Experimental evidence for trait utility of gill raker number in adaptive radiation of a north temperate fish

North temperate fish in post‐glacial lakes are textbook examples for rapid parallel adaptive radiation into multiple trophic specialists within individual lakes. Speciation repeatedly proceeded along the benthic–limnetic habitat axis, and benthic–limnetic sister species diverge in the number of gill rakers. Yet, the utility of different numbers of gill rakers for consuming benthic vs. limnetic food has only very rarely been experimentally demonstrated. We bred and raised families of a benthic–limnetic species pair of whitefish under common garden conditions to test whether these species (i) show heritable differentiation in feeding efficiency on zooplankton, and (ii) whether variation in feeding efficiency is predicted by variation in gill raker numbers. We used zooplankton of three different size classes to investigate prey size dependency of divergence in feeding efficiency and to investigate the effect strength of variation in the number of gill rakers. Our results show strong interspecific differences in feeding efficiency. These differences are largest when fish were tested with the smallest zooplankton. Importantly, feeding efficiency is significantly positively correlated with the number of gill rakers when using small zooplankton, also when species identity is statistically controlled for. Our results support the hypothesis that a larger number of gill rakers are of adaptive significance for feeding on zooplankton and provide one of the first experimental demonstrations of trait utility of gill raker number when fish feed on zooplankton. These results are consistent with the suggested importance of divergent selection driven feeding adaptation during adaptive radiation of fish in post‐glacial lakes.     

Molecular genetic evidence for reproductive isolation between sympatric populations of smelt Osmerus in Lake Utopia, south-western New Brunswick, Canada

Two morphologically and ecologically distinct forms of smelt, Osmerus, reside sympatrically in Lake Utopia, south-western New Brunswick, Canada. The ‘normal-sized’ form matures at greater than 200 mm standard length, averages about 31–33 gill rakers, and spawns in lake outlets. By contrast, the ‘dwarf-sized’ form matures at less than 150 mm standard length, averages 34–36 gill rakers, and spawns in small streams 3–5 weeks later than the normal form. We tested whether these sympatric forms represented ecological polymorphism within a single population or two reproductively isolated demes by assaying variation within and between forms by mitochondrial DNA (mtDNA) restriction site and nuclear minisatellite DNA analyses. Analysis of smelt mtDNA with twelve restriction enzymes resolved ten composite genotypes (differing by an average 0.27% sequence divergence) which differed markedly in frequency between the forms. Net percentage sequence divergence between the forms was O.l6%. A Wagner parsimony/ bootstrapping analysis of the restriction site presence/absence matrix, however, suggested that there were no significant distinctions between dwarf and normal smelt based on the phylogeny of composite genotypes. Hybridization studies of genomic DNA digests with a minisatellite probe indicated both that nuclear restriction fragment differentiation and the frequency of specific fragments differed significantly between the forms. Significant genetic differentiation between the sympatric forms demonstrates that they are distinct gene pools and reproductively isolated. Our molecular evidence for reproductive isolation between dwarf and normal smelt in Lake Utopia, coupled with the persistent morphological and ecological differentiation between them, argues strongly that they are behaving as distinct species. The Lake Utopia Osmerus populations provide further illustration of the potential for rapid differentiation to the level of biological species in postglacial environments.     

Diel and seasonal habitat and food segregation of three sympatric Coregonus lavaretus forms in a subarctic lake

Habitat use of three sympatric whitefish Coregonus lavaretus forms was determined using hydroacoustics, pelagic trawling and epibenthic gillnetting in the subarctic Lake Muddusjärvi during the day and night in June, August and September. Whitefish constituted 97% of the numerical catches and whitefish with high number of gill rakers (DR) were the most abundant whitefish form. Forms with low numbers of gill rakers used only epibenthic habitats during both the day and night in all study periods: large whitefish with low numbers of gill rakers (LSR) dwelled mainly at depths 0–10 m, whereas small whitefish with low numbers of gill rakers (SSR) used deeper (>10 m) habitats. LSR and SSR whitefish consumed mainly benthic macroinvertebrates during all study occasions. The planktivorous DR whitefish used both epibenthic and pelagic habitats, but vertical habitat selection varied both over time of day and season. In June, when light intensity was continuously high, DR whitefish did not perform diel vertical migrations. In August and September, when dark nights were distinguishable, DR whitefish ascended from the bottom to the pelagic at dusk to feed on zooplankton, and descended at dawn. DR whitefish used pelagic habitats only at the lowest light intensities during the night, which was probably related to the high predation risk from brown trout Salmo trutta .     

Different growth trends of whitefish (Coregonus lavaretus) forms in the northern Baltic Sea

The sea growth of two whitefish forms, anadromous (Coregonus lavaretus lavaretus) and sea‐spawning (Coregonus lavaretus widegreni), was analysed using samples collected from the commercial sea catch in the Gulf of Bothnia (GoB) in the northern Baltic Sea during 1998–2014. In the GoB area, these two forms are possible to identify because the gill‐raker number and size at maturity vary between forms. The growth rate of the forms is linked to their feeding area. Sea‐spawning whitefish, which has a feeding migration near its home site, was shorter in the northern GoB (66°N–64°N) at the ages of 3–11 than those in the southern GoB (64°N–60°30′N). In the data, most whitefish were caught with gill nets in the GoB. The mesh sizes of gill nets capturing the anadromous form were mostly 35–45 mm, while those capturing the sea‐spawning form were <35 mm in the northern GoB. It is likely that the different growth trends for small and large whitefish were connected with differences in their recruitment for fishing. The length of anadromous females at the age of four sea years increased significantly, but the length of six‐year‐old anadromous female whitefish decreased over the catch years from 1998–2014. In contrast, the length of slow‐growing sea‐spawning whitefish of six years or older increased significantly in relation to the catch year in the gill‐net catch. The increase in the growth of young age groups in both forms was probably associated with the increasing temperature and the low fishing pressure on small fish. The decreasing age at capture for both forms and the depression of the mean size of old anadromous whitefish are signs of high fishing pressure with a high gill‐net effort that selectively removes the largest and oldest individuals of both forms.     

Collapse of Lake Whitefish Coregonus clupeaformis (Mitchill, 1818) species pair in Como Lake,Ontario

Lake Whitefish (Coregonus clupeaformis) species pairs exist in a small number of northern temperate lakes in North America. Sympatric populations of dwarf and normal‐sized whitefish represent a continuum of morphological and genetic divergence among these lakes. In this study, we used morphological and age data to assess the status of the Lake Whitefish species pair in Como Lake, Ontario (Canada). Whitefish were collected with gill‐nets from spawning shoals during the fall of 2012 and 2014 and compared with archived (circa 1989) dwarf and normal‐sized specimens. Our study indicates that the two previously collected forms have been replaced by a single form with a different morphology and age structure. In comparison to archived specimens, contemporary Lake Whitefish were larger, deeper‐bodied and older; with more gill rakers, lateral line scales, smaller eyes and a wider interorbital width. We hypothesize that the recent invasion of Como Lake by Spiny Waterflea (Bythotrephes longimanus) has affected whitefish growth rates and changed the ecosystem conditions that previously maintained the species pair.     

Patterns of morphological changes and hybridization between sympatric whitefish morphs (Coregonus spp.) in a Swiss lake: a role for eutrophication?

Whitefish, genus Coregonus, show exceptional levels of phenotypic diversity with sympatric morphs occurring in numerous postglacial lakes in the northern hemisphere. Here, we studied the effects of human‐induced eutrophication on sympatric whitefish morphs in the Swiss lake, Lake Thun. In particular, we addressed the questions whether eutrophication (i) induced hybridization between two ecologically divergent summer‐spawning morphs through a loss of environmental heterogeneity, and (ii) induced rapid adaptive morphological changes through changes in the food web structure. Genetic analysis based on 11 microsatellite loci of 282 spawners revealed that the pelagic and the benthic morph represent highly distinct gene pools occurring at different relative proportions on all seven known spawning sites. Gill raker counts, a highly heritable trait, showed nearly discrete distributions for the two morphs. Multilocus genotypes characteristic of the pelagic morph had more gill rakers than genotypes characteristic of benthic morph. Using Bayesian methods, we found indications of recent but limited introgressive hybridization. Comparisons with historical gill raker data yielded median evolutionary rates of 0.24 haldanes and median selection intensities of 0.27 for this trait in both morphs for 1948–2004 suggesting rapid evolution through directional selection at this trait. However, phenotypic plasticity as an alternative explanation for this phenotypic change cannot be discarded. We hypothesize that both the temporal shifts in mean gill raker counts and the recent hybridization reflect responses to changes in the trophic state of the lake induced by pollution in the 1960s, which created novel selection pressures with respect to feeding niches and spawning site preferences.     

Ecological speciation in postglacial European whitefish: rapid adaptive radiations into the littoral,pelagic, and profundal lake habitats

Understanding how a monophyletic lineage of a species diverges into several adaptive forms has received increased attention in recent years, but the underlying mechanisms in this process are still under debate. Postglacial fishes are excellent model organisms for exploring this process, especially the initial stages of ecological speciation, as postglacial lakes represent replicated discrete environments with variation in available niches. Here, we combine data of niche utilization, trophic morphology, and 17 microsatellite loci to investigate the diversification process of three sympatric European whitefish morphs from three northern Fennoscandian lakes. The morphological divergence in the gill raker number among the whitefish morphs was related to the utilization of different trophic niches and was associated with reproductive isolation within and across lakes. The intralacustrine comparison of whitefish morphs showed that these systems represent two levels of adaptive divergence: (1) a consistent littoral–pelagic resource axis; and (2) a more variable littoral–profundal resource axis. The results also indicate that the profundal whitefish morph has diverged repeatedly from the ancestral littoral whitefish morph in sympatry in two different watercourses. In contrast, all the analyses performed revealed clustering of the pelagic whitefish morphs across lakes suggesting parallel postglacial immigration with the littoral whitefish morph into each lake. Finally, the analyses strongly suggested that the trophic adaptive trait, number of gill rakers, was under diversifying selection in the different whitefish morphs. Together, the results support a complex evolutionary scenario where ecological speciation acts, but where both allopatric (colonization history) and sympatric (within watercourse divergence) processes are involved.     

Generation of a neutral FST baseline for testing local adaptation on gill raker number within and between European whitefish ecotypes in the Baltic Sea basin

Divergent selection at ecologically important traits is thought to be a major factor driving phenotypic differentiation between populations. To elucidate the role of different evolutionary processes shaping the variation in gill raker number of European whitefish (Coregonus lavaretus sensu lato) in the Baltic Sea basin, we assessed the relationships between genetic and phenotypic variation among and within three whitefish ecotypes (sea spawners, river spawners and lake spawners). To generate expected neutral distribution of F_ST and to evaluate whether highly variable microsatellite loci resulted in deflated F_ST estimates compared to less variable markers, we performed population genetic simulations under finite island and hierarchical island models. The genetic divergence observed among (F_CT = 0.010) and within (F_ST = 0.014–0.041) ecotypes was rather low. The divergence in gill raker number, however, was substantially higher between sea and river spawners compared to observed microsatellite data and simulated neutral baseline (P_CT > F_CT). This suggests that the differences in gill raker number between sea and river spawners are likely driven by divergent natural selection. We also found strong support for divergent selection on gill raker number among different populations of sea spawners (P_ST > F_ST), most likely caused by highly variable habitat use and diverse diet. The putative role of divergent selection within lake spawners initially inferred from empirical microsatellite data was not supported by simulated F_ST distributions. This work provides a first formal test of divergent selection on gill raker number in Baltic whitefish, and demonstrates the usefulness of population genetic simulations to generate informative neutral baselines for P_ST–F_ST analyses helping to disentangle the effects of stochastic evolutionary processes from natural selection.     

The role of gill raker number variability in adaptive radiation of coregonid fish

Gill raker divergence is a general pattern in adaptive radiations of postglacial fish, but few studies have addressed the adaptive significance of this morphological trait in foraging and eco-evolutionary interactions among predator and prey. Here, a set of subarctic lakes along a diversifying gradient of coregonids was used as the natural setting to explore correlations between gill raker numbers and planktivory as well as the impact of coregonid radiation on zooplankton communities. Results from 19 populations covering most of the total gill raker number gradient of the genus Coregonus, confirm that the number of gill rakers has a central role in determining the foraging ability towards zooplankton prey. Both at the individual and population levels, gill raker number was correlated with pelagic niche use and the size of utilized zooplankton prey. Furthermore, the average body size and the abundance and diversity of the zooplankton community decreased with the increasing diversity of coregonids. We argue that zooplankton feeding leads to an eco-evolutionary feedback loop that may further shape the gill raker morphology since natural selection intensifies under resource competition for depleted prey communities. Eco-evolutionary interactions may thus have a central role creating and maintaining the divergence of coregonid morphs in postglacial lakes.     

Gill raker structure and selective predation on zooplankton by particulate feeding fish

The relationship between the gill raker structure of planktivorous fish (number, distance between gill rakers and length) and selective feeding on different species and size classes of zooplankton was studied. Gill raker structure was measured for brown trout Salmo trutta , Arctic charr Salvelinus alpinus , whitefish Coregonus lavaretus , roach Rutilus rutilus , bleak Alburnus alburnus , and three-spined stickleback Gasterosteus aculeatus . All species are facultative planktivorous fish and occur commonly in Scandinavian lakes. The effect of gill raker structure was studied by comparing prey found in fish stomachs with the availability of zooplankton from several lakes. Gill raker length and distance were significantly correlated with fish length. Although gill raker structure differed among species, all fish species selected the larger zooplankters. The minimum size of cladoceran species found in fish stomachs was much smaller than the distance between gill rakers. Despite great differences in gill raker spacing, the minimum size ingested of Daphnia galeata and Bosmina longispina was similar for all predators. The hypothesis that small zooplankton are strained and retained by the gill rakers in particulate feeding planktivorous fish, particularly in salmonids and roach, is rejected.     

The effect of core compression on the measurement of zinc concentrations and anthropogenic burdens in lake sediments

The whitefish (Coregonus lavaretus) in Lake Tyrifjorden, southeastern Norway, is monomorphic with regard to gill raker number: (x=32), and exhibits a bimodal length distribution. The small whitefish (<27 cm) were distributed mainly in the littoral zone and the deeper layers of the pelagic zone. The larger whitefish (>27 cm) were mainly distributed in the upper layers of the pelagic zone. Both small and large whitefish fed on zooplankton in the pelagic zone. In the littoral zone the diet of the two size groups was significantly different; the small whitefish fed mainly on zooplankton whereas the large whitefish fed mainly on zoobenthos.     

Variation in number of vertebrae and gill rakers of sockeye salmon,Oncorhynchus nerka,in North America

Synopsis The latitudinal, regional, and annual variation in number of vertebrae and number of gill rakers present in sockeye salmon, Oncorhynchus nerka, stocks in North America was examined. Stocks in more northern areas had higher numbers of vertebrae and gill rakers than did those in more southern ones. Significant annual variability in the frequencies of these meristic characters within stocks was observed. When stocks were grouped into three regions (southern, central, and northern), heterogeneity in vertebral and gill raker frequencies was greater among regions than among stocks within the regions. Similarly, heterogeneity was greater among stocks than among sampling years within stocks. Differences in vertebral and gill raker frequencies are only useful for stock identification of sockeye salmon on a broad regional basis.     

Diversifying selection drives parallel evolution of gill raker number and body size along the speciation continuum of European whitefish

Adaptive radiation is the evolution of ecological and phenotypical diversity. It arises via ecological opportunity that promotes the exploration of underutilized or novel niches mediating specialization and reproductive isolation. The assumed precondition for rapid local adaptation is diversifying natural selection, but random genetic drift could also be a major driver of this process. We used 27 populations of European whitefish (Coregonus lavaretus) from nine lakes distributed in three neighboring subarctic watercourses in northern Fennoscandia as a model to test the importance of random drift versus diversifying natural selection for parallel evolution of adaptive phenotypic traits. We contrasted variation for two key adaptive phenotypic traits correlated with resource utilization of polymorphic fish; the number of gill rakers and the total length of fish, with the posterior distribution of neutral genetic differentiation from 13 microsatellite loci, to test whether the observed phenotypic divergence could be achieved by random genetic drift alone. Our results show that both traits have been under diversifying selection and that the evolution of these morphs has been driven by isolation through habitat adaptations. We conclude that diversifying selection acting on gill raker number and body size has played a significant role in the ongoing adaptive radiation of European whitefish morphs in this region.     

Adaptive divergence between lake and stream populations of an East African cichlid fish

Divergent natural selection acting in different habitats may build up barriers to gene flow and initiate speciation. This speciation continuum can range from weak or no divergence to strong genetic differentiation between populations. Here, we focus on the early phases of adaptive divergence in the East African cichlid fish Astatotilapia burtoni, which occurs in both Lake Tanganyika (LT) and inflowing rivers. We first assessed the population structure and morphological differences in A. burtoni from southern LT. We then focused on four lake–stream systems and quantified body shape, ecologically relevant traits (gill raker and lower pharyngeal jaw) as well as stomach contents. Our study revealed the presence of several divergent lake–stream populations that rest at different stages of the speciation continuum, but show the same morphological and ecological trajectories along the lake–stream gradient. Lake fish have higher bodies, a more superior mouth position, longer gill rakers and more slender pharyngeal jaws, and they show a plant/algae and zooplankton‐biased diet, whereas stream fish feed more on snails, insects and plant seeds. A test for reproductive isolation between closely related lake and stream populations did not detect population‐assortative mating. Analyses of F1 offspring reared under common garden conditions indicate that the detected differences in body shape and gill raker length do not constitute pure plastic responses to different environmental conditions, but also have a genetic basis. Taken together, the A. burtoni lake–stream system constitutes a new model to study the factors that enhance and constrain progress towards speciation in cichlid fishes.     

Genetic differentiation in gill raker number and length in sympatric anadromous and nonanadromous morphs of sockeye salmon, Oncorhynchus nerka

The genetic and environmental basis for polymorphism in gill raker number and length in sympatric anadromous and nonanadromous morphs of sockeye salmon, Oncorhynchus nerka, was investigated. Analysis of 30 full sib families involving pure types and reciprocal hybrids revealed that the variation was partitioned significantly among families within cross types and among cross types in both traits. As in the wild, kokanee displayed more gill rakers than sockeye; reciprocal hybrids displayed intermediate counts. Gill raker length also varied markedly among cross types, with pure sockeye displaying 19% longer gill rakers than comparable sized kokanee. This difference was in the opposite direction predicted, given the common positive association between gill raker number and length in sympatric morphs of the same species in fishes. Gill raker number and length were generally not correlated within cross types, suggesting independent divergence of the traits. The results are discussed in relation to genetic and trophic divergence of the morphs and to factors selecting for differentiation in the two gill raker traits.     

Developmental progression of gill rakers as a post-hatch developmental marker in pink salmon, Oncorhynchus gorbuscha

We evaluate the usefulness of gill rakers as a post-hatch developmental marker in salmon by tracking development in undisturbed and stressed yolk-bearing salmon embryos. Native pink salmon (Oncorhynchus gorbuscha) from Auke Creek, Juneau, Alaska and genetically stressed outbred hybrids between Auke Creek and Pillar Creek (Kodiak Island) salmon were incubated in ambient-temperature Auke Creek water. Environmentally stressed native embryos were reared in water that was 2 to 4 °C warmer than ambient. The sum of rakers on the first left and right branchial arches of natives reared at ambient temperatures averaged 23.20 (SD?±?1.64) per embryo when post-hatch sampling began. The subsequent increase in raker counts was linear and positively correlated with the accumulation of thermal units until counts reached maxima 223 days after fertilization, which coincided with the complete consumption of yolk reserves. The average maximum raker count was 39.55 (SD?±?1.76) per embryo, which is substantially fewer than the 60 rakers typically observed in adults. Neither raker development nor yolk consumption patterns were affected by hybridization. Elevated incubation temperatures accelerated raker formation and yolk consumption in native embryos, but the number of rakers associated with a given amount of yolk was the same regardless of temperature suggesting that changes in yolk consumption rate and raker development rate did not influence raker counts. These results indicate that rakers are easily observed and counted, grow in a predictable sequence, and are developmentally stable in the face of both genetic and environmental stress, thereby making them potentially reliable post-hatch developmental markers.     

Trophic ecology and gill raker morphology of seven catostomid species in Iowa rivers

Understanding the trophic ecology of closely‐related species is important for providing insight on inter‐specific competition and resource partitioning. Although catostomids often dominate fish assemblages in lotic systems, little research has been conducted on their ecology. This study was developed to provide information on the trophic ecology of catostomids in several Iowa rivers. Food habits, diet overlap, and gill raker morphology were examined for highfin carpsucker Carpiodes velifer, quillback C. cyprinus, river carpsucker C. carpio, golden redhorse Moxostoma erythrurum, shorthead redhorse M. macrolepidotum, silver redhorse M. anisurum, and northern hogsucker Hypentelium nigricans sampled from four Iowa rivers (2009). Diet overlap among all species was calculated with Morista’s index (C). Food habit niche width was quantified with Levin’s index (B) and similarity in gill raker morphology was compared with analysis of covariance. Values from Morista’s index suggested significant overlap in the diets of highfin carpsucker and river carpsucker (C = 0.81), quillback and river carpsucker (C = 0.66), and shorthead redhorse and silver redhorse (C = 0.67). Levin’s index indicated that golden redhorse (B = 0.32), quillback (B = 0.53), and river carpsucker (B = 0.41) had the most generalized feeding strategies as their food niche widths were substantially wider than the other species. Gill raker length and spacing were positively correlated with the standard length of the fish for all species (gill raker length: r² = 0.67–0.88, P ≤ 0.01; gill raker spacing: r² = 0.63–0.73, P ≤ 0.01). Slopes of regression of gill raker length and spacing to standard lengths were significantly (P ≤ 0.05) different among species, indicating that rates of change in gill raker morphology with body length varied among species. Differences in gill raker morphology likely allow catostomids to partition resources and reduce competitive interactions.     

Structure,development and function of the branchial sieve of the common bream,Abramis brama,white bream,Blicca bjoerkna and roach,Rutilus rutilus

Synopsis The filter feeding organ of cyprinid fishes is the branchial sieve, which consists of a mesh formed by gill rakers and tiny channels on the gill arches. In order to establish its possible role during growth we measured the following morphological gill raker parameters over a range of sizes in three cyprinid fishes, bream, white bream and roach: inter raker distance, bony raker length, raker width, cushion length and channel width. At any given standard length common bream has the largest inter raker distance, roach the lowest and white bream is intermediate. In the comb model of filter feeding the inter raker distance is considered to be a direct measure of the mesh size and retention ability (= minimal size of prey that can be retained) of a filter. For the three species under study there is a conflict between the comb model and experimental data on particle retention. Lammens et al. (1987) found that common bream has a large retention ability whereas roach and white bream have a much smaller one. A new model, the channel model (Hoogenboezem et al. 1991) has been developed for common bream; in this model the lateral gill rakers can regulate the mesh size of the medial channels on the other side of the gill slit. The present data indicate that this model is not appropriate for white bream and roach. At any given standard length white bream and roach only reach 70% of the raker length of common bream, which means that in this model the gill slits should to be very narrow during filter feeding. The gill rakers consist of a bony raker and a fleshy cushion. The bony rakers have a rather long needle-like part outside the cushion in bream, but not in white bream and roach which have blunt gill rakers. Blunt gill rakers are not suited to reduce the diameter of the medial channels. The comb model seems more appropriate for white bream and roach, but doubts about the validity of this simple model remain. The sum of the areas of the medial channels is an approximation of the area through which water flows in the filter. This channel area therefore gives an impression of the capacity or flow rate of the filter. With this capacity estimation and an estimation of energy consumption we calculated an energy ratio of filter feeding. The energy ratio decreases with increasing standard length with an exponent close to the expected exponent of -0.40. The energy ratio is highest in bream, intermediate in white bream and lowest in roach.     

Contemporary trait change in a classic ecological experiment: rapid decrease in alewife gill‐raker spacing following introduction to an inland lake

相似文献   

Lake resident and migratory smelt, Retropinna retropinna (Richardson), of the lower Waikato River system, New Zealand

Retropinna retropinna occurred in two forms in Lake Waahi, North Island, New Zealand. One form was characterized by high mean gill raker counts ( c . 28) and low mean vertebral counts ( c . 52)—similar to those of non-migratory, lake resident, native smelt in some other lakes of the lower Waikato River system. The other form had low mean gill raker counts ( c . 21) and high vertebral counts ( c . 60)—similar to migratory smelt at five sites along the lower Waikato River down to its estuary. The lake resident form was shorter but deeper bodied than the migratory form, as indicated by significant differences in mean length and in weight-length regression analyses.
The two forms overlapped in time of sexual maturity. As in other native lacustrine smelt populations of the lower Waikato system, the resident form had more but smaller eggs than the migratory form. The egg number to body length regression and the mean egg diameter of migratory Lake Waahi smelt were similar to those of lower Waikato River smelt. Evidence for reproductive separation of the two forms is presented and mechanisms controlling their meristic and other differences are considered.