Printer Friendly

Developmental Consequences of Temperature and Salinity Stress in the Sand Dollar Dendraster excentricus.

Introduction

Understanding how organisms develop within the context of their physical environment is a major aim of the field of ecological developmental biology (Gilbert, 2001; Gilbert and Bolker, 2003). While it has long been known that developing organisms are capable of responding to environmental fluctuations (e.g., Pigliucci, 2001), there is renewed interest in understanding how organisms respond to their environment during development (e.g., Sultan, 2003, 2007). In some cases, environmentally induced developmental plasticity may even facilitate the evolution of traits (West-Eberhard, 2003; Moczek et al., 2011) by extending the range of phenotypic variation that organisms may exhibit during early life-history stages (Sultan, 2007), eventually resulting in fixation of the phenotype in the population (Abouheif et al., 2014).

One potentially adaptive form of developmental plasticity is polyembryony, the production of two or more individuals from a single sexually produced embryo (Craig et al., 1997). The phenomenon of polyembryony is predicted to be adaptive in situations in which offspring have more information regarding their environmental conditions than the mother (Craig et al., 1997). Polyembryony may be especially beneficial for organisms whose offspring develop in unpredictable environments for long periods after release from the mother. For example, the parasitoid wasp Copidosoma floridanum inserts 1 or 2 fertilized eggs into a moth egg, and those eggs proliferate into as many as 2000 individual wasps that then eat their way out of the host (Ode and Strand, 1995; Grbic et al., 1998). The quantity of wasps produced from the one or two initial eggs is determined by the level of juvenile hormone present in the developing host moth larva (Craig et al., 1997). In the case of this parasitoid, offspring developing inside of the host are better able to assess environmental conditions than is the mother at the time of oviposition.

Similarly, free-spawning marine invertebrates with extended dispersal during planktonic development also meet the criterion that embryos are better suited than mothers to assess the environmental conditions in which they are developing. Among marine invertebrates, polyembryony is a widespread phenomenon (reviewed by Allen et al., 2018) that has long been recognized to occur in response to both abiotic and biotic cues (Just, 1937). For example, Hey ward and Negri (2012) reported that, in turbulent waters, embryos of the coral Acropora millepora fragmented during early cleavage cycles; but each fragment still yielded normal, although smaller, larvae and, ultimately, smaller juvenile polyps. Polyembryony also occurs in hydrozoan cnidarians (Craig et al., 1997) and cyclostome bryozoans (Hughes et al., 2005). Echinoderms are also reported to exhibit polyembryony during both the embryonic (Mortensen, 1938) and larval (Bosch et al., 1989) phases. Larval polyembryony, or cloning, is best described in asteroids (Jaeckle, 1994; Vickery and McClintock, 2000; Knott et al., 2003; Allen et al., 2019); but it has also been reported in echinoids, holothuroids (Eaves and Palmer, 2003), and ophiuroids (Balser, 1998).

In contrast to larval cloning, reports of environmentally induced polyembryony in early echinoderm embryos are rare. The first description of polyembryony in echinode[pi]ns reported regular formation of twin blastulae in the pencil urchin Prionocidaris baculosa (Mortensen, 1938); however, no details were provided about what environmental conditions might or might not induce this phenotype. More recently, polyembryony has been reported in another pencil urchin (Eucidaris tribuloides) and a sand dollar (Echinarachnius parma) (Allen et al., 2015). In both of these species, embryos were observed to develop into twins, triplets, quadruplets, and so on, in response to elevated temperatures and reduced salinities (Allen et al., 2015). However, the ability of these resulting embryos to complete development to metamorphosis--and the frequency with which polyembryony might occur in nature--remains unknown. In this study we sought to examine how fluctuations in temperature and salinity influence the development of the sand dollar Dendraster excentricus.

In Washington State, the Salish Sea and, particularly, the waters of the San Juan Archipelago periodically experience spikes in temperature and corresponding dips in salinity during the warmer summer months (May-August), due to freshwater inputs from the Fraser River (Sutherland et al., 2011; Pia et al., 2012). These fluctuations in abiotic conditions overlap with the spawning season of many echinoderms, including the sand dollar D. excentricus. We exposed embryos of D. excentricus to various salinity and temperature treatments and followed early development to document the prevalence of polyembryony and other developmental pheno-types. We also tested whether both embryos derived from twin polyembryonic embryos were competent to complete development to the juvenile stage. Additionally, we tested two potential mechanisms by which polyembryonic pheno-types could be generated. Whether polyembryony in echinoderms is adaptive, or merely a developmental abnormality in response to changing environmental conditions, remains to be demonstrated.

Materials and Methods

Adult collection, spawning, and environmental parameters

Adult Dendraster excentricus (Eschscholtz, 1831) were hand collected from sand dollar beds at the head of East Sound on Orcas Island, Washington (48[degrees]41'38"N, 122[degrees]53'46"W), between April and August 2014. Animals were transported by boat back to Friday Harbor Laboratories (FHL) on San Juan Island, Washington (48[degrees]32'45"N, 123[degrees]00'46"W), and placed in large outdoor flow-through seawater tanks with several centimeters of sand coating the bottom. Adults were kept in those tanks for the duration of the experiments and were haphazardly removed to collect gametes for experimental trials. Once spawned, adults were not reused in any other trials.

In order to induce spawning, we injected animals with 1-3 mL of a 0.5 mol [L.sup.-1] potassium chloride (KC1) solution (Strathmann, 1987). After injection, sand dollars were inverted over 100-250-mL glass beakers containing 32 ppt 0.45-[micro]m Millipore-filtered seawater (FSW). During spawning events, gamete viability was assessed by visual inspection under a compound light microscope. Eggs were examined to ensure uniformity of size and completion of germinal vesicle breakdown. Sperm were examined to ensure that they were actively swimming when diluted in FSW. One milliliter of a dilute sperm suspension was then added to a small subsample of eggs to score fertilization success. Fertilization was assessed by the presence of the fertilization envelope (FE) in the first 50 eggs observed. Acceptable fertilization scores were 45 out of 50 eggs or more (i.e., [greater than or equal to]90%); scores lower than that were assumed to reflect poor gamete viability, and additional sand dollars were induced to spawn until a viable pair was found. Typically, fertilization was high, and only rarely was injection of new individuals required. Prior to experimentation, unfertilized eggs were rinsed in FSW, stirred, and then gently pipetted in 2-mL aliquots into experimental glass bowls containing 100 mL of the experimental salinity. Sperm were then pipetted into experimental bowls in 100-[micro]L aliquots, at which point sperm and eggs were gently stirred in each bowl to facilitate gamete contact.

For the fertilization assay, 7 salinity treatments were used, starting at 32 ppt and decreasing by increments of 2 ppt, down to 18 ppt. Each salinity treatment was replicated in three unique bowls, and all bowls were kept on the lab bench to develop under elevated temperature conditions (19-22 [degrees]C). For a single male-female pair, sperm and eggs were combined at each salinity; and the percent fertilization was scored in each experimental bowl after ~15 minutes. Beginning at this time, fertilization envelope diameter (FED) and egg diameter (ED) were also measured, and all diameter measurements were completed prior to any visible signs of the first cleavage cycle (n = 10 embryos per replicate bowl). All scoring and measuring were completed using a compound light microscope at 100 x magnification.

To validate the laboratory conditions to which embryos were exposed, we collected temperature and salinity data from East Sound, Orcas Island, including the water immediately above the sand dollar beds where adults were collected. Data were collected on two different days during high tide, using a YSI Pro 1030 conductivity and temperature probe (Yellow Springs Instrument, Yellow Springs, OH). Data from an array of temperature and salinity sensors deployed at -1.7 m mean lower low water (MLLW) at Cantilever Point, San Juan Island, were provided by Dr. Emily Carrington (FHL) and were used to assess the local water conditions near the intake of the FHL seawater system.

Induction of polyembryony and twin viability

Experiments were conducted to assess the frequency of different developmental morphologies, including the frequency of polyembryonic development in embryos exposed to different environmental conditions. These experiments were run at all combinations of three salinity (32, 29, and 26 ppt) and two temperature (12-15 [degrees]C and 19-23 [degrees]C) treatments. The first temperature treatment, hereafter known as "ambient," was set up in a flow-through sea table in which the water temperature ranged from 12 [degrees]C to 15 [degrees]C (14.4 [+ or -] 2.2 [degrees]C, mean [+ or -] SD). The second temperature treatment, hereafter known as "elevated," was set up in a water bath ranging from 19[degrees]C to 23 [degrees]C (21.2 [+ or -] 0.4 [degrees]C, mean [+ or -] SD). Each temperature and salinity combination was replicated three times. The entire experiment was replicated across three independent male-female parental pairs. For each replicate, embryos were scored in one of four categories: (1) normal (Fig. 1A); (2) polyembryonic (Fig. 1B-E); (3) pseudo-polyembryonic (Fig. 1F); and (4) abnormal (Fig. 1G). Pseudo-polyembryonic phenotypes were defined as exhibiting a clear ingression during the blastula stage of development, thus reflecting an incomplete separation of the blastomeres earlier in development (Figs. IF, 2B, C). The pseudo-polyembryonic phenotype appeared to be equivalent to the "half-twin" morphology of Prionocidaris baculosa, figured in Mortensen (1938). In a second, related set of experiments, we examined the degree to which polyembryony was present in an additional 13 male-female pairs. The same conditions were used as described above, but only polyembryonic development and pseudo-polyembryonic development were scored. For analysis of the frequency of polyembryony and pseudo-polyembryony, all 16 male-female pairs across both experiments were used.

In order to test the developmental potential of polyembryonic zygotes, twin and normally developing embryos from the above experiments were isolated via mouth pipette. Isolated embryos were transferred to petri dishes for about 24 hours post-fertilization (hpf), after which time embryos were observed under a dissecting microscope (10 x magnification) to confirm normal larval development and swimming. Pairs of twins were isolated and cultured together to determine whether both members of a pair of twins could settle and metamorphose into the juvenile stage. To do this, each pair of twins was isolated and maintained in a 2-mL cuvette held in place in a test tube rack, which was then placed into the flow-through seawater tank to maintain larvae at ambient seawater temperature. Water changes were conducted every other day by transferring the larvae via mouth pipette into a new cuvette containing clean FSW. After each water change, larvae were fed a mixture of 2500 cells [mL.sup.-1] of each of three algal species: Dunaliella tertiolecta, Isochrysis galbana, and Rhodomonas lens. Algal densities were counted from each culture by using a hemocytometer, after which time appropriate volumes of algae were centrifuged for three to five minutes and the supernatant was pipetted off prior to the algal pellets being re-suspended in the larval cultures. Larvae were maintained until they reached competence (21 days old), at which point a few hundred grains of adult-conditioned sand were added to each cuvette to stimulate larval settlement (Highsmith, 1982).

Mechanisms of polyembryony induction

For fertilized zygotes from three independent male-female pairs, FED and ED were measured in each of three replicate bowls immediately after fertilization in order to determine whether osmotic swelling of the fertilization envelope provided blastomeres developing in reduced-salinity conditions with more physical space in which to split apart during early cleavage cycles. After about 5-7 hpf, all bowls were checked for signs of polyembryony. The development of embryos in each bowl was scored using the following seven categories: normal, twins, triplets, quadruplets, more (>4 embryos within a single fertilization envelope), pseudo-polyembryonic, or failed or arrested development (Fig. 1). Embryos that were identified as being polyembryonic then had a second measure of FED taken at 6 hpf for the elevated temperature treatment and at 12 hpf for the ambient temperature treatment (n = 10 embryos per replicate bowl). These times were chosen to represent a late stage of pre-hatching embryonic development.

We also tested the hypothesis that embryos incubated in reduced-salinity environments cannot maintain adequate cell-cell adhesion interactions during early development and that they subsequently drift apart after early cleavage cycles due to a lack of [Ca.sup.2+]. We termed this the reduced [Ca.sup.2+] hypothesis, and testing it required us to create seawater of pH and salinity equivalent to normal FSW, but with [Ca.sup.2+] replaced with other salts to create [Ca.sup.2+] reduced seawater ([Ca.sup.2+]RSW). To make the [Ca.sup.2+]RSW, we initially made calcium-free seawater, using a recipe detailed in Strathmann (1987) and modified by substituting deionized water for calcium-free seawater when making 26 ppt seawater. When used, [Ca.sup.2+]RSW was treated exactly the same as the standard salinity treatments of 32, 29, and 26 ppt and data were collected from this added salinity treatment as in the other treatments. A single male-female pair was used for this experiment.

In order to better understand at what point in the developmental cycle polyembryony was determined, we conducted transfer experiments where embryos were fertilized at one of two salinities and then periodically transferred to a new salinity. Embryos were fertilized using only 32 and 26 ppt salinity FSW, and they were maintained at elevated temperature conditions (~19-23 [degrees]C). Transfers were conducted at time intervals of 0, 15, 60, 90, and 240 minutes post-fertilization (mpf). Three reservoir bowls were used for each salinity treatment, each containing equal aliquots of 2-3 mL of the eggs obtained from the female used for that experiment. Eggs were fertilized in reservoir bowls and then immediately transferred via mouth pipette to bowls of the other salinity (T = transfer), and also to another bowl at the same salinity (TC = transfer control) for the 0-minute time point. For the subsequent transfer time points, embryos were always transferred by mouth pipette from the reservoir bowls and into the T or TC bowls. After 5-6 hpf, all bowls were scored for polyembryony, using methods described above.

Furthermore, we tested whether pseudo-polyembryonic embryos (Fig. 1F) of Dendraster excentricus could recover from the pseudo-twin phenotype and develop into phenotypically normal larvae. Gametes were obtained from adults and fertilized by using methods detailed above. Fertilized eggs were incubated at elevated temperature (19-23 [degrees]C) on the lab bench, in 9 glass bowls containing 100 mL of 32, 29, or 26 ppt artificial seawater (Instant Ocean salt mixture [Blacksburg, VA] combined with deionized [DI] water; ASW); and they were allowed to develop for about 4-5 hours. After this period, embryos were sorted into one of four developmental categories (Fig. 2): Class 1 = normal, Class 2 = pseudo-twin with <50% ingression, Class 3 = pseudo-twin with >50% ingression, and Class 4 = twin embryos. Embryos in each category were sorted via mouth pipette and individually maintained in 96-well plates with ASW. About 24-48 hpf, sorted larvae were scored for survival and normality of appearance.

Statistical analyses

All statistical analyses were conducted using SPSS Statistics version 22 (IBM, Armonk, NY). Means of ED and FED data obtained from replicate bowls were analyzed using a general linear mixed model. The analysis of ED data was carried out with trial modeled as a random factor, and salinity, temperature, and their interaction modeled as fixed factors. FED data were analyzed similarly to ED data, with the addition of time as another fixed factor, along with its interaction with salinity and temperature. Bonferroni-corrected post hoc pairwise comparisons were conducted to determine whether there were any differences between salinity treatments. Because normality of the data is an assumption of general linear mixed models (and ANOVAs in general), residuals were calculated, plotted, and tested for normality by using the Kolmogorov-Smirnov and Shapiro-Wilk tests. Data reported are assumed to be insignificant at P > 0.05 for both normality tests unless stated otherwise. Data that significantly deviated from both tests were transformed as described in the Results in order to meet the assumption of a normal distribution. In situations in which interaction terms were included in mixed models, interactions with P > 0.250 were excluded from the final analysis; and the model was re-run as a reduced model with only main effects (Quinn and Keough, 2002). Where appropriate, we also calculated measures of effect size (eta") for each fixed factor in our ANOVA models (Lakens 2013).

For analysis of the proportion of polyembryony and pseudopolyembryony, all data were arcsin-square root transformed to meet the assumptions of ANOVA. In the case of pseudopolyembryony data, the residuals of the transformed data were not significantly different from normal (Kolmogorov-Smirnov test, P = 0.200), while in the case of polyembryony data, the residuals of the transformed data remained significantly different from a normal distribution (Kolmogorov-Smirnov test, P < 0.001). However, we still present the ANOVA results for both because ANOVA is known to be robust to deviations from normality (Quinn and Keough, 2002).

Results

Field data and observations

Temperature and salinity data from the underwater monitoring station at FHL are reported in Figure 3. There were three major temperature spikes during episodes of freshwater input from the Fraser River over the course of the summer of 2014. These temperature increases occurred synchronously with declines in salinity as the relatively warm, fresh waters of the Fraser River plume spread south through the San Juan Archipelago. Temperatures reached a maximum of 15.5 [degrees]C, and salinity reached a minimum of 21.3 ppt during these events. Minimum salinities in the field are well below the 26 ppt minimum used in our experimental trials, but temperature data suggest that the maximum temperatures used in our lab data may have been unreasonably high (21.2 [degrees]C vs. 15.5 [degrees]C). However, the data in Figure 3 were collected about 21 km from the sand dollar beds and the shallow fjord where sand dollar embryos and larvae likely developed (Emlet, 1986) and at a greater depth than the intertidal sand dollar beds. To more directly assess the temperature and salinity conditions at the sand dollar beds, we collected held data en route to and at Orcas Island, San Juan. Field observations in East Sound indicate that sand dollars experience temperature and salinity fluctuations even during lulls in Fraser River episodes (Tables 1A, B; arrows, Fig. 3). Temperature data taken at the sand dollar beds indicate that animals experience temperature conditions that are indeed similar to our laboratory conditions. For example, on July 18, 2014, the mean water temperature directly above the sand dollar bed in East Sound was 21.0 [degrees]C, and the mean salinity was 31.6 ppt (Table 1B). The highest mean temperatures used in experimental laboratory trials were 21.2 [degrees]C, which closely match high temperatures observed in the field.

Induction of polyembryony

There were strong effects of salinity on the fertilization of Dendraster excentricus eggs. We found that fertilization declined from 90% or greater at 30-32 ppt down to about 50% at 26-28 ppt and then to near 0% at salinities of 24 ppt or lower (Fig. 4). For subsequent salinity trials, we selected 26 ppt as representative of an ecologically relevant low-salinity environment at which development was still viable.

Polyembryony data were analyzed across three replicate D. excentricus parental pairs in which embryos were scored as normal, polyembryonic, pseudo-polyembryonic, or failed development. An ANOVA showed that salinity, temperature, and their interaction significantly affected the percentage of normal, polyembryonic, and pseudo-polyembryonic development (Fig. 5A-C; Table 2A-C); but there was no significant effect of either treatment or their interaction on failed development (Fig. 5D; Table 2D). Bonferroni-corrected post hoc comparisons showed a significant difference between heated and ambient temperatures (P = 0.003) across all salinities (P < 0.008), except the 29 and 32 ppt salinity treatments, which did not significantly differ from one another (P = 0.826). Across 13 additional parental pairs, the frequency of polyembryony ranged from 0% to 48.53% (Fig. 6A). A two-way ANOVA in which temperature and salinity were coded as fixed effects, and parental pair as a random effect, showed that the two experimental treatments each had a significant effect on percent polyembryony (2-way ANOVA, [F.sub.1,38] = 4.932, P = 0.032 and [F.sub.2,29] = 4.059, P = 0.028, respectively); but their interaction did not (2-way ANOVA, [F.sub.2,29] = 1-745, P = 0.192). Bonferroni-corrected post hoc comparisons showed no significant differences between salinity treatments of 26 ppt to 29 ppt and 29 ppt to 32 ppt (P = 0.186 and P = 0.965, respectively), but comparing 26 ppt to 32 ppt revealed a significant salinity effect (P = 0.032). When parental pair was modeled as a fixed effect, there were no qualitative changes in the effects of temperature, salinity, and their interaction on percent polyembryony (2-way ANOVA, temperature: [F.sub.1,29] = 4.390, P = 0.045, [eta.sup.2] = 0.036; salinity: [F.sub.2,29] = 4.280, P = 0.024, [eta.sup.2] = 0.070; temperature x salinity: [F.sub.2,29] = 1.859, P = 0.174, [eta.sup.2] = 0.030); but parental pair was shown to have a significant effect as well ([F.sub.15,29] = 4.090, P = 0.001, [eta.sup.2] = 0.501). Across the same 13 parental pairs, the frequency of pseudo-polyembryony ranged from 0% to 68% (Fig. 6B). A two-way ANOVA, with temperature and salinity coded as fixed effects and parental pair coded as a random effect, showed that the two experimental treatments had a significant effect on the frequency of polyembryony (2-way ANOVA, [F.sub.1,44] = 7.190, P = 0.006, and [F.sub.2,20] = 3.271, P = 0.010, respectively). Bonferroni-corrected post hoc comparisons revealed no significant differences between salinities of 26 ppt to 29 ppt and 29 ppt to 32 ppt (P = 0.362 and P = 0.188, respectively), but a comparison of 26 ppt to 32 ppt salinity treatments was significant (P = 0.008). As above for polyembryony, we also modeled parental pair as a fixed effect (2-way ANOVA, temperature: [F.sub.1,29] = 13.155, P = 0.001, [eta.sup.2] = 0.158; salinity: [F.sub.2,29] = 3.046, P = 0.063, [eta.sup.2] = 0.073; temperature x salinity: [F.sub.2,29] = 0.671, P = 0.519, [eta.sup.2] = 0.016), which revealed a significant effect of parentage on the percent of embryos exhibiting pseudo-polyembryony ([F.sub.15,29] = 2.149, P = 0.038, [eta.sup.2] = 0.387)

Both members of several pairs of twin embryos were isolated and successfully cultured through the larval period, to demonstrate that both individuals in a pair of twins were viable and able to complete development to metamorphosis. Polyembryonic embryos appear, therefore, to be completely capable of continuing normal larval development and reaching the benthos as morphologically normal juveniles.

Mechanisms of polyembryony

Egg diameter and FED data were compiled from three replicate parental pairs (Fig. 7), and a two-way ANOVA revealed that salinity significantly affected ED (2-way ANOVA, [F.sub.2,48] = 130.903, P < 0.001) whereas temperature did not (2-way ANOVA, [F.sub.1,48] = 1.052, P = 0.310). When temperature treatments were pooled together, eggs maintained in 26 ppt seawater exhibited a 7-[micro]m increase in ED, equivalent to a 5% increase in ED and a 1.9% increase in egg volume. Bonferroni-corrected post hoc pairwise comparisons indicated that all three salinity treatments were significantly different from one another (P< 0.001).

When normality tests were conducted on FED data, both the Kolmogorov-Smirnov and Shapiro-Wilk tests showed significant departures from normality. To meet the ANOVA assumption of normality, FED data were [log.sub.10] transformed (Kolmogorov-Smirnov, P = 0.2; Shapiro-Wilk, P = 0.269). After the data transformation, a three-way ANOVA revealed that temperature, salinity, time, and their interaction did not significantly affect FED. When nonsignificant interactions (P > 0.250) were removed from the model, salinity and temperature still did not have a significant effect on FED (2-way ANOVA, [F.sub.2,80] = 1.198, P = 0.307, and [F.sub.1,80] = 1.892, P = 0.172, respectively). However, time did have a significant effect (2-way ANOVA, [F.sub.1,80] = 13.239, P < 0.001), such that FED2 was significantly greater than FED1. Taken together, these data suggest that fertilization envelopes continued to swell in the 6-12 hours after our initial observations of fertilization (6 hpf for the elevated temperature treatment and 12 hpf for the ambient temperature treatment).

In order to more directly test for a correlation between FED swelling and the production of this developmental response, FED2 data were collected only in treatments that produced polyembryonic embryos. We hypothesized that polyembryonic embryos would have larger FEDs, either as a result of the extra space required by the presence of multiple embryos or as an underlying cause of polyembryony. However, there was no significant difference between the mean ([+ or -] SE) FED of embryos exhibiting polyembryony (178.76 [+ or -] 1.88 [micro]m) and the FED of those that did not (180.19 [+ or -] 2.8 [micro]m; paired t-test, P = 0.581). FED2 was slightly smaller in polyembryonic embryos in three of the six bowls tested and was slightly larger in the remaining three bowls.

To elucidate at what developmental stage polyembryony was determined, embryos of D. excentricus were fertilized in 32 and 26 ppt seawater and were transferred into the alternate salinity at 4 different time intervals that were representative of key cleavage and/or developmental stages. The results of three experimental trials using three unique female-male pairs are presented for polyembryonic development (Fig. 8A) and for pseudo-polyembryonic development (Fig. 7B). For these 3 parental pairs, incidences of polyembryonic development were generally low, reaching a maximum of 10% in embryos that were control transferred from 26 ppt to 26 ppt seawater 15 mpf (Fig. 9A). Despite the low frequency of polyembryony overall, a two-way ANOVA revealed that there was a significant effect of treatment ([F.sub.4, 34] = 10.031, P < 0.001) but not of time ([F.sub.3, 34] = 0.342, P = 0.795) or of the interaction of treatment and time ([F.sub.*, 34] = 0.509, P = 0.858) on the proportion of embryos exhibiting this phenotype. Bonferroni-corrected post hoc comparisons showed that the 26 ppt--26 ppt transfer control treatment was significantly different from all other treatments except the 26 ppt reservoir. No other pairwise comparisons were significantly different. Similarly, a two-way ANOVA revealed that the incidence of pseudo-polyembryonic development varied significantly with treatment ([F.sub.4, 34] = 12.887, P < 0.001) but not with time ([F.sub.3, 34] = 1-250, P = 0.307) or with the interaction between treatment and time ([F.sub.9, 34] = 0.493, P = 0.869). Again, embryos in the 26 ppt-26 ppt transfer control treatment were significantly different from all other treatments except the 26 ppt reservoir (Bonferroni-corrected multiple pairwise comparisons, P< 0.05). The 26 ppt reservoir treatment was also significantly different from the 32 ppt reservoir and the 32 ppt-32 ppt transfer control treatments (P < 0.05).

At 240 mpf, embryos that were transferred from 26 ppt to 26 ppt exhibited a mean of 19.09% ([+ or -] 11.71%) pseudopolyembryonic development, compared to embryos that were transferred from 26 ppt to 32 ppt at the same time point, which exhibited a mean of 9.56% ([+ or -] 6.41%; Fig. 9B). This represents a 50% decrease in the frequency of this developmental phenotype across sibling embryos maintained under the same conditions, except for the final 2-3 hours prior to hatching. Embryos first fertilized in 32 ppt seawater and then transferred to 32 and 26 ppt seawater only rarely exhibited pseudopolyembryony (0.89% [+ or -] 0.59% and 3.33% [+ or -] 2.69%, respectively; Fig. 9B). Embryos that were never transferred from 26 and 32 ppt seawater exhibited pseudo-polyembryonic development 16.54% ([+ or -] 10.88%) and 0.22% ([+ or -] 0.22%) of the time, respectively (Fig. 9B).

To follow up on the observations obtained from the transfer experiments, experiments were conducted on pseudo-twin embryos of D. excentricus. Class 2 and 3 embryos that exhibited signs of pseudo-polyembryony (Fig. 2) developed into larvae that appeared normal with a high frequency (82.78%). Class 1 embryos (blastulae without ingression; Fig. 2) produced normal larvae with a similarly high frequency (83.93%). In contrast, offspring from Class 4 embryos (twins; Fig. 2) successfully continued to the larval stage at a much lower frequency (27.66%).

In contrast to the effects of reduced salinity on development, embryos fertilized and incubated in [Ca.sup.2+]RSW during early developmental stages did not exhibit any detectable differences in proportions of polyembryonic or pseudo-polyembryonic development, regardless of temperature treatment (Fig. 8). Polyembryony was rare in all treatments, and the frequency of abnormal development was slightly lower in [Ca.sup.2+]RSW treatments than in control treatments at both ambient and elevated temperature.

Discussion

This study reports polyembryony in embryos of Dendraster excentricus in response to elevated temperature and reduced salinity, and it represents one of the few reports of polyembryony within echinoid echinoderms (Mortensen, 1938; Allen et al., 2015). Embryos of D. excentricus were exposed to both increased temperatures (~20 [degrees]C) and decreased salinities (26 ppt), which, in combination, increased the frequency of polyembryony. Conversely, normal development was reported to decrease by 52% under these conditions, which could have great implications for the survival, fecundity, and overall fitness of members of this population.

While environmentally induced zygotic polyembryony is rarely described (reviewed by Allen et al., 2018), it has long been known that multiple embryos can be induced to develop from a single egg in the lab, either via the complete separation of blastomeres during early cell divisions (e.g., Horstadius, 1973) or through application of development-disrupting chemicals (e.g., Mazia, 1958). Mazia (1958) chemically induced a physical separation of blastomeres by exposing fertilized eggs of D. excentricus to mercaptoethanol during the first cell-division cycle. Vacquier and Mazia (1968a, b) demonstrated the effects of dithiothreitol (DTT) on the early development of D. excentricus and of two additional urchin species: Lytechinus pictus and Strongylocentrotus purpuratus. Again, embryos of D. excentricus produced twins at high frequencies (Vacquier and Mazia, 1968a), but the embryos of L. pictus and S. purpuratus exposed to similar experimental manipulations failed to develop properly (Vacquier and Mazia, 1968b). These results suggest that any environment in which embryos develop that decreases cell-cell communication and adhesion via filopodia may lead to polyembryony, but that there are species-specific (and perhaps lineage-specific) differences in response rates.

What, then, is the developmental milieu for embryos of D. excentricus in our study region? The waters of the San Juan Islands have intense tidal flows and are well mixed (Engie and Klinger, 2007). The flow and circulation of water in the San Juan Archipelago have been modeled using drift cards released at different locations and retrieved in diverse and distant areas (Klinger and Ebbesmeyer, 2001). Intense tidal flows, in combination with regular intrusions of freshwater from the Fraser River, may have great implications for the fauna of this region. Data collected in the current study validated experimental conditions used in the laboratory. Water temperatures over beds of adult sand dollars in East Sound, Orcas Island, San Juan, reached temperatures of about 20 [degrees]C, closely resembling experimental temperature treatments. Salinity conditions at an underwater monitoring station at Friday Harbor Laboratory reached minimums (21 ppt) that were well below the salinity treatments used in laboratory conditions (26 ppt). Importantly, during inputs of freshwater from the Fraser River, increases in temperature nearly exactly mirrored decreases in salinity (Fig. 3), suggesting that our two experimental variables closely covary in the field. Ideally, we would have continuously monitored salinity and temperature directly over the sand dollar bed and throughout East Sound, but the data we were able to collect both in East Sound and at FHL suggested that the laboratory conditions to which we subjected embryos approximated (temperature) or were less extreme than (salinity) the range of values experienced in nature. Despite these efforts, the natural frequency of polyembryony and other lab-documented developmental phenomena is generally unknown from field-collected embryos. Future studies should focus on collection of embryos from the field to determine the extent and type of developmental irregularities in the field. This is likely to be a fruitful area of future research for free-spawning animals in general.

Nearshore marine environments may be commonly affected by terrestrial sources of freshwater input. For example, the Great Barrier Reef (GBR) in Australia receives freshwater input from the Burdekin River that can affect flushing of the GBR lagoon and can drastically change salinity conditions experienced by organisms living in the area (Wolanski and Jones, 1981). Similarly, low freshwater input to the Chesapeake Bay in 2002 was associated with increased water quality, while extremely high levels of freshwater input in 2003 corresponded to a decrease in water quality (Acker et al., 2005). Given that nearshore organisms may frequently encounter fluctuating temperature and salinity conditions, it is conceivable that they have experienced strong selection to cope with these environmental changes. Fluctuations in temperature and salinity are widely known to affect fertilization, embryonic development, larval development, and juvenile mortality in echinoderms (Greenwood and Bennett, 1981; Watts et al., 1982; Allen and Pechenik, 2010; for a comprehensive review see Russell, 2013).

If polyembryony is a developmental response to conditions experienced in nature, what are the consequences for embryos, and how might we explain polyembryony mechanistically? The consequences of successful twinning events can be estimated from prior manipulations using experimental embryology. Artificial manipulations of embryos at the two-cell stage have shown that lab-induced twinning yields viable larvae and, ultimately, viable juveniles, but that these offspring take longer to develop and are smaller at settlement (e.g., Sinervo and McEdward, 1988; Hart, 1995; Alcorn and Allen, 2009; Allen, 2012). Our observations, that both embryos from a single egg can complete development to metamorphosis, align well with these and other studies of artificially twinned embryos (e.g., Cameron et al., 1996); and they imply that development is not halted as a consequence of polyembryony. Larvae, however, may still incur costs, such as increased time spent in the plankton, to acquire the necessary nutrients to fuel metamorphosis and post-metamorphic growth (Allen, 2012). This prolonged planktonic period might increase the risk of predation (Vaughn and Allen, 2010), as could the smaller size of larvae associated with polyembryonic development (Allen, 2008). Alternatively, for at least one species of echinoid, larval polyembryony may be an adaptive response to predator cues in the environment (Vaughn and Strathmann, 2008), resulting in smaller larvae that are less vulnerable to visual fish predators (Allen, 2008; Vaughn, 2010).

The mechanisms underlying polyembryony are less clear than the consequences. We hypothesized that a disruption in cell-cell adhesion due to reduced-salinity conditions might be the causative agent. Cadherins, calcium-dependent adhesion proteins, are cell-surface proteins that maintain pathways of communication and adhesion between cells (Lodish et al., 2000). After manipulating blastomere interactions by using chemical treatments, Mazia (1958) turned to investigating embryonic development in the context of a modified ionic environment. Fertilized eggs of D. excentricus were placed in calcium-free seawater and left to divide until the four-cell stage, before being returned to "normal" seawater (Mazia, 1958). Mazia (1958) observed that embryos temporarily exposed to calcium-free seawater did result in the production of twins. However, if embryos were left to continue dividing in calcium-free seawater, embryos would not twin, thus suggesting an embryonic ability to reorganize and continue with normal development (Mazia, 1958). Whereas Mazia (1958) noted a visible effect of the mercaptoethanol treatment described above on the adhesion of blastomeres, he suggested that calcium-free seawater merely softens these connections but does not dissolve them.

We attempted to examine the effects of calcium concentrations in seawater on the development of D. excentricus. We hypothesized that reduced-salinity seawater had reduced concentrations of free calcium ions ([Ca.sup.2+]), which might limit adhesion potential in cells at the two-cell stage, causing them to fall apart and continue on two independent developmental paths. However, embryos cultured in [Ca.sup.2+]RSW did not produce multiples in high frequencies. Therefore, reduction in [Ca.sup.2+] alone does not seem to explain the polyembryony we observed. However, additional experimentation focused on blocking calcium-dependent adhesion molecules is necessary to further test this hypothesis, because these molecules have been shown to play an important role in later embryonic development in sea urchins (Miller and McClay, 1997a, b). Distinguishing the effects of cell-cell adhesion molecules from cell-matrix adhesion molecules will be especially important, given that there are hundreds of genes that have been identified as being involved in adhesion in echinoderms (Whittaker et al., 2006). The result that reduced [Ca.sup.2+] does not induce polyembryony is not in conflict with prior studies demonstrating that calcium-free seawater can be used to artificially create twins in the lab (e.g., Allen, 2008, 2012), because those studies completely removed calcium from seawater rather than merely reducing its concentration. The threshold level of calcium ions at which blastomeres will separate is currently unknown. Thus, much more work can be done at both the molecular and biochemical levels to understand the basic mechanisms underlying polyembryony.

One alternative hypothesis is that the hyaline layer plays some role in preventing or permitting polyembryony. The hyaline layer surrounds the developing embryo and maintains cells' proximity to one another through microvillar processes that extend from the cells' apical surfaces (Citkowitz, 1971; Hall and Vacquier, 1982). What has been qualitatively described as a "weak" interaction between blastomeres and the hyaline layer in irregular echinoids, including D. excentricus (Mazia, 1958; Vacquier and Mazia 1968a, b), potentially allows for polyembryonic phenotypes. In contrast, McClay and Fink (1982) have demonstrated that, in the early development of regular echinoids, cell-hyaline interactions were even stronger than cell-cell interactions. The strength of this interaction weakens as development progresses and as cell-cell interactions become much stronger (McClay and Fink, 1982), but the hyaline layer appears to be a critical substrate for development in early embryos. Embryos of cidaroid urchins that lack a hyaline layer entirely (Schroeder, 1981) exhibit frequent polyembryony (Mortensen, 1938; Allen et al., 2015), as would be expected if hyaline played a role in maintaining embryo integrity.

To help understand the underlying mechanisms of polyembryony, several transfer experiments were conducted to determine the stage at which polyembryony is induced in embryos of D. excentricus. The results from these transfer experiments indicated an unexpected degree of plasticity in pseudo-polyembryonic development. At the beginning of the study, pseudo-polyembryonic individuals were virtually ignored because they were not the primary focus of the study. Embryos that were fertilized at 26 ppt and transferred to 32 ppt seawater at 240 mpf showed an average decrease of roughly 10% pseudo-polyembryonic development when compared to embryos that were control transferred from 26 ppt to 26 ppt seawater. This suggests that embryos are capable of changing their developmental morphology even as late as 4 hpf, which is roughly 2 hours prior to when hatching begins in embryos maintained at elevated temperatures. Preliminary experiments were conducted on embryos of the pencil urchin Eucidaris tribuloides, in which over 90% of sorted Class 2 and Class 3 pseudo-twin embryos were observed to reach the larval stage successfully (STA-R, pers. obs.). The shift in pseudo-polyembryonic embryos to normal embryos of D. excentricus even late in development suggests some sort of shift in cellular organization on the part of pseudo-polyembryonic embryos that is plastic in response to fluctuations in environmental conditions. However, this assertion needs to be investigated in further detail. Additionally confounding this observation, when embryos were transferred from 32 to 26 ppt seawater at 240 mpf, percent pseudo-polyembryony remained quite low (only a 2.44% increase with transfer to decreased salinities). This discrepancy in percent pseudo-polyembryony across different transfer treatments indicated that embryos initially exposed to favorable conditions become fixed in their response, whereas embryos initially exposed to less favorable conditions maintained some level of plasticity.

Embryonic development is not necessarily being halted or impaired by experimental manipulations, which suggests that the embryos are able to cope with their relatively stressful environments. Allen et al. (2015) have shown in embryos of Echinarachnius parma and Eucidaris tribuloides that a combination of decreased salinities and increased temperatures elicits similar polyembryonic responses. In this study, it is important to note that polyembryony is highly variable across D. excentricus parental pairs. For example, we observed about 80% polyembryony in one replicate bowl, and in another experimental trial we observed 0% polyembryony under identical conditions. This indicates that there is significant variation within this population with regard to embryonic propensity to produce multiples. Whether this trait is heritable is unknown. However, if propensity of polyembryony is heritable, then evolutionary forces could alter the abundance of this trait in the population, causing the presence of this trait to increase or decrease, depending on the increase or decrease in fitness this trait might confer. Studying the heritability and underlying mechanisms controlling these responses would further our understanding of the potentially adaptive benefits of these traits.

Acknowledgments

We thank student members of the Allen Lab for their assistance and insight throughout this project, especially Stacy Trackenberg, who was critical to the success of this work. We also thank the William & Mary Charles Center for a Summer Research Grant and Honors Fellowship via the Batten Foundation to STA-R. This work was also funded by a National Science Foundation grant (DEB 1257039) to JDA. Additionally, we thank the University of Washington Friday Harbor Laboratories for allowing access to research facilities and equipment and Dr. Emily Carrington for field temperature and salinity data.

Literature Cited

Abouheif, E., M. J. Fave, A. S. Ibarraran-Viniegra, M. P. Lesoway, A. M. Rafiqi, and R. Rajakumar. 2014. Eco-evo-devo: The time has come. Pp. 107-125 in Ecological Genomics: Ecology and the Evolution of Genes and Genomes, C. R. Landry and N. Aubin-Horth. eds. Springer Science, Dordrecht.

Acker, J. G" L. W. Harding, G. Leptoukh, T. Zhu, and S. Shen. 2005. Remotely-sensed chl a at the Chesapeake Bay mouth is correlated with annual freshwater flow to Chesapeake Bay. Geophys. Res. Lett. 32: 1-4.

Alcorn, N. J., and J. D. Allen. 2009. How do changes in parental investment influence development in echinoid echinoderms? Evol. Dev. 11: 719-727.

Allen, J. D. 2008. Size specific predation on marine invertebrate larvae. Biol. Bull. 214: 42-49.

Allen, J. D. 2012. Effects of egg size reductions on development time and juvenile size in three species of echinoid echinoderms: implications for life history theory. J. Exp. Mar. Biol. Ecol. 422/423: 72-80.

Allen, J. D., and J. A. Pechenik. 2010. Understanding the effects of low salinity on fertilization success and early development in the sand dollar Echinarachnius parma. Biol. Bull. 218: 189-199.

Allen, J. D., A. F. Armstrong, and S. L. Ziegler. 2015. Environmental induction of polyembryony in echinoid echinoderms. Biol. Bull. 229: 221-231.

Allen, J. D., A. M. Reitzel, and W. Jaeckle. 2018. Asexual reproduction of marine invertebrate eggs, embryos and larvae. Pp. 67-81 in Evolutionary Ecology of Marine Invertebrate Larvae, T. J. Carrier, A. M. Reitzel, and A. Heyland. eds. Oxford University Press, Oxford.

Allen, J. D., E. L. Richardson, D. Deaker, A. Aguera, and M. Byrne. 2019. Larval cloning in the crown-of-thorns sea star, a keystone coral predator. Mar. Ecol. Prog. Ser. 609: 271-276.

Balser, E. J. 1998. Cloning by ophiuroid echinoderm larvae. Biol. Bull. 194: 187-193.

Bosch, I., R. B. Rivkin, and S. P. Alexander. 1989. Asexual reproduction by oceanic planktotrophic echinoderm larvae. Nature 337: 169-170.

Cameron, R. A., P. S. Leahy, and E. H. Davidson. 1996. Twins raised from separated blastomeres develop into sexually mature Strongylocentrotus purpuratus. Dev. Biol. 178: 514-519.

Citkowitz, E. 1971. The hyaline layer: its isolation and role in echinoderm development. Dev. Biol. 42: 348-362.

Craig, S. F., L. B. Slobodkin, G. A. Wray, and C. H. Biermann. 1997. The "paradox" of polyembryony: a review of the cases and a hypothesis for its evolution. Evol. Ecol. 11: 127-143.

Eaves, A. A., and A. R. Palmer. 2003. Widespread cloning in echinoderm larvae. Nature 425: 146.

Emlet, R. B. 1986. Larval production, dispersal and growth in a fjord: a case study on larvae of the sand dollar Dendraster excentricus. Mar. Ecol. Prog. Ser. 31: 245-254.

Engie, K., and T. Klinger. 2007. Modeling passive dispersal through a large estuarine system to evaluate marine reserve network connections. Estuar. Coasts 30: 201-213.

Gilbert, S. F. 2001. Ecological developmental biology: Developmental biology meets the real world. Dev. Biol. 233: 1-12.

Gilbert, S. F., and J. A. Bolker. 2003. Ecological developmental biology: preface to the symposium. Evol. Dev. 5: 3-8.

Grbic, M., L. M. Nagy, and M. R. Strand. 1998. Development of polyembryonic insects: a major departure from typical insect embryogenesis. Dev. Genes Evol. 208: 69-81.

Greenwood, P. J., and T. Bennett. 1981. Some effects of temperature-salinity combinations on the early development of the sea urchin Parechinus angulosus (Leske). Fertilization. J. Exp. Mar. Biol. Ecol. 51: 119-131.

Hall, H. G., and V. D. Vacquier. 1982. The apical lamina of the sea urchin embryo: major glycoproteins associated with the hyaline layer. Dev. Biol. 89: 168-178.

Hart, M. W. 1995. What are the costs of small egg size for a marine invertebrate with feeding planktonic larvae? Am. Nat. 146: 415-426.

Heyward, A. J., and A. P. Negri. 2012. Turbulence, cleavage and the naked embryo: a case for coral clones. Science 335: 1064.

Highsmith, R. C. 1982. Induced settlement and metamorphosis of sand dollar (Dendraster excentricus) larvae in predator free sites: adult sand dollar beds. Ecology 63: 329-337.

Horstadius, S. 1973. Experimental Embryology of Echinoderms. Clarendon Press, Oxford.

Hughes, R. N., M. E. D'Amato, J. D. D. Bishop, G. R. Carvalho, S. F. Craig, L. J. Hansson, M. A. Harley, and A. J. Pemberton. 2005. Paradoxical polyembryony? Embryonic cloning in an ancient order of marine bryozoans. Biol. Lett. 1: 178-180.

Jaeckle, W. B. 1994. Multiple modes of asexual reproduction by tropical and subtropical sea star larvae: an unusual adaptation for genet dispersal and survival. Biol. Bull. 186: 62-71.

Just, E. E. 1937. Phenomena of embryogenesis and their significance for a theory of development and heredity. Am. Nat. 71: 97-112.

Klinger, T., and C. Ebbesmeyer. 2001. Using oceanographic linkages to guide marine protected area network design. lOnlineJ. Proceedings of the 2001 Puget Sound Research Conference. Olympia, Washington. Available: https://archives.eopugetsound.org/conf/2001 PS_ResearchConference/sessions/oral/4a_kling.pdf [2019. October 14].

Knott, K. E., E. J. Balser, W. B. Jaeckle, and G. A. Wray. 2003. Iden tification of asteroid genera with species capable of larval cloning. Biol. Bull. 204: 246-255.

Lakens, D. 2013. Calculating and reporting effect sizes to facilitate cumulative science: a practical primer for t-tests and ANOVAs. Front. Psychol. 4: 863.

Lodish, H., A. Berk, S. L. Zipursky, P. Matsudaira, D. Baltimore, and J. Darnell. 2000. Molecular Cell Biology, 4th ed. W.H. Freeman, New York.

Mazia, D. 1958. The production of twin embryos in Dendraster by means of mercaptoethanol (monothioethylene glycol). Biol. Bull. 114: 247-254.

McClay, D. R., and F. D. Fink. 1982. Sea urchin hyaline: appearance and function in development. Dev. Biol. 92: 285-293.

Miller, J. R., and D. R. McClay. 1997a. Changes in the pattern of adherens junction-associated [beta]-catenin accompany morphogenesis in the sea urchin embryo. Dev. Biol. 192: 310-322.

Miller, J. R., and D. R. McClay. 1997b. Characterization of the role of cadherin in regulating cell adhesion during sea urchin development. Dev. Biol. 192: 323-339.

Moczek, A. P., S. Sultan, S. Foster, C. Ledon-Rettig, I. Dworkin, H. F. Nijhout, E. Abouheif, and D. W. Pfennig. 2011. The role of developmental plasticity in evolutionary innovation. Proc. R. Soc. B Biol. Sci. 278: 2705-2713.

Mortensen, T. 1938. Contributions to the study of development and larval forms of echinoderms. IV. Mem. Acad. R. Sci. Lett. Dan., Cph. 7: 1-59.

Ode, P. J., and M. R. Strand. 1995. Progeny and sex allocation decisions of the polyembryonic wasp Copidosoma floridanum. J. Anim. Ecol. 64: 213-224.

Pia, T. S., T. Johnson, and S. B. George. 2012. Salinity-induced morphological changes in Pisaster ochraceus (Echinodermata: Asteroidea) larvae. J. Plankton Res. 34: 590-601.

Pigliucci, M. 2001. Phenotypic Plasticity: Beyond Nature and Nurture. Johns Hopkins University Press, Baltimore.

Quinn, G. P., and M. J. Keough. 2002. Experimental Design and Data Analysis for Biologists. Cambridge University Press, Cambridge.

Russell, M. P. 2013. Echinoderm responses to variation in salinity. Adv. Mar. Biol. 66: 171-212.

Schroeder, T. E. 1981. Development of a "primitive" sea urchin (Eucidaris tribuloides): irregularities in the hyaline layer, micromeres and primary mesenchyme. Biol. Bull. 161: 141-151.

Sinervo, B., and L. R. McEdward. 1988. Developmental consequences of an evolutionary change in egg size: an experimental test. Evolution 42: 885-899.

Strathmann, M. F. 1987. Reproduction and Development of Marine Invertebrates of the Northern Pacific Coast: Data and Methods for the Study of Eggs, Embryos, and Larvae. University of Washington Press, Seattle.

Sultan, S. E. 2003. The promise of ecological developmental biology. J. Exp. Zool. B Mol. Dev. Evol. 296: 1-7.

Sultan, S. E. 2007. Development in context: the timely emergence of ecodevo. Trends Ecol. Evol. 22: 575-582.

Sutherland, D. A., P. MacCready, N. S. Banas, and L. F. Smedstad. 2011. A model study of the Salish Sea estuarine circulation. J. Phys. Oceanogr. 41: 1125-1143.

Vacquier, V. D., and D. Mazia. 1968a. Twinning of sand dollar embryos by means of dithiothreitol: the structural basis of blastomere interactions. Exp. Cell Res. 52: 209-219.

Vacquier, V. D., and D. Mazia. 1968b. Twinning of sea urchin embryos by treatment with dithiothreitol: roles of cell surface interactions and of the hyaline layer. Exp. Cell Res. 52: 459-468.

Vaughn, D. 2010. Why run and hide when you can divide? Evidence for larval cloning and reduced larval size as an adaptive inducible defense. Mar. Biol. 257: 1301-1312.

Vaughn, D., and J. D. Allen. 2010. The peril of the plankton. Integr. Comp. Biol. 50: 552-570.

Vaughn, D., and R. R. Strathmann. 2008. Predators induce cloning in echinoderm larvae. Science 319: 1503.

Vickery, M. S., and J. B. McClintock. 2000. Effects of food concentration and availability on the incidence of cloning in planktotrophic larvae of the sea star Pisaster ochraceus. Biol. Bull. 199: 298-304.

Watts, S. A., R. E. Scheibling, A. G. Marsh, and J. B. McClintock. 1982. Effect of temperature and salinity on larval development of sibling species of Echinaster (Echinodermata: Asteroidea) and their hybrids. Biol. Bull. 163: 348-354.

West-Eberhard, M. J. 2003. Developmental Plasticity and Evolution. Oxford University Press. New York.

Whittaker, C. A., K. F. Bergeron, J. Whittle, B. P. Brandhorst, R. D. Burke, and R. O. Hynes. 2006. The echinoderm adhesome. Dev. Biol. 300: 252-266.

Wolanski, E., and M. Jones. 1981. Physical properties of Great Barrier Reef lagoon waters near Townsville. I. Effects of Burdekin River floods. Aust. J. Mar. Freshw. Res. 32: 305-319.

S. T. ABDEL-RAHEEM AND JONATHAN D. ALLEN (*)

Department of Biology, William & Mary, Williamsburg, Virginia

Received 14 May 2019; Accepted 30 September 2019; Published online 16 December 2019.

(*) To whom correspondence should be addressed. Email: jdallen@wm.edu.

Abbreviations: ASW, artificial seawater; [Ca.sup.2+]RSW, calcium-reduced sea-water; ED, egg diameter; [eta.sup.2], effect size; FE, fertilization envelope; FED, fertilization envelope diameter; FHL, Friday Harbor Laboratories; FSW, filtered seawater; GBR, Great Barrier Reef; hpf, hours post-fertilization; mpf, minutes post-fertilization; T, transfer; TC, transfer control.

DOI: 10.1086/706607
Table 1A
Field data collected on July 18, 2014

Location                 Time (h)   Depth (ft)   Salinity   Temperature
                                                 (ppt)      ([degrees]C)

48[degrees]38.365'N,     15:44       4           31.6       21.0
122[degrees]52.652"W
48[degrees]38.365'N,     15:47       4           31.2       20.2
122[degrees]52.625"W
Madrona Point            15:50      10           31.7       17.5
Middle of East Sound     16:04      40           32.3       15.0
  48[degrees]38.593'N,
  122[degrees]52.995'W
Front of East Sound      16:12      95           32.5       13.2
  48[degrees]37.007'N,
  122[degrees]51.703'W

Table 1B
Field data collected on July 25, 2014

Location                        Time (h)   Salinity   Temperature
                                           (ppt)      ([degrees]C)

White Beach                     09:51      32.4       14.5
Rosario                         09:59      32.3       14.9
Madrona Point                   10:11      32.1       16.4
East Sound (sand dollar bed)    10:51      31.9       19.4
East Sound (sand dollar bed)    10:52      31.9       19.5
East Sound (sand dollar bed)    10:54      31.6       20.3
East Sound (sand dollar bed)    10:57      31.7       19.3
East Sound (sand dollar bed)    10:59      31.8       19.1
East Sound (sand dollar bed)    11:00      31.8       20.2
Midway between sand dollar      11:24      31.9       17.7
bed and Madrona Point
(28-ft depth)
Madrona Point (36.7-ft depth)   11:27      32.0       17.0
Rosario                         11:43      32.3       15.3
White Beach                     11:53      33.0       12.4

Table 2
Analysis of the effects of temperature, salinity, and parental pair on
(A) normal, (B) polyembryonic, (C) pseudo-polyembryonic, and (D) failed
development

                   Numerator  Denominator  F-value  P-value  [eta.sup.2]
                   df         df

(A) Normal
  Temperature      1          10           33.149   0.001    0.195
  Salinity         2          10           47.559   0.001    0.279
  Temperature      2          10            8.504   0.007    0.099
  x salinity
  Parental pair    2          10            7.485   0.010    0.088
(B) Polyembryonic
  Temperature      1          10           15.143   0.003    0.206
  Salinity         2          10           14.420   0.001    0.393
  Temperature      2          10            6.803   0.014    0.186
  x salinity
  Parental pair    2          10            2.763   0.111    0.076
(C) Pseudo
-polyembryonic
  Temperature      1          12           16.360   0.002    0.318
  Salinity         2          12            8.239   0.008    0.321
  Temperature      2          12            3.296   0.080    0.129
  x salinity
  Parental pair    2          12            0.993   0.404    0.038
(D) Failed
development
  Temperature      1          10            0.002   0.962    0.001
  Salinity         2          10            1.889   0.201    0.064
  Temperature      2          10            0.193   0.827    0.007
  x salinity
  Parental pair    2          10            1.314   0.311    0.044

ANOVAs were conducted on embryonic data generated from three
experimental male-female pairs of Dendraster excentricus. Effect sizes
are presented as [eta.sup.2] for each factor tested in our model, df,
degree of freedom.
COPYRIGHT 2019 University of Chicago Press
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2019 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Abdel-Raheem, S.T.; Allen, Jonathan D.
Publication:The Biological Bulletin
Date:Dec 1, 2019
Words:9058
Previous Article:Regulation of Metamorphosis by Environmental Cues and Retinoic Acid Signaling in the Lecithotrophic Larvae of the Starfish Astropecten latespinosus.
Next Article:Arrested Sexual Development in Queen Conch (Lobatus gigas) Linked to Abnormalities in the Cerebral Ganglia.
Topics:

Terms of use | Privacy policy | Copyright © 2020 Farlex, Inc. | Feedback | For webmasters