August 2019

Vol. 47, No. 4

Gleanings: A Leaf of Another Color

David M. Larson

Human eyes have three types of retinal cones for detecting what we call "visible light." Many animals have a wider spectrum of detection, with their visible light extending from the ultraviolet to the infrared, and they have more types of cones for detecting these wavelengths. In the article by Tedore and Nilsson (2019), the authors examine the wavelengths related to the tetrachromatic vision of most terrestrial-foraging birds.

These birds have four types of cones in the retina for detecting colors: red (L), green (M), and blue (S), as in humans, and ultraviolet (UV). Generally speaking, L cones are sensitive in the 600–700nm wavelengths, M cones in the 500–600nm range, and S-cones in the 400–500nm range. Bird taxa have two types of UV-detecting cones: U cones have a peak sensitivity at about 370nm, and V cones at about 410nm. While there have been many studies showing that UV detectors correlate and probably have co-evolved with flower color signals, the utility of UV imaging in more mundane life experiences is unknown. The authors used a multispectral camera, which mimicked the basic spectral sensitivity curves for the four avian cone types, including variants, to image three basic habitats: open rangeland and deciduous forest fragments in Sweden, old-growth wet sclerophyll scrub with open canopy in Australia, and old-growth tropical and subtropical rain forests in Australia. They found that the contrast between upper and lower leaf surfaces was exaggerated in the UV in comparison to the visible wavelengths. Lower leaf surfaces were dark in the UV channels, increasing contrast in the environment and enhancing structural detail.

Interestingly, the green channel (peak 540nm) was quite unhelpful in distinguishing leaf contrast. This is the result of similar reflectance and transmission of green light in leaves. In contrast, for UV light, reflectance is much higher than transmission. Since the primary light sources in forests are from above, the reflection of UV light from the upper surfaces outweighs the reflection from the lower surface. Tedore and Nilsson found that leaves reflect more than 25 times as much UV light (300–400nm) as they transmit, while they and others have shown that the ratios for L, M, and S ranges are in the 0.7–1.5 range. The utility of sharpened contrast and detail in leaves is obvious for birds and other animals that need to forage, hide, and live in a leafy environment.

The authors also tested for an effect of habitat type in the optimal tuning of UV- and blue-sensitive detectors. It is important to note that these birds have UV cone types that are matched with specific blue cones: U cones with a peak sensitivity at ~370nm are matched with S(U) cones at 460nm, and V cones with a peak sensitivity at ~410nm are paired with S(V) cones at 470nm. In terms of leaf contrast, the V-cone was particularly useful in rain forest and wet sclerophyll habitats, while the S(U)-cone was useful in all habitats. Carrying out a one-by-one substitution of U, V, S(U), and S(V) wavelength detectors gave largely intermediate values. The U, S(U) combination found in birds was advantageous for leaf color contrast detection in deciduous and wet sclerophyll habitats. It is possible that changing environmental conditions over evolutionary history explain the repeated differential expression of these cones over time.

The authors optically modelled the system, randomizing up to 12 different habitat and environmental variables in repeated calculations. Using just diffuse reflections and transmittance, modeling did not show conditions where the V cone would show higher leaf contrast than the U cone. When they added specular reflections—reflections of sky or canopy from waxy leaf cuticle—the optical model showed that the V-cones showed higher leaf contrast than the U-cones when the specular light came from overlying leaves. Canopies more than 80% closed shifted the spectrum of light from the upper surface but not lower surface of leaves. At over 80% canopy coverage, leaf contrast was higher above 400nm. At less than 80% canopy, leaf contrast was higher below 400nm, suggesting that V-cones, with long UV wavelengths, would be more effective in very dense forest while U cones, with short UV wavelengths, would be more effective in more open forest.

The authors conclude that the shorter wavelengths—violet and ultraviolet—likely provide a useful discriminator in birds (and probably in other taxa with UV-sensitive visual sensors) for visualizing forest structure. Further, the optical model suggests that the evolution and persistence of cones with different sensitivities to ultraviolet is adaptive during changing environmental conditions. Overall, the utility of UV-sensitive cones goes beyond detection of specialized plants and seems integral to daily life in terrestrial foraging birds.


David M. Larson, PhD, is the Science and Education Coordinator at Mass Audubon's Joppa Flats Education Center in Newburyport, the Director of Mass Audubon's Birder's Certificate Program and the Certificate Program in Bird Ecology (a course for naturalist guides in Belize), a domestic and international tour leader, President of the Nuttall Ornithological Club, and a member of the editorial staff of Bird Observer.

blog comments powered by Disqus
Bird Observer logo

Our mission: to support and promote the observation, understanding, and conservation of the wild birds of New England.

Bird Observer supports the right of all people to enjoy birding and nature in a safe and welcoming environment free from discrimination and harassment, be it sexual, racial, or barriers for people with disabilities.
© Copyright 2023 by Bird Observer, Inc.