<p>There is an immense amount of variation in leaf shape, size, and colouration, both across and within plant species. Leaf shape and colour, in some instances, can be attributed as a physiological response to particular abiotic stressors. However, leaf shape, size, and colour are used by herbivores to identify sources of palatable foliage for food. It is possible, therefore, that an undefended plant might gain protection from herbivores by matching leaf characteristics of a chemically defended species. The matching of defensive signals by an undefended species in order to dupe a predator is known as Batesian mimicry, and whilst believed to be a relatively common phenomenon amongst animals, it has yet to be proven in plants. The foliage of Alseuosmia pusilla (Colenso) A. Cunningham, is strikingly similar to the human eye to that of Pseudowintera colorata (Raoul) Dandy, an unrelated sympatric species found in New Zealand. Unlike the foliage of A. pusilla, that of P. colorata contains a number of secondary metabolites associated with herbivore defence, including a sesquiterpene dialdehyde known as polygodial, a known potent insect antifeedant that imparts a pungent peppery taste when eaten. It has been hypothesised that this similarity evolved under browsing pressure from nine species of large extinct herbivorous birds, collectively known as moa. Whilst moa became extinct soon after the arrival of humans, the large herbivore guild has been effectively replaced by a range of introduced mammalian herbivores including several species of deer, though to what degree remains controversial. In chapter two, I established a robust spatially explicit morphometric analysis method to test how similar the leaves of A. pusilla and P. colorata leaves were, and whether leaf shape was a distinctive trait within their shared habitat. Using the Cartesian coordinates of leaf margins as descriptors of leaf shape, I found that P. colorata leaves were morphologically distinct from all of the neighbouring species except for those of A. pusilla. A. pusilla individuals were more similar to neighbouring than to distant P. colorata, and 90% of leaf shape variation in the two species varied similarly across an elevational gradient. The data are consistent with Batesian mimicry, wherein the conspicuous characteristic of a defended model is replicated by an undefended mimic across its entire growing range. In chapter three, I tested how leaf shape variation within, and between, A. pusilla and P. colorata responded when exposed to high levels of mammalian herbivory. I demonstrated that in a forest population of P. colorata and A. pusilla exposed to high mammalian herbivory pressure, leaf shape variation is reduced in both focal species, but not in other sympatric species. This is consistent with Batesian mimicry, wherein increased herbivory pressure selects for a stronger signal in the distinctive characteristic of the defended plant, and through the selection for mimicry, variation in the mimic’s phenotype converges on the model’s phenotype. Additionally, when alternative palatable food is preferentially targeted, P. colorata increased in abundance along with a proportionate increase in A. pusilla’s abundance. Invertebrate herbivory was estimated to be similar on both species at both sites. In chapter four, I tested the hypothesis that A. pusilla is a Batesian mimic of P. colorata using farmed red deer (Cervus elaphus scoticus) in feeding trials. The deer found A. pusilla more palatable than P. colorata, and after eating a P. colorata individual, they became reluctant to eat another plant. Although the two plants differ significantly in volatile organic compound emissions, deer were equally likely to first eat an A. pusilla as they were a P. colorata, therefore were unable to use olfactory cues, or visually differentiate between the two species. As the relative abundance of P. colorata increased, herbivory damage was lower, both in the defended P. colorata and in the undefended A. pusilla. This study provides the first unequivocal proof of defensive Batesian mimicry in plants. In chapter five, using humans as surrogate herbivores, I tested how leaf shape and colour can be used as cues or signals by herbivores when foraging for food under different conditions. Subjects found leaf size a distracting characteristic, foraging more effectively when A. pusilla and P. colorata individuals were most similar in 94% of their shared shape variation. The trait of leaf colour, whilst unreliable by itself, acted to potentiate the trait of leaf shape, as a signal or cue. Fast feedback on species palatability improved accuracy in identifying A. pusilla, but neither fast nor slow feedback improved discriminability of P. colorata. A. pusilla leaves were harder to discriminate when presented on a “disruptive” backdrop. My results demonstrate that leaf shape can act as a signal or cue. These results indicate why further research into plant-herbivore communication is important and that it could provide powerful insights into the functional significance of leaf morphology. This thesis provides a significant contribution to our understanding of how leaves function as signals or cues to herbivores in three ways: (i) it provides the first detailed and powerful quantitative evidence of leaf shape matching between two species, and demonstrates the importance of using a spatially explicit morphometric method when investigating leaf shape; (ii) it is the first to unequivocally prove defensive Batesian mimicry in plants; and (iii) it demonstrates that leaf traits can act as signals or cues.</p>
Geometric and topological approaches to shape variation in
Ginkgo
leaves
