Darwin's Backyard

by James T. Costa

  Darwin’s concluding exclamation mark is an indication of the importance that he ascribed to this unexpected conclusion. In pitching his argument in the Origin for the reality of evolution by natural selection he needed his readers to appreciate how little we understand about interconnectedness, how individuals and populations and species each have an impact and influence on others in myriad ways we can only dimly envision. This simple food chain is but one tiny thread of connectedness in an extraordinarily complex tapestry of nature, a small exemplar that represents an infinitude of threads, most unseen.

  The clover experiments were conducted in mid-August of 1859 in Great House Meadow just behind Down House, maybe as a pleasant diversion as he was working away correcting the proofs for the Origin. Bordered by the kitchen garden and the sandwalk, Great House Meadow is a classic English meadow that has probably never been violated by a plow, teeming today as in Darwin’s day with grasses and forbs—including lots of red clover, its low patches revealing a scarlet gleam here and there in the grass. In season the meadow swarms with bees, their Doppler hums and buzzes noting their comings and goings. It was another enclosure experiment, as he reported in more detail later in his 1876 Effects of Crossing book: “One hundred flower-heads on plants protected by a net did not produce a single seed, whilst 100 heads on plants growing outside, which were visited by bees, yielded 68 grains weight of seeds; and as eighty seeds weighed two grains, the 100 heads must have yielded 2720 seeds”23—a potent demonstration of the effects of bee visitation.

  As informative as the experiment was for Darwin, by modern standards there are, of course, a number of problems with his design. For one thing the heavy gauze used for his enclosures would have done more than exclude bees; it would have shaded the plants too. Starved of adequate light, perhaps these plants just didn’t have energy enough to set seed even by selfing. Another problem is the single covered plot; this was a one-off experiment, informative enough, but today both enclosed treatments and uncovered control plots (both of standard dimensions) would be replicated, and their placement in the clover field might be randomized for added statistical rigor. Such details of experimental design were not appreciated in Darwin’s day, and it’s not fair to hold him to modern standards. But an even bigger potential problem threatened to break his chain. Darwin maintained that only bumblebees visited red clover since the probosces of honeybees are too short to reach the nectar. If that was not the case, however, then the final link in his causal chain leading from cats to field mice to bumblebees to red clover was in jeopardy. Soon after the Origin came out an old acquaintance wrote Darwin to report that he observed bees of different kinds, including honeybees, visiting red clover, but he qualified this by noting that he observed them on mown clover, which reputedly resprouts with smaller blossoms that are perhaps more accessible to shorter-tongued bees like honeybees.

  Darwin started looking out for cases of honeybees visiting red clover. While he, Emma, and the younger kids visited William at Southampton in the summer of 1862, he naturally made time for fieldwork. He scoped out a field of red clover with lots of honeybees, and found that some dipped into the clover flowers the usual way while others sucked the nectar through holes left by nectar robbers. A thought suddenly struck him: maybe some of these bees had long probosces while others had short probosces; the long ones could reach the nectar from the mouth of the flower, but the short ones had to resort to nectar robbery instead. This could be a form of division of labor—potentially a very interesting discovery. He collected a number of bees of both “types” (at the flower mouth versus nectar robbing) before the family departed Southampton for Bournemouth, where they planned to continue their holiday. He couldn’t find any clover fields there to continue his observations, however, much to his frustration. He was so sure he was onto something that he excitedly wrote to Lubbock, a devotee of ants, bees, and wasps, to observe bees on clover and collect specimens for him: “for Heaven sake catch me some of each & put in spirits & keep them separate—I am almost certain that they belong to two castes, with long & short probosces. This is so curious a point that it seems worth making out.”24 His letter was premature, alas: when he finally got around to examining the bees collected in Southampton he realized there was no difference in proboscis length after all. He hastily fired off another letter to catch the evening post: “I beg a million pardons,” he wrote Lubbock. “I do so hope that you have not wasted any time for my stupid blunder.— I hate myself I hate clover & I hate Bees.”25

  Darwin made mistakes too, and that’s the nature of research. Long- and short-tongued honeybee castes don’t exist, but it was a neat idea. As for the bumblebees, although Darwin was basically correct that these bees play a leading role in the pollination biology of red clover, he seemed to assume that they all had long probosces, whereas some four species are known to occur at Down House, two longs and two shorts. But how is it that sometimes honeybees visit this plant and at other times they ignore it completely? The discrepancy probably lies in flower constancy: the phenomenon whereby honeybees tend to focus their efforts almost exclusively on the most profitable flowers available at a given time. They communicate the location and type of flower of interest through their waggle-dance, calling in reinforcements to work with them. Bees finding a good resource patch will dance to recruit their hive-mates to join them there, but the intensity of the dance is related to the quality of the nectar and pollen available. This sets up a dynamic of competing recruiters, where the bees working the best patches recruit bees away from the others, and soon all the foragers of a colony are working on what they perceive as the best available flower patch. So, sometimes red clover is the best thing going and the honeybees work it like there’s no tomorrow (including taking advantage of the holes left by their nectar-robbing bumblebee cousins), and other times there’s a better deal out there and they won’t give red clover even a passing buzz. Such are the allegiances of honeybees, a bit like a circle of avid bargain hunters who periodically compare notes on the prices they’ve found; before long they have all flocked to where the best deals are to be had and snub stores with yesterday’s bargains. Honeybees, too, shop around. Darwin would have deeply appreciated the marvelous honeybee waggle-dance and its competition dynamic—a kind of selection process, after all—but this astonishing phenomenon was discovered in the twentieth century.

  Insofar as bumblebees are not the sole pollinators of red clover, Darwin’s food chain is not, in this instance, quite the linked series of interdependency that he thought. But nature is more complex than we sometimes think. In 2007–2008 the intrepid trio Norman Carreck, Toby Beasley, and Randal Keynes repeated Darwin’s bee enclosure experiment at Great House Meadow and found bumblebees visiting the clover, just as Darwin did. When it came to setting seed, the unenclosed red clovers of Great House Meadow didn’t fare much better than those in the enclosures. Plants have their good years and their bad years, owing to the vagaries of wind, weather, pollinators, and other exigencies; some years there’s a good seed set, other years are a bust, and the reasons why are not always clear. Although Darwin did not replicate his enclosure experiments across seasons or years, he likely would have viewed season-to-season variation as mere noise. Time is of the essence: it’s the long haul that matters, he would have argued, sure that insect intermediaries are key. In his Natural Selection manuscript Darwin related how Kölreuter was astonished, upon realizing that certain flowers could only be pollinated by insects, that something as important as fertilization would be left to the whims of such creatures. Then Kölreuter thought the better of it, and concluded that with insects the Creator in his wisdom had actually used the surest means possible for transferring pollen. Darwin concurred: “hardly any means, I am convinced, could be surer” than ubiquitous, abundant, ever-active insects.

  Modern evolutionary biologists would largely agree, in view of the fact that the dramatic evolutionary diversification of the flowering plants in the Cretaceous period is largely coincident with the ra
pid diversification of bees and other major insect pollinator groups. Striking specialist insect-floral relationships such as those found in orchids, yuccas, and figs practically scream coevolution, where two organisms become mutually adapted to one another. But the relationship in most cases is more diffuse, with several different kinds of pollinators capable of pollinating a given plant species and several plant species capable of providing nectar and pollen to a given species of pollinator. Even so, features seemingly tailored to entice and (usually) reward insect visitors abound. The most obvious are the flowers themselves, the ones with showy petals, sepals, or bracts, often with nectar guides, those stripes and streaks pointing the way to the reward center like runway lights and painted lines guiding pilots. Tellingly, nectar guides are intended for insects’ eyes, being especially prominent (and in some cases only visible) at ultraviolet wavelengths, which insects can perceive while vertebrates like us cannot. The insightful Sprengel saw the stripes and other patterns on petals as advertisements for insects. Darwin missed this initially, but later he acknowledged that there may be something to it based on a little experiment. “There can be little doubt,” he wrote in Natural Selection, “that C. C. Sprengel has pushed his views to quite a fanciful degree; as for instance when he accounts for all the streaks of colour on the petals, as serving to guide insects to the nectary. Nevertheless some facts could be given in favour of such a view: Thus in a patch of the little blue Lobelia, which was incessantly visited by Hive Bees, I found that the flowers from which the corolla, or the lower streaked petal alone had been cut off, were no longer visited.”26 He wasn’t sure if removing those streaked petals made the flowers seem past-peak to the bees and so not worth visiting, or simply deprived them of a convenient landing place, but it was possible that the flowers no longer had a signpost advertising the nectaries.

  Legitimate and Illegitimate Marriages

  Another of Darwin’s adventures in pollination is the subject of perhaps the most important of the “multiplicity of experiments” that he began in the early 1860s: flower polymorphism, or the occurrence of two or more distinct flower forms in the same plant species. That may not sound momentous, but in fact Darwin succeeded in shedding light on a phenomenon that had puzzled botanists for centuries.

  It all started in the spring of 1860, in the months following the publication of the Origin in November 1859. It was a tumultuous period for Darwin, at least internally. November found him seeking hydropathic treatment in Yorkshire while he waited for the Origin to come out. If his illness was brought on by stress he surely had plenty, judging by his rashes and churning stomach. He thought the Origin would bring down a firestorm of rage and condemnation on his head. That never materialized, but it did touch off controversy as naturalists and clergy weighed in for or against his theory. He was still in Yorkshire when one of his old Cambridge mentors, Rev. Adam Sedgwick, wrote a scathing, if heartfelt, letter filled with pity, outrage, and disappointment. Henslow on the other hand said that Darwin should be given a fair hearing, and other clerics, like Charles Kingsley (author of The Water Babies), wrote to register support and declared that Darwin’s ideas were emphatically not at odds with religion. Darwin took heart; already hard at work on corrections for a second edition, he sent thanks to Kingsley and asked if he could quote him in the new edition. Kingsley was happy to oblige.

  Darwin headed home on December 7, 1859. Emma, the girls, and the youngest of the boys welcomed him—21-year-old Willy was at Cambridge by then, occupying his father’s old rooms at Christ’s College, and 15-year-old Georgy and 12-year old Franky were at Clapham School. (In keeping with the chauvinism of the times, Etty, then 16, and Bessy, just 12, were tutored at home and did not receive the depth or breadth of education that their brothers enjoyed.) He continued to field letter after letter over the newly published Origin and labor over corrections; the second edition came out on January 9, 1860, while Darwin continued to edit material for the first American edition.

  That spring found him investigating how stamens and pistils bend, picking up on his observations from the 1840s. By the end of April 1860 he had sent a lengthy list of flowers to Hooker for review. He wondered if he was correct in thinking that, as a rule, pistils tend to be bent in the direction of the nectaries. “Why I care about it,” he wrote, “is that it shows that visits of insects are so important, that these visits have led to changed structure.”27 But just a few days later he was interrupted in this project when he received a letter from Henry Doubleday, an authority on primroses, the sunshine-yellow tubular flowers of woodland, glade, and garden. There was an ongoing debate at the time over whether the common primrose (Primula) and the similar cowslip and oxslip represented three distinct species or if one of these was the product of hybridization of the other two. In the Origin Darwin cited the group as examples of strongly marked varieties ranked by some naturalists as distinct species—those “doubtful forms” that so blurred the line between species and variety that even botanists disagreed on their status. In fact the genus Primula is complex, with as many as 500 species recognized by some botanists. Many hybridize, and myriad cultivars have been developed. Today the three that were of interest to Darwin are recognized as separate species: common primrose (Primula vulgaris), cowslip (P. veris), and the rare oxslip (P. elatior).

  Evidently Doubleday’s letter prompted Darwin to take another look at his primroses, but this time he noticed something that he was only dimly aware of before: there appeared to be two distinct flower morphs. Some plants had long stamens and a short pistil while others had short stamens and a long pistil. On May 7th he dashed off an excited letter to Hooker:

  I have this morning been looking at my experimental Cowslips & I find some plants have all flowers with long stamens & short pistils which I will call “male plants’’—others with short stamens & long pistils, which I will call “female plants’’. This I have somewhere seen noticed, I think by Henslow; but I find (after looking at only two sets of plants) that the stigma of male & female is of slightly different shape & certainly different degree of roughness, & what has astonished me the pollen of the so-called female plant, though very abundant, is more transparent & each granule is exactly only ⅔ of size of pollen of the so-called male plant.— Has this been observed??28

  It dawned on him that Henslow had pointed this out years and years ago, but he had somehow forgotten about it. In fact, his own children knew about the phenomenon—long-pistiled flowers are called “pin-eyed” as the stigma is round like the head of a pin, poking just outside the corolla tube, while the long-stamened “thrum-eyed” flowers had elongated anthers somehow reminiscent of those bits of weaving-loom yarn waste. Probably for millennia kids had known that the pin-eyed flower morphs were best for stringing the blossoms into flower chains by threading the corollas over the long pistils.

  These flowers made him sit up straight; he dissected several and peered at their pollen under the microscope. “It may turn out all blunder,” he said to Hooker, but he planned on tracking the seed production of each morph. He noticed that the female flowers seemed to produce smaller pollen grains than the male flowers, and it struck him that he may have stumbled upon plants caught in the act, as it were, of evolving into separate sexes. “It would be fine case of gradation between an hermaphrodite & unisexual condition,” he enthused in his letter to Hooker. He wrote to Henslow too, hopeful his old mentor could give him other cases to investigate.

  This phenomenon, called heterostyly, is found in many plant groups but it was a stroke of luck that it occurred in so widely cultivated a group as primroses. Darwin sent the kids (except Etty, who had just come down with typhus) on a mission to collect cowslips. He opened his experiment book entry of May 13th with the results: “My children gathered great bunch of Cowslips 79 stalks were male flowers, & 52 were female flowers.”29 By the next day they had added many more, for a grand total of 281 males and 241 females. His working hypothesis was that these plants were transitioning to dioecy, where male and fem
ale flowers are borne on different individual plants of the species. Long pistils and small pollen grains suggested this morph was becoming more female, while short pistils and larger pollen signaled this one was becoming more male. It was a neat and tidy hypothesis, and an exciting one too since, if correct, these primroses would provide a textbook example of an evolutionary dynamic. Unfortunately, however, the primroses ended up presenting a different kind of textbook case: exemplars of that “great tragedy of Science,” as Huxley so memorably quipped in his 1870 presidential address to the British Association for the Advancement of Science, namely “the slaying of a beautiful hypothesis by an ugly fact.” Darwin expected that the “male” form would exhibit lower seed production, but in fact he found the opposite.

  Undaunted, Darwin soon appreciated what was actually going on: the dimorphism of flowers promoted intercrossing between the morphs—it was a mechanism to minimize selfing and promote crossbreeding. He explored this systematically over the course of the next year, performing crosses and making observations . . . at least, as systematically as he could, given his simultaneous infatuation with orchids, insect pollination mechanisms, tendrils, and sundews! He noted that both morphs produce nectar and are visited by bees and other insects. However, they are not visited by the same insects, and in any case pollen of the two morphs is deposited in different locations on the probosces of those visiting: around the base for insects coming to the long-stamened form, and nearer the tip for visitors coming to the short-stamened form. Thus, at subsequent floral stops any basal, long-stamened pollen is most likely to be left atop long-styled pistils, and apical short-stamened pollen is most likely to be deposited on short-styled pistils. Did this matter? He hand-pollinated in every pairwise arrangement—long stamen/long style, short stamen/short style, long stamen/short style, and short stamen/long style—and then counted up the number of resulting seed capsules, and weighed the mass of tiny seeds they yielded. The results were clear: nearly twice the weight of seed was produced in long-long and short-short fertilizations than in the long-short and short-long cross-bred fertilizations.