In many woody plants, expanding leaves flush red because they accumulate anthocyanin pigments before greening at maturity. A comparative study of 76 subtropical woody species (Chen and Huang 2013, Oikos) showed that species with red young leaves have significantly less mechanical defence — trichomes, enhanced cuticle, multiple epidermis — than species with green young leaves, and phylogenetic reconstructions suggested that gains of mechanical protection co-occur with evolutionary loss of redness, implying that pigmentation and mechanical construction function as alternative defensive strategies during the vulnerable leaf-expansion phase. What has not been established is why this association exists. The causal mechanisms and selective agents that drive gains and losses of red young-leaf pigmentation versus recruitment of mechanical defences remain unidentified; the hypothesized functional trade-off between rapid leaf expansion and early investment in cuticle or other physical defence has not been experimentally quantified; and the relative importance of the several adaptive hypotheses for red coloration — photoprotection, warning signaling, undermining of herbivore camouflage, attraction of natural enemies, and direct chemical or antifungal defence — which may operate simultaneously, has not been disentangled across species that differ in mechanical defence levels. Later work has replicated the negative association across communities and a 2025 phylogenetic meta-analysis confirmed that nongreen leaves are chemically better defended and suffer less herbivory, but explicitly recorded insufficient data to test the mechanical-defence and top-down (enemy-attraction) dimensions, leaving the causal core open. The problem is to determine, experimentally and comparatively, what generates and maintains the pigmentation–mechanical-defence trade-off, including its allocation basis, fitness consequences, and the anatomical localization patterns that constrain candidate mechanisms.

Background

In many woody plants, leaves are not green while they are young: expanding leaves flush red or purple because they synthesize anthocyanins — water-soluble red-to-violet flavonoid pigments — in young leaf tissue before the leaf greens at maturity. In tropical floras many species additionally show delayed greening, in which leaves remain pale and photosynthetically weak until fully expanded; this has been interpreted as an anti-herbivore defence because young pale leaves are poor food for herbivores (Kursar and Coley 1992; Queenborough et al. 2013). Several adaptive hypotheses compete to explain red young-leaf coloration: (i) photoprotection — anthocyanins act as a light-attenuating screen or antioxidant shielding young, not-yet-fully-functional chloroplasts from excess light and reactive oxygen species; (ii) the warning-signal (aposematic or coevolutionary) hypothesis, imported from the autumn-colour literature (Archetti 2000; Hamilton and Brown 2001), in which red honestly signals low palatability — high chemical-defence content — to herbivores; (iii) undermining of herbivore camouflage (crypsis), because herbivorous insects resting on a red leaf become conspicuous to their own predators; (iv) attraction of natural enemies (predators or parasitoids) of herbivores; and (v) antifungal defence (Coley and Aide 1989; Tellez et al. 2016). All of these hypotheses implicitly assume that pigmentation substitutes for other defences during the expansion phase.

Mechanical (physical) defences of young leaves include trichomes (epidermal plant hairs), an enhanced or incrassate cuticle (a thickened waxy layer over the epidermis), and multiple epidermis (more than one epidermal cell layer). Chen and Huang (2013) provided the first broad, phylogenetically aware comparative test of whether colour-based and mechanical defences are alternative strategies. Surveying 76 woody species of a subtropical flora (Wuhan, central China), they found that the 40 species with green young leaves had a significantly higher incidence of enhanced cuticle and trichomes than the 36 species with red young leaves, in both G-tests of association (a likelihood-ratio form of contingency test; e.g. trichomes G = 9.345, p = 0.002; enhanced cuticle G = 33.11, p < 0.001) and in Pagel's phylogenetic correlation test (a likelihood method asking whether two binary characters evolve in a correlated fashion across a phylogeny); the composite trait 'at least one mechanical protection' was strongly negatively associated with redness (G = 51.2, p < 0.001). Ancestral-state reconstructions suggested that green young leaves are the ancestral state, that redness arose at least six times independently, and that gains of mechanical protection frequently co-occur with reconstructed transitions from red back to green young leaves. The authors read this as evidence that red young leaves substitute pigment-based for mechanical defence — additional support for the adaptive interpretation of leaf colour change — while explicitly emphasizing the limits of their evidence: co-occurrence of reconstructed changes on a phylogeny establishes neither temporal order nor the causal mechanisms or selective agents involved; the predominance of anthocyanin located in the mesophyll (the leaf's inner photosynthetic tissue, beneath the epidermis) and concentrated on the abaxial (lower) leaf surface in their sample differs from patterns reported for tropical floras and remains unexplained; multiple epidermis was too rare (seven species) for robust inference; the hypothesized allocation trade-off between rapid leaf expansion and early cuticle investment was not quantified; and the individual adaptive hypotheses, which may operate simultaneously, were not disentangled.

Subsequent work has strengthened the pattern but not resolved its cause. Chen, Huang and Tang (2021) replicated the negative association between redness and mechanical protection across five forest communities in south-central China (151 species, including Pagel correlated-evolution tests) and, using leaf reflectance spectra analysed with a bee-vision colour model (a standard proxy for how leaf colour appears to insect eyes, which lack the red-sensitive receptors of human vision), found that the most common subtropical herbivores (lepidopteran larvae and coleopterans) can probably hardly distinguish red from green young leaves — concluding that anthocyanin likely defends 'in some other way beyond visual warning'. Gong et al. (2020) found in a tropical assemblage that red young leaves suffered less herbivory and contained more tannins and anthocyanins, whereas chlorophyll-fluorescence indices did not support the light-attenuator photoprotection function there. A phylogenetic meta-analysis of 160 registers (Cornelissen et al. 2025) confirmed that nongreen leaves are better defended chemically, less palatable, less attacked, and reduce herbivore fitness, with stronger herbivory effects in tropical plants — but recorded insufficient studies measuring mechanical defence or camouflage to test those mechanisms, and found the predator-attraction hypothesis almost untested, explicitly calling for future experimental work manipulating predator access. Behavioural experiments with generalist grasshoppers support a warning function of red/purple leaf colour under white light (Ide 2022; Ide 2025), although Lev-Yadun (2023) argued that the same data also support the camouflage-undermining hypothesis. Photoprotection by young-leaf anthocyanins is real but season- and context-dependent, operating alongside non-photochemical quenching (the chloroplast's built-in pathway for dissipating excess absorbed light energy as heat) and enzymatic antioxidants (Yu et al. 2020). Demographic work links delayed greening and red young leaves to higher survival at a cost of lower growth (Queenborough et al. 2013), and the negative association between redness and physical defence persists along subtropical successional gradients (Han et al. 2019). Whether the pigmentation–mechanical-defence trade-off is causal — and what generates and maintains it — therefore remains open.

Problem Statement

What causal mechanisms and selective agents generate and maintain the negative association between anthocyanin-based red young-leaf coloration and mechanical defence (trichomes, enhanced cuticle, multiple epidermis) across woody plant lineages? A resolution must address all four of the following: (1) which selective pressures — for example, herbivore regimes, physiological costs of pigment synthesis, and environmental constraints such as light environment — favour red pigmentation over mechanical-defence recruitment during leaf expansion, and which drive the reverse transitions from red to mechanically defended green young leaves; (2) through which mechanism or combination of mechanisms (namely photoprotection, warning signaling, undermining of herbivore camouflage, attraction of natural enemies, and direct chemical or antifungal defence) red pigmentation reduces damage in mechanically undefended expanding leaves, and their relative importance in species that differ in mechanical defence levels; (3) how both energetic and developmental constraints shape the allocation between pigment synthesis and mechanical construction during rapid leaf expansion, and what the consequences of this allocation are for herbivore fitness and plant fitness in situ; and (4) why anthocyanin localization within the leaf (mesophyll, the inner photosynthetic tissue, versus epidermis; abaxial, lower, versus adaxial, upper, leaf surface) differs between subtropical and tropical floras — a pattern that constrains which optical and defensive mechanisms are physiologically plausible. None of the four parts substitutes for another, and each requires a determinate, evidence-supported answer rather than a list of possibilities: the mechanisms in (2) must be shown to operate and their relative importance ranked, and the localization pattern in (4) must be brought to bear as a constraint that discriminates among candidate mechanisms. Purely correlational replications of the association, taken alone, do not answer the question.

References

  1. Chen, Y.-Z. & Huang, S.-Q. (2013). Red young leaves have less mechanical defence than green young leaves. Oikos 122: 1035–1041. DOI 10.1111/j.1600-0706.2012.20852.x. https://doi.org/10.1111/j.1600-0706.2012.20852.x
  2. Chen, Y., Huang, Z. & Tang, L. (2021). Invisible red in young leaves: Anthocyanin likely plays a defensive role in some other way beyond visual warning. Flora 280: 151833. DOI 10.1016/j.flora.2021.151833. https://doi.org/10.1016/j.flora.2021.151833
  3. Cornelissen, T., Silveira, F. A. O., Gomes, S. V., López-Goldar, X., Martin-Eberhardt, S. A. & Wetzel, W. (2025). Signaling defenses with color: a meta-analysis of leaf color variation, palatability, and herbivore damage. New Phytologist 247: 884–896. DOI 10.1111/nph.70243. https://doi.org/10.1111/nph.70243
  4. Gong, W.-C., Liu, Y.-H., Wang, C.-M., Chen, Y.-Q., Martin, K. & Meng, L.-Z. (2020). Why Are There so Many Plant Species That Transiently Flush Young Leaves Red in the Tropics? Frontiers in Plant Science 11: 83. DOI 10.3389/fpls.2020.00083. https://doi.org/10.3389/fpls.2020.00083
  5. Han, T., Wang, J., Ren, H., Yi, H., Zhang, Q. & Guo, Q. (2019). Changes in defense traits of young leaves in subtropical forests succession. Plant Ecology 220: 305–320. DOI 10.1007/s11258-019-00916-1. https://doi.org/10.1007/s11258-019-00916-1
  6. Queenborough, S. A., Metz, M. R., Valencia, R. & Wright, S. J. (2013). Demographic consequences of chromatic leaf defence in tropical tree communities: do red young leaves increase growth and survival? Annals of Botany 112: 677–684. DOI 10.1093/aob/mct144. https://doi.org/10.1093/aob/mct144
  7. Gomes, S. V. & Cornelissen, T. (2025). Shifts in leaf color during ontogeny do not affect herbivory in the tropical tree Lecythis pisonis (Lecythidaceae). Plant Species Biology 40: 500–509. DOI 10.1111/1442-1984.70017. https://doi.org/10.1111/1442-1984.70017
  8. Hughes, N. M. & Lev-Yadun, S. (2023). Review: Why do some plants have leaves with red or purple undersides? Environmental and Experimental Botany 205: 105126. DOI 10.1016/j.envexpbot.2022.105126. https://doi.org/10.1016/j.envexpbot.2022.105126
  9. Ide, J.-Y. (2022). Why do red/purple young leaves suffer less insect herbivory: tests of the warning signal hypothesis and the undermining of insect camouflage hypothesis. Arthropod-Plant Interactions 16: 567–581. DOI 10.1007/s11829-022-09924-x. https://doi.org/10.1007/s11829-022-09924-x
  10. Lev-Yadun, S. (2023). Testing of two of the visual anti-herbivory strategies of red young leaves: aposematic versus undermining herbivorous insect's camouflage. Arthropod-Plant Interactions 17: 615–619. DOI 10.1007/s11829-023-09980-x. https://doi.org/10.1007/s11829-023-09980-x
  11. Ide, J.-Y. (2025). Effect of color on the feeding preference of a generalist grasshopper, Acrida cinerea (Orthoptera: Acrididae). Journal of Asia-Pacific Entomology 28: 102415. DOI 10.1016/j.aspen.2025.102415. https://doi.org/10.1016/j.aspen.2025.102415
  12. Yu, Z.-C., Lin, W., Zheng, X.-T., Chow, W. S., Luo, Y.-N., Cai, M.-L. & Peng, C.-L. (2020). The relationship between anthocyanin accumulation and photoprotection in young leaves of two dominant tree species in subtropical forests in different seasons. Photosynthesis Research 149: 41–55. DOI 10.1007/s11120-020-00781-4. https://doi.org/10.1007/s11120-020-00781-4
  13. Tellez, P., Rojas, E. & Van Bael, S. (2016). Red coloration in young tropical leaves associated with reduced fungal pathogen damage. Biotropica 48: 150–153. DOI 10.1111/btp.12303. https://doi.org/10.1111/btp.12303
  14. Coley, P. D. & Aide, T. M. (1989). Red coloration of tropical young leaves: a possible antifungal defence? Journal of Tropical Ecology 5: 293–300. DOI 10.1017/s0266467400003667. https://doi.org/10.1017/s0266467400003667
  15. Kursar, T. A. & Coley, P. D. (1992). Delayed greening in tropical leaves: an antiherbivore defense? Biotropica 24: 256–262. DOI 10.2307/2388520. https://doi.org/10.2307/2388520
  16. Archetti, M. (2000). The origin of autumn colours by coevolution. Journal of Theoretical Biology 205: 625–630. DOI 10.1006/jtbi.2000.2089. https://doi.org/10.1006/jtbi.2000.2089
  17. Hamilton, W. D. & Brown, S. P. (2001). Autumn tree colours as a handicap signal. Proceedings of the Royal Society of London Series B 268: 1489–1493. DOI 10.1098/rspb.2001.1672. https://doi.org/10.1098/rspb.2001.1672