Breath-hold diving presents air-breathing vertebrates with the challenge of maintaining aerobic respiration while exercising underwater. Adaptive increases in the oxygen (O₂) storage capacity in the lungs, blood, or muscle tissues can enhance these reserves and greatly extend aerobic foraging time underwater. Here, we report blood- and muscle-O₂ storage parameters (blood hemoglobin concentration ([Hb]), hematocrit, and myoglobin concentration ([Mb]) in the pectoralis and gastrocnemius) for 16 species of diving and dabbling ducks found in North America, and investigate which parameters are correlated with the diving behaviors reported in both the sea ducks (Mergini) and the pochards (Aythini). Both [Hb] in the blood and [Mb] in the gastrocnemius, a major leg muscle used in propulsion for these predominantly leg-propelled divers, were significantly higher in the sea ducks compared to the dabblers (Anatini). The pochards also showed a significant increase in [Hb] and were intermediate between the sea ducks and the dabblers in hematocrit and [Mb] in the gastrocnemius. Among these four variables and total body mass, [Mb] in the gastrocnemius was the most significant predictor of mean species dive time, and these two variables were correlated across the phylogeny. Our results indicate that the observed changes in O₂ storage capacity in the blood and muscles are positively correlated with diving behavior in two clades of ducks, such that larger increases are correlated with longer dive times.

Hemoglobin and Hematocrit

Blood was obtained via cardiac puncture from freshly collected ducks.Immediately after extraction, [Hb] of the whole blood was analyzed using a HemoCue 201+ Analyzer (HemoCue America, Brea, CA, USA), which chemically converts Hb to azidemethemoglobin before measuring absorbance at 570 and 880 nm. All values were corrected using an avian correction factor of –1 g/dL (Simmons and Lill 2006). Packed red blood cell volume (Hct) was measured in quadruplicate using 75 mm heparinized capillary tubes. Tubes were spun for five minutes in a ZIPocrit centrifuge (LW Scientific, Lawrenceville, GA, USA), after which Hct was measured. All hematocrit measurements represent the mean of all four tubes, except in cases where one tube broke or was clearly faulty (e.g. 28% when three other tubes were 45%). Mean corpuscular hemoglobin concentration (MCHC), which shows the average [Hb] in red blood cells, was calculated using the following equation: MCHC = (Hb[g/dL] x 100)/Hct[%]). 6–27 individuals were used for hemoglobin and hematocrit measurements. 

Myoglobin

Frozen tissue samples from the pectoralis and gastrocnemius of 16 species of ducks were used for myoglobin assays. Myoglobin concentration was determined using the modified Reynafarje (Reynafarje 1963) method utilized by Dawson et al. (2016, 2020). Frozen tissue was homogenized in 19.25 volumes (1 mL buffer per 1 gram of frozen tissue) of ice-cold homogenization buffer (40 mM potassium phosphate; pH 6.6) and centrifuged for 99 minutes at 13,700x g and 4 °C (Dawson et al. 2016). The resulting supernatant was transferred to a 25 mL rotating boiling flask and exposed to pure carbon monoxide (CO) for 8 minutes. Sodium dithionite was added and CO was bubbled in for a further 2 minutes to ensure complete reduction of Mb. Samples were diluted a further 19.5x with homogenization buffer, transferred to a 1 ml cuvette (10mm optical path length). The optical density of each sample was measured in triplicate with a homogenization buffer blank at 538 and 568 nm using a VWR V1200 Spectrophotometer (VWR, Radnor, PA, USA). Myoglobin concentration ([Mb]; mg/g) was then determined based on the following equation: [Mb] = (OD538 – OD568) x 112.6 (Dawson et al. 2016). Only 10 individuals per species were included in myoglobin assays.