Summary
Reproduction is an energetically demanding life history stage that requires costly physiological and behavioral changes, yet some individuals will invest more into reproduction and breed more successfully than others. To understand variation in reproductive investment, previous studies have evaluated factors during breeding, but conditions outside of this life history stage may also play a role. Using a free-ranging population of black-capped chickadees (Poecile atricapillus), we assessed the repeatability of plastic traits relating to energetic condition (circulating initial corticosterone concentrations and body condition) during the nonbreeding season and evaluated whether these traits predicted reproductive investment in the subsequent breeding season. We found that initial corticosterone concentrations and an index of body condition, but not fat score, were moderately repeatable over a 1-week period in winter. This trait repeatability supports the interpretation that among-individual variation in these phenotypic traits could reflect an intrinsic strategy to cope with challenging conditions across life history stages. We found that females with larger fat reserves during winter laid eggs sooner and tended to spend more time incubating their eggs and feeding their offspring. In contrast, we found that females with higher residual body mass delayed breeding, after controlling for the relationship between fat score and timing of breeding. Additionally, females with higher initial corticosterone in winter laid lighter eggs. Our findings suggest that conditions experienced outside of the breeding season may be important factors explaining variation in reproductive investment.
Methodology
Field methods
We searched forested properties surrounding the Queen’s University Biological Station (near Chaffey’s Lock, Ontario; 44°34′N, 76°19′W) for flocks of overwintering chickadees in December 2015. At 14 sites, we trapped and collected blood from a total of 137 chickadees (48 females, 82 males, 7 unknown sex) between 24 January 2016 and 27 February 2016, using seed-baited walk-in “Potter” traps set on platforms 1.2 meters above the ground. We trapped a second time at the same locations at least 6 days after the initial capture period (mean ± SE = 7.8 ± 0.15 days, range = 6–15 days) and repeated the same procedure as above. Of the 137 individuals, 53 were captured and blood sampled twice.
All trapping occurred after sunrise (between the daylight hours of 07:30 AM and 01:00 PM) to minimize the influence of diel rhythms in CORT secretion (Breuner et al. 1999). We punctured the brachial vein using a 26G needle and collected up to 120 µL of blood into heparinized microcapillary tubes within approximately 3 min of capture (mean ± SE = 2.44 ± 0.04 min, range = 1.05– 4.17 min). We fitted all birds with a uniquely numbered Canadian Wildlife Service aluminum band, a single colored leg band, and a passive integrated transponder (PIT) integrated into a colored plastic leg band (Eccel Technology Ltd, Item No. EM4102 [2.6 mm], Glenfield, Leicester, United Kingdom). The PIT band allowed for behavioral data collection using radio-frequency identification devices (RFID). We also took morphological measurements including wing chord length using a wing ruler (±1.0 mm), body mass using a Pesola spring scale (±0.25 g), and a score for visible fat deposits stored in the inter-clavicular depression (furculum) on a scale from 0 to 5, following Krementz and Pendleton (1990). Studies have shown that this visible fat scoring method is predictive of total body fat content in birds (Seewagen 2008; Labocha and Hayes 2012; McWilliams and Whitman 2013). All body measurements were taken by K. Schoenemann or C. Montreuil-Spencer. We centrifuged blood samples for 9 min within 7 h of capture, separated plasma from red blood cells using a Hamilton syringe, and stored them separately in microcentrifuge tubes at −20°C.