Influence of euthanasia method on blood and gill variables in normoxic and hypoxic Gulf killifish Fundulus grandis

In many experiments, euthanasia, or humane killing, of animals is necessary. Some methods of euthanasia cause death through cessation of respiratory or cardiovascular systems, causing oxygen levels of blood and tissues to drop. For experiments where the goal is to measure the effects of environmental low oxygen (hypoxia), the choice of euthanasia technique, therefore, may confound the results. This study examined the effects of four euthanasia methods commonly used in fish biology (overdose of MS‐222, overdose of clove oil, rapid cooling and blunt trauma to the head) on variables known to be altered during hypoxia (haematocrit, plasma cortisol, blood lactate and blood glucose) or reflecting gill damage (trypan blue exclusion) and energetic status (ATP, ADP and ATP:ADP) in Gulf killifish Fundulus grandis after 24 h exposure to well‐aerated conditions (normoxia, 7·93 mg O₂ l^?1, c. 150 mm Hg or c. 20 kPa) or reduced oxygen levels (0·86 mg O₂ l^?1, c. 17 mm Hg or c. 2·2 kPa). Regardless of oxygen treatment, fish euthanized by an overdose of MS‐222 had higher haematocrit and lower gill ATP:ADP than fish euthanized by other methods. The effects of 24 h hypoxic exposure on these and other variables, however, were equivalent among methods of euthanasia (i.e. there were no significant interactions between euthanasia method and oxygen treatment). The choice of an appropriate euthanasia method, therefore, will depend upon the magnitude of the treatment effects (e.g. hypoxia) relative to potential artefacts caused by euthanasia on the variables of interest.