Preference‐performance relationship in the thistle tortoise beetle: Implications for its realised host range

Abstract The evolutionary constraints on host plant preferences of herbivorous insects are a question of theoretical and practical importance. Since larvae have limited mobility, natural selection pressures should promote host plant choices by adults that optimise larval survival. Such optimal host utilisation is predicted to reinforce the evolution of specialisation. This concept is expressed and tested as the ‘preference‐performance hypothesis’ (PPH), which predicts a positive relationship between adult preference and larval performance. Through a series of choice and no‐choice experiments, the preference‐performance relationship was examined for the oligophagous tortoise beetle, Cassida rubiginosa , used for biocontrol of thistle weeds. It was hypothesised that adult host plant choices based on olfaction, final location, feeding intensity and oviposition are adaptive and reflect the optimal hosts for larval performance (survival). In accordance with the PPH, there was a positive relationship between all measures of adult preference and larval performance. Adult preference based on olfactometer experiments showed a significant but weak correlation, explaining 27.6% of the variation in larval survival. The strongest correlation was found for adult preference measured as the amount of feeding, explaining 72.5% of the variation in larval survival. Adult preference based on location at the end of the experiment showed a similarly strong correlation, explaining 69.4% of the variation in larval survival. Adult preference based on oviposition and the number of hatched larvae was moderately correlated with larval survival, explaining 46.8% and 48.7% of larval survival, respectively. While there is a significant positive relationship between preference and performance, adult host preference is only a moderate predictor of larval performance. The adult beetle showed substantial flexibility in host plant selection, and equal performance across most Cardueae hosts. This is likely due to the larvae being more sensitive to physical and chemical defences, and possibly an evolutionary lag in the host range between the life stages. Any expansion of the realised host range will be limited by the ability of the larva to adapt to suboptimal hosts.