Predation environment predicts divergent life-history phenotypes among populations of the livebearing fish Brachyrhaphis rhabdophora

Abstract

We document a strong association between predation environment and life-history phenotypes in the Costa Rican livebearing fish Brachyrhaphis rhabdophora. Populations that co-occurred with piscine predators attained maturity at a smaller size, and produced more, smaller offspring relative to populations from predator-free environments. These differences persisted over 3 years and between wet and dry seasons within a year. Reproductive allotment did not differ between predation environments, but was greater in the wet season than in the dry season. We also examined the phenotypic covariance structure among life-history traits and found traits to be highly correlated. Based on life-history differences, discriminant analyses showed that populations could be neatly classified by predation category, and could be reasonably classified into wet and dry season categories. Finally, we found that the pattern of predator-associated life-history divergence in B. rhabdophora is remarkably similar to that of the taxonomically distinct Trinidadian guppy (Poecilia reticulata), possibly pointing to an evolutionary convergence between these two systems.

Life-history phenotypes in populations of Brachyrhaphis episcopi (Poeciliidae) with different predator communities

“…Here we test whether life-history phenotypes of another poeciliid, Brachyrhaphis episcopi, differed between populations that did or did not cooccur with larger predatory fish. Our findings are broadly in agreement with those from guppies and recent work on B. rhabdophora (Johnson and Belk 2001), which increases confidence that direct or indirect predator effects generate population variation in poeciliid life histories.…”

“…In poeciliid fish, similar patterns are seen for "predator" and "predator-free" sites in Brachyrhaphis rhabdophora (Johnson and Belk 2001) and "high-risk" and "low-risk" sites in Poecilia reticulata (Reznick 1989). Here we report significant differences in life-history traits between Brachyrhaphis episcopi populations co-occurring with predatory fish (Characin sites) and those without (Rivulus sites).…”

“…Although male poeciliids generally cease growth upon maturation Johnson and Belk 2001), male B. episcopi within a single population can range from 15 to 36 mm in length (Turner 1938). Adult males are smaller than females (range 14.3-30.6 mm versus 19.2-51.7 mm standard length; this study).…”

“…In contrast, no such trend has been reported for guppies or B. rhabdophora (Johnson and Belk 2001), although closer inspection may show that it exists. We compared sites using the across-site average for female size (163 mg dry weight) and values one standard deviation around the average (62 and 264 mg).…”

“…Guppies are, however, unusual because they show considerable population divergence for almost every trait examined (Magurran 1999). Repeatable life-history shifts may be less prevalent in other freshwater fish (but see Bertschy and Fox 1999;Johnson and Belk 2001). Here we test whether life-history phenotypes of another poeciliid, Brachyrhaphis episcopi, differed between populations that did or did not cooccur with larger predatory fish.…”