Brooding and development of <Emphasis Type="Italic">Anasterias minuta</Emphasis> (Asteroidea: Forcipulata) in Patagonia,Argentina

Brooding, embryonic and larval development, and the influence of environmental and biological factors in tidepool habitats were studied in the sea star, Anasterias minuta, at various sites along ~220 km of the Patagonian coast. This species has a benthic, lecithotrophic development that includes eight distinct developmental stages. A larval organ, the connection cord, is developed from a small preoral lobe at early stages of development and becomes larger and thinner at advanced stages. Fecundity and average egg size increased with female body size. The regression of log egg number to log sea-star size and weight at different sites had a slope significantly less than 3.0, resulting in negative allometry and indicating that brood capacity was limited in large females. Development was generally synchronous among sites, but varied within each brood at advanced stages, with more developed brooded larvae located at the periphery of the brood mass. Brooding was synchronous among various populations at different years and spatial scales, and extended over a period of 8 months. The highest proportion of brooding females occurred during May and June (austral winter). Juveniles were released mainly during September. The likelihood of finding brooding sea stars decreased with increasing sea water temperature, tidal height, and wave exposure, and increased with increasing body size. Both body size of brooding females and brooding rate were higher in the infralittoral fringe than at midlittoral levels. A revision of the current model of brooding behavior and development among forcipulate sea stars is given.     

Apparent resource competition among embryos in the brood pouch of a male pipefish

Embryo success was studied in the paternally brooding pipefish Syngnathus typhle. During brooding, which lasts about a month, males provide embryos in their brood pouch with nutrients and oxygen via a placenta-like structure. Egg size depends on female size. In aquaria, males were mated with differently sized females to give the following treatments: M, mixed-egg-size broods of approximately half large and half small eggs; L, single-egg-size broods of large eggs; S, single-egg-size broods of small eggs; and F, field mated males. All males were kept in aquaria for a full brooding period. For each egg-size category, the number of newborn was compared with the number of eggs the male initially fertilized in his brood pouch. Within mixed-egg-size broods, a higher proportion of large eggs survived and large eggs resulted in heavier newborn than small eggs. Indeed, small eggs from a mixed-egg-size brood had significantly lower relative success (proportion of embryos surviving to birth) than those from a brood entirely composed of small eggs. The implication is that embryos compete for resources within the brood pouch, and that competitive success depends on egg size. Given that females produce eggs corresponding in size to their body size, and that females are known to compete indirectly for access to mates (i.e., the sex-roles are reversed), this intrabrood competition could be seen as an extension of female-female competition, but alternative explanations are discussed. Received: 28 April 1995/Accepted after revision: 28 October 1995     

Reproductive biology in two species of deep-sea squids

Deep-sea squids, Moroteuthis ingens and Gonatus antarcticus, were collected in the slope waters off the Falkland Islands and their reproductive systems preserved and investigated onshore. Changes in oocyte length-frequencies at maturation and spawning, and their fecundity were studied. These squids, as well as many other species, are characterised by a synchronous oocyte growth and ovulation. Oviducts are not used for ripe egg accumulation and consequently the universal scale of Lipinski (1979) cannot be applied to assign female maturity. M. ingens spawns near the bottom; its fecundity is 168–297 thousand eggs. Maximum egg size is 1.8–2.7 mm. G. antarcticus spawns midwater; its fecundity is 10–25 thousand eggs. Egg size is 3.2–3.3 mm. In M. ingens spawning takes place in the austral autumn and winter, in G. antarcticus—in austral winter. Our data and the literature data show that the so-called “synchronous ovulation” probably occurs in all deepwater squids. This pattern is very rare among fish, but is quite common among benthic octopods that brood their egg masses.     

Male limitation of female reproductive success in a pipefish: effects of body-size differences

Summary In the pipefish Syngnathus typhle, a species with exclusive male parental care, males limit female reproductive success because of their limited brood pouch space and long pregnancy. Sexual size dimorphism is absent in these 1-year-old animals but increases with age so that older females are larger than similarly aged males. Because fecundity is related to size in both sexes and increases more rapidly with body size in females than in males, the difference in growth increases female fecundity more, relative to male fecundity, as the fish get older. We therefore predicted that male limitation of female reproductive success is even more severe when all age classes are considered. To measure a female's maximum reproductive rate, she was provided with three males. Small 1-year-old females produced as many eggs, or produced eggs at the same rate, as a male of similar size could care for. Small females filled on average 1.06 males within the time span of one male pregnancy and actually produced on average 10 eggs fewer than needed to fill a similarly sized male. Large 2-year-old females, in contrast, produced on average a surplus of 149 eggs and filled 2.7 similarly sized males within the course of one pregnancy. The difference between females of the two size classes was highly significant. Males prefer to mate with larger females if given a choice. In nature sex ratios are equal, and males limit female reproductive success in the whole population. Therefore, small females are more severely constrained by mate availability than are larger females because males choose to mate with larger females. Offprint requests to: A Berglund     

Growth and reproduction in the Antarctic brooding bivalve Adacnarca nitens (Philobryidae) from the Ross Sea

We present information on the reproductive biology, population structure, and growth of the brooding Antarctic bivalve Adacnarca nitens Pelseneer 1903, from the Ross Sea, Antarctica. Individuals ranging from 0.85 to 6.00 mm were found attached to a hydrozoan colony. This species shows low fecundity and large egg size, common to other brooding species. The minimum size at which oogenesis was detected was 2.3 mm and the minimum size at which brooding was evident was 3.9 mm. Embryos of a full range of developmental stages were brooded simultaneously in females. The population showed a log–normal distribution and results suggest non-periodic reproduction with continuous embryonic development. The reproductive traits of A. nitens are discussed in the context of circum-Antarctic species distribution and limitations to dispersal in brooding benthic invertebrates.     

Reproductive biology of a deep-sea brittle star Amphiura carchara (Echinodermata: Ophiuroidea)

 New information on the brooding reproduction of Amphiura carchara, and previous studies of several deep-sea congeners, contradict the prevailing notion that direct development is exceptional among deep-sea echinoderms. Over 500 specimens of A. carchara from 2,850 to 4,100-m depths off the coast of California, USA, were studied. The species was found to be gonochoric, although most brooding species of ophiuroids are hermaphroditic. Females each brooded up to 72 embryos at a time, with up to 10 in a single bursa. The embryos of individual adults were often at different stages of development, but those brooded in a single bursa tended to be at the same stage. Thus, the species has characteristics that breach the distinctions between sequential and simultaneous brooding exhibited by shallow-water ophiuroids. The embryos developed from yolky eggs that are large for an ophiuroid, with the mean largest oocytes 0.45 mm, and ranging to 1.28 mm in diameter. Almost all late-stage embryos were positioned with their mouth and arms pressed against the wall of the bursa, possibly to facilitate the uptake of nutrients from the parent. Thus, A. carchara may be matrotrophic. The largest embryos examined had a rudimentary disk skeleton, and arms with four joints and a terminal plate. Emerging juveniles probably differ in disk diameter. The species appears to brood year round, although differences in gonad size, the incidence of brooding, and the relative numbers of early developmental stages in summer and winter samples indicated that there are seasonal trends in reproduction. Received: 14 February 2000 / Accepted: 3 July 2000     

Long-distance dispersal of a subantarctic brooding bivalve (Gaimardia trapesina) by kelp-rafting

The probability of successful dispersal by sessile benthic invertebrates is thought to strongly influence their geographic distribution and population genetics. Generally, species with long-lived planktonic larvae are expected to exhibit wider distribution patterns than those species which brood their young, due to their presumably greater potential for dispersal. In some cases, however, brooding species exhibit broad distributions and show evidence of genetic exchange with geographically distant populations. One potential factor that has been invoked as an expianation is dispersal by floating and rafting of adults and egg masses. Several studies have shown that it is possible for sessile adults to disperse on the order of several to many thousand kilometers by rafting on debris in ocean currents. With very few exceptions, however, direct evidence of rafting in the open ocean has been lacking. We present evidence of long-distance (1300 to 2000 km) dispersal of a brooding pelecypod,Gaimardia trapesina (Lamarck, 1819), in the Southern Ocean in the vicinity of Cape Horn, the Falkland Islands, and the antarctic island South Georgia (54°S; 37°W). Data on survival and fecundity rates ofG. trapesina and the prevalence of kelp rafts collected during the austral winter of 1993 indicate that dispersal by rafting can occur over ecologically relevant time scales and could potentially serve as a significant means of genetic exchange between populations.     

Reproduction of the invasive slipper limpet, Crepidula fornicata, in the Bay of Brest, France

The reproduction of Crepidula fornicata was studied in the Bay of Brest in order to characterise the first step of the reproductive cycle of this invasive species. The survey was carried out from 2000 to 2003 and different parameters were measured, namely, the percentage of the different sexual stages, the straight length of the shell and the percentage of brooding females using a survey of the embryonic development and the fecundity. The juvenile frequency increases generally from mid-June or mid-August, depending on the year. In 2001 and 2003, a first peak was observed as early as May, but it was followed by a rapid disappearance of the individuals. The sex-ratio female/male increased from 0.22 to 0.46 between 2001 and 2003. The sex change between intermediates and females took place mainly in summer and was well marked in 2001 and 2003. The survey of the embryonic development in the egg capsules brooded by the females provided an annual phenology of the laying and hatching processes. The laying period extends from February to September with three to four major periods of egg-laying per year and corresponding hatching periods about 1 month later. Each female lays two to four times per year on average. The first egg-laying concerned fewer females than subsequent ones, except in 2003, and exhibited a higher fecundity. The annual mean of the number of eggs for each stage was not significantly different, thus indicating no significant mortality rate during embryonic development. For the C. fornicata population in the Bay of Brest, several reproductive characteristics tend to highlight its invasive capacity: (1) a long reproductive period, (2) reproduction in a ‘multi-trials’ process equivalent to a spreading out of the risks and (3) a relatively high fecundity.     

Maturity,fecundity, and reproductive cycle of the spotted ratfish, <Emphasis Type="Italic">Hydrolagus colliei</Emphasis>

Size at maturity, fecundity, and reproductive periodicity were estimated for the spotted ratfish, Hydrolagus colliei (Lay and Bennett, 1839), off the coast of California, Oregon, and Washington. Maximum body size and size at median maturity were greater for females than males. Skeletal muscle concentrations of the steroid hormones testosterone (T) and estradiol (E₂) predicted similar, but slightly smaller sizes at maturity than the morphological criterion. Stage of maturity for males was estimated identically using internal organs or external secondary sexual characters, thus allowing non-lethal maturity assessments. Size at median maturity was greater north of Point Conception for females, and north of Cape Mendocino for males. Peak parturition occurred from May to October, with increased concentrations of E₂ in skeletal muscle of females correlating with ovarian recrudescence during November to February. No significant seasonal trends in female T were apparent, but mean female 11-ketotestosterone (11KT) was 300% greater in April than any other month during the parturition season. There was a marginal evidence for increased number and size of ova with maternal size. Extrapolation of the hypothesized 6 to 8-month egg-laying season to observed mean parturition rates of captive specimens yielded an estimated annual fecundity of 19.5–28.9 egg cases. Differences in fecundity among higher taxonomic classifications of chondrichthyans were detected with chimaeriform fishes more fecund than lamniform, myliobatiform, squaliform, and rhinobatiform fishes, and less fecund than rajiform fishes. Electronic supplementary material  The online version of this article (doi:) contains supplementary material, which is available to authorized users.     

Maturation,fecundity, and spawning strategy of the thornback ray, <Emphasis Type="Italic">Raja clavata</Emphasis>: do reproductive characteristics vary regionally?

In Portuguese waters, thornback ray Raja clavata spawns mainly between May and January, although females and males in spawning condition are found throughout the year. The maturation process can be divided into three main phases by using information on gonad weight, oviducal gland and uterus width in females and on gonad weight, clasper length and sperm duct width in males. Females attain length-at-first-maturity at 784 mm and males at 676 mm, at ages of 7.5 and 5.8 years, respectively. In females larger than length-at-first-maturity, a resting stage was identified characterized by low gonadosomatic index and well-developed oviducal glands and uteri. These results along with the low proportion of adult females that are effectively reproducing per month demonstrated that the thornback ray cannot be considered a continuous spawner as described in other studies. Fecundity was determinate with about 35 eggs released per batch. During the spawning season, a total of four batch episodes occur indicating that the total fecundity was approximately 136 eggs per female. Regional differences may exist in the reproductive strategy of the species, namely on the duration of the spawning season, length-at-first-maturity, and fecundity, which can be related to a more intense fishing pressure in northern European waters.     

Costs of extended brood protection in the Antarctic sea star, Neosmilaster georgianus (Echinodermata: Asteroidea)

Changes in size, morphology, and biochemical composition in adults and embryos of a brooding sea star, Neosmilaster georgianus (Sladen), were studied in a population adjacent to Anvers Island, Antarctic Peninsula (64°46′S; 64°04′W) during the austral spring, 1991. Five morphological stages of development were designated in 24 broods, and for each the weight and biochemical composition of the brooding adults and their embryos were determined. Between Stage 1 and 2, the dry weight (dw) and organic weight (ow) of the embryo did not change. From Stage 2 to 3, the dw and ow increased significantly by 10%. Stage 2 and 3 embryos were in clusters of a few (2 to 10) to as many as 40 individuals. In the smaller clusters, individual embryos were attached by tissue cords to another, sometimes atrophied, brood member. In the larger clusters, they were attached to a central mass of tissue containing remnants of embryos. We interpret these interactions as a form of cannibalism which may account for the weight gains between Stage 2 and 3. During Stages 4 and 5, as juvenile form was approached, the dw and ash weight of the young increased significantly and the ow decreased significantly. The calculated energy content for the juvenile (Stage 5) was not significantly different from the energy content of the earliest undifferentiated stage (Stage 1), an indication that most of the organic matter in the egg is the primary contribution to the large juvenile. In brooding females, pyloric caeca indices declined by 52% from Stage 1 to Stage 5 and pyloric energy stores declined by 63% due to proportionately equivalent declines in protein and lipid. The ovary index was low and increased only slightly during brood protection, while the size of the largest oocytes remained approximately 23% that of ova. Energy stores in the pyloric caeca of brooding N. georgianus thus become depleted over a long period of incubation during which the adults apparently starve. This may delay oocyte development and ultimately limit the frequency of individual reproduction. Received: 15 September 1998 / Accepted: 9 March 1999     

Reproductive habits and life cycle of the small clam Kingiella chilenica (Bivalvia: Cyamiidae) in an estuarine sand flat from the south of Chile

A population of the small clam Kingiella chilenica Soot Ryen 1959 was studied from March 1986 to December 1988 in an intertidal flat at Queule River, in the south of Chile. The life cycle and life history pattern of the bivalve were established contrasting population structure and dynamics to its reproductive habits. Individuals are gonochoristic and semelparous, presenting a typical annual life cycle. The species is a sequential brooder whose embryos undergo direct development. After the brooding season (summer through autumn), the adults disappear gradually (autumn through winter). Recruited juveniles overwinter during a relatively long period, undergoing rapid growth during the spring to attain the adult stage during the summer. The number of brooded embryos increases in proportion to adult length cubed. Life history traits of this bivalve are compared to those reported for other small brooder clams. Some basic tendencies become apparent when traits for semelparous versus iteroparous species are contrasted. As in other semelparous sequential brooders (Transennella tantilla, Gaimardia bahamondei), the relationship between brood size and shell length obseved in K. chilenica does not fit the allometry hypothesis for marine brooding invertebrtes that allometric constraints on the brooding space limit the fecundity of larger individuals. Also contrary to theoretical predictions, small body size does not limit the diversity of a clam's developmental patterns. Possible explanations for this finding are discussed.     

Generalised linear modelling of fecundity at length in southern rock lobsters, <Emphasis Type="Italic">Jasus edwardsii</Emphasis>

Accurate fecundity estimates are central to population modelling of reproductive status and egg production in a fishery. Southern rock lobster (Jasus edwardsii) fisheries are managed with the consideration of performance measures based on the level of current egg production relative to the unfished stock. Egg production is a function of the size at the onset of maturity, the size structure of the stock, and the fecundity at length of lobsters. Fecundity at length of the Tasmania stock of J. edwardsii has not been reported previously and was estimated from seven sites around Tasmania. Data were analysed by classical linear regression and also generalised linear models (GLM) with gamma and Poisson distributions. GLM with a gamma distribution produced the best fit and did not require a correction factor to generate a model on the linear scale. Individual fecundity ranged from 43,918 to 660,156 and could be predicted from carapace length according to the equation F = −1.707 + 2.969 (log length). The coefficient of length approximated three, implying the fecundity at length relationship is cuboidal. Despite the spatial heterogeneity in reproductive parameters in this fishery, the inclusion of spatial information and egg development stage into the analysis only explained an additional 1% of the variance in fecundity. Thus, it appears valid to apply the relationship across the fishery without spatial separation.     

The effects of size-dependent predation risk on the interaction between behavioral and life history traits in a stream-dwelling isopod

In field surveys, laboratory observations and field-based assays of behavior, I examined the effects of size-dependent predation risk on the interaction between size at reproductive maturity and maternal care behavior in the stream-dwelling isopod, Lirceus fontinalis. L. fontinalis exhibit population-specific sizes at reproductive maturity which result in population differences in predation risk during the adult phase. Females from streams containing salamander larvae (that prefer small prey) mature at large sizes and then become relatively safe from predation. Females from streams containing fish (that consume all size classes of prey equally) mature at small sizes and remain at risk. I tested whether these differences in expected survival were reflected in the behavior of females during the maternal phase (i.e., the period during which females exhibit maternal care). Female L. fontinalis carry developing juveniles inside a brood pouch. I simulated predatory attacks on gravid female L. fontinalis from the different population types and found that female behavior correlated with population differences in risk. When “attacked”, females from streams with predatory fish (that experience high risk to adult females) released juveniles from the brood pouch, whereas females from populations with predatory salamander larvae (that pose relatively little risk to adult females) did not release juveniles. I discuss the results with reference to the joint evolution of behavioral and life history traits. Received: 6 March 1996 /Accepted after revision: 12 August 1996     

Reproductive biology and biochemical composition of the brooding echinoid <Emphasis Type="Italic">Amphipneustes lorioli</Emphasis> on the Antarctic continental shelf

The bathyal West Antarctic Peninsula (WAP) shelf experiences intense seasonal variability in primary production, with summer phytoplankton blooms yielding intense pulse of phytodetritus to shelf sediments. Echinoderms form a conspicuous proportion of the deposit-feeding megabenthos on the shelf and of these Amphipneustes lorioli was the most abundant irregular echinoid. To explore the reproductive response of A. lorioli to this seasonal production cycle, A. Lorioli was sampled at one location on the WAP shelf during four separate cruises between March 2000 and March 2001. Reproductive patterns were determined by histological analyses of gonad tissue, and elemental (CHN) analyses were used to estimate the nutritional and energetic status of the body tissues. Histological analysis of the brooding echinoid A. lorioli suggested a quasi-continuous gametogenic pattern in both the ovaries and the testes. Biochemical analysis of the gonads and the gut tissues were consistent with a continuous gametogenic cycle, showing no significant changes in the biochemical composition of the tissues among seasons. Size-frequency distributions of the embryo and juvenile echinoids within the adults’ brood pouches revealed a synchronous recruitment of embryos and juveniles in specific cohorts between different adult specimens. Whilst this occurrence of different cohorts of the developing brood may be an adaptation to limited brood space, there may also be an external factor influencing the synchrony between adult individuals. Nonetheless, a continuous gametogenic cycle and the lack of seasonal variation in the biochemical composition of gonad and gut tissues suggest that this deposit-feeding irregular urchin is exploiting a persistent sediment food bank in WAP shelf sediments throughout much of the year.     

Influence of environmental factors and female size on reproductive output in an invasive temperate marine gastropod <Emphasis Type="Italic">Rapana</Emphasis><Emphasis Type="Italic">venosa</Emphasis> (Muricidae)

Life history and reproductive strategies influence population dynamics at the inter- and intra-specific level. Environmental conditions suitable for gonad development and spawning, the reproductive range, may be a smaller portion of the broader species distribution. The only known breeding population of veined rapa whelks (Rapana venosa) in North America is in Chesapeake Bay, USA. There is considerable interest in the potential reproductive range of this non-indigenous species given the rapa whelk’s negative impacts on commercial shellfish species in both its native and introduced ranges. Weight-specific reproductive output is described for wild caught Chesapeake Bay rapa whelks maintained in flow-through mesocosms for 2 years. Measured reproductive output within and between egg capsule deposition seasons (years) in relation to water temperature, salinity, daylength, and female size is used to describe the rapa whelk’s reproductive range. Egg capsule production is influenced by seasonal and absolute water temperatures as well as seasonal daylength cycles. Egg capsule deposition by Chesapeake Bay rapa whelks begins at water temperatures of approximately 18°C and continues for 11–15 weeks. Forty to 70% of female whelks deposited egg capsules in most weeks during this season, producing 150–200 egg capsules female⁻¹ week⁻¹. Water temperatures >28°C caused reduced egg capsule production relative to temperatures of 20–25°C. Egg capsule production was positively related to seasonal changes in daylength, and two peaks of egg capsule deposition were observed in the 2001 and 2002 deposition seasons. The combination of declining daylength and higher water temperatures in late summer was associated with the cessation of egg capsule deposition. A lower average weight specific reproductive output in 130–145 mm SL rapa whelks (average 12 ± 1%) than in 90–106 mm SL rapa whelks (average 22 ± 1% of body weight) may reflect a life history that balances the physiological costs of maintaining a large body mass with the production of many planktonic larvae from multiple clutches of egg capsules per breeding season over a 10–15-year lifespan. Estimates of the cumulative day-degree requirements corresponding to the annual initiation of egg capsule deposition were 238 and 236 for 2001 and 2002, respectively. Reproductive output and day-degree requirements for Chesapeake Bay rapa whelks were similar to values calculated from previous studies of native muricids (Eupleura caudata and Urosalpinx cinerea). A latitudinal range of 30–41° (N and S) is predicted as the realized reproductive range for rapa whelk populations on the basis of the day-degree requirements for native whelks and reproductively active invasive rapa whelk populations. Electronic supplementary material  The online version of this article (doi:) contains supplementary material, which is available to authorized users.     

Factors affecting the interval between clutches in the hermit crab <Emphasis Type="Italic">Pagurus nigrivittatus</Emphasis>

Interval between reproductive events is an important factor for iteroparous animals because it determines the number of clutches throughout life. This study examined whether female size, clutch size, shell size and prenuptial molting affected the clutch interval in the hermit crab Pagurus nigrivittatus. Precopulatory guarding pairs of P. nigrivittatus were sampled in the field and kept in the laboratory until the female extruded eggs. The clutch interval of each female was assessed as one of two types of relatively “short” and “long” intervals by checking whether the guarded female had eggs and/or egg cases from the preceding brood or not when the guarding pair was collected. The clutch interval was longer in females with prenuptial molting than those without molting and these females usually grew larger at the prenuptial molt. This suggests that female P. nigrivittatus with a long interval might allocate energy into growth at the expense of the number of clutches during the current reproductive season. The allocation to growth is theoretically predicted to decrease with female size. Gastropod shell size is also known to affect the reproductive activity in hermit crabs. However, female size did not significantly affect the clutch interval in P. nigrivittatus, and the effect of gastropod shell size on clutch interval was not consistent with previous empirical studies. These results may be caused by differences in the gastropod species of shell occupied by the females of P. nigrivittatus.     

Problems of the gonad index and what can be done: analysis of the purple sea urchin <Emphasis Type="Italic">Strongylocentrotus purpuratus</Emphasis>

The gonad index, GI, is widely used as a measure of changes in reproductive state. There are, however, problems with its use because it is based on the implicit assumption of an isometric relationship between gonad size and some measure of total size. If, for example, gonad weight and total weight are used, the exponent for an allometric relationship usually is ignored and hence assumed to be 1.0. It is further assumed that this exponent is fixed for all states of the reproductive cycle and that gonads begin to develop at size = 0. Data for the purple sea urchin Strongylocentrotus purpuratus at Gregory Point, Oregon, USA, gathered over a period of 31 months showed that these assumptions cannot be supported. The relationship is better modeled with a function that (1) takes into account size of initial gonad production and (2) allows allometric exponents that vary with site or season. Thus, a better approach is to use a wide range of sizes to estimate size when gonads begin to develop and then, with this correction, ANCOVA to test for differences of gonad size among samples. Gonad changes at Gregory Point were estimated using fixed sizes of 5 cm diameter and 60 g total weight. Publishing means for X and Y, the standard error of the estimate, R ², and slope for each regression are shown to be sufficient to compare our results with results across studies.     

Multi-year observations on the gametogenic ecology of the Antarctic seastar <Emphasis Type="Italic">Odontaster validus</Emphasis>

This study reports the first multi-year observations on the reproductive patterns for an Antarctic predator/scavenger, Odontaster validus (Koehler 1912). Seastars were collected monthly from a shallow site (15–20 m depth) near the British Antarctic Survey (BAS) Rothera Research Station (Adelaide Island, 67°34′S 68°08′W) from July 1997 to January 2001. Reproductive condition, oocyte size frequencies and spermatogenesis were examined in at least ten seastars each month using histological and image analysis techniques. Gonad indices (GI) and pyloric caeca indices (PI) were also examined in the same samples. Female and male GIs varied seasonally, in parallel with a reduction in the proportion of large oocytes and mature sperm in the gonad in August to mid-October following winter spawning. Despite there being remarkable consistency in the timing of spawning from year to year, differences in the reproductive condition of individuals were apparent. Patterns in the digestive tissues also varied with season, peaking in December and reaching a minimum in February in two of the three study years. This weaker annual pattern may partly reflect the varied diet of this predator/scavenger species, which is not directly dependant on the timing and magnitude of the annual phytoplankton bloom. Pooled oocyte size distributions and residual analysis suggested that oogenesis progressed over 18–24 months, with the largest of the two size classes (maximum diameter = 183 μm) being spawned annually. This pattern of oocyte growth and spawning was previously reported in the early 1960s for an O. validus population from McMurdo Sound, which lies south of Rothera by 10° latitude. The extremely catholic diet of this predator/scavenger suggests the reproductive patterns of the seastar will be less susceptible to changes in food supply compared to polar suspension feeders or deposit feeders. Electronic supplementary material  The online version of this article (doi:) contains supplementary material, which is available to authorized users.     

The costs of male parental care and its evolution in a neotropical frog

Summary Parental care is practiced exclusively by males of the Puerto Rican frog, Eleutherodactylus coqui. Males brood clutches of direct-developing eggs in non-aquatic nest sites and defend eggs against cannibalistic nest intruders. Here, I report on energetic and mating costs incurred by males that provide parental care, and suggest how these proximate costs affect male fitness and the evolution of male parental care in this species. Energetic costs are small for brooding males in comparison to non-brooding, calling males. Brooding males had a higher frequency of empty stomachs and lost small, but significant, fractions of their initial body mass during parental care. Abdominal fat bodies of brooding males during the middle third of parental care were significantly smaller than those of calling males; those of males brooding eggs in earlier or later stages were not different. The mating cost of parental care is greater. Most brooding males cease calling during parental care. However, gravid females are available (i.e., known to mate) on most nights during the principal breeding season; hence non-calling males miss potential opportunities to mate. A mating cost was estimated by calculating nightly mating probabilities for calling males in a plot where nightly calling male densities and daily oviposition schedules were known. On average, a male exhibiting normal calling behavior would be expected to obtain a new mate once every 35.7 days. Hence a brooding male that ceased calling for a 20-day parental care period would miss, on average, 0.56 additional mates. Males that were more successful than average in attracting mates could miss up to 1.63 matings. A marginal value model (Fig. 1) is used to analyze the net effect on male fitness of parental care benefits and costs in E. coqui (Fig. 3). The model indicates that males garner the highest reproductive success by providing care from oviposition through hatching. There is no stage during the pre-hatching period at which a desertion strategy would yield higher reproductive success. In fact, the model suggests that males should provide full parental care even in the face of much higher mating costs than currently obtain in the system.