Inter- and intrapopulation genetic variation in species ofPorphyra (Rhodophyta: Bangiales) from British Columbia and adjacent waters

Starch gel electrophoresis of 17 proteins has provided data on inter- and intrapopulation genetic variation in 20 species ofPorphyra occurring in British Columbia and adjacent areas.P. cuneiformis andP. nereocystis showed no within species variation, even over ranges of more than 1000 km. Populations ofP. abbottae, P. fallax, P. fucicola, P. gardneri andP. schizophylla were characterized by fixation for certain alleles. The number of polymorphic loci in a population ranged from zero to nine, depending on the species. Six species had populations that were polymorphic at just a single locus. Only two species (P. mumfordii andP. pseudolanceolata) had populations that were polymorphic at more than three loci. These levels of genetic variation are lower than those reported for populations of JapanesePorphyra species. Eleven taxa were polymorphic for 6-phosphogluconate dehydrogenase, the most variable enzyme. No within species polymorphisms were detected for bromoperoxidase, lactate dehydrogenase, superoxide dismutase or phycoerythrin. Possible evidence for the chimeric nature of the thallus was observed only inP. mumfordii.     

Genetic variation in two widespread species of salamanders,Taricha granulosa and Taricha torosa

Two species of the genus Taricha are widely distributed. T. granulosa ranges from southern Alaska to central California. T. torosa is comprised of two described subspecies, T. t. torosa, which occupies much of the coast ranges of California, and T. t. sierrae, which inhabits the western slopes of the Sierra Nevada Mountains. A starch gel electrophoretic survey for genetic variation at 34 loci in four population samples of T. granulosa and at 40 loci in five population samples of T. torosa reveals differences among these taxa both in amounts of intrapopulational variability and in patterns of geographic variation. Average observed heterozygosity is 9.6%±0.3% in T. granulosa, 3.3%±0.5% in T. t. torosa, and 7.2%±1.2% in T. t. sierrae. Average numbers of alleles per locus and proportions of polymorphic loci are also highest in T. granulosa, intermediate in T. t. sierrae, and lowest in T. t. torosa. Oregon and California granulosa are genetically nearly as different as the subspecies of torosa, but geographic variation is continuous in the former. T. torosa on the other hand is comprised of three distinct gene pools—T. t. sierrae and northern and southern races of T. t. torosa. Strikingly different amounts of intrapopulational genetic variation and patterns of geographic variation may be explained by steadystate species differences, but historical causes may also exist.This work was supported by AEC Research Contract AT(04-3)34 and NSF Grant GB-42246 to F. J. Ayala and by an NIH predoctoral traineeship administered by the Department of Genetics, U.C. Davis.     

Protein Polymorphisms, Segregation in Genetic Crosses and Genetic Distances among Fishes of the Genus Xiphophorus (Poeciliidae)

The products of 49 protein-coding loci were examined by starch gel electrophoresis for populational variation in six species of Xiphophorus fishes and/or segregation in intra- and interspecific backcross and intercross hybrids. Electrophoretic variation was observed for 29 of the 35 locus products in a survey of 42 population samples. The highest frequency of polymorphic loci observed in noninbred populations was 0.143. After ten or more generations of inbreeding, all loci studied were monomorphic. Inbred strains generally exhibited the commonest electrophoretic alleles of the population from which they were derived. An assessment of genetic distances among Xiphophorus populations reflected classical systematic relationships and suggested incipient subspeciation between X. maculatus from different drainages as well as several species groups. Thirty-three loci were analyzed with respect to segregation in hybrids. The goodness of fit of segregations to Mendelian expectations at all loci analyzed (except loci in linkage group I) is interpreted as evidence for high genetic compatibility of the genomes of Xiphophorus species. It is anticipated that these data will result in a rapid expansion of the assignment of protein-coding loci to linkage groups in these lower vertebrate species.     

Island isolation and habitat heterogeneity correlate with DNA variation in a marine snail (Littorina saxatilis)

The null assumption of molecular variation is that most of it is neutral to natural selection. This is in contrast to variation in morphological traits that we generally assume is maintained by selection, and therefore often by selection coupled to environmental heterogeneity in time and space. Examples of molecular variation that vary over habitat-shifts, particularly in allozymes, show that the relative impact of non-neutral variation as compared to neutral variation might be substantial in some systems. To assess the importance of habitat-generated variation in relation to variation generated by random processes in nuclear DNA markers at small spatial scales, we compared the effects of island isolation and habitat heterogeneity on genetic substructuring in a rocky shore snail ( Littorina saxatilis ). This species has a restricted migration among islands owing to the lack of free-floating larvae. Earlier studies show that allozymes vary extensively as a consequence of isolation by water barriers among islands, but also as a consequence of divergent selection among different microhabitats within islands. In the DNA markers we observed genetic differentiation owing to island isolation at three of nine loci. In addition, variation at three loci correlated with habitat type, but the correlation for two of the loci was weak. Overall, isolation contributed slightly more to the genetic variation among populations than did habitat-related factors but the difference was small. It is concluded that both island isolation, which interrupts gene flow, and a heterogeneous habitat cause genetic substructuring at the DNA level in L. saxatilis in the studied area, and thus in this species we need to be somewhat concerned about habitat heterogeneity also at DNA loci.  © 2004 The Linnean Society of London, Biological Journal of the Linnean Society , 2004, 82 , 377–384.     

Enzyme polymorphism in the freshwater mussel species, Aspatharia pfeifferiana (Mollusca, Lamellibranchiata) from southern Africa

Gene products of 31 protein coding loci in Aspatharia pfeifferiana examined by horizontal starch gel electrophoresis revealed genetic variation at 10 (32.26%) of the loci studied. The mean number of alleles per locus was 1±39 (±0±11), and the average heterozygosity value was 0±074 (±0±026). Not only is this the first account of electrophoretic variants of this species, but also the results show that the species has less genetic variation than most other molluscs. Genetic variation compares favourably with values obtained for other invertebrate species and for fish from the same geographical region. A. pfeifferiana has sufficient variation to adapt to environmental changes or to be used in selection programmes. The results are useful for conservation stocking and it will undoubtedly aid in improving the systematics and phylogeny of freshwater clams in general.     

A cline of allozyme variation inAbies mariesii

Genetic variation at 22 allozyme loci was examined for 1,003 trees from 11 isolated natural populations ofAbies mariesii covering all except the southernmost region of its geographic range. Genetic diversity within species (H _es=0.063) was low compared to many other long-lived woody species. Most of the genetic variation is found within populations (G _ST=0.144) despite their isolated distribution. Genetic distance between populations was positively correlated with geographic distance. Genetic diversity within populations was generally low (meanH _ep=0.054), but varied across populations in a clinal fashion such that genetic variation decreased with increasing latitude. These genetic characteristics may reflect the distribution history of this species.     

Allozyme differentiation of populations of the dogwhelk Nucella lapillus, (L.): the relative effects of geographic distance and variation in chromosome number

Geographic patterns of genie differentiation were compared with differentiation between karyotypes in the intertidal snail Nucella lapillus. Samples from 24 sites covering the species range in Europe and North America were analysed for allozyme variation at 16 soluble enzyme loci. Two homokaryotypes have been identified with diploid numbers 2n = 26 and 2n= 36 (variation is Robertsonian and hybrids have intermediate chromosome numbers) and samples were classified (on the basis of published data) according to karyotype. Group 1 consisted of samples from three English Channel populations of higher chromosome number (on average 2n > 32) and Group 2 consisted of the remaining 21 samples (presumed to be 2n= 26). Karyotype variation accounts for roughly the same amount of the absolute allozyme variance as geographic variation (46.3 °, and 53.7°, respectively). Yet the patterns of differentiation seen between karyotypes and with geographic separation are very different. In samples classified as 2n= 26 (Group 2), while there is a significant amount of heterogeneity (F_ST per locus averaged 0.128 for 10 polymorphic loci), allozyme variation occurs independently at different loci so mean genetic identity (Nei) is high: 0.972. There is only a slight decline in genetic identity with distance (genetic identity averaged 0.965 for amphi-atlantic comparisons) indicating that passive transport of juveniles or adults may contribute significantly to gene flow. Conversely, allozyme variation between karyotypes was concordant. High chromosome number populations possessed a suite of alleles at four allozyme loci (Esl-3, Lap-2, Mdh-1 and Pep-2) which were absent or rare in Group 2 samples resulting in high F_ST values for these loci (from 0.294 to 0.472) when karyotypic classes were combined. Consequently the mean genetic identity between these Robertsonian races is low, 0.856, and falls within the range more usually associated with congeneric comparisons than with con-specific comparisons. The mechanisms maintaining this genie difference are unclear. However the distribution of the karyotypes and physiological and morphological differences (in shell shape) between them strongly suggest that karyotypic variation in Nucella is adaptive.     

Isozyme Variation in a Tropical Pioneer Tree Species (Cecropia obtusifolia,Moraceae) with High Contents of Secondary Compounds1

We define 48 allozyme loci for a tropical pioneer tree species, Cecropia obtusifolia Bertol, which has high contents of secondary compounds. Our goals were to find the effects of extraction procedures on artifacts and variation in resolution of enzyme banding patterns; to explore the relationship among the variation of the loci sampled and the enzymes' molecular structure, metabolic function and substrate; to obtain estimates of the genetic variation in this species at Los Tuxtlas rain forest (México) and to explore the variation of allelic frequencies in six successive life-history stages of the species. The resolution of the isozymes bands and the actual banding pattern varied with the type and age of tissue, the collection and storage procedures, the extraction buffer, and other loading and running procedural details. However, artifactual variation was eliminated with a new extraction buffer for species with high contents of secondary compounds. Of the 26 enzymes resolved for C. obtusifolia, we found that enzymes with a greater number of substrates and an oligomeric quaternary structure tended to be more variable than their counterparts, but the relationship was not statistically significant. The proportion of polymorphic systems varied significantly with the metabolic pathway and the function of the enzymes. Enzymes involved in starch synthesis are significantly more variable (p < 0.05) than all others, except those involved in amino acids metabolism and the proportion of polymorphic enzymes is also significantly associated with the hnction of the enzyme, the hydrolases and isomerases are significantly more variable than lyases and oxidoreductases enzymes. The percentage of polymorphic loci for C. obtusifolia was estimated at P = 27.1%. The effective number of alleles was estimated at ne = 1.3 and ne = 2.4 for all loci and only polymorphic ones respectively and the average heterozygosity (H) for all 48 loci was estimated at H = 0.05. Allele frequencies varied throughout the life-cycle of the species, with significant differences for some alleles and loci among some life-cycle stages. “Tree seeds” allele frequencies differ significantly (P < 0.05) from “rainy dispersed seeds” in 7 of 8 loci and from “soil seeds” in Six of eight loci. Allele frequencies of all three seed categories (“tree seeds”, “rainy dispersed seeds”, and “soil seeds”) differed strongly from established individuals (seedlings, juveniles and adults), while allele frequencies of established individuals are relatively similar to one another. Seedling allele frequencies at most loci were also significantly different from those found in seeds collected from trees, seed-rain, and soil. Two alleles (at GOT-2 and FE-2) were only found in soil seeds and one allele (at LAP-2) was only found in seedlings.     

Allozyme variation and evolutionary relationships of grain amaranths (Amaranthus spp.)

Summary Allozyme studies in amaranth provided useful assays of genetic variation in order to verify the patterns inferred from morphological traits, for elucidating the genetic structure of landraces, and for the studies of evolutionary relationships among wild, weedy and crop species. Thirty-four populations of cultivated New World amaranths were surveyed along with 21 weedy New World populations for allozyme variation at nine electrophoretic enzyme loci. Eleven populations of cultivated amaranths from the Indian State of Uttar Pradesh and six from Nepal were also surveyed for a comparison. In the New World populations, heterozygosity was low, and different populations ranged from 0 to 44% polymorphic loci. Adjacent populations were often fixed for different alleles or had very different allele frequencies at certain loci, with no apparent geographical patterns. Diversity index H was partitioned into the intra- and interpopulation as well as the interspecific components of variability. The crop versus weed genetic distances were the largest, whereas the intra- and interpopulation components of H were about equal. Genetic structure of all three species of the New World amaranths together can be described as a collection of distinct populations, each more or less a heterogeneous collection of highly homozygous individuals. The North Indian populations showed relatively less allozyme variability with the most common alleles same as those of Mexican landraces. Alleles at several loci proved to be diagnostic of the crop and weed groups, and of the three individual crop species. Genetic distances based on pooled gene frequencies showed the three crop species to be generally more closely related inter se than they were to their putative weedy progenitor species, respectively (with the exception of the weed-crop pair A. quitensis and A. caudatus). This implies a single domestication event involving A. hybridus as the common ancestor rather than three separate domestication events. Close similarity between A. caudatus and A. quitensis might have resulted from transdomestication based on a weedy or semi-domesticated species having migrated from Meso-America to South America. This preliminary report must now be expanded by further ecogeographical, cytogenetic and population studies on new extensive collections from the areas of early domestication. Some evidence of recent introgression and/or segregation of crop-weed hybrids between A. caudatus and A. retroflexus is available in the form of rare individuals in crop populations with crop allozyme genotypes except for a single homozygous weedy allele.     

Genetic variation in Simocephalus (Anomopoda: Daphniidae) in North America: patterns and consequences

The presumption of intercontinental distributions of many species of Anomopoda has been reinforced by their considerable morphological uniformity and ease of passive dispersal via ephippial eggs. To test the validity of this dogma, genetic variation among taxa in the cladoceran genus Simocephalus was examined on a continental scale. Genetic variability (percentage polymorphic loci = 8.8%, individual heterozygosity = 3.4%) was comparable to that determined on a local scale in Simocephalus and somewhat less than for other zooplankton groups. Four species complexes were distinguished allozymically with unique allelic substitutions found for at least two loci between these taxa. Eight species in North America were differentiated within these complexes (Nei s genetic distance, D>0.30), all of which were clearly separated from two European species. Both S. cf. vetulus and S. cf. serrulatus are broadly distributed in North America, and intraspecific divergence is relatively low. In contrast, S. cf. congener and S. cf. exspinosus represent species complexes, both consisting of several species, differentiated allozymically and morphologically. The concept of generalist species with moderate genetic variation throughout a broad range and specialist species, genetically depauperate, with restricted distributions, is explored.     

Structure of Genetic Variation within and between Populations of Mycophagous Drosophila

Patterns of genetic variation within and between populations of five species of mycophagous Drosophila were examined by gel electrophoresis of several polymorphic loci. Populations of the five species could not be shown to be subdivided into sympatric host-adapted races. Statistically significant, but small, between-host differences in gene frequencies were observed at three of 15 loci. Mean gene frequencies at all loci were similar in New York and Tennessee, and, with one exception, relatively little genetic differentiation was observed among study sites within those two regions. Gene frequencies generally were stable over several years of collecting as well. The unpredictable nature of the fungal hosts may preclude the site fidelity and continuity of diversifying selection necessary for adaptive divergence of populations.     

Genetic structure of Glycine canescens,a perennial relative of soybean

Summary Allozyme variation as detected by starch gel electrophoresis was used to assess the extent and spatial organization of genetic variation across the entire range of Glycine canescens sensu lato. Eleven enzyme systems were assayed in 116 accessions of this taxon and 102 alleles were detected at a total of 31 loci. Eighty-one percent of loci were polymorphic. Most of this variation occurred between and very little within accessions. Three major groupings were detected. These groupings (groups 1, 2, and 3) also differed with respect to mean seed size and their geographic distribution. A further ten accessions stood out from these distinct groups. These accessions were most closely related to group 3 but were variable among themselves. In general, they were collected from highly dissected terrain, often in the remote interior of the continent. A final group of 18 problematic accessions (group X), originally tentatively identified as G. canescens on morphological grounds, was shown to be isozymically distinct from this species and was reclassified as one form of the polytypic species G. clandestina.     

A genomic evaluation of taxonomic trends through time in coast horned lizards (genus Phrynosoma)

Determining the boundaries between species and deciding when to describe new species are challenging practices that are particularly difficult in groups with high levels of geographic variation. The coast horned lizards (Phrynosoma blainvillii, Phrynosoma cerroense and P. coronatum) have an extensive geographic distribution spanning many distinctive ecological regions ranging from northern California to the Cape Region of Baja California, Mexico, and populations differ substantially with respect to external morphology across much of this range. The number of taxa recognized in the group has been reevaluated by herpetologists over 20 times during the last 180 years, and typically without the aid of explicit species delimitation methods, resulting in a turbulent taxonomy containing anywhere from one to seven taxa. In this study, we evaluate taxonomic trends through time by ranking 15 of these species delimitation models (SDMs) using coalescent analyses of nuclear loci and SNPs in a Bayesian model comparison framework. Species delimitation models containing more species were generally favoured by Bayesian model selection; however, several three‐species models outperformed some four‐ and five‐species SDMs, and the top‐ranked model, which contained five species, outperformed all SDMs containing six species. Model performance peaked in the 1950s based on marginal likelihoods estimated from nuclear loci and SNPs. Not surprisingly, SDMs based on genetic data outperformed morphological taxonomies when using genetic data alone to evaluate models. The de novo estimation of population structure favours a three‐population model that matches the currently recognized integrative taxonomy containing three species. We discuss why Bayesian model selection might favour models containing more species, and why recognizing more than three species might be warranted.     

Spectral Diversity of Fluorescent Proteins from the Anthozoan <Emphasis Type="Italic">Corynactis californica</Emphasis>

Color morphs of the temperate, nonsymbiotic corallimorpharian Corynactis californica show variation in pigment pattern and coloring. We collected seven distinct color morphs of C. californica from subtidal locations in Monterey Bay, California, and found that tissue– and color–morph-specific expression of at least six different genes is responsible for this variation. Each morph contains at least three to four distinct genetic loci that code for these colors, and one morph contains at least five loci. These genes encode a subfamily of new GFP-like proteins, which fluoresce across the visible spectrum from green to red, while sharing between 75% to 89% pairwise amino-acid identity. Biophysical characterization reveals interesting spectral properties, including a bright yellow protein, an orange protein, and a red protein exhibiting a “fluorescent timer” phenotype. Phylogenetic analysis indicates that the FP genes from this species evolved together but that diversification of anthozoan fluorescent proteins has taken place outside of phylogenetic constraints, especially within the Corallimorpharia. The discovery of more examples of fluorescent proteins in a non-bioluminescent, nonsymbiotic anthozoan highlights possibilities of adaptive ecological significance unrelated to light regulation for algal symbionts. The patterns and colors of fluorescent proteins in C. californica and similar species may hold meaning for organisms that possess the visual pigments to distinguish them. Christine E. Schnitzler and Robert J. Keenan contributed equally to this work. Data deposition footnote: The GenBank () accession numbers for the genes and gene products discussed in this paper are: ccalRFP1 (AY823226); ccalYFP1 (AY823227); ccalRFP2 (DQ065851); ccalGFP1 (DQ065852); ccalOFP1 (DQ065853); ccalGFP3 (DQ899732)     

Isoenzyme Variation in Asian Beans

Electrophoretic variation of isoenzyme characters encoded by 20 gene loci has been studied in 9 species of Asian beans, comprising subgenus Ceratotropis of the genus Vigna. Four species of hypogeal germination type, V. angularis, V. umbellata, V. dalzelliana and V. minima, are shown to form a distinct monophyletic group in the subgenus, sharing common electromorphs of two major isoenzymes of aspartate aminotransferase. V. glabrescens is confirmed to be an allotetraploid derivate of two diploids closely related to V. radiata and V. umbellata, but divergent from their contemporary accessions analyzed. V. trilobata is the most variable species of the subgenus in isoenzyme characters. It reveals intraspecific differentiation at several isoenzyme loci correlated with the seed morphology and geographical localization.     

Capturing Genetic Variation during Ecological Restorations: An Example from Kankakee Sands in Indiana

Genetic variation in populations, both natural and restored, is usually considered crucial for response to short‐term environmental stresses and for long‐term evolutionary change. To have the best chance of successful long‐term survival, restored populations should reflect the extant variation found in remnants, but restored sites may suffer from genetic bottlenecks as a result of founder effects. Kankakee Sands is a large‐scale restoration being conducted by The Nature Conservancy (TNC) in northwestern Indiana. Our goal was to test for loss of genetic variation in restored plant populations by comparing them with TNC’s seed source nursery and with local remnant populations that were the source of nursery seed and of the first few restored sites. Allozyme analysis of Baptisia leucantha, Asclepias incarnata, Coreopsis tripteris, and Zizia aurea showed low levels of allozyme diversity within all species and reductions in polymorphism, alleles per locus, and expected heterozygosity between remnants and restorations for all species except A. incarnata. Almost all lost alleles were rare; restored populations contained almost 90% of alleles at polymorphic loci that occurred in remnants at frequencies greater than 1%. Allele frequencies for most loci did not differ between remnants and restored sites. Most species showed significant allele frequency differentiation among remnant populations and among restored sites. Our results indicate that seed collection techniques used at Kankakee Sands captured the great majority of allozyme variation present in seed source remnant populations.     

Geographic distribution and multilocus organization of isozyme variation of rice (Oryza sativa L.)

Genetic organization of isozyme variation in rice (Oryza sativa L.) was investigated based on 17 polymorphic isozyme loci using a sample of 511 accessions of worldwide origin. The genetic diversity within the species was very high (H=0.36 with 4.82 alleles per locus), as compared with most selfing plant species. Three diversity centers were detected for isozyme variation including South Asia, China and Southeast Asia. The accessions were classified into three well-differentiated cultivar groups corresponding to the indica and japonica subspecies, and a new unnamed group. Variation within the cultivar groups accounted for 80% of the total isozyme variation. Within-country variation accounted for 58% of the total variation while among-region and among-country variation within the cultivar groups accounted for only 14% and 8% of the total variation. Analyses using log-linear models revealed that pronounced non-random associations between and among alleles at many unlinked isozyme loci were organized in a non-hierarchical pattern, and subspecific and macro-geographic differentiation was much more pronounced in multilocus phenotype frequencies than in allelic frequencies at individual loci. These results suggest that selection on multilocus gene complexes was largely responsible for the maintenance of the extensive isozyme variation within the species and the indica-japonica differentiation. Our results further suggest the independent domestication of indica and japonica, the dual origins of the indica rice from China and South Asia (India), and the differentiation of the ecotypes ’javanica’ and the ’temperate japonica’ within the japonica subspecies. Received: 5 August 1999 / Accepted: 13 December 1999     

Genetic diversity and population structure of<Emphasis Type="Italic">Kalopanax pictus</Emphasis> (Araliaceae)

Kalopanax pictus (Araliaceae) is a long-lived woody species primarily distributed throughout East Asia. This species is regarded as medically and ecologically important in Korea. We used starch gel electrophoresis to investigate its allozyme variation and genetic structure in samples from Korean populations (both natural and cultivated). Our analysis of 10 enzymes revealed 18 loci, of which 10 were polymorphic (35.6%). Genetic diversity at the species and population levels were 0.200 and 0.149, respectively, with the mean for cultivated populations (0.124) being lower than for natural (wild) populations (0.181). Asexual and sexual reproduction modes, perennial habitat, and longevity all were possible factors contributing to such high diversity. An indirect estimate of the number of migrants per generation (N _m = 1.08) indicated that gene flow was not extensive among these Korean populations. Therefore, we suggest that geographical distance as well as reproductive isolation between wild and cultivated plants may play roles in shaping the population structure of this species.     

Ten polymorphic microsatellite loci for the endangered Buena Vista Lake shrew (Sorex ornatus relictus)

The ornate shrew (Sorex ornatus) is restricted to the vanishing wetlands of California, USA and Baja California, Mexico. Several subspecies of ornate shrews are considered ‘mammal species of special concern’ in California by the Department of Fish and Game, and one (Sorex ornatus relictus) has recently been listed as endangered. Populations of shrews around Buena Vista Lake have been diminished or extirpated due to habitat deterioration and human development. In order to study the patterns of genetic variation in isolated populations of Buena Vista Lake shrews, we developed 10 polymorphic microsatellite loci. There were 6–27 alleles per locus, and the loci had heterozygosity values that ranged from 20 to 80%. In addition, we screened 20 different populations of S. ornatus, eight species within two subfamilies of shrews (Soricinae and Crocidurinae), as well as in a mole (Talpidae, Neurotrichus gibbsii), to determine if these loci could be informative in other species as well.     

Aspects of the population genetics of Mytilus (Mytilidae; Mollusca) in the British Isles

Starch gel electrophoresis was used to study variation at 11 loci in mussels sampled mainly from British coastal sites. Two types of mussel were identified, Mytilus edulis, the common mussel and its southern relative Mytilus galloprovincialis. Several partially diagnostic loci were used to map the distribution of the two forms. Mytilus edulis was present at all sites sampled in Britain and Ireland but was at low frequency in SW England; M. galloprovincialis was detected in SW England, the south and west of Ireland. Scotland and NE England, but was absent from south Wales, the Irish sea coasts of Wales and Ireland, and SE England. Apart from the occurrence of M. galloprovincialis in NE England, this distribution conforms with the results of studies using morphological characters and parallels the distribution of many other southern species in Britain. At the microgeographical level, M. edulis was found to prefer more sheltered and estuarine conditions than M. galloprovincialis. Analysis using the best diagnostic loci showed that hybridization is occurring between M. edulis and M. galloprovincialis at all localities where they occur sympatrically but that the extent of hybridization varies considerably between localities. The distribution of localities having high proportions of hybrid individuals is best interpreted by assuming that hybrids have higher fitness than parental types at these localities. A study was made of variation within and between those localities where only M. edulis individuals were observed. Little significant geographic variation in allele frequency was detected, but significant deficits of heterozygotes compared with Hardy-Weinberg expectations were seen for most loci. Analysis suggests that the Wahlund effect is not involved and that the most likely cause of the deficit is low frequencies of null alleles. In M. edulis no differences in phenotypic variance in shell height and width were observed between samples of multiply heterozygous and multiply homozygous individuals and no genetic differences were found between juveniles and adults. Overall little evidence was found that balancing selection is responsible for maintenance of the polymorphisms studied in M. edulis. The pattern of geographic variation in gene frequencies in Mytilus in the British Isles is discussed in relation to variation in the south and north of Europe and North America. It is concluded that steep clines in gene frequencies in M. edulis observed by other workers in the Baltic and in Long Island Sound cannot be attributed to the presence of M. galloprovincialis.