• O‘Connor, Constance M.
  • Gilmour, Kathleen M.
  • Van Der Kraak, Glen
  • Cooke, Steven J.


While social interactions influence vertebrate endocrine regulation, the dynamics of regulation in relation to specific behaviors have not been clearly elucidated. In the current study, we investigated whether androgens (testosterone) or glucocorticoids (cortisol) play a functional role in aggressive offspring defense behavior in wild smallmouth bass (Micropterus dolomieu), a teleost fish with sole paternal care. We measured circulating testosterone and cortisol concentrations in plasma samples taken from parental males following a simulated nest intrusion by a common nest predator, the bluegill sunfish (Lepomis macrochirus). To understand whether endocrine regulation changes across the parental care period, we looked both at males guarding fresh eggs and at males guarding hatched embryos. Plasma testosterone levels increased in males subjected to a simulated nest intrusion when compared to sham controls. Testosterone concentrations in males guarding embryos were lower than in males guarding fresh eggs, but circulating testosterone was positively correlated with the level of aggression towards the nest predator at both offspring development stages. However, there was no increase in cortisol levels following a simulated nest intrusion, and no relationship between cortisol and any measured parameter. These results suggest that androgens play an important role in promoting aggressive nest defense behavior in teleost fish.


On May 25 and 26, 2008 (water temperature 15–16°C), male smallmouth bass guarding nests with fresh eggs (0–1 day after laying and fertilization) were identified on Charleston Lake, a public lake that is part of the Gananoque River system in eastern Ontario (44°32′N, 75°59″W). All nests were individually marked with a numbered tile, and the brood size of each nest was classified according to egg score. The egg score is a standard and highly repeatable measure of the relative number of eggs within a nest, and ranges from 1 (low, <500) to 5 (high, >4,000; Philipp et al. 1997; Suski et al. 2003). For standardization, only males guarding nests with intermediate egg scores of 3–4 were included in this study.

A subset of parental fish was sampled immediately. Fish were first subjected to a simulated nest predator intrusion. Bluegill sunfish (Lepomis macrochirus) are common nest predators of smallmouth bass, and simulated nest intrusions of bluegill sunfish elicit aggressive nest defense behaviors (e.g., Hanson et al. 2009). A bluegill sunfish (150–200 mm) in a 3.8-L glass jar was placed in the nest by a snorkeler who then retreated to a distance of approximately 2 m for 3 min of behavioral observations. Aggressive behaviors were defined as by O’Connor et al. (2009). Briefly, a “yawn” is when the parent opens and closes its mouth in the direction of the glass jar, a “charge” is when the parent moves rapidly towards the jar but without making physical contact, and a “hit” refers to the smallmouth bass making physical contact with the jar. Snorkelers recorded the sum of all three aggressive behaviors for each individual during the 3 min period. After 3 min, the snorkeler removed the jar. After 25 min, a standard time to allow circulating cortisol concentrations to rise following a stressor (e.g., O’Connor et al. 2011) and also sufficient for androgens to increase following a simulated territorial intrusion in cichlid fish (e.g., Desjardins et al. 2005), fish were quickly captured by rod-and-reel angling and sampled for 1 mL of blood by caudal puncture using lithium-heparinized vacutainers (B.D. Vacutainer, Franklin Lakes, NJ) and 21 gauge, 38 mm needles. Only fish that could be captured and blood sampled within 5 min of the first angling attempt were included in the study. Fish were measured (total length, TL) and released. In total, blood samples were obtained from 24 males [TL = 399.1 ± 8.7 mm (mean ± SE)] guarding fresh eggs. Blood samples were also obtained from 6 males (TL = 413.8 ± 17.4 mm) selected for a “sham” treatment, in which an empty jar was placed in the nest. On 5 June 2008 (water temperature 17.5°C), smallmouth bass guarding newly hatched embryos (TL = 399.2 ± 9.8 mm, N = 19) were subjected to a simulated nest intrusion and sampled for blood as described above.

Blood samples were held in water–ice slurries for no more than 1 h, and then centrifuged at 10,000g for 5 min (Compact II Centrifuge, Clay Adams, NJ). Plasma samples were flash frozen in liquid nitrogen and stored at −80°C until analysis.