<p>Initial maternal provisioning has pervasive ecological and evolutionary implications for species with direct development, influencing offspring size and energetic content, with subsequent effects on performance, and consequences in fitness for both offspring and mother. Here, using three sympatric marine intertidal direct developing gastropods as model organisms (Cominella virgata, Cominella maculosa and Haustrum scobina) I examined how contrasting strategies of maternal investment influenced development, hatchling size, maternal provisioning and juvenile performance. In these sympatric whelks, duration of intra-capsular development was similar among species (i.e. 10 wk until hatching); nonetheless, differences in provisioning and allocation were observed. Cominella virgata (1 embryo per capsule; ~3 mm shell length [SL]) and C. maculosa (7.7 ± 0.3 embryos per capsule; ~1.5 mm SL) provided their embryos with a jelly-like albumen matrix and all embryos developed. Haustrum scobina encapsulated on average 235 ± 17 embryos per capsule but only ~10 reached the hatching stage (~1.2 mm SL), with the remaining siblings being consumed as nurse embryos, mainly during the first 4 wk of development. Similar chronology in the developmental stages was recognizable among species. Higher growth rates and evident juvenile structures became clear by the second half of development and larval characteristics were less frequently observed. Even after 10 weeks of encapsulation and despite emergent crawling juveniles, some hatchling H. scobina still retained “larval” traits, suggesting that this nurse embryo-based provisioning could result in intracapsular asynchrony of development, and that female of this species would be able to bet-hedge in a higher extent compared with female C. maculosa or C. virgata. Maternal investment in newly laid egg capsules differed among the three study species. The structural lipids phospholipid (PL) and cholesterol (ST) and the energetic lipids aliphatic hydrocarbon (AH), triglycerides (TG), diglycerides (DG) and free fatty acids (FFA) occurred in all three species. Only eggs (and also hatchlings) of the multiencapsulated embryos C. maculosa and H. scobina were provisioned with the energy lipids wax ester (WE) and methyl ester (ME), suggesting an interesting similarity with pelagic larvae of other invertebrates and fish where those lipid classes have also been recorded. Despite differences in hatchling size, the small H. scobina had significantly higher amounts of the energy storage lipid TG compared with C. maculosa and C. virgata, suggesting interesting trade-offs between offspring size and offspring energy resources. H. scobina was the only species that suffered a complete depletion of FFA during development (5th wk), suggesting an additional role of this energetic lipid during the early stages of development. Differences in the amount of lipids among newly laid capsules and siblings within capsules were also detected within species. In both species with multiple embryos per capsule, C. maculosa and H. scobina, these differences were largely explained by variation in TG and PL, enhancing the important role of the major structural (PL) and energy (TG) lipids during the early stages of these whelks, and also providing an integrative approach for evaluating maternally-derived lipids on a perindividual basis in direct developing species with contrasting provisioning and offspring size. Because in direct developers maternal provisioning to the embryos is the primary source of nutrition until offspring enter juvenile life, differences in performance should be closely related with initial provisioning, which in turn may reflect maternal nutritional conditions. Field-based surveys and manipulative experiments in the laboratory showed that different maternal environments (i.e. locations and sites) and contrasting offspring size influenced juvenile performance in different ways for C. virgata and C. maculosa. Despite the large differences in conditions and available resources between the Wellington Harbour and the nearby South Coast, the two locations did not influence the hatchling size of either species, and the most important source of variation was at the smallest scale (i.e. among sites), with substantial variation also occurring within and among females. Between and within species differences in hatching size reflected juvenile performance when fed, regardless of whether subjected to desiccation stress. When starved however, species-specific and size differences in performance were less significant. As has been described for many taxa, large offspring often perform better than small conspecifics; however, because this performance is likely to be context-dependent, understanding the importance of the different scales of variation is crucial for determining how variation in size reflects an organism’s performance. Despite the long recognized role of intra-specific variation in offspring size in mediating subsequent performance, the consequences of inter-specific variation in peroffspring maternal investment for co-occurring taxa have been rarely examined in a predator-prey context. Manipulative experiments in the laboratory with hatchling and juvenile C. virgata and C. maculosa revealed that vulnerability of their early life-stages to common crab predators (i.e. the shore crab Cyclograpsus lavauxi) is highly size-dependent. When predator size was evaluated, small crabs were unable to eat hatchlings of either whelk species. Medium and large shore crabs consumed both prey species; however, hatchlings of C. virgata were less vulnerable to predation by medium crabs than large ones, and C. maculosa were equally vulnerable to both sizes of crabs. In hatchlings of both prey species the shell length and shell thickness increased over time; however, only C. virgata reached a size refuge from predation after two months posthatch. Results showed that vulnerability to predators can be mitigated by larger sizes and thicker shells at hatch; nonetheless, other species-specific traits such as juvenile growth rates, may also play key roles in determining the vulnerability of hatchling and juvenile snails when exposed to shell-crushing predators. Overall, these findings suggest that when defining offspring size, provisioning and performance relationships, many context-dependent scenarios are likely to arise. Therefore examining the early life-history stages of direct developing whelks with contrasting maternal investment under an integrative morphological, physiological and experimental approach, allowed a better understanding of how these complex relationships arises and how mediated the species life-history in terms of offspring size, maternal provisioning and subsequent juvenile performance.</p>