Vertebrate growth plasticity in response to variation in a mutualistic interaction

Abstract

Vertebrate growth can be phenotypically plastic in response to predator-prey and competitive interactions. It is unknown however, if it can be plastic in response to mutualistic interactions. Here we investigate plasticity of vertebrate growth in response to variation in mutualistic interactions, using clown anemonefish and their anemone hosts. In the wild, there is a positive correlation between the size of the fish and the size of the anemone, but the cause of this correlation is unknown. Plausible hypotheses are that fish exhibit growth plasticity in response to variation in food or space provided by the host. In the lab, we pair individuals with real anemones of various sizes and show that fish on larger anemones grow faster than fish on smaller anemones. By feeding the fish a constant food ration, we exclude variation in food availability as a cause. By pairing juveniles with artificial anemones of various sizes, we exclude variation in space availability as a single cause. We argue that variation in space availability in conjunction with host cues cause the variability in fish growth. By adjusting their growth, anemonefish likely maximize their reproductive value given their anemone context. More generally, we demonstrate vertebrate growth plasticity in response to variation in mutualistic interactions.

Figure 1

Relationship between the standard length of rank 1 A. percula (mm) and the size (ln cm²) of their H. magnifica host for 202 groups in Kimbe Bay, Papua New Guinea.

Figure 2

Results for two Bayesian linear mixed models testing the effect of initial standard length (SL) and Ln anemone area on growth of Amphiprion percula: (a) Posterior distributions with medians and 95% intervals for experiment 1, where fish were paired with Entacmaea quadricolor; (b) Posterior distributions with medians and 95% intervals for experiment 2, where fish were paired with artificial anemones; (c) Predicted change in standard length (mm) dependent on Ln anemone area (cm²) for A. percula on E. quadricolor; (d) Predicted change in standard length (mm) dependent on Ln anemone area (cm²) for A. percula on artificial anemones.

Figure 3

(a) Predicted sizes of rank 1 Amphiprion percula over 12 months in anemones of varying sizes (ln 5–ln 10 cm²). Using the mean probability estimates of the Bayesian mixed model analyses (Fig. 2a), change in SL per month was calculated as a function of initial SL and ln anemone area: ${\mathrm{\Delta }SL}_{t0-1}={initialSL}_{t0}+6.42+\left(0.81,,\ast ,\ln ,\left(a,n,e,m,o,n,e,,a,r,e,a\right)\right)-\left(0.25,,\ast ,{initialSL}_{t0}\right)$. (b) Distribution of rank 1 A. percula SL (mm) and the size of the anemone they reside in (binned, Ln cm²) from a population in Kimbe Bay, Papua New Guinea.