Crustacean Research 2021 Vol.50: 17–27 ©Carcinological Society of Japan. doi: 10.18353/crustacea.50.0_17 Seasonal variability in fecundity and egg size in the porcellanid crab Petrolisthes japonicus (Decapoda: Anomura: Porcellanidae) Katsuyuki Hamasaki, Minako Ishii, Shigeki Dan Abstract.̶ We investigated the seasonal variability in fecundity and egg size in a Petrolisthes japonicus population by monthly sampling for one year on an intertidal cobble and boulder shore on the Boso Peninsula, Japan. Fecundity (number of eggs per brood) was determined for ovigerous females with non-eyed or eyed eggs in early (May and June), middle (July) and late (August and September) breeding periods. The long and short diameters of non-eyed eggs were measured, and the egg volume was calculated. Fecundity was positively correlated with female body size and found to be greater in the middle breeding period than in the early and late breeding periods. Egg loss occurred during embryogenesis, the egg loss rate being calculated as 44.6%, 14.8% and 30.2% from the non-eyed to the eyed egg stages in the early, middle and late breeding periods, respectively. Egg size decreased linearly from the early to the late breeding periods. It is inferred that the seasonal variability in reproductive output, fecundity and egg size in P. japonicus may be regulated by female nutritional condi- tions and seasonal fluctuations of food production for larvae, as well as a trade-off be- tween the number and size of eggs in a brood. Key words: reproductive seasonality, reproductive output, reproductive traits, phenotypic plasticity, egg loss Electronic supplementary material. The online version of this article contains supplementary material at https://www.jstage.jst.go.jp/article/crustacea/50/0/50_17/_article ■ Introduction 1991; Reid & Corey, 1991). The fecundity (number of eggs per brood) is determined by Decapod crustaceans belonging to the Subor- the average egg size of a brood, owing to a der Pleocyemata, such as lobsters, crayfishes, trade-off between egg size and number within caridean shrimps, anomuran crabs and brachy- an ovary with a designated biomass for produc- uran crabs, brood eggs underneath the pleon, ing a brood (Hines, 1982). In general, egg size whereas those belonging to the Suborder Den- is positively correlated with the size of newly drobranchiata, such as penaeid prawns, spawn hatched larvae (Hines, 1986; Hancock, 1998; eggs directly into the water (Fernández et al., Hamasaki et al., 2003; Bas et al., 2007; Gebau- 2020). Reproductive output (the amount of en- er et al., 2013). ergy allocated to egg production in terms of The life-history trait of producing few and biomass) is principally determined by female large eggs has often been considered as an evo- body size because of allometric constraints on lutionary adaptation to aquatic environments space available for yolk accumulation within with insufficient, unpredictable or seasonally the cephalothorax (Hines, 1982; Corey & Reid, short production of food for larvae (Anger, Received: 23 Nov 2020. Accepted: 7 Jan 2021. Published online: 23 Feb 2021. 17 KATSUYUKI HAMASAKI, MINAKO ISHII, SHIGEKI DAN 2001, 2006). Reproductive traits such as fecun- japonicus population by conducting monthly dity and egg size are thus particular to each sampling for one year on an intertidal cobble species; however, the intraspecific variability and boulder shore on the Boso Peninsula, Ja- in reproductive output, fecundity and egg size pan. has been known to occur in several decapod crustacean species, depending on regionally ■ Materials and Methods and/or seasonally varying environmental con- ditions, such as temperature, salinity and food Crab sampling and measurements production (Sampedro et al., 1997; Paschke et We used P. japonicus samples collected in al., 2004; Leme, 2006; Bas et al., 2007; Terossi our previous study for examining the reproduc- et al., 2010; Wehrtmann et al., 2012; Gebauer tive season and population structure of this et al., 2013; Urzúa & Anger, 2013; Guzmán et species (Hamasaki et al., 2020). Briefly, al., 2016; Sato et al., 2017; Baldanzi et al., monthly sampling was conducted from April 2018). Information on the intraspecific vari- 2008 to March 2009 in an intertidal cobble and ability in fecundity and egg size is thus crucial boulder zone (34°58′N, 139°46′E) on the Boso to infer the reproductive potential and future Peninsula, Japan. The crab specimens were stock size of a decapod crustacean population. brought to the laboratory and kept in the freez- Anomuran crabs of the family Porcellanidae er. Haworth, 1825 exhibit a brachyuran crab-like After thawing the frozen specimens, the cara- body shape with broad and dorsoventrally flat- pace width (CW, the widest portion between tened chelipeds and a well-developed, symmet- the lateral margins of the carapace) was mea- rical pleon that is carried bent under the cepha- sured using a digital calliper (CW＞～3 mm, lothorax and held against the thorax (Jones, accuracy 0.05 mm) or a stereomicroscope with 1977; Osawa & McLaughlin, 2010). Petrolis- the aid of a micrometer (CW＜～3 mm, accu- thes japonicus (De Haan, 1849), one of the racy 0.01 mm). The crabs with＞2.3 mm CW most common porcellanid crabs in Japan, is were sexed based on the position of the gono- found on intertidal cobble and boulder shores pores (i.e. at the coxae of the third pereiopods in temperate and tropical regions (Miyake, in females and at the coxae of the fifth pereio- 1998). It has been reported that the main repro- pods in males) (Osawa & McLaughlin, 2010) ductive season of P. japonicus lasts for approx- because the minimum body size of females has imately 4–5 months during the spring and sum- been found to be 2.3 mm CW (Nakasone, mer seasons, based on the proportions of 1972; Hamasaki et al., 2020). ovigerous females occurring in the populations Each female was classified according to the (Nakasone, 1972; Hamasaki et al., 2020). presence or absence of eggs attached to the It is thus expected that the fecundity and egg pleopods (ovigerous or non-ovigerous females, size of P. japonicus may vary during the rela- respectively). To determine the fecundity of fe- tively longer reproductive season. However, no males, i.e. the number of eggs attached to the information is available for fecundity and egg female pleopods, we carefully removed the size in P. japonicus, except for that provided eggs from each ovigerous female using for- by Nakasone (1972), who reported the mean ceps, and their embryonic development was number of eggs in five body-size classes as 30– classified into three stages under the stereomi- 381 eggs and the mean long and short diame- croscope according to the following character- ters of eggs as 0.75 × 0.67 mm. The aim of istics: non-eyed eggs, embryos without eye the present study was to investigate the season- pigmentation; eyed eggs, embryos with eye al variability in fecundity and egg size in a P. pigmentation; and hatching eggs, fully devel- 18 Crustacean Research 50 REPRODUCTION OF PETROLISTHES JAPONICUS oped embryos with little or no yolk and any water temperatures near the sampling site mea- larvae that had emerged from the egg mem- sured in the early and middle days of each brane. All the non-eyed and eyed eggs were month): early breeding period, May and June counted, but the hatching eggs were not count- (～18–20°C); middle breeding period, July (～ ed because some embryos might have been lost 20–23°C); and late breeding period, August during the sampling operations. and September (～26–27°C). An ovigerous fe- Egg sizes were determined for the non-eyed male collected in February was included tem- stage because decapod crustacean eggs in- porarily in the late period. The outlier was ex- crease their volume during embryonic develop- cluded from the regression analyses with the ment, particularly during the eyed egg stage aid of Cook’s distance (Cook, 1977), using the (e.g. Wear, 1974; Hamasaki et al., 2003). The cutoff point of 4/(n－k－1), where n is the long (L) and short (S) diameters (mm) of 10 number of samples and k is the number of ex- non-eyed eggs from each specimen of a sub- planatory variables (Fox, 2008). The Cook’s sample of the monthly collected ovigerous fe- distance was calculated using the cooks.dis- males were measured under a microscope with tance function. the aid of a micrometer (accuracy 0.001 mm); Analysis of covariance (ANCOVA) was em- the number of female specimens was 17, 28 ployed to compare differences in slopes and and 7 individuals in June, July and August adjusted means of the response variable (lnNE) 2008, respectively. The egg volume (mm3) was among/between log-transformed regression calculated as (1/6 × L × S2 × π). equations using the covar.test function (http:// aoki2.si.gunma-u.ac.jp/R/covar-test.html). Be- Data analyses cause significant differences were not detected Statistical analyses were performed with R in the slopes of regression equations among statistical software version 4.0.2 (R Core Team, any comparisons, adjusted mean values with 2020) at a 5% significance level. The relation- 95% confidence intervals of the response vari- ship between the number of eggs attached to able (lnNE) were calculated using the lsmeans the pleopods (NE) and female body size (CW) function (Lenth, 2016), based on the log-trans- was determined using a power function, NE＝ formed equations with the same slope and dif- aCWb, as applied for many decapod crustacean ferent intercepts. These adjusted means (lnNE) species (Somers, 1991), including porcellanid were converted to antilogarithmic numbers. In crabs (e.g. Jones, 1977; Gebauer et al., 2007; decapod crustacean species, eggs are lost from Pinheiro et al., 2017).