Fitness consequences of larval traits persist across the metamorphic boundary

Metamorphosis is thought to provide an adaptive decoupling between traits specialized for each life-history stage in species with complex life cycles. However, an increasing number of studies are finding that larval traits can carry-over to influence postmetamorphic performance, suggesting that these life-history stages may not be free to evolve independently of each other. We used a phenotypic selection framework to compare the relative and interactive effects of larval size, time to hatching, and time to settlement on postmetamorphic survival and growth in a marine invertebrate, Styela plicata. Time to hatching was the only larval trait found to be under directional selection, individuals that took more time to hatch into larvae survived better after metamorphosis but grew more slowly. Nonlinear selection was found to act on multivariate trait combinations, once again acting in opposite directions for selection acting via survival and growth. Individuals with above average values of larval traits were most likely to survive, but surviving individuals with intermediate larval traits grew to the largest size. These results demonstrate that larval traits can have multiple, complex fitness consequences that persist across the metamorphic boundary; and thus postmetamorphic selection pressures may constrain the evolution of larval traits.     

An integrated analysis of phenotypic selection on insect body size and development time

Most studies of phenotypic selection do not estimate selection or fitness surfaces for multiple components of fitness within a unified statistical framework. This makes it difficult or impossible to assess how selection operates on traits through variation in multiple components of fitness. We describe a new generation of aster models that can evaluate phenotypic selection by accounting for timing of life‐history transitions and their effect on population growth rate, in addition to survival and reproductive output. We use this approach to estimate selection on body size and development time for a field population of the herbivorous insect, Manduca sexta (Lepidoptera: Sphingidae). Estimated fitness surfaces revealed strong and significant directional selection favoring both larger adult size (via effects on egg counts) and more rapid rates of early larval development (via effects on larval survival). Incorporating the timing of reproduction and its influence on population growth rate into the analysis resulted in larger values for size in early larval development at which fitness is maximized, and weaker selection on size in early larval development. These results illustrate how the interplay of different components of fitness can influence selection on size and development time. This integrated modeling framework can be readily applied to studies of phenotypic selection via multiple fitness components in other systems.     

Amphibian lipid levels at metamorphosis correlate to post-metamorphic terrestrial survival

In organisms that have complex life cycles, factors in the larval environment may affect both larval and adult traits. For amphibians, the postmetamorphic transition from the aquatic environment to terrestrial habitat may be a period of high juvenile mortality. We hypothesized that lipid stores at metamorphosis may affect an animal’s success during this critical transition period. We examined variation in total lipid levels among years and sites in recently metamorphosed individuals of two pond-breeding salamander species, the marbled salamander (Ambystoma opacum) and the mole salamander (A. talpoideum), with limited data for one anuran species (southern leopard frog, Rana sphenocephala). Lipid levels were allometrically related to body size and ranged from 1.9 to 23.8% of body dry mass. The two salamander species differed in lipid allocation patterns, with A. opacum apportioning a higher percentage of total lipid reserves into fat bodies than A. talpoideum. Species differences in lipid allocation patterns may primarily reflect that large metamorphs will mature as one-year olds, and, regardless of species, will alter lipid compartmentalization accordingly. We used mark–recapture data obtained at drift fences encircling breeding ponds for 13 A. opacum cohorts to estimate the proportion of postmetamorphic individuals that survived to breed (age 1–4) and the mean age at first reproduction. Regression models indicated that size-corrected lipid level at metamorphosis (i.e., lipid residuals), and to a lesser extent rainfall following metamorphosis, was positively related to adult survival. Snout-vent length at metamorphosis was negatively related to age at first reproduction. We suggest that lipid stores at metamorphosis are vital to juvenile survival in the months following the transition from aquatic to terrestrial habitat, and that a trade-off shaped by postmetamorphic selection in the terrestrial habitat exists between allocation to energy stores versus structural growth in the larval environment.     

Fitness consequences of the timing of metamorphosis in a freshwater crustacean

Metamorphosis is a common life-cycle transition in organisms as diverse as amphibians, insects, fishes and crustaceans, and the timing of this transition often affects an individual's fitness. Here, we measured age and size at metamorphosis in laboratory-reared individuals of the freshwater copepod, Diaptomus leptopus , and then followed individuals over their entire life cycle to assess the fitness consequences of variation in age and size at metamorphosis. In 3 separate experiments, individuals were raised in different food conditions: low food (0.2 μg C/ml) switched to high food (0.7 μg C/ml), or high food switched to low food, at several different larval and juvenile stages. Control individuals were reared on high or low food concentrations over their entire life cycles. For each individual, we measured age and size at metamorphosis and age and size at maturity; for females, we also measured total lifetime egg production, longevity, and calculated a composite fitness measure, λ. Statistical analyses showed no significant effects of age or size at metamorphosis on these same traits measured at maturity, or on the fitness components we estimated. The first individuals to mature had the highest total egg production and individual fitness; differences in body size at maturation explained none of the variation observed in fitness components. Our results show that metamorphosis was uncoupled from maturity and from fitness components by growth and development achieved during the juvenile phase of the life cycle, and support the conclusion that fitness consequences of metamorphosis depend fundamentally on the organization of an organism's life cycle. They also suggest that body size plays a different life-history role in these organisms than is recognized in most poikilotherms, and suggest the hypothesis, based on laboratory experiments, that selection may act primarily on juvenile developmental rates in field populations.     

Sequential predator effects across three life stages of the African tree frog, <Emphasis Type="Italic">Hyperolius spinigularis</Emphasis>

While theoretical studies of the timing of key switch points in complex life cycles such as hatching and metamorphosis have stressed the importance of considering multiple stages, most empirical work has focused on a single life stage. However, the relationship between the fitness components of different life stages may be complex. Ontogenetic switch points such as hatching and metamorphosis do not represent new beginnings—carryover effects across stages can arise when environmental effects on the density and/or traits of early ontogenetic stages subsequently alter mortality or growth in later stages. In this study, I examine the effects of egg- and larval-stage predators on larval performance, size at metamorphosis, and post-metamorphic predation in the African tree frog Hyperolius spinigularis. I monitored the density and survival of arboreal H. spinigularis clutches in the field to estimate how much egg-stage predation reduced the input of tadpoles into the pond. I then conducted experiments to determine: (1) how reductions in initial larval density due to egg predators affect larval survival and mass and age at metamorphosis in the presence and absence of aquatic larval predators, dragonfly larvae, and (2) how differences in mass or age at metamorphosis arising from predation in the embryonic and larval environments affect encounters with post-metamorphic predators, fishing spiders. Reduction in larval densities due to egg predation tended to increase per capita larval survival, decrease larval duration and increase mass at metamorphosis. Larval predators decreased larval survival and had density-dependent effects on larval duration and mass at metamorphosis. The combined effects of embryonic and larval-stage predators increased mass at metamorphosis of survivors by 91%. Larger mass at metamorphosis may have immediate fitness benefits, as larger metamorphs had higher survival in encounters with fishing spiders. Thus, the effects of predators early in ontogeny can alter predation risk even two life stages later.     

Carry‐over Effects of Size at Metamorphosis in Red‐eyed Treefrogs: Higher Survival but Slower Growth of Larger Metamorphs

Most animals have complex life histories, composed of a series of ecologically distinct stages, and the transitions between stages are often plastic. Anurans are models for research on complex life cycles. Many species exhibit plastic timing of and size at metamorphosis, due to both environmental constraints on larval growth and development and adaptive plastic responses to environmental variation. Models predicting optimal timing of metamorphosis balance cost/benefit ratios across stages, assuming that size affects growth and mortality rates in each stage. Much research has documented such effects in the larval period, but we lack an equal understanding of juvenile growth and mortality. Here, we examine how variation in size at metamorphosis in the Neotropical red‐eyed treefrog, Agalychnis callidryas, affects post‐metamorphic growth, foraging, and behavior in the lab as well as growth and survival in the field. Surprisingly, many individuals lost mass for weeks after metamorphosis. In the lab, larger metamorphs lost more mass following metamorphosis, ate similar amounts, had lower food conversion efficiencies, and grew more slowly after mass loss ceased than did smaller ones. In field cages larger metamorphs were more likely to survive than smaller ones; just one froglet died in the lab. Our data suggest that size‐specific differences in physiology and behavior influence these trends. Comparing across species and studies, large size at metamorphosis generally confers higher survival; size effects on growth rates vary substantially among species, in both magnitude and direction, but may be stronger in the tropics.     

Genetic polymorphism and trade-offs in the early life-history strategy of the Pacific oyster, Crassostrea gigas (Thunberg, 1795): a quantitative genetic study

We investigated genetic variability and genetic correlations in early life-history traits of Crassostrea gigas. Larval survival, larval development rate, size at settlement and metamorphosis success were found to be substantially heritable, whereas larval growth rate and juvenile traits were not. We identified a strong positive genetic correlation between larval development rate and size at settlement, and argue that selection could optimize both age and size at settlement. However, trade-offs, resulting in costs of metamorphosing early and large, were suggested by negative genetic correlations or covariances between larval development rate/size at settlement and both metamorphosis success and juvenile survival. Moreover, size advantage at settlement disappeared with time during the juvenile stage. Finally, we observed no genetic correlations between larval and juvenile stages, implying genetic independence of life-history traits between life-stages. We suggest two possible scenarios for the maintenance of genetic polymorphism in the early life-history strategy of C. gigas.     

GENETIC VARIATION FOR PHENOTYPIC PLASTICITY IN THE LARVAL LIFE HISTORY OF SPADEFOOT TOADS (SCAPHIOPUS COUCHII)

Phenotypic plasticity in life-history traits is common. The relationship between phenotype and environment, or reaction norm, associated with life-history plasticity can evolve by natural selection if there is genetic variation within a population for the reaction norm and if the traits involved affect fitness. As with other traits, selection on plasticity in a particular trait or in response to a particular environmental factor may be constrained by trade-offs with other traits that affect fitness. In this paper, I experimentally evaluated broad-sense genetic variation in the reaction norms of age and size at metamorphosis in response to two environmental factors, food level and temperature. Differences among full-sib families in one or both traits were evident in all treatments. However, variation among families in their responses to each treatment (genotype-environment interaction) resulted in variation among treatments in estimated heritabilities and genetic correlations. Age at metamorphosis was equally sensitive to temperature in all families, but size at metamorphosis was more sensitive to temperature in some families than in others. Size at metamorphosis was equally sensitive to food level in all families, but age at metamorphosis was sensitive to food in some families but not in others. At high temperature or low food, the genetic correlation between age and size at metamorphosis was positive, generating a potential trade-off between metamorphosing early to attain higher larval survival and metamorphosing later to achieve larger size. This trade-off extends across treatments: families with the largest average size at metamorphosis achieved larger size with the longest average and greatest plasticity in age at metamorphosis. Other families achieved shorter average larval periods by exhibiting greater plasticity in size at metamorphosis but had the smallest average size at metamorphosis. This trade-off may reflect an underlying functional constraint on the ability to respond optimally to all environments, resulting in persistent genetic variation in reaction norms.     

HYBRIDIZATION IN LEOPARD FROGS (RANA PIPIENS COMPLEX): LARVAL FITNESS COMPONENTS IN SINGLE-GENOTYPE POPULATIONS AND MIXTURES

Recognizing the predominant mode of selection in hybrid systems is important in predicting the evolutionary fate of recombinant genotypes. Natural selection is endogenous if hybrid genotypes are at a disadvantage relative to parental species independent of environment. Alternatively, relative fitness can vary in response to environmental variation (exogenous selection), and hybrid genotypes can possess fitness values equal to or greater than that of parental species. I investigated the nature of natural selection in a leopard frog hybrid system by rearing larvae of hybrid and parental genotypes between Rana blairi and R. sphenocephala in 1000-L outdoor experimental ponds. Three hybrid (F₁, backcrossj [B₁], backcross₂ [B₂]) and two parental (R. blairi [BB] and R. sphenocephala [SS]) larval genotypes were produced by artificial fertilzations using adult frogs from a natural population in central Missouri. Resultant larvae were reared in single-genotype populations and two-way mixtures at equal total numbers from hatching to metamorphosis. In single-genotype ponds, F₁ hybrid larvae had highest survival and BB were largest at metamorphosis. When F₁ and SS larvae were mixed together, F₁ hybrids had reduced survival and both F₁ and SS larvae metamorphosed at larger body masses than when reared separately. When mixed, both B₁ and SS larvae had shorter larval period lengths than when reared alone. Higher proportion of B₁ metamorphs were produced when larvae were mixed with either parental species than when reared alone. Larval fitness components as measured by survival, body mass at metamorphosis, proportion of survivors metamorphosing, and larval period length for B₂ hybrid and BB larvae were similar in single-genotype populations and mixtures. Comparison of composite fitness component estimates indicated hybrid genotypes possess equivalent or higher larval fitness relative to both parental species for the life-history fitness components measured. Despite reduced survival of F₁ hybrids in mixtures, backcross-generation hybrid genotypes demonstrated high levels of larval growth, survival, and metamorphosis in mixtures with parental species. Consequently, this study suggests natural hybridization and subsequent backcrossing between R. blairi and R. sphenocephala can produce novel and relatively fit hybrid genotypes capable of successful existence with parental species larvae. Thus, the evolutionary fate of hybrid and parental genotypes in this system may be influenced by exogenous selection mediated by genotypic composition of larval assemblages.     

Heterochrony during skeletal development of Pseudis platensis (anura,hylidae) and the early offset of skeleton development and growth

The aquatic frog Pseudis platensis has a giant tadpole, long developmental time, and dissociated metamorphic events that include later offset of larval somatic morphologies. Moreover, when the tadpole metamorphoses, the young frog is nearly the size of an adult, suggesting that this species has low rates of postmetamorphic growth. Herein, we study the development of the skeleton during larval development up to the end of metamorphosis, which is denoted by the complete lost of the tail in P. platensis. Our study revealed heterochronic differences in skeletal development compared with that of most anurans; these involve the complete differentiation of skull bones and the extensive ossification of the postcranial skeleton before completion of metamorphosis. The skull of metamorphosing P. platensis has an ossified sphenethmoid and a fully formed plectral apparatus, thus differing with regard to the pattern observed in most anurans in which both developmental events take place during the postmetamorphic life. Despite the fact that the iliosacral articulation and the urostyle are present at the end of metamorphosis as in most anurans, ossification/calcification of carpus, tarsus, and limb epihyses during metamorphosis of P. platensis suggests that the postcranial skeleton lacks postmetamorphic growth. This study also includes a discussion of the pattern of development of the plectral apparatus, which allows us to propose a new hypothesis regarding pars externa plectri homology. J. Morphol., 2009. © 2008 Wiley‐Liss, Inc.     

Warmer winters reduce frog fecundity and shift breeding phenology,which consequently alters larval development and metamorphic timing

One widely documented phenological response to climate change is the earlier occurrence of spring‐breeding events. While such climate change‐driven shifts in phenology are common, their consequences for individuals and populations have rarely been investigated. I addressed this gap in our knowledge by using a multi‐year observational study of six wood frog (Rana sylvatica) populations near the southern edge of their range. I tested first if winter temperature or precipitation affected the date of breeding and female fecundity, and second if timing of breeding affected subsequent larval development rate, mass at metamorphosis, date of metamorphosis, and survival. Warmer winters were associated with earlier breeding but reduced female fecundity. Winter precipitation did not affect breeding date, but was positively associated with female fecundity. There was no association between earlier breeding and larval survival or mass at metamorphosis, but earlier breeding was associated with delayed larval development. The delay in larval development was explained through a counterintuitive correlation between breeding date and temperature during larval development. Warmer winters led to earlier breeding, which in turn was associated with cooler post‐breeding temperatures that slowed larval development. The delay in larval development did not fully compensate for the earlier breeding, such that for every 2 days earlier that breeding took place, the average date of metamorphosis was 1 day earlier. Other studies have found that earlier metamorphosis is associated with increased postmetamorphic growth and survival, suggesting that earlier breeding has beneficial effects on wood frog populations.     

Influence of density and salinity on larval development of salt‐adapted and salt‐naïve frog populations

Environmental change and habitat fragmentation will affect population densities for many species. For those species that have locally adapted to persist in changed or stressful habitats, it is uncertain how density dependence will affect adaptive responses. Anurans (frogs and toads) are typically freshwater organisms, but some coastal populations of green treefrogs (Hyla cinerea) have adapted to brackish, coastal wetlands. Tadpoles from coastal populations metamorphose sooner and demonstrate faster growth rates than inland populations when reared solitarily. Although saltwater exposure has adaptively reduced the duration of the larval period for coastal populations, increases in densities during larval development typically increase time to metamorphosis and reduce rates of growth and survival. We test how combined stressors of density and salinity affect larval development between salt‐adapted (“coastal”) and nonsalt‐adapted (“inland”) populations by measuring various developmental and metamorphic phenotypes. We found that increased tadpole density strongly affected coastal and inland tadpole populations similarly. In high‐density treatments, both coastal and inland populations had reduced growth rates, greater exponential decay of growth, a smaller size at metamorphosis, took longer to reach metamorphosis, and had lower survivorship at metamorphosis. Salinity only exaggerated the effects of density on the time to reach metamorphosis and exponential decay of growth. Location of origin affected length at metamorphosis, with coastal tadpoles metamorphosing slightly longer than inland tadpoles across densities and salinities. These findings confirm that density has a strong and central influence on larval development even across divergent populations and habitat types and may mitigate the expression (and therefore detection) of locally adapted phenotypes.     

The impact of larval predators and competitors on the morphology and fitness of juvenile treefrogs

Studies of phenotypic plasticity typically focus on traits in single ontogenetic stages. However, plastic responses can be induced in multiple ontogenetic stages and traits induced early in ontogeny may have lasting effects. We examined how gray treefrog larvae altered their morphology in four different larval environments and whether different larval environments affected the survival, growth, development, and morphology of juvenile frogs at metamorphosis. We then reared these juveniles in terrestrial environments under high and low intraspecific competition to determine whether the initial differences in traits at metamorphosis affected subsequent survival and growth, whether the initial phenotypic differences converged over time, and whether competition in the terrestrial environment induced further phenotypic changes. Larval and juvenile environments both affected treefrog traits. Larval predators induced relatively deep tail fins and short bodies, but there was no impact on larval development. In contrast, larval competitors induced relatively short tails and long bodies, reduced larval growth, and slowed larval development. At metamorphosis, larval predators had no effect on juvenile growth or relative morphology while larval competitors produced juveniles that were smaller and possessed relatively shorter limbs and shorter bodies. After 1 month of terrestrial competition among the juvenile frogs, the initial differences in juvenile morphology did not converge. There were no differences in growth due to larval treatment but there were differences in survival. Individuals that experienced low competition as tadpoles experienced near perfect survival as juvenile frogs but individuals that experienced high competition as tadpoles suffered an 18% decrease in survival as juvenile frogs. There were also morphological responses to juvenile competition, but these changes appear to be due, at least in part, to allometric effects. Collectively, these results demonstrate that larval environments can have profound impacts on the traits and fitness of organisms later in ontogeny.     

Carry-over effects of the larval environment on post-metamorphic performance in two hylid frogs

Life history theory and empirical studies suggest that large size or earlier metamorphosis are suitable proxies for increased lifetime fitness. Thus, across a gradient of larval habitat quality, individuals with similar phenotypes for these traits should exhibit similar post-metamorphic performance. Here we examine this paradigm by testing for differences in post-metamorphic growth and survival independent of metamorphic size in a temperate (spring peeper, Pseudacris crucifer) and tropical (red-eyed treefrog, Agalychnis callidryas) anuran reared under differing larval conditions. For spring peepers, increased food in the larval environment increased post-metamorphic growth efficiency more than predicted by metamorphic phenotype and led to increased mass. Similarly, red-eyed treefrogs reared at low larval density ended the experiment at a higher mass than predicted by metamorphic phenotype. These results show that larval environments can have delayed effects not captured by examining only metamorphic phenotype. These delayed effects for the larval environment link larval and juvenile life history stages and could be important in the population dynamics of organisms with complex life cycles.     

Phenotypic plasticity and optimal timing of metamorphosis under uncertain time constraints

Life-history theory suggests that optimal timing of metamorphosis should depend on growth conditions and time constraints under which individuals develop. Current models cannot make reliable predictions for species in ephemeral habitats where individuals often face an increasing mortality risk over time because these models assume time-invariant mortality rates (i.e., daily mortality rates remain constant) and fixed seasons. We examined the plasticity of growth, development, and body mass at metamorphosis in tadpoles of the tree-hole breeding frog Phrynobatrachus guineensis in relation to an unpredictable time constraint in the field and in controlled experiments along a fixed density and food gradient. Mean mass and age at metamorphosis of sibships were positively correlated with per capita food level. Based on our results, we developed a simple model of the optimal timing of metamorphosis under time-dependent mortality rates showing that development rates are not only adjusted to growth conditions but also to time-variant mortality rates. The increasing mortality rate represents a time constraint that favors a reduced larval period, but because it is based on probabilities of survival it allows a trade-off between development time and mass. We extend this model to different types of time constraints and show that it can predict the range of documented reaction norms. Differences between species in␣the correlation of age and mass at metamorphosis may have evolved due to differences in their time-variant mortality rates.     

Effects of induced variation in anuran larval development on postmetamorphic energy reserves and locomotion

Anuran larvae exhibit high levels of phenotypic plasticity in growth and developmental rates in response to variation in temperature and food availability. We tested the hypothesis that alteration of developmental pathways during the aquatic larval stage should affect the postmetamorphic performance of the Iberian painted frog (Discoglossus galganoi). We exposed tadpoles to different temperatures and food types (animal- vs. plant-based diets) to induce variation in the length of the larval period and body size at metamorphosis. In this species, larval period varied with temperature but was unaffected by diet composition. In contrast, size at metamorphosis was shaped by the interaction between food quality and temperature; tadpoles fed on an animal-based diet became bulkier metamorphs than those fed on plant-based food at high (22°C) but not at low (12°C) temperature. Body condition of newly metamorphosed frogs was unrelated to the temperature or food type experienced during the premetamorphic stage. Frogs maintained at high temperature during the larval period showed reduced jumping ability, especially when fed on the plant-based diet. However, when considering size-independent jumping ability, cold-reared individuals exhibited the lowest performance, and herbivores reared at 17°C the highest. Cold-reared (12°C) frogs accumulated larger amounts of energy reserves than individuals raised at 17°C or 22°C. This was still the case after correction for differences in body mass, thus indicating some size-independent effect of developmental temperature. Despite the higher lipid content of the carnivorous diet, the differences in energy reserves between herbivores and carnivores were relatively weak and associated with differences in body size. These results suggest that the consequences of environmental variation in the larval habitat can extend to the terrestrial phase and influence juvenile growth and survival.     

GENETIC CORRELATIONS AMONG MORPHOMETRIC TRAITS AND RATES OF GROWTH AND DIFFERENTIATION IN THE GREEN TREE FROG,HYLA CINEREA

It is often proposed that the morphometric shape of animals often evolves as a correlated response to selection on life-history traits such as whole-body growth and differentiation rates. However, there exists little empirical information on whether selection on rates of growth or differentiation in animals could generate correlated response in morphometric shape beyond that owing to the correlation between these rates and body size. In this study genetic correlations were estimated among growth rate, differentiation rate, and body-size-adjusted head width in the green tree frog, Hyla cinerea. Head width was adjusted for size by using the residuals from log-log regressions of head width on snout-vent length. Size-adjusted head width at metamorphosis was positively genetically correlated with larval period length. Thus, size-independent shape might evolve as a correlated response to selection on a larval life-history trait. Larval growth rate was not significantly genetically correlated with size-adjusted head width. An additional morphometric trait, size-adjusted tibiofibula length, had a nonnormal distribution of breeding values, and so was not included in the analysis of genetic correlations (offspring from one sire had unusually short legs). This result is interesting because, although using genetic covariance matrices to predict long-term multivariate response to selection depends on the assumption that all loci follow a multivariate Gaussian distribution of allelic effects, few data are available on the distribution of breeding values for traits in wild populations. Size at metamorphosis was positively genetically correlated with larval period and larval growth rate. Quickly growing larvae that delay metamorphosis therefore emerge at a large size. The genetic correlation between larval growth rate and juvenile (postmetamorphic) growth rate was near zero. Growth rate may therefore be an example of a fitness-related trait that is free to evolve in one stage of a complex life cycle without pleiotropic constraints on the same trait expressed in the other stage.     

Larval habitat duration and size at metamorphosis in frogs

The relationship between size at metamorphosis and adult size was studied in 12 closely-related species of frog from Malawi (Central Africa). These species of frogs breed in water of different durations, and occupy different habitats as adults. We could demonstrate no correlation between size at metamorphosis and size of adults when frogs were divided into groups on the basis of occupying similar habitats as adults, but when frogs were divided into groups on the basis of similar duration of larval habitat we demonstrated a strong correlation between size at metamorphosis and adult size. Thus we suggest that duration of the larval habitat is a major determinant of size at metamorphosis, with species which breed in the more temporary habitats metamorphosing at smaller size than species which breed in more permanent habitats, but which are of similar size as adults. Such manipulation of the life cycle appears to be adaptive since it results in individuals becoming independent of water earlier when the likelyhood of early loss of larval habitat is high.     

An extraordinarily long larval duration of 4.5 years from hatching to metamorphosis for teleplanic veligers of Fusitriton oregonensis

Veliger larvae of the NE Pacific snail Fusitriton oregonensis were reared in culture for 4.5 to 4.6 years from hatching to metamorphosis and through postlarval growth to reproduction. Larval shells grew in length from 0.20 to 3.9 mm. Late veligers grew slowly, but shell sizes increased even in the 4th and 5th years. Widths of larval shells at late stages equaled or exceeded those of the protoconchs of two juveniles from the field. Cultured larvae did not metamorphose until presented with subtidal rocks and associated biota. There was no indication of larval senescence: the first 2 years of postmetamorphic shell growth were slightly faster, and time from metamorphosis to first reproduction (3.3 years) was slightly less than for an individual that had developed to metamorphic competence in the plankton. A 4.5-year larval phase exceeds previous estimates for teleplanic larval durations and greatly exceeds estimates of the time for transport across oceans. This extraordinarily long larval period may exceed the usual duration in nature but shows that larval periods can be much longer than previously suspected without complete stasis in growth and with little if any loss of viability.     

The evolution of metamorphosis in temnospondyls

Temnospondyls, possible relatives of extant amphibians and crudely similar to recent salamanders, are known from larval, neotenic and metamorphosed stages. Here, ontogenetic data of various temnospondyl taxa are analysed in order to recognize metamorphosis. Here, metamorphosis is strictly defined as a shift from an aquatic to a terrestrial existence. Following a check-list of criteria, the most likely metamorphosis-induced changes are proved in three temnospondyl lineages: eryopids, zatrachydids and dissorophoids. In a few other, unrelated taxa, terrestrial adults are known but no larval or metamorphosing forms. The distribution of metamorphosis among the Temnospondyli does not strictly correlate with phylogeny, which highlights the widespread occurrence of neoteny. In each group, characteristic patterns of metamorphosis are described and compared. Among temnospondyls, dissorophoids had the most intensive type of metamorphosis, characterized by a condensed ontogeny and a relatively small body size. The result was a distinct transformed morphotype with far-reaching terrestrial adaptations.