70 
Fishery Bulletin 119(1) 
Linear growth during the larval stage (Powell et al., 2002; 
Pace et al., 2006) was estimated with this equation: 
y=mx+b, 
where y = length (in micrometers); 
m = slope (in micrometers per day); 
x = time (in days); and 
b = intercept (initial length in micrometers). 
For each bucket, mean rates of shell growth (in microme- 
ters per day, with standard errors of the mean [SE]) were 
determined by the slope of shell length versus age. An 
analysis of covariance was used to compare the regression 
lines and to discern if they were statistically significant 
(P<0.05). If a significant difference was detected between 
slopes, further examination was done by using analysis of 
variance. Data for lipid concentration (in micrograms per 
gram of wet weight), for percent lipid (lipid concentration 
divided by total lipid concentration multiplied by100) and 
for percentages of all larvae in the sample that survived 
and that had successfully completed metamorphosis were 
collected at the end of the experiment. All data collected as 
percentages were transformed with arcsine (square root) 
prior to statistical analysis and prior to analysis of vari- 
ance. Least square difference was used to determine dif- 
ferent homogeneous groups. For all statistical analyses, 
degrees of freedom equaled 6. 
Results 
Seawater physicochemical variables 
Carbonate chemistry and environmental 
variables in the 3 pCO, treatments 
are shown in Table 1. Values for mea- 
sured pH were 7.972 (SE 0.003), 7.629 
(SE 0.003), and 7.462 (SE 0.003) for the 
treatments with low, medium, and high 
levels of pCOg, respectively. Values of DIC 
were 1800.66 pmol/kg (SE 2.35) for the 
low-pCO, treatment, 1903.74 pmol/kg 
(SE 2.39) for the medium-pCO, treat- 
ment, and 1953.32 pmol/kg (SE 2.37) 
for the high-pCO, treatment. Other car- 
bonate system characteristics (pCObs, 
Qealciter Qaragonite. 2nd pH in situ) were 
calculated at the in situ temperature and 
are reported in Table 1. The calculated 
levels of pCO, were 343.9 patm (SE 5.9), 
820.8 patm (SE 6.0), and 1242.9 patm 
(SE 5.9) for the 3 treatments with different 
experimental conditions. There was a sig- 
nificant difference in pCO, (F ratio=2945, 
P<0.01) between the low-, medium-, and 
high-pCO, treatments. Mean seawater 
temperature (18.9°C [SE 0.1]) and salin- 
ity (25.73 [SE 0.08]) remained stable 
throughout the study period (Table 1). 
= 
Ss 
2 
D> 
o 
fe 
@ 
f= 
ep) 
Larval growth, survival, metamorphosis, and lipids 
Significant differences were observed in the final shell 
height of Atlantic surfclam from fertilization to metamor- 
phosis at the conclusion of the experiment, with greater 
shell height (in micrometers) for individuals in the medi- 
um-pCO, treatment than for those in the treatments 
with low and high pCO, levels (F ratio=6.986, P=0.03; 
Fig. 1). The specimens in the medium-pCO, treatment 
had a final shell height that measured 13% larger than 
those in the the high- and low-pCO, treatments, but 
there was no difference in final shell height between the 
individuals exposed to low and high pCO, levels. Larvae 
in the treatment with a medium pCO, level had a mean 
growth rate of 7.16 pm/d (SE 0.25) that was signifi- 
cantly greater (F ratio=5.84, P=0.04), 15% higher, than 
the mean rate observed for larvae in the low-pCO, treat- 
ment, 8.26 pm/d (SE 0.25). Growth rates of larvae under 
the high-pCO, treatment, with a mean rate of 7.25 ym/d 
(SE 0.25), were similar to those of larvae in the low-pCO, 
treatment. 
Although larval survival did not differ significantly 
(F ratio=3.23, P=0.17) between treatments, survival rates 
in the medium-pCO, treatment (29% [SE 5]) were twice 
those observed in the treatments with high (17% [SE 5]) 
and low (11% [SE 5]) pCO, levels. After 28 d, significantly 
more larvae (F ratio=12.89, P<0.01) in the medium-pCO, 
treatment successfully completed metamorphosis (88% 
[SE 1]) than in the low- and high-pCO, treatments, at 
65% (SE 6) and 61% (SE 5), respectively. There was no 
—S— Low 
see¥%-* Medium 
-£#- High 
Metamorphosis 
15 
Time (d) 
Figure 1 
Mean shell height of laboratory-reared larval Atlantic surfclam (Spisula 
solidissima) over the 28 d of the experiment conducted in 2015 with 3 treat- 
ments that exposed larvae to low (circles, 344 patm), medium (triangles, 
821 atm), and high (squares, 1243 patm) levels of partial pressure of car- 
bon dioxide (pCO,). Differences in the letters at the end of regression lines 
indicate significant differences in final shell height between treatments 
(P=0.03). The adult brood stock used to rear individuals for this study were 
collected in 2014 from Long Island Sound near Stratford and Norwalk, 
Connecticut. 
