Adult purple sea urchins, Strongylocentrotus purpuratus (Stimpson, 1857), were collected on SCUBA subtidally near Goleta Bay, CA, in September 2020. Urchins were collected under the California Department of Fish and Wildlife Scientific Collecting permit #SC-9228 to the Santa Barbara Coastal LTER, and all protocols followed the guidelines of animal care and use at University of California, Santa Barbara. The mean temperature at the site was 16.8 ± 0.9°C on the day of collection. Following collection, urchins were held in a flow-through seawater table at ambient temperature (~16.4°C) for 24 hours before being placed in the adult acclimation treatments. During this period urchins were monitored for any disease or premature spawning and removed if needed.

To determine how maternal conditioning influenced provisioning and subsequent offspring performance, adult urchins were acclimated in either a warm (W: ~18.3°C) or cool (C: ~13.1°C) temperature condition representing MHW and non-MHW conditions, respectively (see Supplementary Material Figure 1 ). Urchins were held in the acclimation treatments from October 2020 to January 2021, which coincides with the window of gametogenesis for S. purpuratus populations in the SBC (Strathmann, 1987). Adult acclimation treatments consisted of three 37L replicate tanks held in a seawater table, acting as a water bath, to maintain temperature. Seawater was supplied to the treatment tanks via a dripper irrigation system with a flow rate of 12L/hr. The temperature of the seawater table was controlled using a Delta Star® heat pump with a Nema 4x digital temperature controller (AquaLogic, San Diego, CA, USA). The temperature of each tank was monitored daily using a wire thermocouple (Omega HH81a). Aquarium pumps (Aqua-Supreme) were installed in each tank to maintain proper water flow. Ten adult urchins were randomly selected and added to each treatment tank at the beginning of the acclimation period (N=60 total, 10 urchins per tank replicate for each temperature condition) ( Figure 2 ). Each week seven fronds of kelp (Macrocystis pyrifera) were added to each tank and urchins fed ab libitum. Urchins were inspected daily, and any individuals that showed signs of disease or death were removed.

Following the four-month conditioning period, urchins were induced to spawn by injecting 0.53M KCl into the coelom (Strathmann, 1987). Sperm was collected “dry” via pipette, and then stored on ice in 1.5mL microcentrifuge tubes until activated. Eggs were collected by inverting females over a beaker containing UV sterilized 0.35micron filtered seawater (FSW). Sperm and egg quality were checked under a compound microscope. Females with the highest quality eggs (determined via visual inspection of size, shape, and transparency) from each adult acclimation treatment were selected for test fertilizations with potential sires. To test for gamete compatibility, a sample of eggs from each female were combined with diluted sperm from four different non-MHW males. The male with highest sperm quality and compatibility (e.g., good motility and >95% fertilization) was chosen as the sire. Eggs from five females from each acclimation treatment, with the highest quality eggs and without fertilization incompatibilities, were then pooled in two separate beakers, creating two egg pools (i.e., a pool of eggs from MHW acclimated females, and a pool from non-MHW acclimated females). Each female contributed approximately 72,000 eggs to each pool. Beakers containing pooled eggs were gently mixed while diluted, activated sperm (at approx. 10⁵ cells per mL) from the chosen sire was added in small increments to avoid polyspermy. Since this study focused on examining maternal effects, only one sire from the non-MHW group was chosen to reduce any variation in offspring phenotype attributed to paternal genetic variation. Once pooled eggs reached 95% fertilization, embryos were deposited in larval culturing containers for the rearing phase of the experiment.

For the reciprocal culturing experiment embryos were raised at two different developmental temperatures: a MHW temperature (~18.5°C) hereafter denoted as ‘W’ and a non-MHW temperature (~12.7°C) hereafter denoted as ‘C’. This experiment created four offspring combination treatments that are denoted as: WW, WC, CW, CC, where the first letter represents the condition for the adult, and the second letter indicates the temperature at which embryos underwent development. Each of the four groups were raised in three replicate culture containers ( Figure 2 ). Culturing containers consisted of two nested 19L buckets. The inner bucket, which held the embryos, had cutouts that had been covered and sealed with 35μm mesh. The mesh covered cutouts allowed water to leave the bucket without losing embryos. Filtered seawater was fed into a reservoir tank, allowing the water to come to the treatment temperature prior to being fed into culture container using irrigation drippers at a rate of 6L/hr. To maintain treatment temperatures, reservoir tanks and larval culturing containers were held in seawater tables, acting as water baths, in the same set up described for adult treatment tanks. A paddle attached to a motor on each lid of the containers gently stirred the seawater to suspend embryos. Each container contained 12L of FSW and was stocked with 122,000 embryos. Seawater temperature in each culturing container was closely monitored using a wire thermocouple.

Embryo sampling was conducted at three stages during development: hatched blastula (HB), early gastrula (EG), and an early form of echinopluteus larvae, the prism stage (PR). Each stage was sampled when embryos displayed distinguishing developmental characteristics described in Strathmann (1987). Specifically, hatched blastula stage were characterized by hatching, where the embryo is no longer enclosed in the embryonic envelope; HB were sampled at 16 hours post-fertilization (hpf) for WW and CW treatments and 23hpf for WC and CC treatments. The early gastrula stage was characterized as the embryo displaying invagination at the vegetal plate and was sampled at 21hpf for WW and CW treatments and 34hpf for WC and CC treatments. The prism stage was characterized as the full formation of the tripartite archenteron, in addition to growth of the body rods, and was sampled at 44hpf for WW and CW treatments, 58hpf for WC, and 55hpf for CC treatments.

To determine the effects and interaction of maternal acclimation and developmental temperature on the performance of the four treatments (WW, WC, CC, CW), subsets of hatched blastula embryos from each treatment were tested in an acute thermal tolerance trial. Methods for this trial were as described by Wong and Hofmann (2020) with some modification. Briefly, using a temperature gradient heat block, scintillation vials containing 3.5 mL of FSW were brought to seven temperatures (17.5, 20.5, 26.4, 24.7, 28.0, 29.8, 31.1°C). Embryos from replicate culturing containers in each treatment were pooled so that each container contributed equally to the thermal tolerance trial. For the trials, 1,000 embryos in 500μL of FSW were added to the preheated scintillation vials, and were capped to prevent evaporation. Vials containing the embryos were maintained in the heat block for 1hr, at which point additional cool FSW was added to each vial and the vials were transferred to a 14°C environmental chamber for a recovery phase. During the recovery phase, embryos were allowed to proceed through development to the prism stage, an early pre-feeding larval stage. The timing of the appearance of embryos progressing to the prism stage was assessed visually using microscopy; approximately 43 hours elapsed from the start of the recovery period and to the beginning of the scoring process. Periodic water changes were conducted to make sure water in vials did not become hypoxic. Using approximately 100 individuals from each vial, we screened for advancement to the prism. Larvae were scored as alive if swimming or cilia movement was observed. LT₅₀ (the temperature at which the population experiences 50% mortality) was calculated for each treatment. While previous studies have measured mortality immediately following exposure to an acute thermal stress, for this experiment mortality was measured once the embryos reached the prism stage. This allowed for a more accurate measure of survival by adding a recovery phase to the acute temperature trial.

For HB and EG, 500 embryos from each culturing container were preserved in 2% formalin in FSW. In addition, 500 PR larvae from each culturing container were preserved in 2% formalin in FSW buffered with 100mM sodium perborate to a pH of 8.7, a formulation that reduces degradation of the skeletal rods. For each stage, preservation in 2% formalin was achieved by adding an equal volume of 4% formalin (in FSW or buffered FSW) to the sample volume.

Preserved embryos were promptly photographed for morphometric analysis using a compound microscope (Olympus BX50) with a 10x objective and attached digital camera (Motic 10 MP and Motic Image Plus software). Images were measured using ImageJ (National Institutes of Health, USA) and calibrated using a stage micrometer. For HB and EG images, individual embryos (N=35) from each treatment replicate were oriented so the full lateral view was visible, 2D area was measured and calculated by tracing around the perimeter of the embryo. For PR images, individuals (N=35) from each treatment replicate were oriented in a lateral view, so that the side profile of the archenteron was visible, 2D area was similarly measured by tracing around the perimeter of the embryo. At the prism stage, skeletal rod length was also determined by measuring from the tip of the body rod to the tip of the postoral rod.

To assess maternal investment in females from the different acclimation treatments, we measured egg traits including egg size and biochemical composition. Eggs were preserved and measured using the same methods as described above for embryos. For imaging, 500 eggs from each contributing female (N=10) were collected promptly after spawning, and preserved in 2% formalin in FSW for morphometric analysis. Individual eggs (N=35) from each female were measured by calculating the average of three diameters and two-dimensional (2D) area.

For lipid quantification, 2,000 eggs were deposited in a 1.5mL cryovial. Each sample was centrifugated to concentrate eggs (approximately 30s at 16,000 RCF) (Eppendorf 5415C Centrifuge), the water layer was carefully removed, and the sample was flash frozen in liquid nitrogen and stored at -80°C until analysis. Two tubes from each contributing female were used to quantify total lipid concentration. Total lipid was extracted using methods described in Wong et al. (2019), a method based on Sewell (2005) with some modification. The concentrated eggs from each sample were suspended in 250μL of Milli-Q water and sonicated at 2amps for a total of 45s at 15s intervals (Fisher Scientific Sonic Dismembrator 500). Each homogenized sample was then transferred to a 5mL glass V-vial (Wheaton) and combined with 125μL of chloroform (HPLC grade) and 250μL of methanol (HPLC grade). Samples were shaken vigorously and then centrifuged at 2,680 RCF for 5min at 4°C (Beckman Coulter Allegra™ 21R Centrifuge). Using a pulled Pasteur pipette the aqueous and chloroform layers were transferred to a clean V-vial; chloroform and water were then added to the final volume ratio of 4:3:2 of water:chloroform:methanol. Samples were then hand shaken and centrifuged again at 2,680 RCF for 5min at 4°C. After centrifuging, the chloroform layer was carefully transferred to a clean glass vial. All samples were flushed with N₂ and stored at -20°C prior to analysis.

Total lipid was quantified spectrophotometrically using modified methods based on Marsh and Weinstein (1966). Briefly, samples were placed under a flow of N₂ to allow chloroform to evaporate (Reacti-Vap™ Evaporators TS-18826, Thermo Scientific™). Once dried, 500μL of sulfuric acid (ACS grade) was added to each sample, covered with aluminum foil, and dried in a furnace at 200°C for 15min. Samples were allowed to cool for 10min prior to the addition of 2.5mL of deionized water to each vial. Samples were allowed to cool for an additional 10min and then absorbance of each sample was measured on a spectrophotometer (Shimadzu UV-1800) at 375nm. A standard curve of a known mass of lipid within a range of 30-300μg was also prepared and run in parallel with each batch of samples (R ² = 0.98-0.99). The lipid profile used to generate the standard curve consisted of major lipid classes found in S. purpuratus eggs which were previously reported in Wong et al. (2019); specifically, the standard was comprised of 51% triacylglycerol (Glyceryl tripalmitate, Sigma-Aldrich, Catalog No. T5888), 38% phospholipid (L-α-Phosphatidylcholine, Sigma-Aldrich, Catalog No. P3556), and 11% sterol (Cholesterol, Sigma-Aldrich, Catalog No. C8667).

Eggs used for protein quantification were prepared using the same process as described for lipid quantification. Three tubes from each contributing female were used to quantify total protein concentration. Protein was extracted from 2,000 eggs from each females using methods described in Wong et al. (2019). Briefly, 100μL of homogenization buffer (20mM Tris-HCl, (pH 7.6); 130mM NaCl; 5mM EDTA) with 1% Triton-X and 1% Protease Inhibitor Cocktail (Catalog number P8340, Sigma-Aldrich) was added to each sample of concentrated eggs and sonicated on ice for 20s. Samples were then shaken (Thermolyne M71735 Slow Speed Roto Mix) on ice for 15min, followed by centrifugation at 16,000 RCF for 20min (Eppendorf 5415C Centrifuge). The supernatant was transferred to a 1.5mL microcentrifuge tube and stored at -20°C until analysis. Total protein content was quantified using the Pierce™ BCA Protein Assay Kit following manufacturer’s instructions (Catalog No. 23225, Pierce Biotechnology, Rockford, IL, USA). The standard curve was generated using bovine serum albumin (R ² = 0.99).

All statistical analyses were conducted in R Studio (2022.12.0 + 353). Adult survival data was generated using the Kaplan-Meier method (Rich et al., 2010) and analyzed using a log-rank test. For the thermal tolerance trial, the influence of maternal acclimation temperature and offspring developmental temperature on survivorship in response to temperature was analyzed using a generalized linear mixed effect model (lme4 package) (Bates et al., 2015). The binary data (alive vs. dead) were fitted using a logit link. Models were compared, and the best fit was determined using standard model reduction approach for fixed effects, their interactions, and random effects; where the Akaike information criterion (AIC) was used to determine the best fit model. Maternal and offspring temperature were set as fixed effects and vial number was set as a random effect. A Wald chi-squared test was conducted to determine the significance of each factor on survivorship. The LT₅₀ value was calculated using a logistic regression for each treatment. The influence of maternal condition and the developmental temperature on body size at each stage (i.e., HB, EG, and PR), was similarly analyzed using a linear mixed-effect model. Maternal and offspring temperature were set as fixed effects and individual culturing containers were set as a random effect. For egg traits, the influence of maternal acclimation on egg size (i.e., average diameter and 2D area), total lipid, and total protein content was analyzed using a linear mixed-effect model. For egg size, total lipid, and total protein content, maternal temperature condition was set as a fixed effect while each individual mother was set as a random effect.

Adult S. purpuratus were successfully acclimated in the laboratory to two experimental temperature conditions, MHW (18°C) and non-MHW (13°C) conditions, with some mortality. Throughout the four-month acclimation period, mortality was observed in both treatments with a trend of higher survival in the non-MHW treatment (90%), as compared to the MHW treatment (73%). However, the difference in survival probability between treatments was non-significant ( Figure 3 ) (log-rank test: χ² = 2.8, df=1, p=0.10). Following the acclimation, surviving urchins from each treatment were spawned, providing ample material to create the embryo families and raise cultures. However, we observed decreased spawning success in MHW-acclimated adults as compared to non-MHW-acclimated adults where 50% of surviving urchins in the MHW treatment spawned as compared to the non-MHW treatment where all adults were induced to spawn.

Our study did detect a carry-over effect of maternal thermal history to the progeny. In the thermal tolerance trial, progeny from females acclimated to MHW temperatures (denoted as ‘W’) had a higher thermal tolerance to acute thermal stress than progeny from non-MHW-acclimated females (denoted as ‘C’), with higher percentages of larvae alive following the recovery period. Embryos from females acclimated to 18°C, the MHW condition, had higher LT₅₀ values than those acclimated to 13°C, the non-MHW condition ( Figure 4 ). Across the experiment, calculated LT₅₀ ranged from 27.2 - 28.4°C. For each treatment, LT₅₀ values for WW, WC, CW, and CC were 28.2, 28.4, 27.9, 27.2°C, respectively. When comparing the influence of maternal condition, a comparison of WW vs. CW treatments suggested that embryos, when reared at higher temperatures, gained a modest increase in tolerance (+0.3°C) if mothers experienced MHW temperatures during the period of oogenesis. We also found evidence that embryo environmental conditions influenced thermal tolerance. Here, elevated developmental temperature increased thermal tolerance in progeny from females naïve to MHW exposure, where LT₅₀ of the CW family was 0.7°C higher than the CC family.

A generalized linear mixed-effect model found that maternal acclimation, temperature at which offspring were raised, and their interaction all had a significant effect on embryo thermal tolerance ( Table 1 ). These results indicated that the thermal history of the females and the developmental temperature of the embryo both influenced thermal tolerance of early-stage purple sea urchins in the thermal tolerance trial.

The influence of maternal acclimation and offspring developmental temperature on body size differed as a function of stage, where maternal acclimation significantly influenced embryo body size at the early gastrula and prism stage, while offspring developmental temperature had a significant influence only at the prism stage ( Table 2 ). Difference in body size amongst treatments were most evident at the prism stage, where on average, offspring from MHW-acclimated females were 7.1% larger than those from non-MHW-acclimated females. Additionally, offspring raised at the MHW temperature were on average 14.3% larger (2D area) than those raised at the non-MHW temperature and had skeletal rod lengths that was on average 53.5% longer than those raised in the non-MHW condition ( Figure 5 ). Results from the best fit linear mixed effect models are summarized in Table 2 .

In general, maternal temperature acclimation had no effect on egg size. Although eggs from non-MHW-acclimated females trended larger in size, maternal environmental history during oogenesis did not significantly influence average egg diameter (F _1,8 = 0.1272, p= 0.7286) or 2D area (F _1,8 = 0.5882, p= 0.4608). Average egg diameter for MHW- and non-MHW-acclimated mothers was 0.0798 ± 0.023 and 0.0801 ± 0.023mm, respectively. The average 2D area of eggs from MHW- and non-MHW-acclimated females was 0.0051 ± 0.003 and 0.0052 ± 0.003mm², respectively ( Figures 6A, B ).

In contrast to morphometric traits, the biochemical composition of the eggs did vary in a comparison of females acclimated to MHW vs. non-MHW temperatures. Specifically, protein concentrations were higher in eggs from MHW-acclimated females as compared to eggs from non-MHW-acclimated females (F _1,8 = 11.184, p<0.01). Total protein content from MHW and non-MHW acclimated mothers was 40.4 ± 3.38 and 34.6 ± 3.23 ng/egg, respectively ( Figure 6C ). In contrast, maternal thermal history did not have a significant effect on total lipid content of eggs (F _1,8 = 2.3268, p =0.1575), although trending differences in lipid content were observed. Total lipid content in eggs from MHW- acclimated females was on average 17.6 ± 5.19 as compared to 14.0 ± 2.62 ng/egg in MHW- acclimated females ( Figure 6D ).

The main objective of the study was to examine whether adult environmental experience in MHW conditions would influence the performance of their progeny. Ecologically, this is significant because MHW events can occur throughout the period in which purple urchins are broadcast spawning and thus, early development could occur at anomalously high temperatures in situ ( Figure 1 ). Using laboratory experiments, we queried whether adults could “prime” their progeny during periods of anomalously elevated temperatures, or whether MHW stress during gametogenesis would have deleterious effects on offspring performance. Using acute temperature trial assays used previously on early stage echinoderms (Wong and Hofmann, 2020), we observed that maternal exposure to MHW temperatures increased thermal tolerance in progeny ( Figure 4 ), as demonstrated by higher LT₅₀ values.

These results join other studies that examined TGP in marine invertebrates in response to thermal stress, where the offspring response varied with species, timing, and nature of parental conditioning (Putnam and Gates, 2015; Shama, 2015; Morley et al., 2017; Rivera et al., 2021; Bernal et al., 2022; Waite and Sorte, 2022). For example, in a study on the marine polychaetae, Ophryotrocha labronica, researchers found that the thermal tolerance of the progeny tracked the timing of the temperature exposure of the female, where tolerance of offspring differed depending on whether maternal acclimation occurred during early or late oogenesis (Massamba-N'Siala et al., 2014). These results match the outcome of our experiment in S. purpuratus. Together, these findings underscore the significance of the timing of the warming events associated with MHWs and their subsequent effects on offspring resilience. Since the nature of MHWs can vary from prolonged to short-term events (Oliver et al., 2021), transgenerational responses may differ depending on duration of maternal exposure.

In addition to measuring upper thermal tolerance as a metric of offspring performance, we examined body size, an important functional trait of echinoderms in larval ecology and physiology (Strathmann, 1971; Allen and Marshall, 2014) that has shown to correlate to performance in studies of local adaptation and plasticity in S. purpuratus (Kelly et al., 2013; Strader et al., 2022). In this study using MHW-relevant temperatures, we found that maternal thermal history significantly influenced body size (measured as 2D area) for early gastrula embryos and prism larvae ( Figure 5 ), where offspring from MHW-acclimated females were larger in size, potentially indicating a positive transgenerational effect, since larger size in larvae is often associated with higher fitness (Strathmann, 1971; Allen and McAlister, 2007). Offspring developmental temperature also influenced body size for prism-stage larvae, which was evident in the observed differences in skeletal rod length. An interaction between maternal thermal history and developmental temperature was seen only at the early gastrula stage, where differences in body size within each developmental treatment were dependent on the maternal experience.

Other studies examining transgenerational effects in response to elevated temperature conditions in echinoderms have reported mixed effects on larval size. For example, studies on Strongylocentrotus intermedius and Tripneustes gratilla, show that adults exposed to long-term elevated temperatures, 15 months and 6 months respectively, produced progeny that were smaller in size compared to progeny from adults conditioned to ambient temperatures (Zhao et al., 2018; Karelitz et al., 2019). In addition, adult long-term acclimation to +2°C temperature conditions did not affect body size of the progeny in Heliocidaris erythrogamma (Harianto et al., 2021). Overall, measurement of body size at different stages in development may contribute to the differences seen amongst studies. Furthermore, length of adult exposure, differences in species plasticity, and population-level responses likely contribute to reported differences in transgenerational responses.

Mechanistically, maternal contributions to shifts in tolerance of the progeny could manifest in two ways - 1) as maternal energetic investment in the eggs, and/or 2) as maternally-derived egg transcripts and proteins working to reduce cellular damage due to heat stress during early development. Historically maternal energetic investment has been assessed through measurements of egg size and biochemical composition (i.e., lipids and proteins) where studies have shown that egg size and nutritional macromolecules are significant in determining conditions and performance of developing progeny in echinoderms (Jaeckle, 1995; Prowse et al., 2008; Moran et al., 2013; Peters-Didier and Sewell, 2017). Given these data on echinoderm eggs, we hypothesized that observed differences in offspring phenotype would be due to differences in maternal energetic provisioning (i.e., egg size and biochemical composition). In testing this hypothesis, we found that acclimation of females to MHW temperatures during oogenesis resulted in significantly altered egg characteristics. Specifically, although egg size was unaffected by maternal acclimation, eggs from MHW- acclimated females had higher protein content and had total lipid content that trended upward ( Figure 6 ).

For marine invertebrates, egg size is often used a standard measurement of maternal investment as it has been linked to numerous adaptive traits in developing embryos and larvae such as fertilization success, larval development, and post-metamorphosis survival (Moran and McAlister, 2009; Allen and Marshall, 2014). While differences in egg size were not observed in our study, past research examining the effects of maternal acclimation temperature on egg size report varying outcomes (Foo and Byrne, 2017). In combination, data in the literature suggest that species and the duration of thermal exposure are likely significant determinants of egg size (Skadsheim, 1989; Suckling et al., 2015; Guillaume et al., 2016; Karelitz et al., 2019). However, when examining provisioning of eggs, focusing solely on size may not be the best predictor of energetic content. In echinoid species specifically, while egg size generally scales with energy content, at the intraspecific level this relationship is weakened (McEdward and Morgan, 2001; Moran and McAlister, 2009).

In this light, to further assess differences in egg quality, we measured total protein and lipid concentrations. Protein concentrations were significantly higher in eggs from females with the history of high temperature conditioning ( Figure 6C ). In addition, although there was a trend indicating higher lipid levels as well, there was not a significant difference between eggs of MHW- and non-MHW-acclimated females. In general, the levels of protein and lipid measured here are consistent with other measurements in S. purpuratus (Matson et al., 2012; Wong et al., 2019). Studies examining how these macromolecules are used throughout development show that lipids (specifically triglycerides) are major energy fuels, while structural lipids (e.g., phospholipids) and proteins are used to construct the larval body (Byrne et al., 2008). These patterns of biochemical utilization support our results where larger larvae developed from more protein rich eggs (i.e., eggs from MHW-acclimated females).

While not evaluated in our study, differences in egg protein characteristics may potentially influence thermal tolerance of early-stage embryos via maternal transmission of heat shock proteins. In support of this hypothesis, a study examining gonad tissue of acclimated adult urchins (20 vs 26°C) found higher levels of Hsp70 protein in the warm-acclimated urchins (Harianto et al., 2021). While measurements of gonad tissue do not identify standing stocks of defensome proteins in spawned eggs, this study does highlight the potential for maternal acclimation via changes in egg protein characteristics to alter tolerance of developing embryos. In addition to proteins, maternally sourced transcripts may also contribute to differences in thermal tolerance at early embryonic stages. Together these inherited molecules are necessary for driving early developmental processes, and it is not until the maternal-to-zygotic transition that embryos are able to activate their own genome (Hamdoun and Epel, 2007; Schier, 2007). With this in mind, we chose to study hatched blastula embryos, specifically, as studies have shown that these early-stage embryos still rely on maternal transcripts and are unable to completely activate elements of the embryonic defensome, e.g., genes for molecular chaperones and heat-shock proteins (Sconzo et al., 1995).

To further highlight physiological mechanisms that contribute to TGP, transcriptomic studies on urchins have shown that parental acclimation to different temperatures influence gene expression in offspring during early development (Shi et al., 2020). Similar results have been observed in S. purpuratus under variable conditions of temperature and _pCO₂ (Wong et al., 2018). While transcriptomic techniques have aided our ability to determine the molecular underpinning of TGP, to shed light on maternal provisioning mechanisms, future studies should incorporate long-term acclimation during gametogenic windows followed by examination of maternally sourced RNA and proteins in eggs and early embryos. Assessment of the temperature-related egg changes in a climate-change context could be further investigated by examining the translatome, an approach that has been used in other echinoderms (Chassé et al., 2018).

In addition to addressing transgenerational processes, the factorial design of the experiment allowed us to examine how the temperature at which embryos developed influenced their phenotypic and functional traits (i.e., body size and thermal tolerance). We found that progeny from non-MHW acclimated females reared in non-MHW conditions had the lowest LT₅₀ value. However, thermal tolerance of these progeny increased when reared at the higher MHW temperature ( Figure 4 ). This outcome has been observed in other echinoderm species. In a study on red urchins (Mesocentrotus franciscanus), embryos reared at 17°C had an increased thermal tolerance as compared to embryos reared at 13°C (Wong and Hofmann, 2020), indicating that if rearing temperatures do not exceed upper thermal limits there is a potential for a resistance response in early-stage embryos. In our study, developmental temperature also had a significant effect on body size at the latest stage evaluated (i.e., PR), where development in MHW conditions was positively associated with body size and body rod length ( Figure 5 ).