THE FACTORS WHICH AFFECT ANIMALS So far we have considered the surroundings of animals from a broad point of view, in order to reach a general orientation about them. It is necessary now to go a stage farther and try to analyze these general habitats (e.g., "The Alpine zone," "an Oak wood," "the bottom of a lake") into their constituent factors. We are immediately brought up with a jerk against the fact already pointed out, that the animal ecologist requires a varie gated and rather queerly assorted knowledge of subjects which are not biological at all, and many of which may never have been encountered during his training as a scientist. He is in the posi tion of having to correlate the behaviour and occurrence of animals with botanical or inanimate phenomena which are studied by specialists, highly trained in their own lines of work. There is in existence a rather rich literature on the subject of the physical and chemical environments of animals, much of which has been accumulated by ecologists (see "Ecology" passim, and such works as Whipple's Microscopy of Drinking Water and Murray's Depths of the Ocean). The distribution and numbers of animals can be correlated with climatic factors, the gas content of waters, the soil temperature, and other such factors. Theoretically a very large amount of this work lies within the province of the meteor ologist, the chemist, or the hydrographer, but unfortunately the data which are accumulated by these specialists are seldom exactly those required by the biologist. For this reason the animal ecolo gist is frequently delayed and even completely sidetracked on fundamental work which has little direct bearing on animals, although it is necessary for the ultimate solution of the problems upon which he started to work when he was younger. This obstacle is an important one from the point of view of progress in animal ecology; it explains the peculiar (and to the outsider, rather pointless and disconnected) nature of the literature upon the subject, the fact being that animal ecologists are really trying to do everyone else's work as well as their own, without realizing that such a feat is impossible. To attempt this is also undesirable, since there are specialists studying these other questions, while there is no one but the ecologist to study the equally urgent purely biological problems met with in the field. Progress in animal ecology will therefore probably lie along the lines of friendly co-operation with other specialists, so that ecolo gists themselves will have time left to study animals.
The factors which influence animals may be divided conven iently into three main groups (a) physical and chemical, (b) plants, (c) other animals. Any of these factors may affect the distribution or numbers of a species, and much scattered work has been done on different animals. This is often of great value in indicating the nature of the environment of animals and the processes going on in it, but, as far as the animals themselves are concerned, only one or two principles of importance have emerged, partly owing to a failure to appreciate the difference between ordi nary factors and limiting factors. An example will make this difference clear. It has been shown that in certain parts of France the distribution of the freshwater crayfish is determined by the amount of calcium carbonate in the waters of the streams which form its habitat. This in turn is determined by the nature of the rocks which go to form the country through which the streams flow. Granite areas support few or no crayfish, while regions with enough calcium carbonate in the rocks are able to supply the material required by the crayfish for the formation of its skeleton. In this case calcium carbonate is the limiting factor to the distribution of the crayfish. No amount of work upon the oxygen requirements of the crayfish would have thrown any light upon its distribution, although it would tell us something about its physiology. Another example is that of the elf-owl (Micro pallas whitneyi) which occurs in the deserts of California and Arizona. This bird makes its nests exclusively in the holes made by two woodpeckers (Centurus uropygialis and Colaptes chryso ides mearnsi) in the stems of a giant cactus (Cereus giganteus). Its range in the desert is determined by these two biotic factors —cactus and woodpeckers. But the woodpeckers are not like the owl confined to the one cactus for nesting; they can also use other plants or trees, so that to them the cactus is not a limiting factor. Returning to the owl, it will also be obvious that in part of its range it might be limited by the absence of cactus (even though woodpeckers were there) ; while, elsewhere, the cactus might be present but not available for nesting purposes owing to the absence of the woodpeckers. This idea of limiting factors is one of the most important ones which is used by ecologists in the course of field work; failure to realize its significance may often result (and has done in the past) in a great deal of work which is useless, ecologically speaking. (We owe the discovery of this idea to the plant ecologists.) It has often been the custom for animal ecologists to tabulate a list of the factors which act on animals, and then to pick out any one of them and study their exact quantitative mode of operation. From what has already been said it is clear that such a method will not necessarily throw any light upon the distribution of the animal under consideration, although it may reveal interesting facts about its physiology, and about the environment itself. After all, the object of. ecology should be mainly to discover why animals occur in particular numbers at particular places : this can only be done by finding out the limiting factors and studying them.
At this point it is desirable to go back for a moment in order to reconsider the question of habitats and the best method of recording the data about them, when engaged on primary eco logical surveys. What has just been said about limiting factors puts the idea of a habitat in rather a different light. Before, we were considering it as a complex one composed of numerous factors, all of equal importance or value to the animals living among them. But although all these factors exist, from the point of view of the animal's distribution and numbers the limiting fac tors are the ones which are of paramount importance. But in practice it is not possible to tell at a glance what are the limiting factors at work on each species, so that what actually happens is that we try to record any feature in the animal's environment with which it is constantly associated. This feature may or may not be the real limiting factor; usually it is only correlated with it. When we say that an aphid occurs or the under sides of oak leaves at a certain distance from the ground, we are only describ ing in a rough shorthand way the fact that it is confined to that place by, say, conditions of humidity or the nature of the plant juices produced in the leaves at that point. It should always be remembered, however, that the ultimate aim of ecological work of this sort is to be able to describe exactly the limiting factors acting on an animal, so that we may be in a position to predict what it will do under any given circumstances. It is at this point that ecology links up with physiology; for there is really no sudden break between the chain of causes leading from the soil to the oak tree, to its leaf juices, to the aphid digestion, and then to the utilization and excretion of the matter that it absorbs. Of course, in practice, it is not usually necessary to follow up the action of environmental factors upon animals any further than the rough correlation between the outer limiting factor and the animal's behaviour. Elaborate physiological explanations only become necessary in the case of man and his domestic animals, where we wish to apply medical measures. The confusion between physiological and ecological work is one of the many snags which have held up the progress of ecology a good deal during the last zo years, and clear thinking on the matter is desirable.
We divided the environmental factors into three groups: the first two groups (physical and chemical, and plants) form great sciences of their own, and will not be treated here in detail. The influence of animals upon one another demands a much fuller consideration, and will, in fact, occupy most of the remainder of this article. The subject has been comparatively neglected by professional ecologists although it has been the chief interest of countless naturalists, especially of the older school, and more recently of a fairly considerable army of economic zoologists. The reason for this one-sided development of ecology appears to have been that plant ecology was made into a science before animal ecology had got properly under way, so that most of the earlier animal ecologists automatically followed the tendencies of plant ecology, and their interest came to be focussed upon the effects of soil, climate, and water upon species, and also upon the effects of plants, especially those due to succession. Hence, the centre of ecological interest drifted away from animal inter-relationships, which, however, carry with them the key to the understanding of animal numbers. Indeed, the subject of animal numbers has hardly figured at all in ecological literature until quite recently, although it was being intently investigated by biologists who did not term themselves ecologists, and who were working often in isolation, perhaps at some remote tropical station. It was Carr Saunders's Population Problem (a study of the numbers in man) which helped to direct attention again to the regulation of num bers in animals and, of course, more particularly in man.
The fascinating problems presented by the inter-relations of animals have attracted naturalists since the earliest times. There has been in consequence a colossal amount of literature published on this subject, which has been approached from a great many points of view. Such work falls conveniently into two sections (I) the inter-relations of members of the same species (e.g., the social habits of ants or of antelopes, the courtship of birds or of insects, co-operation in all its stages of evolution, etc.) ; (2) the inter relations between animals belonging to different species (e.g., food relations, parasitism, mimetic associations). We do not propose in this article to go over in any detail the work which has been done in the first sphere, although it comes quite logically under the heading of ecology. The fact is that with most animals, mem bers of the same species do not act as limiting factors to one another's distribution, although in a sense they can act as limiting factors to numbers (see BIRDS, Reproductive Habits). Therefore much of the work on the subject bears very little relation to the questions of distribution and numbers with which the ecologist is mainly occupied. This statement applies to a great deal of the work on flock tactics, on the pairing habits of birds and insects, and on caste-systems in insects like ants and bees.
(a) It must be sufficient to mention some of the more important sources of information on these questions, and to indicate the points at which they illuminate general ecological problems. The problems concerning the inter-relations of animals of the same species again fall into two well-defined groups : (I) relations be tween the sexes, and (2) other relationships (flocking, co-opera tion, caste-systems, etc.). Most of these complicated inter-relation ships have been brought about by the tendency of animals to evolve means of producing some division of labour which enables the species as a whole to maintain itself more effectively in the face of competition. The formation of two sexes with a suitable division of labour, is the most widespread example of this tendency, and since this division of labour has as its result the more successful production or upbringing of the young, we find that among many of the higher forms of animals the sexual re lationship has become mixed up with more or less elaborate sys tems of territory which are required for the proper provision of a sufficient supply of food for the developing young. This phe nomenon is especially well-marked in birds and mammals, and in some of the social insects. Eliot Howard (Territory in Bird Life) gives a good summary of the problem as seen in birds, and his theories have been criticized and developed further by Nicholson (How Birds Live). These two books give a clear idea of the ways in which sex relations may have a direct bearing on the population problems of a group of animals. The problems pre sented by the sex relations of insects have been fully reviewed from certain points of view by Richards, but it is at present un certain what precise part is played by these relations in the other aspects of insect economics.
(b) Division of labour takes other forms in some of the more specialized social animals. Definite castes (workers, often soldiers, occasionally more remarkable types like living honey-pots, thread spinners, etc.) occur amongst many ants, termites, and bees. This subject has been very fully treated by Wheeler (Ants; Social Life among Insects) and more briefly by Alverdes (Social Life in the Animal World). The division of labour may also take the form of temporary occupation of different jobs by different individuals (e.g., sentries in antelopes and flamingos; creche-minders in giraffes and penguins). The book by Alverdes sums up a number of facts bearing on these and other aspects of co-operation among social animals. All these points are mentioned here because it has been shown (e.g., by Wheeler, for insects) that increased co operation and division of labour between members of the same species usually have the effect of enabling a denser population to exist on a given area, and this fact has, of course, a direct bearing on the problems concerning the numbers of animals, which we shall have to consider later on. Finally, it is worth mentioning that a perversion of the caste system has in certain cases re sulted in the existence of robbers within the species. This is an ecological problem, to which little attention has been paid, but which has an important bearing upon the evolution of certain kinds of parasitism.
(c) We now come to the second type of animal inter-relation, that between members of different species. The inter-relations be tween different animals have been studied very intensively,but with the exception of work upon parasitism, organic symbiosis, mimicry, and colour adaptation, the facts discovered have been chiefly of a rather disconnected and unorganized nature. Also they have usu ally been recorded with one particular purpose in view—to prove or at any rate to illustrate Darwin's theory of adaptation through natural selection. In consequence, there has been a tendency to focus attention upon the strange and unlikely habit, the curious adaptation, the attractive resemblance, or the triumph of some successful parasite over its host. This tendency was a natural re sult of the wave of enthusiasm which followed the fruitful ideas of Darwin and Wallace, but at the present time ecological work upon the inter-relations of animals is tending in rather a different direction. It is being realized gradually that these remarkable examples of inter-relations between different species are only special cases of a very common phenomenon—the fact that animals frequently depend for their living upon their ability to catch other animals for food, a fact referred to by Darwin as "the almost universal law of `consume or be consumed'." Animal communities are not simply unorganized assemblages of animals which happen to live in the same habitat—they are intimately connected together in a most complex manner, and the tie that binds them together is the tie of food. Food and eating are so familiar that it is possible to miss their enormous importance in ecology, and to forget that most of the problems which centre around animal numbers depend for their solution upon an adequate knowledge of the food-relationships of animals ; since it is in this way that the numbers of one species are kept down by another. We shall not consider in any detail here the work which has been done upon animal coloration, commensalism, and sym biosis, since these subjects form only a comparatively small part of the whole study of food-relations among animals, and since, further, many of the conclusions which can be drawn from such work must be left in a rather indefinite state until a more settled idea is obtained of the exact way in which species arise and the part which adaptation really plays in this process.