Effect of Sperm Ratio and Temperature on Fertilization and Early Larval Development of the Surf Clam Mesodesma donacium (Bivalvia:Mesodesmatidae)

Simple Summary The surf clam Mesodesma donacium, known in the Southeastern Pacific as “macha”, has experienced decreased availability for local consumption due to high levels of extraction for many years in Chile and the effect of El Niño Southern Oscillation (ENSO) in Perú. To compensate the population decline, the advancement of the development of technology for cultivation and commercial production is proposed. Therefore, the development of small-scale aquaculture with the participation of artisanal fishermen and small fish farmers is considered. In this work, progress is made in determination of the conditions for fertilization and the effect of temperature on early embryonic and larval development. Abstract The effect of sperm ratio on fertilization was evaluated in five sperm:oocytes treatments (10:1, 50:1, 100:1, 500:1 and 1000:1), the effect of temperature on embryonic and larval development in three temperature treatments (13 °C, 16 °C and 19 °C) was recorded and the duration of each stage, the growth rate and survival rate were registered. The oocytes were spherical (67.5 ± 4.2 μm) with a defined nucleus. Spermatozoa had a circular head (2 μm) and a fusiform flagellum (12 μm). The 500:1 sperm:oocytes treatment presented the lowest % of unfertilized oocytes, and lysis was observed in the 1000:1 treatment. An inverse relationship was observed between temperature and the duration of the stages of embryonic development. At 16 °C, veliger D larvae were observed at 41 h 45′ pf (88 ± 13.0 μm). Umbonate larvae were obtained at day 16 in the 13 °C culture and at day 10 in the 16 °C and 19 °C treatment (140 μm). On day 16 of culture, advanced umbonate larvae with a well-defined stomach (235 μm) were observed. The larval growth rate was higher in the 19 °C treatment (3.6 μm day−1) than the 13 °C and 16 °C treatment (2, 2.2 μm day−1). The mortality was higher in the 19 °C treatment (91%). These results are an initial contribution towards the culture of M. donacium as part of small-scale aquaculture in South America.


Introduction
Mesodesma donacium (Lamarck, 1818) (Bivalvia:Mesodesmatidae), commonly referred to as "macha" in Chile and Perú, is an endemic surf clam species in the Chile-Perú Malacological Zoogeographic Province [1], originally distributed in South America from Sechura Bay in northern Perú (5 • S) to the Inio River at the southern tip of Chiloé Island in Chile (43 • S) [2,3]. Nevertheless, after the extraordinary El Niño Southern Oscillation (ENSO) Chiloé Island during the maturation period in austral winter (June 2018). The specimens were transported to the hatchery of Copulhue SpA, located on Quinchao Island, and maintained in laboratory conditions until spawning. This was performed in a close water flow system with filtered seawater at 10 µm at a temperature of 13.3 ± 2.3 • C. The broodstock was randomly divided into two 1000 L tanks. A 30 L tray was placed at the top of each tank, and a 10 cm layer of fine sand was added as a substrate to each tray. Specimens were fed daily with a diet based on Isochrysis lutea and Chaetoceros muelleri in a ratio of 1:1 with a final concentration of 250,000 mL −1 cells. A visual examination was performed every week to determine changes in the size of the gonads in terms of sexual maturation, as per [28].
For the experiments, both female and male gametes were obtained by the "stripping" method. Successive cuts with a scalpel were performed on the gonad with caution so as not to cut the digestive gland and to avoid contamination. Subsequently, the gametes were dragged with a wash bottle with seawater filtered at 1 µm and stored in a 250 mL beaker where they were hydrated for 15 min for observation and measurement through a trinocular optical microscope (AMSCOPE 40C-200 x, AmScope, Irvine, CA, USA) with graduated eyepiece equipped with a digital camera using IPCapture software 9.1 (IMT, Burnaby, BC, Canada). For the evaluation of the quality of the gametes, motility of sperm and shape and size of oocytes were observed. Oocytes were counted in a Sedgewick Rafter egg count chamber (1 mL) and sperm in a 0.1 mL Neubauer count chamber.

Effect of Sperm Ratio on Fertilization
To assess the effect of sperm concentration on fertilization, increasing proportions of sperm per oocyte were used. Five treatments were used in a ratio of 10:1, 50:1, 100:1, 500:1 and 1000:1 sperm:oocytes, respectively. Glass containers of 250 mL with 200 mL of seawater filtered at 1 µm with a concentration of 20 oocytes per ml were used as per Helm, Bourne and Lovatelli [17]. Each treatment considered three replicates that were kept in a thermoregulated bath at 12 • C. Twenty-four hours post fertilization (pf), 3 samples of 1 mL were taken for every replica in each treatment, and the number of fertilized and non-fertilized oocytes was determined by trinocular optical microscope.

Effect of Temperature on Embryonic and Larval Development
After obtaining and quantifying gametes, the fertilization process was performed in plastic containers of 1200 mL with 1000 mL of filtered seawater in thermoregulated baths with a sperm/oocyte ratio of 100:1. Three temperature treatments were used, a 13 • C (12.73 ± 1.23) (autumn-winter temperature), a 16 • C (16.03 ± 0.72) (spring-summer temperature) and, finally, a high temperature in the zone of 19 • C (18.91 ± 0.65). Embryonic development was observed from fertilization until the presence of D larvae, approximately 48 h post fertilization (pf), by taking 1 mL subsamples for each treatment. The observations were made every 15 min during the first 3 h and every 30 min until D larvae were obtained. The duration of each stage was determined when 50% of the organisms attained a certain stage. The stages of development were observed and photographed under an optical microscope.
The same temperature treatments used in embryonic development in triplicate were used for larval development (from D larvae to umbonate larvae). For larval culture, an initial density of 4 larvae per mL was used with the seawater filtered (1 µm) and water exchanged every other day. The larvae were maintained through mixed feeding with microalgae (flagellates and diatoms) at a concentration of 20,000 cel/mL −1 [18], and growth rate and survival were evaluated by taking subsamples of 1 mL.

Data Analysis and Statistical Tests
The normality of the data was determined by a Shapiro-Wilk test and the homoscedasticity by a Levene test. The fertilization data (%) were transformed to the arcsine value for subsequent analysis. A one-way analysis of variance was used to identify significant

Data Analysis and Statistical Tests
The normality of the data was determined by a Shapiro-Wilk test and the homoscedasticity by a Levene test. The fertilization data (%) were transformed to the arcsine value for subsequent analysis. A one-way analysis of variance was used to identify significant differences between fertilized gametes in the five sperm:oocyte treatments. A significance level of α = 0.05 was used for all statistical analyses.

Results
Of the gametes extracted using the stripping technique, an average of 322,528 ± 202,987 oocytes (n = 4) and 1,583,333 ± 418,911 spermatozoa (n = 3) was obtained. The female and male gonads did not differ in color or shape according to the naked eye, so microscopic analysis was required. The oocytes were spherical, with an average size of 67.5 ± 4.2 μm (n = 30), a defined nucleus and without a jelly coat. Spermatozoa were characterized by a rounded head (2 μm) and a fusiform flagellum (12 μm) ( Figure 1).

Effect of Sperm Ratio on Fertilization Success
The five sperm:oocyte ratio treatments showed significant differences in fertilization percentage (one-way ANOVA, df = 35, F = 27.75, p < 0.001). A Tukey's a posteriori analysis (unequal N HSD) showed that the fertilization percentage ratios of 10:1 and 50:1 were significantly lower than the other sperm:oocyte ratios (≥100:1 sperm:oocyte ratio) ( Figure  2). Treatments with the highest ratio of sperm per oocyte showed the lowest percentages of unfertilized oocytes. The presence of oocytes in lysis was observed in one replicate of the treatment with a ratio of 1000:1 sperm:oocyte.

Effect of Sperm Ratio on Fertilization Success
The five sperm:oocyte ratio treatments showed significant differences in fertilization percentage (one-way ANOVA, df = 35, F = 27.75, p < 0.001). A Tukey's a posteriori analysis (unequal N HSD) showed that the fertilization percentage ratios of 10:1 and 50:1 were significantly lower than the other sperm:oocyte ratios (≥100:1 sperm:oocyte ratio) ( Figure 2). Treatments with the highest ratio of sperm per oocyte showed the lowest percentages of unfertilized oocytes. The presence of oocytes in lysis was observed in one replicate of the treatment with a ratio of 1000:1 sperm:oocyte.

Effect of Temperature on Embryonic and Larval Development
The embryonic development of M. donacium showed differences in the duration of the stages in the different temperature treatments. In general, an increase in temperature decreased the length of the different stages. Fecundation was characterized by the pres-

Effect of Temperature on Embryonic and Larval Development
The embryonic development of M. donacium showed differences in the duration of the stages in the different temperature treatments. In general, an increase in temperature decreased the length of the different stages. Fecundation was characterized by the presence of a fertilization membrane with a diameter of 68 ± 4.5 µm at 15 and 30 pf in the treatment at 16 • C and 19 • C, respectively. At 13 • C, the rotatory blastula stage was observed at 10 h pf, while, at 16 • C and 19 • C, this phase was reached in 8 h 30 and 7 h 30 pf, respectively. Finally, type-D veliger larvae with a length of 88 ± 13.0 µm were obtained in the treatments at 13 • C and 16 • C. At 16 • C, the larval phase was achieved at 41 h 45 pf and, at 13 • C, at 43 h 30 pf ( Table 1). The survival at the 14th day of culture was higher at 13 • C (22%) than at 16 • C (21%) and 19 • C (9%). The embryonic development was characterized by the presence of a fertilization membrane and, later, the presence of the first polar corpuscle, giving rise to successive segmentations until it reached the morula stage. Subsequently, there was a blastula during the first invagination, a rotating blastula with cilia around it, a gastrula in the second invagination, trochophore larva with the presence of a flagellum and great mobility until the formation of the early veliger larva stage with the presence of ciliated veil and, finally, veliger larva, type D, with a straight hinge and defined stomach ( Figure 3).
Temperature treatments showed differences in the duration of stages during larval development. Larval development from D larvae to umbonate larvae lasted 16 days at 13 • C but lasted 10 days at 16 • C and 19 • C (140 µm). In general, during day 6 and 8, no differences in the size or morphology of the larvae were observed. Later, on day 12 of culture, the umbonate larvae (180 µm) began to be observed in the treatment at 16 • C; on day 16 of culture, larvae with semi-curved hinges corresponding to advanced umbonate larvae with a well-defined stomach and a length of 235 µm ( Figure 4) were observed. The larval growth rate was 2 µm day −1 at 13 • C, 2.2 µm day −1 at 16 • C and 3.6 µm day −1 at 19 • C.
segmentations until it reached the morula stage. Subsequently, there was a blastula during the first invagination, a rotating blastula with cilia around it, a gastrula in the second invagination, trochophore larva with the presence of a flagellum and great mobility until the formation of the early veliger larva stage with the presence of ciliated veil and, finally, veliger larva, type D, with a straight hinge and defined stomach ( Figure 3). Temperature treatments showed differences in the duration of stages during larval development. Larval development from D larvae to umbonate larvae lasted 16 days at 13 °C but lasted 10 days at 16 °C and 19 °C (140 μm). In general, during day 6 and 8, no differences in the size or morphology of the larvae were observed. Later, on day 12 of

Discussion
The "stripping" method [17] for obtaining gametes in Mesodesma donacium has been described as the fastest and most appropriate when mass fertilization is required [16]. The use of this technique has been applied by several researchers for this same species [11,12,29,30]. Of the gametes extracted by the "stripping" technique in this work, an aver-

Discussion
The "stripping" method [17] for obtaining gametes in Mesodesma donacium has been described as the fastest and most appropriate when mass fertilization is required [16]. The use of this technique has been applied by several researchers for this same species [11,12,29,30]. Of the gametes extracted by the "stripping" technique in this work, an average of 322,528 ± 202,987 oocytes and 1,583,333 ± 418,911 sperm cells was obtained. This number of oocytes was lower than that obtained by Illanes [31] using the same technique (3,330,401 ± 4,182,911 oocytes); the broodstock used by Illanes was collected in Caleta San Pedro (La Serena, Chile), with a total length and weight of 77.0 ± 4.7 mm and , respectively, unlike those used in this experiment, which measured 63.7 ± 2.8 mm and 25.0 ± 3.0 g. This could be explained by the differences in M. donacium population along the Chilean coast [32].
Regarding fertilization, in our experience, the ratio of ≥ 100:1 sperm:oocytes showed over 75% success in fecundation, and, in the 1000:1 sperm:oocytes treatment, oocytes in lysis were registered. Similarly, Le et al. [21] established an optimal ratio < 500:1 sperm:oocytes for Panopea zelandica, finding no significant differences in the percentage of embryos at 18 h pf between the ratio of 50:1 and 500:1; however, the ratio 50:1 showed a higher percentage (96%) of normal embryos. In our study, the result was the opposite, with greater fertilization obtained at higher sperm:oocyte ratios.
The triggers that modulate the reproductive cycle are both exogenous, such as temperature and food availability, and endogenous (genetic, hormonal, energy reserves) [35]. The diversity of environments and life habits of bivalves has promoted the development of different reproductive tactics, such as short or extended reproductive periods, adopting "r" or "k" reproductive strategies and producing a large or small number of offspring [36]. The satisfactory development of embryos under artificial conditions (hatchery) is essential for reliable larval production. The success of fertilization depends primarily on a good gametic ratio of sperm:oocytes, and this ratio depends on the endogenous and exogenous factors that regulate different processes in the reproductive cycle. Low sperm concentrations reduce the likelihood of oocyte encounters, and high sperm concentrations can increase the risk of polyspermy, which can lead to abnormal embryo development, affecting larval survival [19][20][21]23]. This optimal ratio varies between species of bivalves. In the clam Clinocardium nuttallii, an optimal sperm:oocyte ratio was found at a high value of 10,000:1 according to the results of Liu et al. [22]. In the Spisula solidissima clam, the best result was recorded in a range between 50:0 and 100:1 with 95% success in fertilization [23]. On the other hand, Dong et al. [19] observed an increase in the rate of polyspermy with a higher sperm:oocyte ratio and negative effects on larval survival. A concentration of < 200:1 was the most appropriate for the bivalve Tegillarca granosa. In Chilean clams with commercial value, the variation was very high between a ratio of 10:1 for Gari solida [37], a ratio of 100:1 for Ensis macha [38] and, finally, a ratio of 500:1 for Tawera elliptica [39].
It is difficult to make comparisons between different clams mainly due to the initial conditions of the broodstock at the beginning of the experiments and, thus, in the quality of the oocytes [40]. Embryogenesis and early larval development depend on the endogenous energy reserves supplied to the oocytes, and any energy deficiency can have serious consequences on subsequent success [41]. Cerviño-Otero [42] investigated the combined effect of seasonality and the geographical origin of broodstock on the number of eggs released and the larval yield of Venerupis corrugata, reporting significant differences in the biochemical composition of the gonads during the different seasons in all the studied populations.
It is important to note that the success of fertilization in bivalves is largely based on the sperm:oocyte ratio; however, there are other factors that can affect the production of D larvae. The density of the oocytes, the quality of the spermatozoa measured as motility, the contact time and the age of the gametes are also factors to be considered [20,43]. On the other hand, environmental factors, such as temperature, pH and salinity, also play an important role in the feasibility of restocking populations with a high distribution range [2,12,16,18,23]. In this study, the emission of gametes was through the "stripping" method, and comparisons with other species of clams where spontaneous spawning methods [22], application of desiccation stress and temperature [19,39] or injection of neurotransmitters [20,21,38] were used were difficult. Clotteau and Dubé [23] used the stripping method to obtain the gametes and required a high density of ml −1 oocytes (>10,000) and also a sperm:oocyte ratio > 100:1, which could be indicative that, depending on the type of spawning induction used [20,21], the process may require a greater or lesser ratio of sperm per oocyte and also different densities of oocytes.
Embryonic and larval development observed in M. donacium followed the patterns described for bivalve molluscs [7,17,37,39,[44][45][46]. The effect of temperature on the first divisions, the stages of morula, blastula, gastrula and trochophore larva, was observed, with shorter developing times in higher-temperature treatments. The duration of the stages of development was similar to the results of Tarifeño [33] and Zaro [12]. They observed the trochophore larval stage at 24 h pf and veliger larva, type D, with a typical planktotrophic development of between 110 and about 250 µm, respectively, values similar to those obtained in this work (90 µm larva veliger, type D, and 235 µm umbonated 16-day-old larvae cultured at 16 • C). The diameter increased after the appearance of the fertilization membrane from 68 ± 4.5 µm to 88 ± 13.0 µm, and, after the formation of veliger larva, type D, the umbonate larva presented at 12 days pf in the treatment at 16 • C with a diameter of 140 µm, similar to that described for the family Mesodesmatidae (Table 2). nd = no data.
The larval growth observed in M. donacium suggests a direct increase in relation to temperature; the growth rate of the larvae (maximum valve length) was 2 µm day −1 at 13 • C, 2.2 µm day −1 at 16 • C and 3.6 µm day −1 at 19 • C. This value was lower than that obtained by Lépez [48] at 17 • C (10.52 µm day −1 ) but similar to that obtained by Guisado [52] with a valve growth of 3.3 µm day −1 in the same species. As mentioned, differences in larval culture results can be explained by the genetic origin, the sexual condition of the broodstock and the time of year, as well as the temperature and food availability. In this study, the broodstock was collected in austral winter (June 2018) when the breeders were in maturation stage. Reverol et al. [24] observed in Paphies ventricosa (Mesodesmatidae) that larval shell growth during the first 10 days was slow, with an increase in mean total length of 5.5 µm per day. On the other hand, Gadomski et al. [25] identified in P. ventricosa that the morphology of the larva does not depend on temperature; they only indicated a slowdown of larval development rather than physiological damage and abnormal development resulting from the increase in temperature, in the same way as observed in Tivela mactroides, where the increase in temperature promoted the accelerated development of the larvae with a greater growth rate. However, they pointed out that an increase in temperature could increase the possibility of occurrence of bacterial proliferations [24], which could be an explanation for the decrease in survival in higher temperature treatments in the present work.
Temperature accelerates metabolic processes, food activity and also growth in bivalve molluscs [17,53,54]. Temperature accelerates embryonic development in molluscs. Clotteau and Dubé [23] suggested that an adequate temperature range for early embryos in the clam S. solidissima is achieved between 15 • C and 20 • C. Lee and Rho [55] established that the time required to reach the stages of blastula, trochophora and D larva decreased to a third by increasing temperature from 8 • C to 17 • C in the Panope japonica clam. However, no survival data were presented. On the other hand, a lower temperature during the storage of the gametes before fertilization helped to extend the period in which the sperm remained motile, generating better management of the gametes in the hatchery [22,43].
In M. donacium, temperature generally decreases the duration of stages in embryonic development, increases growth and inversely influences survival. Zaro [12] determined that, above 21 • C, the larvae stage did not exceed 5 days and that, below 19 • C, the pediveliger larva stage was reached at 22 days with a 1.9% survival rate. Reverol et al. [24] observed that in Tivela mactroides the temperature (22 • C, 25 • C, 28 • C) was inversely proportional to the duration of the stages of larval development and directly proportional to survival rate. Albentosa et al. [53] mentioned that the general model of the effect of temperature can vary widely not only between species but within the same species. However, Reverol et al. [24] indicated that larvae, when approaching the juvenile stage, are more sensitive to temperature changes. In our experiment, M. donacium presented a high thermal sensitivity at the end of embryonic development at high temperatures (19 • C), with larvae after the stage of rotary blastula not being observed in that treatment: an important result highlighting the current climatic and environmental contingency.
M. donacium is a stenothermal species with a low tolerance to high temperatures related to its biogeographic origin and associated with the minimum oxygen zone and a high productivity associated with low temperatures in the upwelling ecosystem of the Humboldt Current [13,56]. Environmental phenomena, such as El Niño, which modifies large-scale environmental patterns in the Humboldt Current System, generate massive mortalities of species, which could affect the stability of the surf clam in the environment, especially in the meroplanktonic larval stages [57]. M. donacium also presented a high thermal sensitivity in the larval stages, mainly limiting growth.
It should be noted that the action of bacterial contamination or accumulation of materials and microorganisms (fibers, feces and protozoa) [24,58] are factors that affect larval survival and are enhanced by the increase in temperature in treatments. Reverol et al. [58] suggested that larval stages are especially sensitive to environmental changes, which could lead to larval mortality in the final phase of the cycle. Temperature influences survival and larval growth in bivalves; this factor is related to the ability to capture food (filterers), which modifies the filtration rate by factors such as temperature, the speed of water movement and the concentration of particles [24,59].

Conclusions
This study showed that the "stripping" method is effective for obtaining viable gametes in the surf clam. For the success of fertilization, it is recommended to use a ratio between ≥100:1 and ≤1000:1 sperm:oocytes. A temperature of 16 • C is suggested for the incubation of oocytes and for optimal embryonic and larval development for broodstock from the Cucao bank (Chiloé). The information obtained is fundamental for the development of a replicable culture technology that allows the scaling of larval crops of this species and optimizes the early stages of cultivation, enhancing it as an alternative for the sustainable development of raw materials with the support of small-scale aquaculture and contributing to food security and family income especially in rural areas. The main challenges to be addressed in the surf clam are the basic variables of larval culture such as larval density and feeding. It is also essential to study the phenomenon of metamorphosis in sandy substrate conditions and its subsequent cultivation in artificial and/or natural environments to finalize the cultivation technology of the species.