“The falling leaves
Drift by my window
The falling leaves
Of red and gold”
Red, green and gold, all on one tree
When Frank Sinatra sang Autumn Leaves he was almost certainly not thinking of aphids and I am pretty certain that the English lyricist, Johnny Mercer, who translated the words from the original French by Jacques Prévert wasn’t either 🙂
The colours we see in autumn are mainly due to two classes of pigment, the carotenoids (yellow-orange; think carrot) and the anthocyanins (red-purple). Carotenoids are present in the leaves all year round but are masked by the green chlorophyll. Chlorophyll breaks down in autumn, leaving the yellow carotenes visible. The anthocyanins on the other hand are not formed until autumn (Sanger, 1971; Lee & Gould, 2002) and this mixture of pigments give us the colours that have inspired so many artists.
Autumn Leaves Georgia O’Keeffe (1924) Tate Modern
To many, autumn starts with the appearance of the first turning leaves, to me it is the arrival of gynoparae* of the bird cherry-oat aphid (Rhopalosiphum padi) on my bird cherry (Prunus padus) trees.
Bird cherry, Prunus padus, leaves on the turn.
Little did I know when I started my PhD in 1977 that almost thirty years later I would be part of a raging debate about the function of autumn colouration in woody plants. At the time I was interested in the colonisation patterns (or as I pretentiously termed it in my thesis ‘remigration’) of bird cherry aphids from their secondary grass and cereal host plants to their primary host bird cherry. My study system was 30 bird cherry saplings divided between two cold frames in the Biology Compound at the University of East Anglia (Norwich). Every day from the middle of August until leaf fall I checked every leaf of each tree, for gynoparae, males and oviparae, carefully noting the position of each leaf, its phenological stage and giving it a unique number. I repeated this in the autumns of 1978 and 1979. The phenological stage was based on the leaf colour: green, mature; yellow, mature to senescent; red, senescent. What I reported was that more gynoparae landed on green and yellow leaves than on red and that the gynoparae on green and yellow leaves survived for longer and produced more offspring (oviparae), than those on red leaves (Leather, 1981). The gynoparae of the bird cherry aphid are quite special in that although as adults they do not feed (Leather, 1982), they do not land on bird cherry trees at random (Leather & Lehti, 1982), but choose trees that not only do their offspring (the oviparae) do better on, but that also favour those aphids hatching from eggs in the spring (Leather, 1986). It should not have come as a surprise then, that when I analysed some of the data I had collected all those years ago, their preference for green and yellow leaves over red ones, is linked to how long those
Figure 1. Length of time leaves remained on tree after first colonisation by gynoparae of Rhopalosiphum padi (F = 30.1 df 2/77, P <0.001)
leaves have left to live (Figure 1). The timing of events at this time of year, has, of necessity, got to be very precise. The egg-laying females (oviparae) are unable to develop on mature bird cherry leaves (Leather & Dixon, 1981), but it seems that the bird cherry aphid has this under control, making its decisions about the timing of the production of autumn forms (morphs) sometime in August (Ward et al., 1984). All very sensible as far as I was concerned and that was as far as I took things. Subsequent work by Furuta (1986) supported this in that he showed that maple aphids settled on and reproduced on green-yellow and yellow-orange leaves but avoided red leaves which had shorter life spans.
Jump forward fifteen years or so, and in a paper, that at the time, had somehow passed me by, the late great Bill Hamilton and Sam Brown (Hamilton & Brown, 2001) hypothesised that trees with an intense autumn display, similarly to those brightly coloured animals that signal their distastefulness with yellows, blacks and reds, were signalling their unsuitability as a host plant to aphids. Like the costs imposed on insects that sequester plant toxins to protect themselves against predators, the production of anthocyanins responsible for the red autumn colouration is expensive, especially when you consider that the leaves have only a short time left to live (Hoch et al., 2001). In autumn, trees and woody shrubs are normally mobilising resources in the leaves and moving them back into themselves ready to be used again the following spring (Dixon, 1963). Ecologists and evolutionary biologists were thus keen to explain the phenomenon in terms of trade-offs, for example, fruit flags that advertise the position of fruits for those trees that rely on seed dispersal by vertebrates (Stiles, 1982) or as ultra-violet screens to prevent tissue damage (Merzlyak & Gittelson, 1995). Hamilton & Brown felt that these hypotheses were either, in the case of the fruit flag, only applicable to trees with fruit present and, in the latter, untenable. Instead they advocated the ‘signalling hypothesis’ which was based on the premise that trees that suffer from a lot of aphids (attacked by more than one species rather than by large numbers of a single species), invest in greater levels of defence and in autumn advertise this using bright warning colours. The premise being, that although it is metabolically expensive for the plants to produce these colours, it is worth the investment if they result in a reduction in aphid attack.
This hypothesis was not without its detractors. Others suggested, that far from avoiding red colours, aphids were attracted to yellow or green as an indicator of host nutrition (Wilkinson et al., (2002). Holopainen & Peltonen (2002) also suggested that birch aphids use the onset of autumn colours to pick out those trees where nutrient retranslocation was happening, and thus with higher levels of soluble nitrogen in the leaves. This was of course, what I was trying to confirm back when I was doing my PhD. Conversely, supporters of the signalling hypothesis, argued that trees (birch again) that could ‘afford’ to produce bright autumn colours were fitter, so more resistant in general and that they were warning potential herbivores of this by a bright autumn display (Hagen et al 2004).
Round about this time (2002), I was approached by a young Swiss researcher, Marco Archetti, who knew that I had a plot of sixty bird cherry trees that I had planted up when I arrived at Silwood in 1992, originally designed to follow-up some work that I had begun whilst at the Forestry Commission looking at the effects of early season defoliation on subsequent tree growth (Leather, 1993, 1995). Marco convinced me that I had the ideal set-up to test the ‘signalling hypothesis’ and what was to be a very fruitful collaboration began.
We counted arriving gynoparae and their offspring (oviparae) throughout October (Marco making trips over from Oxford where he was then based**) noting leaf colour before and after each count. As with my PhD work we found that the greener trees were preferentially colonised by the gynoparae and that more oviparae were produced on those trees and that given what I had found earlier that bird cherry aphid gynoparae chose trees that are good hosts in spring (Leather, 1986), Marco felt that we were able to support the honest signalling hypothesis (Archetti & Leather, 2005). I was slightly less comfortable about this, as there are only two species of aphid that attack bird cherry and one of those is very rare and the original signalling hypothesis was based on the premise that it was trees that were attacked by a lot of aphid species that used the red colouration as a keep clear signal. Anyway, it was published 🙂
That said, others agreed with us, for example, Schaefer & Rolshausen (2006) who called it the defence indication hypothesis, arguing that bright colours advertise high levels of plant defence and that the herbivores would do well to stay away from those plants displaying them. On the other hand, Sinkkonen (2006) suggested that reproductively active plants produce autumn colours early to deter insects from feeding on them and thus reduce their seed set.
Chittka & Döring (2007) on the other hand, suggested that there is no need to look further than yellow carotenoids acting as integral components of photosynthesis and protection against light damage and red anthocyanins preventing photo-inhibition (Hoch et al., 2001) as to why trees turn colourful in autumn. In other words, nothing to do with the insects at all. A couple of years later however, Thomas Döring and Marco got together with another former colleague of mine from Silwood Park, Jim Hardie, and changed their minds slightly. This time, whilst conceding that red leaves are not attractive to aphids but noting that yellow leaves are even more attractive than green ones, suggested that the red colour could be being used to mask yellow (Döring et al., 2009).
Others have their own pet theories. In recent years, veteran Australian entomologist Tom White has become interested in the concept of insect species that specifically feed on senescent plant tissue (White, 2002, 2015) and added to the debate by suggesting that aphids in general are senescence feeders and thus choose green and yellow as they have longest time to live and that the red leaves are also nitrogen depleted (White, 2009) which is supported by my PhD data (Figure 1). This resulted in a spirited response by Lev-Yadun & Holopainen (2011) who claimed that he had misunderstood the scenario in thinking that leaves go sequentially from green to yellow to red, which they suggest is rare (I question this) and that actually in trees that go from green to red, the leaves still contain significant amounts of nitrogen, so a deterrent signal is still required.
Maple, green to yellow in this case
Spindle, Euonymus europaeus, green to red
What about those trees and other plants that have red or purple leaves in the spring or all year round and not just in autumn?
Some trees have red foliage all year
Trees like some of the ornamental cherries or copper beech? I haven’t been able to find any papers that suggest that red or purple-leaved varieties of beech and cherries are less susceptible to aphid attack. My own observations, probably imperfectly recalled, are that copper beech is regularly infested by the beech woolly aphid, Phyllaphis fagi , and just as heavily, if not more so than the normal green-leaved beech trees. That of course may just be a reflection that the white waxy wool covering the aphid stands out more against the red leaves. Perhaps someone out here might like to check this out? Some work that my friend and former colleague, Allan Watt, (sadly unpublished) did many years ago in Scotland looking at the effect of beech species and cultivar on infestation levels by the beech leaf mining weevil, Rhynchaenus fagi, did not indicate any differences between copper and green cultivars. It does seem however, that in cabbages, leaf colour can tell the specialist cabbage aphid, Brevicoryne brassciae, if plants are well defended or not, the bluer the cabbage, the nastier it is (Green et al, 2015).
- Red leaves are produced by the trees in autumn to reduce ultraviolet damage and protect metabolic processes in the leaf.
- Red leaves are deliberately produced by the tree to warn aphids that their leaves are well defended – honest signalling.
- Red leaves are produced by the tree to ‘fool’ the herbivores that the leaves are likely to drop soon and warn them to keep away so as to safeguard their fruit – dishonest signalling.
- The tree is blissfully unaware of the aphids and the aphids are exploiting the intensity of the autumn colours produced by the trees to select which are the best trees to colonise in terms of nutrition and length of time left on the tree.
As I write, the debate still goes on and we seem no nearer to arriving at a definitive answer to the riddle of why trees produce bright leaves in autumn. If nothing else however, the debate has generated a lot of interest and enabled people to sneak some amusing titles into the scientific literature. Do make the effort to read the titles of some of the references below.
Archetti, M. (2009) Phylogenetic analysis reveals a scattered distribution of autumn colours. Annals of Botany, 103, 703-713.
Archetti, M. & Leather, S.R. (2005) A test of the coevolution theory of autumn colours: colour preference of Rhopalosiphum padi on Prunus padus. Oikos, 110, 339-343.
Chittka, L. & Döring, T.F. (2007) Are autumn foliage colors red signals to aphids? PLoS Biology , 5(8): e187. Doi:10.1371/journal.pbio.0050187.
Dixon, A.F.G. (1963) Reproductive activity of the sycamore aphid, Drepanosiphum platanoides (Schr) (Hemiptera, Aphididae). Journal of Animal Ecology, 32, 33-48.
Döring, T.F., Archetti, M. & Hardie, J. (2009) Autumn leaves seen through herbivore eyes. Proceedings of the Royal Society London B., 276, 121-127.
Furuta, K. (1986) Host preferences and population dynamics in an autumnal population of the maple aphid, Periphyllus californiensis Shinji (Homoptera: Aphididae). Zeitschrift fur Angewandte Entomologie, 102, 93-100.
Green, J.P., Foster, R., Wilkins, L., Osorio, D. & Hartley, S.E. (2015) Leaf colour as a signal of chemical defence to insect herbivores in wild cabbage (Brassica oleracea). PLoS ONE, 10(9): e0136884.doi:10.1371/journal.pone.0136884.
Hagen, S.B. (2004) Autumn coloration as a signal of tree condition. Proceedings of the Royal Society London B, 271, S184-S185.
Hamilton, W.D. & Brown, S.P. (2001) Autumn tree colours as handicap signal. Proceedings of the Royal Society London B, 268, 1489-1493.
Hoch , W.A., Zeldin, E.L. & McCown, B.H. (2001) Physiological significance of anthocyanins during autumnal leaf senescence. Tree Physiology, 21, 1-8.
Holopainen, J.K. & Peltonen, P. (2002) Bright colours of deciduous trees attract aphids: nutrient retranslocation hypothesis. Oikos, 99, 184-188.
Leather, S.R. (1981) Reproduction and survival: a field study of the gynoparae of the bird cherry-oat aphid, Rhopalosiphum padi (L.). Annales Entomologici Fennici, 47, 131-135.
Leather, S.R. (1982) Do gynoparae and males need to feed? An attempt to allocate resources in the bird cherry-oat aphid Rhopalosiphum padi. Entomologia experimentalis et applicata, 31, 386-390.
Leather, S.R. (1986) Host monitoring by aphid migrants: do gynoparae maximise offspring fitness? Oecologia, 68, 367-369.
Leather, S.R. (1993) Early season defoliation of bird cherry influences autumn colonization by the bird cherry aphid, Rhopalosiphum padi. Oikos, 66, 43-47.
Leather, S.R. (1995) Medium term effects of early season defoliation on the colonisation of bird cherry (Prunus padus L.). European Journal of Entomology, 92, 623-631.
Leather, S.R. & Dixon, A.F.G. (1981) Growth, survival and reproduction of the bird-cherry aphid, Rhopalosiphum padi, on its primary host. Annals of Applied Biology, 99, 115-118.
Leather, S.R. & Lehti, J.P. (1982) Field studies on the factors affecting the population dynamics of the bird cherry-oat aphid, Rhopalosiphum padi (L.) in Finland. Annales Agriculturae Fenniae, 21, 20-31.
Lee, D.W. & Gould, K.S. (2002) Anthocyanins in leaves and other vegetative organs: An introduction. Advances in Botanical Research, 37, 1-16.
Lev-Yadun, S. & Holopainen, J.K. (2011) How red is the red autumn leaf herring and did it lose its red color? Plant Signalling & Behavior, 6, 1879-1880.
Merzlyak, W.N. & Gittelson, A. (1995) Why and what for the leaves are yellow in autumn? On the interpretation of optical spectra of senescing leaves (Acer platanoides L.). Journal of Plant Physiology, 145, 315-320.
Sanger, J.E. (1971) Quantitative investigations of leaf pigments from their Inception in buds through autumn coloration to decomposition in falling leaves. Ecology, 52, 1075-1089.
Schaefer, H.M. & Rolshausen, G. (2006) Plants on red alert – do insects pay attention? BioEssays, 28, 65-71.
Sinkkonen, A. (2006) Do autumn leaf colours serve as reproductive insurance against sucking herbivores? Oikos, 113, 557-562.
Stiles, E.W. (1982) Fruit flags: two hypotheses. American Naturalist, 120, 500-509.
Ward, S.A., Leather, S.R., & Dixon, A.F.G. (1984) Temperature prediction and the timing of sex in aphids. Oecologia, 62, 230-233.
White, T.C.R. (2003) Nutrient translocation hypothesis: a subsect of the flush-feeding/senescence-feeding hypothesis. Oikos, 103, 217.
White, T.C.R. (2009) Catching a red herring: autumn colours and aphids. Oikos, 118, 1610-1612.
White, T.C.R. (2015) Senescence-feesders: a new trophic subguild of insect herbivore. Journal of Applied Entomology, 139, 11-22.
Wilkinson, D.M., Sherratt, T.N., Phillip, D.M., Wratten, S.D., Dixon, A.F.G. & Young, A.J. (2002) The adaptive significance of autumn colours. Oikos, 99, 402-407.
*for a detailed account of the wonderful terminology associated with aphid life cycles read here
**coincidentally he is now a Lecturer at the University of East Anglia in the same Department where I did my PhD