The Balance of Nature

by Dr. Lawrence B. Slobodkin
Our earth houses literally millions of plant and animal species. A preponderance of species maintain a consistent population, a small number are increasing in abundance and some are in danger of extinction in response to human activity. Why? This must depend on the balance for each species between the characteristics that increase and decrease their individual numbers. For example, predators reduce their prey’s abundance but do not usually eat the last specimen.
Essentially all plants are edible to some herbivore. Look very closely at a mature leaf and you will find that bits have been eaten off the edge, or holes have been eaten through, or the underside of the leaf has white or grayish patterns where tiny animals have eaten tunnels into the leaf’s tissue. On Long Island, omnivorous tent caterpillars may strip an apple tree of all its leaves; while herbivores do not completely strip the leaves-it is as if their meal was interrupted. Further, all animals that eat plants are edible to some meat eater: birds consume myriads of insects, while bigger predators eat deer and zebra.
Why don’t the carnivores eat all the herbivores and then, without any food left, starve to death? Why don’t the herbivores eat all the plants and then starve to death? It sometimes happens, but not often. What regulates the abundance of organisms in nature and what is the mechanism of that regulation? What accounts for exceptions, like violent numerical changes or extinctions? Some exceptions include owls, cats and foxes, which have wiped out the quail population in parts of Suffolk County, New York. Why are there still quails in other parts of the country?
Conversely, there are occasional cases of enormous increases. For example, the vast rabbit population of Australia began with twenty rabbits imported in the mid-nineteenth century. The Pacific fishery for “Rock fish”, called striped bass on the east coast, started with less than a hundred baby fish that survived a transcontinental train trip in a milk can of water in the late 19th century and now yields many tons of fish per year (Merriman 1941).
These sorts of increases do not often occur, although all organisms can increase rapidly in proper circumstances. For example, if the animals in a population are very well fed their number will increase from year to year and generation to generation. If a species becomes very abundant, it may cause deterioration to its environment, leading to shortages of food and other resources, and deterioration in the nutritional supply for the species. This in turn produces an increase in death rate and a decrease in birth rate, or diseases may break out in more crowded populations. Conversely, decrease in density may increase the available resources per individual, permitting greater survival and reproduction.
In nature, somehow the number of offspring combines with the number of deaths so that, on average, the organism in nature just replaces itself and the population remains constant. A wild cherry tree may live for fifty years, producing thousands of seeds each year, but only one of them will grow into a replacement for its parent.
When dramatic changes of abundance do occur, it is very often related to an “exotic” species-an organism removed from one location, intentionally or accidentally, and set free in a (very) different location.
There have been several theories in the 20th century explaining why some populations grow, others decrease, and others stay constant. In the mid 20th century, texts asserted that the actual number of organisms in natural populations was random. Populations could be simulated by a “random walk” or “drunkard’s walk” theory, according to Andrewartha (Andrewartha and Birch 1954). In this theory, whether a population increases or decreases is a random event, like the walk of a person too drunk to tell right from left.
Populations cannot maintain a random walk. “Random walks” occur only if there is no limiting resource (Population sizes are limited in the downward direction by the absorbing barrier at zero.) At the very least, random walk assertions could only hold within a narrow set of conditions. Something other than randomness was occurring. If the fluctuations were not random, how were they controlled?
In contrast to the “drunkard’s walk” theory is the “density dependence” theory, in which a population increases if it is small and decreases if it is large. The loudest advocate for density dependence was the experimental population ecologist A.J. Nicholson.
Nicholson and Andrewartha were both correct, but overstated various cases. One of Andrewartha’s central examples was the abundance of rose thrips (insects that eat roses). Fred Smith felt that thrips in rose gardens in Australia are not really natural (Smith 1961). Smith showed, using Andrewartha’s data, that the changes in the thrip’s abundance were not random. Any population rate will decrease when a resource becomes unavailable to the individual organisms in growth promoting quantities. That resource is said to be “limiting.” (Blackman 1905)
In approximately 1959, after seven years of daily arguments about ecology, Nelson Hairston (Sr.), Fred Smith and I wrote a short, explosive paper that consolidated the aspects and data of ecological systems we agreed upon. (Hairston et al. 1960) It crystallized basic properties of ecological systems for the scientific community by calling attention to certain well-known facts:

  • In many terrestrial situations living vegetation not completely consumed by herbivores withers and falls to the ground. Usually withered plant material does not simply accumulate. A whole suite of decomposer organisms eats it. In temperate forests the oldest vegetable detritus on the ground surface is only two or three years old. Older material is either completely absent from the forest floor or perhaps there might be a slight brown stain indicating the last leftovers of carbon from leaves that grew more than three or four years ago.
  • More than 99% of the photosynthetic organic material produced in any year is consumed, either by herbivores eating fresh vegetation or by the animals, molds, protozoa, and bacteria that live on dead leaves, roots, and twigs. The carbon that is not consumed is stored in black and brown soils and muds, and fossilized in coal and oil. This carbon comes from from carbon that was produced photosynthetically (by plants), but has not been reduced to carbon dioxide by animals, fungi or bacteria.
  • Because almost all organic material is consumed, we asserted that if there had been more organic material (dried leaves, etc.), the consumers of leftover digestible organic material (called detritovores) would have increased.
  • Energy, mineral resources or both limit entire ecological systems, although this does not imply that all organisms are limited in the same way.
  • There is always some kind of biological control beyond random fluctuations.

We very briefly suggested how biological control could work. We hypothesized an alteration of limiting processes so that if a certain level of the food chain, such as herbivores, were limited by the predators, those predators must be limited by food (the herbivore prey). We suggested several “trophic” levels where, if one level was limited by predation, the next higher level was limited by food. In regard to plants, they are generally not limited by energy but they do have a chronic shortage of mineral nutrients. In a later paper we discussed how populations could be limited simultaneously by nutrition and predation (Slobodkin et al. 1967).
We did not, in our 1500 word paper, review the complex details. Ever since, there has been vast and sometimes highly polemical literature that referred back to these papers – not concerned with the importance of randomness in ecology but with statements made in our proof that randomness could not totally determine abundances. Are particular species controlled by food or by predators? Sometimes this curious question is treated as if these are opposite natural laws. Clearly it can be sometimes one, sometimes the other, depending on the organisms and circumstances. The literature has also accumulated cases in which herbivores have been food limited, or predator resource limited but these exceptions do not destroy our paper’s generalizations.
Several alternatives to our explanation of abundance control are particularly interesting. For example, Professor Glen Lopez (Personal communication, 2005) and his colleagues are studying the clam population of the Great South Bay of Long Island. There has been an enormous clam decline correlated with increasing fishing intensity. This change has not been accompanied by any increment in apparent food for the surviving clams. They suggested that the clams, instead of feeding only on flagellates and phytoplankton are also feeding on copepods and invertebrate larva. Food supply being released by the reduction of clams is not resulting in better nutrition for surviving clams but in better nutrition of other small invertebrates. In effect, the structure rather than just the quantities in the community is apparently irrevocably changed by removal of clams. The suggested remedy is to add large numbers of clams back into the Great South Bay. (See also the discussion of the interaction between food web complexity and trophic levels by Hairston and Hairston 1991.)
In conclusion, our short paper affirmed what we thought should have been obvious; that the ecological world is complex, but that there are mechanisms working that prevent population abundance from being just randomness. This turned out to be much more important than we anticipated. There is, in fact, a set of mechanisms that place broad limits on changes in organism abundance, but there is no externally imposed balance for ecosystems; the balance is within nature itself.
Dr. Slobodkin is a Professor Emeritus at the State University of New York-Stony Brook, Fellow of the American Academy of Arts and Sciences, Guggenhiem Fellow and Woodrow Wilson Institute Fellow. His book, “A Citizens Guide to Ecology,” published by Oxford Press, is now available as an affordable paperback.


Andrewartha, H., and L. Birch. 1954. The Distribution and Abundance of Animals. University of Chicago Press, Chicago.
Blackman, F. F. 1905. Optima and limiting factors. Annals of Botany 19:281-295.
Hairston, N., F. Smith, and L. Slobodkin. 1960. Community structure, population control and competition. American Naturalist 94:421-425.
Hairston, N. S., and N. J. Hairston. 1991. Ecological Experiments: Purpose, design and execution. Cambridge University Press, Cambridge.
Merriman, D. 1941. Studies of the striped bass, Roccus saxatilus of the Atlantic coast. Fisheries Bulletin, U.S. Fish and Wildlife Service 50:ii+77.
Nicholson, A. J., editor. 1957. The self-adjustment of populations to change., Cold Spring Harbor.
Slobodkin, L. 1980. Growth and Regulation of Animal Populations, Second edition. Dover Publications, New York.
Slobzdkin, L., F. Smith, and N. Hairston. 1967. Regulation in terrestrial ecosystems and the implied balance of nature. American Naturalist 101:109-124.

Originally posted in “On Eagles’ ¬†Wings” September 16th 2005