Three queen morphs with alternative nest-founding behaviors in the ant, Temnothorax longispinosus

In many ant species, multiple modes of founding new colonies occur in the same population. These modes include dependent founding, independent founding by haplometrosis (single queen), and independent founding by pleometrosis (multiple queens). In several cases, a dimorphism in queen size has been found, such that each morph specializes in a particular nest-founding behavior. I investigated queen size in the ant Temnothorax longispinosus in several southern Wisconsin populations and found three distinct queen morphs: small queens with very low fat content and short wings, large queens with low fat and long wings, and large queens with high fat and very long wings. Several traits associated with founding behavior correlated with these queen sizes. Small queens were produced in lower numbers, were more common in polygynous nests, and returned to the nest in higher proportions than both large queen morphs. The size ranges and fat levels of each queen morph were similar to those of other species that specialize in either haplometrosis (very large, high fat), pleometrosis (large, low fat), or dependent founding (small and low fat). However, there was extensive overlap in several of the founding behaviors, suggesting that the morphs in these populations have some flexibility in founding behavior. The queen morphs in these populations of T. longispinosus may resemble early stages in the evolution of more specialized dispersal polymorphisms found in other ant species. Received 11 January 2006; revised 15 September 2006; accepted 18 September 2006.     

Alternative migration and host parasitism strategies and their long-term stability in river lampreys from the River Endrick, Scotland

The stability of a discrete body size dimorphism of sexually mature river lamprey Lampetra fluviatilis from the River Endrick, Scotland, was examined over a 21 year period. Stable isotope analysis was used to test the hypothesis that the two size forms comprise individuals with differing migration and parasitic foraging strategies. Maturing river lamprey and the brook lamprey Lampetra planeri were trapped over 3 months each year in the periods 1983–1984 and 2004–2005. Brook lamprey catches and catches of both species combined showed no significant trend in catch rate with time. The catch rate of small body size river lamprey declined between 1983–1984 and 2004–2005 (although the difference did not reach statistical significance; P = 0·055). In contrast, there was a significant increase in the catch rate of the large body size river lamprey and as a consequence, a significant change in the relative proportion of each of the two river lamprey morphs over the study period. Analysis of the stable isotopes of C and N in muscle tissue showed that brook lamprey tissue derived its carbon from a freshwater source and had a δ¹³C more consistent with that of the River Endrick than with Loch Lomond. δ¹⁵N values for this species showed it to be feeding at the base of the food chain, consistent with filter feeding as an ammocoete. The large body size and the small body size river lamprey adults differed substantially in their δ¹³C values, with the small body size δ¹³C signature indicative of a freshwater carbon source and the large body size morph of a marine source. The small body size morph had a δ¹³C signature that was consistent with that of Loch Lomond powan Coregonus lavaretus suggesting that they share a common carbon source. The large body size morph was clearly feeding at a higher trophic level than the small body size morph. A single small body size river lamprey individual with typical morphology for that group, however, had C and N signatures that clustered with those of the large body size morphs. This individual had either migrated to sea to forage, as is typical for the species, or had been feeding on an anadromous fish with a strong marine C signature in fresh water. It is concluded that the body size dimorphism is indicative of a differential migration and foraging strategy in the parasitic phase of the life cycle of river lamprey at this site.     

A new morph of Telmatochromis temporalis (Cichlidae; Cichliformes) from Lake Tanganyika

Telmatochromis temporalis is a cichlid fish endemic to Lake Tanganyika. Two morphs of this species, normal and dwarf, form a good model for the study of ecological speciation through divergent natural selection on body size. This study reports a third morph of this species, slender morph, which was collected from deep waters off Kasenga, Zambia, whereas the normal morph inhabits shallow waters of the same locality. This study examined morphological characters and mitochondrial DNA sequences in 18 populations of the three T. temporalis morphs and two closely related species. The slender morph was morphologically similar to the normal and dwarf morphs of the same species, but clearly differed from closely related species. Genetic analyses showed that the slender morph was closest to but significantly different from the parapatric normal morph, suggesting reproductive isolation between them. Due to the lack of colour differences between morphs and of obvious geographical barriers between habitats, reproductive isolation between these morphs may be attributed to ecological factors, rather than sexual or geographical segregation. Further studies examining the evolution of the slender morph may deepen our knowledge of initial stages of speciation, like in the dwarf morph.

     

Ecological mechanisms for coexistence of colour polymorphism in a coral-reef fish: an experimental evaluation

The evolution of different colour morphs and how they are maintained in animal populations is poorly understood. We investigated the mechanisms maintaining yellow and brown morphs of a coral-reef fish, Pseudochromis fuscus, at Lizard Island, on the Great Barrier Reef. Histological examination of the gonads revealed that colour morphs were not sex-limited, therefore sexual selection does not appear to promote dichromatism in this species. The field distributions of the two colour morphs were spatially segregated, limiting the opportunity for negative frequency-dependent selection to operate. Our results support another ecological mechanism of coexistence. The yellow morph occurred in deeper areas, usually close to the reef edge, where there was a proportionally high cover of live branching corals. In contrast, the brown morph occurred in shallower areas, more distant from the reef edge, that were proportionally low in live branching corals. Within these habitats, each colour morph of P. fuscus displayed a close association with similar coloured damselfishes from the genus Pomacentrus. The yellow morph was associated with predominantly yellow damselfishes (P. moluccensis and P. amboinensis) and the brown morph with darker coloured species (P. adelus and P. chrysurus). Multiple-choice experiments in the laboratory revealed that: (1) each colour morph of P. fuscus preferentially selected habitat patches occupied by damselfishes with the same colouration; and (2) differences in microhabitat use between the two colour morphs of P. fuscus were due to the presence of different coloured damselfishes in these microhabitats. P. fuscus is a predator of newly recruited damselfishes and the striking resemblance between each morph of P. fuscus and the damselfish with which it was associated, suggests that aggressive mimicry may promote coexistence of P. fuscus colour morphs.Due to an error in the citation line, this revised PDF (published in December 2003) deviates from the printed version, and is the correct and authoritative version of the paper.     

Morphological variation in Bosmina longirostris (O.F. Müller, 1785) (Crustacea: Cladocera): consequence of cyclomorphosis or indication of cryptic species?

Samples of the polymorphic cladoceran Bosmina longirostris were collected in several lakes and ponds in Western Germany between 1993 and 1999. The shape of the first antennule was used as a qualitative criterion to classify 1241 individuals to three morphs (`longirostris', `cornuta', `pellucida') according to Lieder (Crustacea: Cladocera: Bosminidae. Stuttgart, Germany: Gustav. Fischer Verlag, 1996). Eight morphometric variables were subsequently measured in each individual to quantify size‐ and shape‐related features. Principal component and discriminant analyses were used to describe the morphological variation within and among the three morphs. Seasonal environmental changes modified several traits and thus, cyclomorphosis accounted for a high proportion of within‐morph variation. However, there was no gradual change from one morph to another. As well among‐morph variation was considerably greater than within‐morph variation. Consequently, the disagreement between the qualitative pre‐classification and the assignment basing on morphometric discriminant functions was low (4.6–12.0%). Considering that each morph is morphologically well‐defined, and that the different morphs coexisted over several generations in the same lake we conclude that they represent different species rather than variants of one polymorphic species. Therefore, we recognize beside the nominal taxon Bosmina longirostris ( 26 ) two more species: Bosmina cornuta ( 10 ) and Bosmina pellucida 30 .     

Parallel divergence of sympatric genetic and body size forms of Arctic charr,Salvelinus alpinus,from two Scottish lakes

F _ST and R_ST estimates for Arctic charr from six microsatelite markers collected from two neighbouring Scottish lakes, Loch Maree and Loch Stack, confirm the presence of two distinct genetic groupings representing separate populations within each lake. In both lakes, there was also a clear body size dimorphism, with large and small body size forms that segregated according to genetic grouping. There was evidence of only subtle foraging ecology differences between morphs, with the small body size morph in both lakes being more generalist in its foraging in the summer (consuming mostly plankton but also some macrobenthos) than the large body size morph, which specialized on planktonic prey. Trophic morphology (head and mouth shape) did not differ significantly between morphs (although the small sample size for Maree makes this a preliminary finding). Cluster analysis of the microsatelite data and the presence of private alleles showed that morphologically similar forms in different lakes were not genetically similar, as would be expected under a multiple invasion hypothesis. Thus, the data do not support a hypothesis of a dual invasion of both lakes by two common ancestors but instead suggest an independent origin of the two forms in each lake. Thus parallel sympatric divergence as a result of common selection pressures in both lakes is the most parsimonious explanation of the evolutionary origin of these polymorphisms. © 2008 The Linnean Society of London, Biological Journal of the Linnean Society, 2008, 95 , 748–757.     

The trade‐off between fecundity and egg size in a polymorphic population of Arctic charr (Salvelinus alpinus (L.)) in Skogsfjordvatn,subarctic Norway

Reproductive traits differ between intralacustrine Arctic charr morphs. Here, we examine three sympatric lacustrine Arctic charr morphs with respect to fecundity, egg size and spawning time/site to assess reproductive investments and trade‐offs, and possible fitness consequences. The littoral omnivore morph (LO‐morph) utilizes the upper water for feeding and reproduction and spawn early in October. The large profundal piscivore morph (PP‐morph) and the small profundal benthivore morph (PB‐morph) utilize the profundal habitat for feeding and reproduction and spawn in December and November, respectively. Females from all morphs were sampled for fecundity and egg‐size analysis. There were large differences between the morphs. The PB‐morph had the lowest fecundity (mean = 45, SD = 13) and smallest egg size (mean = 3.2 mm, SD = 0.32 mm). In contrast, the PP‐morph had the highest fecundity (mean = 859.5, SD = 462) and the largest egg size (mean = 4.5 mm, SD = 0.46 mm), whereas the LO‐morph had intermediate fecundity (mean = 580, SD = 225) and egg size (mean = 4.3, SD = 0.24 mm). Fecundity increased with increasing body size within each morph. This was not the case for egg size, which was independent of body sizes within morph. Different adaptations to feeding and habitat utilization have apparently led to a difference in the trade‐off between fecundity and egg size among the three different morphs.     

Convergent evolutionary processes driven by foraging opportunity in two sympatric morph pairs of Arctic charr with contrasting post‐glacial origins

The expression of two or more discrete phenotypes amongst individuals within a species (morphs) provides multiple modes upon which selection can act semi‐independently, and thus may be an important stage in speciation. In the present study, we compared two sympatric morph systems aiming to address hypotheses related to their evolutionary origin. Arctic charr in sympatry in Loch Tay, Scotland, exhibit one of two discrete, alternative body size phenotypes at maturity (large or small body size). Arctic charr in Loch Awe segregate into two temporally segregated spawning groups (breeding in either spring or autumn). Mitochondrial DNA restriction fragment length polymorphism analysis showed that the morph pairs in both lakes comprise separate gene pools, although segregation of the Loch Awe morphs is more subtle than that of Loch Tay. We conclude that the Loch Awe morphs diverged in situ (within the lake), whereas Loch Tay morphs most likely arose through multiple invasions by different ancestral groups that segregated before post‐glacial invasion (i.e. in allopatry). Both morph pairs showed clear trophic segregation between planktonic and benthic resources (measured by stable isotope analysis) but this was significantly less distinct in Loch Tay than in Loch Awe. By contrast, both inter‐morph morphological and life‐history differences were more subtle in Loch Awe than in Loch Tay. The strong ecological but relatively weak morphological and life‐history divergence of the in situ derived morphs compared to morphs with allopatric origins indicates a strong link between early ecological and subsequent genetic divergence of sympatric origin emerging species pairs. The emergence of parallel specialisms despite distinct genetic origins of these morph pairs suggests that the effect of available foraging opportunities may be at least as important as genetic origin in structuring sympatric divergence in post‐glacial fishes with high levels of phenotypic plasticity. © 2012 The Linnean Society of London, Biological Journal of the Linnean Society, 2012, ?? , ??–??.     

Skewed morph ratios lead to lower genetic diversity of the heterostylous Primula veris in fragmented grasslands

• Populations of heterostylous plant species are ideally composed of equal frequencies of two (distylous) or three (tristylous) morphologically different floral morphs. Intra-morph incompatibility helps to avoid inbreeding and to maintain genetic diversity, supporting plant fitness and long-term viability. Habitat fragmentation can lead to skewed morph ratios and thereby reduce the abundance of compatible mates. This, in turn, can result in a loss of genetic diversity. We tested whether the genetic diversity of heterostylous plants is affected by morph ratio bias using populations of the distylous grassland plant Primula veris in recently fragmented grasslands.
• We recorded morph frequencies and population sizes in 30 study populations of P. veris on two Estonian islands characterised by different degrees of habitat fragmentation. Examining variation of thousands of single nucleotide polymorphisms (SNPs) and heterostyly-specific genetic markers, we quantified overall and morph-specific genetic diversity and differentiation in these populations.
• Morph frequencies deviated more in smaller populations. Skewed morph ratios had a negative effect on the genetic diversity of P. veris in more fragmented grasslands. In the populations of better-connected grassland systems, genetic differentiation among S-morphs was higher than among L-morphs.
• Our study shows that deviations from morph balance are stronger in small populations and have a negative impact on the genetic diversity of the distylous plant P. veris. Together with the direct negative effects of habitat loss and decreased population size on the genetic diversity of plants, morph ratio bias may intensify the process of genetic erosion, thus exacerbating the local extinction of heterostylous species.

     

Phenotypic plasticity in the Antarctic fish <Emphasis Type="Italic">Trematomus newnesi</Emphasis> (Nototheniidae) from the South Shetland Islands

The presence of the two morphs, “typical” and “large mouth”, in the Antarctic fish species Trematomus newnesi (Perciformes, Notothenioidei) was recorded for the first time in nearshore waters of the South Shetland Islands (Potter Cove) and western Antarctic Peninsula (Petermann Island). The two morphs were distinguishable in specimens of 60–241 mm total length (TL); about 30% of the specimens constituted intermediate forms. In addition to the previously known characters separating the morphs, we found that the “relative size of the eye” can also be used to identify smaller and larger fish of the typical morph. The ecological significance of the two morphs remains unclear. Ratios of diagnostic characters for identification of the species at two size ranges (60–131 and 132–241 mm TL) are provided.     

Substrate preference in a colonial spider: is substrate choice affected by color morph?

Animals can show preference for a particular background as a way of decreasing visibility. Species with color polymorphism may have morph‐dependent background preference. I test this hypothesis on the orb‐weaving spider Parawixia bistriata. Adult females of P. bistriata present two distinct morphs characterized by brown and yellow opisthosomata. This nocturnal spider can be found in its retreat on the vegetation during the day. In order to examine whether females exhibit substrate preference dependent on their color morph, I first recorded the distribution of color morphs on different substrates (leaf and branch) and then performed a mark and release experiment. Field censuses indicated that the yellow morph was associated with leaves while the brown morph was found on either substrate type. The results of a mark and reciprocal release experiment agreed with the censuses and suggest that the two morphs differ in their association to substrate type: yellow females were associated with the leaf substrate, while brown females showed no association to a particular substrate type. Possible forces behind these differences in substrate choice are discussed.     

GENETIC AND MORPHOLOGICAL DIVERGENCE AMONG MORPHOTYPES OF THE ISOPOD EXCIROLANA ON THE TWO SIDES OF THE ISTHMUS OF PANAMA

Excirolana braziliensis is a dioecious marine isopod that lives in the high intertidal zone of sandy beaches on both sides of Central and South America. It possesses no larval stage and has only limited means of adult dispersal. Indirect estimates of gene flow have indicated that populations from each beach exchange less than one propagule per generation. Multivariate morphometrics have discovered three morphs of this species in Panama, two of them closely related and found on opposite sides of Central America (“C morph” in the Caribbean and “C′ morph” in the eastern Pacific), the third found predominantly in the eastern Pacific (“P morph”). Though the P and C′ morphs are seldom found on the same beach, they have overlapping latitudinal ranges in the eastern Pacific. A related species, Excirolana chamensis, has been described from the Pacific coast of Panama. Each beach contains populations that remain morphologically and genetically stable, but a single drastic change in both isozymes and morphology has been documented. We studied isozymes and multivariate morphology of 10 populations of E. braziliensis and of one population of E. chamensis. Our objective was to assess the degree of genetic and morphological variation, the correlation of divergence on these two levels of integration, the phylogenetic relationships between morphs, and the possible contributions of low vagility, low gene flow, and occasional extinction and recolonization to the genetic structuring of populations. Genetic distance between the P morph, on one hand, and the other two morphotypes of E. braziliensis, on the other, was as high as the distance between E. braziliensis and E. chamensis. Several lines of evidence agree that E. chamensis and the P morph had diverged from other morphs of E. braziliensis before the rise of the Panama Isthmus separated the C and C′ forms, and that the P morph constitutes a different species. A high degree of genetic differentiation also exists between populations of the same morph. On the isozyme level, every population can be differentiated from every other on the basis of at least one diagnostically different locus, regardless of geographical distance or morphological affiliation. Morphological and genetic distances between populations are highly correlated. However, despite the high degree of local variation, evolution of E. braziliensis as a whole has not been particularly rapid; divergence between the C and C′ morphs isolated for 3 million yr by the Isthmus of Panama is not high by the standard of within-morph differentiation or by comparison with other organisms similarly separated. Alleles that are common in one population may be absent from another of the same morph, yet they appear in a different morph in a separate ocean. The high degree of local differentiation, the exclusive occupation of a beach by one genotype with rare arrival of foreign individuals that cannot interbreed freely with the residents, the genetic stability of populations with infrequent complete replacement by another genetic population, and the sharing by morphs of polymorphisms that are not shared by local populations, all suggest a mode of evolution concentrated in rare episodes of extinction and recolonization, possibly coupled with exceptional events of gene flow that help preserve ancestral variability in both oceans.     

Colour morph related performance in the meadow grasshopper Chorthippus parallelus (Orthoptera,Acrididae)

Abstract 1. Polymorphism has been described for a number of herbivorous insects, but little is known about whether differences in body colour cause fitness differences. In Chorthippus parallelus, three main colour morphs occur, namely brown, green, and dorsally striped. 2. The present study examined colour morph abundances and morph‐related differences in body size, oviposition rate, and offspring numbers in females of C. parallelus collected in 15 montane grasslands. The study also examined the effect of plant species richness, composition, community productivity, and solar radiation on colour morph frequency and fitness. 3. The relative frequencies of the three colour morphs was 31.7% (brown), 33.1% (green), and 35.2% (dorsally striped), but the morphs were not evenly distributed across the 15 sites. 4. There was no effect of the habitat variables on the distribution of the green and the striped morph in the study sites, however 80% of the variation in the abundances of the brown morph was explained by plant species richness and composition. 5. Grasshopper size was equal among the morphs. Brown females laid significantly more egg pods than the green and dorsally striped morphs. There were no significant differences in offspring numbers among the colour morphs. 6. Body colour in C. parallelus seems to be a fitness‐relevant trait, raising the question of the evolutionary maintenance of polymorphism.     

Resource polymorphism and diversity of Arctic charr Salvelinus alpinus in a series of isolated lakes

Morphological, dietary and life‐history variation in Arctic charr Salvelinus alpinus were characterized from three geographically proximate, but isolated lakes and one large lake into which they drain in south‐western Alaska. Polymorphism was predicted to occur in the first three lakes because S. alpinus tend to become polymorphic in deep, isolated lakes with few co‐occurring species. Only one morph was evident in the large lake and two of the three isolated lakes. In the third isolated lake, Lower Tazimina Lake, small and large morphs were found, the latter including two forms differing in growth rate. The small morph additionally differed from the two large forms by having more gill rakers and a deeper body than same‐sized individuals of the large morph, consuming more limnetic and fewer benthic resources, having a greater gonado‐somatic index and maturing at a smaller size. The two large forms consumed only slightly different foods (more terrestrial insects were consumed by the medium‐growth form; more snails by the high‐growth form). Trends in consumption of resources with body shape also differed between lakes. Variability in life history of S. alpinus in these Alaskan lakes was as broad as that found elsewhere. This variability is important for understanding lake ecosystems of remote regions where this species is commonly dominant.     

Spatial analyses of two color polymorphisms in an alpine grasshopper reveal a role of small‐scale heterogeneity

Discrete color polymorphisms represent a fascinating aspect of intraspecific diversity. Color morph ratios often vary clinally, but in some cases, there are no marked clines and mixes of different morphs occur at appreciable frequencies in most populations. This poses the questions of how polymorphisms are maintained. We here study the spatial and temporal distribution of a very conspicuous color polymorphism in the club‐legged grasshopper Gomphocerus sibiricus. The species occurs in a green and a nongreen (predominately brown) morph, a green–brown polymorphism that is common among Orthopteran insects. We sampled color morph ratios at 42 sites across the alpine range of the species and related color morph ratios to local habitat parameters and climatic conditions. Green morphs occurred in both sexes, and their morph ratios were highly correlated among sites, suggesting shared control of the polymorphism in females and males. We found that in at least 40 of 42 sites green and brown morphs co‐occurred with proportions of green ranging from 0% to 70% with significant spatial heterogeneity. The proportion of green individuals tended to increase with decreasing summer and winter precipitations. Nongreen individuals can be further distinguished into brown and pied individuals, and again, this polymorphism is shared with other grasshopper species. We found pied individuals at all sites with proportions ranging from 3% to 75%, with slight, but significant variation between years. Pied morphs show a clinal increase in frequency from east to west and decreased with altitude and lower temperatures and were more common on grazed sites. The results suggest that both small‐scale and large‐scale spatial heterogeneity affects color morph ratios. The almost universal co‐occurrence of all three color morphs argues against strong effects of genetic drift. Instead, the data suggest that small‐scale migration–selection balance and/or local balancing selection maintain populations polymorphic.     

Adaptive loss of color polymorphism and character displacements in sympatric <Emphasis Type="Italic">Mnais</Emphasis> damselflies

A geographical survey of two Mnais damselfly species in the Kinki area of Japan showed evidence for character displacements when the two species were found in sympatry. Mnais costalis, a species that has polymorphic male mating types of orange-winged territorial and clear-winged non-territorial morphs (hereafter abbreviated to orange and clear morphs respectively) in allopatry often shifted to having monomorphic orange morphs in sympatry. The mean body size of orange morphs was consistently larger than that of clear morph in allopatry. The mean body size of the sympatric orange morphs was even larger than that of allopatric orange morphs. By contrast, Mnais pruinosa, a species that also has two morphs of large orange and small clear morphs in allopatry, shifted to having monomorphic clear morphs in sympatry. The mean body size of the sympatric clear morphs was smaller than that of allopatric clear morphs. Divergence was also detected in the preference for habitat insolation conditions between sympatric Mnais damselflies. Both species in allopatric regions prefer half-light forest habitats, while in sympatric regions they showed diversified habitat preference: M. costalis preferred sunny habitats while M. pruinosa preferred shady habitats. Multiple character displacements in signal traits and habitat preference emerged in heterogeneous forest light environments are likely to have synergistic effects on the reproductive isolation of the two species.     

Taxonomic status of the six-band morph of Cyphotilapia frontosa (Perciformes: Cichlidae) from Lake Tanganyika, Africa

Six- and seven-band morphs have been identified in a cichlid, Cyphotilapia frontosa, that is endemic to Lake Tanganyika. These color morphs have allopatric distributions; the six-band morph is widespread in the northern half of the lake while the seven-band morph is restricted to Kigoma on the east coast of the lake. Because no specimens of the seven-band morph have been available for taxonomic study except for the holotype of C. frontosa, the taxonomic status of these morphs has not been discussed. In a recent survey at the lake, 21 specimens of the seven-band morph were collected. A comparison of these with existing collection specimens of the six-band morph showed significant differences in morphometric and meristic characters; however, because all characters largely overlapped between these morphs, they are regarded as conspecific.     

A Comparative Study of Two Corbicula Morphs (Bivalvia,Corbiculidae) Inhabiting River Danube

In Europe, at least two Corbicula morphs (generally referred as C. fluminea and C. fluminalis) are known. However, their taxonomic and systematic position is controversial. In order to ascertain their taxonomy and systematics, we studied their characteristics in River Danube. Morphometric analysis confirmed the existence of two distinguishable morphs. Based on mtCOI sequence, morph‐1 is identical to C. fluminea, while morph‐2 is an undefined species. According to the median joining analysis, morph‐1 belongs to the C. fluminea clade, whereas morph‐2 forms a distinct clade, which is known only from invasive populations. Differences were recorded in population structure and reproductive biology, too. It is probably correct to treat the two morphs as distinct taxa, but as long as the African and western Asian parts of the genus’ range are poorly studied, taxonomic position of morph‐2 remains unclear. (© 2011 WILEY‐VCH Verlag GmbH & Co. KGaA, Weinheim)     

Body size evolution of a shell‐brooding cichlid fish from Lake Tanganyika

The substrate‐brooding cichlid fish Telmatochromis temporalis in Lake Tanganyika demonstrates a simple example of ecological speciation between normal and dwarf morphs through divergent natural selection on body size. The dwarf morph most likely evolved from the ancestral normal morph; therefore, elucidating the evolution of its small body size is a key to understanding this ecological speciation event. Previous studies suggest that the small body size of the dwarf morph is an adaptation to the use of empty snail shells as shelters (males) and spawning sites (females), but this idea has not been fully evaluated. Combining original and previously published information, this study compared likelihood values to determine the primary factor that would be responsible for regulating the body size of the dwarf morph. Male body size is most likely regulated by the ability to turn within shells, which may influence the predation avoidance of adult fish. Females are smaller than males, and their body size is most likely regulated by the ability to lay eggs in the small spaces within shells close to the shell apices where predation risk on eggs is lower. This study provides new evidence supporting the hypothesis that different natural selection factors affected body size of the different sexes of the dwarf morph, which has not been reported in other animal species.