Many populations have intraspecific diversity in phenotype and ecological strategy, but the mechanisms maintaining such diversity are not fully understood. Multiple behaviors can be maintained either as a conditional strategy, where fitness depends on an individual’s phenotype, or as a mixed strategy, where alternative behaviors have similar fitness independent of phenotype. Using high-resolution depth and time sampling, we characterize 2 distinct diel vertical migration behaviors in a population of freshwater zooplankton (Daphnia pulicaria). Individuals in this population differ in their color phenotype and migratory behavior with red morphs upregulating hemoglobin and undergoing a deep migration and pale morphs not producing hemoglobin and undergoing a shallow migration. We experimentally manipulated the behavior of each phenotype in the field and measured population growth in their natural migration behavior as well as population growth in their alternative behaviors. Experimental populations of pale and red morphs under their natural migrations had roughly equal fitness, despite vast differences in environmental conditions. When forced to switch behaviors, pale morphs suffered reduced fitness, whereas red morphs had similar fitness compared with their natural migration. Our results suggest that although behavioral diversity may be promoted by the opportunity for alternative behaviors of equal fitness, the distinct physiological conditions required for survival in alternative behaviors limit the capacity for individual behavioral switching and likely maintain behavioral diversity as a conditional strategy.
Sampling and experiments were done in Round Lake, which is a relatively deep meromictic lake with strong environmental gradients in temperature and oxygen (Supplementary Appendix S1). To characterize migration behavior of D. pulicaria, the density of individuals by depth was estimated every 3 h over a 24-h sampling period by taking duplicate live zooplankton samples from a depth of 1–26 m at 1-m intervals using Schindler traps (35 and 19.5 L traps). Each depth-density profile took 90 min to complete and replicate samples were taken simultaneously at 9:00, 12:00, 15:00, 18:00, 21:00, 0:00, 3:00, and 6:00. Samples were stored live in cold, filtered (80 µm) lake water pumped from 10 m, held in a chilled cooler, and transferred to a refrigerator on return to the field station. Sample enumeration was done using size-calibrated photographs because other preservation methods changed the redness of individuals. Photos were taken within 36 h of sampling using a Samsung WB350F camera (sensor resolution 16.3 Megapixels) in a purpose-built lightbox to ensure consistent lighting and photo quality. We estimate that samples retained 90% of their Hb in this time, based on rates of Hb loss in Daphnia magna (Fox and Phear 1953) and synthesis in Daphnia pulex (Kring and O’Brien 1976; Silverman 1981). Daphnia were counted and classified as either adult or juvenile using size-class circles. Daphnia longer than 1.4 mm (tail spine base to top of head) were considered adults (Gliwicz and Boavida 1996). Samples taken at 3:00 were sized with greater resolution (S1: <1.0 mm; S2: 1.0–1.4 mm; S3: 1.4–1.85 mm; S4: 1.85–2.3 mm; S5: >2.3 mm) to examine the correlation between size and migration behavior.