ESTERASE VARIABILITY IN THE GASTROPODS LITTORINA SAXATILIS (OLIVI) AND L. ARCANA ELLIS

The esterase system has been studied using isoelectric focusingin two closely related rough periwinkles, Littorina saxatilisand L. arcana, in an investigation of possible differences betweenthe two species at this level of discrimination. Collectionsof both species were obtained from Yorkshire, north Cornwalland south Devon; L. saxatilis was also obtained from sites insouth Cornwall (in and near the estuary of the Tamar), Dorsetand Scotland, where L. arcana was not present. There is considerablevariation both within and between sites which tends to maskany overall differences between the two species. However, ingeneral there is a much greater degree of heterogeneity withinsamples of L. saxatilis than within samples of L: arcana whenboth species are found together; this is reflected in the ShannonWiener diversity index for bands. Also the number of esterasebands is greater for L. saxatilis than for L. arcana at eachof the five sites where both occurred. A dendrogram derivedfrom the percentage occurrence of each band present in eachsample produced a reasonable degree of separation of the twospecies (Received 8 January 1988; accepted 24 February 1988)     

ANNUAL AND LUNAR REPRODUCTIVE CYCLES IN LITTORINA SAXATILIS (OLIVI) AND DIFFERENCES BETWEEN BREEDING IN THE MARINE FIRTH OF FORTH AND THE FORTH ESTUARY

Gonads were most commonly reduced in October and fullest inMar-Jun with more pronounced seasonal differences in the estuarythan in the firth. Brood-pouches accordingly were fullest inspring-early summer and emptiest in autumn-early winter. Eggsand early embryos predominated in brood-pouches of early springbut were few in Aug-Oct whereas well-developed juveniles werefew in early spring, increased in spring-summer and came topredominate in Jun-Nov. Brood-pouch loads varied with size offemale but, for a standard 10 mm female, were significantlylower in the 2 estuarine sites than in the firth. Changes inmean brood-pouch count varied only 2.5-fold during the yearin the firth but 4.10-fold in the estuary. In vitro development of eggs to shelled juveniles with the velumresorbed took some 68 d at 10°C and 32%. Eggs and late brood-pouchjuveniles were significantly heavier in the estuary than inthe firth (dry weight, ash-free dry weight (AFDW), ash weight).AFDW changed little during development while ash weight (mostlyshell) increased 3.5-4.4-fold. In a laboratory tide tank, femalesreleased 60 d means of <1.3 juveniles d^–1; more inmid- and late year than in Jan-Mar, more from upshore firthfemales than from low shore or estuary females, more when heldat high tide-tank levels than at lower levels, and generallyat rates comparable to release in containers held at the siteof origin. Egg production in laboratory females did not keeppace with release of young and, at the end of trials, brood-pouchcounts were mostly lower than in contemporary free winkles.Females that had released most (e.g. at high tank levels) containedleast brood-pouch embryos and young subsequently. Release of young in the tide tank was markedly faster at timesnear new moon (often reaching or exceeding 2 female^–1d^–1) than near full moon (often <1 female^–1 d^–1).This lunar fluctuation was clearest in fast-releasing sets heldat high- or mid-tide tank levels. It was not observed in Jan-Mar(when release from most females was slow) nor in females atlow tank levels. (Received 3 September 1990; accepted 8 November 1990)     

A GENETIC ANALYSIS OF SYMPATRIC SUBPOPULATIONS OF THE SIBLING SPECIES LITTORINA SAXATILIS (OLIVI) AND LITTORINA ARCANA HANNAFORD ELLIS

A detailed analysis of shell characteristics (four parameters)and allozyme allele frequencies (eight to nine polymorphic loci)of sympatric Littorina saxatilis and Littorina arcana subpopulationsfrom Filey Brigg on the east coast of England supports theiridentification as closely related but separate species. Oneenzyme system, esterase-2, was nearly species diagnostic. Resultswere generally comparable with an earlier study of the two speciesfrom East Lothian, Scotland. Littorina saxatilis was again foundto be more variable than L. arcana. (Received 14 December 1983;     

EMBRYONIC ABNORMALITIES IN POPULATIONS OF LITTORINA SAXATILIS (OLIVI) (GASTROPODA: PROSOBRANCHIA) IN THE WHITE SEA

The incidence of abnormally developing embryos in brood pouchesof females was studied in five populations of Littorina saxatilisliving under different salinity conditions. The proportion ofabnormals decreased during embryonic development. Age of femaleand infection with the ciliate Protophrya ovi-cola had no effecton the frequency of abnormals. The proportion of abnormal eggs(PAE) increased at the end of the reproductive season; a similartrend was found for the proportion of abnormal veligers (PAV)although it was not always significant. The proportion of abnormaljuveniles (PAJ) showed no seasonality. The increase of PAE coincidedwith decrease of water temperature in September. Environmentalsalinity was significantly correlated with frequency of abnormalsin L. saxatilis populations: the highest values of PAV and PAJwere observed in estuarine population (salinity range 5–8%),intermediate in populations living under conditions of 10–20%osalinityand lowest in control populations (23–24%). Itis suggested that critical salinity (sensu Khle-bovich, 1974)(approximately 8–10%) may also be a crucial point forstability of embryonic development in the White Sea populationsof L. saxatilis. (Received 28 June 1994; accepted 28 November 1994)     

CHROMOSOME NUMBER IN TWO PHENOTYPICALLY DISTINCT POPULATIONS OF LITTORINA SAXATILIS OLIVI, AND IN SPECIMENS OF THE LITTORINA OBTUSATA (L.) SPECIES-COMPLEX

Chromosome numbers counted from embryo cells of the marine gastropodLittorina saxatilis (=L. rudis), were the same, 2n = 34, inpopulations of two different phenotypes from a small area inSweden. Chromosome number in the related species-complex, Littorinaobtusata, could not be determined definitely, but the most adequatefigure seemed to be 2n = 34. (Received 9 September 1982;     

INCIPIENT REPRODUCTIVE ISOLATION BETWEEN TWO SYMPATRIC MORPHS OF THE INTERTIDAL SNAIL LITTORINA SAXATILIS

The study of speciation in recent populations is essentially a study of the evolution of reproductive isolation mechanisms between sub-groups of a species. Prezygotic isolation can be of central importance to models of speciation, either being a consequence of reinforcement of assortative mating in hybrid zones, or a pleiotropic effect of morphological or behavioral adaptation to different environments. To suggest speciation by reinforcement between incipient species one must at least know that gene flow occurs, or have recently occurred, and that assortative mating has been established in the hybrid zone. In Galician populations of the marine snail Littorina saxatilis, two main morphs appear on the same shores, one on the upper-shore barnacle belt and the other in the lower-shore mussel belt. The two morphs overlap in distribution in the midshore where hybrids are found together with pure forms. Allozyme variation indicates that the two parental morphs share a common gene pool, although within shores, gene flow between morphs is less than gene flow within morphs. In this study, we observed mating behavior in the field, and we found that mating was not random in midshore sites, with a deficiency of heterotypic pairs. Habitat selection, assortative mating, and possibly sexual selection among females contributed to the partial reproductive isolation between the pure morphs. Sizes of mates were often positively correlated, in particular, in the upper shore, indicating size-assortative mating too. However, this seemed to be a consequence of nonrandom microdistributions of snails of different sizes. Because we also argue that the hybrid zone is of primary rather than secondary origin, this seems to be an example of sympatric reproductive isolation, either established by means of reinforcement or as a by-product to divergent selection acting on other characters.     

FUNCTIONAL MORPHOLOGY OF THE RECTUM IN THE MARINE GASTROPOD LITTORINA SAXATILIS (OLIVI) (PROSOBRANCHIA:LITTORINOIDEA)

The ultrastructure of the rectum of the marine prosobranch gastropod,Littorina saxatilis suggests that it may have an additionalrole to the voiding of faecal pellets. It consists of two epitheliallayers, separated by a blood space. The inner epithelial layeris greatly folded and consists of metabolically highly activecells containing a wide variety of inclusions, and showing evidenceof secretory activity at their luminal surfaces. The outer epitheliumcontains a greater diversity of cell types, also with a largenumber of different inclusions present. Unlike some other prosobranchgastropods, a distinct rectal gland is absent in L. saxatilis.The entire rectal epithelium is glandular, and it appears thatthe rectal epithelial cells produce a number of secretions.This may indicate a role in absorption, the removal of materialfrom the blood, or the addition of a final covering to the faecalpellets. (Received 6 February 1989; accepted 6 March 1989)     

PHENOTYPIC DIFFERENTIATION IN LITTORINA SAXATILIS OLIVI (MOLLUSCA, PROSOBRANCHIA) IN A SMALL AREA ON THE SWEDISH WEST COAST

The Littorina saxatilis species-complex exhibits two distinctmorphs in the area investigated. Differences in shell morphologyand number of penial glands are statistically significant (P<0.02)between populations from exposed and sheltered habitats. However,in intermediate environments intermediate specimens as wellas typical specimens of both morphs are found. The differencesare considered intraspecific and at least partly geneticallydetermined, and possible selection pressures responsible arediscussed. (Received 8 February 1981;     

VARIATION IN THE OCCURRENCE OF ABNORMAL EMBRYOS IN FEMALES OF THE INTERTIDAL GASTROPOD LITTORINA SAXATILIS OLIVI

The incidence of abnormally developed embryos within the brood-pouchof female L. saxatilis was very variable, ranging from 0–100%of embryos being abnormal. A minor part of the total variationwas attributed to between-shore variation, while 82% reflectedwithin sample differences between females. Possible factorsresponsible for the variation in embryo abnormality shown bythis and other investigations, are discussed and it is concludedthat genetic rather than non-genetic factors are important. (Received 8 November 1983;     

HYBRID FITNESS SEEMS NOT TO BE AN EXPLANATION FOR THE PARTIAL REPRODUCTIVE ISOLATION BETWEEN ECOTYPES OF GALICIAN LITTORINA SAXATILIS

Partial reproductive isolation between closely related groupsof organisms is suggested to be of central importance duringspeciation. Galician populations of Littorina saxatilis arephenotypically differentiated into an upper-shore and a lower-shoremorph. These mate assortatively in the mid-shore zone of overlap,and genetic assessment shows an impeded gene flow between theparental morphs. The traditional explanation as to why assortativemating occurs is that reproductive isolation is reinforced dueto hybrid unfitness. Earlier studies have, however, not foundhybrids to be less viable. Likewise, growth and migratory behavioursare merely intermediate between those of the parental morphs.In the present study we compared male and female fertilitycomponents of the parental morphs and the hybrids to test hypothesesof decreased hybrid fertility. The results showed that hybridmales were as fertile as other males, and hybrid females didnot produce fewer embryos, nor aborted embryos at a higher rate,than the parental morphs. ^* To whom correspondence should be addressed. (Received 19 March 1999; accepted 7 July 1999)     

MORPHOLOGICAL DIFFERENTIATION AND GENETIC COHESIVENESS OVER A MICROENVIRONMENTAL GRADIENT IN THE MARINE SNAIL LITTORINA SAXATILIS

The marine gastropod Littorina saxatilis has different ecotypes in shores only a few meters apart. This has both taxonomic and evolutionary implications. Here we report on an extreme type of within-shore dimorphism in shell characters. In the wave-exposed rocky shores in northwestern Spain, we found one form of L. saxatilis in the upper-level barnacle zone. It had a white, ridged shell, with black bands in the grooves. Another form confined to the lower-shore mussel belt had a smooth shell that was either white and tessellated or darkly colored. These two forms cooccured in a narrow midshore zone together with individuals that had combined characters, but were present in low frequencies (11%–29%). We used principal-component analysis of metric shell characters to study variation in shell size and shape. We found that the upper-shore form was larger than the lower-shore form. We also found small but significant differences in shell shape. Experiments in a common laboratory environment suggested the differences in shell ornamentation and color are inherited, but the individuals did not develop the morph-specific characters until a shell height of about 3 mm. The occurrence of mainly two distinct forms may suggest the presence of two species that hybridize. An analysis of five polymorphic enzyme loci in populations of snails from three geographically separated sites indicated, however, that there was no positive correlation between morphological distances and genetic distances among populations on a geographic scale (tens of kilometers). Thus, we rejected the hypothesis of two species. However, on a microgeographic scale (meters), genetic differentiation between groups with the same form was less than differentiation between forms. This indicated a partial barrier to gene flow between the two forms, and preliminary mate choice data suggested this was caused by nonrandom mating in the midshore zone of overlap.     

GENETIC EVIDENCE FOR REPRODUCTIVE ISOLATION AND MULTIPLE ORIGINS OF SYMPATRIC TROPHIC ECOTYPES OF WHITEFISH (COREGONUS)

We assessed variation in mitochondrial DNA (mtDNA) by restriction fragment length polymorphism (RFLP) analysis and in nuclear genes by allozyme analysis among sympatric pairs of limnetic and benthic ecotypes of whitefish (Coregonus) coexisting in three lakes of southern Yukon to address three evolutionary questions regarding their origins. Are sympatric low and high gill-raker count ecotypes genetically differentiated? Are they issued from monophyletic or polyphyletic evolutionary events? If they are polyphyletic in origins, did they originate from multiple allopatric speciation events or intralacustrine radiation? Our results corroborated previous genetic and ecological studies of these ecotypes, indicating that they represent genetically distinct reproductive units, and therefore refuting the hypothesis of phenotypic polymorphism within a single population. However, the amount of gene flow between ecotypes varied among lakes, correlating with the extent of morphological differentiation and the potential for premating reproductive isolation. The results indicated a polyphyletic origin of ecotypes whereby each of them have been expressed independently more than once. In the two lakes of Squanga Creek drainage, the existence of sympatric pairs was best explained by the secondary contact of two monophyletic whitefish groups that evolved in allopatry during the last glaciation events. In Dezadeash L. of Alsek R. drainage, our results could not verify either sympatric or allopatric (or microallopatric) origin of ecotypes. Regardless of the mode of speciation involved in their origins, these sympatric whitefish populations provided further evidence that Pleistocene glaciation events created conditions favoring rapid divergence and phenotypic differentiation among northern freshwater fishes.     

MORPHOLOGICAL AND ALLOZYMIC EVIDENCE FOR SABATIA FORMOSA (GENTIANACEAE) IN THE SECTION CAMPESTRIA

An investigation of eleven populations of Sabatia, utilizing twelve morphological traits and fifteen allozymic loci, revealed the presence of three species in the section Campestria. Currently, the section has two recognized species, S. campestris Nutt. and S. arenicola Greenm. A stepwise discriminant analysis of the morphological data indicated the presence of the third previously described species, S. formosa Buckl. Eight of the loci were either monomorphic or there were no allozyme frequency differences among the three species. An analysis of the allozyme frequencies of the seven polymorphic loci substantiated the presence of S. formosa. Ecological and phenological phenomena further separate the three species. A distribution map of the populations analyzed, populations observed, and populations represented in herbarium specimens indicates that the ranges of the three species are separate although sympatry does occur. Thus, the morphological form described and designated S. formosa by Buckley in 1862 is a separate species.     

DISTRIBUTION,MORPHOLOGICAL DIVERSITY AND EVIDENCE FOR POLYPLOIDY IN NORTH AMERICAN ZYGNEMATACEAE (CHLOROPHYTA)1

Large-scale collections of Zygnemataceae in the continental United States of America were made between March and August in 1982, 1983, and 1984. Collections were made on a 31000-km transect through 35 states. Zygnemataceae were found at 318 sites was inspected. Temperature average 19°C and p^H averaged 6.1 over all sites. Algal strains in collections were identified to genus, characterized for filament width, chloroplast number, and end wall type, then photographed and isolated into unialgal culture. Spirogyra was the most common genus collected(632 strains), followed in abundance by Zygnema (174 Strains) and Mougeotia (135 strains). These three genera contained 95% of the strains collected and were equally widely distributed. Strains of the three genera frequently occurred together; no genus displayed evidence of habitat specialization among the three habitat types: flowing water, permanent ponds or lakes, and temporary pools. In Spirogyra, strains with plane (flat) end walls were four times more abundant than those with replicate (interlocking) end walls. Spirogyra with plane end walls showed more variation in filament width than Zygnema, Mougeotia, or Spirogyra with replicate end walls. In Spirogyra with plane end walls, filament width was correlted with nuclear DNA content and number of strains found per collection site was twice that of other genera or Spirogyra, with replicate end walls. Spirogyra strains wider than 70 μm were more frequent on the northern part of the transect. It is proposed that polyploidy may be of widespread occurrence in Spirogyra with plane end walls and that associated morphological plasticity may account for the high apparent specied diversity and survival of the genus in a wider variety of microhabitats than occupied by other Zygnemataceae.     

ALLOZYME VARIATION IN A SNAIL (LITTORINA SAXATILIS)—DECONFOUNDING THE EFFECTS OF MICROHABITAT AND GENE FLOW

It is commonly observed that a restricted gene flow among populations of a species generates genetic differentiation in, for example, allozyme markers. However, recent studies suggest that microhabitat-specific variation may contribute to the total differentiation. To appreciate the relative contributions of geographic variation and habitat-specific variation, we sampled 42 subpopulations of the intertidal snail Littorina saxatilis from three different microhabitats (boulders, low and high rocky intertidal) on five small islands within a distance of 15 km. We used a modified orthogonal version of Nei's gene diversity analysis with a modified analysis of variance (ANOVA) that estimated the significance of habitat and geographic separation and the interaction between them. Between subpopulation differentiation (G_ST) was usually in the range of 5% to 10% but was exceptionally high in one locus (Aat; 53%). Genetic differentiation attributable to different habitats accounted for 10% to 81% (mean 35%) of the between subpopulation variation and was significant (P < 0.05) in six loci. Differentiation due to geographic separation accounted for 11% to 61% (mean 36%) and was significant (P < 0.05) in seven loci. Furthermore, three loci showed interactions between habitat and island, suggesting varying effects of habitat in different islands. Microhabitat-specific variation, probably through spatially varying fitness, seems particularly likely in Aat and Pgm-2. Moderate levels of habitat associated variation added to the observed differentiation due to gene flow in Pgi, Pnp, and Pgm-1, whereas in the remaining three loci either the habitat effect was confused by strong habitat-island interaction (Ark) or was virtually absent (Pep and Mpi).     

EVIDENCE FOR MULTIPLE ORIGINS AND SYMPATRIC DIVERGENCE OF TROPHIC ECOTYPES OF SMELT (OSMERUS) IN NORTHEASTERN NORTH AMERICA

The rainbow smelt, Osmerus mordax (Mitchill), is an osmerid fish that exhibits extensive life-history diversity throughout watersheds of northeastern North America. There are both ???sea-run (anadromous) and lake-resident (lacustrine) populations and the latter have diversified further into “dwarf-” and “normal-sized” life-history types. Anadromous and lacustrine smelt may inhabit the same watershed and there are several instances where dwarf and normal populations reside within the same lake. We assayed variation among smelt for morphological traits linked to feeding performance in fishes to see if trophic ecology might promote life-history diversity in Osmerus. We also examined mitochondrial DNA (mtDNA) restriction site variation among forms to assess their evolutionary interrelationships. Dwarf smelt had significantly more gill rakers, larger eyes, but shorter upper jaws than normal lake and anadromous smelt. The populations clustered into two trophic “morphotypes”; an anadromous/normal lake group of populations and a group consisting only of dwarf smelt. The mtDNAs of 444 smelt from 16 populations were digested with 12 restriction enzymes revealing 93 composite mtDNA genotypes that clustered (UPGMA) into two major phylogenetic groups differing by 0.78% in sequence. Both genetic groups were present in dwarf and normal smelt as well as in anadromous fish. Further, geographic proximity, rather than trophic morphotype, appeared to be the major determinant of genetic affinities among populations. In two lakes, however, dwarf and normal smelt populations had significantly different mtDNA genotype frequency distributions indicating that the forms are reproductively isolated within both lakes. A clustering analysis of population affinities suggested that the divergence of sympatric dwarf and normal populations had occurred independently in the two lakes. We concluded (1) that trophic ecology is an important factor promoting differentiation in smelt life histories; (2) that smelt ecotypes are polyphyletic and there have been multiple, independent divergences of Osmerus life-history types throughout northeastern North America; and (3) that the biological and mtDNA differences between coexisting dwarf and normal lake smelt argue strongly that their genetic isolation may have developed sympatrically.