A Buzz in the Meadow (11 page)

I started my PhD in April 1988. My PhD supervisor, Denis Owen, was an odd character. He had spent many years working out in Africa, and would often come to work in colourful East African gowns. He smoked like a chimney in his office, so much so that the room filled with smoke and he became a blurred, dimly visible silhouette when viewed through the glass window of his office door. He would spend his time in there hammering away on an old-fashioned manual typewriter so loudly that the thin office walls would gently shake. Denis barely spoke to me when I started, merely telling me to go and spend a lot of time in the field, looking at butterflies. I was rather intimidated and so didn't push him for any more pointers, although I was in desperate need of them. I hadn't really got a clue what to do. A PhD requires an idea, a hypothesis, an interesting question to pursue. I had none of these, so I went every day to Bernwood and wandered about looking for butterflies.

Bernwood has an interesting history. One thousand years ago it was a vast royal hunting forest, covering many hundreds of square kilometres, and it is said to have been a favourite hunting ground of Edward the Confessor. Over time it was gradually cleared and converted to arable land and pasture, as successive monarchs sold off chunks of the forest to raise cash. By the early twentieth century Bernwood had shrunk to occupy just 300 hectares, but despite its diminished size it had become well known as a hunting ground for butterfly-collectors. It supported many rare woodland species, including the spectacular purple emperor and the black hairstreak. Sadly, it was purchased by the Forestry Commission in the 1950s, and at the time this organisation had little interest in butterflies or in conservation. Virtually the entire wood was clear-felled and sprayed with herbicides from the air to kill off any regeneration. Dense stands of conifers were planted, and then sprayed with DDT to kill beetle pests that attacked the conifers. Remarkably, the forest remained a hotspot for rare butterflies through the 1960s and 1970s. Despite the herbicide sprays, woodland flowers and shrubs somehow survived, and so did most of the butterflies. When I first visited Bernwood in the mid-1980s, it still had many rare butterflies; it remains the only place in the UK where I have seen wood whites, brown hairstreaks and purple emperors. It was thought to have more breeding butterfly species than any other site in Britain. Sadly, though, most were in decline. The conifers were growing larger and larger, casting a dense shadow and thus shading out all of the native flora that provide butterflies with nectar and food plants for their caterpillars. My undergraduate project was on the effects of increasing shade on woodland butterflies, and its conclusion was that all but the most shade-tolerant butterflies, such as the speckled wood butterfly, would soon disappear.

Luckily, the Forestry Commission has evolved over time. Where once it was concerned solely with efficient timber production and with generating maximum profit, now it is far more interested in conserving biodiversity and managing forests in a way that balances commercial needs with those of wildlife. Changes were evident in the late 1980s at Bernwood, where the Forestry Commission began widening rides and cutting small clearings to combat the gloom cast by the towering spruce-tree plantations. It was pretty small-scale stuff, but it was a start.

I spent the spring and summer of 1988 walking these rides, looking for butterflies, and trying to come up with an idea as to what I should study for my PhD. Having a strong interest in conservation, I decided to study the ecology of a rare species, the Duke of Burgundy fritillary. It isn't actually a fritillary, and it has no known connection to the Duke of Burgundy, so its name is hard to fathom, but it is a lovely little creature with a chequered orange-and-brown upperside and silver spots on the underside of its wings. The butterfly is on the wing in May, while for much of the rest of the year the caterpillars can be found on the leaves of their food plants, cowslips and primroses. At least in theory they can. In practice I walked through Bernwood every day in May when it wasn't raining and saw one Duke of Burgundy, a female that promptly flew away. I searched diligently for eggs and caterpillars over the following months, but found none, although I did map out the locations of every primrose in Bernwood. By September I had exactly one data point, and there really wasn't a great deal I could say based on that.

I spent the autumn and winter reading everything I could on butterflies, in search of inspiration. It felt as if time was slipping away. My PhD funding was for three years and there are, self-evidently, just three field seasons in three years; I had squandered one already. If my second field season was similarly unsuccessful I would be doomed to fail my PhD. It dawned on me, not before time, that studying rare creatures is rather difficult. For safety, I decided instead to study something very common, something that I could not fail to find by the thousand. And this is where we return to the meadow brown. It isn't a woodland butterfly, but there were quite a few in the woodland rides and I had seen countless numbers of them in an adjacent meadow, known appropriately enough as Bernwood Meadows. But what exactly should I study? I came across Ford's work on meadow browns and became intrigued. I knew of Ford, for he was author of the New Naturalist book
Butterflies
, which I had been given as a child. I'd also met him briefly when I was an undergraduate student at Oxford, at which time he had been a very old but nonetheless rather intimidating figure.

Ford worked closely with Wilfrid Hogarth ‘Bunny' Dowdeswell, a former Oxford undergraduate who became an academic at Winchester College, and together they counted the spots on the hindwings of meadow brown butterflies for twenty-five years, from the late 1940s to the 1970s. They obtained samples from all over Europe, and particularly from the Scilly Isles, which they visited repeatedly over many summers, camping together on the uninhabited islands.
1
In most populations in Britain and Europe, the majority of male meadow browns have two spots, while most females have none. In some peripheral populations, such as some of the Isles of Scilly, females with one or two spots were very common. On some of the smaller of these islands the frequency of spotted and unspotted butterflies varied sharply from year to year, far more than could be explained by chance (or genetic drift, as Wright termed it). This, Ford declared once again, was clear evidence for natural selection, and proof that drift was inconsequential. What is more, Ford described a mysterious boundary phenomenon along a line roughly following the border between Devon and Cornwall, where the frequency of spotted and unspotted butterflies changed markedly from one side of a hedge to the other. Given that meadow brown butterflies can readily fly over hedges, this seemed odd, but was again attributed to some powerful selective force at play.

What interested me was whether these spots actually served any function that could explain the selective forces that seemed to change the frequency of more or less spotty individuals from place to place and year to year. Some butterflies, such as the peacock, have very prominent eye-spots near the tips of their forewings, which are thought to scare away potential predators by mimicking the eyes of some much larger beast. The eye-spots of a peacock are hidden when it sits with its wings closed, but if it is disturbed, it has only to open its wings to reveal huge eyes comprising concentric circles of blue, red and yellow. Owl butterflies – enormous creatures that live in the rainforests of South America – have eye-spots and associated wing patterns that convincingly resemble the face of an owl. Meadow browns also have far less impressive eye-spots near the tips of the forewing, on both the upper and lower sides, formed from a large black spot with a small white spot in the centre, resembling the glint of light reflected from the black pupil of an eye. These spots might serve to frighten predators, but this wasn't especially interesting to either Ford or me, as all meadow browns were the same in this respect.

Could the small spots on the underside of the hindwing have a similar purpose, relating to predation? Many hairstreak butterflies have an elaborate false eye near the edge of their hindwing, often with slender tails that resemble antennae, creating the impression that the back of the insect is in fact the front. To further enhance this impression, hairsteaks often perform a swift 180-degree turn as soon as they land on a perch or flower. The idea is that this may deflect the peck of a predatory bird from the real head of the insect to the margin of the wing, which readily snaps off, allowing the butterfly to escape. It is common to find butterflies with peck-marks nipped out of the edges of their wings.

In the early 1980s Paul Brakefield, an English scientist based in Leiden, proposed an explanation for the variation in spot numbers in meadow browns and their relatives. He suggested that the spots did serve to deflect attacks, but that the trade-off was that they might make the butterfly more obvious to predators when sitting still. When flying about, visiting flowers and looking for mates, butterflies are very obvious to predators such as birds, so deflective spots might be helpful. In contrast, if the spots make them less well camouflaged when stationary, they would be best not to show them when sitting still. Female butterflies spend much more time sitting still than males, who fly around a lot, which might explain why females tended to have no spots whilst the majority of males had two. Brakefield went on to predict that in populations inhabiting cold, wet climates, where the butterflies spend a lot of time sitting around waiting for the sun to come out, spots ought to be a disadvantage and hence rare, whilst in sunnier places the spots ought to be more common. The latter certainly appears to be true in the large heath butterfly, a relative of the meadow brown in which spottiness decreases in populations that are further north or west or at higher altitudes.

This theory was quite neat, but no one had ever shown that these spots actually did deflect the attacks of predators. I decided to try to find out. I also became interested in the genitalia of meadow browns. The genitalia of male butterflies consist of a pair of rather barbaric claspers, which lock on to the female during mating, and between which the tube-like phallus protrudes. Place your wrists together with the palms of your hands facing one another and you get a rough idea of the design of the claspers. They are hinged at the base, so that they can open out and then close on the female like grappling irons. In some butterflies the claspers are armed with sharp, incurved, talon-like hooks to help them grip. Pity the female speckled wood butterfly. In the meadow brown there are no sharp hooks, but there is a bristly ‘thumb' protruding from their upper edge. The genitalia of male butterflies, and of male insects in general, tend to be very useful in identifying species that are otherwise similar. They usually differ in obvious ways between species, whilst being fairly uniform within species. One theory that had been put forward to explain this was that the male and female genitalia are like a key and lock; they only fit together if the male and female are of the same species, helping to prevent unfruitful mating between members of different species. The exception is the meadow brown, in which the male claspers are very variable. In some the ‘thumb' is no more than a gentle bump, while in others it is very pronounced with a swollen end capped in bristles, resembling a tiny pollarded willow tree. So far as I could tell, no one had ever looked at the fit between female and male insect genitalia before, or explained why meadow browns have such variable genitalia, and this seemed worth exploring.

Questioning the function of tiny spots on butterfly wings, or explaining why their genitalia come in various shapes and sizes, might seem fantastically obscure subjects to pursue. Answering these questions was never likely to change the world and, with the benefit of hindsight and experience, I can think of many more profound topics that I might have tackled, but in the spring of 1989 this was all I had.

To get to grips with the function of wing-spots, I counted their number in meadow brown populations all over southern England. I had a lovely blue 650cc Suzuki motorbike on which I roared about the countryside, with a net, notebook and countless small cardboard pots in my rucksack. I wanted to test whether butterflies had fewer spots in cooler, shadier sites and more on sunny, south-facing sites, as Brakefield's theory would predict, but disappointingly this didn't seem to be true. I went down to Cornwall to look for Ford's boundary, where the proportion of spotted butterflies was said to change suddenly, but it was no longer there, so far as I could tell. I didn't seem to be getting very far, so I tried more experimental approaches.

I killed a range of spotted and unspotted butterflies, dried them with their wings open or closed, and then tied them to grass seedheads in Bernwood Meadows and waited to see whether they were eaten by birds. Butterflies with their wings open were more likely to be scoffed, but whether or not they had spots on their hindwings appeared not to make the slightest difference.

I tried a different tack. I caught 600 butterflies from Bernwood Meadows and changed their spots artificially. It is quite easy to remove these small spots by simply brushing away the dark wing scales with the tip of a damp paintbrush, and equally it is easy to add small black spots with a marker pen. So I randomly assigned individuals to either spotted or unspotted groups, made the necessary changes, gave them an additional discreet mark on the topside of their wings, so that I could subsequently identify them, and released them back where I had found them. Three days later I went back and hunted for my marked butterflies, to see which ones were still alive. In theory, if females spend most of their time sitting still, spots should make them more conspicuous and hence more likely to be eaten, while males that spend most of their time flying around should benefit from having spots to deflect attacks. So I predicted that females with added spots and males with spots removed should be predated more highly that unspotted females and spotty males. Of my 600 butterflies I managed to recover ninety-one – it was hard going because there are countless thousands of meadow browns in that meadow, so finding the ones I had marked was like looking for a needle in a haystack. To my slight surprise, the results followed the prediction. In particular, adding a spot to females did seem to make them more likely to be eaten.