Comp. Biochem. Physiol., 1976, Vol. 54B, pp. 243 to 248. Pergamon Press. Printed in Great Britain

BIOCHEMICAL VARIATION AND GENIC SIMILARITY OF MYOTIS VELIFER AND MACROTUS CALIFORNICUS DONALD O. STRANEY,1 MICHAEL H. SMITH,1 ROBERT J. BAKER2 AND IRA F. GREENBAUM2 ~Savannah River Ecology Laboratory, Drawer E, Aiken, SC 29801 and 2Department of Biology, Texas Tech University, Lubbock, TX 79409, U.S.A.

(Received 15 April 1975) Abstract--1. Allozymic data for 17 gene loci were used to derive heterozygosity estimates for two bat species, Myotis velifer (H = 0.144) and Macrotus californicus (H = 0.030). 2. The high heterozygosity value for M. velifer can probably be best attributed to differential selection between sexes. 3. The two species, representatives of two distinct families, share no alleles at the loci studied. INTRODUCTION GENIC heterozygosity has been estimated from allozymic data for a relatively large number of mammalian species. Most species studied have been from one order (Rodentia; Selander & Johnson, 1973). Individual heterozygosity averages 0.056 and ranges from 0.008 for Thomomys talpoides (Nevo et al., 1974) to 0"110 for Mus musculus brevirostris (Selander & Yang, 1969). Despite this narrow range, there is geographical intraspecific variation in heterozygosity (Selander et al., 1971 ; Johnson et al., 1972) and some indication that higher values may be characteristic of certain mammals other than rodents (Ramsey, personal communication). Our primary objective was to extend these studies to include representatives of the Chiroptera. Two species of bats, Myotis velifer and Macrotus californicus (sensu Davis & Baker, 1974), were examined in the present study. They represent two families of bats (Vespertilionidae and Phyllostomatidae) and would be expected to show a low degree of genic similarity (Smith et al., 1973). Both species are colonial, inhabiting caves in large numbers, unlike other bat species studied which are solitary roosting forms (Straney et al., 1976). In addition, Macrotus is the first truly homeothermic species of bat assayed for aUozymic variation. MATERIALS AND METHODS

Techniques of tissue preparation, electrophoresis and biochemical staining were similar to those described in Selander et al. (1971). One exception, not studied by Selander et al., was alkaline phosphatase (AKP) which was obtained from kidney extract and processed using a discontinuous Tris-citrate (Poulik) buffer. AKP was stained with Na-/~-naphthyl-acid-phosphate (25 mg), Fast Blue RR salt (25 mg), and MgSO, (60 mg), in 0.2 M Tris--hydrochloric acid buffer, pH 8'0 (50 ml). Seven gel types (Electrostarch Lot 371, Otto Hiller, Madison, Wisconsin) were used and stained as follows: (1) Phosphate (liver)--phosphoglucose isomerase (PGI); (2) Continuous Tris-citrate I (kidney)---lactate dehydrogenase (LDH-1), isocitrate dehydrogenase (IDH-1), malate dehydrogenases (MDH-1 and MDH-2) and phosphoglucomutase (PGM-1); (3) Continuous Tris-citrate II (liver)--~tglycerophosphate dehydrogenase (ot-GPD), glutamic oxalacetic transaminase (GOT-l) sorbitol dehydrogenase (SDH) and alcohol dehydrogenase (ADH); with kidney--isocitrate dehydrogenase (IDH-2); (4) Tris-hydrochloric acid (liver)--general protein (GPT-I) and albumin (ALB); with hemolysate-hemoglobin (HB); (5) Discontinuous Triscitrate (Poulik) (kidney)--alkaline phosphotases (AKP-I and AKP-2), esterases (ES-1, ES-2 and ES-4); (6) Trismaleate (kidney)--6-phosphogluconate dehydrogenase (6PGD); (7)lithium hydroxide (plasma)--transferrin (TRF) and plasma protein (PP-1); with liver--indolephenol oxidases (IPO-1 and IPO-2), lactate dehydrogenase (LDH-2) and glutamic oxalacetic transaminase (GOT-l). Due to the nature of electrophoretic techniques (Smith et al., 1973) direct side-by-side comparison of mobilities of allozymes on the same gel was used for comparison of alleles. For each locus, the allele occurring in the highest frequency in Myotis velifer was designated 100 if migration was anodal or -100 if cathodal. Other allozymic bands and their corresponding alleles were designated numerically as percentages of distances migrated relative to that of the 100 allele. The most anodal locus in a system was numbered "1"; more cathodal loci received progressively higher numerical designations.

One hundred and sixty five bats including 45 Macrotus californicus (303' and 15~) and 116 Myotis velifer (653 and 51~) were analyzed for allozymic variation. Individuals were prepared as museum specimens and are on deposit in The Museum, Texas Tech University. Localities and sample sizes are as follows: Macrotus californicus. Arizona: Pima Co.; 25 mi. S Casa Grande, Old Mamman Mine (n = 45). Myotis velifer. Arizona: Pima Co.; 25 mi. S Casa Grande, Old Mamman Mine (n = 33). Texas-1 : ColRESULTS lingsworth Co.; 2"3 mi. N, 2 mi. E Wellington (n = 43). Texas-2: Childress Co.; 14.5 mi. S Wellington (n = 40). The alleles found at each locus and their frequenHeart, liver and kidney of each specimen (except Myotis velifer from Texas) were frozen on dry ice immediately after cies are presented in Table 1. In Myot~s velifer, 10 sacrifice and shipped by air to the Savannah River Ecology loci (62'5%) were segregating for more than one allele, whereas in Macrotus californicus there were seven Laboratory. Heart and kidney extracts were processed together. The Myotis velifer from Texas were flown alive such loci (44%). N o locus deviated significantly from to the laboratory, bled, and tissues removed immediately. Hardy-Weinberg expectations and sexes within a 243

244

DONALD O. SfRANEY, MI('HAt£L H. SMIIH, R()BERI ,I. BAKER AND IRA I. (JRI!LNF~:\I Xl

population did not differ significantly in allele frequencies or heterozygosity. The banding patterns of the allozymes at each locus in our study are similar to those found by Selander et al. (1971) in Peromyscu,s. We infer a similar genetic base for these loci. Alkaline phosphatase (AKP), not studied by Selander et aL, was monomorphic within the two species studied here. Banding patterns for MDH, and LDH differ considerably from those reported by Carmody et al. (1971) for Myotis luc(fugus. We cannot interpret their results for these loci within a genetic model. Esterase phenotypes, however, correspond quite closely to those we have found in Myotis t:eli/br. A number of systems deserve special mention. On gels overstained for PGM a faint set of bands appeared anodal to what we have labeled here as PGM-1. We were unable to consistently score this alternate locus. In a like manner, malic enzyme (ME: Smith et al., 1973; Shows et al., 1970) appeared on gels overstained for MDH. This enzyme is NADP-dependent in both species and lies between MDH-1 and the origin in Macrotus californicus. In Myotis l,eli[er, the M E band partly overlaps that of Mdh-1 zoo Four esterase loci were evident in this study. Both ES-1 and ES-4 were eserine sensitive. ES-2 was the only esterase present in Macrotus cali/brnicus. It may be homologous to "ES-Y' in Myotis ~,elifer which is slightly cathodal to ES-2 on gels where both species were included. We could not score "'ES-Y' consistently and have not included it here. It may represent the products of more than one locus. ES-4 in M.votis

t'el(ler migrates only a short distance anodal tiom the origin. Originally, four loci were scored on buffer systems different from those reported above (LDH-2, IDH-2 with continuous Tris-citrate I; PGM-1 with Tris maleate: GOT-2 with continuous Tris-citrate II). At the LDH-2, IDH-2 and GOT-2 loci, mobilities of alleles appeared to be the same. The P qm-21"° allele of Myotis L'el([er appeared to be the same as the Pore-21°5 allele of Macrotus cal!lbrnicus. Since these data represented a Rogers' similarity value of between 0.267 and 0.281 for interspecific (in this case also interfamilial) comparisons, samples were re-run with all possible buffer types. These allozymes do differ in mobility on the buffer systems reporled in the methods above, as indicated in Table 1. Nine loci scored in Texas populations of Mvotis vel!/~'r could not be scored in the two Arizona samples. We did not obtain blood samples from the latter, eliminating transferrin (TRFJ. plasma protein (PP-I) and hemoglobin (HB). In addition, these samples arrived in an unfrozen state and it appears that the other six enzymes (ADH, AKP-2, IDH-I, SDH, IPO-I and IPO-2) denatured before they were processed. Of these six, only ADH gave scorable results for a short while, but then only for Macroms cal(/brnicus. These nine loci were not included in any of the calculations. However, if heterozygosity values (H) for the Texas populations are calculated both inclusive and exclusive of these loci, the values differ by only ± 0.0l. This indicates that values for H calcu-

Table 1. Alleles and frequency (in parentheses) at each locus in three populations of Mvotis veliJbr and Macrotus caliJbrnicus, h is the proportion of individuals heterozygous at a given locus and is averaged over the three M. l~el!fer populations, ns indicates the locus was not scorable but was present (see textt M~otis velifer Locus

1

~-GPD

Mac rotus calTfornicus

Texas-i

Texas-2

Arizona

h

Arizona

h

i00(.919)

i00(.012)

100(.900)

.194

168(.950)

.i00

97(.012)

117(.088)

115(.040)

127(.762)

124(.060)

127(.069)

184(.050)

137(.138) AKP-I

i00(I.00)

i00(i.00)

i00(i.00)

0

120(1.00)

ALB

100(.870)

i00(.825)

100(.950)

.200

114(.980)

91(.130)

91(.175)

91(.050)

ES-I 2

100(.670)

100(.760)

100(.548)

.397

ABSENT

107(.100)

107(.060)

i07(.258)

107.5(.190)

107.5(.130)

i07.5(.129)

114(.030)

114(.050)

114(.065)

ES-2 2

ABSENT

ABSENT

ABSENT

ES-4 2

100(1.00)

100(1.00)

100(1.00)

GOT-I

100(.942)

100(.850)

100(.890)

68(.058)

68(.150)

68(.110)

GOT-2

-i00(.558)

-100(.500)

-100(.530)

-79(.442)

-79(.500)

-79(.470)

GPT-I

-100(.977)

-i00(i.00)

IDH-2

-i00(i.00) i00(i.00)

0

.044

103(.020)

100(.8L0)

.200

118(.140) 0 .193

ABSENT 82(.990)

.022

128(.010) .044

.427

-100(.030

-100(1.00)

.015

-225(1.00)

0

-100(l.00)

-i00(i.00)

0

-102(1.00)

0

I00(i.00)

I00(i.00)

0

I17(I.C0)

0

-89(.970)

-84(.023)

LDII-I

Biochemical variation in bats Table i.

245

Continued.

Macrotus californicus

Myotis velifer h

Arizona

h

Locus

Texas-I

Texas-2

Arizona

LDH-2

-i00(i.00)

-i00(1.00)

-100(1.00)

0

-ii0(i.00)

0

MDH-I

100(.954)

i00(.960)

i00(i.00)

0.056

148(1.00)

0

90(.046)

90(.040)

-100(1.00)

-100(l.O0)

-100(1,00)

0

-114(.990)

MDH-2

.022

-120(.010) PGI

-i00(.942)

-i00(.950)

-89(.058)

-89(.038)

-100(1.00)

.072

-141(1.00)

100(.970)

.i01

105(.980)

0

-85(.012) PGM-1

100(.953)

100(.925)

70(.023)

70(.025)

200(.012)

200(.050)

.044

64(.020)

200(.030)

210(.012) 6PGD

82(,634) 85(.366)

82(.870)

i00(.780::

85(.280)

85(.220)

73(1.00)

.414

100(.050) ADH

AKP-2

-100(.849)

-100(.872)

-80(.151)

-80(.128)

i00(I.00)

i00(i.00)

-79(1.00)

ns

ns

ns

HB

100(1.00)

100(1.00)

ns

ns

IDH-I

100(.744)

100(.800)

ns

ns

105(.256)

105(.200)

100(.663)

100(.725)

ns

ns

75(.302)

80(.262)

IPO-1

139(.035)

139(,~13)

IPO-2

i00(i.00)

i00(i.00)

ns

ns

PP-I

i00(I.00)

i00(I.00)

ns

ns

SDH

i00(.849)

80(.762)

ns

ns

200(.151)

200(.225)

1O0(l.00)

i00(i.00)

ns

ns

62(.013) TRF

i.

Abbreviations are as follows:

2.

Esterases scored assu~ing

uGPD = ~-Glycerophosphate dehydrogenase

the following structures:

AKP-1 = Alkaline phosphatase-i

Esterase-l, monomer;

AL_BB = Albumin

Esterase-2# monomer;

ES-I r -2, and -4 = Esterase i, 2, and 4

Esterase-4, monomorphic.

GOT-I and -2 = Glutamic oxalacetic transaminase 1 and 2 GPT-I = General protein 1 IDH-I and -2 = Isocitrate dehydrogenase 1 and 2 LDH-I and -2 = Lactate dehydrogenase 1 and 2 MDH-1 and -2 = Malate dehydrogenase 1 and 2 PG__~I= Phosphoglucose isomerase PC@~-I = Pho3phoglucomutase 1 6PG____D= 6 phosphogluconate dehydrogenaae AD}___!= Alcohol dehydrogenase HB = Hemoglobin IPO-I and -2 = Indolephenol oxidase 1 and 2 PP~I = Plasma protein 1 SDH=

Sorbitol dehydrogenase

T~=

Transferrin

246

DONALD 0 . STRANEY, MICHAEL H. SMITH, ROBER'I J. BAKER AND IRA |;. GREI~NBA[",1

Table 2. Heterozygosity estimates (H; in parentheses) and genie similarity between Myotis veli~er and Macrotus cal(['ornicus based upon Rogers' S (Rogers, 1972). H is the mean number of loci heterozygous per individual M[otis velifer

Texas-I Texas-2

Macrotus californicus

Texas-i

Texas-2

Arizona

(0.125)

(0.163)

(0.i01)

l. O00

(0.030)

.922

.922

1.000

.871

0.000

1.000

0.000

Arizona

lated from the full data set of 16 loci are fairly robust (Table 2). We cannot judge what effect these alleles would have had on our figures for genic similarity (Table 2), although hemoglobin is known to differ between the two species (Mitchell, 1970). DISCUSSION

Genic variability. Selander & Kaufman (1973) have suggested that patterns of heterozygosity values reported to date are intimately related to the environmental grain (Levins, 1968) to which populations or species are exposed. They found that organisms living in coarse grained environments (high effective environmental variance) had higher levels of genic heterozygosity than do organisms within fine grained environments. Bryant (1974) reports that up to 70% of geographic variation in heterozygosity in a number of organisms can be explained by measures of environmental variability. An organism's ability to exercise some form of homeostatic control over its internal physiology in response to environmental changes will reduce the effective environmental variance and make the grain finer. Most vertebrates function in this way and, according to Levins' model, it is not necessarily adaptive for them to maintain high genetic variance. The opposite is true for small invertebrates which lack as effective homeostatic mechanisms. Macrotus californieus has a low heterozygosity value, even for a mammal (Table 2). Leitner & Ray (1964) report that individuals of this species show good control of body temperature and will not enter hypothermia even in the laboratory. In addition, Macrotus californicus roosts only in caves well protected from climatic elements and occupies the warmest caves available (Leitner & Ray, 1964). By behaviorally buffering the variance of ambient temperature, the environmental grain has been effectively reduced. Thus, Maerotus californicus is stenothermal and fits the low heterozygosity mode described by Selander & Kaufman (1973). Myotis velifer is the most genetically variable vertebrate yet reported (Table 2). Indeed, H for this species (0.144) is approximately equal to that for 19 species of Drosophila (H = 0-145; Selander & Kaufrnan, 1973) and is within the range found for other invertebrates (Krepp & Smith, 1974). Myotis velifer is thermolabile, and unlike Macrotus californicus, it will enter torpor and either hibernates (Texas; Kunz, 1973) or migrates (Arizona; Hayward,

0.000

1970) in the winter. Exposure of thermolabile individuals to different environmental conditions in migration effectively increases environmental variance (see Villa, 1967). Although hibernating individuals spend most of their time within fairly constant-temperature hibernacula, in species like Myotis velifer which are capable of flight at low temperature, individuals are exposed to considerable thermal heterogeneity (Studier & O'Farrell, 1972). Only four other species of mammalian hibernators have been assayed for heterozygosity levels. Eutamias panamintinus (Kaufman et al., 1973) and Perognathus hispidus (Johnson, 1970), however, have values of H within the range of other mammals. It is possible that other factors are operating to reduce heterozygosity in these species. Eutamias panamintinus is disjunctly distributed on mountains in southwestern Nevada and California. Avise et al. (1974) have discussed the decrease in genetic variability observed in Peromyscus on islands in the Gulf of California. It is possible that a similar situation exists in the isolated Eutamias population studied by Kaufman et al. (1973). Perognathus hibernates in underground burrows where temperature fluctuation is less than that above ground (Kenagy, 1973). Hence, effective environmental variance may be lower for hibernating Perognathus than for cave hibernating Myotis. Myotis californicus and Pipistrellus hesperus hibernate in essentially the same manner as does M. velifer, yet show great differences in levels of heterozygosity (Straney et al., 1976). There is thus little reason to expect hibernating mammals to show high levels of heterozygosity. Any relationship that does exist is probably a weak one, with strong factors such as drift or selection capable of masking any effect hibernation may have on increasing genetic variability. In another study (Straney et al., 1976) we examined two species of bats known to be heterothermic. Pipistrellus hesperus and Myotis ealifornicus differed considerably in heterozygosity (H = 0.026 and 0.126, respectively). We concluded that there was no relationship between thermoregulatory ability and level of genic heterozygosity in these species. Thus, although Macrotus californicus and Myotis velifer fit Selander & Kaufman's (1973) model, we doubt that the magnitude of heterozygosity in either species is due solely to thermoregulatory ability. Our data do indicate that another mechanism could be contributing to the high level of genic variability found in Myotis vel!fer..

Biochemical variation in rats Populations of Myotis velifer exhibit sharp discontinuities in gene frequencies over short distances. An analysis of variance of individual heterozygosity in relation to location indicates that the three Myotis velifer populations differ significantly from each other in mean heterozygosity (P < 0.05). Contingency table analysis indicates that this difference is paralleled by significant allele frequency/location heterogeneity at three loci: 6PGD (P < 0.01), ctGPD (P < 0.01) and ES-1 (P = 0.02). Though the Texas-1 and Texas-2 populations are only approximately 30 km apart, they could be parts of separate genetic populations. The Texas-1 population was an all male colony whereas Texas-2 and Arizona samples were maternity colonies with males representing respectively 27 and 30% of the population. Hayward (1970) has discussed the dynamics of male/female segregation in this species. Male Myotis velifer inhabit marginal regions of a roost, usually near the entrance (Kunz, 1973), migrate first, or hibernate further north and use hibernacula colder than those used by females (Hayward, 1970). Levene's (1953) model of selection in spatially heterogeneous environments seems applicable to the situation found here and predicts the maintenance of variability within certain limits. Giesel (1972) has suggested that differential habitat selection and concomitant selection differences between sexes allows species such as Myotis austroriparius to better track short term environmental changes genetically. Exposure of males to selective regimes different from those of females should produce differences between the sexes in levels of heterozygosity and/or allele frequencies. Neither trend is apparent in our data for males and females within maternity colonies. We do not know what the differences would be in the all male colonies which form breeding colonies with these females. If they are as different from the maternity colonies as is the Texas-1 population, it is possible that differential habitat selection of the sexes could maintain the high levels of genic variability observed in Myotis velifer. While this is an attractive hypothesis, there is a gross lack of population data necessary to test it. However, differences in sex ratio and heterozygosity between Macrotus californicus ("regulator"; 2~:1~) and Myotis velifer ("conformer"; 13' :29) do support Giesel's (1972) argument that differential selection between sexes is an effective means of tracking short term environmental changes genetically. Genic similarity. Values of genic similarity for the three populations of Myotis velifer fall within the range for intraspecific populations found in rodents (Selander & Johnson, 1973). The lack of similarity between Myotis velifer and Macrotus californicus is not unexpected. Johnson & Selander (1971) found that two genera within the rodent family Heteromyidae were similar at the 0.16 level and other evidence from rodents (Smith et al., 1973) and lizards (Webster et al., 1973) suggests that species of different families should have genie similarity values well below 0-10. Fossils of early bat faunas are rare but limited data are available. Bats were well diversified in the Eocene (Russell & Sige, 1970; Jepsen, 1970). The first fossils assigned to the vespertilionid line of evolution is Stehlinia Revilliod from the Eocene-Oligocene boundary (Russell & Sige, 1970) and the genus Myotis is known

247

from the lower Oligocene (Quinet, 1965) or mid-Oligocene (Romer, 1966). The first fossil phyllostomatid, Notonycteris is from the late Miocene (Savage, 1951). Although fossils of phyUostomatid bats have not been found in earlier geological strata, workers have indicated that the lines that gave rise to the Phyllostomatidae were diverged from the vespertilionid stock at least in the Eocene and probably earlier (Russell & Sige, 1970; Smith, 1972). Data to support such interpretations are from comparative morphology and fossils from other lines of bat phylogeny. These two families have belonged to distinct lineages for at least 30 million years. It is not surprising, therefore, that the representative species we have studied share no alleles. SUMMARY

Myotis velifer and Macrotus californicus from Texas and Arizona were assayed for electrophoretically demonstrable allozymic variation at 17 loci. The heterozygosity estimate for Macrotus californicus was low for a mammal (H = 0.030). This species has behavioral mechanisms which effectively reduce the variance in environmental temperature to which individuals are exposed. In contrast, Myotis velifer is the most genetically variable mammal studied to date (H = 0.144; range, 0.101-0.163). Because members of this species migrate or hibernate in the winter, Myotis velifer is exposed to both coarse and fine grained environments. Levels of heterozygosity appear to be at least partially related to environmental grain in these species. Differential habitat selection by sex may also contribute to the high genetic variability exhibited by this species. Conspecific genic comparisons resulted in similarity (S) values within the range observed for other mammals. Interspecific comparisons in this study were also interfamilial comparisons and indicate that the species share none of the sampled alleles. Acknowledgments---This research was supported by Contracts AT(38-1)-310 and AT (38-1)-819 between the U.S. Atomic Energy Commission and the University of Georgia. John Avise, Brent Davis, Mary Elam, Susan Fuller and Steven Tennison helped in the collection of the specimens or the data. Drs. W. E. Johnson and M. J. O'Farrell provided helpful comments on some of the results reported herein. REFERENCES

Axqs~ J. C., SMITHM. H., SELANDERR. K., LAWLORT. E. & RAMSEYP. R. (1974) Biochemical polymorphism and systematics in the genus Perornyscus--V. Evolutionary genetics of mainland and island species of the subgenus Haplomylomys. Syst. Zool. 23, 226-238. BRYANT E. H. (1974) On the adaptive significance of enzyme polymorphisms in relation to environmental variability. Am. Nat. 197, 1-19. CARMODYG. R., FENTONM. B. & LEE D. S. K. (1971) Variation in body weight and proteins in three Ontario populations of hibernating Myotis lucifugus (Le Conte) (Chiroptera: Vespertilionidae). Can. J. Zool. 49, 1535-1540. DAVISB. L. & BAKERR. J. (1974) Morphometrics, evolution, and cytogenetics of mainland bats of the genus Macrotus Gray (Chiroptera: Phyllostomatidae). Syst. Zool. 23, 26-39.

248

DONALD O. STRANEY, MICHAEL H. SMITH, ROBER1 J. BAKH~, AND IRA 1:. GRt-F;NBAIM

GIESEL 3. T. (1972) Sex ratio, rate of evolution and environmental heterogeneity. Am. Nat. 106, 380-387. HAVWARD B. J. (1970) The natural history of the cave bat Myotis velifer. West. New Mex. Univ. Res. Sei. 1(11, 1 74. JEPSEN G. L. (19701 Bat origins and evolution. In Biology of Bats (Edited by WIMSATT W. A.), pp. 1 64 Academic Press, New York. JOHNSON W. E. (19701 Protein variation and systematics in heteromyid rodents. Ph.D. dissertation, University of Texas at Austin. 154 + v p. JOHNSON W. E. & SELANDER R. K. (19711 Protein variation and systematics in kangaroo rats (genus Dipodomys), Syst. Zool. 20(41, 377-405. JOHNSON W. E., SELANDER R. K., SMITH M. H. & KIM Y. J. (1972) Biochemical genetics of sibling species of the cotton rat (Sigmodon). Stud. Genet. VII (Univ. Texas Publ. 7213), 207-305. KAUFMAN D. W., SELANDER R. K. & SMITH M. H. (19731 Genic heterozygosity in a population of Eutamias panamintinus. J. Mammal. 54(3), 776-778. KENAGY G. J. (1973) Daily and seasonal patterns of activity and energetics in a heteromyid rodent community. Ecology 54(6), 1201-1219. KREPP S. R. & SMITH M. H. (19741 Genic hetcrozygosity in the 13-year cicada, Magicicada. Evolution 28, 396-401, KUNZ T. H. (1973) Population studies of the cave bat (Myotis velifer): Reproduction, growth, and development. Occ. Pap. Mus. nat. Hist. Unit'. Kansas. 15, 1 43. LEITNER P. & RAY S. P. (1964) Body temperature regulation of the California leaf-nosed bat, Macrotus cal!fornicus. Am. Zool. 4(3), 105. LEVENE H. (1953) Genetic equilibrium when more than one niche is available. Am. Nat. 87, 331 ~333. LEVINS R. (1968) Evolution in Changin{t Environments. Princeton University Press, Princeton, New Jersey. 120 + ix p. MITCHELL G. C. (19701 An clectrophoretic comparison of hemoglobins in bats. Comp. Biochem. Physiol. 35, 667-677. NEro E., K ~ Y. J., Shaw C. R. & TnAELER C. S. JR. (1974) Genetic variation, selection and speciation in Thomomys talpoides pocket gophers. Evolution 28, 1--23. QUINET G. E. (1965) Myotis misonnei N. sp. Chiroptere de L'Oligocene de Hoogbutsel. Institut royal des Sciences naturelles de Belgique. Bulletin; Tome XLI, no. 20. ROGERS J. S. (1972) Measures of genetic similarity and genetic distance. Stud. Genet. VII (Univ. Texas Publ. 72131, 145 153.

ROMt~R A. S. (1966) Vertebrate Paleontology. Lnivcrsity of Chicago Press, Chicago, IlL 468+ viii p. RttSSELL D. E. & SIGE B. (19701 Revision des Chiropteres lutetiens de Messel (Hesse, Allemagne). Palaeovertehrata Vol. 3, Fasc. 4. SAVAGE D. E. (1951) A miocene phyllostomatid bat from Colombia, South America. Unit. of Calf Pubis 28, (12) 357 366. SELANDER R. K. & KAUFIVIAND. W. (19731 Genic variability and strategies of adaptation in animals, Proc. hath. Acud. Sci. U.S.A. 70(6t, 1875 1877. SELANDER R. K. & JOHYSON W. E. [19731 Genetic variation among vertebrate species. A. Rev. Ecol. Syst. 4, 75 91. SELANDER R. K. & YAN(; S. Y. (19691 Protein polymorphism and genic heterozygosity in a wild population of the house mouse (Mus musculus). Genetics 63, 653 667. SELANDER R. K., SMITH M. H., YANG S. Y., JOHNSON W. E. & GENTRY J. B. (19711 Biochemical polymorphism and systematics in the genus Peromyscus I. Variation in the old-field mouse (Peromyscus polionotus). Stud. Genet. VI (Univ. Texas Publ. 71031, 49 90. SHOWS T. B., CHAPMAN V. M. ~¢~ RUDDLE F. H. 11970) Mitochondrial malate dehydrogenase and malic enzyme: Mendelian inherited electrophoretic variants in the mouse. Biochem. Genet. 4, 707 718. SMITrt J. D. (1972) Systematics of the Chiropteran family Mormoopidae. Uniz. Kans. Pubis. Mus. Nat. Hist. Misc'. 56, 132 pp. SMITH M. H., SELANDER R. K. & JOHNSON W. E. (19731 Biochemical polymorphism and systematics in the genus Peromyscus- -III. Variation in the Florida deer mouse (Peromyscus floridanus), a Pleistocene relict. J. Mamnu~l. 54(1), 1 J3. SmANEY D. O., O'FARRELL M. J. and SMITH M. H. (19761 Biochemical genetics of Mvotis californicus and Pipistrellus hesperus. J. Mammall (In press). STUDIER E. H. 8¢ O'FARRELL M. J. (1972) Biology of M yotis thysanodes and M. lucifugus (Chiroptera: Vespertilionidae)- I. Thermoregulation. Comp. Biochem. Physiol, 41A, 567 595. VILLA B. 11967) Los Murcielagos de Mexico. Univ. Nac. Aut. Mexico, Mexico, D. F. 491 p. WEBSTER, T. P., SELANDERR. K., YANG S. Y. (1973) Genetic variability and similarity in the Anolis li7~ards of Bimini. Evolution 26(4), 523 535.