Summary
Stable isotope analysis is frequently used to examine resource use in wild populations, but it often involves invasive or lethal methods of collecting tissue samples. The development of less invasive or nonlethal sampling techniques will expand the possible uses of stable isotopes. We examined whether fish eggs meet three basic requirements for inferring female resource use from them: (1) the isotope composition of the eggs is correlated with that of other maternal tissues for which isotope composition is known to be related to diet; (2) the isotope composition remains constant over the egg development period; and (3) dietary inferences using eggs are similar to those for other maternal tissues. Using artificial crosses, we tested the relationship between eggs and two commonly sampled maternal tissues (white muscle and liver) in wild‐caught Bluegills Lepomis macrochirus. We found that egg isotope composition was strongly correlated with that of other maternal tissues, particularly liver, and remained constant from prefertilization to the day of hatch, with no change in 13C and an increase of only 0.3‰ in 15N. Furthermore, the results of SIAR (Stable Isotope Analysis in R) mixing models indicated a large degree of overlap in diet estimates between eggs and the other maternal tissues. Overall, eggs can be reliably used to infer the prebreeding foraging ecology of female Bluegills throughout the egg development period.
Methodology
By means of daily snorkel surveys of the littoral habitat of Lake Opinicon, Ontario (44º34′N, 76º19′W), 40 Bluegills (20 males and 20 females) were collected using a dip net on the day of spawning. Collections were made from two separate colonies on June 7 and 8, 2011. Spawning pairs were identified when in a nest together and the female began “dipping” behaviors indicating spawning (Gross 1982). Within 1 h of capture, the males and females were transported to the Queen's University Biological Station on the shore of Lake Opinicon and housed in an on‐site aquarium facility for no longer than an additional 3 h.
Artificial crosses of males and females were completed using the in vitro fertilization technique described by Neff and Lister (2007). Each male and female collected was used once to make an artificial cross, for a total of 20 crosses. Briefly, eggs were obtained from each female by applying gentle pressure to the abdomen and collecting the eggs in a 500‐mL glass jar containing 50 mL of lake water drawn directly from Lake Opinicon. A subsample of the eggs from each female was also placed in a 1.5‐mL microcentrifuge tube and stored at −20°C for subsequent isotopic analysis. Sperm was then collected from each male by applying gentle pressure to the abdomen and gathering the released sperm in a 2‐mL syringe. It was then gently mixed with the eggs, following which each jar was filled with lake water. Air stones were attached to the top of each jar to maintain oxygen levels in the water. Each day, 50% of the water was replaced with freshly drawn lake water.
After egg collection, the females were euthanized with an overdose of clove oil and a sample of white muscle tissue was removed from below the posterior portion of the dorsal fin on the right side of the fish. The liver was also removed and both tissue samples were stored at −20°C. The males were then released at the site of their original collection. Subsamples of the eggs were taken at 24, 48, and 72 h postfertilization, representing the entire period of egg development (Gross 1982; Neff 2003). These eggs were collected using a plastic pipette to gently loosen them from the jar, followed by the retrieval of 50–75 eggs. The eggs were stored in 1.5‐mL microcentrifuge tubes, drained of excess water, and frozen at −20°C. The overall health of the eggs in each cross was monitored; if the majority of the eggs in a jar changed from their usual grey coloration to white and lost their attachment to the jar surface, they were considered unhealthy or dead and collections were stopped for that cross.