For more information on the dataset acquistion and processing methods, see Malvezzi et al. (2015) (doi: 10.1111/eva.12248). A summary follows:
Species origin:
We collected ripe, adult M. menidia at the beginning of the spawning season (25 April 2013) from a tidal salt marsh on the North shore of Long Island (Poquot, 40°58.12'N, 73°5.28'W). Males and females were caught with a 30 × 2 m beach seine, then transported to our laboratory facility (Flax Pond Marine Laboratory) and held overnight in separate temperature-controlled baths (200 L, 21 degrees C). The next morning, 29 females were strip-spawned by gently squeezing their hydrated eggs into a large shallow container containing seawater-activated sperm of 42 males and a large sheet of window screen (1 mm mesh).
Estimation of 'Days survived at high CO2' (DS):
To quantify DS for each hatched larva, the bottom of every high CO2 rearing container was gently siphoned twice daily, and all dead individuals were removed, recorded, and individually transferred to a tissue lysis solution (100 uL tissue lysis buffer + 12 uL proteinase K) for subsequent genomic DNA extraction. Frequent siphoning was critical, because fish larvae decompose beyond recognition within hours after death (at 24 degrees C). The pattern of daily posthatch mortality was typical for early larval fish, with mortality peaking soon after hatch (1–4 dph), but then declining exponentially over the remaining days of the experiment. After two consecutive days without any mortality (days 14 and 15 posthatch), the experiment was terminated and all surviving larvae (i.e., DS = 15) were individually transferred to tissue lysis solution for DNA extraction. To specifically evaluate the resistance of the population to elevated CO2, only hatched larvae of the 10 high CO2 replicates were sampled for genetic analyses.
DNA extraction and amplification:
Adult spawner DNA was extracted from tail clips (15–35 mg) following the animal tissue protocol of Qiagen DNeasy kits. Larval DNA (DS ranging from 1 to 15 dph) was extracted using a more cost-effective ‘salting out’ protocol that yielded useable DNA from larval silversides. Ten polymorphic microsatellite loci for M. menidia were amplified for all parents and offspring in a 10 uL reaction containing 1× PCR buffer, 10× bovine serum albumin, 1.5–3.5 mm MgCl2, 0.12 mm dNTPs, 0.16 um of the reverse primer and fluorescently labeled M13 primer, 0.04 um of the species-specific forward primer and 1 unit Taq polymerase, and 10–40 ng of genomic DNA. Thermal cycling consisted of 5 min at 94 degrees C followed by 35 cycles of 94 degrees C for 30 s, primer specific annealing temperature (Ta) for 30 s and 72 degrees C for 60 s with a final extension at 72 degrees C for 10 min. Fluorescently labeled PCR products were electrophoresed on an ABI 3730 DNA analyzer along with an internal fluorescent ladder (LIZ-500; Applied Biosystems). Alleles were scored by a single analyst (AJM) using the software Peakscanner 1.0 (Applied Biosystems, Life Technologies, Grand Island, NY, USA). A subset of approximately 10% of genotypes was verified by a second analyst (KAF) using GENEMAPPER v4.0 (Applied Biosystems). One locus (Mm09) did not amplify and another one did not yield easily scored peaks (Mm119), hence only 8 of 10 loci were eventually used for parentage assignment and genetic analyses.