How to investigate a putative signal? Stick to the right method when assessing the response of a receiver.

Reply How to investigate a putative signal? Stick to the right method when assessing the response of a receiver The evolution of autumn leaf coloration moved into the focus of ecologists after the seminal publication by Hamilton & Brown (2001). In that article, Hamilton & Brown proposed the coevolutionary hypothesis, suggesting that red and yellow leaves in autumn evolved as warning signals to repel herbivor-ous insects, particularly aphids. Despite strong interest in the topic, experimental studies supporting this hypothesis are lacking. In a recent article, we describe the first experiment reporting that autumnal leaf colour does not influence host selection in aphids (Schaefer & Rolshausen 2007). Instead, we found a strong positive correlation between aphid numbers and fruit production in a cohort of same-aged mountain ash (Sorbus aucuparia) trees. In our experiment, we used artificial green or red leaf colours and caught aphids with glue. Dö ring & Hardie (2007) question the use of glue, because aphids will not necessarily remain where they first land. Instead, aphids will gather additional cues before settling on a plant or rejecting it. According to Dö ring & Hardie (2007), post-landing decisions might counterbalance host choices made during flight, which remains unaccounted for in our experiment. Obviously, post-landing decisions are important , but this comment is confused about the stage at which aphids might react to a warning signal. When searching for suitable hosts, aphids face a hierarchy of decisions: (i) on which individual to land (decision among plant individuals), (ii) where to sample a plant (decision within plant individuals), and (iii) whether or not to vacate the plant after sampling. Importantly, the coevolutionary hypothesis makes predictions about aphid behaviour only at the first level, the decision among plant individuals. This is because the general adaptive significance of warning signals (here leaf colour) is that they ensure mutual avoidance of well-defended prey (individual trees) and predators (aphids). Thus, both the partners would benefit from a warning signal, because it enables aphids to select suitable hosts among other plants more efficiently, and well-defended plants to reduce infection risks caused by sampling (see Ng & Perry 2004). To test the coevolutionary hypothesis, it is therefore crucial to establish whether leaf colour functions as a visual signal that reduces initial landing rate before post-landing decisions can occur. Instead of using glue, Dö ring & Hardie (2007) advocate observations of aphid behaviour, but studies relying on observations have …