Great Basin Naturalist

Volume 47 Number 4 Article 16

10-31-1987

Genetic variation and population structure in the cliff , dorsalis, in the Great Basin of western Utah

Martin L. Dobson Cody, Wyoming

Clyde L. Pritchett Brigham Young University

Jack W. Sites Jr. Brigham Young University

Follow this and additional works at: https://scholarsarchive.byu.edu/gbn

Recommended Citation Dobson, Martin L.; Pritchett, Clyde L.; and Sites, Jack W. Jr. (1987) "Genetic variation and population structure in the cliff chipmunk, Eutamias dorsalis, in the Great Basin of western Utah," Great Basin Naturalist: Vol. 47 : No. 4 , Article 16. Available at: https://scholarsarchive.byu.edu/gbn/vol47/iss4/16

This Article is brought to you for free and open access by the Western North American Naturalist Publications at BYU ScholarsArchive. It has been accepted for inclusion in Great Basin Naturalist by an authorized editor of BYU ScholarsArchive. For more information, please contact [email protected], [email protected]. GENETIC VARIATION AND POPULATION STRUCTURE IN THE

CLIFF CHIPMUNK, EUTAMIAS DORSALIS , IN THE GREAT BASIN OF WESTERN UTAH

Martin L. Dobson', Clyde L. Pritchett-, and Jack W. Sites, Jr.^

Abstract —Allelic variation at 21 of 39 electrophoretically resolved enzyme loci was used to examine patterns of geographic differentiation and population structure in six allopatric samples of Eutamias dorsalis. Coefficients of genetic similarity for paired combinations of £. dorsalis samples ranged from 0.955 to 0.975, except for one population that was 0.900. Conservative genie divergence among five populations is proposed to be the result of relatively recent isolation events. High positive F,s values and chi-square analyses confirm a significant excess of homozygotes at several loci at the five localities for which sample sizes were statistically adequate. This may be partly attributable to inbreeding, a Wahlund effect, linkage disequilibrium, posttranslational modification, or some combination of these; but at present some of these alternatives cannot be excluded in favor of a single explanation. Some samples were collected across altitudinal gradients of over 800 m, suggesting that a Wahlund effect may be the most likely explanation for low levels of heterozygosity in these populations.

The distribution of montane in pinyon-juniper began to replace sagebrush the Great Basin of western Utah is disjunct, communities from the south (Van Devender with populations isolated by low-elevation, and Spaulding 1979). Continued warming cold desert valleys (Brown 1971a). The ob- during the Postpluvial (7,500-5,000 years served pattern has been explained by Pleis- B.P.) allowed range expansion of xeric mam- tocene retreat (Late glacial to Late mal species in the low-elevation deserts, pleniglacial) of montane elements from plu- while ranges of small montane mammals fol- vial valleys to higher elevation and more lowed vegetation shifts north and to montane northern latitudes (Currey and James 1982, uplands. Wells 1983). At least four major glacial events Recent biogeographic theory (Brown occurred during the Pleistocene. The most 1971a, 1978, Patterson 1980, 1982) suggests recent, the Wisconsin, is suspected of having that distributions of small mammals can be the greatest influence on existing boreal mam- explained as nonequilibrium extinctions with- mal faunas. Maximum glaciation occurred out recolonization. Thus, the Great Basin from the end of Early Pluvial (23,000 years environment and its insular montane B.P.) to Late Pluvial (12,500 years B.P.). Dur- faunas offer interesting evolutionary "experi- ing this time coniferous forests covered the ments" in which to assess the effects of isola- foothills and piedmont, while low-elevation tion and possible recent population bottle- areas not covered by Lake Bonneville were necking on levels of genetic divergence dominated by sagebrush and juniper commu- among conspecific montane mammal popula- nities. Coniferous forests offered favorable tions. The possibility of occasionally severe dispersal habitat (Thompson and Mead 1982, reductions in the sizes of insular mammal pop- Van Devender and King 1971, Wells and ulations would be conducive to rapid fixation Berger 1967) in the intermountain valleys and of alternate alleles and loss of overall genetic low passes, which may have allowed exchange variability due to sampling error (Nei et al. of montane faunal elements across the Great 1975, Kilpatrick 1981), and this would facili- Basin. tate divergence between populations despite The onset of xeric conditions during the their very recent isolation. This study reports Late Pluvial (12,500-7500 years B.P.) ini- on levels of genetic variability within and tiated major vegetation changes. Coniferous among samples of cliff {Eutamias forests retreated upward in elevation and dorsalis) from six isolated mountain ranges in

2437 Central Avenue. Cody. Wyoming 82414 Department of Zoology. Brigham Young University. Prove, Utah 84602.

551 552 Great Basin Naturalist Vol. 47, No. 4

in each case specimens were assumed to be from one breeding population because of site proximity and habitat uniformity. Heart, liver, and blood tissues were immediately re- moved from live-trapped specimens (killed by cervical dislocation) and transported in liquid nitrogen to the laboratory. Tissues were then homogenized in an equal volume of buffer (0.01 M Tris, 0.001 M EDTA, 5 x 10 ' M NADP, pH adjusted to 7.0 with HCl), cen- trifuged for 20 min at 4 C, and stored at —80 C. Hemolysate was maintained at 0-5 C until assayed. Methods of horizontal starch gel electrophoresis and biochemical staining were similar to those described by Selander et

al. (1971) and Harris and Hopkinson (1976), with minor modifications. Gels were pre- pared using a 14% concentration of hy- drolysed starch, which consisted of a T.l mix of starch from Sigma Chemical Co. (lot 31F- 0135) and Otto Hillers's Electrostarch (lot 307). A total of 39 presumptive gene loci was consistently resolved across all populations,

Fig. 1. Collection localities for Eutamias dorsalis from and the buffer/stain combinations used are six mountain ranges in western Utah. Locality abbrevia- summarized in Table 2. Enzyme nomencla- tions are as in Table 1; stippled areas represent mountain ture follows recommendations of the Nomen- ranges above 2,000 m; hatched region represents the Rockv Mountains. clature Committee of the International Union of Biochemistry (1984), with locus abbrevia- tions following those suggested for lower ver- western Utah, in an attempt to evaluate the tebrates by Murphy and Crabtree (1985). We effects of drift and recent insularization. recognize that our nomenclature will depart from that used in most conventional mammal Materials and Methods studies, but virtually all of these loci are either

known (Fisher et al. 1980) or suspected of A total of 90 specimens representing six homologous across all tetrapods (Harris allopatric populations oi Eutamias dorsalis in being and Hopkinson 1976). the Great Basin of western Utah (Fig. 1) was collected from May through September 1983. Multilocus enzyme systems in which ho- All 90 voucher specimens were deposited in mologies are uncertain were simply desig- the Brigham Young University mammal col- nated numerically from most to least anodal lection as standard museum mounts. Collec- (Est-'T", -"2", etc.). Alleles were designated tion location, population abbreviations and numerically, with the most common allele as- sample sizes, and voucher specimen numbers signed a value of 100 for anodal and - 100 for are presented in Table 1. cathodal migrants. Other allozymic bands and Preferred habitat of cliff chipmunks is open their corresponding alleles were designated canopy pinyon-juniper complex on granite as percentages of distances migrated relative substrate (Brown 1971b). Chipmunk densities to that of the 100 allele. Individual genotypes were low in all sites except two: Indian Farm were inferred from enzyme phenotypes and Canyon of the Deep Creek Mountains and statistically analyzed with the BIOSYS-1 pro- Painter Creek of the House Range. Low den- gram (SwofiFord and Selander 1981). Measures sity in the Stansbury Range reflects the small of genetic variability computed for each popu- area of suitable habitat. Two locations were lation include average locus heterozygosity sampled from both the House Range and the (H, direct count), percent loci polymorphic

Deep Creek Mountains (Table 1, Fig. 1), but (P), and mean number of alleles per locus (A). October 1987 Dobson etal: Cliff Chipmunk Variation 553

Table 1. Summary oi Etitamias dorsalis samples used in the study.

Locality 554 Great Basin Naturalist Vol. 47, No. 4

Table 2. Enzymes and electrophoretic conditions used in the analysis of Eutamias dorsalis populations. Locus prefixes M and S refer to mitochondrial and supernatant (= cytosolic) loci, respectively; and tissue abbreviations H, He, L, and P refer to heart, hemolysate, liver, and plasma, respectively. Abbreviations of enzymes in parentheses are older names found in most mammal literature.

Enzyme Enzyme commission Locus Buffer Tissue number conditions^

Aconitate hydratase 4.2.1.3 Alcohol dehydrogenase

Aminopeptidase ("Lap ") Aspartate aminotransferase ("Got-2") Aspartate aminotransferase ("Got-1") Dipeptidase Dipeptidase Dipeptidase Esterases (non-specific) Fumarate hydratase Glucose dehydrogenase Glucose-6-phosphate isomerase ("Pgi") Glucose-6-phosphate dehydrogenase Glutamate dehydrogenase

Glycerol-3-phosphate dehydrogenase ("-Gpd ")

L-Iditol dehydrogenase ("Sdh ") Isocitrate dehydrogenase ("Idh-2") Isocitrate dehydrogenase ("Idh-1") Lactate dehydrogenase ("Ldh-2") Lactate dehydrogenase ("Ldh-l") Malate dehydrogenase ("Mdh-2")

Malate dehydrogenase ("Mdh-1 ") "Malic enzyme" ("Me-2") "Malic enzyme" ("Me-l") Mannose-6-phosphate isomerase Phosphoglucomutase Phosphogluconate dehydrogenase Superoxide dismutase ("Ipo") Superoxide dismutase ("Ipo") Xanthine dehydrogenase

General proteins: Albumin Hemoglobin Post-albumin Transferin October 1987 Dobson etal: Cliff Chipmunk Variation 555

Table 3. Allele frequencies and estimates of genie variability in six samples of Eutamias dorsalis. Locality abbreviations are from Table 1 and Figure L 556 A

October 1987 DoBSON ETAL; Cliff Chipmunk Variation 557

Wah Wah Mlns Tablk 5. Summary of F-statistics for all variable loci r Oquirrh Mlns across all examined samples oi Eiitamias dorsalis except Stansbury Mountains. Deep Creek Mlns

Raft River Mlns Locus F,s Fn

t House Range M-Aeon- 1.000 1.000 0.034 Est-".3" -0.259 -0.002 0.196 Slansbury Mlns Est-""6" -0.211 -0.148 0.052

1 1 1 T 1 1 1 1 Funi-A 0.517 0.544 0.056 10 09 0( 07 06 05 04 03 02 01 00 Gcdh-A 0.033 -0.006 0.026 G-6-pdh-A 1.000 1.000 0.027 S-Icdh-A -0.021 -0.004 0.017 Iddh-A -0,0.56 -0.021 0.034 Wah Wah Mlns Ldh-A 1.000 1.000 0.034 S-Mdh-A -0.061 0.012 0.047 Deep Creek Mlns M-Me-A 0.933 0.937 0.066

Oquirrh Mlns S-Me-A 0.714 0.750 0.127 Mpi-A -0.060 -0.022 0.035 Rail River Mlns Pep-A -0.016 -0.003 0.013

House Range Pgm-A 1.000 1.000 0.068 Pgdh-A 1.000 1.000 0.286 Slansbury Mlns Sod-"l" 1.000 1.000 0.068

I I 1 I 1 1 1 1 1 1 1 Xdh-A 1,000 1.000 0.083 90 91 92 93 94 95 96 97 98 99 100 Alb -0,125 -0.023 0.091 Roger's (1972) S Hb-"1" -0,072 -0.058 0.013 Trf 0,752 0.770 0.071

Fig. 2. UPGMA dendrograms of genetic distance val- Mean 0.,320 0.,384 0.094 ues (Nei 1978), A, and similarity values (Rogers 1972), B, for six samples oi Eiitamias dorsalis. Sample localities are

those shown in Figure 1; cophenetic correlation values are 0.991 and 0.975, respectively. phism observed in the other samples (e.g., DC and HR), which presumably were also

subject to the same severe conditions. It is very small, as evidenced by very low capture unlikely that the observed polymorphism of success per unit effort compared to other sam- alternate alleles could have been accumulated

ples, and it appears to be restricted to one in each population in the short time since the canyon. Thus, the relatively large level of ge- Postpluvial, 7500 years B.P., when desert ad- netic divergence may also reflect the vancement last isolated mountain ranges. influence of a recent population bottleneck Two alternate explanations are proposed. and/or pronounced genetic drift. First, one large, genetically variable popula- The overall mean Fsj value of 0.094 (Table tion may have been widely distributed across 5) suggests an appreciable level of subdivision the Great Basin and subsequently became between the montane populations, although fragmented and restricted to mountain ranges

much higher levels are known in other small by the Pleistocene climatic shifts. This is the mammals (Fsj = 0.412 for Thomomys bottae, vicariance explanation proposed by Patterson for example; see Patton and Yang 1977). Ap- (1980, 1982) for montane mammal popula- preciable substructuring in populations may tions in New Mexico. This hypothesis would result from population bottlenecks and the predict near genetic uniformity and very low ensuing influence of drift (Schwartz and Ar- between-population divergence in the ab- mitage 1980), and the winter of 1982-83 was sence of drift, isolation by distance (Wright one of the most severe on record in Utah 1965), or some behavioral mechanism con- (NOAA 1983). This may have reduced popula- tributing to small, effective breeding sizes and tion sizes, forcing inbreeding and fostering a nonrandom mating. Alternately, since chip- breeding structure in which drift could have a munks are reported from the Pliocene of pronounced influence. However, if we invoke North America (Black 1972), E. dorsalis as a an explanation of differentiation by climati- species may predate the Pleistocene and may cally caused population bottlenecking and have entered the Great Basin from the Rocky subsequent drift for the Stansbury sample, we Mountains or some other center of origin. must also account for the extensive polymor- Pleistocene ice ages repeatedly forced floral 558 Great Basin Naturalist Vol. 47, No. 4 and faunal elements to lower elevations and and S-Me-A in the Deep Creek sample, and may have facilitated intermittent gene flow Sod-"r in the Oquirrh Mountain sample) also among chipmunk populations. This may have suggests the possibility of linkage disequi- been sufficient to maintain allelic variants in librium in small, nonrandom mating popula- most populations. Without additional genetic tions. Several other studies have shown that information from hypothesized source popu- small population size per se is not always ac- lations (i.e., Wasatch Range) and others more companied by strong inbreeding, as various distantly isolated in Great Basin mountain species of mammals avoid consanguinous mat- ranges, we cannot choose among these alter- ings by a number of behavioral mechanisms natives. (Foltz and Hoogland 1983, Hoogland 1982, Ecological and behavioral factors may be as Patton and Feder 1981, Schwartz and Ar- important as historical events in determining mitage 1980). Patton and Feder (1981), for the genetic structure of chipmunk popula- example, found a paradoxical situation in tions. For example, in addition to the disper- which high heterozygosity was maintained in sal barriers between populations (i.e., desert apparently very small breeding units of the valleys, lakes, rivers, and distance), chip- gopher Thomomijs hottae, and this was ex- munks also face problems of short-distance plained as an equilibrium achieved between dispersal imposed by complex, interspecific the rate of migration (either recolonization competition, interspecific territoriality (Broad- following extinction or individual recruitment books 1970, Brown 1971b, Heller 1971), habi- into groups) and the effective number of indi- tat requirements (Sharpies 1983), predation, viduals that are contributing to the breeding altitudinal zonation (Chappell 1978, Heller effort each year. We do not have the ecological 1971), and philopatry to home range (Broad- or pedigree information necessary to evaluate books 1970, Martinsen 1968, Sheppard 1972). the importance of these factors in E. dorsalis, Broadbooks (1970), Martinsen (1968), and but their prevalence in other , and the Sheppard (1971) found three significant be- previously mentioned behavioral traits of havioral characteristics of yellow-pine chip- other Eiitamias, collectively suggest that in- munks (£. amoentis) and least (£. lywwnus) breeding alone cannot explain all of the ob- chipmunks that would influence the geo- served heterozygote absences in these popu- graphic distribution of allele frequencies: (1) lations. If it did, it should have a more or less chipmunks have a well-defined home range in equal influence across all variable loci, and which they remain from year to year, (2) a high this is not the case (Table 3). percentage (8 of 11) of the offspring remain in Alternatively, the high frequency of fixed the area of the parent, and (3) 67.4% of chip- allelic differences among different individuals munks released .4 km from their home range within the same sample suggests that we may returned within 1-3 days after release. well have pooled breeding units that differed

If similar behavior is typical of £. dorsalis drastically in their allelic composition populations, then breeding units may be char- (Wahlund effect). The Deep Creek sample acterized by high incidences of parent-off- displayed heterozygote deficiencies at eight spring or sib matings. Some evidence of in- loci and was comprised of collections from two

breeding is given by the F-statistics. For different localities (Table 1, Fig. 1), but the example, when averaged across all samples, excess number of homozygotes in the total F,s values were mostly high and positive, an sample did not correlate with the numbers of indication of heterozygote deficiency for individuals from either of these two sites. In

many loci (Table 5). This is due to the com- other words, this effect did not disappear plete absence of heterozygotes at some loci in when these samples were analyzed sepa- the five localities for which sample sizes were rately. Similarly, the House Range sample statistically "adequate" (all but Stansbury). was collected from two localities and showed The Fjs values may reflect either high levels of heterozygote deficiencies at five loci; again inbreeding or further levels of subdivision the phenomenon was independent of sample within our "samples" of £. dorsalis, but other localities. The Oquirrh and Raft River sam- explanations are possible. For example, the ples were collected from one canyon each, frequent occurrence of double homozygotes and both samples showed heterozygote in some loci segregating three alleles (Xdh-A deficiencies at the same three loci (S-Me-A, October 1987 DoBSONETAL: Cliff Chipmunk Variation 559

Sod-"l", and Xdh-A). Chesser (1983) has evaluated rates of protein inactivation (for 27 shown that important patterns of genetic vari- enzymes) under controlled conditions in four abihty may be obscured when breeding units species of mammals of varying body size {An- are pooled together, and we suspect that our tilocapra amcricana, Plecotiis townsendii, "samples ' of £. dorsalis may include separate Dipodomijs ordii, and Peromyscus boijlii) and Mendelian units that may differ drastically in found that 95% of the proteins routinely ex- allelic composition at some loci. amined electrophoretically are still stable We recognize the risk of over-analyzing (i.e., not denatured and showing mobilities these data in light of the small sample sizes but identical to controls) in unfrozen tissues for a feel that at least some other possible explana- minimum of 12 hrs after death. The locus tions for the complete absence of het- Sod-'l" had no heterozygotes in all five of the erozygotes at many loci can be ruled out. The E. dorsalis samples but was one of the most possibilities include: (1) inadvertant inclusion stable systems studied by Moore and Yates of a second species of Eutamias in the sam- (1983); the least stable system examined by ples, (2) scoring of multiple loci for some en- them, ADH, was not included in our protocol zyme systems in only select individuals from (Table 2). Further, our method of obtaining each sample, and (3) enzyme denaturation from the field insured that tissues and/or posttranslational modification of gene were taken from specimens and frozen in liq- products in select individuals. uid nitrogen within 30 min of capture. Labo-

Eutamias minimus is sympatric with E. dor- ratory protocol for homogenizing and storing salis at all localities sampled, but the latter is tissue samples was consistent throughout the very distinct, and CLP and MLD have had study, so there seems to have been little op- considerable experience with both species. portunity for extensive contamination or de- Museum voucher specimens were prepared naturation of the samples. for all individuals used in this study, and a The possibility of epigenetically or post- recheck confirmed their identification as E. translationally modified electromorph mobili- dorsalis. We conclude that there is almost no ties (see Lebherz 1983) in some E. dorsalis chance of "mistaken identity and that this specimens is one that we cannot evaluate with explanation would not, by itself, account for the information we have. Some classes of the different locus combinations displaying these alterations are known to have a genetic heterozygote absence at the five localities basis in some organisms (Womack 1983, tested. Dykhuizen et al. 1985), and in at least one Second, we can rule out the likelihood of species, mobility differences in two scoring different loci from a multilocus en- different loci (Trf and Ap-A) seem to vary with zyme in different individuals from the same the physiological state of the (McGov- populations, because the number of loci en- ern and Tracy 1981). If this is the explanation coding the enzymes used in this study is well for most or all of the rare homozygotes we known in mammalian systems (Harris and encountered, then the physiologically or ge- Hopkinson 1976). A single tissue type was netically based phenomenon for such electro- used in most electrophoretic runs (liver, see morph mobility alterations must be wide-

Table 2), but even when others were used, spread in £. (forsa/js populations. Elimination multilocus systems were evident either as two of these individuals from our analyses would zones of activity on the same gel, or as differ- lower the mean inbreeding coefficient (F,s) ent patterns of variability evident in different and perhaps slightly decrease mean D and Fst tissues of the same individual. The rare ho- values, although our conclusions about a mod- mozygotes we resolved were scored as such erate level of population subdivision and min- from zones of different mobility in one or a few imum genetic divergence would be virtually individuals on gels that otherwise contained a unaltered. single electromorph common to all other We suggest that the montane mammal pop- specimens, with the same tissue type being ulations of the Great Basin offer excellent used throughout. model systems for addressing issues in island The problem of enzyme denaturation and/ biogeography and population biology, but or posttranslational modification is more that future sampling strategies be designed to difficult to assess. Moore and Yates (1983) collect an adequate number of individuals 560 Great Basin Naturalist Vol. 47, No. 4

FoLTZ, D. W., 1983. evi- (n = 25, if possible) from a single ecologically and J L Hoogland Genetic dence of outbreeding in the black-tailed prairie homogenous site, and that, for larger moun- dog (Cynoinijs ludovicianns). Evolution 37: tain ranges at least, two or more localities be 273-281. collected analyzed as separate population and Harris, H , and D A Hopkinson 1976. Handbook of samples in order to assess within, as well as enzyme electrophoresis in human genetics. between, mountain range divergence. Such North-Holland Publ. Co., Amsterdam. Heller, H C 1971. Altitudinal zonation of chipmunks control in sampling will allow for a more rigor- (Eutamias): interspecific aggression. Ecology 52: ous assessment of macrogeographic patterns 312-319. of gene flow and population structure. Hoogland, J. L. 1982. Prairie dogs avoid extreme in- breeding. Science 215: 1639-1641. International Union of Biochemistry Nomenclature Acknowledgments Committee. 1984. Enzyme nomenclature, 1984. Academic Press, Orlando, Florida.

Jensen, N , R Merkle, and D I Rasmussen 1975. We thank Drs. J. N. Jensen, D. E. Jeffery, J J Locus number and multiple hemoglobins in Eu- and D. K. Shiozawa for helpful suggestion, tamias and Peroimjscus. Biochem. Genet. 14: stimulus, and critically reviewing previous ,541 -.545. drafts of this manuscript. We also thank R. Kaufman, D K , R K Selander, and M H Smith 1973. Dean for assistance with data analysis and D. Genie heterozygosity in a population oi Eutamias 54: 776-777. Currey for discussion about paleobiogeogra- panamintus. J. Mamm. Kilpatrick. C 1981. Genetic structure of insular pop- phy of the Great Basin. Permits for collection W ulations. Pages 28-59m M. H. Smithand Joule, were granted by the Utah Division of Wildlife J. eds.. Mammalian population genetics. University Resources. This work was supported in part of Georgia Press, Athens. by the Department of Zoology and the Asso- Lebherz. H, G 1983. On epigenetically generated ciated Students of Brigham Young University. isozymes ("pseudoisozyrnes") and their possible biological relevance. Pages 203-219 in M. C. Rat- S. eds.. tazzi, J. G. Scandalios, and G. Whitt, Literature Cited Isozymes: current topics in biological and medical research. Vol. 7. A, R. Liss, New York. Levene. H 1949. On a matching problem arising in ge- Black, C. C. 1972. Holarctic evolution and dispersal of netics. Ann. Math. Stat. 20: 91-94. (Rodentia: Scuridae). Evol. Biol. 7; Martinsen. D L. 1968. Temporal patterns in the home 305-322. range of chipmunks (Eutamias). Mamm. 49: Broadbooks. H. E. 1970. Home ranges and territorial J. 83-91. behavior of the yellow-pine chipmunks, Etttamias

McGovERN. M , andC R Tracy 1981. Phenotypic varia- amoenus. J. Mamm. 51: 310-326. tion in electromorphs previously considered to be Brown, J. H 1971a. Mammals on mountaintops: non- genetic markers in Microtus ochrogaster. Oecolo- equilibrium insular biogeography. Amer. Nat. gia 51: 276-280, 105: 467-478. varia- Mewaldt. T , AND S H Jenkins 1986. Genetic 1971b. Mechanisms of competitive exclusion be- W

tion of woodrats (Neotoma cinerea ) and deer mice tween two species of chipmunks. Ecology 52: on montane island 305-311. (Peromyscus maniculatus) habitats in the Great Basin. Great Basin Nat. 46: 1978. The theory of insular biogeography and the .577-.580, distribution of boreal birds and mammals. Great

Moore, D \V , and T L Yates 1983. Rate of protein Basin Nat. Mem. 2; 209-227. inactivation in selected mammals following death. Chappell, M a. 1978. Behavioral factors in the altitudi- Wildl. Manage. 47: 1166-1169. nal zonation of chipmunks (Eutamias). Ecologv J,

, 1985. Evolutionary 59: 565-579. Murphy, R W andC B Crabtree patterns, of loci, and Chesser, R. K. 1983. Genetic variability within and aspects of isozyme number in the prairie rat- among populations of the black-tailed . tissue-specific gene expression viridis. Herpetologica Evolution 37: 320-331. tlesnake, Crotalus viridis .30: 451-470, Currey, D R., and S. R. James 1982. Paleoenvironments of the northeastern Great Basin and northeastern Nei, M 1972. Genetic distance between populations. Basin Rim region: a review of geological and bio- Amer. Nat, 106: 283-292. logical evidence. Man and environment in the 1978. Estimation of average heterozygosity and Great Basin. Soc. Amer. Arch. 2; 27-52. genetic distance from a small number of individu- als. Genetics 89: ,583-590, Dykhuizen, D E , C MuDD, A Honeycltt, and D L 1975. The Hartl. 1985. Polymorphic post-translational Nei, M , T Maruyama, and R Chakraborty' modification of alkaline phosphatase in £s- bottleneck effect and genetic variability in popula-

cherichia coli. Evolution 39: 1-7. tions. Evolution 29: 1-10. (National and Atmospheric Adminis- Fisher. S E . J B Shaklee, S D Ferris, and G S NOAA Oceanic Whitt. 1980. Evolutionof five multilocus isozyme tration). 1983. Climatological data— Utah. 85 systems in the . Genetica 52/53: 73-85. (12, 13). October 1987 DoBsoN ETAL: Cliff Chipmunk Variation 561

Patton, J L . AND J H Feder. 1981. Microspatial genetic SwoFFORD, D L , and R. B. Selander. 1981. BIOSYS-1: a heterogeneity in pocket gophers: non-random FORTRAN program for the comprehensive analy- breeding and drift. Evolution 35: 912-920. sis of electrophoretic data in population genetics

Paiton. . 1977. variation J L AND S Y Yang Genetic in and systematics. J. Hered. 72: 281-283.

Thomomys bottae pocket gophers: macrogeo- Tho.mpson, R S , AND J I. Mead. 1982. Late Quaternary graphic patterns. Evolution 31: 697-720. environments and biogeography in the Great P.atter.S()N, B D 1980. Montane mammalian biogeogra- Basin. Quat. Res. 17: 39-55.

phy in New Mexico. Southwest Nat. 25: 33-40. Van Devender, T R , and J E King 1971. Late Pleis- 1982. Pleistocene vicariance islands montane and tocene records in western Arizona. J. Arizona the evolutionary divergence of some chipmunks Acad. Sci. 6: 240-244.

(genus Eidamias). J. Mamm. 63: 387-398. Van Devender, T R., and W G. Spaulding. 1979. De-

Rogers, J. S 1972. Measures of genetic similarities and velopment of vegetation and climate in the south- genetic distance. Univ. Texas Publ. Genet. 7213: western United States. Science 204: 701-710. 145-153. Wells, P V 1983. Paleobiogeography of montane islands

Schwartz. O A , and K B. Armitage. 1980. Genetic vari- in the Great Basin since the last Glacial pluvial. ation in social mammals: the model. Sci- Ecol. Monogr. 53: 341-382. 665-667. ence 207: Wells, P H , and R Berger 1967. Late Pleistocene

Selander, R K , M H Smith, S Y Yang, W E Johnson, history of coniferous woodland in the Mohave

and J B Gentry 1971. Biochemical polymor- Desert. Science 155: 1640-1647.

phism and systematics in the genus Peromijscus. 1. Womack, J E 1983. Post-translational modification of Variation in the old-field mouse [Peromijscus po- enzymes: processing genes. Pages 175-186 in M. lionutus). 49-90. Univ. Te.xas Publ. Genet. 7103: C. Rattazzi, J. G. Scandalios, and G. S. Whitt, Sharples, F E 1983. Habitat use by synipatric species of eds.. Isozymes: current topics in biological and

Eutamias. J. Mamm. 64: 574-579. medical research. Vol. 7. A. R. Liss, New York. 1971. Sheppard, D H Competition between two chip- Workman, P. L., and J. D. Nisvvander. 1970. Population

munk species (Eutamias). J. Mamm. 52: 320-329. studies on southwestern Indian tribes. 11. Local 1972. ranges Home of chipmunks {Eutamias) in genetic differentiation in the Papago. Amer. J.

Alberta. J. Mamm. 53: 379-380. Hum. Genet. 22:24-49.

Sne.iiTH, P H a , andR R Sokal 1973. Numerical taxon- Wright, S 1965. The interpretation of population struc- omy. W. H. Freeman, San Francisco. ture by F-statistics with special regard to systems Sullivan, R M 1985. Phyletic, biogeographic, and eco- of mating. Evolution 19:395-420. logical relationships among montane populations 1978. Evolution and the genetics of populations. of least chipmunks (Eutamias minimus) in the Vol. 4. Variability within and among natural popu- Southwest. Svst. Zool. 34: 419-448. lations. University of Ghicago Press, Chicago.