Rudolf Harmsen 

Caterpillars and Coyotes

Thoughts on ecological stability

Yesterday I put up our new tent on the east lawn, and admired its high-tech, ultra low-weight, ingenious design. In the evening, we went out for curry with Tom Mawhinney to one of Kingston’s East Indian restaurants. Then, back at home, we sat in the hot, humid evening air, talking and listening to the last birdsong of the day, and to the crickets starting to announce the coming of night. The moon, only a day or two from full, lit up the garden and surrounding fields, creating a strange yellow glow emanating from the tall dead grass beyond the garden fence. I loved the contrast between our perfectly still, moonlit surroundings and the brief points of bioluminescence, coming from different, unexpected directions, as hidden in the darkness between their mating advertisements, a chorus line of fireflies entertained us with their ‘fire’. As bedtime approached, I had a deliciously cooling swim, and crawled into the tent, expecting to fall asleep immediately. But before I dozed off, I became aware of little noises, which alerted my curiosity. There was, for instance a constant gentle tapping on the tent. As I focused on this sound, I realized that very small items of two distinctly different sizes were dropping at brief, but random intervals onto the flysheet above me. It took me a while to see the larger picture, the lawn, the tent, and the silver maple with its branches above the tent. Then it struck me: of course, in those branches was an entire population of caterpillars of two predominant sizes, eating foliage, and dropping their frass. Then I heard equally minute noises, but somehow, recognizable as coming from far away: a barking dog, a late car on the highway, a whistle of a bird, a jet at 30,000 feet on its way to Europe. I started to think about our upcoming Siberia trip, and must have fallen asleep, because I woke to the sound of wind in the tree above me, and as I was dozing off again, I perceived, through my closed eyelids a sudden flash of light. Some twenty seconds later a gentle, but foreboding rumble confirmed my suspicion.

Half an hour later, after the lightning flashes and thunder had come closer and closer, the storm broke over my little tent, and I lay there, wide awake, listening, and expecting our beautiful, expensive tent to be as leaky as any other tent we have ever possessed. But after the storm rumbled eastwards, and the last drops from the tree pattered down onto the tent, I was as dry as I had been before. When moonlight returned to the garden, I curled up in my lair and looked forward to a few more hours of sleep. But once again, it was a sound that caught my ear, and brought me back to full wakefulness. Out of the abandoned orchard to the east of us came first a few yaps, and then an entire chorus of howling coyotes. Lying there, stark naked, inside our lightweight tent made me feel somewhat vulnerable, especially when after a brief period of silence, the canine gang struck up again much closer than the first time. Were they aware that I was trespassing into what they must consider their domain during hours of darkness? Or was it just me, suffering from a deeply ingrained Euro-cultural fear of the wolf? When I woke again, it was raining, and the early morning light stimulated me to start the day, ending my first night in the tent. I will try again tomorrow, and Jeri will hopefully join me, so that by the time we go to Siberia, we will be used to getting a good sleep under camping conditions.

When I live through an experience such as last night’s, it is the immediate, personal and spiritual level that dominates my feelings and short term response, but later, my thoughts turn to the more worldly aspects of the events that I witnessed. Today, I have been thinking about caterpillars and coyotes, not in terms of noises in the night, but in terms of ecological equilibrium theory; ‘the balance of nature’ if you will. A few years ago, I visited Mark Woolhouse in Oxford. Mark is an ex-student of mine, who at that stage was an up-and-coming young scientist rapidly making a name for himself as a critical and original thinker in the fields of ecology and epidemiology. I had not seen him for a couple of years, and made a point of stopping over in the UK for a day on a flight from Amsterdam to Toronto. What I really like about Mark is that he can make things happen, and so he did on that April day. I had arrived in Oxford the night before, and Mark and I had dined out in a very pleasant Chinese restaurant. We tried to catch up on our respective lives, and bounced new ideas off one another over a large pot of green tea until the place closed, and then walked back to Mark’s flat. The late Oxford streets brought back all sorts of memories – our footsteps echoing off the old stone walls, the damp air, the cobble stones, gnarled chestnut trees just bursting into leaf – England.

The next day, Mark suggested that we pay homage to Gilbert White, referring to him as the 18th Century ‘father of ecology’. White’s 1789 book ‘The Natural history of Selborne‘ is the first recorded example of someone asking questions about quantitative and interactive ecological phenomena. Needless to say, this man was not a scientist by our standards, he knew not our terminology, nor would he have known anything about ecological theory, statistics, or experimental design. He was a churchman, a curate; he had read divinity and natural history (a popular combination in those pre-Darwinian days) at Oxford. After his university days, he spent the rest of his life tending the flock at Selborne, and writing a voluminous diary, mostly in the form of letters to other naturalists. A selection of these letters was published as ‘The Natural History’. What was so remarkable about this pastor was that he not just recorded the natural history of his parish, but he thought about why things were as they were. He searched for explanations for such phenomena as the relative stability of the number of swifts that nested every year in the church, or why the sex ratio of flocks of chaffinches in winter differed greatly between different parts of the country. These are questions for which only now in the early 21st Century we are starting to find satisfactory answers. I have owned an illustrated 19th Century copy of ‘The Natural History’ since 1963, but had never been to Selborne; I immediately accepted Mark’s suggestion as to how to spend this precious day with my friend.

Two of Mark’s research students joined us for the day’s journey; Francesca and Kristina were exceptionally bright and observant young women, who embraced the challenge of the day with enthusiasm. We navigated a typical maze of winding country lanes, which eventually took us to Selborne. It was one of those English spring days that make you feel as if the sunlight tickles the very bottom of your heart. Gorgeous clouds with silver linings threw the occasional short-lived but chilling shower our way, followed by bursts of sunlight glittering off the wet landscape, yet, a gentle green wash softened all the trees and shrubs, and there were flowers everywhere: bluebells, daffodils, cowslips, violets, primroses, daisies. Sheep and lambs of-the-year added contrast to the greener than green pastures, and the wood-lots were alive with raucous rooks quibbling over the best nesting sites in their rookeries. Once in Selborne, we ‘did’ the museum, admired the church, visited White’s grave in the Churchyard, and then went for a wonderful hike up and into the beech hanger, from where we enjoyed a great view of the village and its surrounds. Describing the general lay-of-the-land in the first section of ‘The Natural History’, White writes: “The high part of the south-west consists of a vast hill of chalk, rising three hundred feet above the village, and is divided into a sheep-down, the high wood and a long hanging wood, called The Hanger. The covert of this eminence is altogether beech, the most lovely of all forest trees…”. Walking among the centuries old beech trees, and coming upon some of the birds that White wrote about in considerable detail, such as robins, blackbirds and a tawny owl, and realizing that these birds were in all probability descendants of the ones White had seen, described and sketched, gave me a wonderful feeling of connection with that 19th Century naturalist and his world. A less heartwarming realization was that several animals described by White are no longer found in that part of the world. The high wood and the hanger are still there, but much of the surrounding woodland and hedgerows have been cleared, the fields treated with pesticides, and migration routes and wintering grounds for the migratory birds are much diminished.

We ended our afternoon with one of those delicious English ‘cream teas’. We sat under a bower of sprouting grapevines, on a sun-warmed patio, looking over a flower filled garden in which a blackbird sang its melodious love song. We talked about the world of a curate with a passion for nature, who lived in a small English town in the Age of the Enlightenment, and compared it with our own technologically driven lives. Somehow, during the progress of the past two centuries, for better or for worse, the values of Gilbert White’s era got lost.

It is still a scientific mystery that in many ecosystems, (even some anthropogenically modified or disturbed ones), a robust stability in the relative abundances of the species of plants and animals tends to persist over long periods of time. This gives the observer the impression of a balance of nature, whereas at the scale of the individual organism birth and death are common phenomena; on the scale of the ecosystem or population little ever changes. Writing about swifts and swallows, Gilbert White noticed1 not only that the numbers of breeding birds was remarkably constant from year to year, but also that the number fledglings in one year did not seem to affect the number returning birds the next year. Ecologists have created or appropriated terms such as stability, resilience, resistance and ecological homeostasis. I will not delve into the theoretical fine points of this fascinating branch of science, as it depends on a lot of complex and controversial mathematical models. But, this aspect of quantitative ecology is critical not only for our understanding of how nature functions, but it also underlies the disciplines of wildlife management, conservation biology, fisheries management, and even agriculture and forestry. Hence, I will dig into it a bit, and give you a feeling for what processes are considered to be the driving forces in maintaining diverse and stable ecosystems.

At the Queen’s University Biology Station are some very fine stands of old-growth sugar maple forest, one of which I have adopted as a sacred grove to heal my spirit when I feel the need for solitude. I specifically go there every now and then to walk and feel the age and richness of that ecosystem. The old, but mostly healthy trees form a dense green canopy, which seems to protect the lower layers of the forest and everything that lives there. Tens of thousands of different organisms from soil bacteria to maples and moose have co-evolved into a community that remains remarkably stable over many centuries. My favorite time of year in the maple forest is early spring, with the snow gone, and the spring flowers starting to coming out, hopefully just ahead of the blackflies. I walk around looking at patches of trout lilies, hepaticas or even some of the early flowering sedges, but after a while, I sit down quietly and let nature come to me. I especially remember one day when I sat leaning against an old gnarled maple listening to two chickadees trying to outdo one another with their simple but insistent two-note song, quarreling over the boundary between their territories. I was about to confuse them by joining in with my best imitative version of their song, when I heard a ruffling sound of disturbing leaf litter. A deer was approaching me very slowly; I sat as still as I could, and admired the delicate manner by which the deer picked the place for her footfalls. It suddenly dawned upon me that there was a smaller animal closely following the doe, which turned out to be a very young fawn, which in a most endearing manner followed its mother with clumsy, wobbly steps and nervous, ineffective jumping moves. The pair walked within twenty feet of me, so close in fact that I smelled the distinct odor of the doe, but because I was completely motionless, and downwind from her, she never knew that I was there.

The fawn I saw that day looked terribly vulnerable, and the doe could not possibly have saved it from the attack by a coyote or even a fox. Yet, deer are a common species in the woodlands of eastern Ontario. In all stable ecosystems ‘standoff’ relationships evolve between prey and predators as well as between competitors. The first theoretical models that tried to explain these stable relationships were published over a century ago, and they have grown tremendously in complexity and diversity of approaches, but in essence they all depend on interacting positive and negative feedbacks. The growth of a deer population represents a typical positive feedback, in that the number of fawns born in a year depends on the number of does, but in turn, the birth of fawns increases the number of does for the next year. Generation after generation the population grows, faster and faster. Until, of course something will limit it. That can be a growing shortage of fodder; as the deer population gets larger, it will eat more of the available forage, and this causes the deer to be less well fed, resulting in fewer fawns being born, more deer dying of disease, etc, etc. That is an example of negative feedback, i.e. the growth of the population results in an increase in a process (consumption of fodder) which in turn results in a food shortage, which stops or even reverses the population growth. This is a fairly simple concept, and makes us also understand how prey/predator relationships can be self-stabilizing. As prey get killed, they become rare, the predators starve, the prey can hide better, and a stable relationship is established. Just imagine, thousands of such self-stabilizing relationships interacting with one another produce a stable ecosystem like the maple forest.

When Jeri and I bought the derelict farm we now live on, we became the custodians of forty hectares of mostly heavily overgrazed land with shallow, stony soils on fractured limestone bedrock. Scattered trees and shrubbery protruded above the sparse grass and weed pasture. People had long ago stopped trying to make a living off the land, but keeping a herd of miserable, undernourished beef cattle put some steaks in the freezer and some dollars in the bank. As soon as we took possession of the land, we apologized to the poor people who had cleared the original forest a century and a half ago, and planted thousands of tree seedlings. In the early years, when the little trees started to emerge above the now tall, un-grazed grass, weeds and shrubbery, we ran into some interesting problems. Every fourth of fifth winter, we had a very high population of cottontail rabbits, which did a lot of damage to our trees by chewing on the bark. I borrowed a small shotgun, and in those winters we ate a lot of rabbit stew. I have fond memories of sliding silently on my skis between the young trees, looking for one of my nemeses, and coming home with dinner. Jeri was not so keen on my keeping the gun in the bedroom, so that I could pot the odd rabbit on moonlit nights. I would very quietly open the window, and ‘bang’, instantly wake up my dear wife. The following year there would be a lot of foxes, and nearly all the rabbits would disappear. The third year we would have a rabies outbreak among the foxes, and after that we would see neither rabbits nor foxes, until the cycle repeated itself.

Is a cycle, like the rabbit/fox/rabies one, a kind of stability, or is it evidence of instability? All the feedbacks are there, but in terms of population sizes from year to year, one could not call the system stable. However, since the instability was internally driven, cyclical and hence predictable, we did recognize a kind of theoretical ‘stability’. The system, we hypothesized, was simply a three species prey/predator one, which could not stabilize, since the three species had very different reproductive rates as well as different mortality effects. Our hypothetical scenario ran like this. The foxes could not control the rabbit population until the rabbits ran out of food (as indicated by their chewing unpalatable tree bark). At that stage, the rabbit population was so high that the foxes could raise large litters. However, because the rabbits were starving, they had small litters, and the foxes soon got most of them and started to starve themselves. The dense, sickly fox population was now highly susceptible to disease. The rabies virus, which is endemic in eastern Ontario, would reach the critical threshold, and switch into epidemic status, essentially wiping out 99% or more of the local foxes. The next stage would allow the vegetation and then the rabbits to recover. Since the foxes have a lower reproductive potential than the rabbits, but especially since they started from an extremely low beginning population, the next rabbit outbreak was inevitable. We therefore concluded that it was the high rabies mortality, more then any other factor, which drove the system to cycling rather than stabilize. We constructed a fine computer model, which confirmed our loose verbal version, but short of performing a well designed experiment to test our hypothesis, that is what it would remain: a hypothesis.

For reasons not at all related to trying to test our interpretation of the three species system’s dynamics, the Government of Ontario performed the perfect experiment, allowing us to come to the conclusion that out hypothesis is most probably the right one. The frequent rabies epidemics, which spread from foxes to skunks and domestic animals, was seen as a serious health problem. This I personally experienced when I was bitten by a stray cat, and had to undergo the annoying series of shots to prevent me from developing what for humans is invariably a fatal disease. The Ministry of Natural Resources developed an oral vaccine against rabies that could be included in small bait pellets. The bait could be dispersed from airplanes over sparsely populated areas. In 1989 the Ministry started an annual bait-dropping program in our area. The prediction was that this would result in a fair proportion of the fox population becoming immunized against rabies, making epidemics much less likely, and less severe. We added an extra prediction to this, that it would cause the end of rabbit outbreaks as well.

Now, with our forest being well established, we still have rabbits, but they do no longer have population explosions, and neither do the foxes. The system has stabilized very nicely, and to our delight, has had some interesting side affects as well. Three other, rarer predators, bobcat, coyote and fisher, have moved in, taking advantage of a stable supply of rabbits. I am sorry that I did not actually make a prediction that the rabies control project would increase predator biodiversity. I should have seen that coming, but then, I am willing to live with knowing that I can at least explain the phenomenon, and fit it post facto into our model. Other side effects are also welcome, such as much less damage to young trees, and no need anymore to shoot bunnies. This example of a human activity not specifically aimed at increasing ecological stability, actually resulting in increased stability is a very rare one. Far more frequently are examples of the opposite. A good example of that turned out to be the rapid increase of the snow goose population at La Pérouse Bay. As the population kept increasing, and the vegetation was being over-grazed, we had to give up the idea that a goose colony was a self-stabilizing sub-population. We next toyed with the theory that all colonies start small, get larger, and as the conditions for raising the goslings get worse and worse, the adults finally abandon the site, and start again at a new site. However, it soon became obvious that the overgrazing was not a local effect, but happened along the entire Hudson Bay lowlands from James Bay to Southampton Island. The most likely explanation for this phenomenon is that in the past the snow goose population was limited by suitable over-wintering habitat, leaving the arctic breeding grounds essentially under-populated. However, since American farmers have adopted no-tilling management of the corn growing areas of the Midwest, snow geese over-winter in the recently harvested cornfields, where there is enough spilled corn to keep the geese fat and healthy till spring. More healthy birds than ever before return to the arctic, creating an ecological crisis thousands of kilometers from the anthropogenic activity that is the cause of the crisis.

As an ecologist, I became aware early in my career, that most of the North American ecosystems I studied were either young or disturbed or both. By young, I mean post-glacial. The ice ages totally destroyed any pre-glacial ecosystems, and caused the extinction of a lot of species. When the current post-glacial period started, whatever species were left or came back had to adapt to the new conditions and to one another. It was to be expected that this readjustment was in all probability not complete as yet, and on top of that, we humans are interfering on a major scale. A lot of the instability we see in our North American ecosystems, could very well be the result of a lack of co-evolved complexity due to the legacy of the ice ages, and made worse by our destructive tendencies when dealing with our environment. We are all familiar with such cases of instability as lemming explosions in the arctic, and outbreaks of introduced insects such as the gypsy moth. Several years ago, I decided to study two closely related native moth species that have distinct and dramatic population explosions: the forest tent-caterpillar2 and the American tent-caterpillar3.

It was early on a late April morning, on a day of the purest sunshine, that I set off with two students to start our field season. When studying forest tent-caterpillars, you have a very small window of opportunity to find the egg masses of these insects in the field. The moths lay their eggs in August, glued in clusters of over a hundred, onto branches of trembling aspens, usually some four to eight meters up in the tree. It is impossible to find them until the leaves are out of the trees, and the Ontario winter weather is not particularly conducive to collecting insect eggs. The eggs hatch around May first, and within a few days the new leaves appear. We wanted to study the forest tent-caterpillar over an entire cycle from the population low right through the outbreak to the next low, studying a lot of different aspects of its biology, to challenge a hypothesis called the genetic feedback theory. We had decided that we needed to find at least forty egg masses. Despite the fact that we were looking for them at the low point in the population cycle, we had lulled ourselves into believing that such a common and ubiquitous species should be able to find. So, we drove off northwards in my VW beetle, three biologists happy to leave the university and the city behind us. In the bush, the sweet smell of breaking buds and the buzz of the first wild bees greeted us. Soon the discovery of a patch of trout lilies just coming into bloom and the first butterfly of the season filled us with excitement. The skittish mourning cloak4 flitted from one sunning spot to another, and then suddenly took off upwards, in hot pursuit of another one, which had flown into his territory. The two butterflies danced in spirals around one another high among the upper branches of a tall tree, and then as quickly as he had taken off, he was back on his favorite basking spot. The spring weather and all the new life exploding all around us made it difficult to settle down to a methodical search for caterpillar eggs, but that was what we had come for. Soon we were looking up into the crowns of trembling aspen trees, searching for the uniquely shaped, small brown bulges on terminal twigs.

By the time we found our first egg mass, lunchtime had already come and gone, and I walked around with the worst painful neck I have ever had, not to mention eye strain and wet feet. Looking up into trees trying to locate little dark brown blobs on terminal branches above you, stresses muscles in your neck that you don’t use in everyday life, staring into a bright sky challenges your eyes badly, and sooner or later you step into a snow-melt hole. Yet, despite these problems and the fact that we had 39 more to go, we were the happiest people in the world. I climbed the tree and triumphantly returned safely to ground level with a beautiful egg mass. Then, to our delight, we found another one in an adjacent tree, and two more within the next five minutes. We had stopped at several typical upland stands of trembling aspen, and walked for hours through these areas looking up at the trees, and had found nothing. But when we took a shortcut through a wet valley we found four of them. I remember our sitting on a log, and discussing the odd non-randomness of the distribution of the egg-masses, and wondering if perhaps we should check out a few other wet areas. I felt torn between my desire to lie down in a sunny spot for a siesta, and satisfying my curiosity by checking out the next wet area. We decided to continue, and by the end of the afternoon we had several more egg-masses, and, although we did not as yet appreciate it, we also had discovered one of the secrets of the extreme cyclical behavior of the of the forest tent-caterpillar.

After a few more days of collecting, we did get our forty samples, and the project was underway. At that stage we had established that at this low point in the population, 99% of the egg-masses were concentrated in the less than 5% of the forested area that early in the season had aspen trees growing in areas flooded by snow-melt water. From previous years’ experience, and from discussing our project with other scientists, we knew that during an outbreak the population is found mainly in the dryer, upland forest. We suspected that after a population collapse, at the end of an outbreak, the agents that were killing off the tent-caterpillars were sufficiently effective to essentially wipe out the entire population, except for the few colonies that happened to be in the wet areas. What were those agents, and why did the wet areas prevent the agents from being equally effective there? We started a field experiment, where we established colonies in both wet and dry sites, and in each site type, we had some colonies in the open, and some in large screen bags.

It wasn’t until a few weeks later, that the next clue was dropped into our lap. That day in the field was another memorable one, but very different from our fist field day. I was alone, and I was checking some of the colonies: counting the live caterpillars, cleaning out bags, and just carefully observing what was happening. By now the trees were fully clad in their exquisite bright green early season foliage, spring flowers were starting to fade, and a rich and diverse population of sanguivorous insects were desperately trying to increase their fitness at my expense. Yet, the world around me was serene, and peaceable. I was standing next to a small aspen, observing a tight cluster of tent-caterpillars, which were voraciously feeding on fresh aspen leaves, when I noticed a small, pale brown insect hovering nearby. It was a parasitic wasp5, and right in front of me she stealthily approached the caterpillars, landed on one of them, and jabbed her syringe-like ovipositor into the caterpillar. Leaving a microscopically small egg behind, she flew upwards. After hovering for a few seconds, she repeated her attack on another victim. Then I noticed a second wasp, and despite the squirming action of the caterpillars, most of them were parasitized before the attackers flew away. I walked around, checking several tent-caterpillar colonies, and to my delight, I found wasps attacking the cater-pillars at all upland colonies, but hardly at all at the wetland colonies. What deterred the wasps from flying into the airspace of the wetland aspen copse I could not figure out, and remains a mystery till today. But as that summer’s fieldwork progressed we found a few other types of mortality that were more severe in the upland than in the wetland sites. Predation by ants was a major one, and much more easily explained, in that the ants that did the raiding of the caterpillar colonies were subterranean nesters, which could not survive in areas that frequently flood.



1 – Letter XXXIX of The Natural History of Selborne.

2 – Malacosoma disstria – the caterpillar feeds preferentially on trembling aspen.

3 – Malacosoma americana – the caterpillar feed mostly on different cherry species and other plants in the rose family.

4 – Nymphalis antiopa – the butterfly overwinters, and in early spring the males establish territories to attract females, and try to keep other males away.

5 – It turned out to be a fairly common species, but at that time it was still un-described. On the basis of specimens collected by me, it was later described by Dr. Mason, who appropriately named it …........malacosomae, and placed it in the family Braconidae.



  dolf@harmsen.net +1 613 544 3626