Polymorphism of Blood Plasma Proteins in the Anser and Branta Genera

Transcription

1 Biochemical Genetics, Vol. 33, Nos. 3/4, 1995 Polymorphism of Blood Plasma Proteins in the Anser and Branta Genera S. B. Kuznetsov 1 Received 21 June 1994 Final 29 Dec An electrophoretical analysis of blood plasma proteins of eight Anser and two Branta species was performed. Ten polymorphic proteins in blood plasma pattern were distinguished and described." four prealbumin proteins, albumin, three postalbumin proteins, transferrin, and a single posttransferrin protein. Genus-specific and species-specific variants of Pr-1, Al, Pa-3, Pa-X, and Tf proteins were found. The species of Branta differed in Pr-1, Pa-3, Pa-X, and Tf loci. The Anser species differed, apparently, in allele frequencies of described gene loci. A single species-specific protein marker was found in swan geese only. The electrophoretic mobilities of Pr-1, TfB, and PtfA, B, and C were similar for several species of Anser and Branta genera. KEY WORDS: blood plasma proteins; polymorphism; geese. INTRODUCTION There have been many studies of protein polymorphism in different classes of animals, but only recently has much work been performed on birds. In a review of protein polymorphism among several classes of animals presented by Nevo in 1978, only 10 species of birds were listed. Some years later Corbin (1983) listed 71 species, and this number is now much higher. Describing genetic variety and measuring it in quantitative terms are central to much of population genetics and systematics. Traditionally, such investigations have been carried out using physical characters: color of plumage or body dimensions. To reveal the hidden variety among heterozy- 1 Institute of Cytology and Genetics, Siberian Department of Russian Academy of Science, pr. Lavrentiev 10, Novosibirsk , Russia /95/ / Plenum Publishing Corporation

2 124 Kuznetsov gotes, special breeding experiments are needed. Performing such experiments in wild populations is very difficult. Many species of birds have no visible variations, which makes it difficult to study genetic variety and establish relationships between populations and subspecies. In the case of the two genera Anser and Branta, applying classical systematic characters (morphological, ethological, ecological, physiological) does not always provide a complete account of intergeneric and, especially, interspecific relationships. The Anser and Branta phylogeny is problematic. Hitherto, there has been no common opinion about generic accessories of such species as snow goose, bar-headed goose, emperor goose, swan goose, and red-breasted goose. The relationships of the black brant subspecies are unclear also (Stepanian, 1983). The same can be said about subspecies of bean goose, greylag goose, and snow goose. The study of systematics and phylogeny by physical characters in these two genera is complicated by widespread hybridization and the fertility of most hybrids (Johnsgard, 1960b; Sherer and Hilsberg, 1982). Therefore, additional problems appear in establishing relationships between species and genera (Panov, 1989). Use of biochemical characters decreases these difficulties, facilitates the study of genetic differentiation of wild populations, and allows the application of statistical methods to estimate such differentiation (Manwell and Baker, 1975; Lucotte and Kaminski, 1976a,b). The direct relationship between a protein and a gene allows the measurement of genetic variation among individuals, populations, and species in a relatively unbiased and unambiguous manner (Evans, 1989). Clear heritability, phenotypical codominance, and independence from environmental factors are advantages of biochemical characters (Witt, 1983). Blood plasma proteins are suitable for biochemical systematics, because the samples can be collected easily, considerable numbers of gene loci can be analyzed, the functions of many proteins can be revealed (Sibley, 1970), and birds stay alive after collection of samples. This is important in population studies requiring large samples. The present paper is devoted to describing polymorphic blood plasma proteins of the Anser and Branta genera. MATERIALS AND METHODS Blood samples from Anser and Branta species were obtained from ornithological collections of zoos in various cities, such as Askania-Nova (Ukraine), Yerevan (Armenia), Moscow, St. Petersburg (Russia), and Tallinn (Estonia) and from wild populations of Pacific black brant (Branta bernicla nigricans) and the lesser snow goose (Anser caerulescens caerulescens) on

3 Blood Plasma Proteins in Anser and Branta 125 Wrangel Island (Russia). Blood samples were collected from the following species: greylag goose (Anser anser), snow goose (.4. c. caerulescens), bean goose (A. fabalis), bar-headed goose (A. indicus), swan goose (A. cygnoides), emperor goose (A. canagicus), white-fronted goose (A. albifrons), lesser white-fronted goose (A. erythropus), Canada goose (Branta canadensis), and black brant (B. bernicla nigricans). The amounts of samples analyzed are given in Table I. The subspecies status of greylag goose, bean goose, and Canada goose is not registered. In most zoos the waterfowl are kept in common flocks, where hybridization takes place. I believe that some of my samples come from birds that could be hybrids of different subspecies. For this reason, I give the species name only. About 5 ml of blood was drawn from the underwing vein into a heparinized test tube. The samples were centrifuged at room temperature for 10 min at 600g. The plasma was separated and frozen in liquid nitrogen until used. Table I. Genetic Variability inanser and Branta Species (Standard Errors in Parentheses) Mean heterozygosity Mean No. Percentage Species Sample of alleles of loci Direct Hdy-Wbg of goose size per locus polymorphic a count expected a Snow (0.5) (0.031) (0.032) Greylag (0.5) (0.088) (0.101) Bar-headed (0.3) (0.032) (0.068) Swan (0.2) (0.035) (0.065) Emperor (0.3) (0.063) (0.107) White-fronted (0.2) (0.134) (0.103) Lesser white-fronted i000 (0.0) (0.000) (0.000) Bean (0.5) (0.088) (0.082) Brent (0.3) (0.017) (0.016) Canada (0.3) (0.057) (0.071) aa locus is considered polymorphic if the frequency of the most common allele does not exceed aunbiased estimate (see Nei, 1978).

4 126 Kuznetsov The electrophoretic procedure for plasma proteins was that of Gahne et al. (1977). We used gel and electrode buffers at ph 8.5. Polyacrylamide gel slides were 300 x 200 x 1 mm in size. The gradients of acrylamide concentrations in gels were 7-16 and 10-16%. Electrophoresis was carried out at a fixed voltage (250 V) and an initial current of 58 ma, within 6-14 hr. Undiluted plasma was mixed with sucrose (up to 40%), and 6-15 t~1 of the resulting solution was put into the gel pockets. For the electrophoresis of albumin, the plasma was diluted with deionized water (1:15). Identification of transferrins was made by means of the rivanol (6,9-diamino-2-ethoxyacridine lactate) test. The plasma was diluted with a saturated rivanol solution in 0.2 M Tris-HC1 buffer, ph 8.0, then it was shaken for 1-2 hr. After removal of precipitated proteins by centrifuging, the supernatant was refrigerated at 4 C for 3-5 hr. Then it was centrifuged again, mixed with sucrose (see above), and put into the gel. As a control, cattle plasma specimens of well-known transferrin genotypes were used. After electrophoresis, the gels were stained with a solution of 1% Coomassie brilliant blue G-250 (Serva) in ethanol-hcio4-h20 at 1.25:1: 26.5, and the background was destained in 7% acetic acid. For revealing haptoglobin phenotypes, gels were stained with a solution of benzidine in hot 0.5% acetic acid and 0.3% hydrogen peroxide. Staining was fixed by a thin amido black solution. Then the gels were soaked in glycerol solution and dried between two cellophane sheets. RESULTS A number of protein bands (30-35) were revealed on electropherograms. The classification of plasma proteins given here is based on analogy with plasma protein patterns of mammals and birds (Desaurer and Fox, 1964; Baker et al., 1966). If proteins were not identified, they were lettered based on their location with respect to albumin and transferrin (Fig. 1). The allelic designation was alphabetic. The more anodal allele is labeled "A" and others "B," "C," etc., corresponding to their anodal mobilities. Since my recent paper (Kuznetsov, 1991), I have found some more alleles in particular proteins. If their mobilities were close to the allele described, I denoted them by the same letter with a number index (1 being anodal to 2). I use the term "genotype" rather than "phenotype" because alleles are indeed inherited in a typical Mendelian manner. The inheritance data were obtained partially from the zoo samples and partially from my own breeding experiments on domestic geese. The results show that all blood plasma proteins described have a monomeric quaternary structure. In heterozygous individuals, there are no heteropolymeric bands of any protein studied. The alleles of blood plasma proteins analyzed are listed in Table II.

5 Blood Plasma Proteins in Anser and Branta 127 f2 El Fig. 1. Blood plasma protein patterns of geese species. Locations and designations of proteins are indicated on the right. The left side of the pherogram shows transferrin patterns of geese after rivanol treatment. (B) Pattern ofb. bernicla; (C) Pattern orb. canadensis. Prealbumin-1 (Pr-1) is the fastest protein in the blood plasma protein pattern. It is revealed on pherograms in three bands. No intraspecific polymorphism was found, but an interspecific difference was observed. Namely, in Canada goose Pr-1 migrates more slowly than in others (Fig. lc). These Canada geese are probably hybrids between the larger subspecies occidentalis and maxima, because of their zoo origin. The Canada goose Pr-1 variant can be considered species specific, if it is the same in the other subspecies. Prealbumin-2 (Pr-2), following Pr-1, is a well-identified protein. It can be seen as a single band on electropherograms. Six variants of this protein--a, B, B1, C, D, and E--were found. The variant E was found only in bean goose. The mobility of this variant coincides with that of the fast variant of Pr-3. The absence of staining in the Pr-2 zone in the pattern of this bird and the presence of the more brightly stained band in the Pr-3 zone can be explained by the presence of the slow variant Pr-2E. GenotypesAA, BB, B1B1, CC, DD, and EE and all possible combinations were found. It is C

6 128 Kuznetsov Table II. The Occurrence of Variants of Anser and Branta Genera Blood Plasma Proteins Species of goose a Allele Pr-2 A1 + A BI + B + C D E + + Pr-3 F M + + S + + Pr-4 A + + B + + C D Al F + + S Pa-3 A + + B~ + B B3 + C~ + G G + D + + E + X + Y K + L + Pa-X A B C D + G + H + Tf A + + B C + D E + K + L + M + Ptf-2 A B C al, Greylag goose; 2, snow goose; 3, bar-headed goose; 4, bean goose; 5, swan goose; 6, emperor goose; 7, white-fronted goose; 8, lesser white-fronted goose; 9, Canada goose; 10, brent goose.

7 Blood Plasma Proteins in Anser and Branta 129 concluded that the Pr-2 electrophoretic variants are under genetic control of a single autosomic locus with six codominant alleles. Prealbumin-3 (Pr-3) is revealed on the pherograms as a single band. Three electrophoretic variants of Pr-3 were found: fast (F), with the same mobility as Pr-2E, as mentioned above; slow (S); and middle (M), with a mobility intermediate between that of F and that of S. In the samples examined, birds with the genotypes FF, MM, SS, FM, FS, and MS were found. Prealbumin-4 (Pr-4) is the fourth protein of the prealbumin zone, according to its mobility. It can be seen as a single band on the pherograms. Four variants of this protein--a, B, C, and D--where found. Birds were found to have the genotypes AB, AD, AC, BC, CD, and BD, as well as AA, BB, CC, and DD. Thus, electrophoretic variants Pr-3 and Pr-4 are coded by single autosomic loci with the corresponding set of codominant alleles. Albumin (Al) is revealed on electropherograms by two bands--major and minor--in Anser species and by three bands---one major and two minor--in Branta species. Two electrophoretic albumin variants were found inanser, fast (F) and slow (S). The major band of the F variant has the same mobility as the minor band of S. The fast variant was found in heterozygotes FS only in greylag geese and swan geese. The S variant was found in allanser species. Thus, one can conclude that Anser albumin variants are controlled by two codominant alleles at a single autosomic locus, The major albumin band of all Branta species has a mobility intermediate between that of the F and that of the S major albumin bands of Anser (Fig. 1). No polymorphism of albumin in Branta was found. This is preliminary evidence that Branta albumin is genus specific, but the other species of the genus (B. leucopsis, B. rufibrenta, and B. sandvicensis) should be analyzed. A few protein zones, designated postalbumin, were distinguished between albumin and transferrin on a pherogram, Stained by Coomassie, postalbumin-1 (Pa-1) protein is partially covered by albumin, and it becomes clearly visible only after benzidine staining (Fig. 2). This suggested that Pa-1 has haptoglobin (Hp) properties. The locations of the bands are the same after both stainings. I found that Hp in all species studied appears in three bands and has four alleles: A, B, C, and D. No species- or genus-specific variants were found. The species of both genera differ only in their allele frequencies at the Hp locus. Postalbumin-2 (Pa-2) is polymorphic in Anser and monomorphic in Branta. I failed, however, to distinguish any variants of Pa-2, because of its complex pattern. Besides Hp, two more proteins can be seen in this zone. Postalbumin-3 (Pa-3) is the most polymorphic protein in the geese blood plasma protein pattern. Thirteen electrophoretic Pa-3 variants, revealed on pherograms in four bands, were found. In Anser I found nine

8 130 Kuznetsov Fig. 2. The haptoglobin patterns of geese. 1, HbD; 2, HbC; 3, HbAC; 4, HbB; 5, HbA. The arrowhead indicates albumins. variants--a, B1, B2, B3, C1, C2, C3, D, and E--which differ only in mobility, and X and Y, which differ from the rest both in mobility and in arrangement of bands (Fig. 3). In Branta, two variants are analogous toanser A, B, C, D, and E variants, but they have species-specific electrophoretic mobilities. In my samples, there are individuals with almost all homozygous and heterozygous genotypes. These data allow me to consider that Pa-3 electrophoretic variants are under genetic control by a single autosomic locus with a set of codominant alleles. In the postalbumin zone, I distinguished one more protein, appearing on the pherograms as a single diffuse spot under the conditions given (Fig. 4). This protein is unstable in developing. Its appearance was more consistent when the anodal and cathodal buffers were mixed. The protein has the same mobility as Pa-3 and was preliminarily labeled postalbumin X (Pa-X)..,..,,,,,,.,,,- m., ~,,L AA BIB I B2B 2 B3B 3 ClC I C2C 2 C~3 DD EE CIC 2 XX YY B~Y Fig. 3. Pa-3 alleles of geese of theanser genus.

9 Blood Plasma Proteins in Anser and Branta 131 Fig. 4. Pa-X patterns of geese plasma. (1) Transferrins; (2) albumins; (3) postalbumin X. Four variants in Anser and two in Branta were found. My previous report (Kuznetsov, 1991) described only two Pa-X variants in the Anser genus. Analysis of large samples from the Wrange! Island snow goose population has revealed two more electrophoretic variants, necessitating the change in lettering from F and S to A, B, C, and D. Pa-X genotypesa, B, C, D, AB,AC, BC, and BD were found in in the Anser genus. Pa-X variants are under genetic control and are coded by four codominant alleles at a single autosomic locus. In Branta, fast (G) and slow (H) variants were found, G in brent geese only and H in Canada geese. So G and H alleles may be species specific. There are a few more polymorphic proteins between the albumin and the transferrin zones, but I failed to distinguish the electrophoretic variants of these proteins. Transferrin (Tf) is revealed on pherograms in two bands in all species of the Anser genus and in Canada goose. In brent goose, transferrin is revealed as five bands--two major and three minor (Fig. 1B). According to my data, in Anser species there are five electrophoretic variants of transferrin: A, B, C, D, and E. The most common variant, TfB, was found in Canada goose too. The major bands of TfM of brent goose have the same mobility as

10 132 Kuznetsov TfB. InAnser species samples, there were birds with genotypesaa, AB, BB, AD, DD, CD, and BE. Canada geese examined have no polymorphism of transferrin. Among brent geese I found three transferrin variants, L, M, and N. TfM is the most common; L and N were found only in heterozygotes LM and MN. According to these data, electrophoretic variants of Anser and Branta transferrin are genetically controlled and coded by a single autosomic locus with a specific number of alleles. In the posttransferrin zone, I distinguished a few proteins, but only one of them, Posttransferrin-2 (Ptf-2), had electrophoretic variants. Revealed in two bands, three Ptf-2 variants--a, B, and C--were found in bothanser and Branta genera, Besides the birds withaa, BB, and CC genotypes, there were birds with AB, AC, and BC genotypes. I conclude that Ptf-2 is genetically controlled and coded by a single autosomic locus with three codominant alleles. DISCUSSION Through protein electrophoresis, quantitative estimates of the degree of genetic variety in wild populations are possible. To estimate correctly, we need unbiased samples of loci in the genome and disclosure of all alleles at each locus (Ajala and Kiger, 1984). In addition, the samples must be numerous, In this study, the first two points but not the last one hold. As a result, almost all proteins in blood plasma of Anser and Branta species studied are polymorphic (see Table II). The high degree of genetic variety in the loci studied can be explained by their ancient origin. Since then; the species have accumulated a lot of mutations detectable by electrophoretic methods, In some species--lesser white-fronted goose, whitefronted goose, emperor goose--most proteins were monomorphic, probably because of the small samples tested. According to recent data, the average heterozygosity (H) in birds equals 0.047, and the polymorphism (P), (Nevo et al., 1984). For comparison, the percentage of polymorphic loci in the species of geese analyzed in this study varied from 12.5 to 62.5 (0.95 criterion) and the mean heterozygosity (direct count) varied from to (Table I). I explain these data by the specificity of captive breeding, characterized by bottleneck effects, and the different origins of birds from native populations. For example, among 10 snow geese sampled at the Askania-Nova Zoo, TfE was found in four individuals, and among 200 geese from the Wrangel Island wild population, this allele was found only once. Again, Pa-3C2 was not detected in 30 snow geese sampled at different zoos but was seen at a high frequency in the wild population. As for the mean heterozygosity, 0.089, of Wrangel Island snow geese, I explain this by intensive gene flow between

11 Blood Plasma Proteins in Anser and Branta 133 different populations of this species maintained through males (Cooke et al., 1975; our data on Wrangel Island snow geese, in press). An insufficient number of samples for most species and the reasons mentioned above do not allow me to calculate reliable genetic distances between species and genera, but I hope these limitations can be overcome in the future. The present report is a description of polymorphic blood plasma proteins in geese, but these data can be used to estimate relationships in these genera. Differences were revealed in Pr-1, Pa-3, Pa-X, and Tf alleles between two Branta species. As for the Anser genus, only one species-specific allele was found, namely, Pa-3B3 in swan goose. Probably the rest of the Anser species can be differentiated from each other only by the frequencies of alleles of certain proteins. It is possible that some species will be marked by rare alleles. For instance, only snow goose has the very rare alleles TfE and Pa-3C1. TheAnser and Branta genera differ mainly in albumin, Pa-3, and Pa-X, which have electrophoretic mobilities typical only for Branta, and not foranser species. In spite of all the differences determined, many similarities can be seen in plasma protein patterns. Electrophoretic variants of the same mobilities of some proteins can be found in all or most species. For instance, Pr-1 is the same for allanser species and brent goose; Pr-2C, Pr-3F, Pr-4C, and A1S are similar for all Anser species; and TfB is common to all species of the Anser genus and to Canada goose. All these facts demonstrate the close relationship of the species and genera. Taking into account the high degree of genome homology of close species (Altuchov, 1989) and the identical electrophoretic mobility of some variants of homologous proteins in all or most species, one can consider that genes coding these proteins in different species have homologous origins. Comparing electrophoretic data should give us information about changes in these genes and elucidate systematical relationships between the studied taxa. According to zoological description, five genera of true geese--anser, Chen, Cygnopsis, Eulabeia, Philacte--and two genera of brants--branta and Rufibrenta--have been classified. Stepanian (1990) supports this classification. Delacour and Mayr (1945), Delacour (1954), Johnsgard (1960a,b), Kartashev (1974), and Sherer and Hilsberg (1982) revised modern Anseriform systematics and suggested unification of all geese species into the single genus Anser and all brant species into the genus Branta. The systematics of these two genera by Johnsgard (1978) is the most traditional version. Nevertheless, in the Field Guide to the Birds of North America (National Geographic Society, 1988), the genus name Chen is given for three species of geese: snow, Ross's, and emperor. The relationships within the family Anserini and, especially, within genera are still far from clear and are subject to constant alterations. Such a situation with the systematics of geese will

12 134 Kuznetsov force us to use more modern methods to resolve the problem, for example, DNA-DNA hybridization, RFLP, and electrophoretic analysis of proteins and enzymes. The latter is applicable to the intragenera systematics and to population genetics studies. Baker and Hanson (1966) reported results of analyses of four proteins of plasma in 11 Anser and Branta species. They did not find significant intergeneric and interspecific differences. This may be because they used starch gel electrophoresis with a low resolution. A considerable number of enzyme systems in a set of subspecies of Canada goose were studied by Van Wagner and Baker (1986). Several plasma proteins of Canada goose (Morgan et al., 1977) and brant (Novak et al., 1989) were described. Based on the analysis of isozymes of six loci, Cooke et al. (1988) came to the conclusion about former allopatry of blue and white morphs of lesser snow goose. ACKNOWLEDGMENTS During the course of this research, I have been assisted by many people, and I wish to record my gratitude to them. I thank collaborators at the Moscow, St. Petersburg, Yerevan, Tallinn, and Askania-Nova zoos and the Wrangel Island Native Reserve. I thank K. Litvin and S. Charitonov, from the Russian Ringing Center, and American and Canadian scientists J. Takekawa, S. Boid, and R. Kerbes for assistance in collecting blood samples. REFERENCES Ajala, F. J., and Kiger, J. A., Jr. (1984). Modem Genetics, 2nd ed., Benjamin/Cummings, Menlo Park, CA. Altuchov, Yu. P. (1989). Genetical Processes in Populations, Nauka, Moscow (Russia). Baker, C. M. A., and Hanson, H. C. (1966). Molecular genetics of avian proteins. VI. Evolutionary implications of blood proteins of eleven species of geese. Comp. Biochem. Physiol. 17:997. Baker, C. M. A., Manwell, C., Labisky, R. F., and Harper, J. A. (1966). Molecular genetics of avian proteins. V. Egg, blood and tissue proteins of the Ring-necked Pheasant, Phasianus colchicus L. Comp. Biochem. Physiol. 17:467. Cooke, F., Macinnes, C. D., and Prevett, J. P. (1975). Gene flow between breeding populations of lesser snow geese. Auk 93:493. Cooke, F., Parkin, D. T., and Rockwell, R. F. (1988). Evidence of former allopatry of the two color phases of lesser snow geese (Chen caerulescens). Auk 105:467. Corbin, K. W. (1983). Genetic structure and avian systematics. In Johnston, R. F. (ed.), Current Ornithology, Vol. 1, Plenum Press, New York, pp Delacour, J. (1954). The Waterfowl of the World, VoL 1, Country Life, London. Delacour, J., and Mayr, E. (1945). The family Anatidae. Wilson Bull. 57:3. Desaurer, H. C., and Fox, W. (1964). Electrophoresis in taxonomic studies illustrated by analysis of blood proteins. In Leone, C. A. (ed.), Tax~nornic Biochemistry and Serology, Ronald Press, New York, pp Evans, P. G. H. (1989). Electrophoretic variability of gene products. In Cooke, F., and Buckley,

13 Blood Plasma Proteins in Anser and Branta 135 P. A. (eds.), Avian Genetics: A Population and Ecological Approach, Academic Press, London, pp Gahne, Bo R., Juneja, K., and Grolmus, J. (1977). Horizontal polyacrylamide gradient gel electrophoresis for the simultaneous phenotyping of transferrin, post-transferrin, and post-albumin in the blood plasma of cattle. Anim. Blood. Grps. Biochem. Genet. 8:127. Johnsgard, P. A. (1960a). Comparative behavior of Anatidae and its evolutionary implications. Wildlife 11:31. Johnsgard, P. A. (1960b). Hybridization in the Anatidae and its taxonomic implications. Condor 63:25. Johnsgard, P. A. (1978). Ducks, Swans, and Geese of the World, University of Nebraska Press, Lincoln. Kartashev, N. N. (1974). The Systematics of Birds, Nauka, Moscow. Kuznetsov, S. B. (199l). The polymorphism of Anser and Branta blood plasma proteins. Siber. Biol. J. 5:3 (Russia). Lucotte, G., and Kaminski, M. (1976a). Polychromatisme et polymorphisme biochemique chez cing especes de phasianides. Biochem. Syst. Ecol. 4:69. Lucotte, G., and Kaminski, M. (1976b). Polymorphisme biochemique chez le faisan commun (Phasianus colchicus). Biochem. Syst. Ecol. 4:223. Manwell, C., and Baker, C. M. (1975). Molecular genetics of avian proteins XIII. Protein polymorphism in the three species of Australian passerines.austral J. Biol. Sci. 28:547. Morgan, R. P., Sulkin, S. T., and Hemmy, C. J. (1977). Serum proteins of Canada goose (Branta canadensis) subspecies. Condor 79:275. National Geographic Society (1988). Field Guide to the Birds of North America, 2nd ed. Nei, M. (1978). Estimation of average heterozygosity and genetic distance from a small number of individuals. Genetics 89:583. Nevo, E. (1978). Genetic variation in natural populations: Patterns and theory. Theoret. Pop. BioL 13:121. Nevo, E., Beilis, A., and Ben-Shlomo, R. (1984). The evolutionary significance of genetic diversity: Ecological, demographic and life history correlates. In Mani, G. S. (ed.), Evolutionary Dynamics of Genetic Diversity, Springer, New York, pp Novak, J. M, Smith, L. M., and Vangilder, L. D. (1989). Genetic variability within and among wintering populations of brant. J. Hered. 80(2):160. Panov, E. M. (1989). The Hybridization and Ethological Isolation of Birds, Nauka, Moscow (Russia). Sherer, S., and Hilsberg, T. (1982). Hibridisierung und verwandtschaftgrade Innerhalb der Anatidae--eine systematische und evolutionstheoretische Betrachtung. J. ornithol. 123: 257. Sibley, C. G. (1970). A comparative study of egg-white proteins of passerine birds. Peabody Mus. Nat. Hist. Bull. 32:1. Stepanian, L. S. (1983). Superspecies and Sibling-Species in the Aviafauna of USSR, Nauka, Moscow (Russia). Stepanian, L. S. (1990). The Conspectus of Ornithological Fauna of the USSR, Nauka, Moscow (Russia). Van Wagner, C. E., and Baker, A. J. (1986). Genetic differentiation in populations of Canada geese (Branta canadensis). Can. J. Zool. 64:940. Witt, G. S. (1983). Isozymes as probes and participants in developmental and evolutionary genetics. In Rattazzi, M. C., Scandalios, J. G., and Witt, G. S. (eds.), Isozymes: Current Topics in Biological and Medical Research, VoL 7, Alan R. Liss, New York, pp