Plant Spines and Caterpillar Feeding

A tobacco hornworm, Manduca sexta, struggles with the spines of a purple devil, Solanum atropurpureum. Photo by Rupesh Kariyat from the cover of Biology Letters.

This week's paper is in Biology Letters, lead authored by Rupesh R. Kariyat at ETH Zürich. Kariyat and his colleagues examined the interactions between a moth, the tobacco hornworm, Manduca sexta, and horsenettle plants, Solanum carolinense. In previous research, they had shown that horsenettle fed upon by moth caterpillars grows more spines. In this paper they examine how spines impact the caterpillars. 

What they did

They examined the effect of spines on caterpillar performance three ways. First they used inbred and outbred lines of S. carolinense, because inbred lines have less spines than outbred lines. Second, they used three closely related plant species that vary in the number of spines: S. carolinense, S. aethiopicum, and S. atropurpureum. Finally they experimentally removed spines (cut them off with a razor blade) in both the inbred/outbred experiment and the three species experiment. They then examined how long it took a caterpillar placed on the soil at the bottom of the plant to reach a single leaf at the top of the plant, and how long it took a caterpillar to totally defoliate the plant. 

What they found

Kariyat and his colleagues consistently found that plants caterpillars took longer to travel up plants and defoliate plants when plants had more spines. This effect disappeared when the spines were removed. High spine densities not only appeared to slow caterpillars down, but also made them fall off the plants more often. 

The takeaway

This result is intuitive and supports the function that many of us would have hypothesized for plant spines. But measuring the effect of spines is tricky, because plants that vary in spine number likely vary in many traits, and if you remove spines it may cause the plants to induce defenses (plants can become better defended when damaged as noted above with plants that grew more spines, but they can also increase their concentration of toxic chemicals etc). This paper handles these problems nicely by testing the hypothesis with multiple experimental variations. None of them are perfect, but together they tell a convincing story.