, 1940, Occ. Pap. Mus. Zool. Univ. Michigan 412:8-10. Fig. 4. Color pattern of R. h. hesperioides as seen in specimen from south of Guachochic (BYU 22483). Rhadinaea hesperia hesperioides Smith, 1942a, Proc. Biol. Soc. Washington 55:186. Rhadinaea hesperia Mvers, 1974, Bull. Amer. Mus. Nat. Hist. 153:81. 30 mi (by road) S Guachochic, 1 (BYU 22483). This female specimen, taken in the Rio San Miguel Valley in southwestern Chihuahua, is a new record for the state. Based on pubhshed records available, it is at least 250 miles north of the previous records at Plomosas and Santa Lucia in southern Sinaloa. The specimen was taken in the lower edge of the pine forest well above the lowlands along the Rio San Miguel. Except that the ventrals (159) are more than any specimen thus far reported for Sinaloa (149-154), the scale counts are similar: 17-17- 17 scale rows, 121 caudals, 8-8 supralabials, 10-10 infralabials, 1-1 preoculars and 1-1 pre- suboculars, 2-2 postoculars, 1-1 loreals and 1-2 temporals. The specimen is a subadult, 250 mm total length and, of this, 90 mm is tail (36% of total length). The high ventral count may be an indication of a south to north cline, a reversal of the north-south clinal trend from west central Mexico (Jalisco- Michoacan) into Guerrero and Oaxaca. The color pattern is in variance to the de- scription of Myers (1974) for specimens from southern Sinaloa (Fig. 4). At midbody the dor- solateral stripe is on rows 5-6-7, with only the edges of rows 5 and 7 involved, but with more of row 5 included than row 7. On each scale of row 5 below the light stripe is a series of dark spots involving most of the ventral part of the scale and forming, in contrast to the light 640 Great Basin Naturalist Vol. 45, No. 4 stripe, a broken dark line. There are no dis- cernible dark stripes on rows 7 or 8. The dor- sal area between the dorsolateral stripes is pigmented with brown and with dark flecking in the center of row 9, but without a distinct solid, dark stripe. A distinct, light line on the adjacent halves of rows 2 and 3 is clearly visi- ble and is bordered both below and above by narrow, dark stripes. The area between this lateral stripe and the dorsolateral stripe is pig- mented as the dorsal rows (8 and 9). The supralabials are edged above with dark stripe, but also with a series of irregular, dark spots near their ventral center, forming a broken line extending to the spots on the edge of the ventrals. The temporal stripe is distinct and extends only to the nasal scale. This stripe is two plus scales anterior from the dorsolateral stripe. Although the color pattern differs from both hesperia and hesperioides as described by Myers (1974), it does conform closely to the latter as described by Smith (1942a). A series from southwestern Chihuahua and/or north- ern Sinaloa is needed before a final judgment can be made as to the final disposition of the subspecies hesperioides. Smith (1942a), in de- scribing hesperioides, makes the following statement: "A light stripe, most distinct ante- riorly, on adjacent halves of second and third scale rows." This, along with other character- istics, seemingly places not only this speci- men in the subspecies R. h. hesperioides but also seems to justify the recognition of hespe- rioides as a valid taxon. A review of the literature (Smith 1942a, Hardy and McDiarmid 1969, Myers 1974) in- dicated that I should at least examine some specimens from Sinaloa for comparisons be- fore concluding this species report. The three KU specimens (75629, 80870-1) seen, fig- ured, and discussed by Myers were exam- ined. In both 80871 and 75629, the lateral, light line on rows 2-3 (noted by Smith in the type series oi hesperioides) is clearly evident. Myers (1974, Fig. 11-L) illustrates the body color pattern of KU 80871 but represents it as not having a lateral, light stripe on rows 2-3. However, the stripe is present not only on KU 80871 but also on KU 75629, and it extends the length of the bodv. In the larger specimen (KU 80870) from 12.3 km (by Highway 40) SW of Santa Lucia, the lateral line is faint but discernible. The question arises as to whether there is a fading or blending of the lateral pattern in older individuals. In any event, the few specimens available from Sinaloa and Chi- huahua suggest that the subspecies R. h. hes- perioides should be recognized based on its distinct color pattern (Fig. 3). Rhinocheilus lecontei tessellatus Carman Rhinocheilus lecontei tessellatus Garman, 1883, Mem. Mus. Coinp. Zool. 8(3):74. Rhinocheilus lecontei tessellatus Medica, 1975, (^at. Amer. Aniph. and Rept., p. 175. Type locality, Monclova, Coahuila; Klauber, 1941. San Diry;o Soc. Nat. Hist. 9(29):302-.308; Smith and Ta\ lor, 1945. Bull. U.S. Nat. Mus. 187:121. 6.7 mi NW Chihuahua City, 2 (BYU 15284 and 15318). 33 mi N Chihuahua City, 1 (BYU 1.5285). 37 mi N Chihuahua City, 1 (BYU 15286). 17 mi S Sueco, 1 (BYU 15.343). 9 mi N EI Sauz, 1 (UTEP 1309). 3 mi VV Jimenez, 1 (UTEP 4096). 13 mi NW Jimenez, 1 (UTEP 4227). 18 mi S Villa Ahumada, 1 (UTEP 4545). 6 mi S Moctezuma, 1 (UTEP 4546). The scale counts and color patterns were consistently within the range limits listed by Klauber (1941) for this subspecies. One speci- men (BYU 15318), compared with Ridgeway color Plate I, was nearly equal to the color listed as Ceranium Pink. Unfortunately, most of the specimens were DOR. However, the distribution records do place this species within the central part of Chihuahua from south of Juarez City along Highway 45 to the Durango border. The greatest concentration seems to be between Villa Ahumada and Ciudad Chihuahua. Why, with our many trips from El Sueco to Casas Grandes, we did not see this species is not explainable; it is suspected to be present in most of the desert valleys between the west- ern mountains and the eastern deserts. Scdvadora hihuahua City was given to me while at a motel in the northern part of the city. It had been killed and was partly eaten by ants. The only real characters for identification are the extension of the head cap at least four scales posterior to the pari- etals and the wide separation by the first pair of infralabials of the mental from the anterior pair of genials. Cole and Hardy (1981) list five specimens from Chihuahua along Highway 45 (S Samalayuca and \' ilia Ahumada) and west along Highway 10 into the Nuevo Casas Grandes area. According to Cole and Hardy (1981), the geographic ranges of T. nigriceps and T. ho- bartsmithi broadly overlap in Chihuahua. I have not found this to be the case, nor are there actual distributional data in their report (1981 or 1983) to support such a conclusion. Their range map places T. hobartsmithi well to the east of any known localit\' records for T. nigriceps if we base their conclusions on the data presented. There may be an overlapping of ranges in this area, but as yet this has not been demonstrated. In view of the wide range of T. nigriceps, it would be more convincing if additional data would have been presented to justify the ex- tremely narrow extension of T. hobartsmithi into eastern Chihuahua. Based on the data they presented, females from that area could belong to either species. Reynolds and Scott (1977) list three specimens of T. nigricept from along Highway 16. TantiUa wilcoxi Stejneger Tantilla wilcoxi Stejneger, 1902, Proc. U.S. Nat. Mus 25:156. Ft. Huachuca, Arizona. Red Rock, 12 mi up Tinaja Canvon, 2 (BYl 13847-8). .50 mi W Chihuahua City (Highway 30), 1 (BYl 13849). 25 mi SE Creel (La Bufa Road), 1 (BYU 16863V The following data apply to the specinu'n> listed above. Males: ventrals 149-158, fe- males 153-159. Only one male and one fe- male had a complete tail, with 72 and 69 cau- dals, respectively. By including the data ol four specimens reported by Ta\lor anci Knobloch (1940), the following range for three males and five females is: ventrals 149(154)15h and 153(158.8)164. The caudals in two male.s are 62 and 72 and in three females 64, 67, and 69. Head scales are imiform and there are 15 dorsal rows in all specimens. October 1985 TANNER: Snakes oeWestehn Chihuahua 645 Color pattern in the four specimens listed above is uniform. The nape ring involves the tips of the parietals and one or two posterior scales. Only a few specimens of Tantilla ivilcuxi have been collected in Chihuahua. Taylor and Knobloch (1940) reported four specimens from the Sierra Madre Occidental, pre- sumably taken in the vicinity of Majorachic. In a recent study by Cole and Hardy (1981), a single specimen is listed from the vicinity of Stacion Barbara. During the years spent in Chihuahua, four additional specimens were collected (as listed above). The specimen taken south of Creel represents the most southern extent of the known range in Chi- huahua. There is little variation in the scale or color patterns exhibited by the specimens taken over the rather wide range in Chi- huahua. Those on each side of the Continental Divide show little variation in contrast to some of the other species listed in this report. Since the above was written, the catalog account by Liner (1983) appeared; it does not include those records listed above from cen- tral and northern Chihuahua. Liner's distri- bution map and the new records in this report suggest that this species may occur in the mountains of central and western Chihuahua from the northern end of the Sierra Madre Occidental south through at least western Chihuahua to perhaps extreme eastern Sonora, Sinaloa, and into most of Durango. Our records indicate a habitat in Chihuahua above 1500 m (5000 ft) and in the area of the foothills on the eastern edge of the Sierra Madre above the desert valleys and extending into the forested areas above 2250 m (7500 ft). Tantilla ijaquia Smith Tantilla ijaquia Smith, 1942, Zoologica 27:41. Only the type specimen (MCZ 43274, not examined) has been taken in Chihuahua (Guasaremos, Rio Mayo). The foothills ex- tending south and then perhaps eastward into the barrancas of the Rio El Fuerte may also be included in its geographical distribution. Thamnophis Fitzinger Except for the crotalids, the garter snakes are represented in Chihuahua by more spe- cies than any other group of serpents. We collected six species with three additional subspecies, and the literature (Smith and Tay- lor 1945, Fitch 1965) cites a seventh. Since the report by Smith and Taylor (1945), a number of studies dealing with this genus have ap- peared (Bogert and Oliver 1945, Thompson 1957, Tanner 1959a, Conant 1963, Fitch 1965, Webb 1966, Hardy and McDiarmid 1969, Rossman 1971, and others). In none of the studies listed above are all the species of Thamnophis, known to occur in Chihuahua, examined as a group and their distribution and variations considered. Thamnophis rufipunctatus (Cope) In recent years this species has been re- viewed by several authors, each adding new data. However, the problems of generic des- ignation and the possibility of subspecies within this widespread species warrants addi- tional consideration. The reexamination of its generic position is again evaluated, based on additional material from most of the known areas of distribution. I have examined for this report 148 specimens and have reviewed data by others. Field notes, based on extensive work in Chihuahua and limited travels in Ari- zona and Durango, provide data on habits and habitats. As data were evaluated, it became apparent that the species T. rufipunctatus has, for rea- sons of isolation and/or dispersal, developed significant character modifications that war- rant the recognition of subspecies. This was most obvious when comparisons were made between T. nigronuchalis of Durango and the populations of T. rufipunctatus in Chihuahua, as well as the isolated population of the latter in Arizona and New Mexico. These compari- sons aided materially in a better understand- ing of the relationships between the northern, isolated segment of the species and those pop- ulations occurring in Mexico. Although Thamnophis rufipunctatus is widely dis- persed, it has maintained a surprising unifor- mity in most characters. The scale rows, num- ber of loreals, preoculars, postoculars, labials, and, to a degree, ventrals and caudals, are rather uniform. Only in size and position of some scales is variation present. Color pattern also shows some variation and is with certain scale variations discussed below. 646 Table 2. Ventral and caudal variation Great Basin Naturalist in Thamnophis rufipunctatus. Vol. 45, No. 4 Subspecies No. Sex Ventrals Caudals Ventral-Caudals Rufipunctatus Vnilahialis Nigronuchalis 32 25 37 27 14 15 M F M F M F 159(171.0)179 156(164.68)171 155(161.51)169 151(156.68)163 157(161.92)167 152(156.53)160 76(81.93)88 67(70.21)79 71(76.2)82 65(69.95)76 68(72.54)77 62(66.57)72 240(253.45)263 232(239. 1)243 232(239.68)249 219(226.43)239 225(234.63)240 215(223.25)230 A comparison of data for the Arizona-New Mexico, Chihuahua, and Durango popula- tions indicates that three subspecies exist. In each case, these populations occur in widely separated drainage systems, with the Arizona and New Mexico specimens coming primarily from tributaries of the Salt River Basin ex- tending along the southern edge of the Mogol- lon Rim and draining to the south and west. Between this population and that of Chi- huahua lie the desert flats, extending from west of Phoenix east and south past Tucson and into southwestern New Mexico. The dis- tance between these two populations is ap- proximately 320 km (200 or more mi). The geographical differences between the Chi- huahua rufipunctatus and Durango {ni- gronuchalis) populations are not as readily apparent, but they seemingly intergrade with each other somewhere in the highlands of northwestern Durango and/or southwestern Chihuahua. Data now available clearly indicate that the Chihuahua and northeastern Durango popu- lations represent a distinct group from either those in Arizona or west central Durango. Based on scalation (Table 2) and other charac- ters, these subspecies are distinguished and described below. Thamnophis rufipunctatus rufipunctatus (Cope) Chilopoma rufipunctatum Cope, 1875, In Yarrow, Wheelers Rept. Geog. Geol. Expl. Surv. W. 100th Mer. Zool. 5:544. Type locality, southern Arizona. Eutaenia angustirostris Kennicott, 1860, Proc. Acad. Nat. Sci. Philadelphia 12;332. Type locality, Parras, Coahuila, Mexico. Thamnophis angustirostris Ruthven, 1908, Bull. U.S. Nat. Mus. 61:120. Atomarchus multimaculatus Cope, 1883, Amer. Nat., p. 1300. Type locality, San Francisco River, New Mexico. Thamnophis rufipunctatus Smith, 1942c, Zoologica 27(3-4): 120. Natrix angustirostris Lowe, 1955, Copeia 1955(4): 307-309. Thamnophis rufipunctatus Thompson, 1957. Occ. Pap. Mus. Zool. Univ. Michigan 584:1-10; Tanner, 1959a, Herpetologica 15(4): 165; Conant, 1963, Copeia 3:480. Diagnosis. — A moderate- to large-sized subspecies usually with 21-21-17 scale rows, head elongate and compressed laterally; 1 lo- real, 2 preoculars, usually 3 postoculars with the inferior one being narrowly separated from or barely contacting the 4th supralabial in 75% of the specimens; 2 or occasionally only 1 supralabial entering orbit; ventrals 156 to 179, caudals 67-88, ventral-caudals average 239 ? , 254 ^ ; (Table 2); ground color brown to olive brown, with 6 rows of dark brown or rust colored spots. Distribution. — Central Arizona east from Yavapai County into west central New Mexico. Specimens e.xamined. — Arizona: UAZ 26543, 30944, 31384, 41344, Oak Creek Canyon; 26454-55, Big Bonito Creek at bridge to Maverick, Navajo Co.; 26456, Eagle Creek School, Greenlee Co.; 26457, East Fork of White River, 3 mi E Fort Apache, Navajo Co.; 26458, Point of Pines, Graham Co.; 26459, 15 mi SW Flagstaff, Coconino Co.; 26460, Black River at Diamond Fork, Greenlee Co.; 30951, Slide Park, Oak Creek Canyon, Coconino Co. ; 30955, Oak Creek 1/2 mi up from Slide Rock; 31392, 31396 and 41343, East fork of Black River at Diamond Rock Camp, Apache Co.; 34157, 23 mi S Flagstaff; 37035 S of Sedona; 37824, Oak Creek 5-7 mi N Sonoma; 41342, 7 mi S Noury Ranch, Yavapai Co.; 41345-47, Kitbridge, Oak Creek Canvon; 41348-51, 18-20 mi S Flagstaff; 41352-55, 8-9.7 mi above Sedona; 44775, 2 mi bv river below white crossing, Apache Co. ; BYU 11465, Black Canyon, Yava- pai Co.; ASU 10542, Gila Co.; NMMZ 8465, McNary fish cultural station; "17923, Black range at lOof diamond ranch"; 13567-9, FAI, October 1985 TANNER: Snakes of Western Chihuahua 647 reservation, Black River at old military cross- ing, Apache Co. New Mexico: NMMZ 387, Rio San Francisco, 2 mi above Frisco Hot Springs; 385-6, Beaver Creek, Catron Co.; 7442, near Gila Cliff Dwellings Nat. Park, Grant Co.; 4582, Mogollon Mts. 2.5 mi W Wall Lake, Middle fork of Gila; 4616-7, 6023, E of Luna, Wall Dake; 6809, E of Luna, Glen- wood Canyon; 10832, approx. 5 mi E Glen- wood; 32017, Pleasanton E of Luna; 41625, West fork of Gila River 9 mi upstream from Gila Cliff Dwellings, Catron Co. Remarks. — In this subspecies, as in the Chihuahua form, only rarely does the mid- body scale rows vary from 21. When variation occurs, an increase is to 22 or 23 rows (two specimens of each). In 53 specimens (106 counts), there are 2-5 postoculars with 80 having 3, 21 having 4, 4 with 2 and one with 5; those with only 2 postoculars have 2 suprala- bials widely contacting the orbit. In those with 3-3 postoculars, approximately 75% have a narrow contact of the 5th supralabial and the eye on one or both sides (Fig. 5A), or only reaching the 4th labial (Fig. 5B). There is a distinct difference between T. r. rufipunctatus and specimens of the species from Chihuahua, the former having a greater number of both ventrals and caudals. There is only a slight overlap in ventral count and note- worthy differences when these scale counts are combined and averaged (Table 2). In the subspecies rufipunctatus, the di- vided anal variation occurs primarily in popu- lations of Oak Creek in Yavapai Co., Arizona, whereas this variation occurs in the Chi- huahua and Durango subspecies only rarely. I am aware of only two specimens, AUZ 26465 from Garcia, Chihuahua, and LSUMZ 16459 from 8.3 mi W El Salto, Durango. Lowe (1955) placed T. rufipunctatus in the genus Natrix, based on the divided anal in some specimens of the Arizona population and on habits, color pattern, and its "water snake "-like habitude. Thompson (1957) reex- amined the series seen by Lowe and included 14 additional specimens, in which 7 of the series had divided or partly divided anals (25% of the series). From these data he chose to retain rufipunctatus in the genus Tham- nophis. Tanner (1959a) examined 52 speci- mens from Chihuahua and concurred with Thompson in retaining rufipunctatus in the Fig. 5. The relationship of the supralabials and lower postoculars to the orbit in: A, (AUZ 31392); and B, (NMMZ 6023) relationship as seen in Thamnophis r. rufipunctatus: C, (BYU 14218) as seen in the Chihuahua- northeastern Durango populations (Thamnophis r. unila- bialis); and D, (UTEP 36.54 or LSUMZ 16450) as seen in Thamnophis r. nigronuchalis from west central Durango. genus Thamnophis. Conant (1963) also re- viewed the previous studies, examined addi- tional specimens from Chihuahua and Durango, and concurred in retaining ru- fipunctatus in the genus Thamnophis. Smith (1955) placed Thamnophis multimaculatus Cope, which was given species status by Tay- lor and Knobloch (1940), as a synonym of T. rufipunctatus. Thompson ^^1957) also re- viewed the synonymy associated with ru- fipunctatus and, after examining four speci- mens from Parras, Coahuila, concluded that the type series of T. angustirostris Kennicott was in fact more closely related to T. mar- cianus than to Cope's T. rufipunctatus. He states that the name angustirostris was misap- plied by Ruthven (1908) and should now be- long in the synonymy of T. marcianus. The placing of rufipunctatus in the genus Natrix, as suggested by Lowe (1955), is seem- ingly based primarily on the occurrence of a divided anal in some Arizona specimens. That anomaly has also been reported for T. e. va- grans (Tanner, 1950). The divided anal was 648 Great Basin Naturalist Vol. 45, No. 4 brought to my attention when students began classifying specimens of our local Thomnophis (Provo, Utah) in the genus Natrix. There is no question that nifipunctatus is an unusual spe- cies within the genus; with the elongate snout and narrow head, it appears to have become adapted for feeding on small, aquatic verte- brates— particularly fish and tadpoles. How- ever, other species in the genus that are aquatic have not evolved the same head modi- fication, indicating only that great flexibility exists within the genus Thamnophis. Considering all of these factors, it would seem unjustifiable to place this species in a different genus based on a few anomalous specimens found in localized populations. I have seen 37 specimens from Arizona, and of these 5 have divided anals and 3 show a groov- ing. This is 21.6% with divided or grooved anals in contrast to a population in central Utah that reached 30%. A series of 25 speci- mens from eastern Arizona (Apache, Graham, Greenlee, and Navajo counties) and Grant and Catron counties in New Mexico do not have divided anals. Those with this character are generally confined to the local populations in Oak Creek and its tributaries in Yavapai and Gila counties of Arizona. There may be other characteristics that re- late T. rufipunctatus to the genus Nerodia. An in-depth study may reveal this relation- ship and justify a taxonomic adjustment as suggested by Lowe (1955). Such a study is beyond the scope of this report, even though it seems an important step. One character not included in previous studies is the structure of the hemipenis. Everted hemipenes of T. ru- fipunctatus are capitate. I note that the hemipenis of one of the spotted Nerodia (liar- teri), figured by Trapido (1941), is bilobate and appears from the figure to have a similar spine arrangement; that is, the enlarged spines on the proximal part of the structure occur in two series, with the sperm duct lying between them. Three specimens oi rufipunc- tatus were examined (UTEP 3386, UAZ 34158, and LSUMZ 16451), and in each the organ is basically capitate and has two en- larged spines on the outer anterior edge and three on the posterior edge. It is a much dif- ferent structure than that oi N. harteri. A comparison of other Thamnophis {ele^ans, cyrtopsis, and cqucs) to rufipunctatus, and the figure for N. harteri, suggest a much closer overall structural relationship to Thamnophis than to the figure oi Nerodia har- teri. If, as indicated by Cole and Hardy (1981), the structure of the hemipenis is an important taxonomic character, then perhaps the generic status of rufipunctatus will not be satisfactorily resolved until an extensive com- parative study is made for both Thamnophis and Nerodia. Another character peculiar to T. r. ru- fipunctatus is the small, azygous scale sepa- rating the rostral from the internasals (Fig. 6). This scale appears randomly throughout most, if not the entire range and may be con- sidered to be a unique character, particularly in the Arizona and some Chihuahua popula- tions. The type of T. niultimacukitus Cope (type locality San Francisco River, New Mex- ico) was stated to have this preinternasal scale, and it occurs randomly in specimens from Arizona, New Mexico, and Chihuahua. In the series from Arizona and New Mexico (53 spec- imens), 40% have the azygous preinternasal. In Chihuahua populations, only 24% of 67 specimens show this character, and the per- centage is lower in Durango. Also, more spec- imens seen from Arizona have 10 infralabials (80.14%), as is also the case for most Chi- huahua specimens. Excepted are those from San Pedro on the Papagochic and Bocoyna, which have 9 infralabials in 77.5% of the 20 specimens examined. This is in contrast to a series of 26 from Yepomera (about 80 km or 50 mi N and also on the Rio Papagochic), in which nearly 80% (43 of 52) have 10 in- fralabials. The Chihuahua to northeastern Durango populations have both scale and color pattern variations that are different from those in Ari- zona and southwestern Durango, and com- pose a new subspecies described as follows. T}\amnop}\is rufipunctatus unihd)ialis, n. subsp. HoLOTVFE. — BYU 14217, an adult female from .5 mi SW of Bocox na, C'liihuahua, Mex- ico. Collected 11 July 1958 by W. W. Tanner and W. G. Robison. ¥m\at\'pe^.— Chihuahua: BYU 14213-16, 14218, 14224, 17085-6, topotvpes; 14368- 14375 and 14485-6, San Pedro on Rio Pa- October 1985 TANNER: Snakes of Western Chihuahua 649 Fig. 6. Relationship of preinternasal scale to nasal and internasals: A, single azygous scale (BYU 14220); and azygous scales (BYU 13797); and C, usual pattern (BYU 14214). 5, paired pagochic; BYU 13797, 2 mi N Chiihuichupa; BYU 14219-20, Black Canyon, 8 mi W Chu- huichupa; BYU 14207, Rio Bavispe, below Three Rivers; ASU 5304-6 and 5334-6, 3.8 mi SSE Galeana; ASU 17042, Rio Tutuaca; AUZ 26461-2, 35922 and 36290, 4.5 mi SE Galeana; UAZ 34158-63, 34265-79, Ye- pomera; AUZ 35236, El Norte, 3 mi N Chu- huichupa; AUZ 26463-5, Garcia; AMNH 73754-5, Noragachic; NMMZ 31256, Willys on Rio Piedras Verdes; NMMZ 33463-5, ojo de Los Reyes; NMMZ 33478, 4.7 mi SE Galeana. Sonora: UAZ 35235, Yecora. Durango: UTEP 9078, 6 mi SW Los Frailes. Diagnosis. — A subspecies oi nifipunctatus that differs from both r. rufipunctatiis and r. nigronuchalis in having only the 4th suprala- bial contacting the eye by reason of the lower postocular having a firm contact with the 4th labial; ventrals reduced and ventral-caudal av- erages 13-15 scales fewer than in r. riifipunc- tatus. Ventral color pattern with anterior mar- gins of scales usually pigmented, forming alternating dark and light transverse bands. Description of type. — Head and body length 412 mm; tail at 96 mm, 21.8% of total length; head elongate and compressed later- ally, dorsal head scales normal for the species; no azygous preinternasal scale, loreal single, preoculars 2-2, postoculars 3-3, temporals 1- 3, scale rows 21-21-17, ventrals 156, caudals 67, anal single, lower postocular with a firm contact on 4th supralabial (Fig. 5C). A longitu- dinal series in six rows of light rust or dark brown spots (rust spots faded to light spots in preservative) on a dark grey ground color; ventrals and caudals with dark pigmentation on anterior margins, forming dark and light cross bars on ventrals; caudals with reduced pigment forming a dark speckling on a light grey gound color; dorsal head plates dark greyish brown, grading to lighter shades on labials; latter with dark stripes crossing upper third or half of each scale and forming a dark, posterior margin on all but the last 2 suprala- bials; lower labials with posterior margins edged in dark brown; gulars light grey and without spots. Variation. — In this subspecies, there is lit- tle variation among scale and color pattern in the specimens examined. The 67 specimens seen show a small variation of only 8 to 14 scales difference in ventral plus caudal counts, with the greatest variation occurring in the ventrals of males. The one noticeable differ- ence is in the presence of the preinternasal azygous scale, which is seemingly present in most, if not all individuals in the headwaters of the Rio Bavispe. Five specimens from Chuhuichupa and its environs all have this 650 Great Basin Naturalist Vol. 45, No. 4 scale in several variations (Fig. 6), whereas few specimens on the east side of the Conti- nental Divide and those in the Rio Papagochic and Rio El Fuerte drainage do not (3 of 45 specimens). The most noticeable variation in the color pattern is the light, rust-colored spots in some specimens, whereas in others the spots are a dark brown; this difference persists in pre- served specimens as faded light spots. Conant (1963), referring to the field notes of R. G. Zweifel, indicates that the live snakes at Miiiaca had "reddish spots. " The population at San Pedro also had individuals with bright spots, but we determined them to be a bright yellowish rust, quite in contrast to the red colors seen in kingsnakes or bicolored Sonora. That this color pattern was not observed in all specimens collected or seen may suggest an expression of a seasonal or sex-induced char- acter. In a few specimens, the two dorsal rows of spots anteriorly contact each other dorsally, giving the appearance of only 5 rows. The rufipunctatus in northern and eastern Durango, Chihuahua, and Arizona have a di- vided nuchal blotch. This is best seen in young or juvenile specimens in which the nape has a narrow (usually one or a fraction of a scale) light stripe extending from the parietals to divide the dark, irregular blotches. In older specimens, there is a fading or perhaps a blending of the nape color pattern, making it more difficult to identify the divided blotches, especially in preserved specimens. In the young, the spots on the body are dark and readily noted. Posterior to the nuchal blotches in the young is a series of 1-3 dark, middorsal spots. The spots posterior to this series usually divide, forming two rows of dor- solateral spots and thus leaving the normal number of six rows of spots at midbody. In most specimens, the anterior margins of the ventrals are heavily pigmented, but with the posterior much lighter. This produces an even or an irregular cross-barring pattern of dark and light pigment. On the basis of other studies and our obser- vations, it is obvious that this subspecies oc- curs throughout the Sierra Madre Occidental of central and western Chihuahua. It is found primarily along streams or in ponds or mead- ows where fish, tadpoles, and presumably other small, aquatic or semiaquatic verte- brates are foraged. At San Pedro (near Mi- iiaca), we observed a subadult catch a small minnow. When we visited this locality (10 July 1958), the stream was low and clear and the small fish were abundant in the pools along the rocky stream bed. It was also evi- dent that the snakes were feeding on the fish from the fish odor we encountered when the snakes were handled. At a pond 8 km (5 mi) SW of Bocoyna, this species was feeding on tadpoles. The pond was formed between the road and railroad grades with rocks forming the walls around the pond. Along the water's edge, large num- bers of snakes were seen; when disturbed, they descended into the rock pile or the wa- ter. When they were handled, the fish odor was not present. The following information, noted as the se- ries was examined, may be of interest. Speci- men UAZ 26461 collected 7 December 1958 at Yepomera was heavily infested with Acanthocephala. Their spiny heads protruded from between the scales, presumably an at- tempt to escape the preservative solutions. Two males from Bocoyna (BYU 14213 and 17085) and one from San Pedro (BYU 14368), Chihuahua, had everted hemipenes. In each case there were three large spines on the base of the lateral surface and two on the posterior surface. Small spines, but descending in size, covered the organ from the large proximal spines to the enlarged capitate distal end. Specimen UAZ 26460 from Black River at Diamond Fork, Greenlee County, Arizona, did not differ in this character. Specimen UAZ 34335 from Yepomera, Chihuahua, was gravid, with 7 or 8 nearly mature young. It was taken 17 July 1971. ThamnopJiis rufipunctatus nigronuchalis Thompson Thomnophis nigronuchalis Thompson, 1957, Occ. Pap. Mus. Zool. Univ. Mich. 584:1-10. Diagnosis. — A subspecies o{ rufipunctatus with supralabials 4 and 5 having a wide contact with the eye and with a single, dark, median nuchal spot. Distribution. — Type locality San Luis, Durango. Specimens examined: UTEP 3386-7, 6 mi SW El Salto; UTEP 3653-4, 1.5 October 1985 TANNER: Snakes of Western Chihuahua 651 mi W San Luis; UAZ 37709, 17 mi N Coyotes; MVZ 59235, 33 mi ENE El Salto; LSUMZ 11637, 6 mi SW El Salto; LSUMZ 16488-16460, 8.3 mi W El Salto; LSUMZ 33100, 8 mi WSW El Salto; LSUMZ 40830-40834, 40849, 5.6 mi W El Salto; NMMZ 32511, 0.3 mi W El Salto, Durango. Remarks. — The above series is unusual in having few variations. The scale rows were 21-21-17 in most, the only variation occurring at the neck, where 5 specimens had 23 and 3 had 22 rows; supralabials 8-8 in most, but with 9 9 or 9-8 in a few; infralabials usually 10-10 and a few with 9-9 or 9-10; loreals 1-1; preocu- lars 2-2, rarely 3; postoculars 3-3 in all speci- mens examined; temporals 1-2 or 1-3 in about equal numbers. The color pattern consists of a single nape blotch usually wider than long, lying just pos- terior to the parietals. Posterior to the nape blotch is a series of smaller, middorsal spots that may separate toward midbody in two rows of dorsolateral spots; if this occurs, there are 6 rows of spots, othei-wise only 5 rows. Thompson (1957) states that there are 5-10 rows of spots. With some fusions and others splitting, one may at one point across the body count more than 6 rows. An examination of the spots along the entire body indicates that the basic number of rows is 6, reduced to 5 when the two dorsal rows remain fused. The range of this subspecies suggests that it is confined to the high altitude basins that flow from near the Continental Divide westward. Considerably more material must be obtained before its total range will be known and also whether it occurs only west of the Continental Divide and/or if it intergrades with r. unila- hialis in drainage basins north and east of the Continental Divide in Durango. Subspecies relationships and variation. — The recognition of three subspecies in T. rufipunctatus is based primarily on variation in color pattern, ventral-caudal scale counts, and the relationship of the 4th or 5th labials contacting the eye (Fig. 5). There is little vari- ation in the color pattern except that, in T. r. nigronuchalis, a single, large, nuchal blotch is present, whereas the northern subspecies have a divided blotch, at least in the young. When the spotting is apparent, there are usu- ally 5 or 6 rows of body spots, and the head is similarly patterned and colored in all three subspecies. In some specimens of each sub- species, there are bright, yellowish rust spots. In the subspecies nigronuchalis^ the two most dorsal rows of spots are often fused at least anteriorly, thus reducing the number of rows to 5, whereas in the other subspecies, the usual number of rows is 6. The five-row pat- tern is apparently a continuation along the dorsum of the central nape spot pattern in contrast to the divided nape spot pattern in the northern subspecies. The spotting in all subspecies is more apparent in the young and subadults, with some adults having a faded and less apparent spotted appearance, espe- cially in preserved specimens. There is some variation in the ventral color pattern. In the Chihuahua subspecies, the anterior edges of the ventrals are usually mar- gined with dark pigment forming dark cross bars. In r. rufipunctatus, the ventral pigmen- tation tends to form two rows of ventrolateral spots, with only small amounts of pigmenta- tion midventrally and in areas between the spots that are not pigmented. At least the barring effect is not obvious as in r. unila- hialis. In the series of nigronuchalis from 8.9 km (5.6 mi) W of El Salto, there is no dis- cernible ventral pattern. On about the ante- rior third of the body the ventrals are only flecked, but posterior ventrals show more and larger spots. In this series of specimens there isn't a uniform spotting or bar pattern. Variation in the ventrals and caudals is indi- cated in Table 2, and the variation in the labial-eye character is seen in Figure 5. In all specimens of r. nigronuchalis examined, there is a wide contact between the 4th and 5th supralabials and the eye, whereas in r. rufipunctatus, about 25% of the specimens have the lower postocular making contact with the 4th labial and thus excluding the 5th labial from narrowly contacting the orbit. However, in most specimens of the latter, there is usually not a firm or broad contact between the lower postocular and the 4th la- bial, leaving a small area of contact between the 5th labial and the eye or merely with the tip of the lower postocular touching the 4th labial. In both subspecies {nigronuchalis and rufipunctatus), there are usually 3 postocu- lars. Only in the subspecies rufipunctatus and unilahialis have I observed 4 postoculars (in rufipunctatus, 16 of 74 counts, 21.6%). An 652 Great Basin Naturalist Vol. 45, No. 4 increase in postoculars occurs in the Chi- huahua subspecies (43 of 128 counts, 33.6%). However, it is not the number of postoculars that prevents the 5th labial from contacting the orbit, but the size of the postocular scales. In Chihuahua specimens, the lower postocu- lar, whether it is the 3rd or 4th scale, extends around the eye to form a firm contact with the 4th labial, thus permitting only the 4th to enter the orbit (Fig. 5). In specimens with only 2 postoculars, there are always 2 suprala- bials (4 and 5) that enter the orbit. In the series from Arizona and Chihuahua, 7 speci- mens (of 101) had 2 postoculars on one or both sides. There are, however, small populational segments that do not fit all characteristics of these subspecies. An example is seen in the divided anal scale in the Oak Creek popula- tion of r. rufipunctatus. In r. unilabialis , a population in and around Galeana, Chi- huahua has some individuals with a ventral color pattern similar to individuals in New Mexico. Also, the few specimens seen south of the Rio San Miguel near the Durango-Chi- huahua border have a ventral-caudal count noticeably higher than the other Chihuahua specimens. In each case, these small popula- tions are widely separated from the main pop- ulations. Each of these allopatric subspecies occurs in a separate drainage basin. T. r. rufipunc- tatus is primarily in the Gila River Basin, occurring principally in the tributaries drain- ing the Mogollon Rim and the headwaters in east central Arizona and adjoining New Mex- ico. The greatest departure is in Chihuahua and northeastern Durango, in which such basins as the Rio Bavispe, Rio Papagochic, and Rio El Fuerte drain from the Continental Divide to the west and the Rio Conchos and other mainly northern small streams drain to the east. There is little variation in the popula- tions that inhabit these basins, seemingly be- cause they occur in the headwaters where contact, across the open basins on each side of the Continental Divide, is possible. There is, however, wide separation between the popu- lations in Arizona and Chihuahua. Specimens from the headwaters of the Rio Ba\ ispe re- semble the northern subspecies in that a high percentage of specimens have the azygous preinternasal scale. Since the Bavispe flows north and is part of the drainage basin in southern Arizona north from Douglas, the high frequency of this scale character in north- ern Chihuahua may represent a recent past contact with populations northward in east central Arizona and adjoining western New Mexico. The distribution of the subspecies ni- gronuchalis is as yet not well defined. Records available seem to place it primarily in the upper basin of the Rio Presidio, draining to the west. Intergrading specimens are not available. Some character variation, such as in the azygous scale of rufipunctatus and unila- bialis, were not observed in nigronuchalis. Conant (1963) states that the azygous scale character is present in two specimens from the Rio Nazas drainage; however, in these, there is seemingly an anomalous condition where corners of internasals are cut off to form an extra scale. The azygous scale as described above is not a fragment, but it represents a median scale fitted in between the rostral and the anterior ends of the two internasals. As far as is known, this character (Fig. 6) does not occur in any Durango population, including T. r. nigronuchalis. A specimen (UAZ 37709) from 37 km (17 mi) N of Coyotes has all the characters of ni- gronuchalis except that the nuchal blotch is lobate, that is, divided anteriorly but with the lobes connected posteriorly. In some speci- mens of unilabialis, the nape spots are nar- rowly divided, with only a fraction of a scale between them showing a light color. Speci- mens throughout the entire distribution of the species show color pattern similarities except for the single nape blotch in r. nigronuchalis and the ventral bar character in r. unihibialis. Only in a few characters is the variation suff^i- cient to permit a separation into subspecies. Data available suggest that subspeciation is recent, perhaps occurring after the last pluvial period when isolation between the northern (Arizona-New Mexico) and the southern (Chihuahua and Durango) segments of the species occurred (Fig. 7). Ke\ to the Subspecies I. A lar^f, single nuchal blotch and two suprala- hials (4 and 5) wideK' contacting orbit r. ni^ronuchali.s ( )ctober 1985 Tanner; Snakes of Western Chihuahua 653 Fig. 7. Distribution map indicating the approximate distribution of the subspecies oiThamnophis rufipunctatus : A, T. r. rufipunctatus, B, Thamnophis r. unilahialis ; and C, T. r. nigronuchalis. — Nuchal blotch divided, two supralabials may or may not contact the orbit 2 2(1). Lower postocular forming a firm contact with the 4th supralabial (Fig. 5C); ventral + caudal averages less than 240; ventrals with anterior margin pigmented to form a series of dark crossbars r. unilahialis — Lower postocular not forming a firm contact with the 4th supralabial, usually with one or both sides failing to contact the 4th supralabial; ventral + caudal averages more than 239; ven rals spotted but rarely with anterior margins evenly pigmented to form light and dark cross bands r. rufipunctatus Thamnophis melanogaster chihuahuaensis Tanner Tropidonotus melanogaster Peters, 1864, Monatsb. Acad. Wiss. Berlin, pp. 389-390. Thamnophis melanogaster melanogaster Smith, 1942 (part), Zoologica 27:116-120. Thamnophis melanogaster chihuahuaensis Tanner, 1959a, Herpetologica 15:170-172. 654 Great Basin Naturalist Vol. 45, No. 4 Bavispe River below Three Rivers, Chihuahua- Sonora line (BYU 14197, tvpe specimen; BYU 13451, 13505-6, 13493, 13496, 13371, 13373, 14198-14202, 14208-9, topotypes-paratypes). Cuiteco (BYU 14293). 6 mi SE Maguarichic (BYU 16914). Rio Urique approx. 10 mi below La Bufa Bridge (BYU 22696, 22698). Rio San Miguel (BYU 38331-37). Since the original description (Tanner 1959a), 10 specimens have been added to our series: 1 (BYU 16914), taken 6 mi SE of Maguarichic; 2 (BYU 22696 and 22698) fiom the Urique River approximately 10 mi below the La Bufa Road Bridge, and 7 from the junction of the Rio Verde and Rio Loera west to the junction of the Rio San Miguel and the Rio Urique. Nine are females, one the largest of the entire series at 862 mm, and one the smallest at 204 mm. The small one taken on 1 October 1963 is undoubtedly a newborn and exhibits on each side three rows of alternating spots that are separated by a dark brown dor- sal area. In none is there a middorsal light stripe. The larger specimens are a uniform olive brown dorsally and light slate with no dark spots ventrally. The scalation and color pattern are well within parameters listed for the type series. Conant (1963) discusses at length the subspe- cies T. m. canescens from Durango and Za- catecas. His material from the Rio Nazas and the Rio Florida in northern Durango indicates that canescens may occur in southeastern Chi- huahua in the southern tributaries of the Rio Conchos. All specimens of T. m. chi- huahiiaensis have been taken in the Bavispe and El Fuerte River basins, both of which drain to the west. If intergrading populations occur, it will seemingly be in the headwaters of the Nazas southwest of Hidalgo del Parral or west in the upper tributaries of the Rio San Miguel. This species is found along permanent streams or ponds, not very far from water. Its aquatic habits were first observed as we col- lected the type series below Three Rivers on the Bavispe River. The snakes would climb onto the lower branches of a willow that hung over the river; any disturbance and they would fall into the water. We succeeded in catching them by getting into the river near the overhanging willows, and, as the snakes came up for air, we could catch them. Fortu- nately, the river was riled from recent rains — an aid, since the snakes could not see us. The large specimen from the Rio Urique had eaten a fish. TJiamnopJiis cyrtopsis cyi-topsis (Kennicott) Eiitaeniacyrtopsis Kennicott, 1860, Proc. Acad. Nat. Sci. Phila. 12:.333. Type locality, Rinconada, Coahuila, Mexico. Thamnophis cyt-topsis cyrtopsis Smith, 1951, Copeia 1951:140. Thamnophis cyrtopsis cyrtopsis Webb, 1980, Cat. Amer. Amph. and Rept., p. 245. Bavispe River below Three Rivers and near the Sonora-Chihuahua line, 11 (BYU 13367, 13.370, 13372, 13583, and 14471-77). Black Canvon 8 mi W Chuhuichupa, 1 (BYU 14222). 4.5 mi NE Colonia Juarez, 2 (BYU 14509 and 15464). 30 mi NNW Colonia Juarez (Tinaja Canyon), 1 (BYU 1.3443). Colonia Juarez, 1 (BYU 17696). 22 mi S Creel, 2 (BYU 16953-4). 25.5 mi S Creel, 4 (BYU 17687-90). Cuiteco, 2 (BYU 14269 and 1.5666). 6 mi S Hidalgo del Parral, 1 (BYU 13925). Maguarichic, 3 (BYU 16909-11). 4.5 mi SE Maguarichic, 1 (BYU 17073). 10 mi SE Nuevo Casas Grandes, 2 (BYU 1.5316 and 17111). 14 mi W San Francisco del Oro, 1 (BYU 15711). Tejaban, on rim of Barranca del Cobre near air strip, 1 (BYU 32038). Rio San Miguel, 5 (BYU ,38,339-43). Webb (1966) lists 22 specimens from Chi- huahua, 12 from Durango, 5 from Sonora, and 4 from Zacatecas. The range from Zacatecas north through Durango and Chihuahua is pri- marily in the mountains, the foothills, and the high plains to the east of the mountains. Only in the north does its range reach the Gulf Coast in Sonora. In the north it ranges into New Mexico, Arizona, western Texas, south- ern Colorado, and the upper basin of the Col- orado River in eastern Utah. This species was reviewed by Webb (1966), who listed three subspecies distributed in the southwestern United States and Mexico. The subspecies cyrtopsis is found throughout Chi- huahua, except for the deep canyons of the southwest and is one of, if not the most, wide- ' spread and common snakes. Although it is commonly found along streams and mesic ar- eas, it does occur in rather remote desert i areas where springs occur. Scale counts are based on 38 Chihuahua specimens taken from most areas west of October 1985 TANNER: Snakes of Western Chihuahua 655 Highway 45. Ventrals range from 164 to 185, with the males averaging 177.3 and the fe- males 171.1. Caudals range from 73 to 105, with the males averaging 97. 1 and the females 86.5. These averages are approximately four scales higher than those reported by Webb (1966). Some of this discrepancy may have resulted from the fact that most of our speci- mens were secured in the higher elevations, where it is noted that scale counts are higher. The dorsal rows are consistently 19-19-17, ex- cept for one specimen from 4.5 mi SE of Maguarichic, which has 21-21-19 rows. This specimen is otherwise quite average for other mountain specimens. There is little variation in the head scales, there being with few ex- ceptions 1 loreal; 1 preocular; 3 postoculars, with 2 or 4 occurring occasionally; temporals 1-2 or 1-3; supralabials 8, rarely 7 or 9; infrala- bials usually 10 but with 9 out of 80 counts having 11. The color pattern is variable and fits gener- ally the description given by Webb (1966). The middorsal stripe may be on one row, or it may involve the para-vertebral rows to as much as one-half of each scale. Anteriorly this stripe reaches to the parietals or terminates 1 or 2 scales posteriorly. If it reaches the pari- etals, the dark nape spot is divided by it. A wide middorsal stripe may involve 3 or 4 rows just posterior to the nape spot. Nevertheless in those specimens with only 1 row involved, little or no increase in size occurs at any point from nape to tail. Specimens with the narrow middorsal stripe may be dark between the lateral and dorsal stripes, thus resembling the mountain forms of Thamnophis eqties . In such specimens only the reduced scale rows (19) and the lateral stripe on rows 2 and 3 sei-ve to distinguish them from eques. Specimens vary not only in the width of the dorsal stripe, but also in the amount of pigment between the dorsal and lateral stripes (Webb 1980). As noted above, this may result in some misiden- tifications. Whether this dark color pattern is genetically or altitudinally (environmentally) induced convergence in eques virgatenuis or c. cyrtopsis is an interesting yet unsolved speculation. In those specimens with distinct wide stripes and lighter ground color, a distinct spotting occurs above as well as below the lateral stripe. This is particularly true for young and juvenile specimens. In all speci- mens the sutures of the supralabials are edged with pigment. The sutures of the infralabials may or may not be edged with pigment, but never to the same extent as with the suprala- bials. Thamnophis cyrtopsis collaris (Jan) TropiHonotuscollarisjSLn, 1863, Elenco Sist. Ofidi, p. 69. Thamnophis cyrtopsis cyclides Smith, 19.51, Copeia 1951(2): 140. Thamnophis cyrtopsis collaris Webb, 1966, Tulane Stu- dies in Zoology 13(2):55-70. 12 mi above Pitahaye, 1 (BYU 22685). 30 mi below (S) Guaruchic, 1 (BYU 22684). 2 mi N Cerocouhui, 3 (BYU 14478, 14.595, 15651). 10 mi down Urique River from La Bufa Road Bridge, 2 (BYU 22690 and 22699). Piedras Verdes, (near mouth of Rio San Miguel) 1 (BYU 22683). Webb (1966) lists 5 specimens from Urique. The most distinguishing character is the dark (nearly black) nape band, which is not divided by the middorsal stripe (Webb 1980). In specimens from Cerocouhui, the dorsal stripe indents the collar but is several scales from the parietals. There is an increase in the caudals 91(101.2)110, based on 5 male speci- mens. This is 4 more than is present in 10 male Chihuahua specimens of cyrtopsis. Other- wise, the scalation is similar. Specimens taken on the high plateau areas, that is, out of the deep barrancas, are cyrtop- sis. This suggests that collaris is generally confined to the low canyon habitats of south- western Chihuahua. Thamnophis marcianus marcianus (Baird and Girard) Eutaenia marciana Baird and Girard, 1853, Cat. N. Amer. Reptiles, pp. 36-37. Thamnophis marciana Ruthven, 1908, U.S. Nat. Mus. Bull. 61:849-8.52. Thamnophis marcianus marcianus Rossman, 1971, Occ. Papers Mus. Zool. Louisiana State Univ. Bull. 41:11. Colonia Juarez, 2 (BYU 15466-7). Only two juvenile specimens were col- lected from Juarez Creek, a tributary of the Rio Casas Grandes. Distribution of this spe- cies in Chihuahua is apparently restricted to the northeastern part of the state, and seem- ingly in the drainage of those streams flowing northward and at present terminating in the desert lakes (now playas) in northern Chi- 656 Great Basin Naturalist Vol. 45, No. 4 huahua. It is assumed that their occurrence in this area is an extension of their range south- ward from Texas and New Mexico. The scalation and color pattern fit within the ranges established by Rossman (1971) for both male and female specimens. However, the ground color and spotting is identical to specimens seen from Brewster and Jim Hogg counties in Texas. The light, parietal spot edged in black is very prominent and is thus in contrast to a series from Charleston, Cochise County, Arizona, a further indication that the Chihuahua population is an extension of the Texas population during the last Pluvial Pe- riod. The mesic areas between the closed basins in northern Chihuahua and the Rio Grande basin has been ecologically divided since perhaps the late Pleistocene. Thamnophis elegans errans Smith Thamnophis oriUnoides errans Smith, 1942c, Zoologica 27:112. Thamnophis ordinoides Fitch, 1948 (part), Copeia 1948:121-126. Thamnophis elegans errans Tanner, 1959a, Herpetolog- ica 15:168. Thamnophis elegans errans Wehh, 1976, Nat. Hist. Mus. of Los Angeles Countv Bull. 284:1-1.3. Bocoyna, 1 (BYU 1.5742). Black Canvon, 8 mi W of Chuhuichupa, 1 (BYU 14225). Colonia Chuhuichupa, 21 (BYU 1.3889-96. 13921-23, 14479-81, 14492-95, 14501, 14505, 15721, and 15776). 2 mi S Creel, 5 (BYU 14381, 14511, 15644, and 17076-7). The series listed above was seen by Webb (1976). His review of this elegans subspecies cannot be added to at this time, except to note that Tanner (1959a) arrived at essentially the same conclusion, namely that errans was in- deed a southern subspecies of the widespread species Thamnophis elegans . Tanner (1959a) and Webb (1976, 1980) dis- cussed the characteristics of the specimens available from northwestern Chihuahua, that is, areas north of the Rio Papigochic. Speci- mens from this northern area show consider- able overlap in characters with T. elegans va- grans in Arizona, New Mexico, and Utah. This is particularly evident in the dorsal scale rows, which may be 21-21-17 or 19-19-17 or a combination. South of the Rio Papigochic the scale row formula is 19-19-17 and there is an increase in the caudals (Creel area 89.75 and Chuhuichupa area 82.66). A reexamination of the BYU errans specimens suggests a strong relationship to the elegans vagrans population in central and southern Arizona. Although no T. elegans are known from the intervening area, the terrain is favorable for contact through the Bavispe River and its tributaries extending from southern Arizona (near Dou- glas) south, by way of the Rio Bavispe, into western Chihuahua (Chuhuichupa area). Fitch (1948) raised errans to a full species on the basis of the reduced scale rows (19) and the reduced infralabials (9-9). Both these scale patterns occur in many, if not most, popula- tions of vagrans , and, since the ventral-caudal counts are essentially the same, there are no scale character diflPerences sufficient for con- sidering errans to be other than a disjunct subspecies of elegans . In addition the color patterns are similar, but with the southern populations of vagrans (central Arizona) and errans having a more distinct dorsal stripe, usually involving three dorsal rows, and with little or no invasion of the middorsal row by dark spots. The stripe is thus more uniform and distinct and unlike the undulating stripe in elegans vagrans from central Utah. Fitch (1980) considered again the status of Thamnophis elegans errans. 1 would add only two additional comments. First, the species Thamnophis elegans is widely dispersed and as such has developed clinal variations be- cause of its wide dispersion and isolation in some of the desert basins. The color pattern of the populations in the upper Colorado River basin of southeastern Utah is clearly distinct i from those seen in the Little Colorado basin of Arizona. Specimens from Joseph City and near Holbrook, Arizona, have a wide dorsal stripe in which three rows of scales are in- volved; no dark spots invade it — or at least they do not disrupt the stripe as they do in the upper basin specimens. The color patterns in errans is thus similar to the Arizona T. e. vagrans. Fitch (1983) did not include errans in his report and distribution map of Thamnophis elegans, choosing to include only those^populations north of Mexico. A perusal of my field notes indicates that errans has essentially the same wandering habits as does vagrans. We found them along streams or ponds, in meadows, and in fields ( )ctober 1985 Tanneh: Snakes of Western Chihuahua 657 during the rainy season. The distribution of Thamnophis dedans in southeastern Utah, Arizona, and Chihuahua is somewhat similar to that ofLampropeltis pyromelana , that is, in chsjunct populations. Thamnophis eh'^ans oc- curs in separate basins with wide desert areas separating them. In Chihuahua the extent of the deserts, partieularly between those in Ari- zona and Chihuahua, has required the popu- lations either to inhabit streamside habitats or to move into cooler, moist, moimtain areas where habitat restrictions are not as severe. The wide desert areas between errans and va^rans suggest again that, in the recent geo- logical past, more favorable climatic condi- tions existed, permitting T. elegans to be widely dispersed over areas now desert and uninhabitable. However, as a consequence of this former distribution, those populations in northern Chihuahua would be expected to show remnants of intergrading or intermedi- ate characteristics with those in Arizona not only in scalation, but also in color pattern. These factors suggest that the contact be- tween them has been in the recent past. As indicated above, the population of errans south of the Rio Papigochic has apparently been sufficiently isolated to have developed the basic characteristics of the subspecies. Nevertheless, on the basis of the material available to me, the type series (Smith 1942c) and the report by Webb (1976), it now appears that the more typical errans are found on the east side of the Continental Divide in north- ern Chihuahua (Garcia area) and south of the Papigochic River in southwestern Chihuahua and Durango, with the population in the headwaters of the Bavispe River (Chu- huichupa area) showing intergrading charac- teristics with the elegans vagrans to the north. In 1957 and 1958 collecting was done dur- ing July (7-20), and each year the large females were gravid. Two (BYU 14493 and 15721) from Chuchuichupa had 6 and 7 nearly ma- ture embryos, and two (BYU 17076 and 17077) from Creel had 10 and 8 embryos. The latter two specimens were 590 and 498 mm in total length. The largest specimen in the se- ries is a female, 635 mm in total length. Thamnophis eques megalops (Kennicott) Eiitaenia megalops Kennicott, 1860, Proc. Acad. Sci. Phila. 12:330-331. Thamnophis macrostemma megalops Smith, 1939, Piibl. Field Mus. Nat. Hi.st., Zool. Ser. 24:30-31. Thamnophis eques megalops Smith, 1951, Copeia 1951:139-140. Conant, 1963, Copeia 1963:487. Basiguare, 1 (BYU 22701). Black Canvon, 8 mi W of Chuhuichupa, 6 (BYU 14221, 14223, 15744-46 and 15769). Colonia Duhlan, 1 (BYU 13731). Colonia Juarez, 2 (BYU 243 and 1134). 5 mi S Gomez Faria.s, 1 (BYU 15747). Guachochic at spring, 1 (BYU 22687). 10 mi S of Hidalgo del Parral, 1 (BYU 13924). 2 mi N Casas Grandes, 2 (BYU 14136-37). 4.5 mi S of Palomas, 1 (BYU 14504). 2 mi SE La Junta, 1 (BYU 17084). San Pedro on Rio Papigochic, 2 (BYU 14376 and 14487). Conant (1963) listed 10 additional speci- mens from the areas west and south of Hidalgo del Parral, and the following are at the Uni- versity of Texas at El Paso: El Sauz, Rio Sauz, UTEP 3648; Ricardo Flores Magon, 3659-60; and Yepomera, 2053. I have examined only the specimens available to me at BYU. These (19) vary little in scalation from that of Conant (1963:488) and are as follows: ventrals males 166(168.7)170; females 160(164.4)171; caudals males 83(84.8)87; females 69(73.5)78. Scalation of the head is similar in all details to that indicated by Conant for both subspecies megalops and virgatenuis. I have not undertaken a systematic review of this species covering its entire area of distri- bution, although such a study is clearly war- ranted as suggested by Conant (1963:488). The color pattern varies on an altitudinal cline, with the wide, middorsal stripe occur- ring at low altitudes, that is, the low valleys and foothills primarily to the east of the west- ern mountains. The narrow, one-scale-wide, middorsal stripe occurs only in populations at the higher elevations. There is, based on the above series, a grad- ual reduction in the width of the dorsal stripe from those specimens from low elevations (formula 1/2-1-1/2) to the point that some specimens in the highlands (Bocoyna and Chuhuichupa) have an invasion of pigment into the middorsal scale row, reducing the stripe to a fraction of the row (- 1-) (Conant's formula, 1963:490). If we consider any pig- ment encroachment into the paravertebral part of the middorsal stripe as an indication of intergradation between the subspecies, then a wide area in Chihuahua would lie in the zone 658 Great Basin Naturalist Vol. 45, No. 4 of intergradation. We find "true" megalops only in the lower valleys and virgatenuis only in the highest elevations. There are two color pattern variables: first the size of the middorsal stripe depends on the degree of involvement of the paraverte- bral rows in the dorsal stripe, and second the melanistic pigment gradually increases to pro- duce a nearly uniform dark pattern between the ventrolateral stripes and the middorsal stripe. All the variations figured by Conant (1963) are present in the Chihuahua speci- mens, but the above color pattern phe- nomenon is most noticeable in those speci- mens from the area of 6000 feet and higher. A few specimens from Black Canyon just west of Chuhuichupa and at about the same elevation range from little involvement of the paraver- tebral rows to a full one-half scale row involve- ment. There is little variation in color pattern between the Black Canyon and Chuhuichupa specimens, perhaps a little less involvement of the paravertebral rows in the latter. How- ever, Conant (1963) considered the Chu- huichupa specimens to belong to the subspe- cies virgatenuis even though the formula for this population would be +1+ in all speci- mens. In both populations the area between the dorsal stripe and the lateral stripes is also nearly black, with only fine, light flecks. Ap- parently the only useful key characters to dif- ferentiate virgatenuis from megalops are for the dorsal stripe to occur only on the middor- sal row and for the color between the lateral and dorsal stripe to be black or nearly so, with fine, white flecks showing between the scales. Thamnophis eques virgatenuis Conant Thamnophis eques virgatenuis Conant, 1963, Copeia 1963:490. Bocoyna, 12 (BYU 14377-9, 1.57.35-41, 17087-90). 4 mi S Bocoyna, 6 (BYU 17091-95, 14380). 1 mi S Chuhuichupa, 2 (BYU 13919-20). Chuhuichupa, 7 (BYU 13896-97, 14482, 14497-500). 2 mi SE Creel, 5 (BYU 15641-43, 17074-75). 16 mi NE San Juanito, 2 (BYU 17028-29). San Juanito, 1 (BYU 32039). Colonia Ciarcia, 1 (BYU 246). 5 mi WSW Colonia Carcia, 1 (UTEP 4850). Conant (1963) also lists 1 from Bocoyna (AM Nil 74465) and 2 from Sisoguichi (AMNH 57389-90). Although there is considerable intergra- dation in the color and color patterns of the populations of Thamnophis eques in Chi- huahua and little variation in their scalation and body proportions, it is seemingly wise at this time to let the subspecies T. e. vir- gatenuis stand pending a detailed study of the species. The color pattern changes in the Chi- huahua populations are gradual, with speci- mens taken between 6000 and 8000 feet show- ing the variants one might consider to represent the area as well as the range of intergradation. In a few specimens from Bo- coyna, there is an invasion of the middorsal stripe by pigment to the point of reducing the stripe to only a fraction of a scale row. In this character, the evolutionary process may thus be moving to eliminate the middorsal stripe. At least the effect of the color pattern changes is toward a sustained though gradual reduc- tion of the size of the middorsal stripe. Specimens with the virgatenuis color pat- tern were not found south of Creel in the Barranca del Cobre and adjoining areas. It now appears that the population in the Bo- coyna area is not only separated from the high- land popidations in Durango, but it is also separated from the Chuhuichupa population by the low areas in the basin of the Rio Papigo- chic. Thus, in Chihuahua, we now recognize two highland islands of the subspecies T. eques virgatenuis, both of which are sur- rounded by the subspecies T. eques megalops in the lower mountains, foothills, and valleys of western and southwestern Chihuahua, and with intergrading population in between. The highlands of southern Chihuahua and northern Dinango are not well known and may hold the answers to many distribution patterns not only for the several species of Thamnophis , but also for other species. Thamnophis sirtalis dorsalis Baird & Girard Eutaenia dorsalis Baird and Girard, 1853, Cat. N. Amer. Reptiles, p. 31. Eutaenia ornata Baird, 18.59, Rep. U.S. and Me,\. Boundary Surv. 2:1-35. Tfianmopliis sirtalis ornata F'itchand Ma.slin, 1961, Uni\. Kansas Puhl. Nat. Hist. 13(5):297-299. Tliainn(>i)his sirtalis dorsalis Fitch, 1980, Cat. .\mer. , Amph. and Rept., p. 270. 6.2 km SE Galeana, Ojo de los Reves 2 (UAZ 32780 and 36291). O.S km N Nuevas Casas (irandes 1 (UAZ 34434). Yepomera vicinity 13, (UAZ 34066-71, 34149, 34230, 34399, 34879-82). October 1985 TANNER: SNAKES OF WESTERN CHIMUAHUA 659 The specimen taken at Casas Grandes by Nelson and Goldman (USNM 46371) was re- ported by Smith (1942c), and a brief descrip- tion was inclnded. It is a female with the following scale counts: 19-19-17 scale rows, 152 ventrals, tail incomplete, 7-7 supralabials, 10-11 infralabials, 1-1 preoculars, and 3-3 postoculars. A specimen from New Mexico (E. D. Flaherty, no. 560, 1 mi W and 1/2 mi S of Isleta, Bernalillo County) reported by Fitch and Maslin (1961) is stated as having only 8-8 infralabials. This seems most unusual for any specimen o(sirtalis. Of 114 counts (57 speci- mens) of s. sirtalis and s. parietalis (from Utah and Kansas), only one had a count of 8-9; all others were 9, 10, or 11, with 10 being the predominant count. Based on the characteris- tics of the few specimens previously reported, the normal lepidosis for this subspecies re- mains obscure or wanting for additional data. An examination of the scalation and color patterns of the series reported by Van Deven- der and Lowe (1977) suggests that the popula- tion in the two northern valleys (Rio Casas Grande and Rio Santa Maria) corresponds closely to those specimens previously re- ported by Smith (1942c) and Fitch and Maslin (1961). There are still two few specimens, to establish basic data for this southern popula- tion. The two specimens from near Galeana (UAZ 32780 and 36291) have fewer ventrals than the other specimens (from N Casas Grande and Yepomera) and are similar in scalation (ventrals 151 and 152, labials 7-7 and 10-10) to the one reported by Smith (1942c) from Casas Grandes. Since Nelson and Gold- man collected extensively north and east of Casas Grande but returned to that city or San Diego, where they reorganized for further trips, it is possible that the specimen (USUM 46371) was taken in the drainage of the Rio Santa Maria. It has the basic characteristics of those seen from near Galeana. The single specimen from N Casas Grandes is a male with 166 ventrals and 84 caudals. These char- acters relate it to the Yepomera population and justify the suggestion of Van Devender and Lowe (1977) that the Yepomera popula- tion came into its present habitat by way of the Rio Casas Grande. The Yepomera population has developed a few characters not seen in most other popula- tions of sirtalis. These will be discussed in a forthcoming study. The occurrence of sirtalis in the valleys of northern Chihuahua is not surprising and is, undoubtedly, a relict population isolated after the desiccation following the last ice age severed the former water courses directly connected with the Rio Grande. This is yet another example of the effects of the desicca- tion that occurred as the Pluvial Lakes such as Palomas disappeared and rendered the low- lands uninhabitable. We can now assign such species as Lampropeltis pijromelana, Opheodreys vernalis, Thamnophis elegans, and now Thamnophis sirtalis as relict populations. Genus Trimorphodon There are two species that occur in Chi- huahua, with only one specimen of b. vilkin- soni and one of b. lambda having been col- lected; their distribution in the state is poorly documented. The secretive and nocturnal habits may account for the scarcity of collected specimens. Trimorphodon tan occurs in southwestern Chihuahua, inhabiting the deep canyons of the barrancas. Trimorphodon biscutatus vilkinsoni Cope Trimorphodon vilkinsoni Cope, 1886, Proc. Amer. Philos. 23:285-286; Tavlor, 1939, Univ. Kansas Sci. Bull. 25:361-363; Smith, 1941b, Proc. U.S. Nat. Mus. 91:151-167. Trimorphodon biscutatus vilkinsoni Gehlbach, 1971, Herpetologica 27:209. Trimorphodon biscutatus vilkinsoni Scott and McDi- armid, 1984a, Cat. Amer. Amph. and Rept. 353:1-4. Ciudad Chihuahua, 1 (USNM 14268, type speci- men). A description of the type is in the report by Taylor (1939b), and Smith ((1941b) provides a diagram of the possible phylogeny of the genus. The above reports dealt with only one or two specimens. Klauber (1940b) reported on three, Reynolds and Scott (1977) list three taken on Chihuahua Highway 16 between Villa Aldama and El Pastor, and Banicki and Webb (1982) have described in detail a series of 22 specimens from the Franklin Mountains in El Paso, Texas. However, the scarcity of specimens for Chihuahua still exists. Our many trips into Chihuahua produced no spec- imens, but we do have one additional female specimen from Green Gulch, Chisos Moun- 660 Great Basin Naturalist Vol. 45, No. 4 tains, Brewster County, Texas. It has the fol- lowing scalation: scale rows 21-23-17, ventrals 231, caudals 74, supralabials 9-9, infralabials 12-13, loreals 2-2, preoculars 3-3, postoculars 2-3, and temporals 2-2-3 and 3-3-4. There are 24 body spots and 9 on the tail. These patterns correspond closely to those reported by Klauber (1940b) and Banicki and Webb (1982), but they do indicate that additional specimens from central Chihuahua may provide more variation than that seen in the specimens now available. Trimorphodon hiscutahis lambda Cope Trimorphodon lambda Cope, 1886, Proc. Anier. Philos. 23:285-286. Trimorphodon biscutatus lambda Gehlbach, 1971, Her- petologica 27(2);208. Maguarachic, 1 (UMMZ 118926). This female specimen has an unusually high ventral count for this subspecies. Other scale counts and the color pattern are, however, well within the limits of the subspecies. The scale counts are as follows: ventrals 255, cau- dals 67, anal divided, dorsal scale rows 19-23- 16, supralabials 9-9, infralabials 12-12, preoc- ulars 3-3, loreals 3-3, postoculars 3-4, and temporals 3-4. There are 26 dorsal body spots and 12 on the tail. At or near midbody the dorsal spots involve 12-15 rows of scales, with a dark spot extending from the lateral point of the spot to the first row above the ventrals. Remarks. — Using only the specimens available to me in the Monte L. Bean Life Science Museum (BYU), I find the series from Utah, Nevada, Arizona, and Sonora (16 speci- mens) to have a ventral range of 218-(221)-226 in males 227-(232.57)-246 in females. By adding the ventrals and caudals, the eight southern Arizona and Sonoran specimens and the Chihuahua specimen average 320 and range from 303 to 322, but the more northern specimens range from 292 to 301. This em- phasizes the clinal decrease in these counts from south to north. All the specimens listed under T. bisciitata have the chevron head pattern and a similar body pattern. Hardy and McDiarmid (1969) list Trimor- phodon lambda paiicimaculata as occurring in northern Sinaloa. An examination of our se- ries of Trimorphodon iVom Sonora (12) and Sinaloa (2), and 10 specimens from Arizona, Nevada, and Utah, suggests that only two spe- cies of Trimorphodon occur in Sonora and perhaps also in Sinaloa, T. biscutatus and T. tau . Two male specimens from southern Sinaloa (6.7 mi N Rio Quelite) are T. b. biscu- tatus and have low ventral counts (192-197), suggesting a reversed cline from south to north. It is also suspected that any specimen previously identified as lambda or perhaps paucimaculata actually belongs to either the species T. biscutatus or T. tau (see Scott and McDiarmid 19S4b:353, 1984:354). Trimorphodon tau tau Cope Trimorphodon tau Cope, 1869 (1870), Proc. Amer. Phil. Soc. 11:152. Trimorphodon tau McDiarmid & Scott, 1970, Contribu- tions in Sci., Los Angeles Countv Museum, no. 179:29. Trimorphodon tau tau Scott & McDiarmid, 1984b, Cat. Amer. Amph. and Rept. 354:1-2. Batopilas, 1 (USNM). Only a single specimen is listed for Chi- huahua. The species was collected by Edward Wilkinson and reported by Cope (1900:1105). The reports by McDiarmid and Scott (1970, 1984b) cite the locality on their range maps. The occurrence of T. tau in Chihuahua is to be expected, since a number of specimens have been taken in northeastern Sinaloa only a few miles from Chihuahua. Additional col- lecting in the low valleys of southwestern Chi- huahua will undoubtedly pro\ide additional material. Since this report was written, the catalog reports by Scott and McDiarmid (1984b) confirm the distribution of the above species in southwestern Chihuahua, Sinaloa, and Sonora. Family Elapidae On the basis of published reports (Cope 1900:1126, Bogert and Oliver 1945:407, Smith and Taylor 1945:169, Hardy and McDi- armid 1969:211, Zweifel and Norris 1955:245-8), there are seemingly only three | records of elapids for Chihuahua. Both genera {Micruroides and Micrurus) occur in the southwestern part of the state, with their ranges extending westward into Sinaloa and Sonora. ( )ctober 1985 TANNER: Snakes of Western Chihuahua 661 Micruroides euryxanthus australis Zweifel & Norris Elaps etin/xdnthus Kennicott, 1860, Proc. Acad. Nat. Sci. Philadelphia, p. 336. Micruroides euryxanthus australis Zweifel and Norris, 1955, Anier. Midi. Nat. 54:245-248. Batopilas, 1 (USNM S850- Wm. Grant, collector). Cope's record from Batopilas is referred to the subspecies e. australis by Zweifel and Norris (1955) and Roze (1974).' This designa- tion is most logical because of the nearness of Batopilas to the type locality (Guirocoba, Sonora), and because both are in the drainage of the Rio El Fuerte. The record (USNM 8566) from the north- west corner of Chihuahua is recorded as M. e. euryxanthus by Zweifel and Norris (1955:246) and Roze (1974:163). The subspecies referred to above undoubtedly have a wider distribu- tion in western and northwestern Chihuahua. They are poorly represented in collections primarily because little collecting has been done in the fringe areas of their distribution. Micrurus distans distans (Kennicott) Elaps distans Kennicott, 1860, Proc. Acad. Nat. Sci. Philadc-lphia, 12:338. Micrurus diastema distans Schmidt, 1933, Puhl. Field Mus. Nat. Hist., Zool. Ser. 20:39. Batosegachic (= Batosegachie of Smith and Tavlor 1945), 1 (USNM 1144). This locality is in southwestern Chihuahua, west of Cuiteco and approximately half the distance from Cuiteco to the Sonoran line. It is in the drainage of the Rio Oteros, a northern tributary of the Rio El Fuerte. Roze (1983) gives the distribution as southwestern Chi- huahua and southern Sonora to Sinaloa and northwestern Nayarit, intergrading with the subspecies zweifeli in central Nayarit, Mexico. The few records now available for the two genera of coral snakes indicate that they have overlapping ranges in southwestern Chi- huahua and the adjoining terrain in eastern Sonora and northeastern Sinaloa. Family Crotalidae Seven species of this family are known to occur in Chihuahua. Four species or their subspecies occur in the lower valleys and foothills of central Chihuahua, with four spe- cies or their subspecies more commonly found in the western mountains. The single published record for C. v. viridis is for a speci- men taken in the 185()s and is apparently valid. A second specimen, CAS-SU 14361 col- lected by Figg-Hoblyn et al. on 24 August 1950, 25 km (16 mi) E Chihuahua-Sonora bor- der on the road between the towns of Bavispe (Sonora) and Nuevo Casas Grandes, is a con- firmation of the distribution of Crotalus v. viridis occurring in northwestern Chihuahua. It is actually a slight extension southward of the range indicated by Klauber (1936). Inten- sive collecting in northern Chihuahua and Sonora may extend the range of viridis farther south and west. If additional species of this family are in Chihuahua, the most likely area would be in river valleys extending into the mountains of southwestern Chihuahua. Other species may occur in the low valleys of the Urique, El Fuerte, and Oteros rivers that are not found elsewhere and represent species ascending these valleys from the coastal plains. We might expect that C. basiliscus will be found in one or all of these valleys. In summary, there are seven rattlesnake species in the state of Chihuahua, and of these four are more commonly found in the moun- tains; that is, wiUardi, pricei, lepidus, and molossus. Of these, lepidus and rnolossus (Price 1982) do reach the desert foothills east of the mountains, but atrox and scutulatus are seemingly restricted to the deserts east and north of the mountains. Crotalus molossus molossus Baird & Girard Crotalus molossus Baird & Girard, 1853, Cat. N. Amer. Kept., p. 10. Crotalus molossus molossus Gloyd, 1936a, Occ. Papers, Mus. Zool. Univ. Michigan 325:2. 23 km SW Colonia Juarez (on road to Chuhuichupa), 2 (BYU 13875-6). 3 mi N Chuhuichupa, 5 (BYU 13873-4 and 15382-4). 3 mi E Colonia Garcia, 2 (BYU 15380-1). 11.3 mi W R. F. Magon, 1 (BYU 15251). 11 mi S Creel (on La Bufa Road), 1 (BYU 15399). Head of Arrovo Guachochic (southwestern Chi- huahua), 1 (BYU 22688). Gloyd (1940) lists the following locahties: Colonia Garcia, District of Guerrero, Pache- co, and San Bias Mountains. The series available to me does not suggest that, contrary to the distribution maps of Price (1980), C. m. nigre^cens occurs in Chihuahua. 662 Great Basin Naturalist Vol. 45, No. 4 There is no doubt that the color (dark brown to black on tail and posterior one-fourth to one- third of body) indicates a modification of the color pattern in contrast to most specimens seen from Arizona and New Mexico. How- ever, the scalation of the Chihuahua series does not fit that of m. nigrescens . The sum- mary of scale rows, ventrals, and caudals fits well within those listed for C. m. molossus bv Gloyd (1940) and by Klauber (1952, 1956). The scale rows are consistently 27 or 29 at midbody in contrast to 23 or 25 in m. nigres- cens south in Durango. Ventrals are 185-194, with males averaging 188 and females 193, much higher than in m. nigrescens , but aver- age for m. 77wlossus. A key based only on color pattern would place some of the specimens from the western mountains either as m. ni- grescens or as intergrades of tnolossiis x ni- grescens, depending on the specimen at hand. A key based on scalation would place all Chihuahua specimens we have taken with ?n. molossus . All indications are that the color and color pattern show intergradation between the sub- species molossus and nigrescens in the moun- tains of southwestern Chihuahua, but with scalation remaining as in m. molossus. A se- ries from northern Durango may indicate where the scale variation approaches that of nigrescens and thus establish the area of in- tergradation. It now appears that intergrada- tion is occurring over a much wider area than has been indicated by previous studies, per- haps in extreme southern Chihuahua and northern Durango. However, most speci- mens we have collected in southwestern Chi- huahua (Chuhuichupa and south) do not show consistently the closed blotches above the ventrals as indicated by Klauber (1936, Fig. 64). Quite to the contrary, most have the dark edge of the blotches extending to the ventrals (his Fig. 63). Price (1980) indicates by map that a finger-shaped zone of intergradation extends north from Durango into the moun- tains of western Chihuahua. This area of in- tergradation would include Creel and north to near Chuhuichupa. We have collected six specimens from these areas, all of which are m. molossus. Only the specimen from Arroyo Guachochic in southwestern Chihuahua (near Rio San Miguel) could be considered an inter- grade, and this is based only on the increased pigmentation on the posterior third of the body. All scalation in this male also remains with the subspecies molossus: scale rows at midbodv 27, ventrals 190, and labials 18-19 and 17-i7. Based on literature records (Gloyd 1940, Klauber 1936, 1952, Hardy and McDiarmid 1969) and the series available (at BYU), it is surmised that the area of intergradation is south of the Rio San Miguel in Chihuahua, in northern Durango, and perhaps in the moun- tains of northeastern Sinaloa. Crotahis pricei pricei Van Denburgh Crotalus pricei Van Denburgh, 189.5, Proc. California Acad. Sci., Ser. 2, 5:856. Crotalus triseriatus Amaral, 1927, Bull. Antivenin Inst. Amer. 1:52. Crotalus pricei pricei Smith, 1946, Univ. Kansas Sci. Bull. 31(1)(3):79. 3 mi N Chuhuichupa, 6 (BYU 15773-5, 15389, 1.3796 and 13888). 4 mi SW Chuhuichupa, 1 (BYU 15415). 16 mi SW Creel, 1 (BYU 14470). 12 mi W San Juanito, 1 (BYU 17082). 22.5 mi S of Creel on La Bufa Road, 1 (BYU 16951). Gloyd (1940) lists specimens for the follow- ing localities: Casas Grandes, Colonia Garcia, Guadalupe y Calvo, Galeana, 24 km (15 mi) N of Chuhuichupa, Guerrero, Miiiaca, Pacheco, Rio Piedras Verdes, Samachique, and San Bias Mountains. This small, spotted species is seemingly common along the rock\' crests of the river- banks south and north of Chuhuichupa. Aside from those taken, several were seen disap- pearing in the porous rock formations. The scalation is all within the limits of variation set forth by Glo\d (1940) and Mc- Cranie (1980). For the 10 specimens, the ven- trals were 150(158.4)162 in males and 155(159.4)161 in females, caudals in males 24(26)30 and 19(22.2)24 in females. In 20 supralabial counts, 12 were 9, 6 were 10, and 2 were 8; the infralabials were 13 with 10, 5 and 9, and 2 with 11. The color and color patterns do not varv from the description given In Gloyd (1940). Two specimens (BYU 15773-4), from 5 km (3 mi) N of Chuhuichupa, taken 27 August 1957, were gravid. The embryos were fully formed and numbered 5 in each clutch. October 1985 Tanner: Snakes of Western Chihuahua 663 Crotalus lepidus klauheri Gloyd Caudisona Icpida Kennicott, 1861, Proc. Acad. Nat. Sci., Philadelphia 13:206. Crotalus lepidus Cope, 1883, Proc. Acad. Nat. Sci., Philadelphia 35:13. Crotalus lepidus klauheri Gloyd, 1936b, Occ. Papers, Mus. Zool. Univ. Michigan 337:2. 2 mi W Colonia Juarez, 1 (BYU 13416). 5.5-6.5 mi NE Colonia Juarez, 3 (BYU 14247, 15281-2). Pacheco, 1 (BYU 33614). 2 mi E Cerocouhui, 1 (BYU 14244). 25.5 mi SE Creel (on road to La Bufo), 1 and 6 mature embryos (BYU 17114 and 37241-6). Majorachic, 1 (EHT-HMS 23014). On road to Sierra del Nido, 1 (UTEP 2570). 5 km S Yepomera, 1 (UTEP 2541). This report includes 14 specimens seen from Chihuahua, 9 of which were taken in the mountains of western Chihuahua. Because there are considerable variations in color and color pattern between those taken in the mountains and those from the foothills on the eastern front, and because the systematics of Crotalus semicornutus is not fully resolved, the data on variation in this series may be of interest. All specimens were taken west of Highway 45, with those from the foothills west of Nuevas Casas Grandes and near Colonia Juarez having the color and pattern of typical C. /. klauheri (Klauber 1956:56, Fig. 2:281). The montane specimens from Pacheco and south, to near the Barranca del Cobre, are darker in color with secondary spots and nu- merous flecks between the primary dark cross bars. Most are surprisingly similar in the body pattern to that figured by Klauber (1956:67, Fig. 2:27) for C. /. lepidus, except that the parietal-nape spot is single and large, and the crossbars are large and dark as in /. klauheri . An adult female taken 40.8 km (25.5 mi) SE of Creel has the nape spot single, but with di- vided posterior extensions and the anterior- cross bars showing some dorsal enlargement. A clutch of six nearly mature young and one infertile egg were removed from her; all color patterns are mature and resemble the adult. A specimen from 3 km (2 mi) E of Cero- couhui (approximately 120 km or 75 mi SW of Creel) is, except for the darker color, a typical /. klauheri. We failed to secure specimens from the Majorachic-Maguarichic area. Those taken from south of the type locality of Cro- talus semicornutus Taylor (1944) do not show the pattern oi semicornutus, retaining the ba- sic pattern oi' klauheri, with the strong sec- ondary spots and flecking of/, lepidus. A comparison of the scalation shows great similarity. Ventrals 158-170; males 159-163.3- 170 and females 158-161-169. Caudals 19-28; males 19-25-28 and females 19-21.6-25. Scale rows are consistently 23 at midbody, and the head scales are consistent except that the lore- als may vary from 1 to 5. When these varia- tions are compared to the averages of the larger series reported by Gloyd (1940) and Klauber (1956), no difference is apparent. The number of transverse body blotches is also approximately the same, at 17.5. When our data and those presented by Gloyd (1940) and Klauber (1956) are com- pared to the characters reported for Crotalus .semicornutus Taylor, it is seemingly apparent that C. semicornutus is a pattern variant of C. /. klauheri Gloyd. The large vertebral spots in the type o{ semicornutus and the increase in pigmentation and secondary spotting be- tween the body blotches of other montane specimens does suggest that the population in the mountains of western Chihuahua does have color pattern deviations from that of typi- cal C. /. klauheri, as was observed in speci- mens from near Colonia Juarez. Except that the specimen taken south of Creel has a single nape spot, it could pass as a C. /. lepidus. The strong lepidus influence in the mountain populations from southwestern Chihuahua to Jalisco is now apparent and was suggested by Klauber (1952). Furthermore, C. /. maculosus (Tanner, Dixon, and Harris 1972), with its dark pigmentation and divided nape spot, is apparently more easily derived from lepidus than from klauheri . As indicated above, there appears to be a strong color and pattern influence of /. lepidus remaining in the mountain populations of western Chihuahua. The presence of sec- ondary spotting, the dark pigmentation both dorsal and ventral, and apparently the geo- graphic isolation (separated from /. lepidus by typical /. klauheri along the eastern foothills of the Sierra Madre) may have led Taylor to describe C. semicornutus. One may speculate that C. /. klauheri is a recent, vigorous form now extending its range southward and replacing C. /. lepidus, or that perhaps the influence of lepidus has been ex- 664 Great Basin Naturalist Vol. 45, No. 4 tended northward from the area of intergrada- tion, now occurring apparently in the south- ern areas of their distribution. It may also be possible that lepidus is less easily displaced in the mountains than along the foothills, and thus the residual color pattern. The evidence of a retention of lepidus characters, particu- larly color and color pattern, is best exhibited in C. /. maculosiis. The narrow crossbars, often broken on the sides in lepidus, the di- vided nape spot, and the dark pigmentation can be related to lepidus but not to klauberi. The present data give strong evidence that lepidus was present in the western mountains of Chihuahua before klauberi arrived and seemingly is slowly being replaced by a more vigorous form. Or is it possible that the color patterns and coloration is influenced by the environment and that the scalation in the sub- species lepidus and/or klauberi is more mean- ingful in determining taxa? Vincent (1982) ex- amined a large series of two populations from western Texas and observed that significant differences in ground color and color patterns occurred. He concluded that these color and pattern differences resulted from the domi- nant substrate colors that differ between ranges. His conclusion appears true of the series from Chihuahua, with no real differ- ence in scalation and with color and color pattern showing localized variation that places, for the present, all C. lepidus speci- mens from Chihuahua in the subspecies klauberi. Thus, Crotalus semicornutus Tay- lor is apparently only a color morph within the subspecies C. /. klauberi. All our specimens have been taken on rocky hillsides and usually from under rocks. On one occasion a specimen (BYU 15281) was disturbed while out foraging; at least it was disturbed while not under cover, and imme- diately it began rattling and moved rapidly down the hill, parting the grass as it moved (12 September 1959). I have observed many rat- tlesnakes in the open, but never before one with such speed and agility. Crotalus scutulatus scutulatus Kennicott Caudisona scutulata Kfiiiiicott, 1861, Proc. Acad. Nat. Sci., Philadelphia, p. 207. Crotalus .scutulatus Klauher, 19.30, Trans. San Diego Soc. Nat. ni.st. 6:1-17. Crotalus scutulatus scutulatus Price, 1982, Cat. Amer. Amph. and Kept., p. 291. I mi W Sueco, 1 (BYU 19133). 9 mi E Ricardo Flores Magon, 1 (BYU 17109). 4 mi E Ricardo Flores Magon, 1 (BYU 153.51). 4 mi E Buenaventura, 1 (BYU 1.5314). 24.5 mi W Sueco, 1 (BYU 15313). 23 mi E Buenaventura, 1 (BYU 15261). 11.5 mi N Colonia Duhlan, 2 (BYU 13872 and 15479). 6.8 mi SE Nueva Casas Grandes, 1 (BYU 153.50). II mi SE Nueva Casas Grandes, 3 (BYU 13871, 15349 and 15371). 25 mi N Ciudad Chihuahua, 1 (BYU 21717). 28 mi S Sueco, 1 (BYU 15344). 38 mi N Ciudad Chihuahua, 1 (BYU 15296). 5-6 mi N Ciudad Chihuahua, 7 (BYU 15295, 1.5297, 15320-1, 15.345, 16987 and 17113). 17 mi N Ciudad Chihuahua, 1 (BYU 17108). 10 mi W San Francisco del Oro, 1 (BYU 15678). 13.5 mi N Jimenez, 1 (BYU 14074). Records provided by Dr. Robert Webb (UTEP) list six specimens taken along High- way 45 from south of Villa Ahumada to 32 km (20 mi) W of Jimenez. Gloyd (1940) and Klauber (1956) show distribution by map shading rather than by locality. Distribution based on the above localities establishes C. s. .scutulatus in the low, rolling hills and desert valleys extending east from the mountains. Price (1982) lists localities that are apparently in the montane areas of west- ern Chihuahua. 1 did not find them in the western mountains. We found C. .scutulatus on the rocky hill- sides and C. atrox in the brushy valleys; thus, these two species occur widely in the low desert ranges and valleys, but there is seem- ingly a difference in the habitat niche that each occupies. The only other species we have taken in this general habitat area is C. m. molossus . The single specimen taken in a rocky arroyo, 17 km (11 mi) W of R. F. Magon, does not suggest that molo.s.sus represents an important competitor for the more common species C. afro.v and C. scutulatus. Color pat- tern and scalation do not \ary from those listed bv Glovd and Klaulier. Crotalus atro.x Baird and Cirard Crotalus atrox Baird and Cirard, 1853, Cat. N. Amer. Rcpt., p. 5. 20 mi S Palomas, 1 (BYU 13869). Sueco, 1 (BYU 15338). 15 mi W Sueco, 1(BYU 17107). 17.7 mi E Ricardo Flores Magon, 1 (BYl' 1.53.56). 12 mi E ButMuu ciitura, 1 (BYU 1,5299). October 1985 TANNER: Snakes of Western Chihuahua 665 9 mi SE Galeana (N of" Buenaventura), 2 (BYU 13870 and 15342). 14 mi N Colonia Dublan, 1 (BYU 13866). 5 mi SE Nueva Casas Grandes, 2 (BYU 13867-8). 11 mi SE Nueva Ca.sas Grande.s, 2 (BYU 15261 and 15278). 7 mi SE Nueva Ca.sas Grandes, 1 (BYU 15322). 6 mi N Chihuahua City, 3 (BYU 15294, 15298, and 15319). Neither Gloyd (1940) nor Klauber (1952) list locality records. Records provided by Dr. Robert Webb show five localities from 20 mi S of Ciudad Juarez along Highway 45 to 10 km (6 mi) N of El Sneco. All other records available indicate that C. atrox occurs in the desert valleys and foothills of northern and eastern Chihuahua. All locality records listed above are below 1800 m (6000 ft), and range from approximately 1300 to 1700 m (4500 to 5500 ft). We found them to be common along the roads and in the valleys between El Sueco and Colonia Juarez. A few badly mashed ones were seen along the road to Chihuahua City. In spite of our many trips into the mountains of western Chihuahua, none have been found above 1800 m (6000 ft). If our records are indicative of the range of this species in Chi- huahua, then the species does not occur over much of western Chihuahua, as was indicated by the range map of Klauber (1956, Fig. 2:1). The wide range of this species does extend across the lower plainlands from Arkansas to California, but it does not apparently include the high elevations of the various mountain ranges in western Chihuahua, preferring to- live (to quote Klauber 1952) "in dry — even arid — country, such as brush-covered plains, dry washes, sandstone outcrops, or mesquite crowned dunes." This is a good description of the terrain south of Ciudad Juarez on both sides of Highway 45 to southern Chihuahua, and westward in the low valleys to the moun- tain foothills. We soon discovered that C. atrox was tem- peramental and, at times, aggressive. On 20 August 1957, while riding horses in a brushy pasture about 8 km (5 mi) SE of Nueva Casas Grandes, a medium-sized (805 mm) C. atrox sprang full length, barely missing the hind legs of the horse just in front of me. That same evening, while road running, a smaller one sprang at me as I approached it. Other species either took a defensive position or moved away to cover. We soon noted whether we were dealing with atrox or scutulatus, the latter in our experience being much less ag- gressive. The color pattern and scalation con- form closely to those published previously by Gloyd (1940) and Klauber (1930 and 1952). Crotalus viridis viridis (Rafinesque) Crotalus viridis Rafinesque, 1818, Amer. Month. Mag. Grit. Rev., 4:41. Crotalus viridis viridis Klauber, 1936, Trans. San Diego, Soc. Nat. Hist. 8:194. El Espia, reported by Smith and Taylor, 1945, 1 (USNM 264. See Klauber 1952, 26:103). 16 mi E of Chihuahua-Sonora border on the Bavispe to Nueva Casas Grandes Road, 1 (CAS-SU 14361). Our collecting south of Ciudad Juarez, Las Palomas, and Antelope Wells (across the bor- der south of Lordsburg) has not provided a specimen of viridis . The California Academy specimen does establish an authentic record verifying the earlier USNM specimen taken in the 1850s. The full extent of the range of viridis in Chihuahua must yet be established. For the present, it is expected to be found only in the northwestern corner of Chi- huahua, perhaps from the vicinity of Las Palo- mas westward in the foothills and into ex- treme northeastern Sonora. Crotalus wiUardi amabilis Anderson Crotalus willardi amabilis Anderson, 1962, Copeia 1:160-163. The type and 10 paratypes, all from the Sierra del Nido, are deposited in the Museum of Vertebrate Zoology (68895-68900, 71015-71016 and 66177, 68894 and 68893). This series was seen, but not examined. The Sierra del Nido are east of the Sierra Madre Occidental and extend from an area slightly southwest of Gallego to just northwest of Ciu- dad Chihuahua. The principal drainage is from the western slopes into the basin of the Rio Santa Clara, which flows north to termi- nate in the desert basin near Villa Ahumada. This range has an elevation above 2400 m (8000 ft), and thus provides for isolation from the western highlands. The fieldwork of Dr. James D. Anderson provides material that is not only indicative of isolation but also of some species not yet known to occur in the western mountains. 666 Great Basin Naturalist Vol. 45, No. 4 Crotalus willardi silus Klauber Crotalus willardi Meek, 1905, Fid. Col. Mus. Pub. 104, Zool. Ser. 7(1):18. Crotalus willardi silus Klauber, 1949, Trans. San Diego Soc. Nat. Hist. 11(8):128. Type locality on Rio Gavilan, 7 mi SW of Pacheco (MVZ 46694). Red Rock, Tinaja Canyon, 12 mi W highway be- tween Casas Grandes and Colonia Juarez, 4 (BYU 13843-6). Upper fork of Nutria Creek (tributary of the Rio Bavispe near Chihuahua-Sonora border), 1 (BYU 13487). On rim trail appro.ximately halfway between Urique and Cerocouhui, I (BYU 14.596). Chuhuichupa, 1 (BYU 15480). 1 mi W Chuhuichupa, 3 (BYU 1.5720, 1.5722-3). 5 mi W Colonia Garcia, 1 (BYU 15388). Gloyd (1940) lists specimens for the follow- ing localities: Colonia Garcia, Dist. Guerrero, Rio Piedras Verdes (head of canyon). Sierra Madre, Tamarino, and Majorachic. Klauber (1949) lists the following new localities: Rio Gavilan 11 km (or 7 mi SW of Pacheco) and halfway between Majuarachic and Las Varas. The most southern locality previously re- ported was Majorachic. The specimen (BYU 14596) taken near the canyon rim west of Urique is the most southern C. w. silus avail- able to me. Its color pattern, and particularly that of the head, does not vary from those seen from northern localities. Two specimens (BYU 15480 and 15722) each had recently eaten a small rodent, and two (BYU 15388 and 15720) had large oviducal eggs, 7 and 6, respectively. Scale and color patterns do not vary appre- ciably from published reports by Gloyd (1940) and Klauber (1972). The only significant varia- tion occurs in the caudals, which average higher than those listed by Klauber (1972, Table 2:7): caudals, males \5), 31-32-35; fe- males (6), 27-29.7-34. Harris and Simmons (1976) list the subspe- cies C. w. ohsciirus as occurring in the Sierra de San Luis of Northwestern Chihuahua. This is based on a single specimen (UA 27943) from the west slope of that range. We did not col- lect in these mountains and assume the speci- men to represent an addition to the fauna of Chihuahua. In their report is a lengthy and useful discussion of the habitats within the range of the species C. willardi and their un- derstanding of the isolation factors that have resulted in the subspeciation within the spe- cies. A table of meristic characters for the willardi subspecies is provided, as well as a suggested phylogeny. Acknowledgments 1 am indebted to a number of individuals and families who assisted during the fieldwork conducted in various parts of Chihuahua. We were accepted not only by the American colonists, but also by the Mexican and Indian individuals with whom we came in close con- tact upon mmierous occasions. At Colonia Juarez we were fortunate in hav- ing an opportunity to stay at the homes of Mr. and Mrs. David Johnson and Mr. and Mrs. Irvin Romney. Mr. George Turley was alsc , very helpful, but perhaps those most under- i standing of our aims in the gathering of mate- I rial were the Hatch brothers, Herman, Roy. and Seville. We were particularly grateful tc the Herman Hatch family, who not only per- mitted us to bed down under the old apple tree at any hour of the day or night, but of fered us their hospitality and served as infor- mation agents for our travels in much of north- ern and central Chihuahua. Our first trip into the mountains was witl the Colonia Juarez Scout troop to the Rk Bavispe (just below Three Rivers), and the next year with Amilio Borgous to Chu huichupa. These trips introduced us to th( mountains and prepared us, so we thought for the barrancas of southwestern Chihuahua In Colonia Dublan Mr. Alma Jarvis, wh( was the postmaster, provided us with xaluabh information concerning areas for which he wa well informed, and Mr. and Mrs. Keith Bow man upon a niunber of occasions provided u: with meals and a place to stay. In Giudad Chihuahua, we were fortiuiate ii becoming acquainted with Mr. Harold Pratt who was the Chihuahua agent for the Allis Chalmers E(iuipment Company. Mr. anc Mrs. Pratt opened their door to us, provide( an opportimity for us to recoup after haxini been in the mountains for a time, and als( gave us an opportunity to reassemble our col- lections in preparation for the trip home. Also, it was from their estate that we were ablt to spend collecting time in the Giudad Chi- huahua area. \Vv wcmc also fortunate to hav( met a friend, Mr. Ra\' Thane, in San Franciscd October 1985 TANNER: Snakes ok Western Chihuahua 667 del Oro, who provided us with intorniation and an opportiniity to visit with a Mexican family. During the years spent in Mexico, I had the good fortune of having as companions a num- ber of capable faculty and graduate students. The first trip (1956) involved Mr. Verle All- man, a biology teacher, and my son Lynn. For the next four years I had as my companion Dr. Gerald W. Robison. We were accompanied upon one occasion (1958) by Dr. and Mrs. Irving W. Knobloch, he a professor of botany from Michigan State University, and upon another occasion with Dr. Stephen L. Wood, an entomologist from BYU, and his graduate student Dr. Jay B. Karran. The trip to Urique with Dr. Knobloch was a highlight, as was the trip with Dr. Wood to Maguarichic. Each trip added to our species list and seemed to compel us to plan the next trip. In October 1963 I was a member of the John Cross expedition into the Barranca del Cobre. Although we could not run the river as planned, we did get considerable publicity in both Chihuahua and U.S. newspapers and had the opportunity to secure additional ma- terial and data. Mr. John Cross is an accom- plished adventurer and river runner, having been interested in commercial river expedi- tions during much of his life. His interest in the rivers of southwestern Chihuahua was thus not only a part of his vocation but also an adventure for him into a new river system. Mr. Cross made at least two additional trips into the barrancas of southwest Chihuahua: one down the Rio Urique, from just above the Divisadero to Urique (Figs. 8 and 9A-F), and one from the Rio Verde, south of Guachochic into the Rio San Miguel and to the junction of the Rio Urique. Although herpetology was [not their prime interest, a number of new ! records for Chihuahua were obtained and de- posited in the BYU collection. During the next few years 1 had as my companions either Dr. Glen T. Moore, a botany professor at BYU, or Dr. Kenneth R. Larson, a graduate j student at that time. Upon occasion we solicited the aid of some of the Mexican people, particularly when we were short-handed in the mountain areas. We found them to be very cooperative, and at no time were we ever made to feel unwelcome. For example, when we were at Creel the La Fig. 8. Map of the area traversed by J. L. Cross during his river-running expeditions in southwest Chihuahua. Bufa mining superintendent helped us to se- cure the necessary fuel to complete our trip into the southeastern barranca area. At Cero- couhui we parked our truck and all of our contents by the side of the Catholic church, where it remained for over a week totally unmolested. We found that the Mexican peo- ple in the mountains were not only friendly, but very trustworthy. Specimen Materials Except for a few exchanges made with other museums and university collections, the ma- terials gathered during the 15 years of collect- ing are deposited in the herpetological collec- tion at Brigham Young University. We have sent a few specimens to the University of Mexico, and have, in all of our collecting trips, tried to observe the collecting regulations set forth by the Mexican government. During our collecting it was not our intent to take large samples of any one species at any one locality, but we were interested primarily in determin- ing, as far as possible, the species and subspe- cies that actually occur within the state of Chihuahua. For the loan of comparative specimens, I am indebted to the following individuals and institutions: Dr. Charles C. Carpenter, Uni- versity of Oklahoma (UO); Dr. William G. Degenhardt, Dr. Roger Conant, and Mr. Lee 668 Great Basin Naturalist Vol. 45, No. 4 Fig. 9. A series of photographs from southwestern Chihuahua showing the rugged terrain in and around the Barranca del Cob re and the Rio Urique. A. Looking northwest across the Barranca del Cohre toward the Divisadero. Tarahuniara dwelling is in the foreground. Fitzgerald, Museum of Southwestern Biol- ogy, University of New Mexieo (NMMZ); Dr. William E. Duellman and Joseph T. Collins, Museum of Natural History, University of Kansas (KU); Dr. M. J. Fouquette, Jr., Ari- zona State University (ASU); Dr. Arnold G. Kluge, University of Michigan, Museum of Zoology (UMMZ); Dr. Alan Leviton and Dr. Robert C. Drewes, California Academv of Sci- ences (CAS and CAS-SU); Dr. Charles H. Lowe, University of Arizona (UAZ); Dr. Roy W. McDiarmid, U.S. National Museum (USNM); Dr. Douglas A. Rossman, Museum of Zoologv, Louisiana State University (LSUMZ); Dr. David B. Wake, Museum o'f Vertebrate Zoologv, Universitv of California (MVZ); Dr. Robert G. Webb,' University of Texas at El Paso (UTEP); and Monte L. Bean Life Science Museum, Brigham Young Uni- versity (BYU). I greatly appreciate information provided by Dr. Richard D. Worthingtcm, Dr. Roger Conant, Dr. Robert Webb, and also, for his cooperation and the loan of Arizona and Chi- huahua specimens. Dr. C. H. Lowe. I am indebted to Mr. John L. Cross for his kindness in permitting me to accompany him on one of his trips in southwestern Chihuahua (October 1963) and for the use of his field journals from his other trips into the Rio Uricjue Barranca and the valle\' of the Rio San Miguel. From his interest in our project, a number of genera {Leptodeira, Leptophis, Oxyheli.'i, and Rliadinaea) were added to the fauna. Other specimens extended ranges and helped in a better understanding of system- atic relationships. I was also the recipient of his entire herpetological collection, without which this stud\' would have been greatb wanting for material from southwestern Chi- huahua. The drawings arc b\ Mrs. Diane Mellor, library and reference work was done by Mrs. Jody (Chandler, and the manuscript was typed by Mrs. C'olleen Ta\ lor. 'l\) each I am indeed grateful for their help. Tannek: Snakes oe Western Chihuahua B. Looking north from south rim of Barranca del Cobre. The manuscript was reviewed by Dr. Ho- bart M. Smith and Dr. Carl S. Lieb. Literature Cited Amaral, a. D. 1927. Studies of Nearctic Ophidia. II. Crotalus pricei Van Denburgh, 1896, a synonym of C. triseriattis (Wagler, 1830). Bull. Antivenin Inst. Amer. 1:48-54. Anderson, J. D. 1960. Storeria storerioides in western Mexico. Herpetologica 16:63-66. 1962. A new subspecies of the ridge-nosed rat- tlesnake, Crotalus willardi, from Chihuahua, Mexico. Copeia 1962(1);160-163. Bailey, J. R. 1940. The Mexican snakes of the genus Rhad- inaea. Occ. Pap. Mus. Zool. Univ. Michigan 412:1-19. Baird, S. F. 1859. Reptiles of the boundary. Rep. U.S. Mex. Bound. Surv. 3(2): 1-35. Baird, S. F., and C. Girard. 1853. Catalogue of North American reptiles in the museum of the Smithso- nian Institution. Part I. Serpents. Washington: Smithsonian Inst., xvi, 172 pp. Banicki, L. H.. and R. G. Webb. 1982. Morphological variation of the Texas lyre snake {Trimorphodon bisctitatus vilkinsoni) from the Franklin Mountains, west Texas, Southwestern Nat. 27(3):321-324. Bell, T. 1825. On Leptophina, a group of serpents com- prising the genus Drijinus of Merrem, and a newly formed genus proposed to be named Leptophis. Zool. J. 2(7):322-329. Blanchard, F. N. 1920. Three new snakes of the genus Lampropeltis . Occ. Pap. Mus. Zool. Univ. Michi- gan 81:1-10. 1921. A revision of the kingsnakes: genus Lampro- peltis. U.S. Nat. Mus. Bull. 114:1-260. Blaney, R. M. 1977. Systematics of the common king snake Lampropeltis getulus (Linnaeus). Tulane Studies in Zool. and Bot. 19(3-4):84. BOCOURT, F. 1881. Description d'un ophidien oporterodonte, appartenant au genre Catodon. Bull. Soc. Philom. Paris (ser. 7) 5:81-82. Boettger, O. 1883. Herpetologische Mittheilungen. I. Kurze Notizen iiber Reptilien und Amphibien in der Heidelberger Universitiits-Sammlung. Ber. Offenbacher Ver. Nat. 22-23:147-156. Bogert, C. M. 19.39. A study of the genus Salvadora , the patch-nosed snake. Publ. Univ. California at Los Angeles Biol. Sci. 1:177-236. Bogert, C. M., and J. A. Oliver. 1945. A preliminary analysis of the herpetofauna of Sonora. Bull. Amer. Mus. Nat. Hist. 83(6):297-426. Brown, A. E 1901. A new species oiColuher from west- ern Texas. Proc. Acad. Nat. Sci. Philadelphia .53:492-495. Great Basin Naturalist C. Just below the Divisadero looking north up the Rio Urique. Part way up the slope and looking down at the boats in the river below. October 1985 TANNER: Snakes of Western Chihuahua 671 D. Rio Urique just below Divisadero, looking downstream. 672 Great H \si\' \ \turalist Vol. 45, No. 4 E. Photographs showing our attempt to lun the Rio Uiique Top photos showing poitage aiound two waterfalls Bottom photos showing the boulder-filled channel. Cole. C J., and L. M. Hahdy 1981. Systematics of North American colubrid snakes related to Tantilla plan- iceps (Blainville). Bull. Amer. Mus. Nat. Hist. 171(.3);199-2S4. 198.3. Tantilla hohartstnithi Taylor. Cat. Anier. Amph. andRept. 318:1-2. CoNANT, R. 1963. Seniiaquatic snakes of the genus Thamnophis from the isolated drainage system of the Rio Nazas and the adjacent areas in Mexico. Copeia 1963 (3):473-499. Cope, E. D. 1860 (1861). Catalogue of the Coluhridae in the Museum of the Academy of Natural Sciences in Philadelphia. Proc. Acad. Nat. Sci. Philadel- phia 12:241-266. 1865 (1866). Contribution to the hcrpctolog)- of tropical America. Proc. .\cad. Nat. Sci. Phiiaticl- phia 17:185-198. 1866 (1867). On the Reptilia and Batrachia of tiic Sonoran province of the Nearctic region. Proc. Acad. Nat. Sci. Philadelphia 18:300-314. 1869 (1870). Seventh contribution to the herpetol- ogv of tropical America. Proc. Amer. Phil. Soc. Ih 147- 169. 1875. In Yarrow, Wheek-rs Rept. (Jeogr, (ieol. Expl. Surv. W. 100th Mer, 5;.544, 1879. Eleventh contribution to the hcrpetology of tropical America. Proc. .Amer. I'hilos. Soc. 18:261-277. 1883. A new snake from New Mexico. .Amer. Nat. 17:1.300-1301. 1883. Notes on the geographical distribution ol Batrachia and Reptilia in western North America. Proc. Acad. Nat. Sci. Philadelphia 35:10-35. 1886. Thirteenth contribution to the herpetolog>- Phi Soc. from New Mexico. lizards, and snakes U.S. Nat. Mus. for of tropica! America. 1 23:271-287. 1896. On a new CAau( Amer. Nat. 30:753. 1900(1898). The crocodilians of North America. .\nn. Rept 1898:153-1270. Dixox, J R . M Sabbath, and R. Worthincton. 1962. Comments on snakes from central and western Mexico. Ilerpetologica 18(2):91-100. Di.xoN. J. R., AND R. R. Fleet. 1976. Arizona Kennicott. Cat. Amer. Amph. and Rept. 179:1-4. Dowi.iNC, II. C. 1952. A ta.xononiic stud\ of the rat- snakes, genus t7«/>/i£' Fitzinger. 1\ . .\ check list ol the .\nierican forms. Occ. Paji. Mus. Zool. Univ. Michigan 541:1-12. Di ELLNL\N, \V. E. 1958. A inonograplii colubrid snake genus Leptodcira Mus. Nat. Hist. 114:1-152. 1961. The amphibians and rcptiU's of Michoacan, Mexico. Univ. Kan.sas Publ. Mus. Nat. Hist. 15:1-148. Di'NN. E. R. 1936. Notes on North American Leptodcira. Proc. Nat. Acad. Sci. 22:289-298. Fncil II S 1948. Further remarks conccrniiiu rli(iiiii}opliis ordinoiclcs and its relati\fs. (lopci.i 1948:121-126. studv of the] Hull. Amer. October 1985 Tanner; Snakes of Western Chihuahua 673 F. Continuing our attempt to the end, October 1963. .. 1965. An ecological study of the garter snake, Thamnophis sirtalis. Univ. Kansas Publ. Mus. Nat. Hist. 15(10):493-564. .. 1980. Remarks concerning certain western garter snakes of the Thamnophis elegans complex. Trans. Kansas Acad. Sci. 83(3): 106- 113. 1983. Thamnophis elegans (Baird and Girard). Cat. Am. Amph. and Kept. 320:1-4. Fitch. H S , andT. P. Maslin. 1961. Occurrence of the garter snake, Thamnophis sirtalis, in the Great Plains and Rockv Mountains. Univ. Kansas Publ. Mus. Nat. Hist. "l3(5):289-308. 674 Great Basin Naturalist Vol. 45, No. 4 Frost. D. R. 1983. Sonora semiannulata Baird and Girard. Cat. Amer. Amph. and Rept. 333:1-4. Carman. S. W. 1883. The reptiles and batrachians of North America. Part I. Serpents. Mem. Mus. Comp. Zool. 8(3):i-xxiv, 1-185 pp. Cehlbach.F. R. 1965. Herpetology of the Zuni Mountain region, northwestern New Mexico. Proc. U.S. Nat. Mus. 116:243-332. 1971. Lyre snakes of the Trimorphodonbiscutatus complex: A taxonomic resume. Herpetologica 27(2):200-211. Cloyd. H. K. 1936a. A Mexican subspecies of Crotahis mo/os.su.s Baird and Cirard. Occ. Pap. Uni\ . Mich- igan Mus. Zool. 325:1-5. 1936b. The subspecies oiCrotalus lepidus. Occ. Pap. Univ. Michigan Mus. Zool. 337:1-6. 1940. The rattlesnakes, genera Sistrurus and Cro- taltis. A study in zoogeography and evolution. Chicago Acad. Sci. Special Publ. 4:i-viii. 1-70 pp. Goldman. E. A 1951. Biological investigations in Mexico. Smithsonian Misc. Collns., Washington, D. C. 115. xii + 476pp. Grobman, a. B. 1941. A contribution to the knowledge of variation in Opheodrys vernalis (Harlan), with the description of a new subspecies. Misc. Publ., Mus. Zool. Univ. Michigan 50:7-37. GUNTHER, A. C. 1858. Catalogue of colubrine snakes in the collection of the British Museum. Taylor & Francis, London, xvi, 281 pp. 1872. Seventh account of new species of snakes in the collection of the British Museum. Ann. Mag. Nat. Hist. (4)9:13-37. 1885-1902. Biologia Centrali- Americana. Rep- tiles and Batrachia. R. H. Porter and Dulan and Co., London, .xx + 326pp. Hahn, D E. 1979a. Leptotyphlops humilis (Baird and Girard). Cat. Amer. Amph. and Rept. 232:1-4. 1979b. Leptotyphlops dulcis (Baird and Girard). Cat. Amer. Amph. and Rept. 321:1-2. Hallovvell, E. 1852. Description of new species of rep- tiles inhabiting North America. Proc. Acad. Nat. Sci. Philadelphia 6:177-182. Hardy, L. M. 1976. Gyalopion Cope. Cat. Amer. Amph. and Rept. 182:1-4. Hardy, L M , and R W McDiarmid. 1969. The amphib- ians and reptiles of Sinaloa, Mexico. Univ. Kansas Publ., Mus. Nat. Hist. 18(3):39-252. Harris, H. S., Jr., and R. S. Simmons. 1976. The paleo- geography and evolution of Crotalus willardi, with a formal description of a new subspecies from New Mexico, United States. Bull. Maryland Her- petological Soc. 12(l):l-22. Hensley, M. M 1966. A new subspecies of the Mexican snake St/mpholis Ii])i)icn.s Cope. Hi-rpetologica 22:48-55. Hovey, E. O. 1907. A geological reconnaissance in the western Sierra Madre of the state of CJhihuahua, Mexico. Bull. Amer. Mus. Nat. Hist. 23:401-443. Jan, G. 1863. Elenco sistematico degli ofidi descritti e disegnati per I'iconografiia generale. Milan, Lom- bardi. vii f 143 pp. Johnson. J D 1977. The taxonomy and distribution of the Neotropical whipsnake Masticophis nwntovarius (Reptilia, Serpentes, Colubridae). J. Herpetol. ll(3):287-309. 1982. Masticophis mcntov a hus (Dumeril, Bibron, and Dumeril). Cat. Amer. Amph. and Rept. 295:1-4. Keiser, E. D. 1970. Oxybelis WAg\er. Bull. 297 U.S. Nat. Mus., Catalogue Neotropical Squamata: Part 1. Snakes-Peters and Orejas-Miranda, p. 227. 1982. Oxybelis aeneus (Wagler). Cat. Amer. Amph. and Rept. 305:1-4. Kennicott, R 1860. Descriptions of North American serpents in the museum of the Smithsonian Insti- tution, Washington, D.C. Proc. Acad. Nat. Sci. Philadelphia 12:328-338. 1861. On three new forms of rattlesnakes. Proc. Acad. Nat. Sci. Philadelphia 13:204-208. Klauber, L. M 1930. Differential characteristics of southwestern rattlesnakes allied to Crotalus atrox. Bull. Zool. Soc. San Diego 6:1-72. 1936. A key to the rattlesnakes with summary of characteristics. Trans. San Diego Soc. Nat. Hist. 8(20): 185-276. 1940a. The worm snakes of the genus Leptoty- phlops in the Lhiited States and northern Mexico. Trans. San Diego Soc. Nat. Hist. 9:87-162. 1940b. The lyre snakes (genus Trimorphodon) of the United States. Trans. San Diego Soc. Nat. Hist. 9(9): 163- 194. 1941. The long-nosed snakes of the genus Rhi- nocheiliis. Trans. San Diego Soc. Nat. Hist. 9(29): 289-332. 1946. The glossy snake, Arizona, with descrip- tions of new subspecies. Trans. San Diego Soc. Nat. Hist. 10(17):311-393. 1947. Classification and ranges of the gopher snakes of the genus Pituophis in the western United States. Bull. Zool. Soc. San Diego 22:1-81. 1949. The subspecies of the ridge-nosed rat- tlesnake, Crotalus willardi. Trans. San Diego Soc. Nat. Hist. 11(8): 121- 140. 1952. Taxonomic studies of the rattlesnakes of mainland Mexico. Bull. Zool. Soc. San Diego 26:1-143. 1956. Rattlesnakes: their habits, life histories, and influence on mankind. Univ. California Press, Berkeley and Los Angeles. 2 vols. 1476 pp. 1972. Rattlesnakes: their habits, life histories, and influence on mankind. 2d ed. Univ. California Press, Berkeley and Los Angeles. 2 vols, xxx + 1533 pp. Knobloch. I. W., AND D S CoRRELL 1962. Ferns and fern allies of Chihuahua, Mexico. Texas Research Foundation, Renner, Te.xas. xiv -I- 198 pp. Ll'.ci.h.H J M 1959. A new snake of the genus Gcopliis from (Chihuahua, Mexico. Uni\ . Kansas Publ. Mus. Nat. Hist. ll(4):327-334. Leopold, A S. 1950. Vegetation zones of Mexico. Ecol- ogy 31:507-518. Le\iton. a. E.. AND B H Banta I960. Description of a new species of snake of the colubrid genus Gyalop- ion from the state of Zacatecas, Mexico. Occ. Pap. C:alifornia Acad. Sci. 26:1-4. October 1985 Tanner: Snakes oe Western Chihuahua 675 Liner, E H. 1983. Tantilla wilcoxi Stejneger. Cat. Anier. Aniph. and Rept. 345:1-2. Lowe, C H 1955. Generic .status of the aquatic snake Thamnophis anfiustirostris . Copeia 1955 (4): 307-309. McCranie, J. R. 1980. Drymarchon Fitzinger. Cat. Anier. Amph. and Rept. 267:1-4. McDiARMiD, R W , AND N J ScoTT. 1970. Geographic variation and systematic status of Mexican lyre snakes of the Trimorphodon tau group (Colubri- dae). Contr. Sci. Los Angeles County Mus. 179:1-43. Medica. p. a. 1975. Rhhwcheilus Baird and Girard. Cat. Anier. Amph. and Rept. 175:1-4. Meek, S. E. 1905. An annotated list of a collection of reptiles from southern California and northern lower California. Field Mus. Publ., Zool. Ser. 7:1-19. Mertens, R. 1934. Die insel Reptilian, ihre Ausbreitung, Variation and Artbildung. Zoologica (Stuttgart) 32:1-209. Mertens, R . and H G Dowling. 1952. The identity of the snake Pitijophi.s intermedins Boettger, 1883. Senckenbergiana Biol. 33(4/6): 197-201. MORAFKA, D. J. 1977. A biogeographical analysis of the Chihuahuan Desert through its herpetofauna. W. Junk B. V. , Publ. , The Hague, vii + 313 pp. Murphy, R. W 1975. Two new blind snakes (Serpentes Leptotyphlopidae) from Baja California, Mexico, with a contribution to the biogeography of penin- sular and insular herpetofauna. Proc. California Acad. Sci. 60(5):93-107. Myers, C . W 1974. The systematics oiRhadinaea (Colu- bridae), a genus of new world snakes. Bull. Amer. Mus. Nat. Hist. 153(1): 1-262. Oliver, J. A. 1948. The relationships and zoogeography of the genus Thalerophis Oliver. Bull. Amer. Mus. Nat. Hist. 92:157-280. Ortenburger, a. I. 1928. The whip snakes and racers, genera Masticophis and Coluber. Memoirs Univ. Michigan Mus. 1:1-247. Parker, W. S. 1982. Masticophis taeniatus (Hallowell). Cat. Amer. Amph. and Rept. 304:1-4. Peters, W. C. H 1864. Ueber einige neue Saugethiere. Mbr. dt. Akad. Wiss., Berlin, pp. 381-399. Price, A. H 1980. Crotalus molossus Baird and Girard. Cat. Amer. Amph. and Rept. 242:1-2. 1982. Crotalus scutulatus {Kennicoii). Cat. Amer. Amph. and Rept. 291:1-2. Rafinesque, C S 1818. Further accounts of discoveries in natural history, made during a journey through the western regions of the United States. Amer. Month. Mag. Grit. Rev. 4:41-47. Reynolds, R P , and N. J. Scott, Jr 1977. Use of a mam- malian resource by a Chihuahuan snake commu- nity. U.S. Dept. of the Interior, Wildlife Research Rept. 13, Washington, D.C., 1982. Rossman, D. A. 1971. Systematics of the neotropical pop- ulations of Thamnophis marcianus (Serpentes: Colubridae). Occ. Pap. Mus. Zool., Louisiana State Mus. 41:1-13. RoTHWELL, R p., R. W Raymond, and F Hobart 1901. The engineering and mining journal. Engineering and Mining Journal, Inc., New York, vol. Ixxii, xxiv + 872 pp. RozK, J, A. 1974. Micruroides Schmidt. Cat. Amer. Amph. and Rept. 163:1-4. 1983. New world coral snakes (Elapidae): a taxo- nomic and biological summarv. Mem. Inst. Bu- tantan, 46:305-338. Rutiiven, a. G. 1908. Variation and genetic relationships ofthe garter snakes. U.S. Nat. Mus. Bull. 61:xxi + 201 pp. Schmidt, K P. 1933. Preliminary account of the coral snakes of Central America and Mexico. Field Mus. Nat. Hist., Zool. Ser. 20:29-40. 1940. Notes on Texas snakes of the genus Sal- vadora. Field Mus. Nat. Hist., Zool. Ser. 24(12): 143- 150. Schmidt, K P and T F. Smith. 1944. Amphibians and reptiles of the Big Bend region of Texas. Field Mus. Nat. Hist., Zool. Ser. 29(5):75-96. Scott, N. Jr , and R. W McDiakmid. 1984a. Trimor- phodon biscutatus (Dumeril, Bibron, and Dumeril), western lyre snake. Cat. Amer. Amph. and Rept. 353:1-4. ' 1984b. Trimorphodon tau Cope, Mexican lyre snake. Cat. Amer. Amph. and Rept. 354:1-2. Shaw, G 1802. Coluber flagellum. Gen. Zool. or Svstm. Nat. Hist., 3(2):615. London. Shreve, F. 1939. Observations on the vegetation of Chi- huahua. Madrono 5(1):1-13. 1944. Rainfall of northern Mexico. Ecology 25:105-111. Smith, H M 1938. Additions to the herpetofauna of Mex- ico. Copeia 3:149-150. 1939. Notes on Mexican reptiles and amphibians. Field Mus. Nat. Hist., Zool. Ser. 24:15-35. 1941a. A review of the subspecies of the indigo snake (Drymarchon corals). Jour. Washington Acad. Sci. 31(11):466-481. 1941b. Notes on the snake genus Trimorphodon . Proc. U.S. Nat. Mus. 91(3130):149-168. 1941c. Notes on Mexican snakes of the genus Masticophis. J. Washington Acad. Sci., 31(9): 388-398. 1942a. Descriptions ofnew species and subspecies of Mexican snakes ofthe genus Rhadinaea. Proc. Biol. Soc. Washington 55:185-192. 1942b. A resume of Mexican snakes ofthe genus Tantilla. Zoologica 27:33-42. 1942c. The synonymy of the garter snakes (Thamnophis), with notes on Mexican and Central American species. Zoologica 27:97— 123. 1943. Summary ofthe collections of snakes and crocodilians made in Mexico under the Walter Rathbone Bacon Traveling Scholarship. Proc. U.S. Nat. Mus. 93(3169):393-504. 1946. Preliminary notes and speculations on the triseriatus group of rattlesnakes in Mexico. Univ. Kansas Sci. Bull. 31, 1(3):75-101. 1951. The identity ofthe ophidian name Coluber eques Reuss. Copeia 2:138-140. 1955. Effect of preservatives upon pattern in a Mexican garter snake. Herpetologica 11(3): 165-168. Smith, H. M andE. Brodie. 1982. Salvadoradeserticola. Page 194 in A golden field guide. Western Pub- lishing Co. 676 Great Basin Naturalist Vol. 45, No. 4 Smith, H. M and J P Kennedy. 1951. Pituophis nielanoleucus riithveni in eastern Texas, and its bearing on the status of P. catenifer. Herpetolog- ica 7(3):93-96. Smith, H M , and M B. Mittleman. 1943. Notes on the Mansfield Museum's Mexican reptiles col- lected by Wilkinson. Trans. Kansas Acad. Sci. 46:243-249. 1944. A brief biographical note on Edward Wilkin- son. Amer. Midi. Nat. 30(3):803-805. Smith, H. M., and W. W. Tanner. 1944. Description of a new snake from Mexico. Copeia 3:131-136. Smith, H. M., and E. H. Taylor. 1945. An annotated checklist and key to the snakes of Mexico. Bull. U.S. Nat. Mus. 187:1-239. Smith, H M., K. L. William.s, and E O Moll 1963. Her- petological explorations on the Rio Conchas, Chi- huahua, Mexico. Herpetologica 19(3):205-215. Stejneger, L. H. 1893. Annotated list of reptiles and batrachi- ans collected by the Death Valley expedition in 1891, with descriptions of a new species. North Amer. Fauna 7: 159-228. 1902. The reptiles of the Huachuca Mountains, Ari- zona. Proc. U.S. Nat. Mus. 25(1282):149-158. Stejneger, L. H., andT. Barbour 1917. A check list of North American amphibians and reptiles. Harvard Univ. Press, Cambridge, iv, 5-125 pp. Stickel, W. H. 1938. The snakes of the genus Sorwra in the United States and Lower California. Copeia 4:182-190. 1943. The Mexican snakes of the genera Sonora and Chionactis with notes on the status of other colubrid genera. Proc. Biol. Soc. Washington 56:109-128. Tanner, W. W 1950. Variation in the scale and color pattern of the wandering garter snake, in LItah and southern Idaho. Herpetologica 6: 194- 196. 1953. A study of the taxonomy and phylogeny oi'Lain- propeltis pijromelana Cope. Great Basin Nat, 13:47-66. 1957. A new skink of the multivirgatus group from Chihuahua. Gr. Basin Nat., 17 (3-4):59-94. 1959a. A new Thamnophis from western Chihuahua with notes on four other species. Herpetologica 15(4): 165- 172. 1961. A new subspecies oiConopsis nasus from Chi- huahua, Mexico. Herpetologica 17(1):13-18. 1983. Lampropeltis pijromelana (Cope). Cat. Amer. Amph. and Rept. 342:1-2, Tanner, W. W. , J R Dlxon, and H S Harris, Jr 1972. A new subspecies oiCrotahis lepidus from western Mexico, Great Basin Nat. 32(1): 14-22. Tanner, W. W , and C D. Jorgensen 1963, Reptiles of the Nevada Test Site. Brigham Young Univ. Sci. Bull., Biol. Ser. 3(2): 1-35. Tanner, W. W., and W. Gerald Roblson, Jr 1960, A collec- tion of herptiles from Urique, Chihuahua, Great Basin Nat. 19(4):75-82, Taylor, E. H 1936. A taxonomic study of the cosmopolitan scincoid lizards of the genus Eumeces with an account of the distribution and relationships of its species. Univ. Kansas Sci. Bull, 23:1-643. 1939a. Notes on the Mexican snakes of the genus Leptodeira, with a proposal of a new snake genus, Pseudoleptodeira . Univ. Kansas Sci. Bull. 25(15): 315-355. 1939b. On Mexican snakes of the genera Triinor- phodon and Hypsiglcna . Univ. Kansas Sci. Bull, 25(16):357-383. 1939c. On North American snakes of the genus Leptotypldops . Copeia 1:1-7. 1940, A new Lampropeltis from western Mexico, Copeia 4:253-255, 1941. Herpetological miscellanv. No. II. Univ. Kansas Sci. Bull. 27:105-139, 1944, Two new species of crotalid snakes from Mexico. Univ, Kansas Sci. Bull. 30:47-56. T.\YLOR, E. H., AND 1. W. Knobloch, 1940, Report on an herpetological collection from the Sierra Madre Mountains of Chihuahua. Proc, Biol, Soc. Wash- ington 53:125-1.30, Taylor, E H , and H. M. Smith. 1938, Miscellaneous notes on Mexican snakes. Univ. Kansas Sci. Bull. 15(13):249-251. 1942. The snake genera Conopsis and Toluca. Univ. Kansas Sci. Bull. 28(2)(15):325-363. Thompson, F. G 1957. A new Mexican garter snake (genus Thamnophis) with notes on related forms. Occ. Pap. Mus. Zool. Univ. Michigan 584:1-10. Trapido, H 1941. A new species oiNatrix from Texas. Amer. Midi. Nat. 25(3):678-680. VanDenburgh, J, 1895, Description ofa new rattlesnake (Crotalus pricei) from Arizona. Proc. California Acad, Sci, (2)5:856-857. Van Devender.T R , andC. H Lowe 1977. Amphibians and reptiles of Yepomera, Chihuahua, Mexico. J, Herpetology ll(l):41-50, Vincent, J W. 1982, Phenotypic variation in Crotahts lepidus lepidus (Kennicon). J. Herpetology 16(2):189-191. Wagler, J G. 1824. Dryinus aencus Wagler. Serpentum Brasiliensum Species Novas, p. 12. Webb. R G 1966. Resurrected names for Mexican popula- tions of black-nosed garter snakes, Thamnophis cyr- topsis (Kennicott). Tulane studies in Zool. 13:55-70. 1976. A review of the garter snake Thamnophis ele- gans in Mexico, Nat. Hist. Mus, of Los Angeles County Bull. 284:1-13. 1980. Thamnophis cyrtopsis (Kennicott). Cat. Amer. Amph. and Rept., p. 245:1-4. West Te.xas Geological Society. 1964, Field trip guide- book. San Angelo, Texas. Anchor Publishing Co. , 140 pp. 1974. Geologic Field Trip Guidebook. Electrical Log Service, Inc., Midland, Texas. 150pp. Williams, K L 1978. Systematics and natural history of the American milk snake, Lampropeltis triangulum. Mil- waukee Public Mus., Publ. in Biol, and Geol. 2:1-258, Wilson, L D 1973, Masticophis flagellum (Shaw), Cat. Amer. Amph. and Rept. 145:1-14. Worthington, R D 1980. Elaphe subocularis (Brown). Cat, Amer, Amph, and Rept. 268:1-2. Wright, A H., and S. C Bishop. 1915. Snakes sec. 2 in a biological reconnaissance of the Okefenokee Swamp in Georgia. Proc. Acad. Nat. Sci. Philadelphia 67: 192. Zweifel, R. G., and K S Norris, 1955, Contributions to the herpetology of Sonora, Mexico: descriptions of new subspecies of snakes (Micruroides euryxanthus and Lampropeltis getulu.s) and miscellaneous collecting notes. Amer. Midi, Nat, 54:230-249. ELECTROPHORETIC STUDY OF CUTTHROAT TROUT POPULATIONS IN UTAH Mark A. Martin'", Dennis K. Shiozawa', Eric J. Loudenslager', and J. Neil Jensen'' Abstract. — Thirty-nine Utah streams were sampled for cutthroat trout. Of these, 31 contain cutthroat or cutthroat/ rainhow hyhrid populations. By using starch gel electrophoresis, these populations were segregated into three groups. One group consisted predominately offish from the Sevier River (of the Bonneville Basin) and Colorado drainages. A second was primarily populations from the Bear River Drainage (Bonneville Basin) as well as some scattered populations along the Wasatch Front (Bonneville Basin). The third consisted of Wasatch Front populations and populations that have hybridized with rainbow trout. Since different subspecies of cutthroat trout are native to the Colorado and Bonneville drainages, one would expect the populations from within the Bonneville Basin to be more similar to one another and less similar to the Colorado River populations. That this did not occur raises questions concerning the evolutionary relationships of the subspecies and the populations. It is clear that at least a northern (Bear River) and southern (Sevier River) form of the Bonneville cutthroat exists. The Wasatch Front may represent an intermediate zone where these two forms intergrade. Salnio clarki, the cutthroat trout, had the most extensive continental distribution of the western North American native trout (Sahnonidae, Salmu). Behnke (1981) tenta- tively recognized 15 subspecies of cutthroat trout associated with three major phyletic groups: a coastal cutthroat trout, S. clarki clarki, characterized by 68 to 70 chromo- somes (Gold et al. 1977); an interior cutthroat trout, S. c. leivisi, native to the upper Colum- bia River, upper Missouri River, and the South Saskatchewan drainages, characterized by 66 chromosomes (Loudenslager and Thor- gaard 1979); and a group of subspecies derived from the Yellowstone cutthroat trout, S. c. botwieri, which inhabit the upper Snake River, Yellowstone River, the Great Basin, Colorado River, South Platte River, and Rio Grande drainages. These are characterized by 64 chromosomes (Loudenslager and Thor- gaard 1979). Utah's waters originally supported three cutthroat trout subspecies — the Yellowstone, S. c. bouvieri , the Colorado River, S. c. pleu- riticus, and the Bonneville, S. c. Utah. The Yellowstone cutthroat is native in the Raft River drainage of northwestern Utah but has now been introduced throughout Utah. The headwaters of the Colorado River Basin (the Green River) downstream to the Dirty Devil River, Utah, on the west and the San Juan drainage of Colorado, New Mexico, and Ari- zona on the east composed the original range of the Colorado River cutthroat (Fig. 1). This trout has been severely impacted by man and is now considered threatened (Miller 1972). The Bonneville Basin (Fig. 1), situated on the eastern edge of the Great Basin, represents the drainage l)asin of Pleistocene Lake Bon- neville. This basin comprises the original range of the Bonneville cutthroat trout, S. c. Utah . Until recently the Bonneville cutthroat was thought to be extinct or so hybridized with introduced trout that it was unrecogniz- able. However, Hickman (1978) located 15 relict populations in Utah, Nevada, and Wyo- ming, and a sizable sport fishery has now been developed on what may be a native population in Bear Lake at the Utah-Idaho border. The present distribution of cutthroat trout within the Bonneville Basin is restricted to isolated lakes and tributaries where suitable habitat remained following the desiccation of pluvial Lake Bonneville. Three morphologi- cally and ecologically differentiated groups of populations, associated with the Snake Valley region on the Nevada-Utah border, the Bear River drainage in Wyoming, Idaho, and Utah, and the central Bonneville Basin proper, are currently recognized (Hickman and DuflP 1978, Behnke 1981). In addition to the ecolog- ical and morphological differentiation of these Department ofZoology, Brighani Young University, Provo, Utah 84602. "Present address: Environmental Health Specialist, Utah County Health Department, Provo, Utah 84601. ^Present address: Department of Fisheries, College of Natural Resources, Humboldt State University. Areata, California 95521. 'Present address; Department of Zoology, Weber State College, Ogden, Utah 84408. 677 678 Great Basin Naturalist Vol. 45, No. 4 SNAKE RIVER DRAINAGE MISSOURI RIVER DRAINAGE Fig. 1. Major drainage basins in tlie Utah ar Bonneville during the Wisconsin glacial period. The cross-iiatched aiea lepresents the B()nne\ ille stage of Lake October 1985 Martin et al.; Utah Cutthroat Trout 679 population groups, there is evidence of ge- netic divergence. Klar and Stalnaker (1979) reported a distinctive LDH allele in the Snake Valley population group. Gall and Lou- denslager (1981), using 36 protein loci, com- pared three populations from the Bear River drainage and four populations from the Snake Valley with each other and representative S. j[ c. bouvieri, S. c. pleiiriticus , and S. c. hen- shawi . They reported little genetic differenti- ation within the Bear River or Snake Valley population groups but substantial differentia- tion between them. Moreover, the Bear River populations were more similar genetically to S. c. bouvieri, and the Snake Valley popula- tions were more similar to S. c. pleuriticus than the Bear River and Snake Valley groups were to each other. In this paper we present results of an elec- trophoretic analysis of Utah cutthroat trout populations from drainage systems not previ- ously surveyed, using the protein systems that distinguish Snake Valley and Bear River cuttthroat trout from each other and rainbow trout, S. gairdneri (Gall and Loudenslager ' 1981). The objectives were to evaluate the genetic relatedness of these populations and identify hybridization between native cut- throat and introduced rainbow trout. Methods Thirty-nine Utah streams located in the Wasatch-Cache, Uinta, Manti-La Sal and Fish Lake National Forests were examined (Fig. 2, Table 1). Both electrofishing and hook and line were used to collect fish. Eight streams lacked cutthroat trout populations. A total of 550 trout from the remaining 31 streams were examined. Fish were frozen in the field on dry ice and returned to Brigham Young University for processing. Following processing, specimens were preserved in for- malin and stored in 40% isopropyl alcohol. Tissue samples were homogenized in 0.25 M sucrose and centrifuged at 30,000 x g for 15 minutes. The resulting supernatant was ana- lyzed with horizontal starch-gel electrophore- sis. Four protein systems encoded by six loci were examined: tripeptide aminopeptidase (LGG; EC 3.4. 11.4) from muscle tissue, isoci- trate dehydrogenase (IDH-3,4; EC 1.1.1.42) from liver tissue, malic enzyme (ME; EC 1. 1. 1.40) from liver tissue, and sorbitol dehy- drogenase (SDH-1,2; EC 1.1.1.14) from liver tissue (Gall and Loudenslager 1981). Loci are designated using the nomenclature of Allendorf and Utter (1978). An abbreviation that corresponds to the name of a protein designates each locus. Multiple forms of a protein are designated with the least anodally migrating locus as - 1, the next -2, and so on. Allelic variants are designated according to the relative mobility of their products, with the most common allele in S. gairdneri desig- nated 100. Allelic frequencies were determined from the protein bands. A matrix of similarities be- tween populations based on Nei's genetic identity index (Nei 1972) was clustered with the NTSYS statistical package. The un- weighted pair-group method using arithmetic averages (UPGMA), cluster algorithm was used (Sneath and Sokal 1973). Results and Discussion Polymorphism was found in five of the six loci examined: GCP, lDH-3, ME, and SDH- 1,2. Allelic frequencies for these loci are given in Table 2. All of the polymorphisms have been previously described in cutthroat trout (Loudenslager and Gall, 1980; Gall and Lou- denslager, 1981). Evidence of hybridization with hatchery rainbow trout, Salmo gairdneri. — If parental species are monomorphic for different alleles at a locus, or are polymorphic but share no alleles, then that locus can be used to distin- guish the parental species and their hybrids (Gall and Loudenslager 1981). Two loci, GCP and ME, examined in the present study can be used to distinguish cutthroat trout, rain- bow trout, and their hybrids. The GCP locus had two alleles, GCP (160) and GCP (100). The GCP (160) allele was previously reported to be: monomorphic in S. c. bouvieri, S. c. Utah, and S. c. pleuriticus and absent in S. gairdneri (Gall and Loudenslager 1981), whereas the GCP (100) allele is the common allele in hatchery S. gairdneri (Gall and Lou- denslager 1981). Similarly, the ME locus had two alleles, ME (125) and ME (100). ME (125) is monomorphic in S. c. bouvieri, S. c. Utah, and S. c. pleuriticus and absent in hatchery S. gairdneri, whereas ME (100) is the common 680 Great Basin Natuiulist GREAT SALT LAKE SEVIER LAKE KILOMETERS 0 50 100 Mi"i'l I I ' 0 50 MILES Fig. 2. Location of the 39 streams examined in this study. See Tahle 1 for the stream name and drainage basin October 1985 Martin etai..: Utah Cutthroat Trout 681 Table 1. Localities and number.s of trout collected. Sample Major Number of number Sample site Drainage drainage specimens 1. Kabell Creek Green River Colorado River 4 2. Thompson Creek Green River Colorado River 16 3. M. Fk. Beaver Creek Green River Colorado River 9 4. W. Fk. Beaver Creek Green River Colorado River 20 5. Joulious Creek Green River Colorado River 17 6. M. Fk. Blacks Creek Green River Colorado River 24 7. Brush Creek Green River Colorado River 22 8. McKenzie Creek Bear River Bonneville Basin 12 9. Mill Creek Bear River Bonneville Basin 22 10. Carter Creek Bear River Bonneville Basin 17 11. Boundary Creek Bear River Bonneville Basin 20 12. Meadow Creek Bear River Bonneville Basin 19 13. Moffit Creek Weber River Bonneville Basin 18 14. Sugarpine Creek Bear River Bonneville Basin 19 15. Bunchgrass Creek Logan River Bonneville Basin 19 16. Durfee Creek Ogden River Bonneville Basin 0 17. Greetsen Creek Ogden River Bonneville Basin 3 18. Red Pine Creek Weber River Bonneville Basin 18 19. N. Fk. Amer. Fk. River Utah Lake Bonneville Basin 5 20. Silver Creek Utah Lake Bonneville Basin 0 21. L. Fk. Hobble Creek Utah Lake Bonneville Basin 21 22. Strawberry River Green River Colorado River 60 23. Shinglemill Creek Spanish Fork Bonneville Basin 16 24. Chase Creek Spanish Fork Bonneville Basin 4 25. Fifth Water Creek Spanish Fork Bonneville Basin 11 26. Indian Creek Spanish Fork Bonneville Basin 0 27. Wanrhodes Creek Spanish Fork Bonneville Basin 11 28. Little Diamond Creek Spanish Fork Bonneville Basin 17 29. Tie Fork Creek Spanish Fork Bonneville Basin 0 30. Holman Creek Spanish Fork Bonneville Basin 27 31. Nebo Creek Spanish Fork Bonneville Basin 23 32. Mendenhall Creek Utah Lake Bonneville Basin 0 33. North Creek Utah Lake Bonneville Basin 0 34. Bear Canvon Creek Utah Lake Bonneville Basin 0 35. Willow Creek Utah Lake Bonneville Basin 0 36. Muddy Creek Dirty Devil River Colorado River 5 37. Deep Creek Sevier River Bonneville Basin 16 38. Hy Hunt Creek Sevier River Bonneville Basin 25 39. N. Fk. North Creek Sevier River Bonneville Basin 30 lallele in hatchery S. gairdneri. Indivickials representative of the parental species will be homozygous for their respective diagnostic al- leles, Fj hybrids will be heterozygous for both iloci, and Fj or backcross individuals will pos- isess a mixture of heterozygous and ho- mozygous diagnostic loci. Evidence for hy- bridization cannot be based on allele frequencies alone but requires classification of individuals based on a composite biochemical phenotype. This is because composite pheno- Itypes could indicate the presence of both parental species without hybridization. Of the Utah cutthroat trout populations sampled, seven were found that had an appar- ent introgression of rainbow trout alleles: Thompson, Mill, Boundary, Bunchgrass, Wanrhodes, Nebo, and Hy-Hunt Creeks. Us- ing the composite enzyme phenotype, no sample included rainbow trout, Salmo gairdneri . Genetic differentiation and relationships among Utah cutthroat trout popidations. — An inspection of allelic frequencies (Table 2) indicates that the SDH-1 locus is primarily responsible for differences among Utah cut- throat trout populations (after hybridization with rainbow trout is considered). Cutthroat trout populations in the Colorado River drainage are dichotomous for SDH-1 allele frequencies. Middle Fork Beaver, West Fork Beaver, Joulious, Middle Fork Blacks, and 682 Great Basin Naturalist Vol. 45, No. Table 2. Allelic frequencies of 6 loci for 31 trout populations. Streams Stream number M. Fk. W. Fk. M. Fk. Locus Kabell Thompson Beaver Beaver Joulious Blacks Brush McKenzie 1 2 3 4 5 6 7 8 SDH-1 100 40 0.63 0.05 0.02 0.18 0.12 0.02 1.00 0 0.37 1.00 0.95 0.98 0.82 0.88 0.98 SDH-2 250 100 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 IDH-3 170 100 1.00 1.00 1.00 1.00 1.00 l.OU 1.00 1.00 60 IDH-4 140 1.00 1.00 1.00 1.00 1.00 1.00 1,00 1.00 LGG 160 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 100 ME 125 1.00 0.97 1.00 1.00 1.00 1.00 1.00 1,00 100 0.03 nti SDH-1 100 40 0,55 1,00 1.00 1.00 0.03 1.00 1.00 0.17 0 0.45 0.97 0.83 SDH-2 2.50 0.13 100 1.00 1.00 1.00 0,87 1,00 1,00 1,00 1,00 IDH-3 170 0.11 0.03 0.08 100 1.00 1.00 0.89 0.97 0,92 1,00 0,97 1.00 60 0.03 IDH-4 140 1.00 1.00 1.00 1,00 1,00 1,00 1,00 1,00 LGG 160 0,95 1,00 0,68 1.00 1.00 1.00 1,00 1,00 100 0,05 0,32 ME 125 1.00 1,00 0.75 1.00 1.00 1.00 0,97 1,00 100 0.25 0.03 Table 2 continued. Streams Stream number Locus Mill 9 Carter 10 Boundarv 11 Meadow Moffit 12 13 Sugarpine 14 Bunchgrass 15 Greetsen 17 Brush creeks have a high frequency of the SDH-1 (0) allele (x - 0.922), whereas Kabell Creek, Strawberry River, and Muddy Creek have intermediate frequencies of the SDH-1 (0) allele (x = 0.49). Gall and Loudenslager (1981) sampled S. c. pleuriticus from two loca- tions in Wyoming and found the populations monomorphic for the SDH-1 (0) allele. The intermediate frequency of SDH-1 (40) in Ka- bell Creek, Strawberry River, and Mudd> Creek could be due to natural selection, ge- netic drift, or hybridization with stocked cut- throat trout. Since Yellowstone cutthroal trout, S. c. bouvieri, are monomorphic for SDH-1 (40) (Loudenslager and Gall 1980), hy- bridization is a probable cause. The Straw- berry River is also a major source of cutthroat eggs for stocking operations throughout the October 1985 Table 2 continued. Martin et al.: Utah Cutthroat Trout 683 Streams Stream number Red N. Fk. L. Fk. Fifth Locus Pine Am. Fk. Hobble Strawberry Shinglemill Chase Water Wanrhodes 18 19 21 22 23 24 25 27 SDH-1 100 I 40 0.25 0.25 1.00 0.50 0.97 1.00 0.50 0.36 0 0.75 0.75 0.50 0.03 0.50 0.64 SDH-2 250 100 1.00 1.00 1.00 1.00 lDH-3 170 0.01 100 1.00 1.00 1.00 0.99 60 1.00 1.00 1.00 1.00 1.00 1.00 1.00 0.95 0.05 IDH-4 140 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1 LGG 160 100 1.00 1.00 1.00 1.00 1.00 1.00 1.00 0.91 0.09 ME 125 100 1.00 1.00 1.00 1.00 1.00 1.00 1.00 0.91 0.09 Table 2 continued. Streams Stream numbers Locus Little Diamond 28 Holman 30 Nebo 31 Muddv 36 Deep 37 Hv Hunt 38 N. Fk. North 39 SDH-1 100 40 0 0.35 0.65 0.85 0.15 0.54 0.46 0.40 0.60 1.00 0.20 0.80 1.00 SDH-2 250 100 1.00 1.00 1.00 1.00 1.00 1.00 1.00 IDH-3 170 100 60 1.00 1.00 1.00 1.00 1.00 0.04 0.96 1.00 lDH-4 140 1.00 1.00 1.00 1.00 1.00 1.00 1.00 LGG 160 100 1.00 1.00 0.87 0.13 1.00 1.00 0.80 0.20 1.00 ME 125 100 1.00 1.00 0.91 0.09 1.00 1.00 0.86 0.14 1.00 state of Utah. The stocking of fish from this population could change allele frequencies in native populations. Within the Bonneville Basin, cutthroat trout populations were sampled from the Bear River drainage, along the Wasatch Front (We- ber and Utah lake drainage), and the Sevier River drainage (Fig. 1). The four Bear River drainage populations not influenced by rain- bow trout hybridization were monomorphic for the SDH-1 (40) allele. This supports previ- ous observations that Bear River drainage cut- throat trout were monomorphic for SDH-1 (40) (Gall and Loudenslager 1981). In con- trast, both the Deep Creek and North Fork of North Creek populations from the Sevier River drainage were monomorphic for the SDH-1 (0) allele. The SDH-1 allele frequen- 684 Great Basin Natl r.\list Vol. 45, No. 4 Table 3. Genetic identit\- and distance \ alues for painvise comparisons of the 31 trout populations sampled. Identity values are above the diagonal, and distance values are below the diagonal. Stream No. 10 11 12 13 Kabell Thompson M. Fk. Beaver \\'. Fk. Bea\ er Joulious M. Fk. Blacks Brush McKenzie Mill Carter Boundar>- Meadow Moffit Sugarpine Bunchgrass Greetsen Red Pine N. Fk. Am. Fk. L. Fk. Hobble Strawbern.' Shinglemill Chase Fifth \\ater Wanrhodes Lt.* Diamond Holman Nebo Muddy Deep Hv Hunt N.' Fk. North 1 — 2 .071 3 .060 4 .066 5 .037 6 .047 7 .066 .023 .002 .023 .060 .028 .067 .023 .024 .038 .026 .026 .023 .003 .020 .023 .003 27 .017 28 .014 30 .008 31 .006 36 .010 37 .071 38 .047 39 .071 .931 .001 .000 .006 .003 .000 .184 .0.54 .184 .235 .192 .001 .184 .186 .005 .011 .011 .184 .044 .173 .184 .044 .024 .021 .132 .055 .028 .000 .014 .000 .942 .936 .999 1.00 — 1.00 .000 — .003 .004 .001 .002 .000 .000 .164 .045 .164 .217 .172 .001 .164 .167 .002 .007 .007 .164 .036 .175 .050 .175 .229 .183 .001 .175 .177 .004 .009 .009 .175 .040 .154 .165 .164 .175 .036 .040 .019 .022 .015 .019 .116 .047 .021 .125 .052 .025 .000 .000 .012 .014 .000 .000 .964 .995 .997 .996 .001 .004 .122 .025 .122 .171 .129 .005 .122 .124 .000 .001 .001 .122 .018 .113 .122 .018 .008 .005 .082 .028 .009 .005 .010 .005 .9.54 .936 .998 1.00 .999 1.00 .998 1.00 .999 .996 .002 .141 .034 .141 .191 .148 .175 .050 .175 .229 .183 .003 .001 . 141 . 175 .143 .177 .000 .004 .003 .009 .003 .009 .141 .175 .026 .040 .131 .165 .141 .175 .026 .040 .012 .022 .009 .019 .097 .125 .036 .052 .014 .025 .002 .000 .010 .014 .002 .000 .832 .848 .839 .885 .869 .8.39 .035 .000 .029 .003 .175 .000 .000 .125 .101 .101 .000 .044 .000 .000 .044 .077 .075 .004 .040 .064 .182 .132 .182 .998 .977 .947 .832 .956 .848 .951 .839 .975 .885 .967 .869 .951 .839 .965 1.00 — .965 .029 .003 .175 .000 .035 .069 .041 .051 .035 .036 .000 .026 .125 .017 .107 .035 .001 .031 .035 .001 .101 .101 .000 .044 .000 .000 .044 .009 .077 .008 .075 .016 .004 .002 .040 .005 .064 .054 .182 .032 .132 .054 .182 .942 .790 .805 .796 .843 .826 .796 .972 .934 .972 .034 .227 .029 .027 .174 .148 .148 .029 .084 .029 .029 .084 .104 .119 .035 .058 .107 .237 .144 .237 .973 .936 .825 .999 .842 .999 .833 .999 .879 .995 .862 .998 .833 .997 .960 .997 .967 .183 .003 .003 .999 .840 .951 .840 .797 .833 .175 .177 .132 .005 .108 .010 .108 .010 .003 .175 .049 .041 .003 .164 .003 .175 .049 .041 .084 .022 .081 .019 .007 .125 .047 .053 .070 .025 .191 .001 .141 .014 .191 .001 cies for Wasatch Front populations were highly variable: MofFit Creek had the highest frequency of the SDH-1 (0) allele (0.97), wheras Chase Creek and the Left Fork Hob- ble Creek were monomorphic for SDH-1 (40). The remaining populations had SDH-1 (0) al- lele frequencies ranging from 0.03 to 0.83. A pattern in the SDH-1 allele frequencies is discernible if one includes Loudenslager and Gall's (1980) and Gall and Loudenslager's (1981a, b) findings of populations monomor- phic for the SDH-1 (0) allele in four popula- tions native to or derived from the Snake Val- ley area in western Utah. Populations inhabiting the south and west extremes of the Bonneville Basin are monomorphic for SDH- 1 (0), and those in the northeastern region are monomorphic for SDH-1 (40). A zone of in- tergradation in allele frequency occurs along the Wasatch Front. Genetic identity and distance were com- puted (Nei 1972) for all painvise comparisons of the 31 populations using the si.x loci sur- veyed (Table 3). The genetic identity index is an estimate of the proportion of sampled al- leles that are identical between paired popu- lations. Genetic distance is an estimate of the net codon differences and a measure of the accumulated allele differences per locus be- tween two populations. Genetic identity in pairwise comparisons of populations ranged from 1.00 in several comparisons to 0.826 be- tween Meadow Creek and North Fork ot North Creek. The average genetic identity tor pair-wise comparisons of Utah's cutthroat trout was 0.944. The genetic identit\' matrix was also used to calculate the mean genetic identity between groups of populations inhabiting different drainage s\'stems (Table 4). In this anahsis. populations thought to be hybridized with rainbow trout or Yellowstone cutthroat trout were excluded. Within the Bear River, Colo- rado River, and Sevier Ri\er drainages, ge- netic identity among localities was high: I 0.998; 0.998'; and 1.00. respectively. In con- October 1985 x\I.\RTINET.\L.: Ut.\h Cutthro.\t Trout 685 Table 3 con tinned 14 15 17 18 19 21 22 23 24 25 27 28 30 31 36 37 38 .955 39 .997 .976 .963 .974 .974 .977 .997 .981 .997 .997 .984 .986 .992 .995 .990 .9.32 .9.32 .832 .831 .995 .989 .989 .832 .957 .842 .832 .957 .977 .979 .876 .946 .973 1.00 .986 1.00 .848 .847 .998 .993 .993 .848 .965 .857 .848 .965 .982 .985 .890 .9.54 .979 1.00 .988 1.00 .839 .838 .996 .991 .991 .839 .961 .848 .839 .961 .979 .982 .882 .949 .976 1.00 .986 1.00 .885 .885 1.00 .999 .999 .885 .982 .893 .885 .982 .992 .995 .922 .973 .992 .995 .990 .995 .869 .867 1.00 .997 .997 .869 .975 .877 .869 .975 .988 .991 .908 .965 .986 .998 .990 .998 .839 .838 .996 .991 .991 .839 .961 .848 .839 .961 .979 .982 .882 .949 .976 1.00 .986 1.00 1.00 1.00 .883 .904 .904 1.00 .957 1.00 1.00 .957 .926 .928 .996 .961 .938 .833 .876 .833 .965 .964 .974 .983 .983 .965 .999 .970 .965 .999 .991 .992 .984 .998 .996 .948 .969 .948 1.00 1.00 .883 .904 .904 1.00 .957 1.00 1.00 .957 .926 .928 .996 .961 .938 .833 .876 .833 .972 .974 .840 .862 .862 .972 .920 .971 .972 .920 .902 .887 .965 .944 .899 .789 .866 .789 .997 .997 .876 .898 .898 .997 .953 .997 .997 .952 .919 .922 .993 .955 .9.33 .827 .869 .827 .840 .838 .996 .991 .991 .840 .960 .848 .840 .960 .978 .981 .882 .948 .975 .999 .986 .999 1.00 .883 .904 .904 1.00 .957 1.00 1.00 .957 .926 .928 .996 .961 .938 .833 .876 .833 .000 — .881 .902 .902 1.00 .956 1.00 1.00 .956 .926 .926 .996 .960 .937 .831 .876 .831 .125 .127 — .999 .999 .883 .981 .890 .883 .981 .991 .994 .919 .972 .991 .995 .990 .995 .101 .103 .001 — 1.00 .904 .989 .911 .904 .989 .995 .998 .937 .981 .996 .990 .990 .990 .101 .103 .001 .000 — .904 .989 .911 .904 .989 .995 .998 .937 .981 .996 .990 .990 .990 .000 .000 .125 .101 .101 — .957 1.00 1.00 .957 .926 .928 .996 .961 .938 .833 .876 .833 .044 .045 .019 .011 .011 .044 — .962 .957 1.00 .993 .996 .978 .996 .998 .957 .972 .957 .000 .000 .116 .094 .094 .000 .039 — 1.00 .962 .932 .934 .998 .965 .944 .842 .884 .842 .000 .000 .125 .101 .101 .000 .044 .000 — .957 .926 .928 .996 .961 .9.38 .833 .876 .833 .044 .045 .019 .011 .011 .044 .000 .039 .044 — .993 .996 .978 .996 .998 .957 .972 .957 .077 .077 .009 .005 .005 .077 .007 .071 .077 .007 .997 .954 .993 .997 .976 .992 .976 .075 .077 .006 .002 .002 .075 .004 .069 .075 .004 .003 — .956 .989 1.00 .980 .985 .980 .004 .004 .084 .065 .065 .004 .022 .002 .004 .022 .047 .045 — .979 .9W .877 .912 .877 .040 .041 .029 .019 .019 .040 .005 .036 .040 .004 .007 .011 .021 .992 .946 .976 .946 .064 .065 .009 .004 .004 .064 .002 .058 .064 .002 .003 .000 .036 .008 — .973 .982 .973 .182 .185 .005 .010 .010 .182 .044 .172 .182 .044 .024 .021 .131 .056 .027 .985 1.00 .132 .133 .010 .011 .011 .132 .028 .123 .132 .028 .008 .015 .092 .025 .018 .015 — .985 .182 .185 .005 .010 .010 .182 .044 .172 .182 .044 .024 .021 .131 .056 .027 .000 .015 — trast, genetic identity among localities along the Wasatch Front was onh 0.928. Pair-wise com- parisons of populations from different drainages indicated a high identit\ between the Sexier River drainage populations and Colorado Ri\er populations (I = 0.998). Identity between Bear River drainage populations and either Se\ier River drainage samples (I = 0.831) or Colorado River drainage samples (I = 0.855) was much lower. The Wasatch population group had mean identities of 0.940 uith the Bear River sites, 0.930 with the Se\ ier Ri\ er sites, and 0.941 with the Colorado River sites. These data are similar to those of Loudenslager and Gall (1980b). In addition, they demonstrated a genetic identity of 0.996 between the Bear River Bonne\ ille and Yellowstone cutthroat trout. The clustering of the genetic identit\ matri.x resulted in three distinct clusters (Fig. 3). Popu- lations in the first cluster were pohmorphic for the SDH-1 locus with intermediate frequencies of the (0) and (40) alleles. Included in this cluster were populations h\ bridized with rainbow and cutthroat populations from the zone of intergra- gradation along the \\'asatch Front. The second cluster contained populations from the Colorado Ri\er, Se\ ier Ri\ er, and \\'asatch Front with a high frequency of the SDH-1 (0) allele. The third cluster contained populations fi-om the Bear Ri\er drainage and Wasatch Front with a high frequency of the SDH-1 (40) allele. The similarit\ between the Colorado and Se\ ier Ri\ er Bonne\ille and between the Bear Ri\er Bonne\ille, and Yellowstone cutthroat strains could be due to common ancestr> (closely related) or it could be a result of con\ergence or drift. Howe\er, the dissimilarit) between the Bear Ri\ er and Sevier Ri\er forms of the Bon- neville cutthroat is definitive. That is, the occur- rence of different allelic frequencies must be due to di\ ergent histories of the populations. For example, the headwaters of Meadow Creek (Bear Ri\ er drainage) and Moffit Creek (Weber Ri\er drainage) are less than a kilometer apart, yet the cutthroat populations ha\e SDH-1 (0) frequencies of 0.00 and 0.97, respectively. Interpretation of the populations along the Wasatch Front is problematic. Urbanization in 686 Great Basin Naturalist Vol. 45, No. 4 0,876 0.896 0.916 0.936 0.956 0.976 0.996 Genetic Identity ■ 1 Kabell Creek . 9 Mill Creek .25 Fitth Water Creek .22 Strawberry River .31 Nebo Creek .27 Wanrhodes Creek .28 Little Diamond Creek -36 Muddy Creek . 2 Ttiompson Creek -37 Deep Creek -39 N. Fk. North Creek - 3 M. Fk. Beaver Creek . 4 W. Fk. Beaver Creek . 7 Brush Creek . 1 3 Moffit Creek - 5 Joulious Creek - 1 7 Greetsen Creek - 6 M. Fk. Blacks Creek - 1 8 Red Pine Creek - 19 N. Fk. American Fk. Riv« - 38 Hy Hunt Creek - 8 McKenzie Creek - 1 0 Carter Creek - 1 4 Sugar Pine Creek -21 L. Fk. Hobble Creek - 24 Chase Creek - 23 Shinglemill Creek - 1 5 Bunchgrass Creek - 30 Holman Creek - 1 2 Meadow Creek - 1 1 Boundary Creek 1.000 Fig. 3. Cluster dendrogram based on UPGMA clustering of the genetic identity matrix. Utah is concentrated along the Wasatch Table 4. Matrix of genetic identity among cutthroat trout populations from drainages within the Bonneville Basin and Colorado River. The number of sample loca- tions for each drainage is in parenthesis, and within drainage population identity is on the diagonal. L Bear R. (4) 2. Sevier R. (2) 3. Wasatch Front (10) 4. Colorado River (5) .831 .940 .855 .000 .930 .998 .928 .941 .998 Front. The stocking of nonnative trout has been intense in this area. Although we can reliably identify hxbridization with rainbow trout, we are unable to confidently assess hy- bridization with nonnati\ e cutthroats because of the close genetic relationship between na- tive Bonnexille Basin trout and cutthroats from contiguous basins. Whether these popu- October 1985 Martin etal.: Utah Cutthroat Trout 687 lations were originally polymorphic or mono- morphic for SDH-1 is unknown. Several populations in the Bonneville Basin near Utah Lake had high SDH-1 (40) frequen- cies. These fish are similar to the Yellowstone cutthroat trout and may have resulted from stocking. The highly polymorphic populations in the area are also likely to have been influ- enced by the activities of man. For instance, the Diamond Fork drainage (Bonneville basin) receives water diverted from the Straw- berry River (Colorado River) drainage. This would allow colonization by Yellowstone-Col- orado cutthroat from the Strawberry River into the Diamond Fork drainage and could influence allele frequencies. Because determining the original geo- graphical variation of the native Utah cut- throat is difficult, all streams that contain cut- throat trout that have not hybridized with rainbow should be given special management consideration. Such streams need not contain monomorphic populations since monomor- phism may represent only the extremes of the species variability of the subspecies. Polymor- phic populations may still represent the na- tive stocks as long as rainbow hybridization is not evident. This study has advanced our knowledge of the native cutthroat, but much remains to be investigated. One focal area should be the Wasatch Front, where the gra- dation between the northeastern and south- western Bonneville forms occurs. Another topic that warrants study is the identification of additional protein systems that separate the Yellowstone from the Bear River Bonneville form and the Snake Valley Bonneville form from the Colorado River cutthroat. These will be instrumental in understanding the taxo- nomic relationships and variability of the na- tive inland cutthroat trout. Acknowledgments We acknowledge Jack W. Sites (Brigham Young University), Boyd Bentley (University of California, Davis), and Eric Zurcher (Utah State University) for making their expertise available to this project. We are also indebted to Doug Sakaguchi, Shawn May, Dave Bur- toch, Allen Kimball, and Louis Billedeaux for their assistance in the field and with labora- tory work. Special thanks goes to Linda Mar- tin for her support throughout the study. We also acknowledge the biologists from the Utah Division of Wildlife Resources and the U.S. Forest Service, whose interest helped make this project possible. Literature Cited Allendorf, F W . AND F M Utter. 1978. Population genetics offish. Pages 407-454 in VV. .S. Hear and D. j. Randall, eds.. Fish physiology, .\cademic Press, New York. Vol. 8, 786 pp. Behnke, R. J 1981. Systematic and zoogeographical in- terpretation of" Great Basin trouts. Pages 2.5.5-262 in R. J, Naiman and D.L. Soltz eds.. Biology of fishes of the great American desert. John Wiley Pub. , New York. 552 pp. Gall, G A. E and E. J. Loudenslager 1981. Biochemi- cal genetics and systematics of Nevada trout popu- lations. Final Report to Nevada Dept. of Wildlife. 53 pp. Gold.J.R.J C.Avise.andG.A E.Gall. 1977. Chromo- some cytology in the cutthroat series Salmo clarki (Salmonidae). Cytologia 42:377-,382. Hickman, T J 1978. Systematic study of the native trout of the Bonneville Basin. Unpubhshed Masters of Science thesis. Colorado State University, Fort Collins. 122 pp. Hickman, T J , and D A. Duff 1978. Current status of cutthroat trout subspecies in the western Bon- neville basin. Great Basin Nat. 38: 193-202. Klar, G. T., and C. B. Stalnaker. 1979. Electrophoretic variation in muscle lactate dehydrogenase in Snake Valley cutthroat trout, Salmo clarki subsp. Comp. Biochem. Physiol. 64 B: 391-394. Loudenslager. E. J., and G. A. E. Gall. 1980a. Geo- graphic patterns of protein variation and subspeci- ation in cutthroat trout, Salmo clarki. Syst. Zool. 29: 27-42. 1980b. Biochemical systematics of the Bonneville Basin and Colorado River cutthroat. Final Report to the Wyoming Department of Fish and Game. 16 pp. Louden-SLAGER, E.J.andG. H.Thorgaard. 1979. Kary- otypic and evolutionary relationships of the Yel- lowstone {Salmo clarki bottvieri) and west-slope (S. c. lewisi) cutthroat trout. J. Fish. Res. Board. Canada 36: 630-635. Miller, R R. 1972. Classification of the native trouts of Arizona with the description of a new species, Salmo apache. Copeia 1972: 401-422. Nei, M. 1972. Genetic distance between population. Amer. Natur. 106: 283-292. Sneath, P H a , and R. R. Sokal. 1973. Numerical Tax- onomy. W. H. Freeman Co., San Francisco, Cali- fornia. 573 pp. SEXUAL SELECTION AND MATING SYSTEM VARIATION IN ANURAN AMPHIBIANS OF THE ARIZONA-SONORAN DESERT Brian K. Sulli\an' Abstract— Mating system variation in anuran amphibians of the Arizona-Sonoran Desert was reviewed. Male density and breeding period duration were negatively correlated in seven bufonids and pelobatids. Variation in male mating behavior and ability of females to freely select their mates unhindered by active-searching males also was related directly to male density. These observations support hypotheses relating ecological factors to mating system organization. It is suggested that male calling behavior, and anuran lek mating sytems in general, may be significantly influenced by predation on vocalizing males. The Arizona-Sonoran Desert boa.sts a surpris- ing diversity of frogs and toads: seven bu- fonids, three pelobatids, three ranids, two hylids, one microhyhd, and one lepto- dactyhd. As expected, the breeding biology of these anurans is significantly influenced by the xeric climate. Most of these species have short, or "explosive," (Wells 1977) breeding periods during the summer rainy season, al- though a few have relatively extended, or "prolonged" breeding periods during the spring (Sullivan 1982a, 1983a, 1984, Sullivan and Sullivan 1985). As suggested by Wells (1977), the mating behavior of anurans such as those of the Arizona-Sonoran Desert appears related to breeding period duration. For ex- ample, Woodhouse's toad {Biifo woodhousei) forms low-density breeding aggregations at sources of permanent water throughout the spring, and all males within these choruses call to attract females (Sullivan 1982a). In con- trast, the Great Plains toad (B. cog,natus) forms high density aggregations at temporary rain pools for only a few nights each summer. Within these short-lived choruses some males adopt satellite positions near vocalizing males and attempt to intercept females approaching the calling males (Sullivan 1982b). This diversity in breeding period duration and reproductive behavior provides an excel- lent opportunity for evaluating hypotheses of sexual selection and mating svstem theorv (Trivers 1972, Emlen and Oring 1977, Wells 1977). It is generally hypothesized that diua- tion of breeding period should directly influ- ence opportunities for female mate selection and, hence, male behavior. Wells (1977) ar- gued that for explosive-breeding anurans fe- male choice should be restricted as a result of: (1) temporal constraints limiting time avail- able for mate selection and (2) high male den- sities and corresponding alternative male mating tactics that typically reduce the ability of females to move freely in breeding aggrega- tions. Conversely, in species with prolonged breeding periods and lower male densities, females should be capable of freely selecting their mates and constitute a potentialK- signif- icant evolutionary force (e.g., Arnold 1983). The variation in breeding biology of anurans of Arizona allows direct test of these hypothe- ses. Here I summarize recent studies of sexual selection in pelobatids and bufonids of Ari- zona and, following Emlen and Oring (1977), document the existence of predictable rela- tionships between the ecology, mating sys-| terns, and reproductive behavior of these spe- cies. I also examine species recognition and acoustic competition among these anurans. Lastly, I discuss the evolution of anuran lek mating sytems. General Methods Methods employed in the studies reviewed here ha\ e been described in detail elsewhere (Sullivan 1983c, 1984, Sullixan and Sullivan 1985); a brief simimary follows. Observations at anuran breeding aggregations in southern Arizona were made with a six-\ olt headlamp Uepartini'iit of Zoolo 78712. sity, Tcnipc. Arizona H5'2H7 688 October 1985 SULLIVAN: SONORAN AMPHIBIANS 689 Table 1. Mean mule density (number of males) and brcedinj; period duration (days) for seven anurans (Bufo and Scap/^iopws) of southern Arizona (N ^ number of breeding aggregations). Densitv Duration Species X N X N S. multiplicatus S. bombifrons B. cognatus S. couchi B. debilis B. woodhousei B. punctatus 43.6 26.7 25.2 23.3 11.4 5.4 3.0 7 6 8 3 5 5 5 1.6 1.8 2.6 1.0 2,6 19.8 20.0 7 6 10 3 5 10 5 Sullivan and Sullivan, 1985 Sullivan and Sullivan, 1985 Sullivan, 1983c Sullivan and Sullivan, 1985 Sullivan, 1984 Sullivan, 1982a Sullivan, 1984 during the evening in the spring and summer from 1980 to 1984. Toads were captured by hand at breeding aggregations and individu- ally marked by toe-clipping. Snout-vent length was measured to the nearest mm with a plastic rule, and density of males at breeding aggregations was determined by direct count on each night of breeding activity. Male mat- ing success was defined as the number of fe- males a male was observed in amplexus with during a season since amplectant males were never displaced by single males (Sullivan 1983c, 1984, Sullivan and Sullivan 1985). Focal animal sampling techniques were used to observe behavior of individual males and females in breeding aggregations (Sulli- van 1983a, 1983b, 1984). Male advertisement calls were recorded in the field and analyzed in the lab with a Kay 6061B Sonagraph. De- termination of pulse rate and frequency of individual calls followed standard methodol- ogy as outlined in Sullivan (1982c). Density, Alternative Male Mating Tactics, and Female Choice There is considerable range in breeding pe- riod durations of Arizona anurans (Table 1). Mean male density at breeding aggregations is negatively correlated (t = -0.71, P = 0.01) with duration of breeding period for seven bufonids and pelobatids. All the species with high density aggregations utilize tempo- rary rain pools for breeding. Presumably, in these explosive-breeding forms the limited temporal availability of water restricts local populations to a pattern of more or less syn- chronous breeding. Wells (1977) postulated that temporal con- straints and high male densities should pro- mote the adoption of alternative mate-locat- ing tactics by males of explosive-breeding anurans. Sullivan (1982b) corroborated Wells' hypothesis: in breeding aggregations of B. co^i^natus the proportion of noncalling, satel- lite males "parasitizing" calling males is corre- lated with chorus size. These satellite males will begin vocalizing if male density de- creases. Similar satellite, as well as active- searching behavior, has been observed in two of the pelobatids of Arizona (Sullivan and Sul- livan 1985). It seems clear that density is the primary factor influencing male adoption of alternative tactics, especially in light of the intraspecific variation in B. cognatus. Fur- thermore, in those anurans that form low den- sity breeding aggregations no noncalling male tactics have been observed independent of breeding period duration (Sullivan 1982a, 1984). Other bufonids and pelobatids of North America also exhibit active-searching and satellite behaviors (Wells 1977). As expected, all these forms are characterized by explosive breeding periods and high male densities. Active-searching and satellite males have been observed successfully amplexing fe- males in high density breeding aggregations of two Arizona anurans, B. cognatus (Sullivan 1982b), and Scaphiopus multiplicatus (Sulli- van and Sullivan 1985). These observations support the notion that females are limited in their ability to freely select mates in high den- sity aggregations in part as a result of some males adopting noncalling, mate-locating tac- tics. Another line of evidence suggesting that mating is random with respect to male pheno- type in explosive-breeding anurans is the lack of relationship between male size and mating success among these forms (Table 2). Many investigators have suggested that female 690 Great Basin Naturalist Vol. 45, No. 4 Table 2. Relationships between male size and mating success Arizona. five anurans (Bufo and Scaphiopus) of southern Species Snout-vent length (mm) Mating males Nonmating males Positive assortative mating X N .X N P r N P Source S. coiichi S. multiplicatus B. cognatus B. debilis B. woodhousei 71.0 74.1 74.4 50.9 51.2 48.0 78.6 41.3 84.1 70.0 6 >.05* 71.1 18 >.05 69.2 18 >.05=* 51.3 28 >.05 50.0 15 >.05 48.0 11 >.05* 77.2 150 >.05 42.3 16 >.05* 85.4 43 >.05 0.07* 12 =.50 Sullivan and Sullivan, 1985 0.07* 20 =.43 -0.40* 10 =.24 0.05 21 >.05 Sullivan and Sullivan, 1985 0.09 24 >.05 Sullivan, 1983a 0.20* 6 >.05 Sullivan, 1984 0.23 47 >.05 Sullivan, 1983b *For small samples (< 10) a Mann-Whitney U test or Kendal's tau was calculated anurans should select large males as mates if size is an indicator of fitness (Wilbur et al. 1978, Fairchild 1981). Others have argued that females might benefit from selecting a male similar in size to themselves if fertiliza- tion success is related to the size difference between mating males and females (Licht 1976, Davies and Halliday 1979). Analysis of male size and mating success in four explo- sive-breeding anurans of Arizona has not sup- ported either of these hypotheses (Table 2.) Male size is positively correlated with mat- ing success in some explosive-breeding bu- fonids and pelobatids of the United States (Gatz 1981, Woodward 1982). However, in these forms large males might achieve greater mating success as a result of size advantages during direct struggles for possession of fe- males (Sullivan 1983a). Davies and Halliday (1979) and Lamb (1985) determined that large male bufonids are more successful in taking over females during male-male disputes. For these species a mating advantage for large males may be independent of any female pref- erence. Woodward (1982) documented a large-male mating advantage in one of four S. couchi and one of two S. multiplicatus breed- ing aggregations in central New Mexico. This interspecific variation suggests that a number of variables such as density and alternative mating tactics of males influence the outcome of sexual selection in nature. Male mating success in prolonged-breed- ing anurans is often correlated with the num- ber of nights a male participates in chorus activity (Kluge 1981, Woodward 1981, Ryan 1983). For prolonged-breeding B. woodhou- sei in Arizona male mating success is weakly correlated (R" = 0 — 25%) with participation in chorus activity (Fig. 1). However, this result is expected since many males are only ob- served at a chorus for one night: a correlation occurs because any male who mates two or three times must be present at the chorus for at least as many nights. Further, many males who are unattractive as mates (low call rate males) are active at choruses for many nights without obtaining any matings. Hence, for this anuran male persistance in chorus activity is not an important determinant of mating success. Female mate selection has been observed in breeding aggregations of the three Arizona bufonids with low-densitv choruses (Sullivan 1983a, 1984). In all of these forms, females initiate amplexus by making physical contact with a calling male and are able to move freely through breeding choruses to select their mates from among the calling males. In B. woodhousei male mating success is correlated with call rate, suggesting that females prefer high call rate males as mates (Sullivan 1983a). Males within a chorus retain a consistent, rel- ative ranking in call rate within and between nights, and there is significantly more varia- tion in call rate between than within males (Sullivan 1983a, Sullivan and Leek in re\iew). Females prefer speakers liroadcasting calls at the higher rate in simple two-speaker discrim- ination experiments, further supporting the female preference hypothesis. Proximately, female choice of high call rate males could be passive if they merely locate such males more readily, or active if genetically determined October 1985 SULLIVAN: SONOHAN AMPHIBIANS 691 1981-1 r= .37 P= .08 • • • • • • m (4) • m . m • • • • • 1981-2 r= .33 P= .09 1 5 1982-1 39)(9) (9) (3) (11) « • 10 r = .38 P= .001 15 20 1982-2 r= .38 • P= .01 • • • ®® • • ® (3) . • • . 1983- ■1 r = .49 3 P=.01 2 • • 1 • • w «• «• 0 (D- • i •• M • 1983-2 r= .03 P=.80 • • • ®®®(3)- - ® - 10 1 NUMBER OF NIGHTS 10 Fig. 1 . Male mating success (number of matings) against number of nights of participation in chorus activity for six ions oiBiifo woodhousei (number of concordant observations given inside circle). preferences exist (Parker 1982, Arak 1983). In B. woodhousei female behavior is consistent with the latter alternative since they often pass by low call rate males when moving through a chorus toward a high call rate male (Sullivan 1983a). However, the ultimate sig- nificance of these alternatives is indistinguish- able: high call rate males derive fitness bene- fits through increased mating success, and fe- males, although deriving no immediate bene- fits, may benefit via increased offspring survivorship or mating success, or both (Lande 1981, Kirkpatrick 1982, Arnold 1983). The importance of these benefits to females 692 Great Basin Naturalist Vol. 45, No. 4 would be determined by the heritability of male call rate and its relation to offspring sur- vivorship and mating success. In any event, mating is clearly nonrandom with respect to male phenotype and female behavior in B. woodhousei . These observations on male behavior and mate selection by females reveal that the pro- longed-breeding species {B. piinctatiis and B. ivoodhotisei), and at least some of the explo- sive-breeding forms (B. cognatiis and B. de- bilis), have lek-like mating systems (Sullivan 1983c). Females typically select their mates freely from among aggregations of displaying males, and males provide only sperm to their mates. Females mate disproportionately of- ten with males possessing high call rates in both B. cognatus and B. woodhousei (Sullivan 1983c); for the other species it is unknown whether females discriminate between con- specific males on the basis of male phenotype (Sullivan 1984). The explosive-breeding pelo- batids exhibit a typical male-dominance mat- ing system in which mate selection by females is somewhat limited (Emlen and Oring 1977). Males may obtain multiple matings within a breeding season if they participate in breed- ing aggregations following every rainstorm each summer (Sullivan 1983b). Although gen- erally consistent with the hypothesis of Emlen and Oring (1977) concerning the significance of breeding period duration to mating system structure, these observations indicate clearly that male density, through its impact on fe- male choice and regardless of breeding period length, influences sexual selection and the organization of the mating systems of these desert anurans. Species Recognition AND Acoustic Competition An important component of female mate selection concerns proper species recogni- tion. For anurans it is generally assumed that interspecific call differences allow sympatri- cally breeding species to avoid hvbridization (e.g., Hodl 1977, Drewry and Rand 1983, Duelhnan and Pyles 1983). For example, among hylid and bufonid forms, rate of ampli- tude modulation (pulse rate) of the male ad- vertisement call is apparently the primary call parameter separating closely related sym- patic species (Gerhardt 1982, Rose and Capranica 1983, Sullivan and Leek, in re- view). Discrimination experiments have re- vealed that female tree frogs (Hylidae) can discriminate between two acoustic stimuli dif- fering only in pulse rate, and they prefer stim- uli similar to advertisement calls of con- specific males (Loftus-Hills and Littlejohn 1971, Straughan 1975). All pelobatids and most bufonids occurring in southern Arizona have been observed in mixed-species breeding aggregations, provid- ing the potential for heterospecific interac- tions (Fig. 2). The ranges in pulse rate and dominant frequency of advertisement calls of these anurans are given in Figure 3. As ex- pected, advertisement calls of the closely re- lated forms, S. boynbifrons and S . mitltiplica- tiis, and B. mic rosea pJuis and B. woodhousei , are clearly dissimilar in pulse rate but not frequency. In the pelobatids, hybrid offspring are sterile; hence, there is presumably strong selection against females who select het- erospecific mates. The divergence in adver- tisement calls of these two species in south- eastern Arizona suggests that species recognition has been a significant selective force acting on female mate choice. Hy- bridization has also been documented be- tween the two bufonids, but nothing is known concerning the evolutionary significance of this interaction (Sullivan 1986b). Sympatri- cally breeding bufonids in Texas have also been found to differ in pulse rate and fre- quency (Blair 1956). Advertisement call varia- tion in two other Arizona anurans is also com- patible with the species recognition hypo- thesis: the closely related but allopatric bu- fonids B. debilis and B. retifonnis do not differ dramatically in pulse rate. Presumably, selec- tion has not favored divergence in advertise- ment call structure in the absence of an oppor- tunity for heterospecific interactions. Aspects of intra- and interspecific acoustic competition between anuran males have re- cently received considerable attention (re- viewed by Sullivan, 1985). A number of work- ers have argued that male anurans increase their locatability for selecting females by calling asynchronously with neighboring con- specifics as well as heterospecifics. In Arizona B. woodhousei follows this pattern: males re- duce their calling activity in response to play- October 1985 SULLIVAN: SONOIUN AMPHIBIANS 693 SYMPATRICALLY BREEDING ANURANS IN ARIZONA Sb Sc Sb - Sc . - Sm • • Ba • Be • Bd • Bm Bp • Br Bw Sm Ba Be Bd Bm Bp Br Bw • • • • • • Fig. 2. Anurans (Biifo and Scaphioptis) observed breeding synipatrieally (•) in southern Arizona (Sb = S. bombifrons , Sc = S. couchi , Sm = S. miiltipUcatus, Ba = B. alvaiius. Be = B. cognatus, Bd = B. debilis, Bm = B. microscaphus , Bp = B. piinctatus, Br =^ B. rctifonnis. Bw = B. icoodhousei). back of conspecific advertisement calls as well as a variety of synthetic stimuli (Sullivan 1985, Sullivan and Leek, in review). Proximately, males avoid calling during playback of any acoustic stimuli with a freqency close to the mean frequency of the species-typical call (Sullivan and Leek, in review). However, male B. piinctatus do not avoid overlap with broadcast of conspecific advertisement calls (Sullivan 1985). These bufonids possess rela- tively long calls (x = 7.0 sec) and appear to prefer to call synchronously with nearby calling males. Such interspecific variation suggests that male calling behavior is influ- enced by factors other than those associated with locatability for females alone. It may be that males of some species call synchronously to reduce the risk of predation; predators who locate anurans acoustically have difficulty cap- turing synchronously chorusing forms (Ryan and Tuttle 1982). Below I develop the hypoth- esis that predation has shaped the evolution of anuran lek mating systems in general. Evolution of Anuran Lek Mating Systems The evolution of lek mating systems has recently received much attention (Bradbury 1981, Loiselle and Barlow 1978, Wrangham 1980). Lek mating systems, in which females select mates based on male phenotype or dominance position on a display arena, have been described for a variety of organisms, including two Arizona bufonids (Sullivan 694 Great Basin Natur.\list Vol. 45, No. 4 r2 ^ T [ Bd Br _ Be Bp t ^ 1 T f s'c ' ^ Bw Sm Bm • Ba 50 100 150 200 250 Pulse rate (p/s) Fig. 3. Range in dominant frequency against range in pulse rate for bufonids and pelobatids of Arizona (see Fig. 1 for abbreviations). 1982a, 1983b). However, there has been con- siderable controversy concerning evolution- ary origins of such female choice systems in which females receive no material benefits from their mates (Lambert et al. 1981, Taylor and Williams 1982). Most authors have ar- gued that female preference for communally displaying males is a necessary condition for the evolution of a lek mating system (e.g., Alexander 1975, Bradbury 1981). It is sug- gested that when selecting a mate females save time and energy by sampling males in aggregations. Females might also benefit by selecting their mate from a group of males if a wider range of males (in terms of genetic qual- ity) is available in larger aggregations (Brad- bury 1981). A prediction of these hypotheses is that females should prefer the largest aggre- gation of males available since time and en- ergy savings, and availability of high (juality mates, should increase with aggregation size (Alexander 1975). Many anurans have lek mating systems (Alexander 1975, Emlen 1976, Sullivan 1982a, 1983b, Wells 1977). Populations of B. cognatus and B. woodhousei in Arizona clearly possess lek mating systems (Sullivan 1982a, 1983b). In both of these bufonids, males aggregate in discrete areas of ponds, and females freely select their mates based on male behavior (calling activity). Males do not defend resources of interest to females or their offspring, because oviposition occurs away from the calling site of the male. Hence, females obtain no material benefits from their mates. These two toad species provide an op- portunity for direct test of the hypothesis con- cerning female preference for larger aggrega- tions because male choruses in both species often occur in close proximity to other con- specific aggregations. Models of lek evolution predict that female B. cognatus and B. wood- housei will visit large aggregations when se- lecting a mate. That is, the relationship be- tween nightly operational sex ratio (OSR) or the proportional number of females axailable and number of males present (chorus size) should be positive. For both B. cognatus and B. woodhousei OSU (no. of females/no. of males) and chorus size were not significantlv correlated {Be: r = - 0.35, ? > 0.05, N = 19; Bw-.r^ 0.04, P > 0.05, N = 144). Females were afforded the opportunity to select between aggregations of October 1985 SULLIVAN: SONORAN AMPHIBIANS 695 different sizes since for both species there were always from 1 to 4 choruses simulta- neously active near the study chorus (<0.5 km) on the night of observation (some females even visited more than one chorus). It is im- portant to note that in this analysis the as- sumption of independence of variables is vio- lated since OSR is in part derived from chorus size; however, the results indicate clearly that for these aggregations OSR does not increase with chorus size. Hence, females do not prefer to visit larger choruses when selecting their mates. Observations on the intensity of sexual selection in B. woodJiousei further sup- port the hypothesis that anuran females do not prefer to visit larger aggregations of males when selecting a mate. Sullivan (1986a) docu- mented that the intensity of sexual selection increases with chorus size (male density) in B. woodhoiisei . That is, the proportional num- ber of females decreases as male density in- creases: proportionally fewer males obtain matings in larger choruses. This result suggests that females do not derive any benefits by selecting their mate from among larger groups of displaying males. I suggest that benefits accruing to individual males who display communally are the pri- mary factors influencing anuran lek behavior. For example, many authors have suggested that males who participate in aggregations suf- fer less predation than males who display in isolation (see reviews in Alexander 1975, Bradbury 1981, Loiselle and Barlow 1978, Wrangham 1980). Recently Ryan et al. (1981) substantiated this claim by showing that in a neotropical leptodactylid frog individual pre- dation risk decreases for males as chorus size increases. Ryan et al. (1981) found that at least three predators located males acoustically; hence males might benefit greatly by displaying communally. Predation may only be important to males since they participate in choruses night after night: most anuran females visit choruses only on the night they oviposit and are inconspic- uous while present in breeding aggregations (e.g., Sullivan 1982a, Wells 1977). If males derive substantial benefits by dis- playing communally, and females incur no (or insignificant) costs by selecting their mate from among such groups, then it is conceiv- able that females play no active role in the evolutionary maintenance of lek behavior. Of course, female choice of individual males within aggregations would remain an impor- tant aspect of sexual selection in species with lek mating systems. Observations on mate selection by females also suggest that the spatial distribution of males within a chorus is unaffected by female behavior. For example, some workers have suggested that in lek breeding forms females prefer males positioned at the center of the display area. The linear, shoreline choruses of B. woodhousei can be readily partitioned into three equal areas with respect to the distribu- tion of males within the aggregations to test this hypothesis. Analysis of mating success and chorus location for all males on 18 nights of chorus activity reveals there was no position effect: centrally located males did not achieve a disproportionate number of matings (X^ = 0.33, P > 0.05, df = 2). Thus, within choruses females do not appear to favor the clustering of males. Under the extremely arid conditions of the Arizona-Sonoran Desert the lek- like mating systems and acoustic courtship behaviors em- ployed by anurans seem efficient means of rapidly bringing the sexes together. How- ever, this review indicates that some diversity in mating systems can occur under these severe ecological constraints. It is clear that environmental factors influencing breeding period duration shape mating system organi- zation as a result of corresponding variation in male density and behavior. Continued study will be necessary to further elucidate the pre- cise action of sexual selection in relation to mating system variation in breeding aggrega- tions of these desert anurans. Acknowledgments This work was supported by grants from Sigma Xi, the American Museum of Natural History, and the Zoology Department, Ari- zona State University. Literature Cited Alexander, R. D. 1975. Natural selection and special- ized chorusing behavior in acoustical insects. Pages 35-77 in D. Pimentel, ed., Insects, science and society. Academic Press, New York. 696 Great Basin Natufl^list Vol. 45, No. 4 Arak, a. 1983. Male-male competition and mate choice in anuran amphibians. Pages 181-210 in P. P. G. Bateson, ed.. Mate choice. Cambridge University Press, Cambridge. Arnold, S. J. 1983. Sexual selection: the interface of theory and empiricism. Pages 67-107 in P. P. G. Bateson, ed.. Mate choice. Cambridge University Press, Cambridge. Bradbury, J. W. 1981. The evolution of leks. Pages 138-169 in R. D. Alexander and D. W. Tinkle, eds.. Natural selection and social behavior: recent research and new ideas. Chiron Press, New York. Davies, N. B. and T R. Halliday. 1979. Competitive mate searching in male common toads, Bufo hitfo . Anim. Behav. 27:1253-1267. Drewry, G. E. and a, S Rand 1983. Characteristics of an acoustic community: Puerto Rican frogs of the genus Eletitherodactylus . Copeia 1983:941-953. Duellman, W. E. and R A Pyles. 1983. Acoustic re- source partitioning in anuran communities. Copeia 1983:639-649. Emlen. S. T. and L. W. Oring 1977. Ecology, sexual selection and the evolution ot mating svstems. Science 197:21.5-223. FairCHILD, L. 1981. Mate selection and behavioral ther- moregulation in Fowler's toads. Science 212:950-951. Gerhardt, H. C. 1982. Sound pattern recognition in some North American treefrogs (Anura: Hylidae): implications for mate choice. Amer. Zool. 22:581-595. HODL, W. 1977. Call differences and call site segregation in anuran species from central Amazonian floating meadows. Oecologia 28:351-363. KiRKPATRiCK, M. 1982. Sexual selection and the evolution of female choice. Evolution .35:1-12. Lambert, D M., P. D. Kingett, and E. Slooten. 1982. Intersexual selection: the problem and a discus- sion of the evidence. Evol. Theory 6:67-78. Lande, R. 1981. Models of speciation by sexual selection on polygenic traits. Proc. Nat. Acad. Sci. 78:3721-3725. Light, L. E. 1976. Sexual selection in toads (Biifo amcri- camis). Canadian J. Zool. 54:1277-1284. LOFTUS-HlLLS, J. J. AND M. J. LiTTLEJOHN. 1971. Pulse repetition rate as the basis for mating call discrimi- nation bv two sympatric species oi Hyla. Copeia 1971:154-156. Loiselle, p. V. AND G. VV Barlow 1978. Do fishes lek like birds? Pages 31-75 in E. S. Reese and F. S. Lightner, eds., Constrasts in behavior. J. Wiley & Sons, New York. Parker, G. A. 1982. Phenotype-limited evolutionarily sta- ble strategies. Pages 173-201 in King's College Sociobiology Group, ed.. Current problems in sociobiology. Cambridge University Press, Cam- bridge. Rose. G.J and R. R. Capraniga 1984. Processing ampli- tude-modulated sounds by the auditory midbrain oftwo species of toads: matched tenijioral filters. J. Comp. Physiol. A 154:211-219. Ryan, M J , M D Tlttle, and L K Taft 1981, The costs and benefits of frog chorusing behavior, Behav. Ecol. Sociob. 8:27,3-278. Str.\ughan, I. R 1975. An analysis of the mechanisms of mating call discrimination in the frogs Hyla regilla and H. cadaierina. Copei 1975:41.5-424. Sullivan. B K 1982a. Sexual selection in Woodhouse's toad (Bitfo woodhousci). I. Chorus organization. Anim. Beha\ , .30:680-686. 1982b. Male mating behavior in the Great Plains toad {Bufo cognatiis). Anim. Behav. 30:9.39-940. 1982c. Significance of size, temperature and call attributes to sexual selection in Bufo woodhousci australis. J. Herpetologv' 16:10.3-106, 1983a, Sexual selection in Woodhouse's toad (Bufo woodJwusei). II, Female choice, Anim, Be- hav, 31:1011-1017. 1983b, Sexual selection in the Great Plains toad {Bugo cognatus). Behaviour 85:58-64. 1983c. Sexual selection and mating system varia- tion in the Great Plains toad {Bufo cognatus Say) and Woodhouse's toad {Bufo woodhousei australis Shannon and Lowe). Unpublished dissertation, Arizona State University, Tempe. 1984. Advertisement call variation and observa- tions on breeding behavior oi Bufo debilis and B. punctatus. J. Herpetology 18:406-411. 1985. Male calling behavior in response to play- back of conspecific ad\ertisement calls in two bu- fonids, J. Herpetology 19:78-83. 1986a. Intra-populational variation in the intensity of sexual selection in breeding aggregations of Woodhouse's toad {Bufo woodhousei). J. Her- petolog> , In press, 1986b, Hybridization between the southwestern toad {Bufo microscaphus) and Woodhouse's toad {Bufo woodhousei) in Arizona: morphological vari- ation, J, Herpetology. In press. Sullivan, B. K. and E. A. Sullivan. 1985. Size-related variation in advertisement calls and breeding be- havior of spadefoot toads {Scaphiopus bombifrons. S. Couchi and S, multiphcatus). Southwestern Natur. 30:349-355. Sullivan, B, K, andM. R. Leek, In review. Acoustic com- munication in Woodhouse's toad {Bufo woodhou- sei). I, Response of calling males to variation in spectral and temporal components of advertise- ment calls. Behaviour. Taylor, P. D. andG. C. Williams. 1982. The lek paradox is not resolved. Theor. Pop. Biol. 22:392-409. Tri\ers.K L 1972, Parental investment and sexual selec- tion. Pages 136-179 in B, Campbell, (ed,). Sexual selection and the descent of man: 1871-1971, .\ldine Press, Chicago, Wi:lls, K D 1977, The social behaviour of anuran am- phibians, Anim. Behav. 25:666-693. Wilbur, H. M , D, L. Rubenstein, and L. Fairghild 1978. Sexual selection in toads: the roles of female choice and male body size. Evolution 32:264-270. Wrangham, R. W 1980. Female choice of least costly males: a possible factor in the evolution of leks. Z. Ti(>rps>chol. 54:357-367. THREE NEW SAUROPOD DINOSAURS FROM THE UPPER JURASSIC OF COLORADO James A. Jensen' Abstract. — From 1972 to 1982 three exceptionally large sauropod scapulocoracoids and other ecjually large sauropod bones were collected from the base of the Brushy Basin Member of the Upper Jurassic, Morrison Formation, in western Colorado. Two of the scapulae are conspecific, but the third represents a second genus and possibly a new family. The two conspecific specimens are described here as; Supersaurus vivianae, the second genus is described as Ultrasaurtis mcintoshi, and a large, robust anterior dorsal vertebra of unique form is described as Dystylosaitrus edwini. Various miscellaneous elements are referred to the three genera. Historical The genus Brachiosaurus, named by E. S. Riggs (1903), was part of an articulated skeleton from the Upper Jurassic Morrison Formation within the present city limits of Grand Junction in western Colorado. Riggs believed the genus represented a land- dwelling animal, rather than one preferring an aquatic habitat. No one took him seriously at the time, but modern interpretations of sauropod habits and paleoenvironments agree with him. The brachiosaurs are now considered to be the largest terrestrial animals to have lived on earth. The first uranium boom during World War II triggered the discovery of the second known North American brachiosaur; it was found by prospectors Eddie and Vivian Jones who were looking for uranium, circa 1943, when they collected a brachiosaur humerus from the Uncompahgre Upwarp and donated it to the Smithsonian Institution, where it was put on display. However, no credit was given to the collectors. That display led to the discovery of the Uncompahgre fauna and the materials des- cribed herein. The author saw the Jones humerus in 1958 and later in Colorado found the Jones family, who took him to the location on Potter Creek. They also took him to three other major fossil localities on the Uncompah- gre Upwarp that together produced a rela- tively new dinosaur fauna, described as the Uncompahgre fauna (Jensen, in preparation). The author later returned to Potter Creek and collected additional brachiosaur material (be- ing described elsewhere). Dry Mesa Quarry Three large sauropod scapulocoracoids were collected from one of the Jones localities near Dry Mesa, Mesa County, 35 miles west of Delta, Colorado. Over a period of 10 years the site proved to be very productive, yield- ing many tons of field blocks and packages of dinosaur material. It was named after Dry Mesa and is located near the base of the Brushy Basin member of the Morrison For- mation and, consequently, is not easily con- fused with the overlying Lower Cretaceous, Cedar Mountain Formation when making simple stratigraphic determinations. The top of the Morrison Formation is easy to follow cross-country because the superior Cedar Mountain sediments are set-back above it, forming a prominent shoulder from 100 to 500 m wide. Sag-ponds are a characteristic feature of this shoulder, being produced by land-flow and slumping that crush and mutilate fossils contained in the moving sediments. Its ben- tonitic clays and mudstones, interbedded with soft sandstones, grits, and fine gravels, respond actively to cyclic wet/dry stressing, accelerating the deterioration rate of fossil Earth Science Museum, Brigham Young University. Provo, Utah 84602. Present address: 2821 North 700 East, Provo, Utah 84604. 697 F 5000 ig 1 A Brac/iK««,iru.ssp., dorsal vertebra. B, Holotype, Ultm.saurus nmcintoshi , posterior do 3. C, Holotype, Di/.s^r/Z.^n/n/scf/a/rH-, anterior dorsal vertebra, BYU 5750. D, El, unidentified ntoshi , posterior dorsal vertebra, BYU manual phalanges. )ctober 1985 JENSEN; New Sauhopod Dinosaurs 699 anterior view. E, posterior view. 700 Great Basin Naturalist Vol. 45, No. view^'c; Jt;^i SS";p!!;rr /W "''''"^'' • ''"'^^^'■'"'- ^^^--^^J vertel,ra: A left later .1 . km October 1985 JENSEN: New Sauropod Dinosauks 701 materials beyond that of calcified sandstone or limestone preservations. One of the most important problems yet to be solved is that of the exact age of the Dry Mesa sediments. It is mapped as Morrison Formation, but the fauna does not match taxa of classical Morrison localities. The assemblage is not only verv diverse but contains many taxa previously unknown in the Upper Jurassic of North Amer- ica. The author believes the Morrison sediments exposed along the eastern monocline of the Un- compahgre Upwarp are younger than the Mor- rison in previously described localities, and that the Uncompahgre fauna may represent the last expression of Jurassic dinosaur evolution. Class Reptilia Order Saurischia Suborder Sauropodomorpha Infraorder Sauropoda Family indeterminate Supersauriis vivianae, n. gen., n. sp. Etymology. — Supersaurus , internationally published vernacular name; vivianae, after Vi- vian Jones, co-discoverer of all the important Late Jurassic fossil localities on the Uncompah- gre Upwarp. HOLOTYPE. — BYU 5500, scapulocoracoid 2.44 m (8') long. Referred material. — BYU 5501, scapuloco- racoid 2.70 m (8,10") long; BYU 5502, ischium; BYU 5503, medial caudal vertebra; BYU 5504, 12 articulated caudal vertebrae. Type locality. — Drv Mesa quarry; E 1/2, Section 23: T50N, R 14W, NMPM. Horizon. — Near the base of the Brushy Basin Member of the Upper Jurassic Morrison Forma- tion, Mesa County, Colorado. Collector. — James A. Jensen 1979. Clarification. — Sauropod scapular termi- nology in the literature is not uniform (Hatcher 1903, Mook 1921, Gilmore 1936), resulting in some confusion. This paper describes the scapula in a somewhat normal orientation; using an external view with the glenoid cavity down and the coracoid on the right end, the right scapula will be described. Descriptive terminol- ogy used: narrow midsection is the "shaft"; left end is the "distal" end; upper edge is the "superior border"; lower edge is the "inferior border"; ventral projection of glenoid area is the "glenoid process"; ridge separating the two muscular fossae and running on a curved diag- onal line up from the glenoid process to the maximum scapidar width is the "transverse ridge." This ridge and the shaft-axis form an angle that varies in different sauropod genera. The great depressions to the left (above) and right (below) of the transverse ridge are the "superior fossa" and "inferior fossa," respec- tively. Description.— (Holotype BYU 5500; right scapulocoracoid) Scapula long but not robust; distal end expanding moderately; shaft not severely constricted in midsection. A shallow outward curve in inferior border slightly proximad to greatest width of scapula, at top of transverse ridge, indicates origin of a liga- ment, possibly M. scapulohumeralis. This pro- cess also present on Diplodociis , occurring con- siderably higher up on Cetiosaurus and most prominently developed on Ultrasaurus , but ab- sent or insignificant in Brachiosaunis, Ap- atosanrus, and Camarasaurus. Inferior border of scapula forming a gentle curve from glenoid process to distal end, resembling Apatosaunis and Diplodociis rather than Brachiosaunis or Camarasaurus. Inferior fossa not broadly ex- panded as in Brachiosanrus and longer than wide, contrasting with opposite design in Ap- atosaurus and Camarasaurus . Coracoid with a subrectangular profile. Referred material. — BYU 5501, scapu- locoracoid 2.70m (8' 10") long. Description same as Holotype, BYU 5500. BYU 5502, 12 articulated caudal vertebrae: each approximately 30 cm long, collected but not yet prepared for study. They were exam- ined closely in the field by the author, and a decision was made to refer them to Supersau- rus on the basis of their massive size and general morphology. They were found paral- lel to, and near the Supersaurus scapula; how- ever, location is not a criterion for association in the Dry Mesa cjuarry because of the exten- sive fluvial transport of all elements prior to final burial. BYU 5503, ischium (Fig. 7A): straight shaft, more robust than Diplodocus; distal end ex- panded dorsally, truncated ventrally. Very similar to Diplodocus . BYU 5504, two medial caudal vertebrae (Figs. 7C, D, Dl). C, double-keeled, diplo- 702 Great Basin Naturalist Vol. 45, No. 4 October 1985 JENSEN: New Sauropod Dinosau ,5 m 704 Great Basin Naturalist Vol. 45, No. 4 coid caudal with broad ventral channel; D, Dl, double-keeled diplocoid caudal with transveisely thick neural spine that is ex- panded dorsoventrally at its summit. The cau- dal neural spines oi Diplodociis are thin, nar- row, and unexpanded at the summit; caudal rib missing, no pleurocoel but a short channel exists below base of caudal rib. Proportion of diameter-to-length would place the specimen as number 12 in a Diplodociis caudal series, but the reduction of the neural arch would place it much further back in the same series. Order Saurischia Suborder Sauropodomorpha Infraorder Sauropoda Family Brachiosauridae Ultrasaurus macintoshi, n. gen., n. sp. Etymology. — Ultrasaurus , internation- ally published vernacular name; macintoshi, in honor of John S. Mcintosh, an enthusiastic, indefatigable student of sauropods who en- courages everyone to greater effort in their behalf HOLOTYPE. — BYU 5000, posterior dorsal vertebra. Referred material. — BYU 5001, scapu- locoracoid; BYU 5002, anterior caudal verte- bra; BYU 5003, medial cervical vertebra; Type LOCALm.— E 1/2 Section 23, T 50N, R 14W, NMPM, Mesa County, Colorado. Horizon. — Base of Brushy Basin Member of the Upper Jurassic, Morrison Formation. Collector. — James A. Jensen 1979. Diagnosis. — A sauropod diflFering from other brachiosaur genera in having a scapula with a moderately expanded distal end; dorsal vertebrae with anteroposteriorly narrow neu- ral spines; midcervical vertebrae lacking pleurocoels; posterior dorsal vertebrae with high neural spines; anterior caudal vertebrae with high neural spines. Description. — Posterior dorsal vertebra (Figs. IB; 3A, B, C). A long centrum charac- teristic of brachiosaurs: ratio of length to di- ameter, 1.2; Potter C^reek brachiosaur, 1.3; Brachiosaurus altitlwrax (Riggs 1903), 1.07. Ultrasaurus shares the family characteristic of a long dorsal centrum with Brachiosaurus , but in other features it has no parallel with that genus. The Potter Creek Brachiosaurus, collected by Jones and Jensen (description in preparation by the author) appears to be al- tithorax (see Figs. lA, B; 3A-D). Figure 3C is a detail of the base of the neural spine and postzygapophyses of BYU 5000. The form of the suprapostzygapophysal laminae is greatly altered by assymetry, with the right side being more robust than the left. The vertebra is crushed transversely, making an anterior view nearly useless. The neural spine is tall and buttressed posteriorly by suprapostzygapophysal laminae 2/3 the height of the spine (Fig. 3B). In contrast, the supraprezygapophysal laminae are almost nonexistent. The general structure of the neu- ral spine is fragile compared to the Potter Creek specimen (Fig. lA) and altithorax (Riggs 1903). Pleurocoels are present, their length being less than twice their height. Referred material. — BYU 5001, scapu- locoradoid. The inferior fossa is broader than in Supersaurus (Figs. 4A, B); glenoid process projects beyond the inferior border of the shaft as in Brachiosaurus; overall length ap- proximately 2.70 m (8' 10"); dorsoventrally, midshaft is constricted to 23 cm (9"). A tabular process occurs on the inferior border, slightly above the transverse ridge and slightly below the minimum dorsoventral diameter of the flattened shaft (Fig. 4A). The function of this tabular process is unknown but appears to support the origin of an intercostal muscle or possibly an m. humeroscapularis. The process is not a gentle convexity below the inferior border, as in Supersaurus, but instead is a distinct tab erupting from the inferior border with a base approximately 100 mm in length, extending approximately 50 mm ventrally from the edge of the scapula (Fig. 4B). BYU 5002, anterior caudal vertebra (Figs. 2D, E, 3E). This is probably No. 3 caudal with a spine taller than those on anterior brachiosaur cau- dals. All zygapophyses are damaged by crush- ing, but two infraprezygapophxsal laminae support the anterior pair. The suprapostzx ga- pophysal laminae diverge about midway up the spine, expanding to expose a broad, rugose development of the postspinal lamina. Viewed anteriorly the upper hallOf the spine presents a modified rectangular profile. Like BYU 5000 dorsal vertebra, there are no supraprezygapophysal laminae. The supradi- apophysal laminae ibrm a wide border enclos- October 1985 JENSEN; New Sauropod Dinosaurs 705 Fig. 6. A, Measuring Supersaurus vivianae scapulocoracoid. D. E., Vivian Jones; J. A. Jensen. B, The author, e'a" tall beside Supersaurus vivianae scapulocoracoid. 706 Great Basin Naturalist Vol. 45, No. 4 5503 sL^;,^ ' f ^P^^^""^^". Supersaurus vivianac ischium. B, Ultrasaurus caudal vertebra. C BYl' 5503, Supersaurus vtvtanae, referred specimen, i.schiuu,. I). 1)1. BVU .5504. Suprrsaurus vivianae referred snc men, caudal vertebra. E, unidentified caudal vertebra ' leterred sjx c ,- October 1985 JENSEN: New Sauropod Dinosaurs 707 ing a narrow, liut prominent, prespinal lam- ina. A small pleurocoel is present and the centrum is not procoelous but both ends are nearly flat. BYU 5003, medial cervical Vertebra (Fig. 2A, B, C). The original referred specimen is not illustrated. What is seen is a life-sized model of BYU 5003, constructed from careful measurements taken from the crushed origi- nal. All brachiosaur cervical vertebrae are ex- tremely fragile in construction and generally found badly crushed. The cervical rib is arbi- trary, but the vertebra was modeled as care- fully as possible from the original. One re- markable feature is the absence of pleur- ocoels, so radically developed in some sauropod families, such as that of the Diplodo- cidae . Also, the postdiapophysal, or horizon- tal, laminae are missing. The spine is single, being slightly lower than the summit of the supraprezygapophysal laminae, which align with the elevated suprapostzygapophysal laminae to provide a channel for the long cer- vical flexor muscles. The anterior convexity is exaggerated approximately 5%, and the postzygapophysal articular facets were not modeled. Order Sauri.schia Suborder Sauropodomorpha Infraorder Sauropoda Family indeterminate Dijstylosatiriis edwini, n. gen., n. sp. Etymology. — Greek: di, two; stylos, beam; sauros, lizard; edwini, in honor of the late Daniel Edwin (Eddie) Jones, who, with his wife, Vivian, brought more new dinosaur taxa to science than any other two amateurs while providing 20 years of logistic support for fieldwork on the Uncompahgre "Plateau." HoLOTYPE. — BYU 5750, anterior dorsal vertebra. Type locality.— E 1/2, Section 23; T 50N, R14W, NMPM. Horizon. — Near the base of the Brushy Basin Member of the Upper Jurassic Mor- rison Formation. Collector. — James A. Jensen 1972. Diagnosis. — A sauropod differing from all described North American sauropod genera in having two parallel, diagonal infraprezyga- pophysal laminae supporting each hy- pantrozygapophysal arch in anterior dorsal vertebrae; lower half of neural arch massive, the neurocentral suture occupying nearly 7/8 the length of the centrum; neural spine fragile, being transversely broad but antero- posteriorly thin; supraprezygapophysal lami- nae not convergent at midshaft as in Bra- chiosaiirus; neural arches of dorsal vertebrae completely pneumatic, including spine, transverse processes, and zygapophyses. Description. — The location of the parapo- physes at the neurocentral suture and the presence of a strongly developed hyposphen/ hypantrum articulation locates the vertebra anterior to No. 3. Lower half of neural arch massive, being characterized on its anterior face by four diagonal infraprezygapophysal laminae, two below each zygapophyses. Each pair of these diagonal supports are spaced well apart and more or less parallel, supporting the hypantrozygapophysal arch on each side (Fig. 5A). The hypantrum consists of two opposing articular faces formed by the down-turned, medial edges of the prezygapophyses. Each zygapophysis thus specialized forms an arch termed a hypantrozygapophysal arch. The ventral ends of these four laminae rise from the anterior surface of the parapophyses; in- ternal lamina of each pair supports the ventral end of the hypantrum; external pair supports the prezygapophyses. No other sauropod fam- ily displays such a well-designed mechanical arrangement for supporting the hypantrozy- gapophysal arches. Thin supradiapophysal laminae rise to the lateral spur near the sum- mit of the neural spine, making the spine transversely broad (5A). However, viewed laterally (5E), the spine is very slender and fragile when compared to the lower half of the neural arch. By comparison the neural spine oi Brachiosaurtis (5F, E) is deep and robust anteroposteriorly, but the base of the neural arch occupies no more than 2/3 the length of the centrum; the neurocentral suture on Su- persaurus extends approximately 7/8 the length of the centrum. The zygapophyses are small and weak, but the hyposphen/hy- pantrum structure is as strongly developed as it is in the anterior dorsal vertebrae of Barosauriis (Lull 1919). This vertebra bears little resemblance to any described genus and no doubt represents a new sauropod family. Many elements be- longing to it were probably collected from the Great Basin Naturalist Fig. 8. For comparison only; not to scale. Profiles of various sainopod scapulae and scapulocoracoidae: A, C, E, G, scapulae; B, U, F, H, I, .scapulocoracoidae. A, Haploratithosaurus sp. B, Supersaurus vivianae , first specimen. C, Cetiosatirussp. D, Diplodocus loitaus . F. Cavuirasaunis sti))n'mus . F, Apato.saunt.slotii.sae . C, Supersaurus vivianae second specimen. H, Brachiosaurus hrancai . I, Vltrasaurus macintoshi . B is 2.44 m (8) long; G is 2.74 m (S'lO") long. October 1985 JENSEN: New Sauropod Dinosaurs 709 Dry Mesa quarry, but correct associatiou is difficult wheu dealing with masses of" disartic- ulated elements belonging to undescribed genera. Discussion. — In 1983 the unprepared di- nosaur materials collected by the author for the Earth Science Museum at Brigham Young University amounted to appro.ximately 100 tons. This mass of material will require many years of careful preparation and, as of this writing, only a very small part of it had been prepared for study. The Uncompahgre fauna came from massive deposits of disarticulated bones, except two more or less complete articulated sauropod skeletons from the Dominguez/Jones quarry, and it will be diffi- cult, if not impossible, to properly associate many of the elements into generic sets. One paper including a faunal list is in press, and a second much larger paper is in preparation illustrating most of the prepared material and giving some brief descriptions. Literature Cited GiLMORE, C. W. 1936. Osteology oi Apatosaurus with special reference to specimens in the Carnegie Museum. Memoirs Carnegie Museum 11(4): 17,5-276. Hatcher, J. B. 1903. Osteology of Haplocanthosaurus with additional remarks on Diplodocus. Memoirs Carnegie Museum 1(2): 1-75. Mock, C. C. 1921. Camarasaurus, Amphicoelous and other sauropods of Cope. Memoirs American Mus. Nat. Hist. (3)3: 247-387. RiGGS, E. S. 1903. Brachiosaurus altithorax , the largest dinosaur known. American Journal of Science (4)15: 299-306. UNCOMPAHGRE DINOSAUR FAUNA: A PRELIMINARY REPORT James A. Jensen Abstract.— A diverse late Jurassic dinosaur fauna, discovered in western Colorado in 1963, contains many unde- scribed taxa that may represent evolutionary trends at the generic level not previously reported from the Morrison Formation. A preliminary faunal hst is given. Bones of the largest known dinosaur, Ultrasaurus , are present as are a variety of small animals, including Pterosaurs, in which one sacrum displays avianlike fused sacral neural spines. A new family, the Torvosauridae , erected, based on the genus Torvosaunis that is redescribed. One of the most diverse Jurassic dinosaur faunas in North America was found on the Uncompahgre Upwarp in western Colorado in 1963. This fauna contains more unde- scribed taxa than has been encountered in any other North American Jurassic assemblage in this century. The first vertebrate fossil collect- ing on the upwarp was by the author in 1964 and continued for the next 20 years. More than 50 tons of dinosaur bone and matrix were collected from an area stretching 35 miles along the upwarp's eastern monocline. Very little systematic work was done during those two decades of collecting for several reasons: (1) all available funds were used for collecting; (2) the most productive quarries being worked contained masses of disarticu- lated bones that could not be separated with confidence into specific sets; (3) the author thinks that any attempt to sort and describe extensive deposits of disarticulated material before the largest possible sample is taken will not produce the most comprehensive results; and (4) the collecting program was so produc- tive that it acquired more material each sea- son than could be prepared for study in five or more years. However, many representative specimens were prepared for study, and one unique carnosaur was described and named Torvosatirus tanneri (Galton and Jensen 1979). The author believes the fauna will demon- strate substantial evolution at the generic level when compared to classical Morrison assemblages. Dodson et. al. (1980) state that their field investigations "failed to find any convincing evidence of evolution at the generic level within the Morrison Forma- tion." There are familiar forms in the Uncom- pahgre fauna, but there is also consistent evi- dence of change, or "evolution at the generic level" as demonstrated by: (1) giganticism in more than one sauropod family; (2) at least a 100% increase in carnosaur genera; (3) the presence of the first relatively abundant pterosaur elements, previously known in the North American Jurassic from one phalangial fragment from Como BluflP; (4) undescribed variations in sauropod skeletal morphology', particularly the axial skeleton; and (5) the presence of ornithischians above the Mor- rison average, plus various other novel differ- ences. A problem of identification in this diverse Un- compahgre fauna is focused on the question "How far must an evolving genus move from parent stock, i.e., change morphologically, be- fore it qualifies as a new genus?" Satisfactory criteria to deal with this question do not exist. Other Morrison (juarries today generally pro- duce specimens that can be confidently identi- fied with described material in genera that are comfortabK' distinct from one another. The Un- compahgre fauna displays so man\' variations on classical niorpholog}- that it probably represents either an ad\ anced or \ ounger fauna. It contains many specimens that look familiar, as if the>' are closely related, yet vary enough in structure to qualify as new taxa. They may represent exolii- tion at the generic level. 'Earth Science Mus: Briuliam Vouns University, Provo, Utali 84602 :2821 North 7(H) East. Pn 710 October 1985 JENSEN: UNCOMPAHCRE DINOSAUHS 711 Detailed studies of the Unconipahgre fauna may at least provide a new window on di- nosaur evolution and possibly shed some light on the time-transgressive evolution of Jinassic dinosaurs into Cretaceous forms. The Uncompahgre fauna will be discussed further and illustrated in a larger paper, "New and Undescribed Dinosaurs of the Southwest- ern Colorado Plateau," now in preparation. Taxonomic Revision Romer (1956) listed four carnosaur families: Palaeosauridae, Teratosauridae, Megalosaur- idae, and Tyrannosauridae. A conservative modern interpretation of the infraorder Carn- osauria, as used by Russell (1984), retains the Megalosauridae and Tyrannosauridae but dis- cards the other two families and adds Cer- atosauridae, Allosauridae, Aublysodontidae, and Dryptosauridae. Galton and Jensen (1979) placed the genus Torvosaurus in the family Megalosauridae, but subsequent laboratory work has doubled the number of elements available for study, revealing a number of diagnostic features in Torvosaurus not seen in the Megalosauridae. Although unique morphological characteris- tics of this genus set it apart from the families listed by Russell, it shares a few common fea- tures with the Megalosauridae. In contrast, Allosaurus and Tyrannosaurus, representing two different families, share common charac- teristics to a much greater degree than either one resembles Torvosaurus. The genus Torvosaurus is best character- ized as being a theropod with both primitive and advanced characteristics: the pubis and ischium are of the prosauropod, brachyiliac type; the ilium is of the coelurosaurian, doli- choiliac type. This combination has not been seen in any North American theropod (Galton and Jensen 1979) and may be the only exam- ple from any age. Torvosauridae, family The new family Torvosauridae is proposed to receive the genus, species, Torvosaurus tanneri. Order Saurischia, Suborder Therop- oda, Infraorder Carnosauria, New Family Torvosauridae. Dia(;n()SIS. — Characterized by robust long bones; skull short, massive, and moderately low; forelimb very short with ratio of maxi- mum length of humerus to radius at more than 2; lachrymal with a 9()-degree angle between maxillary and jugal rami; very distinctive pelvic girdle with ilium dolichoiliac; pubis and ischium brachyiliac; pubis with closed obtura- tor foramen. Pubis with an almost continuous median symphysis; very small non-weight- bearing pubic foot. Type genus. — Torvosaurus Galton & Jensen 1979. Revised description of Torvosaurus: A large, heavily built theropod with a short skull and a total body length of at least 10 m. Three premaxillary teeth with no rectilinear grooves (Figs. 2D, Di); 10 maxillary teeth (Fig. 1, Bi); open foramina along superior border of fused interdental plates exposing germinal teeth; short dentary with 10 teeth (Figs. 3A-B); un- sutured median symphysis; no meckelian groove on medial surface (Fig. SAJ. Lachry- mal horn absent (Fig. lA^); lachrymal vacuity opening forward (Fig. lAj. Dorsoventrally broad jugal with narrow preorbital ramus. Forelimb with heavy himierus and short fore- arm with ratios of maximum length of humerus to radius at 2.2; humerus straight with large deltopectoral crest, broad distal and proximal ends (Fig. 4D); proximal end of ulna massive with ratio of maximum length to maximum proximal width at 2. 1; metacarpal I with square proximolateral corner; first pha- lanx of digit I stout, short, and helically twisted along its length; metacarpal II short but extremely massive with ratio of maximum length to maximum proximal width at 1.5; metacarpal III massive, ratio 2.2. Ilium heavy with low dorsal blade, broad brevis shelf and transversely wide acetabular surface. Pubis with closed obturator foramen and a nearly continuous median symphysis on both pubis and ischium. Pubis with no horizontal weight- bearing ventral plane. Astragalus massive, as- cending ramus thick and truncated toward calcaneum. (See Galton and Jensen 1979 for previously published figures). Metatarsals massive with no distal lateral or medial pits. Cervical vertebrae with subcircular ball-and- socket joints, the vertical axes being shorter than the horizontal axes; large pleurocoels openly communicating with internal pneu- 712 Great Basin Naturalist Fig. 1. A-Al, Torvosaurus tanneri , left lachrymal: A, medial view. Al, lateral view. B-Bl, Torvosauru.s tmrncri. left maxilla. B, lateral view. Bl, medial view. Abbreviations: a, alveoli; ar, anterior ramu.s; dr. de.seendinii nmnis; fip. iused interclintal plati's; idt, niterdentai foramen; If, lachrvmal foramen; 1\', lachrymal vacnit\'; ps, prema.\illar\ sulnre. October 1985 JENSEN; UNCOMPAIICHE DINOSAURS 713 fj 10 cm Fig. 2. A-Al, Torvosaurtis tanncri atlas intercentriiin with coossified left neuropophysis. A, anterior view. Al, posterior view. B, right lateral view. C, ventral view. D-Dl, Torvosaurtis tanneri right prema.xilla. D, medial view. Dl, right, lateral view. Abbreviations: A In, atlas intercentrum; N, neuropophysis; ac, anterior concavity; en, external naris; nf, nutrient foramen; ns, neurocentral suture; pc, posterior convexity; pms, premaxillary symphysis; pz, postzygapophysis. 714 Great Basin Naturalist Fig 3. Torvosaurus tanncri: A-Al, left dentary; A. lateral view. Al. medial Niew. B. simpUtjed tooth diagram, germinal teeth not shown. C, unidentified carnivore tooth. D. Torvosaurus tanncri posterior dorsal vertebra, h, median cross section of torvosaur dorsal vertebra. , t , iJ Abbreviations: a, alveoli; eb, enamel bases; gt, germinal tooth; hp, lupantrnm; ne. .leural canal, .ua, nutrient canal, nf, nutrient formamina; pz, postzygapophyses; s, symphysis; spzl, supraprezygapopinses; spzl 1. sup.apie/.>,MPO- physal flange; ssf, snbspinal fenestra; tr, tooth root. October 1985 JENSEN: UNCOMPAHCRE DlNOSAUHS 715 matic structure of centrum; anterior ends of centrae with radial flange, or collar, around subspherical convexity; posterior ends of cen- trae with subcircular concavity deeper than length of anterior convexity. Dorsal vertebrae with transverse, subspinal fenestrae passing transversely, anterior to and isolating hypo- sphenal pillar; expanded bases of supraprezy- gapophysal laminae on posterior dorsal verte- brae intruding onto posterior, superior surfaces of prezygapophyses with no fusion between ventral surface of intrusion and supe- rior surface of prezygapophyses (Figs. 3D, 4B, Bi). Caudal vertebrae with transverse processes backswept approximately 30 de- grees (Fig. 4C,). Chevrons more subquadran- gular than bladelike in cross-section (Fig. 4F). Type species: Torvosaurus tanned Galton and Jensen 1979. Uncompahgre Faunal List This list is intended to provide a general view of the diversity of the Uncompahgre fauna. It is not certified as being comprehen- sive, correct, or complete. Identifications, for the most part, are tentative. With the excep- tion of one described theropod, very few gen- era are hsted because the comparative re- search necessary to work below the family level will not, and cannot, be done by the author. Additional taxa are doubtless present, but the nature of the specimens and the great amount of material awaiting preparation pre- cludes their recognition at the present time. Class OSTEICHTHYES Subclass Lepidosauria Order Rhynchocephalia Family Sphenodontidae Undescribed genus, species Subclass Archosauria Order Dipnoi Family Ceratodontidae ? Ceratodus sp. Class Reptilia Order Chelonia (Testudinata) Family Pleurosternidae ? Glyptops sp. Undescribed genus and species Order Crqcgdilia Family Crocodylidae Crocodylinae Order Saukisciiia Suborder TllEROPOUA Infraordcr Coeluhosauria Coeluridae Undescribed family Infraordcr Caknosauria Allosauridae new genus, species Torvosauridae Torvosaurus tanneri One or more undescribed families Infraordcr Omithomimosauria Ornithomimidae Suborder Sauropodmorpha Infraordcr Sauropoda Brachiosauridae Ultrasaurus Brachiosaurus ? alius undescribed genera Camarasauridae undescribed genera Diplodocidae undescribed genera One or more undescribed families Order Ornithischia Suborder Ornithopoda Hypsilophodontidae Laosaurus indeterminate species Othnelia indeterminate species Iguanodontidae Camptosaurus unidentified species Suborder Stegosauria Stegosauridae Stegosaurus indeterminate species Order Pterosauria Suborder FxERODACnLOiDEA Pterodactylidae unidentified genera and species Undescribed? suborder (avianlike fused sacral neural spines) Class Mammalia Incertae sedis: distal half of humerus (Probably multituberculate or triconocont) The Uncompahgre fauna includes some of the most spectacular fossil bones ever found. Their size equals or exceeds that of the di- nosaurs from Tendaguru Hill in Tanganyika (Tanzania), Africa, which produced the skele- ton of the great Brachiosaurs brancai, long displayed as the world's largest dinosaur in the Museum fiir Naturkunde in Berlin. It stands 11.87 m tall and 22.65 m long. The British Museum of Natural History also col- lected material from Tanzania, but specula- 716 Great Basin Naturalist Vol. 45, No. 4 Fig. 4. Torvosaurtts tanneri, A-F: A, posterior dorsal vt-rtehra, right lateral \ iew. B-Bl, posterior dorsal vertebra. B, right lateral view. Bl, superior view. C-Cl, articulated medial caudal vertebrae. C, right lateral view. CI, ventral view. F, posterior view of chevron, articulates with C, (^1. D-D2, three theropod forelegs representing three families: D, Torvosauridae, Torvosatirus tanneri. Ul, Tyrannosauridae. uiidescribed geiuis, species. D2- Allosauridae. AUosaurus%\). E-E2, Othnclia sp. ilium: E. left lateral view. El, ventral \ie\v. E2, medial \iew. Abbreviations: hp, hypantnun: pi, plemococl; iipi)/-i)()steri()r i:)n)ccss i)i'i)rez\ gapopliysis; po/, postz\gapoiihvsis; pz, prezygapophysis; spzl f, snpraprezygapophysal laminar llangc; ssf, subspiual fenestra. October 1985 JENSEN: UnCOMPAHCRK DINOSAURS 717 tion as to which locahty produced the largest dinosaur is put to rest by the presence of a cervical and a dorsal vertebra, both nearly 4'6" (1.36 m) long and two scapulocoracoids 8' 10" (2.70 m) long horn Dry Mesa. Because of perpetual universal interest in dinosaurs, the discovery of these materials was given wide exposure by the international news media. The first such discovery was that of a sauropod scapulocoracoid eight feet (2.44 m) long in 1972. It was given the popular name "Supersaurus" by an article in the do- mestic and all international editions of the Readers Digest (George August 1973). A larger scapulocoracoid, 8' 10" (2.70 m) long, was collected from Dry Mesa the same year but was not prepared until several years later, thus remaining unknown to the news media. It displays diplodocid affinities. In 1979, during the filming of a Japanese Television Workshop documentary at the Dry Mesa quarry, a second scapulocoracoid 8' 10" (2.7 m) long was discovered and received ex- tensive international publicity as the "world's largest dinosaur" under the popular name "Ultrasaurus." At the time of the discovery, and in response to persistent questions from the news media asking what the huge creature would be called, the author replied that, since it was probably the ultimate in size for a land animal, unofficially he would name it Ultra- saurus. Subsequent universal usage of this name, ap- plied to the huge dinosaur bones from Dry Mesa, has established it as the "world's largest dinosaur " in many publications such as various textbooks; documentary films; science pam- phlets for school children; popular books written by scientists, e.g., Dinosaurs, an Illustrated Histonj by Dr. Edwin H. Colbert; books by pop- ular writers, such as Dinosaurs of North Amer- ica by H. R. Sattler, The New Dinosaur Dic- tionary by Donald Glut, and Dinosaurs Discovered by John Gilbert; and various other hardcover publications, all of which cite Ultra- saurus as being the world's largest dinosaur. The Ultrasaurus scapulocoracoid represents a family unlike that of the other two large speci- mens. It is characterized by a constricted central shaft 9" wide (23 cm) and a prominent tabular process on the lower anterodorsal border, and it is not greatly expanded dorsoventrally at the distal end as is the case in Brachiosaurus. This huge bone was protected only by a shallow layer of loose dirt and rocks when discovered, having been subjected to deterio- ration from plant-root raiding, frost action, and water leaching for a long time. Its ex- tremely fragile condition suggested a plaster mold be made of the first side exposed to preserve all dimensions in case of mishaps during collection and transportation. A model was subse(}uently developed from the mold, and a cast of it has been circulated internation- ally as part of "Ultrasaurus, the world's largest dinosaur. ' Additional large bones were collected that obviously belong to one of the three large scapulocoracoids, but they cannot be cor- rectly matched up; e.g., a cervical vertebra nearly 4'6" (1.36 m) long was collected, but there is no way to relate it to a particular scapula. Other unusually large sauropod elements collected include: a posterior dorsal vertebra 4'6" (1.36 m) tall; a very robust median dorsal vertebra 1 m tall; several cervical vertebrae more than 1 m long; an unusually large ante- rior caudal vertebra; an articulated posterior caudal series 12 feet (3 m) long; an ischium; various rib materials, including a rib head 18" (45 cm) across the capitulum/tuberculum di- mension; and a pedal phalange. No large sauropod cranial elements or appropriately large limb bones were found. Theropods At least four genera and more than one unidentified family are represented by theropod humeri. An equal number of novel forms can be identified in caudal vertebrae, but since humeri and caudal vertebrae do not articulate there is no way to properly match them up. One humerus displays an anomaly involv- ing the distal half of the shaft. A massive ab- normal growth of bone penetrated by random vascular burrows indicates a serious patholog- ical condition in life. The elbow joint was de- stroyed and the forearm was completely use- less, if not severely atrophied. The proximal half of the bone is moderately robust, with a straight shaft, a moderate expansion of the proximal articular surfaces, and a short deltoid crest. It belongs to an undescribed genus in an 718 Great Basin Naturalist Fig. 5. A-D, Torvosaurus tanneri right premaxilla: A, lateral view. B. niax.lUuy contac o. P^^™^^^^^^^ anterior edge E ty;o shed teeth,? Torvosaurus tanneri. F, H, I, Torvosaurus tanncn. F. letma.x.lla with outline "er^ax Ha. H, Mt dentary. I, four .shed t.eth in dav model for contrast with C. left dentary of Allosaurus sp. October 1985 JENSEN: Unc()mpah(;he Dinosaurs 719 unidentified family. Other al)nornial condi- tions occurring on elements in the fauna in- clude various excellent examples of gnawed bones. A Remarkable Carnosaur The genus and species Torvosaitrus tanneri was erected on the holotype elements consist- ing of the left humerus, radius, and ulna. A matching right forelimb was found nearby but not in close association. The holotype and all referred materials were used in the familial and generic diagnoses. The referred material was selected because it matched the very robust nature of the genus and because there was no evidence of a second robust carnosaur genus present in the de- posit. Based on a comparison of a dozen metatar- sals and an apparent enigma in the distribu- tion of diagnostic characteristics in the verte- bral column, the author feels there may be two torvosaur species present in the fauna. That possibility may be considered by future workers. Small Dinosaurs There is an interesting variety of single ele- ments from small dinosaurs in the fauna. Small, proximally compressed metatarsals are characteristic of both coelurosaurid and or- nithomimid genera making the identification of incomplete pedal elements guesswork. No small cranial elements have been recovered to date, but one very important diagnostic pelvic element was recently uncovered in a plas- tered block in the laboratory. An Othnelia sp. ilium (Fig. 4E) was found marking another occurrence of this character- istically European genus in North America. Galton and Jensen (1975) incorrectly identi- fied the locality producing the first known North American Othnelia skeleton as being in the Lower Cretaceous. Instead, the skeleton was collected by the author from the base of the Brushy Basin Member, Upper Jurassic Morrison Formation, near an important fossil plant locality (Chandler 1966) in central Utah. This small ornithopod belongs to a suborder widely distributed in the Upper Jurassic of North America but poorly known, partly be- cause of a paucity of good study specimens. Its infraorder, Ilypsilophodontia, has undergone various revisions in recent years and at present is in a state of mild confusion. Because of this and the incomplete nature of or- nithopod materials from Dry Mesa, miscella- neous small specimens such as centrae and random fragments of small bones can only be listed as unknown ornithopod or theropod. Pte Pterosaurs were previously known from the Jurassic of North America by one short pha- langial fragment from Como Bluff. Dry Mesa produced the first significant Jurassic pterosaur material, which includes a .sacrum with avianlike fused neural spines and a very reduced caudosacral vertebra. The tail was extremely small, with no important function. The neural spines contrast with all described forms that display short, well-separated spines of equal height. One procoelous dorsal vertebra is present, together with three humeri, various more or less complete man- ual elements, a nearly complete scapulocora- coid, a tiny femur with a spherical head set on a stem nearly parallel to the femoral shaft, and a tiny unidentified cervical vertebra that may be pterosoaurian. Acknowledgments Interested students and aspiring paleontol- ogists who worked on the Uncompahgre fauna project over two decades include Lee Perry, Dennis Belnap, Kevin Maley, Michael Fra- casso, Michael Scheetz, Rodney Scheetz, Brooks Britt, Dee Hall, Richard Erickson, and others. Funding for the project came from a regular Brigham Young University, Earth Science Museum budget plus contributions from private sources such as the National Ge- ographic Society, the Sierra Club, small dona- tions from hundreds of children, and, in par- ticular, the Kenneth Thomson family of Toronto. Logistic and social support were consistently supplied by the late Daniel ("Eddie") Jones and his wife, Vivian, of Delta, Colorado, who directed the author to the best fossil localities worked. The author expresses his sincere thanks to them for knowing where to look and for demonstrating for 20 years an enthusiastic determination to get there. Their contribution to science is invaluable. 720 Great Basin Naturalist Vol. 45, No. 4 Literature Cited Chandler. M. E. J. 1966. Fruiting organisms from the Morrison Formation of Utah, USA. Bull. British Museum (Natural History) Geology 12(4): 139-171. Colbert, E H 1982. Dinosaurs, an illustrated history. Hammond Inc., New Jersey. 224 pp. DoDSON, P , A K Behrensmeyer, R. T. Bakker, J S McIntosh. 1980. Taphonomy and paleoecology of the dinosaur beds of the Jurassic Morrison Forma- tion. Paleobiology 6(2): 208-2.32. Galton, P M , J A Jensen. 1975. Hypsilophodon and Iguanodon from the Lower Cretaceous of North America. Nature 257(5528): 666-69. 1979. A new large theropod dinosaur from the Upper Jurassic of Colorado. Brigham Yoimg Uni- versity Geology Studies. 26(2): 1-12. George, J 1973. Supersaurus, the greatest of them all. Reader's Digest, August. Glut, D F 1982. The new dinosaur dictionary. Citadel Press, New Jersey. 288 pp. Jensen. J. A., and J. H Ostrom. 1977. A second Jurassic pterosaur from North America. Journal of Paleon- tology 51(4): 867-870. Marsh. O. C 1878. New pterodactyl from the Jurassic of the Rockv Mountains. American Journal of Sci- ence. 3(16): 233-234. Prothero. D R., and J A Jensen. 1983. A mammalian humerus from the upper Jurassic of Colorado. Great Basin Nat. 43(4): 551-53. RiGGS, E S. 1904. Structure and relationships of opisthocoelian dinosaurs. Field Columbian Mus. Geol. Series 94(2): 229-247. RoMER, A. S. 1956. Osteology of the reptiles. University of Chicago Press. 772 pp. Russell, D. A. 1984. A checklist of the families and genera of North American dinosaurs. National Museums of Canada. Syllogeus 53: 1-35. FOOD HABITS AND DIETARY OVERLAP OF NONGAME INSECTIVOROUS FISHES IN FLINT CREEK, OKLAHOMA, A WESTERN OZARK FOOTHILLS STREAM' C. Stan Toclcl"" and Kenneth W. Stewart" Abstract. — Insectivorous fishes were sampled from March, 1983 to February 1984, in Flint Creek, Delaware Co., Oklahoma. There was insignificant habitat segregation between Etheostoma spectahile and E. punctulatum and seasonal habitat partitioning between Cottiis carolinae and both darters. Mature £. spectahile ate primarily chirono- mids and mayflies, whereas juveniles fed primarily on microcrustaceans. Mature £. punctulatum consumed fewer Ephemerella and Leptophlebia than E. spectahile , feeding on Stenonema and other crustaceans. Juvenile E. punctula- tum fed mainly on amphipods and mayflies, and juvenile E. spectahile ate primarily microcrustaceans. Cottus carolinae elected primarily mayflies in spring-summer and chironomids in January-February. Coefficients of dietary overlap were highest between larger E. spectahile and juvenile E. punctulatum and lowest between immature E. spectahile and mature E. punctulatum . Overlap between the two darters was significantly correlated with differences in mean prey size (p<0.0005). Overlap between sizes of £. spectahile was also significantly correlated to differences in mean prey sizes. Etheostoma spectahile generally preferred smaller prey than £. punctulatum . All three species avoided Stenelmis . Cottus carolinae avoided microcrustaceans. The study showed that resource partitioning among these three insectivorous fishes is affected by complex interactions of habitat and prey electivity, and prey size selectivity. Darters and other nongame insectivorous fishes, such as sculpins and madtoms, are of- ten found coexisting in the same stream (Dia- ber 1956, Braasch and Smith 1967, Page and Smith 1970, 1971, Novak and Estes 1974, Page 1974, Flynn and Hoyt 1979, Matthews etal. 1982, Wynes and Wissing 1982). In such systems, resource partitioning by coexisting species is expected (Cause 1934, Zaret and Rand 1971). Smart and Cee (1979), Matthews et al (1982) and Wynes and Wissing (1982) have shown varying degrees of segregation of food and habitat use in coexisting darter spe- cies. Northcote (1954) studied the ecology of two sculpin species, and Daiber (1956) com- pared the feeding habits of the mottled sculpin and the fantail darter. Ecological seg- regation has also been shown among several other groups of sympatric freshwater fishes (Zaret and Rand 1971, Moyle 1973, Mendel- son 1975, Werner and Hall 1976, Sural et al. 1980, Baker and Ross 1981). Several factors may influence competition for food between cohabiting fishes. These in- clude primarily: (1) the utilization of different habitats, (2) prey size selectivity, and (3) the selection of specific prey species. Spatial segregation by stream fishes is diffi- cult to assess, and studies based on seine hauls may be deceptive, since the method of cap- ture potentially disturbs them to the point where their distribution may no longer repre- sent natural activities, or the sampling area is so large that important differences in micro- habitat usage are obscured. The food habits of some darters and sculpins change as they mature (Koster 1937, Bailey 1952, Daiber 1956, Braasch and Smith 1967, Page and Smith 1970, 1971, Scalet 1972, Page 1974, Flynn and Robert 1977, Schenec and Whiteside 1977, Layzer and Reed 1978). This suggests an age-related change in the strategies of food partitioning. Different size classes of a species are often found in different habitats. This potentially reduces both interspecific and intraspecific competition. The selection of different prey or diflferent prey sizes may be influenced by many factors such as morphological, physio- logical, and behavioral characteristics of both the predator and prey. Preliminary studies of darter food habits in Flint Creek, Oklahoma, suggested a differ- ence in the magnitude of dietary overlap, de- pending on the size classes of fishes com- pared. Although Smart and Cee (1979) separated fish into age groups, most studies have reported indices of overlap without sub- 'Study supported in part by National Science Foundation Grants #DEB 78-12.56.5 and #BSR 8308422 and the Faculty Research Fund of NTSU. "Department of Biological Sciences, North Texas State University, Denton, Texas, 76203. 721 722 Great Basin Naturalist Vol. 45, No. 4 dividing the population into size classes, pos- sibly overlooking some very important as- pects of darter ecology. At times during the life histories of game fishes there may be competition with noneco- nomic fishes for the available food resources. To better comprehend these relationships, it is necessary to understand the resource re- quirements of the noneconomic species. In Flint Creek, Delaware County, Oklahoma, insectivorous fishes that are in potential com- petition with the primary game fish, the smallmouth bass (Micropterus dolomieui La- cepede), are the orangethroat darter (Etheo- stoma spectabile Agassiz), the stippled darter (Etheostoma punctulatum Agassiz), the banded darter {Etheostoma zonale Cope), the fantail darter {Etheostoma flahellare Rafi- nesque), the greensided darter {Etheostoma blennioides Rafinesque), and the banded sculpin (Cottus carolinae Gill). The food habits and life histories of many darters and sculpins have been studied (Koster 1937, Dineen 1951, Bailey 1952, Braasch and Smith 1967, Page and Smith 1970, 1971, Scalet 1972, Novak and Estes 1974, Page 1974, Flynn and Hoyt 1979, Cordes and Page 1980). Neither the food habits of the orangethroat darter, the stippled darter, or the banded sculpin nor the relation- ships between these species have been re- ported. Therefore, the objectives of this study were: (1) to describe the food habits of the three most abundant insectivorous fishes in Fhnt Creek, Oklahoma, E. spectabile, E. punctulatum, and C. carolinae; (2) to deter- mine the magnitude of dietary overlap be- tween various sizes of the two darter species and the banded sculpin; (3) to conduct field and laboratory studies to determine if the darters and sculpins of each size class are se- lecting prey items. Materials and Methods Study AREA. — Flint Creek is a third order stream running southeasterly through Dela- ware County, Oklahoma, to its confluence with the Illinois River. The study area was a 1-km section approximately 4-km upstream from Oklahoma Highway 33. At this point the stream passes through an open valley with occasional riparian trees. The stream consists of many pools separated by as many riffles. The substrate ranges from detritus and silt to gravel and rubble, with gravel being most common. During summer months, a large portion of the stream becomes covered with water willow {Justicia americana) and water primrose {Ludwigia spp.). Habitat usage. — Fishes were captured monthly from March 1983 to February 1984 using a kick-net placed downstream of habi- tats to be sampled, with the substrate dis- turbed to dislodge fish. A total of 50 kick-sets were made monthly in each of four general habitats: (1) open pools, (2) open riffles, (3) pools with submerged and/or emergent cover, and (4) riffles with submerged and/or emergent cover. The standard length in mm was measured for each fish. An analysis of co-dispersion was made incorporating a log- likelihood ratio test. Although sampling effi- ciencies may have differed in each habitat, an AN OVA was run to test for differences in the average length of each species between the four habitats followed by a Student-Newman- Keuls multiple comparison. Stomach analysis. — In habitats where two or more species were captured, five fish in as many 10-mm size classes as possible were pre- served in 10% formalin for stomach analysis. Roberts and Winn (1962) and Daughtery et al. (1976) found that visual stimuli were neces- sary for normal feeding responses of darters. Several researchers have found darters to ex- hibit diurnal feeding patterns, with peak feed- ing occurring in midday or afternoon, with maximum gut content later in the day (Mathur 1973, Schenec and Whiteside 1977, Lavzer and Reed 1978, Cordes and Page 1980, Matthews et al. 1982). Preliminary studies indicated the two darter species under study both fit this pattern; therefore, all fish were captured between 12:00 noon and 5:00 p.m. Stomach contents of the preserved darters were identified to the lowest possible taxon (genera for most insects). The head capsule width (HCW) of insects or the maximum di- ameter of other taxa was measiu-ed with an ocular micrometer. Food availability. — Within the same habitats where darters were collected for stomach analyses, three 0.05 m" Hess samples were taken at the same monthly sampling times and preserved in 70% isopropanol for October 1985 Todd, Stewart: Oklahoma Fish 723 determination of prey availability. Additional benthic organisms were collected, and their HCW or maximum diameter measured. These organisms were then killed in 4% for- malin, immediately dried for 24 hours at 105 C, and their dry weight was measured to the nearest 0. 1 mg. Regression lines were devel- oped from these data and were used to esti- mate the live dry weight of the prey con- sumed. Dietary overlap. — Dietary overlap was determined with an index of association CX (Horn 1966) which is calculated: 22 X, CX Sx, i r y, = where: total number of food taxa proportion of the total diet of species X taken from taxa i proportion of the total diet of species V taken from taxa i This index ranges from 0.0 to 1.0, with a value of 0.0 indicating no overlap, and 1.0 indicating complete overlap. An index of overlap was calculated for both numbers and biomass of each prey taxa. Prey selection. — A linear index of food selection L (Strauss 1979) was used. It is calcu- lated: Li = !"( - p, where: r; = the relative abundance of item i in the gut Pi = the relative abundance of item i in the environment This index ranges from - 1.0 to 1.0, with posi- tive values indicating preference and negative values indicating avoidance and/or inaccessi- bility. A Students t-test was used to test the null hypothesis of no difference in L from zero. An index of selectivity assesses differences in the proportions of a specific prey item in the diet relative to proportions available in the environment. These indices may or may not express actual election or avoidance of a prey item. The real availability of prey items is usually not known; therefore, unavailability may be mistaken as avoidance. The following experiment was conducted in June 1984 to gain some further insight into selectivity by darters in a situation where no prey had protective cover. Five 18.9-1 aquaria were placed in the stream to minimize stress on the darters during the experiment. A selected group of prey, representative of the stream population and in sufficient numbers (n = 70) to allow feeding without greatly affect- ing the proportions of the prey, were counted into the aquaria. Fish were placed in the aquaria after dark as follows: (1) two mature E. spectabile in each of three aquaria and (2) two mature E. punctidatiim in each of two aquaria. This allowed adjustment to confine- ment prior to dawn, when normal feeding should begin. The darters were allowed to feed until noon, when they were preserved for stomach analysis. Size selectivity. — Size selectivity was as- sessed by comparing mean widths of prey consumed by each species in each size class. A one-way ANOVA followed by a Student-New- man-Keuls multiple range test was used to test for differences in mean widths of prey consumed between different size classes of fishes. Results and Discussion Seasonal abundance. — A total of 992 darters and sculpins were captured over the 12-month study period. Etheostoma spectabile was most abundant (n = 543), fol- lowed by Cottus carolinae (n ^275), E. punc- ttilatiim (n=157), and£. zonale (n = 17). Their seasonal abundance from March 1983 to February 1984 is illustrated in Figure 1, and numbers of £. spectabile and C. carolinae generally followed expected survivorship, ex- cept in August when dense macrophytic vege- tation may have impaired sampling efficiency. These two species are spring and summer spawners, and their numbers were greatest, as expected, during these periods of recruit- ment. Etheostoma zonale was low in abun- dance over the entire year, with the greatest number captured (n = 6) occurring in August. The efficiency of seining as a method of sam- 724 Great Basin Natur.\list Vol. 45, No. 4 100 Q 90 UJ RO ir '-J h- n 70 < u 60 T (f) li bU O 40 E. spectabile E. punctulatum -• E. zonale o o ■ C. carolinae ° ° III 30- en \ / ^ -) 20- V D "V. 10- M Fig. 1. Sea.sonal abundance oi the lour mo.st abundant nonecononiic insectivorous fishes in Fhnt Creek, Oklahoma, March 198.3 to February 1984. pling darters and sculpins is questionable, be- cause of their ability to avoid capture by hid- ing under rocks and in crevices. The use of kicknets in this study probably gave a more discrete sample. Habitat USAGE. — An analysis of co-disper- sion for £. spectabile and E. punctulatum re- vealed no significant habitat segregation. Sig- nificant (p<0.05) positive co-dispersion (overlap) was observed in June and Septem- ber 1983. All values for co-dispersion were positive except in November 1983 and Janu- ary 1984, when numbers collected were low (19 and 13, respectively). Areas sampled can greatly affect the results of an analysis of co-dispersion. Small sampling areas tend to give negative values, whereas large sampling areas tend to result in positive values. Kick-netting resulted in a large per- centage of kick-sets captiuing no fish. The positive values for co-dispersion and a low capture rate suggests that the two darters did not separate themselves on the basis of habitat selection alone. Similar analyses with C. carolinae and both darters showed relatively more habitat segre- gation between mature banded sculpins and both darters during spring (March, April), with significant negative co-dispersion be- tween C. carolinae and £. punctulatum in April. Significant (p<0.05) overlap was shown between C. carolinae and £. spectabile in May, coinciding with the hatching and re- cruitment of both species. Mature C. caroli- nae were absent from the four habitats sam- pled from May to September, when juvenile fish were abundant, probably reducing in- traspecific and interspecific competition. On a yearly basis, an ANOVA revealed a significant difference in the average body lengths of both E. spectabile and E. punctula- tum between habitats (p<().()001), whereas no significant difference in the mean lengths of C carolinae were found. A Student-New- man-Keuls multiple comparison showed E. punctulatum with a significantly larger mean length in riffles with submerged and/or emer- gent cover than in open pools, 62.3 and 43.5 nun, respectively (p<().()5). Etheostoma spectabile had a significantly higher mean length in riffles with and without submerged and/or emergent cover (51.6 and 50.3 mm. October 1985 Todd, StewarT: Oklahoma Fish Etheostoma spectabile (<41nnnn 725 100 80-' CD ^^ g 60 m 50 2 40 LlI ^ O ?T. 30 20' ■ ^ M ^ wm ^1 ?^ iSl <^ FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC JAN FEB D CHIRONOMIDAE LARVAE BAETIS STENONEMA CLADOCERA EPHEMERELLA ASELLUS M COPEPODA r™l LEPTOPHLEBIA El OSTRACODA AMPHIPODA MISCELLANEOUS Fig. 2. Food habits oiEtheostoma spectabile <41 mm February 1983 to February 1984. In this figure and Figures 3-6, each bar equals the percent by number of each selected prey item, numbers above bars are equal to the ratio of fed to total guts examined, numbers within bars are equal to the percent by dry weight of food items, and numbers right of major food categories are equal to the Strauss Selectivity Index. Asterisks with Strauss Index are equal to significance (p<0.05). respectively) than in pools with and without submerged and/or emergent cover (42.3 and 40.8 mm, respectively) (p<0.05). Although not significant, both species were larger in riffles with submerged and/or emergent cover and pools with submerged and/or emergent cover than in open riffles and open pools, respectively. The presence of different sizes of fishes in different habitats may be impor- tant in the reduction of intraspecific competi- tion. Since roughly similar size classes of the two darters were found in similar habitats, the segregation of habitat alone by different size classes probably had little effect on inter- specific competition; however, the combina- tion of habitat selection, temporal segregation of spawning times, prey selection, and differ- ences in prey size probably resulted in low- ered competition between £. spectabile and £. punctulatum. Food Habits. — In both darters a major di- etary shift could be related to maturation, and therefore separation of fish sizes for discus- sion of food habits was made on this basis. Etheostoma spectabile and E. punctulatum were therefore respectively separated into <41 and >40, and <51 and >50 mm sizes (Figs. 2-5). The stomachs of 244 £. spectabile con- tained 35 different prey taxa. Both juveniles and adults fed heavily on chironomids during winter and early spring (January- March, Figs. 2,3). Chironomids continued to consti- tute 16. l%-59.5%, by dry weight, of juvenile fish diets in April to July, as they increased feeding on microcrustaceans and small mayflies (Fig. 2). This shift to planktonic prey items coincided with recruitment of darter fry. In August, diets of these smaller fish were exclusively composed of microcrustaceans, particularly Cladocera and Copepoda, and for the rest of the year microcrustaceans and am- phipods continued to make up the major per- centages by number and dry weight, into De- cember (Fig. 2). Mayfly nymphs {Baetis, Stenonema, Ephemerella, Leptophlebia, and Caenis) made up 49.1%-77.3%, by dry weight, of mature E. spectabile diets in spring 726 Great Basin Naturalist Etheostoma spectabile (>40nnm) Vol. 45, No. 4 FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC JAN FEB CHIRONOMIDAE LARVAE BAETIS STENOMEMA CLADOCERA EPHEMERELLA ASELLUS COPEPODA LEPTOPHLEBIA CAENIS rn OSTRACODA ^553 AMPHIPODA I 1 STENOMEMA F!^ Fig. 3. Food habits oi Etheostoma spectabile >40 mm February 1983 to February 1984 MISCELLANEOUS Etheostoma punctulatum (<51nnnn) 100 90 80 70 60 50 40 30 20 10 5/5 0/0 M m ^ i m 26 6 0' ^^M " m i 1 1 I i FEB MAR APR I CHIRONOMIDAE I LARVAE I BAETIS STFNONEMA MAY JUN JUL g CLADOCERA m EPHEMERELLA ASELLUS AUG SEP OCT mm COPEPODA M LEPTOPHLEBIA mi CAENIS NOV □ I 1 STFNONEMA F^ Fig. 4. Food habits oi Etiicostoina punctulatum <51 mm February 1983 to Fcbruar\ 1984 DEC JAN FEB OSTRACODA AMPHIPODA MiSl,ELLANEOUS October 1985 Todd, Stewart: Oklahoma Fish Etheostoma punctulatum (>50mm) 727 FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DEC JAN CLADOCERA EPHEMERELLA ASELLUS CHIRONOMIDAE LARVAE BAETIS r^ OSTRACODA ^^ AMPHIPODA I I STENONEMA p| ASELLUS mm CAENIS H MISCELLANEOUS Fig. 5. Food habits o( Etheostoma punctulatum >50 mm February 1983 to February 1984. COPEPODA LEPTOPHLEBIA CAENIS and early summer until June (Fig. 3), corre- sponding with the emergence and recruit- ment of these insects. They were ingested at lower levels for the rest of the summer, and none were eaten by 12 fish in October (Fig. 3). Unhke juveniles, mature fish fed on chirono- mids at levels less than 11% by dry weight from April to July, and microcrustacea were not present in mature fish guts in late spring and summer in amounts above 3.1% by dry weight (June, August). Amphipods appeared in diets of mature orangethroat darters at lev- els of 42.5% and 91.0% (dry weight), respec- tively, in September and October, when in- gestion of amphipods generally decreased and chironomid feeding increased in November and December until winter, when chirono- mids again predominated in guts. Mature fish relied more heavily on larger items such as Baetis, Ephenierella, Leptophelebia, Ste- nonema, and amphipods as these items were available. Stomachs of 205 E. punctulatum contained 30 different prey taxa. Both classes of £. punc- tulatum utilized chironomids in the coldest months, January and February (Figs. 4, 5), but relied much less on them in those months, and generally throughout the year, than did E. spectabile {Figs. 2, 3). Feeding by £. punc- tulatum during these months was less concen- trated on chironomids than in E. spectabile, in favor of diverse feeding on mayflies, Asel- lus, and amphipods (Figs. 4, 5). Few juvenile E. punctulatum were cap- tured in spring (March-May). The five fed fish taken in March fed almost exclusively on the mayfly Leptophlebia (94.6% by dry wt.), apparently segregating themselves from small E. spectabile that spread their feeding pre- dominately over chironomids (24.3% by dry wt.), Ephenierella (18.1%), and Lep- tophlebia (21.2%). During summer months (June- September) and into November, juve- nile stippled darters fed mainly on amphipods and the mayflies Stenonema , Baetis, and Lep- tophlebia, further suggesting segregation of themselves from smaller E. spectabile that were feeding more on microcrustaceans dur- ing that time. The single fed immature fish captured in December had only Asellus in the gut (Fig. 4). Mature E. punctulatum showed little pattern in feeding, except in January to March, when chironomids, Leptophlebia, and Ephemerella were ingested in high vol- 728 ICO 90t 80 g60 m 50 §40 U ^30 Q- 20 10 Great Basin Naturalist Cottus carolinae Vol. 45, No. 4 (<71mnn) (=>70mnn) FEB MAR- MAY JUN JUL AUG" OCT NOV DEC JAN FEE APR SEP t CHIRONOMIDAE LARVAE BAETIS CLADOCERA EPHEMERELLA ASELLUS ™ COPEPODA fTl LEPTOPHLEBIA I I STENONEMA F^H Fig. 6. Food habits o(Cottus carolinae February 198.3 to February 1984. FEB MAR ADR MAY- JAN pn OSTRACODA ^^ AMPHIPODA ^m MISCELLANEOUS umes (Fig. 5). During these months larger E. spectabile were feeding predominately on chironomids and the mayflies Ephemerella and Leptophlehia . Ingestion of Stenonema and miscellaneous organisms was greater and more seasonally distributed (Fig. 5) than in larger E. spectabile (Fig. 3), indicating parti- tioning of available food resources by the ma- ture fish of both species. The seemingly er- ratic pattern of prey ingestion (Figs. 4, 5) was generally closely related to the relative abun- dance of the various prey in the environment. A few taxa which were not consumed in sufficient numbers to be important items in the annual diet of both darter species com- posed a large percentage of the diet in months when these items were available. In March 1983 fish eggs made up 25.6%, by dry weight, of the diet of juvenile E. spectabile. In mature fish the following prey made up an important part of the diet: Februarv 1983, Prostoia (12.8%); April 1983, fish eggs (18.3%), Isop- erla (5.1%); August 1983, isomjchia (35.4%). The diet of smaller E. punctulatiim contained such important prey as: Julv 1983, Neoperla (16.2%), Stenacron (9.8%); August 1983, Conjdalus (61.4%); September 1983, Lepto- hijphes (5.2%). The following items were im- portant at times in the diet of mature E. punc- tulatum: May 1983, Isomjchia (39.1%), oligochaeta (27.3%), Psephenus (8.2%), Acroneuria (15.2%); June 1983, decapoda (20.8%), AcroneanV/ " (38.6%), Conjdalus (16.4%); July 1983, Neoperla (17.9%), Psephenus (27.9%), decapoda (18.8%); August 1983, Conjdahis (47.9%), decapoda (34.2%), Psephenus (13.7%); September 1983, Psephenus (42.4%); October 1983, Argia (15.2%), Conjdalus {78.2%h November 1983, Psephenus (86.5%); December 1983, oligochaeta (94.3%). Two size classes of sculpins were selected with similar reasoning as with the two darter species. These size classes were <71 and >70 mm (Fig. 6). During the initial months of the study (Feb- ruary-April), it was not known whether suffi- cient sculpins could be sampled for food habits analysis; therefore, young sculpins were returned to the stream. In the fall, fish were collected for stomach analysis, and 42 specimens were taken during the period Oc- October 1985 Todd, StewarT: Oklahoma Fish 729 tober- February (Fig. 6). Mature fish were not collected in the study area of the stream during May-September, and very few were captured until February, possibly because of migration into unsampled habitats such as deep pools. Therefore, food habits were de- termined from a relatively small sample of nine fish during February-April 1983 and February 1984 (Fig. 6). While studying the movements of sculpins, Bailey (1952) found that, although they usually remained within a radius of 46 m, some moved up to 144 m from the point of initial capture. Deep pools were always within this range in the study area. Stomachs of 53 C. carolinae contained 18 prey taxa. Cladocerans, copepods, and ostra- cods were absent in the diets of both size classes of C. carolinae, suggesting segregation of this species from juvenile £. spectabile dur- ing August- December, when the latter fed heavily on these items. Both size classes of C. carolinae fed on EphemereUa (47.9% and 38.0%, by dry weight) during February 1983, as did mature E. spectabile (Figs. 3, 6). Ma- ture C. carolinae continued to feed on EphemereUa through April, when no juve- niles were collected. At this time, mature E. spectabile and E. piinctulatum were observed feeding on the mavflv EphemereUa (Figs. 3, 5). In May and July the three small fish col- lected fed on chironomids (May, 20.0% by dry weight) and small mayflies, Stenonema (Mav, 80.0%), and Leptophlebia (July, 53.7%) (Fig. 6); in July Hydroptila was also taken (46.3%). During October and November small mayflies such as Baetis (0.7% and 15.8%, re- spectively, by dry weight) were taken, with large numbers oi Psephenus (2.5%), decapods (63.2%), and Conjdalus (29.8%) in October. In November Conjdalus (59.0%) and Psephemis (16.6%) made up a large percent- age of the diet. Heavy feeding by small C. carolinae on chironomids was observed in January (56.0%), increasing to 97.5%, by dry weight, in February (Fig. 6). Although the larger size class consumed some chironomids in this month, Asellus made up 91.3% of the diet of the mature fish (Fig. 6). Only mature E. piinctulatum also fed on Asellus at this time in amounts higher than 2 percent (Fig. 5). Dietary OVERLAP. — Monthly values for di- etary overlap between different size classes of darters and sculpins were highly variable, ranging from 0.00 to 0.99 for numeric data and 0.00 to 0.97 for dry weight data (Table 1). Based on numeric data, the lowest average monthly overlap value was between imma- ture E. spectabile and mature E. punctulatum (0.22), whereas the highest average monthly overlap value was between larger E. spectabile and juvenile E. punctulatum (0.45). Variations in monthly overlap values may be attributed to differences in prey availability, gape size differences, or differ- ences in prey size. No significant correlation between overlap values and prey availability, or differences in gape were shown; however, a significant (p<0.0005) correlation between the difference in mean prey sizes of the groups being compared and overlap values was found. This indicates that prey size selection is important in regulating possible competi- tion for food between these insectivorous fishes in Flint Creek. Indices of dietary overlap between the two size classes of £. spectabile were significantly correlated (p<0.05) to the difference in the mean size of prey selected by the two groups. Although mean prey size selection may be important in reducing intraspecific competi- tion for E. spectabile, the indices of dietary overlap between the two size classes of E. punctulatum were not correlated with the dif- ferences in the mean prey size selected by the two size classes. This suggests that the regula- tion of dietary overlap between the two size classes of E. punctulatum is dependent on some factor besides prey size, such as actual election of specific prey items. Dietary overlap was generally high (.81 -.99) only between small banded sculpins and both classes of the orangethroat darters in January- February 1984, since both species fed heavily on chironomids (Table 1). Overlap values between the banded sculpin and E. punctidatum were generally low. A value of .96 by number, .93 by dry weight (Table 1), however, was found in January between juve- nile C. carolinae and mature E. punctulatum, when both species were feeding on chirono- mids. Size selectivity. — The selection of prey on the basis of size is important in the parti- tioning of Flint Creek food resources by or- angethroat and stippled darters. During most 730 Great Basin Naturalist Vol. 45, No. 4 Table 1. Coefficients of dietary overlap of different size classes of Etheostoma spectabile, E. punctulatum , and Cotttis carolinae from February 1983 to February 1984. Upper values are based on numeric data. Lower values are based on dry weights data. Comparison Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb E. spectabile <41 mm .33 .01 .40 ..58 .84 .16 .03 .00 .17 .87 E. punctulatum <51 mm .37 .00 .49 .02 .97 .88 .63 .00 .04 .55 E. spectabile <41 mm .20 .19 .01 .05 .16 .00 .13 .07 .04 .00 .92 .61 E. punctulatum >50 mm .40 .48 .00 .09 .06 .00 .20 .08 .04 .00 .76 .17 E. spectabile >40 mm .29 .38 .03 .75 .27 .46 .88 ..35 .00 .21 .85 E. punctulatum <51 mm .28 .53 .05 .25 .12 .80 .84 .55 .00 .13 .63 E. spectabile >40 mm .47 ..38 .53 .18 .63 .22 .25 .44 ..34 .06 .04 .98 .58 E. punctulatum >50 mm .43 .67 .44 .12 .25 .20 .01 .15 .06 ,00 .85 .87 .21 E. spectabile <41 mm .94 .42 .73 .08 ..55 .01 .22 .29 .44 .62 .97 .99 E. spectabile >40 mm .78 .76 .18 .51 .60 .03 .77 .94 .47 .07 .92 .96 E. punctulatum <51 mm .48 ..37 .31 .46 .10 ..33 .40 .15 .48 .25 .75 E. punctulatum >50 mm .49 .13 .04 .41 .69 .15 .06 .05 .03 .,33 .27 C. carolinae <71 mm .12 .03 ,01 .02 .94 .96 £. spectabile <41 mm .23 .04 .00 .02 .81 .95 C. carolinae <71 mm .55 .12 .04 .01 .33 .98 .99 E. spectabile >40 mm .67 .14 .04 .00 .06 .87 .90 C. carolinae >70mm .17 .09 .50 E. spectabile <41 mm .31 .07 — .04 C. carolinae >70 mm .64 .32 .43 .50 E. spectabile >40 mm .74 .54 .12 .06 C. carolinae <71 mm .89 .44 C. carolinae >70mm .87 .03 C. carolinae <71 mm .06 .25 .01 .04 .21 .77 E. punctulatum <51 mm .03 .23 .00 .01 .15 .42 C. carolinae <71 mm .15 .43 .00 .05 .32 .96 .50 E. punctulatum >.50 mm .06 .10 .00 .42 .25 .93 .11 C. carolinae >70 mm .32 .01 .55 E. punctulatum <51 mm .25 .00 .04 C. carolinae >70 mm .22 .85 ..38 .80 E. punctulatum >50 mm .08 .96 .05 .36 months there were significant differences (p<0.05) in the mean prey widths selected by E. spectabile and E. punctulatum (Table 2). Also, during many months there were signifi- cant differences in prey widths selected by the different size classes of each darter species (Table 2.). Differences in morphology may be partially responsible for these differences. Etheostoma punctulattwi has the larger gape of the two species. However, in both species the mean prey size selected was not signifi- cantly correlated with gape size (p>0.05). In months when C. carolmae were sampled for stomach analysis, the size of prey selected by this species was generally similar to that con- sumed by immature E. punctulatum . During January and February 1984, when both C. carolinae and E. spectabile were electing chi- ronomids, prey size for these two groups were similar, possibly indicating a period of higher competition. Selectivity', — For each class of fishes, monthly electivity for particular prey items were variable; however, some trends are evi- dent (Figs. 2-6). All classes of fishes avoided Stenelmis. A preference for specific prey items coincided with months when those prey items were abundant. Etheostoma spectabile generally preferred smaller items such as cladocerans and chironomids, whereas E. punctulatum generalK- preferred larger items such as mayflies, Conjdalus . decapods, and oligochaetes. Cottus carolinae showed nega- tive selection for microcrustaceans such as os- October 1985 Todd, Stewart: Oklahoma Fish 731 Table 2. Mean prey size (mm) of difi'erent size classes oi Etheostoma spcctahile, E. punctulatum, and Cottiis carolinae from February 1983 to February 1984. Month E. spectabile E. spectabile C. carolinae C. carolinae E. punctulatum E. punctulatum immature mature immature mature immature mature Feb 0.74 1.40 1.14 1.45 1.09 Mar 0.54 0.67 1.29 1.31 1.35 Apr 0.28 0.92 1.47 1.38 May 0.11 0.36 0.98 2.30 Jun 0.13 0.68 — 0.84 1.49 Jul 0.23 0.55 0.55 0.93 2.14 Aug 0.07 0.65 0.83 2.73 Sep 0.21 0.35 0.36 1.66 Oct. 0.19 0.58 1.33 0.80 1.37 Nov 0.08 0.25 1.41 0.82 2.63 Dec 0.05 0.19 1.15 1.47 Jan 0.24 0.39 0.36 0.81 0.52 Feb 0.32 0.38 0.30 1.21 0.54 0.89 tracods and cladocerans, preferring mayflies, oligochaetes, and decapods. Several researchers have suggested that changes in the diets of darters occur in re- sponse to changing prey densities (Braasch and Smith 1967, Page and Smith 1971, Page and Burr 1976, Schenck and Whiteside 1977). The results of this study support this conclu- sion. However, changes in either morphology or behavior, or both, as darters mature may also contribute to shifts in the diet. For exam- ple, the greater and more seasonally dis- tributed ingestion of Stenonema and other miscellaneous organisms by mature E. punc- tulatum (Fig. 5) than by mature E. spectabile (Fig. 3), and other instances of positive dec- livity (Figs. 2-5), cannot be totally rational- ized on the basis of those food items becoming more available at certain times. In-stream aquaria experiments. — A group of natural prey, totaling 70 individuals in each aquarium, was used for assessing darter selectivity under the condition of no protective cover. Specific potential prey and their predetermined compositions (by num- ber) were Cheumatopsyche (14.3%), Atherix (28.6%), Neoperla (28.6%), Stenelmis larvae (7.1%), Psephenus (7.1%), Isonychia (7.1%), and Stenonema (7.1%). Under these condi- tions in June 1984, prey selection by the ma- ture fish of each species generally corrobo- rated selectivity indices calculated from field studies in June 1983. Positive election for mayflies and negative selection for Psephenus continued. Both species continued to avoid Stenelmis, and none were consumed. Atherix, which was not abundant in the stream in June 1983, was also completely avoided. Etheostoma spectabile had a higher selectivity index for Cheumatopsyche (.11) and Neoperla (.13) than in the field studies (.01 and -.00, respectively). Etheostoma punctulatum also showed a higher index for Neoperla (.21) than in the field (-.00). The inaccessibility oi Cheumatopsyche and Neop- erla , because of their cryptic nature and the fact that the experimental array of prey was not exactly the same as in the field, may have resulted in altered electivities. Conclusions Several factors influence the degree of over- lap in the diet of cohabiting fishes. Differ- ences in the diets of £. spectabile, E. punctu- latum, and C. carolinae in Flint Creek, Oklahoma, are related to a complex of factors, including the utilization of different habitats, prey size selection, and the selection of specific prey items. The absence of mature C. carolinae during the summer and fall suggests their movement into unsampled areas, thereby exhibiting habitat partitioning. Analyses of co-dispersion further revealed partitioning of habitat be- tween the banded sculpin and both darters during months when mature sculpins were captured in the study area. Habitat partition- ing probably reduces both intraspecific and interspecific competition . An analysis of co-dispersion between the two darters showed no habitat segregation: however, different lengths of both darter spe- cies were found in different habitats. This not 732 Great Basin Naturalist Vol. 45, No. 4 only may reduce intraspecific competition, but, together with temporal segregation of spawning times, may reduce interspecific competition. The selection of prey on the basis of size was found to be an important factor in explaining overlap in the diets of insectivorous fishes. Other researchers have shown prey size to be consistent with gape (Northcote 1954, Daiber 1965, Matthews et al. 1982), and the results of this study support this. However, no correla- tions between mean prey size and gape were found. No significant (p<0.05) correlation was found between differences in gape and dietary overlap between corresponding fish groups. This suggests that gape may be important in prey size range, but other ethological or mor- phological factors are involved in the determi- nation of the degree of dietary overlap be- tween species. The selection of prey on the basis of size may reduce both intraspecific and interspecific competition. Although differ- ence in dietary overlap between age groups of E. spectabile are correlated with differences in mean prey size, differences in dietary over- lap between age groups of E. punctidatum and mean prey size are not. Etheostoma spectabile generally fed upon smaller prey than E. punctulatum. Although the selection of prey items may be due to the election of a specific prey size, some selection of specific prey items regardless of size was noted. Both species selected prey at times when the particular item was abundant. The management of stream game fishes is dependent on a knowledge of the species to be managed and their potential competitors. The resources available for management are de- pendent on the entire food web in the stream, and the resource requirements of noneco- nomic fishes are often overlooked, obscuring an important factor in total stream manage- ment. This study reveals food and potential habitat resource use by, and interactions be- tween, nongame fishes in an Ozark stream. These findings not only provide new basic knowledge for these fishes, but they should be helpful reference data for future assessments of resource use, partitioning, and consequent development of management strategies for stream fishes. Acknowledgments We thank M. R. Ernst, who accompanied the authors on sampling trips and assisted in collections. Literature Cited Bailey, J E 1952. Life history and ecology of the sculpin Cotttis bairdi punctulatiis in southwestern Mon- tana. Copeia 1952: 243-255. Baker, J A , and S T Ross 1981. Spatial and temporal resource utilization by southeastern cyprinids. Copeia 1981: 178-189.' Braascu,ME AND P. W.Smith. 1967. Life history of the slough darter, Etheostoma gracile (Pisces, Per- cidae). Illinois Natur. Hist. Surv., Biol. Notes 58: 1-12. CoRDES, L. E, andL. M. Page. 1980. Feeding chronology and diet composition of two darters (Percidae) in the Iroquois River system, Illinois. Amer. Mid. Natur. 104: 202-206.' Daiber. F. C 1956. A comparative analysis of the winter feeding habits of two benthic fishes. Copeia 1956: 141-151. Daughtertv, C H , L B Daughtery. and A. P Blair. 1976. Visual and olfactory stimuli in the feeding behavior of darters (Etheostoma) inhabiting clear and muddy water. Copeia 1976: 380-382. Dineen, C F. 1951. A comparative study of the food habits of Cotttis bairdi and associated species of Salmonidae. Amer. Mid. Natur. 46: 640-645. Flynn, R B , AND R D HoYT 1979. The life history of the teardrop darter, Etheostoma barbouri Kuehne and Small. Amer. Mid, Natur. 101: 127-141. Cause, G F. 1934. The struggle for existence. Williams and Wilkins Co., Baltimore. Horn, H S 1966. Measurement of "overlap" in compara- tive ecological studies. Amer. Natur. 100: 419-424. KosTER, W J 1937. The food of sculpins (Cottidae) in central New York. Trans. Amer. Fish. Soc. 66: 374-382. Layzer, J B., AND R J Reed. 1978. Food, age and growth of the tessellated darter, Etheostoma obmtedis, in Massachusetts. Amer. Mid. Natur. 100: 459-462. M.'VTHUR, D 1973. Food habits and feeding chronology ol the blackbanded darter, Percina nigrofaciatu (Agassiz), in Halawakee Creek, Alabama. Trans Amer. Fish. Soc. 102: 48-55. M.\TTHEVVS, W. J.,J. R. Bek, andE. Sur-vt. 1982. Compar- ative ecology of the darters Etheostoma podoste- mone, E. flabellare, and Percina ronaoka in the upper Roanoke River drainage, N'irginia. Copeia. 1982: 805-814. Mendklson, J 1975. Feeding relationships among spe- cies oiNotropis (Pisces: Cyprinidae) in a Wiscon- sin stream, Ecol. Monogr. 45: 199-230. MoYLE P B 1973. Ecological segregation among three' species of minnows (Cyprinidae) in a Minnesota | lake. Trans, Amer, Fish, Soc, 102: 794-805. I October 1985 Todd. Stevvart: Oklahoma F'ish 733 NoRTIlCOTE, T G. 1954. Observation.s on the comparative ecology of two species of fish, Cottus aspcr and Cottus rhotheus, in British Cohimbia. Copeia 1954: 25-58. Novak, J K . and R D Estes. 1974. Snnimer food habits of the black sculpin, Cottus baileiji, in the Upper South Fork Holston River Drainage. Trans. Amer. Fish. Soc. 103: 270-276. Page. L. M. 1974. The Ufe history of the spottail darter, Etheostoina sqiiamiceps , in Big Creek, lUinois, and Ferguson Creek, Kentucky. lUinois Natur. Hist. Surv., Biol. Notes 89: 1-20. Page.L. M.andPW Smith. 1970. The hfe history of the dusky darter, Percina sclera, in the Embarras River, Illinois. Illinois Natur. Hist. Surv., Biol. Notes 69: 1-15 1971. The life history of the slenderhead darter, Percina phoxocephala , in the Embarras River, Illinois. Illinois Natur. Hist. Surv., Biol. Notes 74: 1-14. Roberts. N. J., and H. E. Winn. 1962. Utilization of the senses in feeding behavior of the johnny darter, Etheostoma nigrum. Copeia 1962: 567-570. SCALET, C. G. 1972. Food habits of the orangebelly darter, Etheostoma radiosum cijanorum (Osteichthyes: Percidae). Amer. Mid. Natur. 87: 515-522. SCHENEC. J. P.. AND B V. WiiiTESiDE. 1977. Food habits and feeding behavior of the fountain darter, Etheostoma fonticola. Southwest. Natur. 21: 487-492. Smart. H. S . and J H Gee 1979. Coexistence and re- source partitioning in two species of darters (Per- cidae) Etheostoma nigrum and Percina maculata. Canadian J. Zool. 57: 2061-2071. Strau.ss, R. a. 1979. Rclial)ility estimates for Ilev's elec- tivity index, the forage ratio, and a proposed linear index of food selection. Trans. Amer. Fish. Soc. 108: 344-352. SURAT, E M , W. J Matthews, and J. R. Bek. 1980. Com- parative ecology ofNotropis albeolus, N. ardens, and N. cerasinus (Cyprinidae) in the upper Roanoke River drainage, Virginia. Amer. Mid. Natur. 107: 13-24. Werner. E. E , and D J Hall 1976. Niche sifts in sun- fishes: experimental evidence and significance. Science 191: 404-406. Wynes, D. L. andT. E. Wissing. 1982. Resource sharing among darters in an Ohio stream. Amer. Mid. Natur. 107: 294-304. Zaret.T. M., and a. S. Rand. 1971. Competition in tropi- cal stream fishes: support for the competitive ex- clusion principle. Ecology 52: 336-342. CHECKLIST OF VASCULAR PLANTS FOR THE BIGHORN CANYON NATIONAL RECREATION AREA, WYOMING AND MONTANA Robert W. Lichvar', Ellen I. Collins', and Dennis H. Knight" Abstract. — The checklist of vascular plants of Bighorn Canyon National Recreation Area is presented based on field collections and herbarium specimens. The checklist treats 656 taxa from 73 families. Bighorn Canyon National Recreation Area (BCNRA) is located near the northern end of the Bighorn Mountains in Wyoming and Montana, and occupies 25,911 ha (64,000 acres). The vegetation ranges from desert shrubland to lower montane forest. Previous to this study, the area had only been collected infrequently by several botanists. Therefore, the lack of knowledge about this area warranted further botanical collections. The objectives of this study were threefold: (1) to conduct an intensive inven- tory of the flora of BCNRA, (2) to bring to- gether existing herbarium data on the plants of the area, and (3) to compile a checklist of the flora. Methods The fieldwork was conducted during the summer of 1983, with collecting trips taking place monthly from May through August. Ap- proximately 1750 collections, in triplicate, were made from throughout the BCNRA. One set of specimens has been deposited in the Rocky Mountain Herbarium (Department of Botany, University of Wyoming) and an- other set is at the BCNRA Visitor Center (Lovell, Wyoming). Specimens were identified during the au- tumn of 1983. To ensure proper identifica- tion, most specimens were compared to mate- rial in the Rocky Mountain Herbarium (RM). In addition to field collections, several herbaria and knowledgeai)le botanists in the region were consulted for information. The RM and two small herbaria at BCNRA were checked for species location data, and Dr. Robert Dorn, who has made many collecting trips to the area, provided a list of species he , had collected there. Major Floristic Elements The flora of BCNRA is composed of the Rocky Mountain, Great Basin, and Great Plains floristic elements (Porter 1962, Dorn 1977). The Great Basin element is the major component of the flora, which includes the following prominent species: Juniperus os- teosperma, Artemisia spinescens, Atriplex confertifolia, Cercocarpus ledifolius, Te- tradymia spinosa, Sarcobatus vermiculatus, and Agropijron spicatum. This element is ap- proaching the northern edge of its range in the BCNRA. The Rocky Mountain element is the second major component of the BCNRA flora, primarily because of the close proximity to the Bighorn and Pryor mountains. Some of the species associated with this element are Abies lasiocarpa, Picea engehnamiii, Pinus pon- derosa, Mahonia repens, Arnica cordifolia, and Calypso bulbosa. The smallest element of the flora represents the Great Plains region. This element is found in the northeast portion of the area and in- cludes Amorpha canescens, Calylophus scr- rulatus, Andropogon gerardii, Bouteloua curtipendida. Echinacea angustifolia, Liatris punctata, and Petalostemon purpureum. The distribution of the three floristic ele- ments is correlated to some extent with cli- matic patterns. The south end of the recre- ation area is the most arid and is dominated by 'Wyoming Natural Heritage Program/The Nature Conservancy, 1603 Capitol Avenue #32.5, Cheyt ^Department of Botany, University of Wyoming, Laramie. Wyoming 82071. 734 Wyoming 82001. 3ctober 1985 LiCHVAR ET AL.: BiCHORN CaNYON PlANTS 735 he Great Basin flora, whereas the Rocky Viountain element dominates at cooler, ligher elevations from near the south end to 3ull Elk Basin. Coniferous forests dominate he higher north slopes. In the area of Fort jmith, the flora is more representative of the jreat Plains element, with higher summer )recipitation. The average annual precipita- ion ranges from 18 cm (7 inches) at the south md (Lovell) to 50 cm (20 inches) at Fort Smith )n the north end (BCNRA, personal commu- lication). Novelties of the Flora Because of the representation of three ma- or elements in the area and numerous en- lemics located within the Bighorn Moun- ains, the checklist for BCNRA has some taxa varranting recognition. Also, these taxa of in- erest include species on the edge of their •anges. A list of these species follows. Near or Regional Endemics Penstemon cariji Penn. SiiUivantia hapemanii (Coult. & Fish.) Coult. Musineon vaginatum Rydb. Townsendia spothulata Nutt. Erigeron allocotus Blake Cryptantha cana (A. Nels.) Pays. Eritrichitim howardii (Gray) Rydb. Kelseya uniflora (Wats.) Rydb. Eriogonum brevicaule Nutt. spp. camim (Stokes) Dorn Astragalus hyalinus Jones Stanleya tomentosa Parry Peripheral Species Ligusticiim porteri Coult. & Rose Logfia arvensis (L.) Holub. Polystichum lonchitis (L.) Roth. Triodanis leptocarpa (Nutt.) Niewl. Liliiim philadelphicum L. Smilax herhacea L. The Checklist Families, genera, and species are listed in alphabetical order, and the nomenclature fol- lows Dorn (1977), Hitchcock and Cronquist 1973), and Cronquist, Holmgren, Holmgren, md Reveal (1972, 1977). A total of 73 families Df vascular plants occur in the BCNRA with 320 genera and 656 taxa of specific or sub- jpecific rank. Because this checklist is based largely on one svmimer of fieldwork, addi- tional species should be expected with addi- tional collections. Acknowledcments We are grateful to the University of Wyo- ming-National Park Service Research Center and Kenneth Diem, its director, for funding and support throughout the project. William Binnewies, Richard Lake, and J. Terry Peters of the BCNRA provided logistical support, information, and encouragement, and Yoshiko Akashi typed herbarium labels and helped with other important tasks. Ronald Hartman and B. Ernie Nelson are thanked for use of the Rocky Mountain Herbarium, and especially for checking and annotating speci- mens to ensure proper identification. Robert Dorn shared his field collections and notes from the region. Equisetaceae Equisetum arvense L. hyemale L. laevigatum A. Br. POLYPODIACEAE Cheilanthes feeiT. Moore Cystopteris fragilis(L.) Bernh. var. fragilis Pellaea occidentaUs (E. Nels. ) Rydb. Polystichum lonchitis (L.) Roth Woodsia scopulina D.C. Eat. Selaginellaceae Selaginella densa Rydb. Cupressaceae Juniperus communis L. var. depressa Pursh osteospenna (Torr.) Little scopulorum Sarg. Pinaceae Abies lasiocarpa (Hook.) Nutt. var. lasiocarpa Picea engelmannii Parry e.x Engelm. 736 Great Basin Naturalist Vol. 45, No. 4 Pimis flexilis James ponderosa Dougl. ex P. & C. Lawson Pseudotsuga inenziesii (Mirb.) Franeo var. glauca (Beissn.) Franco ACERACEAE Acer glabrum Torr. negundo L. var. interuis (Britt.) Sarg. Amar.\nthaceae Amaranthus blit aides Wats. Anacardiaceae Rhus glabra L. trilobata Nutt. var. trilobata Toxicodendron rydbergii (Small ex Rydb.) Greene Apiaceae Berula erecta (Huds.)Cov. var. incisa (Torr. ) Cronq. Biipleurum americamtm Coult. & Rose Coniiim maculatum L. Cijmopterus acaulis (Pursh) Raf. Heracleum sphondylium L. Ligusticwn filicirum Wats. Lonwtitim ecus (Wats.) Coult. & Rose dissectiim (Nutt.) Math. & Const. var. multifidum (Nutt.) Math. & Const. foeniculaceum (Nutt.) Coult. & Rose orientale Coult. & Rose triternatum (Pursh) Coult. & Rose ssp. platycarpum (Torr.) Cronq. Musineon divaricatum (Pursh) Nutt. ex T. 6c G. vaginattim Rydb. Osmorhiza chilensis H. & A. depauperata Phil. longistylis (Torr.) DC. Perideridia gairdneri (H. & A.) Math. ssp. borealis Chaung & Const. Pteryxia terebinthina (Hook.) Coult. & Rose var. calcarea (Jones) Math. Sanicula marilandica L. Apocynaceae Apocynum androsaemifolium L. cannabinum L. Asclepiad.\ceae Asclepias speciosa Torr. Asteraceae Achillea millefolium L. var. lamdosa (Nutt.) Piper Agoseris glauca (Pursh) Raf. var. dasycephala (T. & G.) Jeps. var. glauca var. laciniata (D.C. Eat.) Smiley Ambrosia artemisiifolia L. var. elatior{L.) Desc. psilostachya DC. var. coronopifolia (T. & G.) Farw. trifida L. var. trifida Antcnnaria alpina (L.) Gaertn. var. media (Greene) Jeps. microphylla Rydb. parvifolia Nutt. racemosa Hook. Arctium minus (Hill) Bernh. Arnica cordifolia Hook. fulgens Pursh latifolia Bong. rydbergii Greene sororia Greene Artemisia biennis Willd. campestris L. ssp. borealis (Pall.) Hall & Clem, var. scouleriana (Bess.) Cronq. cana Pursh ssp. cana dracunculus L. frigida Willd. ludoviciana Nutt. var. ludoviciana nova A. Nels. si)incscens D. C. Eat. tridrntata Nutt. Aster chilensis Nees ssp. adscen(hns ( 1 -indl. ) Cronq. ciliolatus Lindl. eatonii {GvAs) Howell falcatus Lindl. foliaceus Lindl. ex DC hcsperius (iray October 1985 LiCHVARETAL.; BiCHORN CaNYON PLANTS 737 Balsamorhiza incana Nutt. sagUtata (Pursh) Nutt. Bidens cernua L. Brickellia grandijlora (Hook.) Nutt. var. grandiflura Centaurea maculosa Lam. repens L. Chaenactis douglasii (Hook.) H. & A. Chnjsanthemum leucanthemum L. Chrysothamnus natiseosus (Pall, ex Pursh) Britt. viscidiflorus (Hook.) Nutt. ssp. viscidiflorus var. latifolius (D.C. Eat.) Rydb. var. viscidiflorus Cichorium inttjbus L. Cirsium arvense (L.) Scop. flodmanii (Rydb.) Arthur tweedyi (Rydb.) Petr. undulaturn (Nutt.) Spreng. vulgare (Savi) Tenore Conyza canadensis (L.) Cronq. Crepis acuminata Nutt. ssp. acuminata atraharba Heller intermedia Gray modocensis Greene ssp. modocensis runcinata (James) T. & G. Echinacea pallida Nutt. var. angustifolia (DC.) Cronq. Erigeron allocotus Blake caespitosus Nutt. compositus Pursh corymbosus Nutt. glabellus Nutt. var. glabellus ochroleucus Nutt. var. ochroleucus var. scribneri (Canby ex Rydb.) Cronq. pumilus Nutt. ssp. pumilus speciosus (Lindl.) DC. strigosus Muhl. e.x Willd. Eupatorium maculatum L. Gaillardia aristata Pursh Grindelia squarrosa (Pursh) Dunal Gutierrezia sarothrae (Pursh) Britt. & Rusby Haplopappus armerioides (Nutt.) Gray Hclianthclla quimiucnervis (Hook.) Gray Hclianthus annuus L. nuttallii T. & G. petiolaris Nutt. ssp. petiolaris rigidus (Cass.) Desf. var. subrhomboideus (Rydb.) Cronq. Heterothcca villosa (Pursh) Shinners Hieracium albiflorus Hook. cynoglossoides Arv. -Touv. Hymenopappus filifolius Hook, var. filifolius var. polycephalus (Osterh.) Turner Hyinenoxys acaulis (Pursh) Parker torreyana (Nutt.) Parker Iva axillaris Pursh xanthifolia Nutt. Lactuca ludoviciana (Nutt.) Ridd. oblongifolia Nutt. serriola L. Liatris punctata Hook. var. punctata Logfia arvensis (h.) Holub Lygodesmia juncea (Pursh) D. Don Machaeranthera grindelioides (Nutt.) Shinners var. grindelioides tanacetifolia (H.B.K.) Nees Malacothrix sonchoides (Nutt.) T. & G. torreyi Gray Microseris nutans (Hook.) Schultz-Bip. Nothocalais cuspidata (Pursh) Greene Picrandeniopsis oppositifolia (Nutt.) Rydb. ex Britt. Platyschkuhria integrifolia (Gray) Rydb. var. integrifolia 738 Great Basin Naturalist Vol. 45, No. 4 Ratibida columnifera (Nutt.) Woot. & Standi. Rudheckia laciniata L. var. ampla (A. Nels.) Cronq. Senecio canus Hook. eremophilus Richards. integerrimiis Nutt. pauperculus Michx. plattensis Nutt. serra Hook. streptanthif alius Greene Solidago canadensis L. var. salehrosa (Piper) Jones gigantea Ait. var. serotina (Kuntze) Cronq. missouriensis Nutt. rigida L. var. humilis Porter spathidata DC. var. nana (Gray) Cronq. Sonchus asper {L.)Hi\\ oleraceus L. uliginosus Bieb. Sphaeromeria capitata Nutt. Stephanomeria runcinata Nutt. Taraxacum laevigatujn{WiM.)DC. officinale Weber Tetradijmia canescens DC. spinosa Hook & Arn. Townsendia hookeri Beaman incana Nutt. parrtji D.C. Eat. spathulata Nutt. Tragopogon dubius Scop. scabra Hook, var. scabra Xanthium struma riuin L. Bf.hberidac:eae Mahonia repens (Lindl. ) G. Don Betulaceae Betula occidentalis Hook. BOKACINACEAE Aspertigo procumhens. L. Coldenia nuttallii Hook. C njptantha cana (A. Nels.) Pays. celosioides (Eastw.) Pays. flavoculata (A. Nels.) Pays. kelsetjana Greene Cijnoglossutn officinale L. Eritrichiurn howardii (Gray) Rydb. Hackelia deflexa (Wahlenb.) Opiz var. americana (Gray) Fern. &I. M. Johnst. floribunda (Lehm.) I. M. Johnst. Lappula redowskii (Horneni.) Greene Lithospermum incisum Lehm. ruderale Dougl. ex Lehm. Mertensia ciliata (James ex Torr.) G. Don var. ciliata oblongifolia (Nutt.) G. Don viridis{A. Nels.) A. Nels. Onosmodium inolle Michx. var. molle var. occidentalc (Mack.) I. M. Johnst. Brassicaceae Alyssum alyssoidcs (L.) L. desertorum Stapf Arabis demissa Greene var. languida Roll. glabra (L.) Bernh. hirsuta (L.) Scop. holbocllii Hornem. var. pendulocarpa (A. Nels.) Roll. var. retrofracta (Grab.) Rydb. lignifera A. Nels. microphylla Nutt. var. saxiinontana Roll. nuttallii Robins. sparsiflora Nutt. Barbarea orthoccras Ledcb. Camelina inicrucarpa .\ndrz. ex DC. Capsella bursa-pastoris (L.) Medic, var. hursa-pastoris Cardaria chalepensis (L.) Hand.-Mazz. pubescens (C. A. Mey.) Jarmol. Chorispora tenellaiVADDC. October 1985 LiCHVAR ETAL.: BiCHORN CaNYON PlANTS 739 Descurainia pinnata (WAt.) Britt. Sophia (L.) Webb ex Prantl Draba crassifolia Grab. var. crassifolia nemo rasa L. oU'fiospenna Hook. pracalta Greene reptans (Lam.) Fern. ssp. reptans stenoloba Ledeb Erysimum asperum (Nutt.) DC. var. asperum cheiranthoides L. ssp. altum Ahti inconspicuum (Wats.) MacM. var. inconspicuum Hesperis matronalis L. Lepidium densiflorum Schrad. perfoliatum L. Lesquerella fl/pjna (Nutt.) Wats. ssp. alpina ludoviciana (Nutt.) Wats. Malcohnia africana (L.) R. Br. Nasturtium officinale R. Br. Physaria acutifolia Rydb. var. acutifolia didymocarpa (Hook.) Gray vary, didymocarpa Rorippa calycina (Engelni.) Rydb. curvipes Greene sinuata (Nutt.) Hitchc. Sisy7nbrium altissimum L. loeselii L. linifolium (Nutt.) Nutt. exT. & G. Smelowskia calycina (Steph. e.x Willd.) G.A. May var. americana (Regel & Herd.) Drury & Roll. Stanleya pinnata (Pursh) Britt. tomentosa Parry Streptanthella longirosfris (Wats.) Rydb. var. longirostris Cactaceae Opuntia polyacantha Haw. Pediocactus simpsonii (Enseli var. simpsonii ) Britt. &Rose Campami.ackak Campanula rotundifolia L. Triodanis leptocarpa (Nutt.) Nieuwl. perfoliata (L.) Nieuwl. var. perfoliata Cannabaceae Humulus lupulus L. V ar. neomexicanus Nels. & Ckll. Capparaceae Cleome lute a Hook, var. lutea serrtdata Pursh Polanisia dodecandra (L.) DC. ssp. trachysperma (T. & G.) litis Caprifoliaceae Lonicera utahensis Wats. Sambuctis canadensis L. var. canadensis cerulea Raf. var. cerulea racemosa L. ssp. pubens (Michx.) House var. melanocarpa (Gray) McMinn Stjmphoricarpos albus (L.) Bhke var. albus occidentalis Hook. oreophilus Gray var. w^a/jensis (Rydb.) A. Nels. Caryophyllaceae Arenaria congesta Nutt. hookeri Nutt. nutt alia Pax ssp. nuttallii obtusiloba (Rydb.) Fern. Cerastium arvense L. nutans Raf. var. nutans Lychnis alba Mill. Paronychia sessiliflora Nutt. Silene menziesii Hook. var. viscosa (Greene) Hitchc. & Maguire 740 Great Basin Naturalist Vol. 45, No. 4 Stella ria media {L.)Vi\\. Chenopodiaceae Atriplex argentea Nutt. ssp. argentea var. argentea canescens (Pursh) Nutt. confertifolia (Torr. & Frem.) Wats. gardneri (Moq.)D. Dietr. heterosperma Bunge rosga L. Ceratoides lanata (Pursh) J. T. Howell var. lanata Chenopodium berlandieri Moq. var. zschackeii]. Murr.)J. Murr. leptophyllum (Moq.) Wats. Corispermum hyssopifoliiim L. Grayia spinosa (Hook.) Moq. Halogeton glome rat us (Bieb.) C. A. Mey. Kochia americana Wats. scoparia (L.) Schrad. Monolepis nuttalliana (Schult.) Greene Salsola kali L. var. tenuifolia Tdusch. Sarcobatus vermiculatus (Hook.) Torr. var. vermiculatus Suaeda fruticosa (L.) Forsk. torreyana Wats. COMMELINACEAE Tradescantia bracteata Small ex Britt. occidentalis (Britt.) Smyth var. occidentalis CONVOLVULACEAE Convolvulus arvensis L. Ipomoea leptophylla Torr. CORNACEAE Cornus stolonifera Michx. Crassulaceae Sedum lanceolatum Torr. Cyperaceae Carex aquatilis Wahlenb. elynoidcs Holm fihfalia Nutt. heliophila Mack. lioodii Boott interior Bailey lanuginosa Michx. microptcra Mack. nehraskcnsis Dewey oederi Retz parry ana Dewey var. parryana praegracilis W. Boott raynoldsii Dewey rostrata Stokes ex With. sprengelii Dewey ex Spreng. viridula Michx. vulpinoidea Michx. Eleocharis palustris (L.) R. & S. Scirpus pallidus (Britt.) Fern. pungens Vahl validus Vahl Elaeagnaceae Elaeagnus angustifolia L. Shepherdia argentea (Pursh) Nutt. canadensis (L.) Nutt. Ericaceae Pyrola asarifolia Michx. secunda L. EUPHORBUCEAE Euphorbia cyparissias L. esula L. glyptosperma Engelm. robusta (Engeliu.) Small Fab.\ceae Astragalus adsurgens Pall. var. robustior Hook. agrestis Dougl. ex G. Don bisulcatus (Hook. ) (iray canadensis L. var. canadensis crraniicus Shclcl. var. /i/i/o/ii/.v (Gray) F.J. H chamaeleuce Gray crassicarpus Nutt. drumnwndii Dougl. ex Hook. gilviflorus Shcld. Jiyalinus ](.mcs ki)itro})luita i\\w\ lotijlorus Hook. October 1985 LiCHVAR ET AL.: BiCHORN CaNYON PLANTS 741 miser Dougl. ex Hook. var. decumbens (Nutt.) Cronq. missouriensis Nutt. var. missouriensis oreganus Nutt. ptirshii Dougl. e.x Hook. var. purshii spcitulafus Sheld. Glycyrrhiza U'pidota Pursh Hedijsarum boreale Nutt. stdphurescens. Rydb. Liipinus arnenteus Pursh \ar. ar(;.cntciis scriceus Pursh icyctliii Wats. Medicago lupulina L. sativa L. Melilotus alba Medic. officinalis (h.)?ii\\. Oxytropis besseyi (Rydb.) Blank. var. besseyi var. fallax Barneby campestris (L.) DC. deflexa{Va\\.)DC. lagopus Nutt. sericea Nutt. Petalostemon occidentale (Heller ex Britt. & Kearn.) Fern. purpureum (Vent.) Rydb. Psoralen esctdenta Pursh lanceolata Pursh tenuiflora Pursh var. tenuiflora Sphaerophysa salsula (PaW.) DC. Thermopsis rhombifolia (Nutt. ex Pursh) Nutt. ex Richards. Trifolium hybridum L. pratcnse L. Vicia americana Muhl. ex Willd. var. minor Hook. FUMARIACEAE Corydalis aurea Willd. Gentianaceae Frasera speciosa Dougl. ex Griseb. Grrani.\(:kae E rod ill m ciciitarium (L.) L'Her. Geranium viscosissimmn Fisch. & Mey. ex Mey. var. viscosissimum Grossulariaceae Ribes americaniim Mill. aureum Pursh var. aureum cereum Dougl. var. inebrians (Lindl.) C. L. Hitchc setosum Lindl. Hydrophyllaceae Phacelia glandulosa Nutt. hastata Dougl. ex Lehm. ivesiana Torr. linearis (Pursh) Holz. sericea (Grab, ex Hook.) Gray Iridaceae Iris missouriensis Nutt. Sisyrinchium angustifolium Mill. montanum Greene Juglandaceae Jiiglans cinerea L. JUNCACEAE Junciis balticus Willd. var. montanus Engelm. ensifolius Wikstr. longistylis Torr. var. longistylis parryi Engelm. regelii Buch. tenuis Willd. var. diidleyi (Wieg.) F. J. Herm. var. tenuis torreyi Gov. Lamiaceae Hedeoma drummondii Benth. Mentha arvensis L. var. glabrata (Benth.) Fern. Monarda fistulosa L. var. menthaefolia (Grab.) Fern. Nepeta cat aria L. Scutellaria galericulata L. Stachys palustris L. var. pilosa (Nutt.) Fern. 742 Great Basin Naturalist Vol. 45, No. 4 Lentibulariaceae Utricularia vulgaris L. LiLIACEAE Allium brevistylum Wats. geyeri Wats. textile Nels. & Macbr. Asparagus officinalis L. Calochortus gunnisonii Wats. nuttallUT. &G. Disporum trachycarpum (Wats.) Benth. & Hook. Fritillaria pudica (Pursh) Spreng. Leucocrinum montanum Nutt. e.x Gray Lilium philadelphicum L. var. andinum (Nutt.) Ker Smilacina racemosa (L.) Desf. stellata (L.) Ded. Stnilax herbacea L. var. lasioneuron (Hook.) A. DC. Yucca glauca Nutt. var. glauca Zygadenus elegans Pursh ssp. elegans venenosus Wats. var. gramineus (Rydb.) Walsh e.x Peck LiNACEAE Linum lewisii Pursh var. lewisii LOASACEAE Mentzelia albicaulis (Dougl. ex Hook.) T. & G. decapetala (Pursh ex Sims) Urban Malvaceae Malva parviflora L. Sphaeralcea coccinea (Nutt.) Rydb. Nyctaginaceae Abronia fragrans Nutt. ex Hook. Mirabilis linearis (Pursh) I Icimerl Onagraceae Cahjlophus serndatus (Nutt.) Raven Catnissionia andina (Nutt.) Raven minor (A. Nels. ) Raven scapoidea (T. & G.) Raven ssp. scapoidea Circaea alpina L. var. alpina Epilobium angustifolium L. ciliatuni Rat. ssp. ciliatum ssp. glandulosum (Lehm.) Hoch & Raven paniculatum Nutt. exT. & G. Gaura coccinea Nutt. ex Pursh parviflora Dougl. ex Hook. Gayophytum ramosissimum T. & G. Oenothera albicaulis Pursh caespitosa Nutt. depressa Greene pallida Lindl. ssp. trichocalyx (Nu\.t. exT. &G.) Munz & Klein Orchid.\ceae Corallorhiza maculata Raf. striata Lindl. Goodyera oblongifolia Raf. Piperia unalascensis (Spreng.) Rydb. Platanthera dilatata (Pursh) Lindl. ex Beck hyperborea (L.) Lindl. Orobanchaceae Orobanche uniflora L. Plantaginaceae Plantago major L. patagonica Jaccj. POACEAE Agropyron cristatum (L.) Gaertn. elongatum (Host) Beauv. repens (L.) Beau\'. riparium Scribn. & Sni. smithii Rydb. spicatum (Pursh) Scribn. & Sm. f. spied til in trachycaulum (Link) Malte ex Lewis October 1985 LiCHVARETAL.: BiCHORN CaNYON PlANTS 743 Agrostis alba L. var. palust ris (Huds.) Pers. exarata Trin. scahra Willd. Andropogon gerardii Vitman scoparius Michx. var. scoparius Aristida fendleriana Steud. longiseta Steud. Bouteloua curtipendula (Michx.) Torr. var. curtipendula gracilis {H. B. K.) Lag. ex Griffiths var. gracilis Bromus carinatus H. & A. ciliatus L. cominutatus Schrad. inennis Leys. japonicus Thunb. tectorum L. Calamagrostis purpurascens R. Br. Calamovilfa longifolia (Hook.) Scribn. var. longifolia Dactylis glomerata L. Deschampsia cespitosa (L.) Beauv. var. cespitosa Distichlis spicata (L.) Greene var. stricta (Torr.) Scribn. Echinochloa crusgalli (L.) Beuav. var. crusgalli Elymus canadensis L. var. canadensis cinereus Scribn. & Merr. glaucus Buckl. var. glaucus virginicus L. var. suhmuticus Hook. Festuca idahoensis Elmer octoflora Walt. Glyceria striata (Lam.) Hitchc. var. stricta (Schribn.) Fern. Hordeum jubatum L. Koeleria macrantha (Ledeb.) Schult. Leucopoa )cm^'/i(Wats.)W. A. Weber Mclica sulndata (Griseb.) Scribn. var. pammelii (Scril)n.) G.L. Hitchc. Muhlenbergia racemosa (Michx.) B. S. P. Oryzopsis hymenoides (R. & S.) Ricker ex Piper rnicrantha (Trin. & Rupr.)Thurb. Phalaris arundinacea L. Phleum pratense L. Phragmites australis (Cav.) Trin. ex Steud. Poa alpina L. var. alpina arida Vasey bidbosa L. compressa L. cusickii Vasey fendleriana (Steud.) Vasey glaucifolia Scribn. & Williams ex Williams interior Rydb. pratensis L. sandbergii Vasey Polypogon monspeliensis (L.) Desf. Setaria viridis{h.) Beauv. Sitanion hystrix (Nutt)]. G. Sm. Spartina pectinata Link Sphenopholis obtusata (Michx.) Scribn. var. major (Torr.) K. S. Erdm. Sporobolus airoides (Torr.) Torr. cryptandrus (Torr.) Gray Stipa Columbiana Macoun comata Trin. & Rupr. lettennanii Vasey viridula Trin. williamsii Scribn. Trisetum spicatum (h.) Richt. POLEMONIACEAE Gilia leptomeria Gray pinnatifida Nutt. ex Gray tweedyi Rydb. 744 Great Basin Naturalist Vol. 45, No. 4 Iponiopsis ptimila (Nutt.) Grant spicata (Nutt.) Grant Leptodactylon caespitosum Nutt. pungens (Torr.) Nutt. ssp. pungens Linanthus septentrionalis Mason Phlox brijoides Nutt. hoodii Richards. midtiflora A. Nels. Polemonium occidentale Greene var. occidentale pulcherrimtun Hook. var. pidchenimum viscosum Nutt. POLYGONACEAE Eriogonum annitum Nutt. brevicaide Nutt. ssp. caniim (Stokes) Dorn flavum Nutt. var. flavum ovalifolium Nutt. var. ovalifolium Polygonum achoreum Blake aviculare L. bistortoides Pursh lapathifolium L. Rumex crispus L. triangulivalvis (Danser) Rech. f. venosus Pursh PORTULACACEAE Claytonia lanceolata Pursh var. lanceolata Lewisia rediviva Pursh ssp. rediviva Montia perfoliata (Donn) Howell var. perfoliata Portidaca oleracea L. POTAMOCETONACEAE Potamogeton filiformis Pers. Primulaceae Androsace septentrionalis L. Dodecatheon pulchellum (Raf.) Merr. Lysimachia ciliata L. Ranunculaceae Actaea rubra (Ait.) WiM. Anemone cylindrica Gray multifida Poir. patens L. Clematis Columbiana (Nutt.) T. & G. var. tenuiloba (Gray) ] . Pringle ligusticifolia Nutt. var. ligusticifolia Delphinium hicolor Nutt. Ranunculus cymbalaha Pursh var. cymbalaria macounii Britt. pensylvanicus L. f. testiculatus Crantz uncinatus D. Don e.\ G. Don var. uncinatus Thalictrum dasycarpum Fisch. et al. fendleri Engelm. ex Gray var. fendleri occidentale Gray var. palousense St. John Rosaceae Agrimonia gryposepala Wallr. Amelanchier alnifolia (Nutt.) Nutt. Cercocarpus ledifolius Nutt. Crataegus douglasii Lindl. Fragaria vesca L. var. bractreata (HeWer) R. J. Davis virginiana Duchn. var. glauca Wats. Geuin aleppicum Jaaj. macro))hylhtm Willd. var. periucisuni (R\cll).) Raup triflorwn Pursh var. triflorum Ivesia gorf/onii (Hook.) T. & G. Kelseya uniflora (Wats.) R\(lb. Petrophytum caespitosum (Nutt.) R\-dh. Physorariius malvaccus (Cvrvuc) Kuntze monogyiuis (Torr.) Goult. October 1985 LiCHVARETAL.: BKJHORN CaNYON PLANTS 745 Potentilla anserina L. hientiis Greene diversifolia Lehm. var. diversifolia fissa Nutt. fruticosa L. gracilis Dougl. ex Hook. var. glabrata (Lehm.) C. L. Hitchc. norvegica L. ssp. monspeliensis (L.) Asch. &Graebn. ovina Macoiin paradoxa Nutt. pensylvanica L. Pruntis americana Marsh virginiana L. var. melanocarpa (A. Nels.) Sarg. Rosa acicularis Lindl. ssp. sayi (Schwein.) W. H. Lewis saiji Schwein. woodsii Lindl. Ruhus idaeus L. ssp. sachalinensis (Levi.) Focke var. gracilipes Jones Spiraea betulifoUa Pall. var. lucida (Dougl. ex Hook.) C. L. Hitchc. RUBIACEAE Galium boreale L. trifidum L. triflorum Michx. Salicaceae Populus X acuminata Rydb. alba L. angustifolia James deltoides Marsh. ssp. monilifera (Ait.) Echenw. tremuloides Michx. Salix amygdaloides Anderss. boothii Dorn drummondiana Barr. ex Hook. exigua Nutt. lutea Nutt. monticola Bebb Santalaceae Co7iiandra UTiibellata (L.) Nutt. var. pallida {A. DC.) Jones Saxifragaceae Boykinia heucheriformis (Rydb.) Rosend. Conimitella williamsii (D. C. Eat.) Rydb. Heuchera parvifolia Nutt. exT. & G. Litlwphragma parviflorum (Hook.) Nutt. ex T. & G. Parnassia palustris L. var. montanensis (Y cm. ik Rydb.) C. L. Hichc. Saxifraga rhomboidea Greene Sullivantia hapemanii (Coult. & Fish) Coult. var. hapemanii SCROPHULARIACEAE Besseya wyomingensis (A. Nels.) Rydb. Castilleja angustifolia (Nutt.) G. Don chromosa A. Nels. linariaefolia Benth. miniata Dougl. ex Hook. pulchclla H\clb. sessiliflora Pursh Collinsia parviflora Lindl. Mimulus guttatus DC. ssp. guttatus suksdorfii Gray Pedicularis cystopteridifolia Rydb. Penstemun a rid us Rydb. cariji Penn. confertus Dougl. var. procerus (Dougl. ex Grab.) Gov. eriantherus Pursh var. eriantherus glaber Pursh humilis Nutt. ex Gray var. humilis laricifolius H. & A. nitidus Dougl. ex Benth. radicosus A. Nels. Verbascum thapsus L. Veronica americana Schwein. ex Benth. peregrina L. var. xalapen.sis{H. B. K.) St. John & Warren SOLANACEAE Physalis heterophylla Nees var. heterophylla 746 Solanum dulcamara L. trijlorum Nutt. Tamaricaceae Tamarix chinensis Lour. Typhaceae Typha angustifolia L. latifolia L. Ulmaceae Celtis occidentalis L. Ulmus pumila L. Urticaceae Parietaria pensylvanica Muhl. Urtica dioica L. Great Basin Naturalist Vol. 45, No. 4 Willd. Valerianaceae Valeriana dioica L. var. sylvatica (Richards.) Wats. edulis Nutt. ex T. & G. var. edulis Verbenaceae Verbena hracteata Lag. & Rodr. hastata L. ViOLACEAE Vto/a adunca Sm. var. adunca Harrington canadensis L. var. canadensis nuttallii Pursh tnllicola A. Nels. VlTACEAE Vifts riparia Michx. Literature Cited Dorn, R. D. 1977. Manual of the Vascular Plants of Wyoming. Garland PubHshing, Inc., New York. CrONQUIST, a , A HOL.MGREN, N HOLMGREN, AND J. Reveal. 1972. Intermountain Flora, Volume 1. Hafner Publishing Company, Inc., New York. 1977. Intermountain Flora, Volume 6. Hafner Publishing Company, Inc., New York. Hitchcock, C. L. and A. Cronquist. 1973. Flora of the Pacific Northwest. University of Washington Press, Seattle. Porter, C L 1962. A Flora of Wyoming, Part 1. Agricul- tural Experiment Station Bulletin 402. University of Wyoming, Laramie. PRESETTLEMENT VEGETATION OF PART OF NORTHWESTERN MOFFAT COUNTY, COLORADO, DESCRIBED FROM REMNANTS William L. Baker' and Susan C. Kennedy" Abstract. — A general botanical inventory of a part of northwestern Moffat County, Colorado, resulted in the location of "remnants" of the presettlement vegetation spectrum that are largely unaltered by grazing, logging, or other recent human-related land uses. The 69 samples taken from these remnants were classified into 22 plant associations. Composition, structure, environmental location, geographical range, and response to disturbance are discussed for each association, and a photograph of each is presented. Seven of the 22 associations are apparently restricted to the study area. Restricted associations occur in the more extreme environments of the study area, such as on calcareous substrata or very xeric sites. More niesic sites along ephemeral creeks, on north-facing slopes, or on sandstones support plant associations that have much wider ranges, many of them extending across the northern Great Basin. The vegetation that occupied the landscape in the western United States prior to settle- ment has been effectively extirpated in some areas by conversion to cultivation or by urban development. In most of the remainder, other kinds of land use have resulted in alteration of the presettlement composition and structure. The most pervasive and most consequential of these other land uses are domestic livestock grazing and logging, though mining and recreation have had substantial effects in more localized areas. Also pervasive has been the replacement of nati^'e plant species by exotics. Additional effects have resulted from fire control, loss or modification of native her- bivore populations, exotic diseases, air pollu- ition, and acid precipitation. In some parts of the western United States, and in many parts of the eastern United States, the composition and structure of the presettlement vegetation can only be known now by reference to histor- ical accounts, early photographs, and other secondary records. Vale (1982) reviewed methods of analyzing these sources. Never- theless, in parts of the West it is still possible to locate remnants of the presettlement vege- itation, which have essentially escaped alter- ation, though such remnants are exceedingly rare at lower elevations or on very productive >ites and' are disappearing as land uses con- tinue or accelerate. These remnants have been widely used in :he forested parts of the western United States to develop "habitat type" classifications (Pfister 1982). Such classifications are irre- placeable records of the detailed composition and structure of the presettlement vegeta- tion. Some of these remnants, occurring on federal lands, have been protected from fur- ther alteration or loss by designation as Re- search Natural Areas under regulations and policies of the U.S. Department of the Inte- rior, Department of Agriculture, and other departments. Perpetuation of remnants in such designated Research Natural Areas means that they will be available in the future for more extensive study. Very few opportuni- ties are available for the study of ecosystem function on unaltered sites. Without such studies it is difficult for land managers to know how to most efficiently manage land uses on similar lands for maximum benefit with mini- mum alteration. Such protected remnants also serve an important role in the long-term perpetuation of their component plants and animals. The natural vegetation of some parts of Col- orado is nearly unknown (Baker 1982a), par- ticularly at lower elevations. An earlier report characterized some of the presettlement veg- etation of the Piceance Basin occurring on Green River and Uinta formations (Baker 1983). This report extends that earlier report to include additional areas of Green River Formation (Fm.) and other geologic substrata occurring in a part of northwestern Moffat Rocky Mountain Heritage Task Force, 1370 Pennsylvania, Suite 190, Denver, Colorado 80203. ^Department of Biology, Western State College, Gunnison, Colorado 81230. 747 748 Great Basin Naturalist Vol. 45, No. 4 ^ r" 1 2 3 4 c:o\ /Con r- 1 2 Co /Coil 1 2 3 4 1 Cov/Con Shrubs Ephedra viridis Artemisia tridejitala ssp. wyomingensis Atriplex confertifolia Opuntia polyacantha Pediocactus simpsonii Artemisia nova Cercocarpus montanus Symphoricarpos oreophilus Cercocarpus ledifolius var. intricatus Chrysothamnus nauseosus Chrysothamnus viscidiflorus Fendlerella utahensis Purshia tridentata tr/50 tr/25 tr/25 tr/25 9.0/99 tr/.50 tr/.50 tr/25 .6/50 tr/25 tr/25 tr/25 Graminoids Agropyron spicatum Poa sandbergii Poa fendleriana Oryzopsis hymenoide Bromus tectorum Sitanion hystrix Koeleria cristata Poa canbyi Carex filifolia Carex pityophila 21. 3/99 1.0/99 3,. 3/99 tr/75 .6/50 tr/25 tr/25 tr/25 tr/25 13.5/99 1..5/.50 .8/99 1,3/99 tr/.50 1.0/.50 tr/75 tr/50 .8/50 tr/25 FORBS Physaria acutifolia Cryptantha flavoculata Erigeron sp. Descurainia richardsonii Phlox hoodii Hedeoma drummondii Gilia sinuata Eriogonum ovalifolium Caulanthus crassicaulis Conringia orientalis Agoseris glauca Arabis demissa Schoenocrambe linifolia Balsamorhiza hookeri var. hispidula Mertensia oblongifolia Crepis occidentalis Senecio integerrimus Haplopappus acaulis Cryptantha fendleri Lappula redowskii Phlox longifolia Astragalus tenellus Trifolium gymnocarpon Astragalus convallarius Arenaria fendleri Eriogonum umbellatum Erigeron eatonii Penstemon fremontii Cryptantha sericea Castilleja chromosa Chaenactis douglasii tr/75 .6/99 tr/25 ,6/99 .8/75 tr/25 tr/50 tr/75 tr/50 tr/25 tr/25 tr/75 tr/50 1.1/75 .6/50 tr/50 tr/25 tr/75 tr/50 tr/50 tr/25 tr/25 tr/25 tr/50 tr/25 tr/25 tr/25 tr/50 tr/25 tr/25 tr/25 ,8/99 tr/.50 tr/.50 tr/99 tr/50 1.0/.50 tr/.50 2.0/99 tr/50 tr/75 tr/25 tr/25 752 Great Basin Naturalist Vol. 45, No. 4 Table 1 continued. Township ION ION ION 10 N 10 N 11 N ION ION ION ION Range Section lOlW lOlW 99W lOOW lOlW lOlW lOlW lOlW lOlW lOOW S23 S15 S9 S27 S.34 S32 S15 S14 S25 S.30 Elevation (meters) 2060 2160 2170 2080 1890 2290 2190 2070 1890 1900 Aspect SW E NE S W SE W NE \V E Slope (degrees) 30 20 15 5 20 20 10 5 5 10 Soil pH Soil EC (mmhos/cm) 7.95 7.85 7.70 7.65 Ave 7.79 7.75 7.95 Ave 7.85 8.15 7.90 7.65 8.(X) Avg-7.93 .37 .86 .33 .36 Avg .48 .34 .34 Avg .;34 .27 .34 40 ,29 Avg= .33 Plant association number Plot number ,3_ r~ , 3 4 C:<)N/Con r- 1 2 C'.iv/Con 1 2 3 4 Cov/Con Astragalus detritalis tr tr/25 Eriogonum sp. tr tr/25 tr tr/50 tr tr/25 Ipomopsis congesta tr tr/25 Lupinus brevicaulis tr tr/25 Haplopappus armerioides tr tr/25 Atriplex sp. tr tr/25 Hymenoxys acauUs tr tr/50 Trifolium longipes ssp. pygmaeum 2 1.0/50 Eriogonum tumulosum tr tr/50 I tr/25 Lesquerella alpina tr tr/50 Penstemon yampaensis tr tr/50 Machaeranthera grindeloides tr tr/50 Cymoptcrus sp (r tr/50 Fetradoriu pumiUi 1 tr/50 Calochortus uutlcilln tr tr/50 Erigerun flagdlaris tr tr/50 Petrophytwn caespitosum 3 .8/25 Cryptantha sp. 1 tr/25 Erigeron nematophyllus tr tr tr/50 Astragalus spatulatus 2 1 .8/50 Cryptantha flava Ir tr/25 Erysimum asperum tr tr tr/50 Cryptantha caespitosa tr tr/25 Arabis pulchra tr tr/25 Oxytropis sericea tr tr tr/50 Stanleya pinnata tr tr/25 Vicia amerinma tr tr/25 Nemophilu hrevifion tr tr/25 Cymopterus fendleri tr tr/25 Lesquerella ludoviciana tr tr/25 Arenaria fendleri tr tr/25 Lappula redowskii tr tr/25 absent. Agropyron spicatum forms a dense sward in stands that are in good condition (Table 1). Stands usually consist of a moder- ately dense (320 — 510 trees/ha) overstory of pure J uniperus osteosperma (Table 2). Pinus edulis, when present, consists primarily of seedlings, but an occasional larger stem may occur. Many stands consistofonly large stems (9-35 in dbh), with very few smaller stems or seedlings and saplings. Domestic livestock grazing results in de- creases in Agropyrun spicatum and Poafend- leriana and increases in shrubs, herbs, and the exotic grass B ramus tectorum. Several stands were observed with Poa sandhergii dominant and only trace quantities of Agropy- ron spicatum remaining. Another gra/.ing-in- duced successional stage has Artemisia tri- dcntata ssp. xcyomingcnsis dominant in the understory, with only small amounts of grass present. On very flat sites Haplopappus acaulis may become very abundant. This association is limited in Colorado to northern and western MoflPat County. It has been observed in Utah in Dinosaur National Monument (Welsh 1957) and in western Wyo- ming (DeSpain 1973, Wight and Fisser 1968). In the study area it is the most common pinyon-juniper woodland, though a large per- centage of stands have been altered by domes- tic grazing or by woodcutting. 2. Juniperus osteosperma -Pinus edulisi Artemisia naval Agropyron spicatum. — This association occurs on a variety of aspects on several parent materials (e.g.. Browns Park Fm., Morgan Fm.) on steep slopes (15-25 degrees) from 1830 to 2375 m in elevation. Soils are similar to those o{ the Juniperus os- teospermal Agropyron spicatum association, ha\ ing an average pH of 7.85 and an electrical conductivity of .34 mmhos/cm (Table 1). It consists of a moderately dense (400-500 tree.s/ha) stand oi Juniperus asteosperma and Pinus edulis (Table 2, Fig. 2). Juniperus stems October 1985 Baker, Kennedy: Colorado Presettlement Vegetation 753 Fig. 2. (a) Limestone Ridge and part of the study area, [h) Juniperus osteosperma "krummholz," (c) Juniperus osteosperma/Agropyron spicatum, (d) J iiniperus osteosperma -Pinus edulisi Artemisia noval Agropyron spicatum, (e) Juniperus osteosperma -Pinus edulislCercocarpus ledifolius var. intricatus, (f) Cercocarpus ledifoliusl Artemisia tn- dentata ssp. wyomingensis-Symphoricarpos oreophilusi Agropyron spicatum, (g) Artemisia nova/ Agropyron spica- tum, (h) Artemisia nova/Stipa comata. 754 Great Basin Naturalist Vol. 45, No. 4 Table 2. Tree diameter size distribution. Tree diameters were measured by 2" size class at breast height (dbh). Seedlings are less than 1" dbh and less than 1 m tall. Saplings are less than 1" dbh and greater than 1 m tall. Size classes are listed by the midpoint of the size class. Entries are the number of stems in each size class within the 375 m' plot. Plot Seed- Sap- No. Species lings lings 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 31 + Juniperus osteosperma/Agropyron spicatu n 1 Junipertis osteosperma 2 1 1 1 1 4 1 1 1 1 Pinus edulis 5 1 1 2 Juniperus osteosperma 3 3 4 2 1 1 2 Pinus edulis 8 1 1 3 Juniperus osteosperma 1 3 3 1 1 1 1 2 1 1 1 1 3 (.34.35,35) 4 Juniperus osteosperma 1 1 1 2 3 2 1 1 Juniperus osteospemui-Pinus edulis/Artei lisia naval Aii ■opyi 0)1 spicati m 1 Juniperus osteospenna 8 2 2 1 1 1 1 1 1 1 1 Pinus edulis 10 3 4 1 2 Juniperus osteospenmi 4 1 1 1 1 1 2 1 2 1 1 1 (35) Pinus edulis 2 2 4 1 Juniperus osteospenna-Pin IS edidis/Ccrcocarpus ledift lius ^ ar. intricatus 1 Juniperus osteosperma 3 1 3 2 2 1 Pinus edulis 4 5 4 3 4 1 1 2 Juniperus osteospenmi 1 1 1 1 1 Pinus edtdis 1 1 4 2 4 1 3 Juniperus osteosperma 2 3 1 1 1 1 1 1 (.35) Pitius edulis 1 2 3 1 4 Juniperus osteosperma 1 1 1 1 2 1 1 Pinus edulis 2 1 2 1 Cercocarpus ledifoliusl Artemisia tridentata ssp wi/omi ig,ensis—Sympli oric arpos 0 reophiluslAgropyronspicatum 1 Cercocarpus ledifolius .5 1 1 1 1 1 1 1 1 1 (34) Juniperus scopulorum 2 Pinus edulis 1 2 Cercocarpus ledifolius 16 1 1 1 2 2 2 2 1 Juniperus scopulorum 1 Pinus edulis 2 3 Cercocarpus ledifolius 14 7 8 7 5 Juniperus scopulorum 1 2 1 Pinus edulis 1 may be more abundant and often larger than Pinus stems. Both species typically have abundant small stems. The shrub layer is sparse (Table 1), with 8%-10% cover of Artemisia nova , which may be nearly hidden by dense Agropyron spicatiim (Fig. 2). Domestic livestock grazing result.s in an in- crease in Poa sandhergii, Haplopappiis acaulis, Balsamorhiza hookeri var. hispidula , and the exotic B ramus tectorum. Agropyron spicatum may decrease and is often nearly absent on heavily grazed sites. Arteinisia nova does not appear to increase appreciably. This association was first located in Colo- rado during this study. It occurs within the study area in only two locations near Lime- stone Ridge. Data collected in Dinosaur Na- tional Monument suggest it may also occur there (S. Wathen, personal communication). It is probably limited in Colorado to north- western Moffat County but has been reported from southeastern Idaho about 600 km north- west of the study area (Johnson and Pfister 1982), and probably occurs in the intervening areas in southwestern Wyoming and north- eastern Utah. 3. Juniperus osteosperma -Pinus edulis/ Cercocarpus ledifolius var. intricatus. — This association is found exclusively on sandstone outcrops of several formations (Weber Sand- stone, Glen Canyon Sandstone, Entrada Sandstone, and Mesa Verde Group Sand- stone) in the study area, typically occurring on rocky ridge tops or abrupt sandstone outcrops where bedrock is extensively exposed and somewhat cracked and jointed. Aspects are variable, slopes range from 0° to 30°, and the association may occur from 1825 to 2300 m in elevation. Soils are poorly developed, but where there is soil it is similar in average pH (7.93) and electrical conductivity (.33 mmhos/ cm) to that of the other pinyon-juniper associ- ations in the study area (Table 1). The association consists oi a moderately dense (290-560 trees/ha) stand oi Juniperus osteosperma and Pinus edulis (Table 2, Fig. 2). Most trees are less than 20 in dbh. Juni- perus often has more and larger stems than Pi- October 1985 Bakek, Kknnedy: Colorado Preseitlhmknt Vegetation 755 nus . No clear pattern in regeneration poten- tial is apparent, with some stands lacking seedlings and saplings of one or both trees. The trees and the shrub layer, characterized by 10%-25% cover o{ Cercocarpus ledifolius var. intricatiis, commonly grow out of cracks and joints in the bedrock. The herb layer con- sists of about 30 species, which have low con- stancy and cover (Table 1). Total herb cover rarely exceeds 5%. It is unlikely that the association receives much use for domestic grazing, which would have little effect in any event because of lack of forage. The exotic Bromus tectorum now ex- ists in some stands. In Colorado the association has been ob- served in western MoflFat County in the study area and in Dinosaur National Monument, as well as in adjoining northwestern Rio Blanco County. It has been reported from Wayne County, Utah, about 250 km southwest of the study area (Dixon 1935) and very likely occurs in other parts of eastern Utah. In the study area the association is limited to the upturned sandstone outcrops east of Limestone Ridge and Irish Canyon. 4. Cercocarpus ledifolius/Artemisia tri- dentata ssp. wyomingensis-Sijmphoricorpos oreophilusi Agropijron spicatum. — This asso- ciation was found in the study area over a narrow elevational range, above the upper limit of pinyon-juniper woodlands from 2440 to 2560 m. It occurs exclusively on Madison Limestone on steep (25°-35°) slopes on a vari- ety of aspects. Some stands occur as long bands on a slope following a particular layer in the Madison Limestone. Soils are not very different from those of adjoining pinyon-ju- niper woodlands in terms of average pH (7.36) and electrical conductivitv (.43 mmho.s/cm) (Table 3). The association consists of a sparse to mod- erately dense (185-530 trees/ha) stand of Cer- cocarpus ledifolius (Table 2). Although Cerco- carpus ledifolius may only reach shrub stature in some areas, it definitely forms woodlands in the study area. Stems as large as 34 in dbh have been observed. Juniperus scopulorum and Pinus edulis are often present as seedlings or saplings and occasionally as trees. There are generally numerous Cercocarpus seedlings and saplings (Table 2). The shrub layer has 10%-25% total cover, with Artemisia and Symphoricarpos co-dominant. Agropyron spicatum dominates the herb layer with 5%-15% cover (Table 3). Domestic livestock grazing may reduce the amount of Agropyron spicatum, resulting in increases in the amount o( Arteinisia triden- tata and herbs. The association is currently not known out- side the study area, where it has been located Table 3. Percent cover and constancy of shrubs and herbs, plot locations, and physical parameters. Plant association number corresponds to that in the text. 4 = Cercocarpus ledifolius/Artemisia tridentata ssp. wyomingen- sis -Symphoricarpos oreophilusi Agropyron spicatum. Table entries under each plot are percent canopy cover. Tr = trace quantities (less than .5% cover). Table entries under Cov/Con are average percent canopy cover left of the slash, and percent constancy right of the slash. 100 is abbreviated to 99. Soil electrical conductivity (soil EC) is discussed in the text. Township Range Section Elevation (meters) Aspect Soil pH Soil EC (mmhos/cm) Plant association number Plot number ION lOlW S9 2470 NE 7.30 .37 ION lOlW 816 2460 E 7.15 .45 ION lOlW S16 2560 SW 7.65 .48 Avg=7.36 Avg= .43 Cov/Con Shrubs Artemisia tridentata ssp. wyomingensis Symphoricarpos oreophilus Artemisia nova Cercocarpus montanus Ribes cereum Amelanchier utahensis Fendlerella utahensis 12.0/99 5.3/99 tr/33 tr/67 1.2/99 tr/33 tr/33 GREAT BASIN NATURALIST Vol. 45, No. 4 756 Table 3 continued. Township Range Section Elevation (meters) Aspect SoilpH . SoilEC(mmhos/cm) Plant association number Plot number Pediocactussimpsonii GRAMINOIDS Agropijronsplcatwn Poafendleriana Onjzopsishiimenoides Bromustectorum p„(i sandhergii Agwpijron smithn Sitanion hystdx Carex piUjophila FORBS Claytonia lanccoUihi Balsamorhizahoukcn var. hispididci Mertensia ohlongijolui Seneciointegerrirntis Petradoria pumila Collinsia parrijlora Erigeron eatumi Heuchera parvifoM Lithofnigma glabrum lomaiiurn triternatwn Androstephiumhrev.florum Antennaria dimorpha Sedwn stenopetalwn Erysimum asperum Linum letvisii Eriogonum sp. Eriogonwnwnhcllatum Penstemon humilus Agoseris glaiica Lomatium urientale Selaginella densa Phlox hoodii CastiUeja chromosii Petrophjtumcacspitoswn Arabissp. , , Haplopappusannenotde. Uthospermiimruderae SchoenocramheHnrfoha Zigadenus paninilants Delphinium nuttullianum Erigeron nematophijUus Comandraumbellata Descurainia richardsonu CryptanthajlavocuUita Ar«/n.s 'iK'ii/t'''" HijmenopappusfihfoUus Penstemoivin^^ 3 4 4 1 3 tr tr 1 1 tr tr tr tr tr tr tr tr tr tr tr tr tr tr tr Avg=7.36 Avg= .43 10 2 2 tr tr 4 tr tr tr tr tr 1 10 tr 1 tr tr 2 2 1 tr tr tr 2 tr tr tr tr 8.7/99 1.7/67 .8/67 tr/33 tr/3.^ tr/33 tr/33 tr/33 1.8/6' October 1985 Baker, Kennedy: Colok.^do Phesetflement Vegetation 75' only on the upper slopes of Limestone Ridge. Cercocarpiis ledifoliiis stands occur across the northern Great Basin to southeastern Ore- gon. Many of these have been classified as belonging to a Cercocarpiis ledifoUiislA^^ro- pijron spicatum association. Cercocarpiis ledtfolius was observed to oc- cur mixed with Pimis ponderosa in the Doug- las Mountain area, south of the studv area. A single stand was located (TUN RIOIW S19 NE4) that has an overstory of Cercocarpiis ledifoliiis with an understory o{ Cercocarpiis montaniis . Scattered individuals oi Cercocar- piis ledifoliiis var. intricatiis also occur in the stand. Because this was the only stand of this sort observed, and no reference to similar vegetation could be found in the literature, it was not described as a separate association. 5. Artemisia noval Agropijron spicatum. — This association occurs exclusively on rela- tively calcareous parent materials (Browns Park Fm., Madison Limestone). It occupies a wide elevational range from 1700 to 625 m. occurring primarily on northerly facing slopes that are often steep (up to 35°). It occurs at lower elevations on mesa sides and the sides of draws, and at higher elevations it may occur in a broad band above the pinyon-juniper zone. Soils, in spite of developing on a calcareous parent material, have an average pH (7.94) and electrical conductivity (.38 mmhos/cm) not much different from those in pinyon-ju- niper woodlands on noncalcareous substrata (Table 4). The association consists of a low shrub layer of sparse (8%-20% cover) Artemisia nova (Table 4). At lower elevations (below about 1900 m) Atriplex confertifolia may commonly occur and occasionally be abundant. The herl) layer, with 15% -30% cover oi Agropyron spi- catum, often overtops the Artemisia nova, giving the association a grassland appearance (Fig. 2). In many stands with abundant domestic livestock grazing signs Koeleria cristata, or occasionally Poa sandbergii, is much more Table 4. Percent cover and constancy of shrubs and herbs, plot locations, and physical parameters. Plant association numbers correspond to those in the text. 5 = Artemisia noval Agropijron spicatum, 6 = Artemisia nova/Stipa comata. Table entries under each plot are percent canopy cover. Tr = trace quantities (less than .5% cover). Table entries under Cov/Con are average percent canopy on the left of the slash and percent constancy to the right of the slash. 100 is abbreviated to 99. Soil EC is soil electrical conductivitv; its measurement is discussed in the text. Township 9N ION 9N ION ION 9N 11 N 9N Range 102W lOlW lOlW lOlW lOlW lOlW lOlW lOlW Section S12 S14 S5 S9 S34 S2 S28 S3 Elevation (meters) 1730 2070 1860 2340 1915 1900 2100 1910 Aspect Slope (degrees) Soil pH N 35 8.10 NE 5 7.85 N 30 8.05 NE 10 7.75 Avg = 7.94 S 2 8.00 S 3 7.90 0 7.75 S 2 7.75 Avg-7.85 Soil EC (mmhos/cm) .38 ..39 .34 .42 Avg= ..38 .42 .37 .73 .,36 Avg= .47 Plant association number j— 5 1 r~ 6 1 Plot number 1 2 3 4 Cov/Con 1 2 3 4 Cov/Con Shrubs Artemisia nova 8 17 17 20 15.5/99 12 10 8 10 10.0/99 Atriplex confertifolia Ceratoides lanata 1 tr tr tr tr tr/50 tr/75 tr 1 tr/50 Gutierrezia sarothrae tr tr/25 Pediocactus simpsonii Artemisia tridentata tr tr tr tr/75 tr tr/25 ssp. wtjomingensis Tetradymia spinosa Opuntia polyacantha Ch rysothamnus viscidiflonis tr tr tr tr tr tr tr tr tr/75 tr/25 tr/50 tr/50 Graminoids Agropyron spicatum Poa sandbergii 25 5 20 3 18 3 20 2 20.8/99 3.3/99 tr 2 3 1.4/75 Vulpia octoflora Oryzopsis hijmenoides Koeleria cristata tr tr 3 tr 1 tr 4 tr/25 tr/75 2.0/75 tr tr 1 tr .6/99 Agropyron smithii tr tr/25 tr 1 tr/50 758 Great Basin Naturalist Vol. 45, No. 4 Table 4 Continued. Township Range 1 Section 9N ION 9N ION 10 N 9N UN 9N 02W 10 IW lOlW lOlW lOlW 10 IW lOlW lOlW S12 514 S5 S9 S34 S2 S28 S3 Elevation (meters) 1730 2070 1860 2340 1915 1900 2100 1910 Aspect N NE N NE S s — S Slope (degrees) SoilpH Soil EC (mmhos/cm) 35 8.10 5 7.85 30 8.05 10 7.75 Avg=7.94 2 8.00 3 7.90 0 7.75 7.75 Avg=7.85 .38 .39 ..34 .42 Avg= .38 .42 ..37 .73 ..36 Avg= .47 Plant association number 1 — 5 1 r" 6 1 Plot number 1 2 3 4 Cov/Con 1 2 3 4 Cov/Con Stipa comata Bromus tectorum 30 1 25 tr 20 tr 25 tr 25.0/99 .6/99 Sitanion htjstrix tr tr 1 tr/75 FORBS Phlox hoodii 4 tr 2 3 2.4/99 tr tr 2 2 1.3/99 tr/25 Arenaria hookeri 1 tr/25 tr Astragalus spatulatus tr tr/25 Arenaria fendleri tr tr/25 Erigeron pumilus tr tr tr tr tr/99 Allium textile tr tr/25 Haplopappus acaulis tr 1 1 .6/75 tr/25 Townsendia incana tr tr/25 tr Cymopterus fendleri tr tr/25 Zigadenus paniculatus tr tr tr/50 Crepis occidentalis tr tr/25 Arabis pulchra tr tr/25 Castilleja chromosa tr 1 tr/50 Eriogonum tumulosum tr tr/25 Astragalus missouriensis tr tr/25 Erysimum asperum tr tr tr/50 1 tr/25 Phtjsaria acutifolia tr tr tr tr/75 Penstemon yampaensis 1 tr/25 tr 1 tr tr/75 tr/25 Astragalus cluimaeleuce tr tr tr/50 tr Crypta n tlui fla va tr tr/25 Hymenoxijs acaulis tr tr/25 Linum lewisii tr tr tr/50 Penstemon osterhoutii tr tr tr/50 Caulanthus crassicaulis tr tr/25 tr tr/25 Agoseris heterophylla tr tr/25 Antennaria dimorpha tr tr/25 Lesquerella alpina tr 1 tr/50 tr/25 Schoenocramhc linifolia tr tr/25 tr Crepis )n(>(l(>cciisis tr tr/25 Sedum stcnopetalum tr tr/25 Arabis demissa tr tr/25 Eriogonum sp. tr tr/25 tr tr/25 Townsendia incana tr tr/25 Petradoria piimila tr tr/25 Androstcphinni hrevijlorum tr tr/25 Descurainia richardsonii 2 tr tr 1.0/99 Lappula redowskii tr tr tr tr tr/99 Sphaeralcea coccinea 1 1 tr ,9/99 Allium textile ti tr tr tr/75 Gilia sinuata ti tr/25 Delphinium nuttallianum ti tr tr/50 Ipomopsis putnila ti tr/25 Eriogonum oiali folium 1 tr/25 Penstemon jrcniontii tr tr tr/50 Lesijiurclla tudoviciana tr tr/25 Orolianche fasciculata tr tr tr/50 Cymopterus purpureus tr tr/25 October 1985 Baker, Kennedy: Colorado Presettlement Vegetation 759 abundant than Agropyron spicotum. Other species that tend to increase under hvestock grazing inchide Viilpia octoflora, Haplopap- pus acaulis, and many other herbs. Moder- ately grazed sites with lower grass cover tend to have more herbs. In Colorado the association has been ob- served only in western Moffat County (the study area), where it occurs from Browns Park north to near the Wyoming border. The asso- ciation occurs across the northern Great Basin from Wyoming (Thatcher 1959, Tweit and Houston 1980) to northern Nevada (Zamora and Tueller 1973), southern Idaho (Hironaka 1978, Johnson and Pfister 1982, Passey et al. 1982, Sharp and Sanders 1978), and California (Barbour and Major 1977). 6. Artemisia nova/Stipa comata. — This as- sociation occurs from 1890 to 2165 m in eleva- tion on nearly flat surfaces primarily on the Browns Park Fm. It may also occur on other parent materials on benches, mesa tops, and flat plains. Soils are similar to those of the A7i:emisia noval Agropyron spicatum associa- tion in terms of average pH (7.85) and electri- cal conductivity (.47 m mhos/cm) (Table 4). The association consists of a sparse stand of Artemisia nova (Table 4) scattered through a dense grass matrix oiStipa comata (Fig. 2). Domestic livestock grazing decreases Stipa comata and results in increases in Poa sand- bergii, Bromus tectorum, and forbs. Most stands now have Poa sandbergii dominant, though others have dense Artemisia nova with very little grass present, or no grass at all. The association has been observed in Col- orado only in western Moffat County, where it has been located only in the vicinity of Lime- stone Ridge, and in North Park in Jackson County some 200 km east of the study area. It is known to occur across the northern Great Basin to Nevada (Blackburn et al. 1969 a,b,c, Zamora and Tueller 1973) and California (Bar- bour and Major 1977). 7. Artemisia tridentata ssp. tridentatal Elymus cinereus. — This association occurs only on relatively flat stream floodplains with or without permanent surface water. Eleva- tions range from about 1980 to 2200 m, though the association has been observed at lower elevations outside the study area. The sub- strate is Quaternary alluvium. Soils are clayey, with an average pH of 7. 75 and electri- cal conductivity of .44 mmhos/cm (Table 5). The association (Fig. 3, Table 5) consists of a tall stand o{ Artemisia tridentata ssp. triden- tata, with 15%-25% cover. Elymus cinereus, with 20%-4()% cover, dominates the herb layer. Agropyron smithii , with 2%-5% cover, is usually present. The association is prone to invasion by ex- otic species. Those commonly present now include Poa pratensis, Bromus tectorum, Thlaspi arvense, Malcomia africana, and Melilotus officinalis . Domestic livestock graz- ing decreases Elymus cinereus and results in increases in the density oi Artemisia triden- tata and these exotic herbs. The environment occupied by the association may be modified from a relatively wide and shallow floodplain to a deep, steep-sided gully if land uses in the stream catchment area result in changes in the timing and intensity of runoff. The association occurs in scattered loca- tions throughout the study area. It is known to occur in scattered locations throughout Moffat and Rio Blanco counties (Baker 1982b) in Col- orado and occurs across the northern Great Basin in Utah (Pammel 1903), northern Ne- vada (Blackburn et al. 1971, Young et al. 1975), southern Idaho (Hironaka 1978, John- son and Pfister 1982), to California (Barbour and Major 1977), Oregon, and Washington (Hironaka 1978). 8. Artemisia tridentata ssp. vaseyanal Agropyron spicatum. — This association oc- curs only above about 2125 m in elevation in the study area, where it occupies upper slopes and ridge tops, occurring on the Laney Mem- ber of the Green River Fm. on flat to moder- ately steep slopes. Though many stands of this association were observed in the study area, only one stand was located that was sufficiently free of grazing effects to be usable for sampling. For this reason, this description must be consid- ered preliminary, but because the association is well known and well described in the litera- ture from other parts of its range, it was felt that inclusion of even minimal data from the study area would be beneficial. The associa- tion (Table 5, Fig. 3) consists of about 15% cover oi Artemisia tridentata ssp. vaseyana, with minor amounts oiCeratoides lanata and Opuntia polyacantha present. Agropyron 760 Great Basin Naturalist Vol. 45, No. 4 Table 5. Percent cover and constancy ofshrubs and herbs, plot locations, and physical parameters. Plant association numbers correspond to those in the text. 7 = Artemisia tridentata ssp. tridentata/Elymiis cinereus, 8 = Artemisia tridentata ssp. vaseijanal Agropyron spicatum, 9 = Artemisia tridentata ssp. toijomingensisi Agropyron smithii, 10 = Artemisia tridentata ssp. wyomingensisi Agropyron spicatum , 11 = Artemisia tridentata ssp. wyomingensis -Atriplex confertifolia-Grayia spinosa*/Stipa comata . Table entries under each plot are percent canopy cover. Tr = trace quantities (less than .5% cover). Table entries under Cov/Con are average percent canopy cover for all the plots in the association on the left of the slash and percent constancy to the right of the slash. 100 is abbreviated to 99. Soil electrical conductivitv (soil EC) is discussed in the text. Township 12N 11 N ION ION ION 10 N 10 N ION 9N 9N 9N 98W 97\V 97W lOlW lOlW 99W lOOW IDOW lOlW 10 IW lOlW Section S34 S3 S3 S12 S2 S9 S27 S35 S12 S18 S17 Elevation (meters) 2100 2030 2220 1885 2060 2165 2130 2100 1750 1790 1790 Aspect NE N N NE \V NE SE SE NE S\V SW Slope (degrees) 2 1 1.5 3 4 5 5 2 3 2 3 SoilpH 7.70 7.80 Avg = 7.75 7.95 8.05 7.85 Avg 7.95 7.80 7.80 7.65 Avg 7.75 8.15 8.20 7.70 Avg = 8.02 Soil EC (mmhos/cm) .45 .42 Avg ■- .44 .43 .19 .17 Avg ,18 .40 .31 .30 Avg ..34 .33 .37 .40 Avg= .37 ber 7 Co /Con 8 1 r~ 1 9. t:o\ /C,iii 1 2 10. 3 Ccn /Con r- 2 11- 3 Plot number r~ Cov/Con .0/99 Shrubs Artemisia tridentata ssp. tridentata 15 25 20.0/99 Chrysothamnus nauseu^us 5 1 3.0/99 Sarcobatus venniculatus tr tr/50 Chrysothamnus viscidiflorus 4 tr 2 3/99 tr Opuntia polyacantha tr tr/50 tr Artemisia tridentata ssp. vaseyana 15 Ceratoides lanata tr tr Artemisia tridentata ssp. wyomingensis 8 Tetradymia spinosa 2 tr 1.3/99 Atriplex confertifolia 2 tr 1.3/99 2 .7/99 4 4 3 3.7/99 Artemisia spinescens tr tr/50 Atriplex gardneri tr tr/50 Grayia spinosa 3 tr 3 2.3/99 Graminoids Elymus cinereus 40 20 30.0/99 Poapratensis tr 1 .8/99 Agropyron smithii 2 5 3.5/99 tr 20 15 17..5/99 3 tr 12/67 Bromus tectorum tr tr/.50 6 tr 1 2.5/99 Sitanion hystrix tr tr/,50 tr tr/.50 tr tr/33 1 tr tr .7/99 Oryzopsis hymenoides tr tr/.50 tr tr tr/.50 1 tr tr/67 tr tr tr/67 Agropyron spicatum 20 15 20 25 20 0/99 Poa sandbergii 3 1 2 1..5/99 3 4 4 3.7/99 2 tr 1 1.2/99 Koeleria cristata tr 2 .7/33 Stipacomata tr tr/50 12 25 25 20.7/99 Poafendleriana tr tr/.33 tr tr/33 Vulpia octoflora 4 tr 5 3.2/99 FORBS Thlaspi arvense 3 tr 1.8/99 Cirsium vulgare tr tr/.50 Cryplantha flavoculata tr tr/50 tr tr tr/67 Mertensia ohlongifolia 3 tr 1.8/99 Schoenocramhe tinifulia 1 tr/50 tr tr/.33 Descurainia sophia tr 1 .8/99 Taraxacum officinale tr tr/50 Atriplex sp. tr tr tr/99 Malcomia africana tr tr tr/99 Melilotus officinalis tr tr/50 Grindelia sp. 3 1.5/50 Penstemon fremontii tr tr/50 tr tr tr/.50 Ipomopsis conge.ita Ir tr/50 tr tr/33 Vicia americana tr tr/50 Stanleya pinnata tr tr/.50 Thelypodiopsis elegans tr tr/,50 Allium textile tr tr tr tr/99 tr tr/.33 tr tr/33 Eriogonum ovalifolium 1 tr tr/3:3 Arenaria fendleri tr Linum lewisii tr Commandra umhellata tr Haplopappus acaulis 1 ,3 tr 1.2/67 Phlox hoodii 2 tr 1 .8/99 tr tr 2 1.0/99 tr tr/33 Draba oligosperma tr Astragalus spatulatu Eriogonum sp. Astragalus purshii Toumsendia incana October 1985 Baker, Kennedy Colorado Presettlement Vegetation 761 Table 5 continued. Town .ship 12 N UN ION ION ION ION ION ION 9N 9N 9N Range 98W 97W 97W lOlW 10 1 W 99W lOOW lOOW lOlW lOlW lOlW Section S34 S3 S3 S12 S2 S9 S27 S35 S12 S18 S17 Elevation (meters) 2100 2030 2220 1885 2060 2165 2130 2100 1750 1790 1790 Aspect NE N N NE W NE SE SE NE SW SW Slope (degrees) 2 1 15 3 4 5 5 2 3 2 3 Soil pH 7.70 7.80 A vg 7.75 7.95 8.05 7.85 A g 7.95 7.80 7.80 7.65 Avg 7.75 8.15 8.20 7.70 Avg 8.02 Soil EC (mmhos/cni) .45 .42 A vg ,44 .43 .19 17 A g .18 40 ..il .30 Avg ..34 .33 .37 40 Avg ..37 Plant association number 1 Y 1 r- 2 ( AKiV.nn 10. Cov/Con r" 1 2 3 Plot number 2 1 C:ov/Con 1 2 3 1 Cov/Con Phlox Umtiifolia tr tr tr tr/67 Arahis dcnmsa tr Erysimum asperum tr Sphaeralcea coccinea tr tr tr/5() tr tr/33 Castilleja chromosa tr tr tr/50 1 tr/33 Descurainia richardsonii tr tr tr/99 tr tr lr/67 tr 1 tr/67 Cymopterus bulbosus tr tr tr/99 Zigadenus panicutatus tr tr/50 Astragalus chanmeleuce tr tr tr/99 Lappula redowskii tr tr/50 tr tr tr/67 Physaria acutifolia tr tr/50 Cryptantha sp. tr tr/50 Chaenactis douglasii tr tr/50 Arabis pulchra tr tr/50 Crepis modocensis tr tr/50 Xylorhiza venusta tr tr/50 Machaeranthera grindelioides tr tr/50 Astragalus detritalis 2 .7/33 Trifolium gymnocarpon 1 tr tr .7/99 Lomatium sp. tr tr/33 Crepis occidentalis tr tr tr tr/99 Balsamorhiza hookeri var. hispidula 4 1.3/33 Erigeron divergens tr tr tr/67 Astragalus megacarpus tr tr/33 Cordylanthus ramosus tr tr/33 Delphinium nuttallianum tr tr/.33 Ipomopsis pumila tr tr tr tr/99 Agoseris heterophylla tr tr tr/67 Erigeron pumilus tr tr/33 Lepidium densiflorum tr 1 tr/67 tr tr/33 Mentzelia albicaulis "■ tr/33 spicatum, with about 20% cover, dominates the understory. Poa sandbergii commonly has 2%-3% cover. Domestic hvestock grazing has altered the composition of nearly all stands in the study area, resulting in the decline or loss of Agropyron spicatum and widespread replace- ment by Poa sandbergii, accompanied by in- creased density of Artemisia tridentata ssp. vaseyana. Herb density may also be very high. The association has been located in Colo- rado only in the study area and in Middle Park, Grand County (Terwilliger and Tiede- man 1978). It is known to occur from western Wyoming (Beetle 1961) to Idaho (Hironaka 1978, Johnson and Pfister 1982) and Oregon (Hironaka 1978). In the study area it has been observed primarily in the Vermillion Bluffs and Sevenmile Ridge area. 9. Artemisia tridentata ssp. wyomingensisi Agropyron smithii . — This association occurs on slight convexities on flat to gently rolling plains formed in shales of the Mancos Fm., Green River Fm. , or Wasatch Fm. It occurs in the study area from 1830 to 2130 m in eleva- tion. Soils have an average pH of 7.95 and electrical conductivity of .19 mmhos/cm (Table 5). Over a broad area the association occurs in an alternating mosaic with the Atriplex gardnerilOryzopsis hymenoides as- sociation, which occurs on more saline, more clayey soils, often in gentle concavities, on the rolling plains. Boundaries between these two associations may be very abrupt, apparently reflecting abrupt changes in soil properties. The association consists of a sparse stand of Artemisia tridentata ssp. wyomingensis (Table 5, Fig. 3) scattered through a grass matrix. Agropyron smithii has 15%-20% cover. Poa sandbergii commonly has l%-2% cover. Domestic livestock grazing may decrease Agropyron and result in increases in 762 Great Basin Naturalist Vol. 45, No. 4 Fig. 3. {a) Artemisia tridentata ssp. tridentata/Elymiis cinereus, (h) Artemisia tridentata ssp. vaseijanalAgropijron spicatum, (c) Artemisia tridentata ssp. wyomingensisl Agropyron smitliii, (d) Artemisia tridentata ssp. wyomingensis/ Agropyron spicatum, (e) Artemisia tridentata ssp. wyomingensis- At riplex confertifolia-Grayia spinosa*/Stipa co- mata , (f) Atriplex confertifolial Agropyron spicatum, (g) Afrip/ex confertifolialElymus salina, (h) Af n>/e.v confertifolia/ Stipa comata. October 1985 Baker, Kennedy: Colorado Presettlement Vegetation 763 Arte7nisia density, but, in general, probably due to the rhizomatous growth form of A^ro- pijron smithii, the association is moderately resistant to alteration from domestic grazing. Many stands, however, have lost Agropyron smithii dominance completely and now have the less palatable grass Poa sandbergii as the understory dominant. This is a very common association within the study area, but it is found only north and east of the Browns Park Formation area that occurs in the vicinity of Limestone Ridge and in Browns Park. It has been located in Colo- rado in the Piceance Basin in Rio Blanco County (Baker 1982b) and in Middle Park in Grand County (Tei-williger and Tiedeman 1978), and it probably occurs in other areas of northwestern Colorado. It also occurs in west- ern Wyoming (Johnson and Pfister 1982) and in western New Mexico (Donart et al. 1978). It has not been reported from other western states. 10. Artemisia tridentato ssp. wyomingen- sislAgropyron spicatum. — This association may occur in the study area from about 1980 to 2440 m in elevation, on gently rolling slopes and flat benches, on a variety of parent materi- als including the Laney Member of the Green River Fm. and the Browns Park Fm. Soils have an average pH of 7.75 and an electrical conductivity of .34 mmhos/cm (Table 5). The association consists of about 10%-20% cover o( Arte7nisia tridentata ssp. wyomingen- sis (Table 5, Fig. 3), with an understory of 15%-25% cover oi Agropyron spicatum. Poa sandbergii typically has 3%-5% cover. Domestic livestock grazing may decrease Agropyron spicatum and result in increases in Poa sandbergii, Bromus tectorum, and Te- tradymia spinosa . Many stands now have Poa sandbergii dominant and appear superficially similar to overgrazed stands of the Artemi- sia tridentata ssp. wyomingensisi Agropyron smithii association. Usually it will be possible to find some plants of Agropyron smithii or Agropyron spicatum in even the most altered stands, allowing identification of the appropri- ate former association. These observations on the effect of domestic grazing are similar to those reported for the association in other areas (Mueggler and Stewart 1980, Tweit and Houston 1980). The association is known in Colorado from northern Larimer County (Hess 1981), North Park in Jackson County (Smith 1966), Middle Park in Grand County (Terwilliger and Tiede- man 1978), and the study area in northern Moffat Covmty and has been observed by the senior author in southern Routt County. Within the study area it was observed primar- ily along Vermillion Bluff's and north toward Powder Wash. The association occurs across the northern Great Basin from western Wyo- ming (Tweit and Houston 1980) and Montana (Mueggler and Stewart 1980) to Idaho (Hi- ronaka 1978, Johnson and Pfister 1982) and Oregon (Hironaka 1978). 11. Arteinisia tridentata ssp. ivyomingen- sis-Atriplex confertifolia-Grayia spinosa*/ Stipa comata. — This association is restricted to sandy soils formed in the Browns Park Fm. It occurs on gently rolling hills, flat benches, and plains from 1700 to 1980 m in elevation. It may also occur on sandy hummocks and con- vexities in an area of finer-textured soils. Soils have an average pH of 8.02 and an electrical conductivity of .37 mmhos/cm (Table 5). The association consists of a mixed shrub layer (Table 5, Fig. 3), with Artemisia triden- tata ssp. wyomingensis generally most abun- dant (10%-20% cover) but Atriplex conferti- folia usually co-dominant. Grayia spinosa may be uncommon or very abundant, but it is always present. The herb layer is dominated by Stipa comata , with 12%-25% cover. Domestic livestock grazing decreases Stipa comata and results in increases in the exotic grass Bromus tectorum and other annual weeds. Most stands in the range of the associa- tion now are dominated by Bromus tectorum. A soil cryptogam layer, which has about 5% cover on relatively ungrazed sites, is absent on more grazed sites, probably due to tram- pling. The association is currently known only from Browns Park in western Moffat County, Colorado. It may extend into Utah in Browns Park. It was, prior to livestock grazing, the predominant vegetation type in Browns Park, occurring over a large area. 12. Atriplex confertifolial Agropyron spi- catum.— This association occurs in the study area from about 1950 to 2200 m in elevation, most often on northerly facing slopes, but also on other aspects. Slopes are shallow to moder- 764 Great Basin Natur.\list Vol. 45, No. 4 ately steep (up to about 25°). It occurs on side-slopes of draws and on gently sloping benches on two parent materials in the study area: the Cathedral Bluffs Tongue of the Wasatch Fm. and the Laney Member of the Green River Fm. Soils have an average pH of 7.90 and an electrical conductivity of .31 (Table 6). The association has a sparse shrub layer com- posed of 4%-8% cover of Atripkx confei-fifolia (Table 6, Fig. 3) and often a small amount of Ceratoides lanata. Agropyron spicatwn domi- nates the herb layer with 15%-30% cover, giv- ing the association a grassland appearance. Domestic livestock grazing decreases Agropyron spicatum, resulting in increases in Poa sandbergii and Tetradymia spinosa . Many of the stands of this association have been grazed primarily by sheep. Often, sheep-grazed stands have low forb density and Atriplex confertifolia plants have very poor vigor, with part of the crowns dead. Agropyron spicatum plants tend to have bet- ter vigor and greater cover on most sheep- grazed areas than on cattle-grazed sites. The association has not been reported to date from outside the study area, where it is found only in the northern part of the area between Vermillion Creek and the Little Snake River. Since the association occurs within a few miles of Wyoming, and similar habitat occurs there, it is likely that it will eventually be found in Sweetwater County. 13. Atriplex confertifolia/ El ymus salina. — This association occurs from 1950 to 2130 m in elevation in the study area on shallow to steep slopes with a northerly aspect. Parent materials include the Cathedral Bluffs Tongue of the Wasatch Fm. and the Laney Member of the Green River Fm. Soils are the most saline observed in the study area, with an average electrical conductivity of 1. 16 mmhos/cm and pH of 7.77 (Table 6). Soils characteristically are shaley, with a surface layer of sandstone fragments. This sandstone surface layer com- monly occurs throughout the range of the as- sociation. The association contains a sparse shrub layer dominated by Atriplex confertifolia with 5%-10% cover (Table 6). Sarcohatus vermicula- tus, Artemisia tridentata ssp. wyomingensis, and Ceratoides lanata are often present in small quantities. Elyimis salina dominates the herb layer with 20% -30% cover. When flowering, this grass may ol)Scure the shrub layer, giving the association a grassland appearance (Fig. 3). Domestic livestock grazing decreases Ely- miis salina and results in increases in Atriplex confertifolia and Poa sandbergii. The association is now known to occur in Colorado in scattered localities from northern Montrose County and Delta and Mesa coun- ties, where it occurs at the boundary between Mancos shale and Mesa Verde Group sand- stones (Baker, unpublished data) to the Piceance Basin in Rio Blanco County on Green River Fm. (Baker 1982b) and north to the study area. Within the study area it occurs in the area between Vermillion Creek and the Little Snake River. It has not been reported outside Colorado to date, but it has been ob- served by the senior author in eastern Grand Table 6. Percent cover and constancy of shrubs and herbs, plot locations, and physical parameters. Plant association numbers correspond to those in the text. 12 = Atriplex confei-tifolial Agropijron spicatum , 13 = Atriplex confertifolial Elymus salina , 14 = Atriplex confertifolia/Stipa comata . Table entries under each plot are percent canopy cover. Tr = trace quantities (less than .5% cover). Table entries under Cov/Con are average percent canopy cover for all the plots in the association on the left of the slash and percent constancy to the right of the slash. 100 is abbreviated to 99. Soil electrical conductivity (soil EC) is discussed in the text. Township ION ION ION ION ION ION ION ION 12 N 12 N 12 N 12 N Range 99W 99W 1(X)W lOOW 96VV lOOW lOOW 96\V 99VV 98VV 98\V 96VV Section S5 S6 S22 S13 S7 S15 S22 S7 S25 830 S25 S20 Elevation (meters) 2070 2050 1960 2190 2090 1965 1980 2065 2100 2090 2120 2170 Aspect NNE N\V NE NW NE NE NW NE _ _ NE N Slope (degrees) 5 10 20 20 15 20 25 5 _ _ 5 3 Soil pH 7.95 7.95 7.90 7.80 _ Avg 7.90 7.85 7.75 7.70 Avg 7.77 7,95 8,05 7,85 7.75 Avg 7.90 Soil EC (mmhos/cn.) .25 .28 ..38 .34 A%g .31 1.43 .36 1 70 Axg 1.16 .34 .32 .46 03 Avg .29 Plant a.ssociation nun her ^_ 12_ 1 — 13 _ 1 1 — 14_ Plot number 1 2 :5 4 5 Co /Con 1 2 .3 Cov/Con 1 2 .3 4 Cov/Con Shrubs Atriplex confertifolia Tetradymia spinosa Artemisia pedatifida Atriplex gardneri Ceratoides lanata .6/80 tr/20 tr/20 tr/60 tr/25 tr/75 1.3/99 October 1985 Baker, Kennedy: Colorado Preseitlement Vecetation (65 Table 6 continued. Township ION ION ION ION ION ION ION ION 12 N 12 N 12 N 12 N Range 99W 99W lOOW lOOW 9fi\\' lOOW lOOVV 96VV 99VV 98W 98\V 96W Section S5 S6 S22 S13 S7 815 S22 S7 S25 S30 S25 S20 Elevation (meters) 2070 2050 1960 2190 2090 1965 1980 2065 2 KM) 2090 2120 2170 Aspect NNE NV\' NE NW NE NE NW NE _ _ NE N Slope (degrees) 5 10 20 20 15 20 25 5 — — 5 3 Soil pH 7.95 7.95 7.90 7.80 — A 'S 7.90 7.85 7.75 7.70 A K " 77 7.95 8.05 7.85 7.75 AvK 7.'anadian prairie provinces. III. Acjuatic and semi-aquatic vegita- tion. Part 2: Freshwater marshes and bogs. Pluto- coenologia. 10(4):401-423. October 1985 Baker, Kennedy: C()ix)r.\do PRESErrLEMENX Vecetation 777 Miller, R F , F A Branson, 1 S McQueen, and C T Snyder. 1982. Water relatioii.s in .soils as related to plant communities in Ruby Valley, Nevada. J. Range Mgmt. 35:462-468. MUEGGLER. VV. F., AND W. L. STEWART. 1980. Grassland and shrubland habitat types of western Montana. USDA For. Serv. Gen' Tech. Rep. INT-66, In- termt. For. and Range Expt. Sta., Ogden, I'tah. 154 pp. NiCHOL, A. A. 1937. The natural vegetation of Arizona. University of Arizona, Agric. Expt. Sta. Tech. Bull. 68:177-222. Nightingale, W. T 1930. Geology of Vermillion Creek gas area in southwest Wyoming and northwest Colorado. Amer. Assoc. Petrol. Geol. Bull. 14:1013-1040. Pammel, L. H 1903. Some ecological notes on the vegeta- tion of the Uintah Mountains. Proc. Iowa Acad. Sci. 10:57-68. Passey, H. B , V. K Hugie, E W Williams, and D. E. Ball. 1982. Relationships between soil, plant community, and climate on rangelands of the In- termountain West. USDA Soil Cons. Serv. Tech. Bull. 1669. 123 pp. Penfound, W. T. 1953. Plant communities of Oklahoma lakes. Ecology 34:561-583. Peterson, J. S., and W L Baker 1983. Inventory of selected areas in Colorado. Unpublished report to the Colorado Natural Areas Prog., Dept. of Nat. Resources, State of Colorado, by Colorado Natural Heritage Inventory, Denver. Pfister, R D. 1982. Designing succession models to meet management needs. Pages 44-53 in J. E. Means, ed.. Forest succession and stand develop- ment research in the northwest. Proceedings of the symposium held 26 March 1981 at Corvallis, Oregon. Forest Research Laboratory, Oregon State University, Corvallis. Pfister, R. D., B. L. Kovalchik, S. F. Arno, and R. C. Presby. 1977. Forest habitat types of Montana. USDA For. Serv. Gen. Tech. Rep. INT-34, In- termt. For. and Range Expt. Sta., Ogden, Utah. 174 pp. Richards, L. A., ed. 1954. Diagnosis and improvement of saline and alkali soils. USDA Handbook 60. 160 pp. Rowley, P D , O Tweto, W R Hansen and P E Car- rara. 1979. Geologic map of the Vernal 1 degree X 2 degree quadrangle, Colorado, Utah, and Wyo- ming, uses Misc. Field Studies Map MF-1163. Sears, J. D. 1924. Geology and oil and gas prospects of part of Moffat County, Colorado, and southern Sweetwater County, Wyoming. USGS Bull. 751-G. Sharp, L A , and K. D Sanders 1978. Rangeland re- sources of Idaho. University of Idaho Wildl. and Range Expt. Sta. Contr. 141. 74 pp. Smuh, E. L., Jr 1966. Soil-vegetation relationships of some Artemisia types in North Park, Colorado. Unpublished dissertation, Colorado State Univer- sity Fort Collins. Soltanpour, p. N., and S. M Workman. 1981. Soil-test- ing methods used at Colorado State University Soil Testing Laboratory for the evaluation of fertil- ity, salinity, sodicity, and trace element toxicity. Colorado State Univ. Expt. Sta. Tech. Bull. 142. 22 pp. SviHLA, R D 1932. The ecological distribution of the manunals on the north slope of the Uinta Moun- tains. Ecol. Monogr. 2:47-73. Terwilliger, C, Jr , and J. A Tiedeman. 1978. Habitat types of the mule deer critical winter range and adjacent steppe region of Middle Park, Colorado. Unpublished report on file at Rocky Mt. For. and Range Expt. Sta,, Fort Collins, Colorado. 108 pp. Thatcher, A. P. 1959. Distribution of sagebrush as re- lated to site difierences in Albany County, Wyo- ming. J. Range Mgmt. 12:55-61. TwEiT. S. J., AND K. E. Houston. 1980. Gras,sland and shrubland habitat types of the Shoshone National Forest. Unpublished report on file at the Shoshone National Forest Headquarters. 143 pp. Vale, T R 1982. Plants and people: vegetation change in North America. Assoc. Amer. Geogr., Resource Publ. in Geogr., Washington, D.C. 88 pp. Welsh, S. L. 1957. An ecological survey of the vegetation of the Dinosaur National Monument, Utah. Un- published thesis. Brigham Young University, Provo, Utah. Wight, J. R., and H, G. Fisser. 1968. Juniperus os- teosperma in northwestern Wyoming, their distri- bution and ecology. University of Wyoming, Agric. Expt. Sta., Science Monogr. 7. 31 pp. Woodbury, A. M., S. D. Durrant, and S Flowers. 1960. A survey of vegetation in the Flaming Gorge Reservoir Basin. Dept. of Anthropology, Univer- sity of Utah, Anthropol. Papers 45. 121pp. Young, J. A., R. A. Evans, and P. T. Tueller. 1975. Great Basin plant communities — pristine and grazed. Pages 187-215 in R. Elston, and P. Headrick, eds., Holocene environmental change in the Great Basin. Nevada Archeol. Surv. Res. Paper 6. Zamora, B., and p. T. Tueller. 1973. Artemisia arbus- cula, A. longiloba, and A. nova habitat types in northern Nevada. Great Basin Nat. 33:225-242. WINTER PREFERENCE, NUTRITIVE VALUE, AND OTHER RANGE USE CHARACTERISTICS OF KOCHIA PROSTRATA (L.) SCHRAD James N. Davis' and Bruce L. Welch' Abstract. — A cafeteria-style study was conducted during the winter for two years with tame mule deer to determine if there were preferential differences between accessions of forage kochia (Kochia prostrata). Deer consumed significantly more of P.I. numbers 314929, 330708, and 356826 than any of the other accessions. Other plant adaptive characteristics and nutritive qualities are also reported. Forage kochia or perennial summer cypress (Kochia prostrata) is a widely distributed shrub native to the arid and semiarid regions of southern Europe and from northern Africa to Manchuria (Moghaddam 1978). Forage kochia was first introduced into the United States from Russia during the early 1960s (Keller and Bleak 1974). In its native Russia, it is commonly associated with Agropyron , es- pecially crested wheatgrass (A. cristatum) (Balyan 1972). There is an increasing interest in forage kochia as a desirable half-shrub for revegeta- tion work on many arid and semiarid western ranges. Ecotypic variation has been noted by many researchers (Balyan 1972, Francois 1976, Keller and Bleak 1974, McArthur and others 1974). Chromosome work indicates that the accessions we worked with included diploids, tetraploids, and hexaploids. The P.I. number 314929 was a diploid (McArthur 1984, per- sonal communication). This same accession has recently been released as "Immigrant" forage kochia for forage and erosion control on greasewood-shadscale, sagebrush-grass, and pinyon-juniper rangelands of the Intermoun- tain West (Stevens et al., in press). Differential preference of wintering mule deer among accessions of big sagebrush {Artemisia tridentata) and black sagebrush (A. nova) has been reported bv Welch et al. (1981). Also, Van Epps and McKell (1978) reported differential preference of domestic sheep for accessions of foui-wing saltbush (Atriplex canescens). The purpose of this study was twofold: first to determine the preference of tame mule deer for 13 accessions of K. prostrata grown in a uniform garden, and second to report the results of research concerned with the nutri- tive value and use of K. prostrata. Methods Four tame mule deer (one buck and three doe) were used in a cafeteria-style preference study for two winters, 1978 and 1979. The second year, three of the four deer were the same as the first year. Throughout the study, the deer were given free choice of their spe- cially formulated and pelleted feed, alfalfa hay, rolled barley, and water. Selected accessions of forage kochia (Table 1) were air dried and clipped into 6 to 10 cm lengths. Samples were randomly assigned to 1 gal plastic buckets placed in a row in a rack in the deer pen. After 24 hours, each bucket was weighed and refilled with 120 g of clipped forage and again randomly placed in the rack. The test ran for 10 consecutive days each win- ter. Analysis of variance was used to determine if there were significant differences between treatment means. Newman-Keuls multiple means test was used to determine the signifi- cant differences between indixidual means. Results Deer consumed significantly more of some accessions tlian others (Table 1). Deer pre- Utah State Division of Wildlife Resources and Ii Laboratory, 735 North 500 East. Provo. Utah 8460 ch Station, USDA Forest Service. Osden, Utah 84401. stationed at the Shrub Sci. 778 October 1985 Davis. Welch: Kochia Phosthata 779 Table 1. Deer preference for selected accessions oi Kochia prostrata , phnit origin of K. prostrate accessions used in this study. rockiction lunnhers, soil types, and Grams/da P.I. number Soil type Location 53. Q''* 314929 43.4^ 330708 39.4" 356826 13.3'' ** 13.2'' 356818 12.8'' 356819 8.6'' 356823 6.8'' 356822 3.4" 356825 2.8" 356820 0.9" 356817 0.9" 356824 0.1" 356821 ** Stavropol, Russia ** Tehran, Iran Salty Actobinsk, Ural Mountains, Russia ** Yini Dudar, Russia ('lay Actobinsk, Aral Sea, Russia Salty Actobinsk, Aral Sea, Russia Sandy Actobinsk, Russia Clay Ural Mountains, Russia Clay Actobinsk, Russia Sandy Actobinsk, Aral Sea, Russia Salty Actobinsk, Aral Sea, Russia Salty Actobinsk, Russia Salty Actobinsk, Aral Sea, Russia *Valvies sharing the same letter superscript are not significantly different at the 9.5% level. ''Information not available. ferred P.I. numbers 314929, 330708, and 356826 over the other 10 accessions. These 3 accessions did not diflfer significantly. The less preferred group of 10 accessions also failed to show significant differences in their means. It should be noted that P.I. numbers 356817, 356824, and 356821 received less than a gram of use per day. Data indicate that preference by tame mule deer for accessions of forage kochia is highly variable. We have no reason to believe that preference of wild and tame mule deer for accessions o{ Kochia differs sig- nificantly (Wallmo and Neff 1970). Highly preferred accessions (P.I. 314929, P.I. 330708, and P.I. 356826) are the ones that should be used in reseeding efforts where grazing is one of the management objectives. Discussion Because of great ecotypic variation, forage kochia appears to be a useful range plant for improvement of our semiarid ranges. Some forage kochia ecotypes are quite salt tolerant. Francois (1976) tested two accessions for three years and found both to be salt tolerant, but one was significantly more productive at all salinity levels. The highest salinity level was twice that normally found in a greasewood community (Gates et al. 1956). Forage kochia is drought tolerant. Moghad- dam (1978) describes transplanting it into ar- eas of Iran where annual precipitation was only 150 mm. He further reported that forage kochia's productivity and persistence was su- perior to fourwing saltbush. In Russia forage kochia is cut as "cypress hay" and fed to sheep, goats, and horses in regions having as little as 165 mm annual precipitation (Balyan 1972). The nutritive value of forage kochia has re- ceived some attention. Davis (1979) reported that the oxalate — a potential animal poison — level in forage kochia was lower than levels in fourwing saltbush and winterfat (Ceratoides lanata). Welch and Davis (1984) reported the mean in vitro digestibility of the 13 accessions used in this study was 32.2% of dry matter (Table 2). In comparison to other winter for- ages, forage kochia ranks low in digestibility. Seasonal crude protein content was also de- termined for the accessions of forage kochia (Table 3). Mean crude protein was highest during July (14.4%) and November (10.7%) for "upper" stems. For the "lower" stems, highest mean protein was May (12.8%) and July (14.0%). Table 4 lists the average winter levels of crude protein o{ Kochia compared to other range plants. Forage kochia tends to green up earlier in the spring than many other range plants. Crude protein levels in new spring growth ranged from 12.1% to 21.8% (Davis and Welch 1984). Forage kochia could be an important and useful subshrub on saline, and alkaline soils of our arid and semiarid ranges in the western United States. It grows well on a wide range of soil textural classes, sandy to fine clays. It is well adapted to areas occupied by juniper- pinyon, big sagebrush, greasewood, and shadscale. It grows fairly rapidly, usually pro- ducing seed the first year. Forage kochia could provide important sources of protein 780 Great Basin Naturalist Vol. 45, No. 4 Table 2. In vitro digestibility of winter range forages. Dry matter Winter forage digestibility/% Range Reference* Aspen .57.4 1 Big sagebrush 57.3 (49.9-67.0) 2, 3, 4, 5, 6, 7, 10 Bud sagebrush 57.0 8 Woods rose .54.5 1 Sand dropseed grass ,53.2 8 Black sagebrush .53.7 (.53.1-54.0) 3, 8, 14 Rose hips 51.1 6 Indian ricegrass 50.0 (45.7-,54.2) 8,10 Bluestem wheatgrass 45.5 10 Curl-leaf mountain mahogany 49.1 (44. 7-53.. 5) 4,6 Galleta 48.2 8 Needle-and-thread 47.0 10 Bluebunch wheatgrass 45.5 10 Common winterfat 44.7 8 Rubber rabbitbrush 44.4 10 Shadscale 43.4 8 Western snowberry 41.0 1 Chokecherry 38.8 (26.,3-51.,3) 1,11 Fourwing saltbush ,38.3 9 Cliffrose 37.6 12 Desert bitterbrush ,35.8 12 Forage kochia (P.I. 330708) 32.4 13 Forage kochia (mean) 32.2 (24.2-36.1) 13 Forage kochia (P. I. 314929) 31.0 13 Apache-plume 29.8 12 Forage kochia (P. I. .3.56826) 28.3 13 Gambel oak 28.1 2 Antelope bitterbrush 25.4 (19.8-30.0) 4, 6, 10, 12 True mountain mahogany 24.3 (20.0-28.5) 4,6 *1. Dietzl972 2. Kufeld et al. 1981 3. Sheehyl975 4. Umess et al. 1977 5. Wallmo et al. 1977 6. Welch and Pederson 1981 7. Pederson and Welch 1982 8. Welch et al. 1983b 9. Welch and Monsen 1984 10. Ward 1971 11. Uresk and Mcssner 1975 12. Welch et al. 1983a 13. Welch and Davis 1984 14. Behan and Welch, in press Table 3. Crude protein content of "upper" and "lower" parts of the same stems oi Kochia prostrata through a year. Data expressed as percent of dry matter. Each data point is a mean of 13 accessions. Month Stem part Dec. Jan. Feb. Mar. Apr. May July Nov. Upper Lower 5.9"* 8.2^ 6.1' 8.3-' 6.r 8.7^ 5.2^ 8.1" 5.7" 9.8- 5.8» 12.8'' 14.4'' 10.7'' 8.6" •Values sharing the same letter superscript are not significantly different at the 95% level. and carotene (Davis 1979, Davis and Welch 1984) and help introduce variety to many monoculture seedings of crested wheatgrass. Otsyina (1983) reported that during a fall graz- ing study sheep showed a high preference for forage kochia in shrub-grass pastures. He also reported that crude protein contents of sheep diets on forage kochia-crested wheatgrass pas- tures were significantK higher than sheep di- ets on pure crested wheatgrass (10.6% vs. 1.5%). Forage kochia shows its greatest po- tential for use with grass ranges in the fall and October 1985 Davis, Welcii: Kochia Phostrata 781 Table 4. Winter crude protein content of selected ranj^e plants Crude protein Range plant {% dry matter) Range Reference* Crested wheatgrass (green regrowth) 15.0 16 Black sagebrush 11.7 13 Big sagebrush 11.4 (9.9-14.2) 1, 2, 3, 4, 6, 8, 9, 10, 13, 16, 19 Curlleaf mountain-mahogany 10.1 (9.6-10.6) 3.7 Fourwing saltbush 9.6 12 Forage kochia (P.I. 330708) 8.9 20 Chokecherrv 8.7 (7.6-9.9) 3,5,11,17 CliflFrose 8.6 (8.4-8.8) 5, 14 Desert bitterbrush 8.5 (8.0-9.0) 3, 14 Rocky mountain juniper 8.4 1 Forage kochia (P. I. 314929) 8.4 20 Antelope bitterbrush 7.8 (6.7-9.1) 1,3,4,7,8,9, 11, 14 True mountain-mahogany 7.8 (7.2-8.4) 1, 5, 9 Rubber rabbitbrush 7.8 (5.9-7.8) 1, 11 Shadscale 7.7 10 Forage kochia (P.I. 356826) 7.3 20 Gardner saltbush 7.2 10 Forage kochia (mean) 7.1 20 Utah juniper 6.6 (5.9-7.6) 3,5,7 Saskatoon serviceberry 5.9 (5.. 5-6. 2) 3,11 Woods rose 5.8 (5.4-6.1) 17, 18 Gambel oak 5.3 (5.1-5.4) 5, 19 Apache-plume 4.8 14 Crested wheatgrass 3.9 11 Native grass 3.6 3 Wildrye 3.2 15 Indian ricegrass 3.0 (2.5-3.5) 11,15 Dietz et al. 1962 Welch and McArthiir 1979 Tueller 1979 Bissell et al. 1955 Smith 1957 Smith 19.50 Smith 1952 Trout and Thiessen 1973 Medin and Anderson 1979 (data converted to dry matter basi National Academy of Sciences 1975 National Academy of Sciences 1958 Welch and Monsen 1981 Sheehy 1975 Welch et al. 1983a National Academy of Sciences 1964 Umess et al. 1983 Dietz 1972 Welch and Andrus 1977 Kufeld et al. 1981 Davis and Welch 1984 would improve forage quantity and quality on extensive crested wheatgrass seedlings in the Intermountain West. plants were grown, is cooperatively main- tained by these two agencies and by Utah State University and Snow College. Acknowledgments Federal funds for wildlife restoration were provided through Pittman-Robertson Project W-82-R, Job 1. Cooperators are the Inter- mountain Research Station of US DA Forest Service and Utah Division of Wildlife Re- sources. The Snow Field Station, where Literature Cited Balyan, G. a. 1972. Prostrate summer cypress and its culture in Kirghizia. Translated from Russian by Ed A. Elias. National Technical Information Ser- vice, U.S. Department of Commerce. 296 pp. BissELL, H D , B Harris, H. Strong, and F. James 1955. Digestibility of certain natural and artificial foods eaten bv deer in California. Calif Fish and Game 41:57-78. 782 Great Basin Naturalist Vol. 45, No. 4 Davis, A. M 1979. Forage quality of prostrate Kochia compared with three browse species. Agron J. 71:822-824. Davis,]. N. AND B L. Welch. 1984. Seasonal variation in crude protein of Kochia prostrata (L.). Pages 145-149 in A. R. Tiedemann, E. D. McArthur, H. C. Stutz, R. Stevens, and K. L. Johnson, compil- ers. Proceedings — Atriplex and related cheno- pods. USDA Forest Service. Gen. Tech. Rep. INT-172. 309 pp. DiETZ, D. R. 1972. Nutritive value of shrubs. Pages 289-302 in C. M. McKell, J. P. Blai.sdell, and J. R. Goodin, eds., Wildland shrubs — their biology and utilization. USDA Forest Service Gen. Tech. Rep. INT-1.494pp. DiETZ. D R . R H Udall, and L E Yeager. 1962. Chemical composition and digestibility by mule deer of selected forage species. Cache la Poudre Range, Colorado. Colorado Game and Fish Dep. Tech. Publ. 14.89 pp. Francois, L. E. 1976. Salt tolerance of prostrate summer cypress (Kochia prostrata). Agron J. 68:45.5-456. Gates, D. H., L. A Stoddart, and C W Cook 1956. Soils as a factor influencing plant distribution on salt-deserts of Utah. Ecol. Monogr. 26:15.5-175. Keller, W., and A T. Bleak. 1974. Kochia prostrata: a shrub for western ranges? Utah Science 34:24—25. Kufeld, R. C , M Stevens, and D C Bowden. 1981. Winter variation in nutrient and fiber content and in vitro digestibility of Gambel oak (Qiiercus gam- belii) and big sagebrush (Artemisia tridentata) from diversified sites in Colorado. J. Range Man- age. 34:149-151. McArthur, E. D., B. C. Giunta, and A P Plummer 1974. Shrubs for restoration of depleted ranges and disturbed areas. Utah Science 35:28-33. Medin, D E , AND A E Anderson. 1979. Modeling the dynamics of a Colorado mule deer population. Wildl. Monogr. 68. 77 pp. Moghaddam, M R 1978. Kochia prostrata — a plant ma- terial for range improvement in arid and semiarid regions. Rangemans J. 5:153-154. National Academy OF Sciences 1958. Composition of cereal grains and forages. Natl. Res. Counc. Publ. 585. 663 pp. 1964. Nutrient requirements of domestic animals. No. 5. Nutrient requirements of sheep. Natl. Res. Counc. Publ. 1193. 40 pp. 1975. Nutrient requirements of domestic animals. No. 5: nutrient requirements of sheep. 5th ed. Natl. Res. Counc. Publ. 74-899. 72 pp. Otsyina, M. R. 1983. Evaluation of shrubs as native sup- plements to cured crested wheatgrass pasture for sheep. Unpublished dissertation. Utah State Uni- versity, Logan. 128 pp. Pederson. J. C. AND B. L. Welch 1982. Effects of monoterpenoid exposure on ability of rumen inoc- ula to digest a set of forages. J. Range Manage. 35:500-502. Sheehy. D P 1975. Relative palatability of seven Artemisia taxa to mule deer and sheep. Unpub- lished thesis, Oregon State Universitv, Corvallis. 147 pp. Smith. A. D 1950. Sagebrush as winter food for nuilc deer. J. Wildl. Manage. 14:285-289. 1952. Digestibilitvofsome native forages for mule deer. J. Wildl. Manage. 16:309-312. 1957. Nutritive value of some browse plants in winter. J. Range Manage. 10:162-164. Trout, L E , and J L Thiessen 1973, Physical condition of range relationships of the Owyhee deer herd. Job Completion Report. Idaho Fish and Game Department, Boise. 37 pp. Tueller, P T 1979, Food habits and nutrition of mule deer on Nevada ranges. University of Nevada, Reno. 104 pp. Uresk. D W . AND H E Messner 1975. Constituents in in vitro solution contribute differently to dry mat- ter digestibility of deer food species. J. Range Manage. 28:419-421. Urness, p. J , A D S.mith, and R. K W.^tkins. 1977. Com- parison of in vivo and in vitro dry matter digestibil- itv of mule deer forages. J. Range Manage. 30:119-121. Urness, P J , D D. Austin, and L C. Fierro. 1983, Nu- tritional value of crested wheatgrass for wintering mule deer. J. Range Manage. 36:225-226. Van Epps, G. A., and C. M. McKell 1978, Major criteria and procedures for selecting and establishing range shrubs as rehabitators of disturbed lands. Pages 352-3.54 in D. N. Hyder, ed.. Proceed- ings— First International Rangeland Congress. 742 pp. Wall.mo, O C . L H Carpenter, W L Regelin, R. B. Gill, and D. L. Baker. 1977, Evaluation of deer habitat on a nutritional basis, J, Range Manage. 30:122-127. Wallmo, O C , and D J. Neff 1970, Direct observation of tamed deer to measure their consumption of natural forage. Pages 105-110 in Range and wildlife habitat evaluation— a research sympo- sium, USDA Misc. Publ. 1147. 220 pp. Ward, A L 1971. In vitro digestibility of elk winter forage in southern Wvoming, J. Wildl. Manage. 35:681-688. Welch, B L , and D Andrus 1977, Rose hips — a possi- ble high-energy food for wintering mule deer? USDA Forest Service Res. Note INT-221. 4 pp. Welch, B, L , and J N Davis. 1984. In vitro digestibility oi Kochia prostrata (L.) Schrad. Great Basin Nat. 44:296-298, Welch, B. L., and E D McArthur 1979, Variation in winter levels of crude protein among Artei7iisia tridentata subspecies grown in a uniform garden. J. Range Manage. 32:467-469. Welch. B, L., E D McArthur, and J N D.wis 1981, Differential preference of wintering mule deer for accessions of big sagebrush and for black sage- brush, J, Range Manage. 34:409-411. Welch, B L , and S B Monsen 1981, Winter crude protein among accessions of foui^wing salthush grown in a uniform garden. Great Basin \,it. 41:343-346. Welch, B. L., and S. B. Monsen, 1984. Winter nutriti\e value of accessions of fourwing ,saltl)ush [Atriph'x canescens) grown on a uniform garden, Paiges 138-144 in A. R. Tiedemann and K. L. John^()n, compilers. Proceedings — biology o{ AtripUx and related chenopods, USDA Forest Service (ien. Tech, Rep, INT-172, 309 pp. October 1985 Davis, Welch: Kochia Prostrata 783 Welch, B. L.. S. B. Monsen, and N L. Shaw 1983a, Nutritive value of antelope and de.sert bitter- brush, Stansbury cliflFrose, and apache-plume. Pages 173-185 in A. R, Tiedemann, and K. L, Johnson, compilers. Proceedings — research and management of bitterbrush and cliffrose in west- ern North America. USDA Forest Service Gen, Tech, Rep. INT-152, 279 pp. Welch, B L., and J. C. Pederson. 1981, In vitro digesti- bility among accessions of big sagebrush by wild mule deer and its relationship to monoterpenoid content. J. Range Manage. 34:497-500. Welch, B L , J. C. Pederson, and W P. Clary. 1983b, Abilitv of different rumen inocula to digest range forages, J, Wildl, Manage. 47:873-877. AGE, GROWTH, AND FOOD HABITS OFTUI CHUB, G7LA BICOLOR, IN WALKER LAKE, NEVADA James J. Cooper Abstract. — At Walker Lake, Nevada, tui chub were collected 1975-1977 for analysis of age, growth rate, and food habits. The fork length (FL)-scale radius (SR) relationship was linear and described by the equation FL = 4.44 + 3. 17 (SR). Age I, II, III, and IV chub were 1 16, 176, 218, and 242 mm fork length, respectively. Ma.ximum longevity was six years. The length weight relationship was defined by the log transformed linear equation log weight = -4.65 + 2.93 (log FL). Chub collected from pelagic regions ate mostly zooplankton, whereas chub collected from littoral areas had a diet of zooplankton and benthic organisms. Tui chub, Gila bicolor, is the most abundant of the three species of fish currently found in Walker Lake, Nevada. It is common to the Walker, Carson, Truckee, and Humboldt river systems of the Lahontan basin (La Rivers 1962). Various subspecies of tui chub occur in other endorheic basins in the drainages of pluvial lakes Railroad, Toiyabe, and Dixie, and lakes in the White Mountains in west central Nevada. Other forms occupy lake basins in California, south- eastern Oregon, and southeastern Washington (Hubbs et al. 1974). In Walker Lake tui chub are an important component of the ecosystem bioenergetics and are preyed upon heavily by the piscivorous Lahontan cutthroat trout, Sahno clarki henshawi (Cooper and Koch, 1984). The vast number of fish-eating birds that annually visit the lake are also predators of tui chub. Most of the life history information reported in the literature for the Lahontan form of tui chub has been collected from Eagle Lake, California, and Pyramid Lake, Nevada. Kucera (1978) and Kennedy (1983) studied the reproductive biology and growth of Pyramid Lake tui chub. Kimsey (1954) described the life history of the Eagle Lake tui chub popula- tion. Cooper (1978, 1982), working on Walker Lake, described various aspects of tui chub life history. Notes on the species can be found in other articles (Snvder 1917, La Rivers 1962, Vigg 1978, 1980, 1981, Galatetal. 1981, Galat and Vucinich 1983a, 1983b). The objectives of this study are to present data on the age, growth rate, and diet of tui chub from Walker Lake. Study Area Walker Lake, a remnant of pluvial Lake La- hontan, is in west central Nevada 209 km southeast of Reno. The lake has a surface area of 15,000 ha, is 25 km long and 9 km wide, and has a maximum and mean depth of 33 and 20 m, respectively. It is the second largest rem- nant of Lake Lahontan. The lake's drainage basin is endorheic and receives water from the eastern Sierra Nevada via the Walker River. Because Walker is a terminal lake, it has a relatively high total dissolved solids (TDS) content of 12,500 mg/1 that has increased rapidly in historic times. During the past 45 years the lake has had an average increase in TDS of 152 mg/1 per year, and the cutthroat trout sport fishery appears to be in jeopardy. The primary factor responsible for the increasing salinity has been surface evaporation exceeding tributary inflow; since 1915 the lake's elevation has dropped at an average rate of 0.58 m per year (Cooper and Koch, 1984). Agricultural and ur- ban diversion of the Walker River is hastening desiccation of the lake. Methods The scale method was used to anahze the age of tui chub at various sizes (Ricker I97I, Everhart et al. 1975). Scales were taken from the left side of the body above the lateral line and below the dorsal fin. In the laboratory scales were placed between two plastic slides and run through a roller press to form an impression. Scales were read using an Eber- Biological Sciences Center, Desert Research Institute, Box ri()22(), Reno, Nt 1 South Fall Street, Carson City, Nevada 89710. 784 October 1985 CoopeRiTuiChub 785 Table 1. Mean calculated fork length and moan cakiiiati-d annual growth increments for tui chub collected June-November 1976, Walker Lake, Nevada. Age class No. of fish Calculated fork length at end of each year of life (mm) I 46 113 II 16 124 187 III 31 117 172 220 IV 8 111 165 214 242 Grand mean 116 176 218 242 Increments of grovv'th 116 60 42 24 Number offish 101 55 39 8 bach microprojector and the distance fioni the focus to each annuhis and from the focus to the scale radius recorded. All age and growth cal- culations were performed using the computer program SHAD II (Nelson 1976). Stomach contents were analyzed two ways: percent occurrence and percent composition. In the former the number of stomachs in which each item occurred was recorded and expressed as a percentage of the total number examined. In the second method a represen- tative sample of a stomach content was placed in a Sedgewick Rafter counting chamber and strip counts made of individual food items at 40X magnification. Percent composition of each food item identified in the counts were calculated for the total sample. Comparisons were made between the feeding habits offish collected from pelagic surface (midlake) areas versus collections from littoral (< 5 m) regions of the lake. Tui chub stomachs appear as an enlargement in the anterior portion of the digestive tract, and it was at this point the contents were removed for food analysis. Fish were collected using multipaneled experi- mental gill nets with meshes of 3.8, 5.1, 6.4, 7.6, 8.9, and 10.2 cm stretched mesh mea- sure. A more complete description of the methods is given in Cooper (1978). Results and Discussion Age and Growth Scale samples from 101 tui chub were exam- ined for age and growth determinations. The fork length (FL) - scale radius (SR X 24) rela- tionship was linear (Fig. 1) and described by the regression equation FL = 4.44 + 3.17 (SR) (r = 0.93). Scale formation begins to oc- cur in Walker Lake tui chub between 25 and 20 30 40 50 60 70 80 80 100 110 120 SCALE RADIUS (mm 24X) Figure 1. Fork length-scale radius relationship for Walker Lake, Nevada, tui chub. 30 mm in length. Kimsey (1954) reported that scale formation occurred in tui chub from Ea- gle Lake, California, when the fry reached 20 to 25 mm in length. Annual growth of tui chub were analyzed from summer hatch to four years (Table 1). The major growth in length is achieved during the first year of life. Young of the year chub captured in late August at approximately 2 to 3 months of age ranged in length from 72 to 115 mm and averaged 91 mm. Kucera et al. (1978) reported that by two months of age Pyramid Lake chub were 48 mm long and by Septem- ber had attained a length of 122 mm. Eagle Lake young of the year only reach a length of 22-41 mm by September (Kimsey 1954). Fol- lowing this rapid growth during the first year of life, annual growth increments become pro- gressively smaller. 786 Great Basin Naturalist Vol. 45, No. 4 Table 2. Tui chub growth rates in five selected waterbodies. Fork length at end of each year of life (mm) Location 1 2 3 4 Reference Walker Lake, Nevada Pyramid Lake, Nevada Eagle Lake, California* Big Sage Reservoir, California East Lake, Oregon 116 123 74 58 43 176 172 125 103 75 218 215 187 140 114 242 259 242 157 149 This studv Kuceraet'al. (1978) Kimsev (1954) Kimsev and Bell (1955) Bird (1975) Standard length converted to fork length by a factor of 1. 12 600i CURVILINEAR / . 2.24 X o-» ft' »3, •/ 500- LINEAR / LOG W = -4. 65 * 2.93 (LOG FL) / 3 400- /. I o 300- 200- 100- .^^ ^,, '"' FORK LENGTH (mm) Figure 2. Length-weight relationship for Walker Lake, Nevada, tui chub. Maximum growth at an immature age can be explained by the fact that this age group does not contend with reproduction and thus all energy is expended in growth. Furthermore, maximum growth in length at an early age with a subsequent decline could be advanta- geous because most Lahontan cutthroat trout can only ingest smaller sized chub. Growth rates of tui chub in Walker and Pyramid lakes are higher than other nonter- minal lakes (Table 2). First-year growth in the two terminal lakes is over three times that of the relatively unproductive, high elevation East Lake, Oregon (Bird 1975). Eagle Lake chub exhibit slow growth their first year but by year IV are the same size as Pyramid and Walker fish. This too may be related to the lake's trophic state because Walker and Pyra- mid are the more productive (Himtsinger and Maslin 1976, Cooper and Koch 1984, Galat et al. 1981). Temperature may also be a factor iiecause winter temperatures are about 6 C warmer in Pyramid and Walker, which do not experience ice cover. Longevity of tui chub from Walker Lake is probably near four or five years of age. Maxi- mum age of 102 fish collected for age examina- tion was one six-year-old individual. This is similar to data collected from Pyramid Lake, where a sample size of 322 produced six age five, one age six, and one age seven (Kucera et al. 1978). The length-weight relationship can best be described by the linear log transformed equa- tion log W = -4.65 + 2.93 (log FL), where W = weight in grams and FL = fork length in millimeters (Fig. 2). Chub used for length- weight analysis ranged in size from 72 to 324 mm in length and 6.3 to 636.8 g in weight. The largest individual caught was a female (324 mm, 636.8 g) that had a young-of-the-year tui chub in its stomach. Walker tui chub are heavier in weight per unit length than Pyramid chub (Kucera et al. 1978), although there was only a slight differ- ence in length at any given age. This was corroborated by the higher mean coefficient of condition (K) of Walker (1.51) versus Pyra- mid (1.31) fish. Kimsey (1954) reported a mean K value of Eagle Lake tui chub of 1.92, the highest of all three lakes. Food Food items identified from 103 stomachs revealed Walker Lake tui chub to be omnivo- rous and highly opportunistic. Zooplankton was the most important item in the diet of chub captured in pelagic water b>' both occur- rence (97.4%) and composition (94.4%) (Table 3). Fish captured in littoral regions also fed on zooplankton but took significant (luantities of benthic material. The attached alga {Chid- ophora ), C'hironomidae larvae, and the gam- marid Utjalella aztcca were evidence of ben- thic feeding activity in the littoral catch. Zooplankton were the most important organ- October 1985 COOPER: Tui Chub 787 Table 3. Food of tui chub by percent frequency of occurrence habitats in Walker Lake, Nevada, in 1976 and 1977*. id percent j^nt composition collected from tw( Food item Occurrence % Composition % 'elagic Littoral Pelagic Littoral 97.4 63.9 94.4 30.6 10.3 88.5 2.5 44.5 5.1 29.5 0.5 12.9 0 0 0 2.9 0 27.9 0 9.1 2.6 0 2.6 0 Zooplankton Cladophora sp. Chironomidae HijaleUa azteca Bottom substrate Gila bicolor Pelagic n - 41 Littoral n =^ 62 isms in the diet of these fish, probably ckie to the small amount of littoral habitat in Walker Lake. These findings are consistent with the litera- ture in categorizing tui chub as opportunistic omnivores (Kimsey 1954, La Rivers 1962, Bird 1975, Langdon 1979, Galat and Vucinich 1983b), although the data suggest zooplank- ton may be more important in the diet of Walker Lake fish. Gill raker counts of Walker chub indicate an abundance of the fine-rak- ered pectinifer form, with only a small repre- sentation of an intermediate between pec- tinifer and the course-rakered obesa form (Cooper 1978, Vigg and Cooper, unpublished data). In Pyramid Lake both forms are abun- dant and exhibit differences in feeding behav- ior (Langdon 1979, Galat and Vucinich 1983a); fine-rakered chub consume more zooplankton and course-rakered chub more macroinverte- brates. However, as in this study, Langdon (1979) found little similarity in the diet of pec- tinifer sampled at the surface and bottom. It seems that obesa is more dependent on a bot- tom-feeding strategy than pectinifer is on pelagic feeding. This theory is supported by Pyramid Lake data showing that only pec- tinifer are found in open water, but both forms can be caught inshore on the bottom (Vigg 1978, Langdon 1979). Although there is little gill raker variation in Walker Lake chub, for whatever reason, pectinifer appear to have successfully filled the trophic niche unoccu- pied by obesa . When compared to lakes sup- porting obesa populations. Walker chub are, in general, feeding on the same food items and can modify their feeding strategy in accor- dance with food availability. This suggests that feeding strategy in lentic Great Basin chub populations may be just as dependent on behavioral and ecological characters as it is on morphological variation. Literature Cited Bird, F. H. 1975. Biology of the blue and tui chubs in East and Paulina lakes, Oregon. Unpublished the- sis. Oregon State University, Corvallis, Oregon. 165 pp. Cooper, J. J. 1978. Contributions to the life history of the Lahontan tui chub, Gila bicolor, in Walker Lake, Nevada. Unpublished thesis. University of Ne- vada, Reno. 89 pp. 1982. Observations of the reproduction and em- bryology of the Lahontan tui chub, Gila bicolor, in Walker Lake, Nevada. Great Basin Nat. 42:60-64. Cooper, J. J and D L. Koch. 1984. Limnology of a deser- tic terminal lake. Walker Lake, Nevada. Hydrobi- ologia 118:275-292. Everhart, H W , a W Eipper, and W D. Youngs. 1975. Principles of fishery science. Comstock Publ. As- soc, Cornell University Press. 288 pp. Galat, D L., E. L. Lider, S, Vigg, and S R. Robertson, 1981. Limnology of a large, deep, North American terminal lake. Pyramid Lake, Nevada, USA. Hy- drobiologia. 82:281-317. Galat, D. L. andN. Vucinich. 1983a. Food partitioning between young of the year of two sympatric tui chub morphs. Trans. Amer. Fish. Soc. 112:486-497. 1983b. Food of larval tui chubs, Gila bicolor, in Pyramid Lake, Nevada. Great Basin Nat. 43:175-178. HuBBS, C L., R. R. Miller, and L. C. Hubbs. 1974. Hy- drographic history of relict fishes of the northcen- tral Great Basin. California Acad. Sci. Memoirs 7:1-259. Huntsinger, K. R, andP. E. Maslin. 1976. Contribution of phytoplankton, periphyton, and macrophytes to primary production in Eagle Lake, California. California Fish and Game J. 62:187-194. Kennedy, J. L. 1983. Seasonal growth of the tui chub, Gila bicolor, in Pyramid Lake, Nevada. Great Basin Nat. 43:713-716. 788 Great Basin Naturalist Vol. 45, No. 4 KiMSEY, J. B. 1954. Life history of the tui chub, Siphatcles bicolor, from Eagle Lake, Cahfornia. CaHfornia Fish and Game J. 40:39.5-410. KiMSEY, J B , .AND R R Bell 1955. Observations on the ecology of largemouth black bass and the tui chub in Big Sage Reservoir, Modoc County. California Dept. Fish and Game, Inland Fish. Branch, Ad- min. Rept. 55-15, Sacramento, California. KuceRA, p. a. 1978. Reproductive biology of the tui chub, Gila bicolor, in Pyramid Lake, Nevada. Great Basin Nat. 38:203-207. KuCERA, P. A.. G. W. WORK.MAN, D. ROBERTSON, S. ViGG. R. Whaley, and R. Langdon 1978. Life history of the tui chub. In W. F. Sigler and J. L. Kennedy, eds. Pyramid Lake Ecological Study. VV. F. Sigler and Assoc. Inc., Logan, Utah. Langdon, R. W. 1979. Food habits of the tui chub, Gila bicolor, in Pyramid Lake, Nevada. Unpublished thesis. Humboldt State University, Areata, Cali- fornia. 45 pp. La Rivers. I. 1962. Fishes and Fisheries of Nevada. Ne- vada Fish and Game Commission. 782 pp.X Nelson. L. 1976. SHAD II, A model for analysis of fish- eries age and growth data. Wildl. Sci. Dept., Utah State University, Logan, Utah. 15 pp. RiCKER, W E. 1971. Methods for assessment offish pro- duction in fresh waters. Inter. Biol. Programme, Black-well Scientific Publ., 348 pp. Snyder, J. O. 1917. The fishes of the Lahontan system of Nevada and northeastern California. Bull. U.S. Bur. Fish. 35:31-86. ViGG. S. 1978. Vertical distribution of adult fish in Pyra- mid Lake, Nevada. Great Basin Nat. 38:417-428. 1980. Seasonal benthic distribution of adult fish in Pyramid Lake, Nevada. California Fish and Game J. 66:49-58. 1981. Species composition and relative abundance of adult fish in Pvramid Lake, Nevada. Great Basin Nat. 41:.395-406. NEW VARIET\' OF YUCCA HARRIMANIAE (AGAVACEAE) FROM UTAH Elizabeth Nc-t\so' and Stanley L. Welsh' Abstkact — Yticca hanimaniac var. stchlis Neese & Welsh is named and deserihed from the Uinta liasin of Utah. The plant is strongly rhizomatous, evidently sterile, and has limht'r, sjiarinuly hliterous leaves that tend to reeline on the ground. During field investigations in the Uinta Basin of Utah in the late 1970s and early 1980s, a phase ofYucca harrimaniaeTve]. was brought to our attention by Mr. Dan Gardner of the Bureau of Land Management in Vernal, Utah, who had observed the plants in the Pariette Bench vicinity of the basin. Several occurrences of this peculiar entity were discovered thereafter by the authors and by other collectors in the region. The plants are characterized by being strongly rhizoma- tous, with the rosettes more or less widely spaced; the leaves are limber and sparingly, if at all, filamentous marginally, and tend to recline in curved fashion on the ground. These characters contrast strongly with typi- cal material of Yucca Jiarmnaniae as it occurs in the Uinta Basin and elsewhere. Further- more, the plants with sprawling, sparingly filamentous, flaccid leaves are not known to produce fruit, even though some populations have been observed over a period of several years, nor has fruit from previous years, often observable in the typical material, been found by us. In typical plants the rosette leaves are stiffly erect-ascending and marginally fila- mentous, the rosettes are clumped to nar- rowly spaced, and fruit is produced routinely. The species is treated as follows. Yucca harrimaniae Trel. Harriman Yucca. [Y. harrimaniae var. gilbertiana Trel, type from Juab County; Y. gilbertiana (Trel.) Rydb]. Plants acaulescent, forming densely to widely spaced rosettes; leaves falcate or straight, lanceolate to spatulate-lanceolate, concavo-convex, deeply striate, rather thick and rigid or limber, pale green, pungent api- cally, 1-5 dm long, 0.7-4 cm wide, the mar- gins white or brown, in age more or less filifer- ous, the fibers, when present, somewhat coarse and curly; inflorescence 3.5-7 dm tall, racemose or rarely with a few short branch- lets, extending from within the foliage to well above; flowers broadly campamilate, pen- dant, yellowish or greenish yellow to cream, tinged with purple, the segments 4-5 (6) cm long, 1.6-3.5 cm broad; ovary 1.5-2 cm long, pale green; style 9-11 mm long, bright green; capsule cylindric, with a short attenuate beak, 3.7-5 (6) cm long, usually deeply constricted toward the center and flaring open when dried, or not developed. Two more or less distinctive varieties are present in this species. 1. Leaves of basal rosette stiffly erect-ascending, conspicuously filamentous along the margin; rosettes often clumped; plants not strongly rhi- zomatous, routinely forming capsules Y. harrimaniae var. harrimaniae — Leaves of basal rosette flaccid, often reclining on the ground, typically curved, not especially fila- mentous marginally; rosettes often widely spaced; plants strongly rhizomatous, not known to form fruit Y. harrimaniae var. sterilis Var. harrimaniae. Warm desert shrub, grasslands, sagebrush, pinyon-juniper, and mountain brush communities at 1200 to 2700 m in Beaver, Carbon (type from near Helper), Duchesne, Emery, Garfield, Grand, Iron, Juab, Millard, Piute, Sevier, San Juan, Uin- tah, and Wayne counties; Nevada, Colorado, Arizona, and New Mexico; 42 (xii). Var. sterilis Neese & Welsh, var. nov. Similis Y. harrimaniae var. harrimaniae in floribus et staturis generalis sed in foliis flacci- dis et floribus sterilis differt. Type: USA Utah. Uintah Co., T6S, R22E, Sec 14, ca 8 km S of Jensen, at mouth of Walker Hollow, at 1470 m elev.. Salt desert shrub community, on bluff 'Life Science Mu id Department of Range Science, Brigham Young University, Prove, Utah 84602. 789 790 Great Basin Natur.\list Vol. 45, No. 4 margin, alluvium over Uinta Formation, 31 May 1979, S. L. Welsh 18461 (Holotype BRY; Isotypes 4, distributed previously as Yucca). Additional specimens: Utah. Uintah Co., 19 km due SSW of Naples, 14 May 1980, S. White & E. Neese 133 (BRY); mouth of Walker Hollow, 31 May 1979, Neese et al. 7479 (BRY); 41 km S of Roosevelt, 29 June 1978, E. Neese &L. England 5899 (BRY); ca 3 km NW of Gusher, 8 June 1979, E. Neese & B. Welsh 7542, 8 June 1979 (BRY). Duchesne Co., ca 11 km NNW of Roosevelt, 19 June 1979, E. Neese 7663 (BRY). Literature Cited HiGGiNS, L. C , AND S. L. Welsh. 1985. Utah flora: Agavaceae. Brigham Young University. Unpub- lished manuscript. 8 pp. REVISION OF THE PHLOX AUSTROMONTANA (POLEMONIACEAE) COMPLEX IN UTAH Stanley L. Welsh' Abstract— The taxa centering around Phlox austromontana Coville are revised. Named as a new variety is P. austromontana var. lutescens Welsh from eastern Garfield County, Utah. A new combination is provided as P. austromontana var.jonesii (Wherry) Welsh. Taxa within the genus Phlox have been re- viewed preparatory to completion of the Utah flora, a summary revision of all indigenous, adventive, and commonly cultivated plant species for Utah. Observations made during that review demonstrated the need for modifi- cation of contemporary concepts within the complex of forms centering around the con- cept of Phlox austromontana sensu lato. The species has been interpreted by contempo- rary authors as consisting of a single polymor- phic taxon, or of a closely allied species pair, i.e., P. austromontana and P.jonesii. Transitional morphological features tend to obscure the populations, which are often more or less geographically or topographically correlated. This is a general problem in the genus, and P. austromontana merely exem- plifies that problem. Thus, it is not surprising that the various taxa represented in herbaria previously should have been subjected to dif- ferential treatment. Some named portions of the variation have been reduced to syn- onymy, when the transitional morphology was interpreted as taxonomically negligible. Phlox austromontana var. prostrata consists of sprawling plants with long internodes and occurs in much of the middle elevation por- tions of Washington County, Utah. The ca- lyces of this variety are ordinarily conspicu- ously hairy. Thejonesii phase is similar to the prostrata phase in having sprawling stems with long internodes, but the calyces are glabrous in extreme examples. Where the two phases meet in Zion Canyon, the type locality for the jonesii phase, there is a mixing of char- acters. Brightly colored pink flowers charac- teristic of the jonesii phase have either glabrous calyces or hairy ones like the pros- trata materials. The overlapping characteris- tics seem to indicate variation that should be recognized within a species in this genus. Therefore, the jonesii, prostrata, and aus- tromontana phases are treated at varietal level. Examination of the large series of speci- mens in this species at the herbarium of Brigham Young University demonstrated the existence of a robust, subligneous specimen taken from crevices in the Cedar Mesa Sand- stone along the margin of Cataract Canyon in eastern Garfield County, Utah. The specimen had creamy yellowish flowers when collected, and, because of its growth habit and flower color, was initially determined as a Lepto- dactylon. The leaves are simple, however, not digitate as in that genus. The flowers have dried a lemon yellow color and are thickly cartillaginous, unlike any other portions of the austromontana complex, but other features, including lax stems, long leaves, and carinate intercostal membranes of the calyx, indicate aff^inity with P. austromontana . The species is, therefore, revised as follows. Phlox austromontana Coville Desert Phlox. Plants caespitose, cushion or matlike, from a pluricipital caudex and taproot, mainly 0.5-3 dm wide; herbage pilose-puberulent to subglabrous or the calyx glabrous to villous externally; leaves opposite, mainly 5-20 mm long, simple, linear-subulate; flowers solitary, sessile or subsessile at branch tips; calyx urce- olate to campanulate, glabrous to villous, the intercostal membranes carinate, the lobes vil- 'Life Science Museum and Department of Botany and Range Science, Brigham Young University. Prove, Utah 84602. 791 792 Great Basin Naturalist Vol. 45, No. 4 lous internally; corolla white, blue, pink, lavender, or yellowish, the tube 8-15 mm long; styles 2-9 mm long. This is a complex assemblage of variants, some sufficiently dis- tinctive and sufficiently geographically corre- lated as to warrant taxonomic recognition. The morphology is, however, wholly conflu- ent. Trends within the diversity are recog- nized at varietal level. 1. Plants more or less open, the internodes typically apparent; plants of western Kane and much of Washington counties 2 — Plants variously open to compact; distribution various 3 2(1). Calyx usually glabrous, the leaves (or some of them) 20-35 mm long; corollas usually bright pink; morphology transitional to the next P. ausfromontana var.jonesii — Calyx at least moderately villous, the leaves typ- ically 10-22 mm long; corolla commonly white P. atistromontana var. prostrata 3(2). Flowers yellowish (fading lemon yellow); leaves 10-25 mm long; calyx campanulate P. atistromontana var. lutescens — Flowers white, pink, or lavender (sometimes fading to cream in color); leaves mostly less than 15 mm long; calyx turbinate to subcylindric. . . . P. austromontana var. atistromontana Var. austromontana [P. densa Brand, type from Frisco]. Mixed desert shrub, salt desert shrub, pinyon -juniper, sagebrush, mountain brush, and ponderosa pine communities at 1525 to 3050 m in Beaver, Carbon, Duchesne, Emery, Garfield, Iron, Juab, Kane, Millard, Piute, Sanpete, Sevier, Tooele, Uintah, Washington (type from Beaverdam Moun- tains), and Wayne counties; Nevada, Califor- nia, and Arizona; 159 (xxiii). Var. jonesii (Wherry) Welsh, comb. nov. [based on: PJiIox jonesii Wherry Notul. Nat. Acad. Nat. Sci. Philadelphia 146:' 8. 1944; holo- type — Washington County, Zion Canyon, 7 May 1923, M. E. Jones sn US!]. Ponderosa pine, pinyon-juniper, and mountain brush communi- ties at 1435 to 2600 m in Kane and Washington counties; endemic; 12 (ii). This variety forms intermediates with both var. prostrata and var. austromontana. It is partially sympatric with both. Var. lutescens Welsh, var. nov. Affinis et similis Phlox austromontana Coville var. aus- tromontana sed in corollis lutescentibus et plus crassis et calvcibus campanulatis differt. Type: USA Utah. Garfield Co., T33S, R14#, SW 1/4 SI, along Orange Cliffs Rd, E of Hvvy 95, 1373 m elev., rimrock-blackbrush, ash, squawbush community, 11 Mav 1983, S. L. Welsh, B. Welsh, M. Chatterley 21972 (Holotype BRY; isotypes 4, distributed previously as Leptodacty- lon watsonii [Gray] Rydb.). The specimens col- lected were taken from a large population of suffrutescent, rounded, cushions growing in crevices in rimrock of Cedar Mesa Sandstone, and, because of the peculiar flower color and growth habit, were mistaken for the superficially similar Leptodactylon watsonii. Var. prostrata E. Nels. Mountain brush and pinyon-juniper communities at 1220 to 2135 m in Washington (tvpe from Silver Reef!) Countv; endemic (?); 14 (0). Literature Cited Welsh, S. L 1985. Utah flora: Polemoniaceae. Brigham Young University. Unpublished manuscript. 48 pp. INDEX TO VOLUME 45 The genera, species, and other taxa described as new to science in this voknne appear in bold type in this index. A fourth species oiOreoxis (Umbelhferae), p. 34. Acanthotomicus ipsiformus, p. 270. Acrantus opimus, p. 270. Addendum to the distribution of two herptiles in Idaho, p. 291. Age, growth, and food habits of Tui chub, Gila hicolor, in Walker Lake, Nevada, p. 784. A new combination and a new variety in Artemisia tridentata, p. 99. Andersen, W. R., Tim D. Davis, N. Sankhk, D. J. Weber, and B. N. Smith, article by, p. 520. Annotated Key to Eriogonum (Polygonaceae) of Nevada, p. 493. Aquatic birds of the White River, Uintah County, Utah, p. 113. Aquatic parameters and life history observa- tions of the Great Basin spadefoot toad in Utah, p. 22. Artemisia trideyitata ssp. vaseijana var. pauci- flora, p. 102. Ashley, John, Samuel R. Rushforth, and Jef- frey R. Johansen, article by, p. 432. Aspects of the biology of the flathead chub (Hybopsis gracilis) in Montana, p. 332. Asfragfl/w.s debequaeus, p. 31. Astragalus piscator, p. 551. Atwood, N. Duane, and Stanley L. Welsh, article by, p. 485. Baker, William L., and Susan C. Kennedy, article by, p. 747. Barnes, James R., and J. Vaun McArthur, article by, p. 117. Barneby, Rupert C, and Stanley L. Welsh, article by, p. 551. Behle, William H., article by, p. 443. Benton, Bob, Peter Hovingh, and Dave Born- holdt, article by, p. 22. Best, Trov L., and A. L. Gennaro, article bv, p. 527.' Bornholdt, Dave, Peter Hovingh, and Bob Benton, article by, p. 22. Bothrosternus hirsutus, p. 271. Brachyprotoma brevimala, p. 366. Bres, Mimi, William F. Sigler, and Steven Vigg, article by, p. 571. Bright, D. E., articles by, pp. 467, 476. Brotherson, Jack D. , and James Callison, arti- cle by, p. 321. Brotherson, Jack D., William E. Evenson, Samuel R. Rushforth, John Fairchild, and Jeffrey R. Johansen, article by, p. 1. Brotherson, Jack D. , and William J. Masslich, article by, p. 535. Brotherson, Jack D., and Samuel R. Rush- forth, article by, p. 542. Brotherson, Jack D., and J. B. Shupe, article by, p. 141. Burrowing Owl foods in Conata Basin, South Dakota, p. 287. Butler, David R., article by, p. 313. Callison, James, and Jack D. Brotherson, arti- cle by, p. 321. Carroll, T. , and R. A. Heckmann, article by, p. 255. Carter, John G., Vincent A. Lamarra, and Marianne C. Lamarra, article by, p. 127. Chatterley, L. Matthew, and Stanley L. Welsh, article by, p. 173. Checklist of the mosses of Grand Teton Na- tional Park and Teton County, Wyoming, p. 124. Checklist of vascular plants for the Bighorn Canyon National Recreation Area, Wyo- ming and Montana, p. 734. Clark, William H., article by, p. 391. C nesinus discretus, p. 271. Cnesinus minor, p. 272. Collins, Ellen 1., Robert W. Lichvar, and Dennis H. Knight, article by, p. 734. Cooper, James J., article by, p. 784. Comparisons of prescribed burning and cut- ting of Utah marsh plants, p. 462. Corthylus truncatus, p. 272. 793 794 Great Basin Naturalist Vol. 45, No. 4 Cryptogamic soil crusts: seasonal variation in algal populations in the Tintic Mountains, Juab County, Utah, p. 14. Davis, James N., and Bruce L. Welch, article by, p. 778. Davis, Tim D., N. Sankhla, W. R. Andersen, D. J. Weber, and B. N. Smith, article by, p. 520. Deutschman, Mark Robert, article by, p. 546. Differential effects of cattle and sheep grazing on high mountain meadows in the Straw- berry Valley of central Utah, p. 141. Dobkin, David S., Jennifer A. Holmes, and Bruce A. Wilcox, article by, p. 483. Dwarf mistletoe-pandora moth interaction and its contribution to ponderosa pine mor- tality in Arizona, p. 423. Dystylosaurus, p. 707. Dystylosaurus edwini, p. 707. Ecological investigation of a suspected spawn- ing site of Colorado squawfish on the Yampa River, Utah, p. 127. Electrophoretic study of cutthroat trout popu- lations in Utah, p. 677. Effectiveness of the seed wing oi Finns flexilis in wind dispersal, p. 318. Elliott, Nancy B., and Frank E. Kurczewski, article by, p. 293. Evenson, William E., Jack D. Brotherson, Samuel R. Rushforth, John Fairchild, and Jeffrey R. Johansen, article by, p. 1. Everett, Richard L., and Steven H. Sharrow, article by, p. 105. Eriogonum lewisii, p. 277. Eriogonum ochrocephalum var. alexanderae, p. 276. Eriogonum tiehmii, p. 277. Eriogonum umbellatum var. furcosum, p. 278. Eriogonum umbellatum var. juniporinum, p. 279. Factors influencing nesting success of bur- rowing owls in southeastern Idaho, p. 81. Fairchild, John, Jack D. Brotherson, William E. Evenson, Samuel R. Rushforth, and Jef- frey R. Johansen, article by, p. 1. First record of Climacia calif o mica (Neu- roptera: Sisyridae) and its host sponge, Ephydatia mulleri (Porifera: Spongillidae), from Idaho with water quality relation- ships, p. 391. Food habits and dietary overlap of nongame insectivorous fishes in Flint Creek, Okla- homa, a western Ozark foothills stream, p. 721. Food habits of the western whiptail lizard {C nemiclophorus tigris) in southeastern New Mexico, p. 527. Food of cougars in the Cascade Range of Ore- gon, p. 77. Gennaro, A. L., and Trov L. Best, article by, p. 527. Gleason, Richard S., and Donald R. Johnson, article by, p. 81. Golightly, Richard T. , Jr., Jeffrey S. Green, Susan Lyndaker Lindsey, and Brad R. Lea- Master, article by, p. 567. Goodrich, Sherel, article by, p. 155. Goodrich, Sherel, E. Durant McArthur, and Alma H. Winward, article by, p. 99. Goodrich, Sherel, and Stanley L. Welsh, arti- cle by, p. 34. Gould, William, article by, p. 332. Grass spider microhabitat use in Organ Pipe Cactus National Monument, Arizona, p. 546. Green, Jeffrey S., Richard T. Golightly, Jr., Susan Lyndaker Lindsey, and Brad R. Lea- Master, article by, p. 567. Growth and reproduction of the flannelmouth sucker, Catostomus latipinnis, in the Up- per Colorado River Basin, 1975-76, p. 281. Guver, Craig, and Allan D. Linder, article bv, p. 607. Habitat relationships of the blackbrush com- munity (Coleogyne ramosissima) of south- western Utah, p. 321. Hansen, Richard M., James G. MacCracken, and Daniel W. Uresk, article by, p. 287. Hartman, Emily L., and Mary Lou Rottman, article by, p. 87. Heaton, Timothy H., article by, p. 337. Heckmann, R. A., and T. Carroll, article bv, p. 255. Heckmann, Richard, aiid Terr\ Otto, article by, p. 427. Helminth parasites of the white-tailed jack- rabbit, Lcpus townscndi , from northwest- ern Colorado and southern Wvoming, p. 604. High rates of photosynthesis in the desert shrub Chnjsotliamnus nauscosus ssp. albi- caulis, p. 520. October 1985 Index 795 Holmes, Jennifer A., David S. Dobkin, and Bruce A. Wilcox, article by, p. 483. Host-parasite studies of Tiichophn/a infest- ing cutthroat trout ( Salmo clarki ) and long- nose suckers (Catostomus catostomus) irom Yellowstone Lake, Wyoming, p. 255. Hovingh, Peter, Bob Benton, and Dave Born- holdt, article by, p. 22. Humphrey, L. David, article by, p. 94. Hylurgus indicus, p. 273. In memoriam: William Wallace Newby (1902-1977), p. 443. Insect communities and faunas of a Rocky Mountain subalpine sere, p. 37. Invasion and stabilization of recent beaches by salt grass (Distichilis spicata) at Mono Lake, Mono County, California, p. 542. Jenni, Donald A., and Roland L. Redmond, article by, p. 85. Jensen, James, A., articles by, pp. 697, 710. Jensen, J. Neil, Mark A. Martin, Dennis K. Shiozawa, and Eric J. Loudenslager, article by, p. 677. Johansen, Jeffrey R., John Ashley, and Sam- uel R. Rushforth, article by, p. 432. Johansen, Jeffrey R., Jack D. Brotherson, William E. Evenson, Samuel R. Rushforth, and John Fairchild, article by, p. 1. Johansen, Jeffrey R., and Samuel R. Rush- forth, article by, p. 14. Johnson, Donald R., and Richard S. Gleason, article by, p. 81. Kadlec, John A., and Loren M. Smith, article by, p. 462. Kass, Ronald J., and Stanley L. Welsh, article by, p. 548. Kennedy, Susan C, and William L. Baker, article by, p. 747. Knight, Dennis H., Robert W. Lichvar, and Ellen I. Collins, article by, p. 734. Koniak, Susan, article by, p. 556. Kurczewski, Frank E., and Nancy B. Elliott, article by, p. 293. Lamarra, Marianne C, Vincent A. Lamarra, and John G. Carter, article by, p. 127. Lamarra, Vincent A., Marianne C. Lamarra, and John G. Carter, article by, p. 127. Lanner, Ronald M., article by, p. 318. Laurance, Wilham F., and Timothy D. Reynolds, article by, p. 291. LeaMaster, Brad R., Jeffrey S. Green, Richard T. Golightly, Jr., and Susan Lyn- daker Lindsey, article by, p. 567. Lept()tyf)hl()))s (hilcis supraocularis, p. 625. Leptotyplilops /iJ///H7/.schihuahuaensis, p. 623. Lichvar, Robert W., Ellen I. Collins, and Den- nis H. Knight, article by, p. 734. Life history of the Cui-ui, Chasmistes cujiis Cope, in Pyramid Lake, Nevada: a review, p. 571. Lindcr, Allan D., and Craig Guyer, article bv, p. 607. Lindsey, Susan Lyndaker, Jeffrey S. Green, Richard T. Golightly, Jr., and Brad R. Lea- Master, article by, p. 567. Loudenslager, Eric J., Mark A. Martin, Dennis K. Shiozawa, and J. Neil Jensen, article by, p. 677. MacCracken, James G., Daniel W. Uresk, and Richard M. Hansen, article by, p. 287. MacMahon, James A., and David J. Schimpf, article by, p. 37. Martin, Mark A., Dennis K. Shiozawa, Eric J. Loudenslager, and J. Neil Jensen, article by, p. 677. Maser, Chris, and Dale E. Toweill, article by, p. 77. Maser, Chris, and John O. Whitaker, Jr., article by, p. 67. Masslich, William J., and Jack D. Brotherson, article by, p. 535. Mathiasen, Robert L., and Michael R. Wagner, article by, p. 423. McAda, Charles W., and Richard S. Wydoski, article by, p. 281. McArthur, E. Durant, Sherel Goodrich, and Alma H. Winward, article by, p. 99. McArthur, J. Vaun, and James R. Barnes, article by, p. 117. McDonald, Jerry N., article by, p. 455. Mites (excluding chiggers) of mammals of Ore- gon, p. 67. Neese, Elizabeth, and Stanley L. Welsh, article by, p. 789. Nesting and predatory behavior of some Tachysphex from the western United States (Hymenoptera: Sphecidae), p. 293. New Nevada entities and combinations in Eri- ogonum (Polygonaceae), p. 276. New records and comprehensive list of the algal taxaof Utah Lake, Utah, USA, p. 237. New Sclerocactus (Cactaceae) from Nevada, p. 553. New species and records of North American Pityophthorus (Coleoptera: Scolytidae), Part IV: The Scriptor group, p. 467. 796 Great Basin Naturalist Vol. 45, No. 4 New species and new records of North Ameri- can Pityophthorus (Coleoptera: Scolyti- dae), Part V: the Juglandis group, p. 476. New species of Astragalus (Leguminosae) from Mesa County, Colorado, p. 31. New species of Astragalus (Leguminosae) from southeastern Utah, p. 551. New species of Primula (Primulaceae) from Utah, p. 548. New species of Talinum (Portulaceae) from Utah, p. 485. New synonymy and new species of bark beetles (Coleoptera: Scolytidae), p. 266. New variety of Yucca harrimaniae (Agavace- ae) from Utah, p. 789. Note on the diet of long-billed Curlew chicks in western Idaho, p. 85. Nutrients in Carex exerta sod and gravel in Sequoia National Park, California, p. 61. Occurrence of anisakid larvae (Nematoda: As- cardidia) in fishes from Alaska and Idaho, p. 427. Oreoxis trotteri, p. 34. Otto, Terry, and Richard Heckmann, article by, p. 427. Pachycotes minor, p. 273. Patterns of macroinvertebrate colonization in an intermittent Rocky Mountain stream in Utah, p. 117. Phloeosinopsoides pumilus, p. 274. Pityophthorus ablusus, p. 476. Pityophthorus atkinsoni, p. 467. Pityophthorus costifera, p. 477. Pityophthorus cracentis, p. 477. Pityophthorus desultorius, p. 478. Pityophthorus diminuiixus, p. 468. Pityophthorus equihuai, p. 469. Pityophthorus insuetus, p. 479. Pityophthorus thamnus, p. 470. Pityophthorus trunculus, p. 470. Pityophthorus zexmenixora, p. 471. Poa L. in New Mexico, with a key to middle and Southern Rockv Mountain species (Poaceae), p. 395. Pollinators oi Astragalus monoensis Barnebx (Fabaceae): new host records; potential im- pact of sheep grazing, p. 299. Presettlement vegetation of part of north- western Moffat County, Colorado, de- scribed from remnants, p. 747. Primula domensis, p. 548. Quaternary paleontology and paleoecology of Crystal Ball Cave, Millard County, Utah: with emphasis on mammals and description of a new species of fossil skunk, p. 337. RatliflF, Raymond D., article by, p. 61. Redmond, Roland L., and Donald A. Jenni, article by, p. 85. Reese, Kerry P., article by, p. 152. Reveal, James L., articles by, pp. 276, 488, 493. Reynolds, Timothy D., and William F. Lau- rance, article by, p. 291. Rickard, LoraC, and Larry M. Shults, article by, p. 604. Rottman, Mary Lou, and Emily L. Hartman, article by, p. 87. Rushforth, Samuel R., John Ashley, and Jef- frey R. Johansen, article by, p. 432. Rushforth, Samuel R., and Jack D. Brother- son, article by, p. 542. Rushforth, Samuel R., Jack D. Brotherson, William E. Evenson, John Fairchild, and Jeffrey R. Johansen, article by, p. 1. Rushforth, Samuel R., and Jeffrey R. Jo- hansen, article by, p. 14. Rushforth, Samuel R., and Lorin E. Squires, article by, p. 237. Sankhla, N., Tim D. Davis, W. R. Andersen, D. J. Weber, and B. N. Smith, article by, p. 520. Schimpf, David J., and James A. MacMahon, article by, p. 37. Sclerocactus blainei, p. 553. Second nesting record and northward ad- vance of the Great-tailed Crackle {Quis- calus mexicanus) in Nevada, p. 483. Sexual selection and mating system variation in anuran amphibians of the Arizona-Sono- ran Desert, p. 688. Sharrow, Steven H., and Richard L. Everett, article by, p. 105. Shiozawa, Dennis K., Mark A. Martin, Eric J. Loudenslager, and J. Neil Jensen, article by, p. 677. Shults, Larry M., and Lora G. Rickard, article by, p. 604. Shupe, J. B., and Jack D. Brotherson, article by, p. 141. Sigler, William F., Steven Vigg, and Mimi Bres, article by, p. 571. Size selection of food by cutthroat trout, Salmo clarki, in an Idaho stream, p. 327. Skinner, William D., article by, p. 327. October 1985 Index 797 Smith, B. N., Tim D. Davis, N. Saiikhla, W. R. Andersen, and D. J. Weber, article by, p. 520. Smith, Loren M., and John A. Kadlec, article by, p. 462. Snakes of western Chihnahua, p. 615. Soil algae of cryptogamic crusts from the Uin- tah Basin, Utah, U.S.A., p. 432. Soreng, Robert J., article by, p. 395. Spatial patterns of plant communities and dif- ferential weathering in Navajo National Monument, Arizona, p. 1. Spence, John R., article by, p. 124. Squries, Lorin E., and Samuel R. Rushforth, article by, p. 237. Steele, Benjamin B., and Stephen B. Vander Wall, article by, p. 113. Stewart, Kenneth W., and C. Stan Todd, arti- cle by, p. 721. Succession in pinyon-juniper woodlands fol- lowing wildfire in the Great Basin, p. 556. Sugden, Evan A., article by, p. 299. Sullivan, Brian K., article by, p. 688. Supersaurus, p. 701. S u pe rsati rus \i\ianae, p. 701. Symbos cavifrons (Artiodactyla: Bovidae) from Delta County, Colorado, p. 455. Talinwn thompsonii, p. 485. Tanner, Wilmer W., article by, p. 615. Tanner- White, Merle, and Clayton M. White, article by, p. 150. Thamtiophis nifipunctatus uni\ahia\is, p. 648. Thermal ecology and activity patterns of the short-horned lizard (Phnjnosoma dou- glassi) and the sagebrush lizard {Scelop- ortis graciosus) in southeastern Idaho, p. 607. Thorne, Kaye Hugie, and Stanley L. Welsh, article by, p. 553. Three additional cases of predation by mag- pies on small mammals, p. 152. Three new Sauropod dinosaurs from the Up- per Jurassic of Colorado, p. 697. Todd, C. Stan, and Kenneth W. Stewart, arti- cle by, p. 721. Torvosauridae, p. 711. Toweill, Dale E. , and Chris Maser, article by, p. 77. Tundra vegetation of three cirque basins in the northern San Juan Mountains, Colo- rado, p. 87. Types oi" Nevada buckwheats {Eriogonum: Polygonaceae), p. 488. Ultrasaurus, p. 704. Vltrasaurus macintoshi, p. 704. Uncompahgre dinosaur fauna: a preliminary report, p. 710. Understory response to tree harvesting of sin- gleleaf pinyon and Utah juniper, p. 105. Unusual social feeding and soaring by the Common Raven (Corviis corax), p. 150. Uresk, Daniel W., James G. MacCracken, and Richard M. Hansen, article by, p. 287. Use of biomass predicted by regression from cover estimates to compare vegetational similarity of sagebrush-grass sites, p. 94. Use of radio transmitter implants in wild canids, p. 567. Utah flora: Saxifragaceae, p. 155. Utah's rare plants revisited, p. 173. Vander Wall, Stephen, B., and Benjamin B. Steele, article by, p. 113. Vegetation patterns in relation to slope posi- tion in the Castle Cliffs area of southern Utah, p. 535. Vegetational and geomorphic change on snow avalanche paths, Glacier National Park, Montana, USA, p. 313. Vigg, Steven, William F. Sigler, and Mimi Bres, article by, p. 571. Wagner, Michael, R., and Robert L. Mathi- asen, article by, p. 423. Weber, D. J., Tim D. Davis, N. Sankhla, W. R. Andersen, and B. N. Smith, article by, p. 520. Welch, Bruce L., and James N. Davis, article by, p. 778. Welsh, Stanley L., article by, p. 31. Welsh, Stanley L., and N. Duane Atwood, article by, p. 485. Welsh, Stanley L., and Rupert C. Barneby, article by, p. 551. Welsh, Stanley L., and L. Matthew Chatter- ley, article by, p. 173. Welsh, Stanley L., and Sherel Goodrich, arti- cle by, p. 34. Welsh, Stanley L., and Ronald J. Kass, article by, p. 548. Welsh, Stanley L., and Elizabeth Neese, arti- cle by, p. 789. Welsh, Stanley L. , and Kaye Hugie Thorne, article by, p. 553. Whitaker, John O. , Jr. , and Chris Maser, arti- cle by, p. 67. 798 Great Basin Naturalist Vol. 45, No. 4 White Clayton M., and Merle Tanner- Winward, Alma H., Sherel Goodrich, and E. White, article by, p. 150. ^ Durant McArthur, article by, p^ 99. Wilcox, Bruce A., Jennifer A. Holmes, and Wood, Stephen L. artic-^e^by, p. 266. David S. Dobkin, article by, p. 483. Wydoski, Richard S. , and Charles W. McAda, Winter preference, nutritive value, and other article by, p. 281 range use characteristics of Kochia pros- Xylechinoso7nus pi\osus, p. 274. ^ ,^ N r, 1 1 rr-70 Yiirrn hnrritn/ininf' var. SteriilS. D. /oy. trata (L. Schrad, p. 778. Yucca harrimaniae var. sterilis, p. 7 0 7ifu86 NOTICE TO CONTRIBUTORS Manuscripts intended for publication in the Great Basin Natttralist or Great Basin Natural- ist Memoirs must meet the criteria outhned in paragraph one on the inside front cover. They should be directed to Brigham Young University, Stephen L. Wood, Editor, Great Basin Naturalist, 290 Life Science Museum, Provo, Utah 84602. Three copies of the manuscript are required. They should be typewritten, double spaced throughout on one side of the paper, with margins of at least one inch on all sides. Use a recent issue of either journal as a format, and the Council of Biology Editors Style Manual, Fourth Edition (AIBS 1978) in preparing the manuscript. An abstract, about 3 percent as long as the text, but not exceeding 200 words, written in accordance with Biological Abstracts guidelines, should precede the introductory paragraph of each article. All manuscripts receive a critical peer review by specialists in the subject area of the manuscript under consideration. Authors may recommend one or two reviewers for their article. Accepted manuscripts less than 100 printed pages in length will automatically be assigned to the Great Basin Naturalist. Those larger than 100 printed pages will be considered for the Memoirs series. Illustrations and Tables. All illustrations and tables should be made with a view toward having them appear within the limits of the printed page. The original illustrations or glossy photoprints of them should accompany the manuscript. Illustrations should be prepared for reduction by the printer to either single-column (2| inches) or double-column (5| inches) width, with the length not exceeding 7| inches. Costs Borne by Contributor. Contributors to the Great Basin Naturalist should be prepared to donate from $10 to $40 per printed page toward publication of their article (in addition to reprint costs). Excessive or complex tables requiring typesetting may be charged to the author at cost. Authors publishing in the Great Basin Naturalist Memoirs may be expected to contribute $40 per printed page in addition to the cost of the printed copies they purchase. No reprints are furnished free of charge. Reprint Schedule for the Great Basin Naturalist 100 copies, minimum cost for 2 pages, $26. Each additional 2 pages, $6. Each additional 100 copies, $4 for each 2 pages. Examples: 300 copies of 10 pages = $82; 200 copies of 13 pages = $86. Great Basin Naturalist Memoirs No. 1 The birds of Utah. ByC. L. Hayward, C. Cottam, A. M. Woodbury, H. H. Frost. $10. No. 2 Intermountain biogeography: a symposium. By K. T. Harper, J. L. Reveal et al. $15. No. 3 The endangered species: a symposium. $6. No. 4 Soil-plant-animal relationships bearing on revegetation and land reclamation in Nevada deserts. $6. No. 5 Utah Lake monograph. $8. No. 6 The bark and ambrosia beetles of North and Central America (Coleptera: Scolytidae), a taxonomic monograph. $60. No. 7 Biology of desert rodents. $8. TABLE OF CONTENTS Life history of the cui-ui, Chasmistes ciijiis Cope, in Pyramid Lake, Nevada: a review. Wilham F. Sigler, Steven Vigg, and Mimi Bres 571 Helminth parasites of the white-tailed jackrabbit, Lepus toivnseiuH, from northwest- ern Colorado and southern Wyoming. Larry M. Shults and Lora G. Rickard 604 Thermal ecology and activity patterns of the short-horned lizard {Phnjnosoma dou- glassi) and the sagebrush lizard {Scelopurus graciosiis) in southeastern Idaho. Craig Guyer and Allan D. Linder 607 Snakes of western Chihuahua. Wilmer W. Tanner 615 Electrophoretic study ofcutthroat trout populations in Utah. Mark A. Martin, Dennis K. Shiozawa, Eric J. Loudenslager, and J. Neil Jensen 677 Sexual selection and mating system variation in anuran amphiliians of the .Arizona- Sonoran Desert. Brian K. Sullivan 688 Three new sauropod dinosaurs from the Upper Jurassic of Ckilorado. James A. Jensen 697 Uncompahgre dinosaur fauna: a preliminary report. James A. Jensen 710 Food habits and dietary overlap of nongame insectivorous fishes in Flint Creek. Oklahoma, a western Ozark foothills stream. C. Stan Todtl and Keimeth W. Stewart 721 Checklist of vascular plants for the Bighorn Canyon National Recreation .\rea. Roliert W. Lichvar, Ellen L Collins, and Dennis H. Knight 734 Presettlement vegetation of part of northwestern Moffat Count)', Colorado, described from remnants. William L. Baker and Susan C. Kennedy 747 Winter preference, nutritive value, and other range use characteristics of Koclna prostrata (L.) Schrad. James N. Davis and Bruce L. Welch 778 Age, growth, and food liabits of tui chub, Gila hicolor, in Walker Lake, Nevada. James J. Cooper 784 New variety of Yucca harnmaniae (Agavaceae) from Utah. Ehzabeth Neese and Stanley L. Welsh 789 Revision of the Phlox austromontana (Polcmoniaceae) complex in Utah. Stanlex L. Welsh '. . . 791 Index 793 1 I 3 2044 072 231