Early studies on genetic differentiation in west-coast bay mussels using electrophoretic analyses of allozyme loci indicate 2 genetically distinct groups.  One type, Mytilus trossulus, is common from Alaska to mid-California, while another, M. galloprovincialis, occurs in southern California.  Both species and their hybrids co-occur in central and northern California.  The southern species, M. galloprovincialis may have been introduced from the Mediterranean.  The authors note that despite the prevalence of references to Mytilus edulis in the scientific literature for this coast, this species of bay mussel is not, nor probably ever has been, present on this coast.  McDonald & Koehn 1988 Mar Biol 99: 111; see also Sarver & Loudenslager 1991 Biochem Syst Ecol 19: 183 for a corroborative study on the same subject; for a more recent review of distribution of Mediterranean bay mussels M. galloprovincialis see Wonham 2004 J Shellf Res 23: 535.

NOTE  one of several forms of an enzyme coded for by different alleles at a locus

A later study on genetic structure of 27 populations of bay mussels Mytilus spp. sited from northern to southern California confirms the presence of M. trossulus in the north and M. galloprovincialis in the south, with mixed degrees of hybrids at sites from Bodega Bay to Monterey Bay (see figure).  The authors also analysed mussels from 12 of these populations for extent of variation in 7 shell characters, including length, height, width of shell, and dimensions of 3 muscle attachment scars, but the separation resulting from this is not as successful.  Measurements at each location of temperature, salinity, tidal height, and degree of wave exposure show that M. trossulus tends to favour locations with lower temperatures and greater salinity variation than M. galloprovincialis.  Effects of tidal height and wave exposure, in contrast, do not significantly affect the macrogeographic distributions of the 2 species.  Sarver & Foltz 1993 Mar Biol 117: 105.

NOTE  examined at 15 allozyme loci using starch-gel electrophoresis

NOTE  an additional population is sampled at Juneau, Alaska

Another study using frequencies of mitochondrial haplotypes confirms the presence of only Mytilus trossulus at sites north of San Francisco Bay, California, but mixed presence of this species and M. galloprovincialis south of there. This is contrary to Research Study 1 above that indicates only the presence at southern sites of M. galloprovincialis. The authors also identify mussels arriving at Coos Bay, Oregon in seawater ballast in ocean-going ships as being M. galloprovincialis.  Thus, despite evidence of continual transport and release, this species has failed to invade the northwestern coast of North America.  The authors caution, however, that invasion success is highly probabilistic, and that future colonisation failure is not a certainty.  Geller et al. 1994 Mar Biol 119: 243.

NOTE  the authors discuss several possible explanationsfor their identification of M. trossulus haplotypes in the southern region, including interbreeding and introduction of the species in ballast water.  In Europe, a similar situation exists, with M. edulis living in the north and M. galloprovincialis in the south, and hybridisation in the overlap area

NOTE  bivalve larvae collected in ballast water are cultured to juvenile stage, then analysed.  The authors note that bivalve larvae, mostly Mytilus spp., were present in ballast water of 70% of 160 ships arriving at Coos Bay from Japan between 1986-1990

Another method for discriminating between Mytilus edulis, M. galloprovincialis, and M. trossulus has been developed by researchers in British Columbia.  This method uses 2 polymerase chain-reaction (PCR) markers that are inherited in Mendelian fashion, show no evidence of close linkage, and have high diagnostic reliability. Apparently, the 2 non-native Atlantic (M. edulis) and Mediterranean (M. galloprovincialis) species have been imported to British Columbia in commercial culture programmes and hybridisation is suspected.  Analysis of genotypes of 482 mussels collected at 12 sites around Vancouver Island and the nearby mainland reveal at least one alien allele at 9 of 12 sites, with an overall frequency of 6%.  This, combined with evidence of extensive hybridisation between native and alien species, suggests that the introduced alleles will persist in the population for some time.  Heath et al. 1995 Can J Fish Aquat Sci 52: 2621.

NOTE  the authors note that use of these PCR-based markers avoids certain interpretive complications inherent in methods based on mitrochondrial DNA (mtDNA) polymorphisms

A later study using diagnostic length differences in a PCR amplified fragment of the gene for byssal adhesive protein confirms the distribution of these 2 species.  Mytilus trossulus is the only bay mussel found in Alaska and is the most abundant species occurring along the coasts of Washington and Oregon (sites in British Columbia are inexplicably omitted from the study).  Mytilus galloprovincialis is the most abundant bay mussel from Tomales Bay, California (north of San Francisco) to San Diego, California.  The authors find evidence of extensive hybridisation in several locations along the coast, most notably at Whidbey Island, Washington, and San Francisco Bay and San Diego Bay, California.  Additional studies in Japan show that M. trossulus is found only in the northern island of Hokkaido, while M. galloprovincialis is the only species found in the southern islands of Honshu and Kyuushu.  The discontinuity between the zoogeographic distributions of the 2 species at about 40^o N latitude on both sides of the Pacific Ocean suggest to the authors that temperature is a factor in determining present distributions. Suchanek et al.1997 Biol Bull 193: 187.

NOTE  near the Golden Gate Bridge, for example, the authors report 34% M. galloprovincialis, 11% M. trossulus, and 55% hybrids

NOTE  only one site is sampled in Hokkaido, while 4 sites are sampled on each of the other Japanese islands

A fascinating glimpse into the history of distributions and competitive takeover in mussel populations in mid and southern California is presented in a genetic study of dried mussel samples from museum collections made in the latter part of 19th and early part of the 20th Centuries.  The genetic origin of the samples is determined from analyses of partial mitochondrial 16S rRNA gene sequences, known to separate out the 3 bay-mussel species Mytilus edulis, M. trossulus, and M. galloprovincialis.  All the samples turn out to be M. trossulus.  Thus, M. trossulus appears to have been the dominant species from Monterey to southern California prior to the invasion of M. galloprovincialis.  Following its introduction in the mid-1900s to California, M. galloprovincialis appears to have completely displaced the indigenous species M. trossulus.  The author amusingly remarks, that owing to the morphological similarity of the two species, the displacement occurred without anyone noticing.  Geller 1999 Conserv Biol 13: 661.

NOTE  the tissues are from bay mussels collected from locations around California (1871 and 1884: Monterey Bay; 1900: Santa Catalina Island) and held in collections at the Los Angeles County Museum of Natural History and the U.S. National Museum of Natural History.  In those days the curators undoubtedly labeled all their specimens as Mytilus edulis

Consequences of hybridization may include: 1) formation of a new species, 2) an ability to colonise new and different habitats, and 3) a reinforcement of reproductive barriers through selection against hybrid individuals. The extent of hybridisation of the 2 species Mytilus trossulus and M. galloprovincialis along the west coast is investigated in this study and discussed relative to the above ideas.  The authors sample mussels from 20 sites in California extending from Crescent City in the north to San Simeon in the south (including 6 sites in San Francisco Bay, not shown here), and analyse for a series of nuclear and mitochondrial markers known from previous work (see Research Study 4 above) to be diagnostic for the 2 species and their hybrids, as well as providing an estimate of extent of introgression. 

Results show that the 2 species are sympatric from Cape Mendocino to the Monterey Peninsula, with hybrids being detected throughout this zone. Note in the illustration that the genotype frequencies line up with the locations on the map. The 2 species are colour-coded, but the hybrids are shown as a faint dotted line. The authors provide data from several locations in the San Francisco Bay region, but these data are not reproduced here.

Introgression at the level of first-generation backcross is identified throughout the hybrid zone, but there is no indication of advanced introgression beyond first-generation.  This suggests that gene flow between the 2 species has been quite limited.  Abrupt declines in frequency of mussels with M. trossulus and hybrid genotypes south of the Monterey Peninsula, and of M. galloprovincialis and hybrid genotypes north of Cape Mendocino indicate that the hybrid zone is positioned between these prominent coastal features with little introgression of alleles into south- and north-bordering parental populations. Rawson et al. 1999 Mar Biol 134: 201; see also Rawson et al. 1996 Heredity 77: 599 for further information on genetic markers used to distinguish Mytilus spp.

NOTE  the magnitude of gene flow from a hybrid into its parent species as a result of backcrossing

NOTE  the hybrid zone for these two mussel species has variable borders depending upon which research paper you are reading (e.g., from Puget Sound to San Diego), perhaps a consequence of introductions via aquaculture or ship-ballast discharge

Although hybrids of the mussels Mytilus trossulus and M. galloprovincialis are rare in Puget Sound, they can be readily hybridised and grown in the laboratory.  Researchers from Oregon State University and University of Washington do reciprocal crosses and compare growth performances of the resulting hybrids with parental crosses under 2 different salinity conditions.  Hybrids show no particular evidence of hybrid vigour, but both reciprocal crosses and the parental cross M. trossulus grow significantly faster than the other parental cross M. galloprovincialis at low salinity (20ppt).  This is expected because M. trossulus grows well in regions of low and variable salinities, as in Puget Sound, while M. galloprovincialis does better in areas of high salinity (e.g., 32ppt).  The overall success of the hybrids through fertilisation, and early growth and survival, prompts the question of why they are so rare in Puget Sound.  The authors discuss this and suggest that the explanation may lie in the only slight overlap in spawning periods of the 2 species in this region.  Matson et al. 2003 J Shellf Res 22(2): 423.

NOTE  the researchers use diagnostic mDNA markers to discriminate between the broodstock and hybrids of the 2 species

NOTE  in Puget Sound M. trossulus spawns Mar-May, while M. galloprovincialis spawns Dec-Mar

The 2 species Mytilus trossulus and M. galloprovincialis are similar morphologically, but what about their habitat preferences?  This question is addressed in a comprehensive investigation at 30 sites throughout Puget Sound, involving identifications using a species-specific genetic marker, measurements of body shape, and comparisons of habitat requirements.  Over 6000 mussels are collected, of which 1460 are genetically typed (see map).  Results show that all 3 genetic types are found throughout the Sound, but with a preponderance of M. trossulus (92%) as compared with M. galloprovincialis (2%) or their hybrids (5%).  Mytilus galloprovincialis and hybrids are generally larger and have a greater height:length ratio than M. trossulus (see histogram showing size-frequency distributions for the 2 species and their hybrids on Left). The authors provide a handy polynomial equation to differentiate the exotic and partial exotic from the indigenous M. trossulus.  Interestingly, the first 2 live somewhat more subtidally and, correspondingly, lower in the intertidal region than than does M. trossulus.  The authors speculate that this could result from a greater intolerance to low salinities in M. galloprovincialis (see also Research Study 2 above).  At more than half the sites the frequencies of the 3 genotypes accord with random mating expectations, suggesting unrestricted gene flow between the species and hybrids.  Based on their habitat-preference results the authors caution other researchers to adopt truly random and stratified sampling protocols in their studies. Elliott et al. 2008 Mar Biol 156: 39.

NOTE  the marker, Me15/16, is a fragment of an adhesive protein that is a component of the byssus thread

NOTE  thus, any mussel with height >0.6L + (-0.0025L2) + 4 is M. galloprovincialis or a hybrid.  The method is successful 95% of the time for mussels 40-80mm in Length