Flowers, Moths, and History of Science

The 7 June issue of Nature has a pretty cool article by Justen B. Whittall and Scott A. Hodges, “Pollinator shifts drive increasingly long nectar spurs in columbine flowers” (pp.706-709). They take up the famous example of the prediction by Charles Darwin that the long nectar spur in Angraecum sesquipedale indicated the existence of a long-tongued pollinator to match. Darwin didn’t live to see his prediction confirmed in 1903, when Xanthopan morgani spp. praedicta was described, complete with a whopping 22 cm tongue.

Whittall and Hodges, though, are concerned not with the prediction of existence of the specific pollinator, but rather with Darwin’s chosen mechanism by which he thought the situation with the flower and matching pollinator came about.

Darwin proposed that a coevolutionary ‘race’ had driven the directional increase in length of a plant’s spur and its pollinator’s tongue.


Under a hypothesis first proposed by Darwin and later elaborated by Wallace, nectar spurs and pollinator tongues are engaged in a one-to-one coevolutionary ‘race’.

Whittall and Hodges propose as an alternative explanation what they describe as a “pollinator shift” model.

Spurs may also evolve to exceptional lengths by way of pollinator shifts as plants adapt to a series of unrelated pollinators, each with a greater tongue length.


Figure 1. Two contrasting hypotheses for the evolution of exceptionally long nectar spurs. Darwin’s coevolutionary race model (a,b), which posits a gradual increase in both the pollinator’s tongue and the plant’s nectar spur, and the pollinator shift model (c,d), where spur length evolves owing to a switch to a new pollinator with a longer tongue. These models differ in whether adaptive peaks are constantly increasing (b), or whether they are relatively fixed optima based on pollinators’ pre-existing tongue lengths (d). They also differ in whether spur-length evolution occurs gradually (b) or in a punctuated fashion (d).

The research looks good, and should prove useful for future studies. Where Whittall and Hodges fall down a bit, though, is in their treatment of the history of ideas. Read on for more.

As this is a letter in Nature, the paper is somewhat telegraphic. The means by which they approach the research is through phylogenetic analysis of 25 North American Aquilegia taxa of columbines, plus another 8 outgroup taxa. What they primarily concentrate upon are apparent shifts in the phylogeny from one “pollinator syndrome” to another. A pollinator syndrome is simply a grouping based on multiple factors. In this case, three pollinator syndromes are identified, grouping flowers that are pollinated by bumble bees, pollinated primarily by hummingbirds, and pollinated primarily by hawkmoths. The bumble bees pollinate columbines with short nectar spurs, hummingbirds do most of the work for columbines with intermediate length nectar spurs, and mostly hawkmoths pollinate columbines with nectar spurs from about 35 cm to 123 cm in length.

The data set supports the alternate model of pollinator shifts, Whittall and Hodges find. Where the phylogenetic analysis indicates that there is a shift between pollinator syndromes, that is strongly directional, going from shorter nectar spurs to longer, and thus transiting from bumble bees to hummingbirds, or from hummingbirds to hawkmoths as the primary pollinator. These transitions also provide evidence for the stability of nectar spur length except for when such a transition occurs. This work is a pretty elegant demonstration of phylogenetic support for punctuated equilibria. There is also the interesting discussion of why columbines of Eurasia have little variation in spur length, where apparently those columbines all fall within the bumble bee pollinator syndrome. Eurasia doesn’t have hummingbirds, though it does have hawkmoths. The lack of the intermediate pollinator syndrome, Whittall and Hodges speculate, may have been what has prevented the establishment of hawkmoth pollination of the Eurasian columbines.

The concluding paragraph is worth a look:

Although Darwin’s coevolutionary race may be responsible for spur-length evolution within species, our comparative phylogenetic evidence indicates that the majority of spur-length evolution in columbines fits the pollinator shift model. Because columbines have experienced a recent and rapid adaptive radiation, it is likely that pollinator tongue lengths were predominantly established before spur-length evolution, which has thus evolved primarily during repeated and directional shifts among pollination syndromes. Of particular note is the finding that shifts in pollination syndrome have occurred without reversals, resulting in the progressive lengthening of nectar spurs. Our results also indicate that large changes in spur length occur disproportionately at speciation events, resulting in ‘punctuated’ morphological changes.

OK, now for some critical commentary.

Whittall and Hodges put the pollinator shift hypothesis of Wasserthal (1997) to empirical test. But while Wasserthal had in mind specifically the case of Angraecum sesquipedale as being better explained by pollinator shift, Whittall and Hodges worked with flowers a lot closer to home. This still leaves the specific case at issue open. How much of A. sesquipedale‘s evolution can be described as “coevolutionary race” and how much is due to “pollinator shift”? That, at least, can be approached empirically, but it will require actual work in Madagascar.

Further, there is the question of just how distinct “coevolutionary race” and “pollinator shift” are as theoretical constructs. I don’t think that the modern distinction being made between the two can be supported as something Darwin and Wallace decided between. When I first read the Whittall and Hodges article, I was thinking along the lines of writing this essay in terms of how Darwin may have been over-eager to apply Occam’s Razor, and that for A. sesquipedale it may be exactly multiple entities (as in multiple pollinators to be shifted between) that provide the best explanation. But since I’ve been burned before by punctuationist mishandling of Darwin, I had to go have a look at the original text by Darwin in his book on fertilization of orchids. Here it is:

What can be the use, it may be asked, of a nectary of such disproportionate length? We shall, I think, see that the fertilisation of the plant depends on this length, and on nectar being contained only within the lower and attenuated extremity. It is, however, surprising that any insect should be able to reach the nectar. Our English sphinxes have proboscides as long as their bodies; but in Madagascar there must be moths with proboscides capable of extension to a length of between ten and eleven inches! This belief of mine has been ridiculed by some entomologists, but we now know from Fritz Müller* that there is a sphinxmoth in South Brazil which has a proboscis of nearly sufficient length, for when dried it was between ten and eleven inches long. When not protruded it is coiled up into a spiral of at least twenty windings.


If the Angræcum in its native forests secretes more nectar than did the vigorous plants sent me by Mr. Bateman, so that the nectary ever becomes filled, small moths might obtain their share, but they would not benefit the plant. The pollinia would not be withdrawn until some huge moth, with a wonderfully long proboscis, tried to drain the last drop.* If such great moths were to become extinct in Madagascar, assuredly the Angræcum would become extinct. On the other hand, as the nectar, at least in the lower part of the nectary, is stored safe from the depredation of other insects, the extinction of the Angræcum would probably be a serious loss to these moths. We can thus understand how the astonishing length of the

[page] 166

nectary had been acquired by successive modifications. As certain moths of Madagascar became larger through natural selection in relation to their general conditions of life, either in the larval or mature state, or as the proboscis alone was lengthened to obtain honey from the Angræcum and other deep tubular flowers, those individual plants of the Angræcum which had the longest nectaries (and the nectary varies much in length in some Orchids), and which, consequently, compelled the moths to insert their proboscides up to the very base, would be best fertilised. These plants would yield most seed, and the seedlings would generally inherit long nectaries; and so it would be in successive generations of the plant and of the moth. Thus it would appear that there has been a race in gaining length between the nectary of the Angræcum and the proboscis of certain moths; but the Angræcum has triumphed, for it flourishes and abounds in the forests of Madagascar, and still troubles each moth to insert its proboscis as deeply as possible in order to drain the last drop of nectar.

The first paragraph is unambiguous in predicting the existence of a moth with the appropriate length of proboscis. That wasn’t at issue, though. The rest, though, can be considered to have a certain fault of vagueness. Certainly, it conforms to a coevolutionary hypothesis, but it appears, infuriatingly, not to exclude a reading that multiple pollinators might be involved, nixing the nifty connection to Occam’s Razor I had already plotted out. I shall have to temper myself to disappointment on that score. But it also means that choosing between a “single pollinator” and “multiple successive pollinators” scenario, as the construction of the “coevolutionary race” and “pollinator shift” models in Whittall and Hodges urges, is not a completely accurate representation of Darwin’s discussion, either. There is nothing in what little Darwin discusses on this score that excludes the essential element of shifting between species of pollinators. Now, it is clear that Darwin was, in this case, concerned only with varieties of moths as pollinators, while Whittall and Hodges expand the discussion to those “pollinator syndromes”, where bees and hummingbirds are also available.

Perhaps, then, Alfred Russel Wallace might be to blame for the bright-line separation of models, as Whittall and Hodges cite him for elaboration of Darwin’s “coevolutionary race” hypothesis. Wallace, indeed, goes on at a good deal greater length, but the case for clear separation of “coevolutionary race” and “pollinator shift” models being an essential part of Wallace’s elaboration takes a major hit along the way. I will bold the relevant bit; it is a small but crucial part of this quote:

“There is a Madagascar Orchis–the Angræcum sesquipedale–with an immensely long and deep nectary. How did such an extraordinary organ come to be developed? Mr. Darwin’s [[p. 475]] explanation is this. The pollen of this flower can only be removed by the proboscis of some very large moths trying to get at the nectar at the bottom of the vessel. The moths with the longest proboscis would do this most effectually; they would be rewarded for their long noses by getting the most nectar; whilst on the other hand, the flowers with the deepest nectaries would be the best fertilized by the largest moths preferring them. Consequently, the deepest nectaried Orchids and the longest nosed moths would each confer on the other a great advantage in the ‘battle of life.’ This would tend to their respective perpetuation and to the constant lengthening of nectar and noses.” The Duke of Argyll then quotes Darwin’s diffident statement “that we can thus partially understand how this astonishing nectary was produced,” and says it is indeed but a partial understanding,–but he does not show what point the explanation given fails to meet. I maintain, on the contrary, that the laws of multiplication, variation, and survival of the fittest, already referred to, would under certain conditions necessarily lead to the production of this extraordinary nectary. Let it be remembered that what we have to account for is only the unusual length of this organ. A nectary is found in many orders of plants and is especially common in the Orchids, but in this one case only is it more than a foot long. How did this arise? We begin with the fact, proved experimentally by Mr. Darwin, that moths do visit Orchids, do thrust their spiral trunks into the nectaries, and do fertilize them by carrying the pollinia of one flower to the stigma of another. He has further explained the exact mechanism by which this is effected, and the Duke of Argyll admits the accuracy of his observations. In our British species, such as Orchis pyramidalis, it is not necessary that there should be any exact adjustment between the length of the nectary and that of the proboscis of the insect, and thus a number of insects of various sizes are found to carry away the pollinia and aid in the fertilization. In the Angræcum sesquipedale, however, it is necessary that the proboscis should be forced down into a particular part of the flower, and this would only be done by a large moth straining to drain the nectar from the bottom of the long tube.1 Now let us start from the time when the nectary was only half its present length or about six inches, and was chiefly fertilized by a species of moth which appeared at the time of the plant’s flowering, and whose proboscis was of the same length. Among the millions of flowers of the Angræcum produced every year some would always be shorter than the average, some longer. The former, owing to the structure of the flower, would not get fertilized, because the moths could get all the nectar without forcing their trunks down to the very base. [[p. 476]] The latter would be well fertilized, and the longest would on the average be the best fertilized of all. By this process alone the average length of the nectary would annually increase, because, the short ones being sterile and the long ones having abundant offspring, exactly the same effect would be produced as if a gardener destroyed the short ones and sowed the seed of the long ones only; and this we know by experience would produce a regular increase of length, since it is this very process which has increased the size and changed the form of our cultivated fruits and flowers.

But this would lead in time to such an increased length of the nectary that many of the moths could only just reach the surface of the nectar, and only the few with exceptionally long trunks be able to suck up a considerable portion.

This would cause many moths to neglect these flowers because they could not get a satisfying supply of nectar, and if these were the only moths in the country the flowers would undoubtedly suffer and the further growth of the nectary be checked by exactly the same process which had led to its increase. But there are an immense variety of moths of various lengths of proboscis, and as the nectary became longer other and larger species would become the fertilizers, and would carry on the process till the largest moths became the sole agents. Now, if not before, the moth would also be affected, for those with the longest probosces would get most food, would be the strongest and most vigorous, would visit and fertilize the greatest number of flowers, and would leave the largest number of descendants. The flowers most completely fertilized by these moths being those which had the longest nectaries, there would in each generation be on the average an increase in the length of the nectaries, and also an average increase in the length of the proboscis of the moths, and this would be a necessary result from the fact that nature ever fluctuates about a mean, or that in every generation there would be flowers with longer and shorter nectaries, and moths with longer and shorter probosces than the average. No doubt there are a hundred causes that might have checked this process before it had reached the point of development at which we find it. If, for instance, the variation in the quantity of nectar had been at any stage greater than the variation in the length of the nectary, then smaller moths could have reached it and have effected the fertilization. Or if the growth of the probosces of the moths had from other causes increased quicker than that of the nectary, or if the increased length of proboscis had been injurious to them in any way, or if the species of moth with the longest proboscis had become much diminished by some enemy or other unfavourable conditions, then in any of these cases the shorter nectaried flowers which would have attracted and could have been fertilized by the smaller kinds of [[p. 477]] moths would have had the advantage. And checks of a similar nature to these no doubt have acted in other parts of the world, and have prevented such an extraordinary development of nectary as has been produced by favourable conditions in Madagascar only and in one single species of Orchid. I may here mention that some of the large Sphinx moths of the tropics have probosces nearly as long as the nectary of Angræcum sesquipedale.2 Now, instead of this beautiful self-acting adjustment, the Duke of Argyll’s theory is, that the Creator of the Universe by a direct act of his Almighty power so disposed the natural forces influencing the growth of this one species of plant as to cause its nectary to increase to this enormous length, and at the same time by an equally special act determined the flow of nourishment in the organization of the moth so as to cause its proboscis to increase in exactly the same proportion, having previously so constructed the Angræcum that it could only be maintained in existence by the agency of this moth. But what proof is given or suggested that this was the mode by which the adjustment took place? None whatever, except a feeling that there is an adjustment of a delicate kind and an inability to see how known causes could have produced such an adjustment. I believe I have shown, however, that such an adjustment is not only possible but inevitable, unless at some point or other we deny the action of those simple laws which we have already admitted to be expressions of existing facts.

So where are we? We’ve learned that in North American columbines, whose longest nectar spur is less than half the length of that found in the A. sesquipedale orchid, a punctuated evolutionary model best fits the phylogenetic evidence, and it is most likely that the long spurs in the group evolved stepwise to accommodate the dimensions of a succession of different pollinators. We see, moreover, that the explanation given by Charles Darwin and later elaborated by A. R. Wallace was specific to the case of A. sesquipedale and its pollinator. Their target was not the modern distinction between phyletic gradualism and punctuated models of evolution, but rather the distinction between evolving and being specially created. As such, we have in Darwin’s case prose that does not exclude the punctuational version of the mechanism, and in Wallace’s case he specifically discusses the evolutionary process using multiple species of pollinators, just as Wasserthal did in 1997 and Whittall and Hodges do in the present paper.

There is an unfortunate tendency among us scientists to inflate the significance of our own work by the expedient of giving past workers short shrift. It would please us to see ourselves as innovators privileged to insights denied great researchers of the past, but it often is the case that we really do stand on the shoulders of giants, whether we choose to ungratefully do so while wearing cleated boots and hopping up and down, or not.

Wesley R. Elsberry

Falconer. Interdisciplinary researcher: biology and computer science. Data scientist in real estate and econometrics. Blogger. Speaker. Photographer. Husband. Christian. Activist.

One thought on “Flowers, Moths, and History of Science

  • 2007/06/27 at 2:41 pm

    I’ve tried to correspond with Whittall about this matter, but haven’t heard back. I’ve sent off a short letter to Nature. I don’t know whether my delay in seeing the article will put it past the time when Nature considers it a timely response. I guess if I don’t see it in the next month there, I’ll post the letter text here.

Comments are closed.