(3) The conditions in the external world may be organic or inorganic. By organic I mean the conditions which depend upon the number of competitors or enemies by which a bird is surrounded. The competitors may include other species which require a similar environment; and the enemies, species which prey upon it, or animals which take its young or its eggs. They vary in different seasons, in different districts, and in nature and extent—the success of one species leads to the failure of another, and the multiplication of the Jay or of the Magpie robs us of many a songster.

By inorganic I refer to the changes in the climate and in the surface of the earth. The nourishment of the young depends upon a regular supply of food, and the supply of food depends upon the climate which alters in different periods; in one decade the temperature falls below, whilst in another it rises above, the normal, and, as the insect life fluctuates, so there is fluctuation in the bird population. The changes in the surface of the earth are manifold. Little by little the alder (Alnus glutinosa) overspreads the marsh. Young shoots spring up here and there, in a few years grow into bushes, and in a few more years are trees; and the dense masses of rush which seemed to choke their growth, yielding their position of importance, slowly disappear. And where formerly the Orchis latifolia, Orchis mascula, and Juncus communis grew in mingled confusion, nothing but water, moss, and the spreading roots of alder cover the ground. As the rush disappears, many birds that for generations have inhabited that marsh must seek accommodation elsewhere. Ancient breeding haunts thus disappear, new ones come into being, and even those which appear to be permanent are almost imperceptibly changing.

Now the bird inherits a nervous system, which works under internal excitation and external stimulation. Given the appropriate organic condition and adequate stimulation, and the impulse to seek isolation comes into functional activity. What the organic condition is and how it arises we do not exactly know; all we know is that organic changes do take place in the breeding season, that these changes profoundly modify character, and that they correspond with the seasonal growth of the sexual organs. And with regard to the question of stimulation, we have again to confess to much ignorance, although certain facts are presented to observation which seem to indicate the direction in which the stimulus lies. For example, it is well known that abnormal climatic conditions influence behaviour; we see migrants retracing their flight along the very course they travelled a short time previously—driven headlong by the blizzard, that at least is what we say. But if the wind, instead of being cold and from the north, is warm and from the west, do they retrace their flight? I have not found it so. And if there be no wind and the temperature is low, are they still affected? Again, I have not found it so. When, as we commonly say, they fly before the storm, some change takes place in their organic complex, some new impulse receives stimulation or the former one lacks it. If, after Lapwings have established themselves in their territories, the weather becomes exceptionally severe, the birds collect together again in flocks and revert to their winter routine; and under similar circumstances, Buntings fail to sing and temporarily desert their territories. In such cases it is clear that the impulse to seek isolation ceases for a time to dominate the situation. The inference, therefore, is that atmospheric changes bear some relation to the functioning of the instinct; but whether it be temperature, or humidity, or the direction and velocity of the wind, or a combination of two or more of these factors that supplies the stimulus, we cannot tell.

The appropriate organic condition and the stimulus have then still to be determined, and we must pursue our inquiry from the point at which the impulse comes into functional activity. We will take a simple case, and one free from complication.

Let us suppose that there is an area bereft of bird life, if it can be so imagined, but in proximity to other inhabited areas. Into this area, whilst in search of isolation, let us imagine that a Yellow Bunting finds its way. After the manner of its race it establishes a territory and occupies, let us say at a low computation, half an acre of ground. It then obtains a mate, breeds, and rears offspring, two of which we will assume are males. Reproduction ended, the birds desert the area, and in the following spring, when the impulse again asserts itself, parents and offspring seek again their former haunts. We now have three males, each of which occupies half an acre, and each of which rears two offspring—that is the position at the close of the second year. In the third year the number will have increased to nine and the area occupied to 4½ acres; and so on in succeeding years, until by the beginning of the eleventh year, we have 59,048 Yellow Buntings occupying 29,524½ acres or 46 square miles. This, then, will be the result of the operation of the impulse, providing that all the individuals survive and that no complications supervene.

But of course complications are numerous, some of which retard while others accelerate the rate of expansion. These complications arise from various sources—in the first place from natural enemies which prey upon the birds or upon their eggs; in the next place from climate which, if it happens to be unfavourable, may mean that food is scarce and that only a small percentage of the young survive; and lastly from rivals—and by rivals I mean closely related forms that require a similar station and similar food—which, by occupying available ground, may check expansion, or, by forcing a continuation of the search, may widen it.

Now when individuals fail as many do fail in their initial attempt to secure territory, the activity of the impulse still persists, and there is no control over the direction in which the bird continues to wander whilst in search of its end. Some therefore seek in this direction, others in that; some wander inwards into inhabited areas and fail to find accommodation, or, according to the relative strength of their impulse, perhaps succeed and so set free a new competitor, others wander outwards into country uninhabited by the species. These latter we will call "pioneers." They may find accommodation within a comparatively short distance of their base, or they may come into competition with rivals and fail, not necessarily on account of any congenital weakness of ability, but because being warned by an alien song, they may be precluded from coming into contact with just the environing conditions which can supply the stimulus and allow behaviour to run its further course—and so be obliged to extend their search into remoter districts. But it must not be overlooked that they will be placed in a most advantageous position so far as the attainment of reproduction is concerned. In their search for territory they will meet with little opposition and will be free to select whatsoever ground they will; and be free also from intrusion by neighbouring males, which is so frequent in occurrence and continues for so long in congested areas. Moreover, in thinly populated districts, the pressure upon the available means of support will not be so great, neither will natural enemies be so plentiful; and since the offspring, guided by prior experience, return to the neighbourhood of their birthplace, the advantages thus gained will be shared by the succeeding generation. It follows, then, that the range of a species will not always be continuous, will not, that is to say, proceed by a series of successive steps, but that sometimes in this direction and at other times in that, the chain of territories will be interrupted and different individuals separated by distances of greater or lesser extent. New colonies will thus come into being; and as the unlimited increase of the population over limited areas gradually reintroduces into them the struggle for territory, new centres of distribution, where the process will repeat itself and from which expansion will proceed afresh, will be formed. Hence, though it is clearly impossible for the progeny of one pair of Yellow Buntings to overspread the whole of the 46 square miles, it is by no means impossible for the limits of their range to exceed even those limits within the eleven years.

To sum up our knowledge regarding this phase. Of the organic condition which renders the impulse responsive to stimulation we know very little; and though certain facts of observation seem to indicate the direction in which the stimulus is to be found, we must here again confess to much ignorance. So far as can be seen, however, the impulse to seek isolation with its correlative territory, leads to constant modification in the breeding range of most species. The occupation of the small space of ground which each individual requires, the extent of which has been gradually adjusted to suit the needs of different species, results in expansion not only in one direction but in every direction, and not only in one season but in every season. And if there were no complications in the external world this expansion would proceed, as we have seen, with astonishing rapidity. But complications, some of which are favourable and others unfavourable, are numerous, and it is difficult to estimate their importance or to indicate their precise effect; the former, however, accelerate the rate of expansion, whilst the latter retard it. Those individuals that wander outwards and seek territory on the outskirts of the range we have called "pioneers." They will have advantages over others that, wandering inwards, seek isolation in congested districts, and will succeed where the latter fail; and since there is in the young an innate ability to return to the district wherein they were reared, the advantages so gained may be said to be handed on from generation to generation.

Let us now turn to the contra-phase, and endeavour to ascertain whether the gregarious instinct bears any relation to the seasonal desertion of the breeding ground. The conclusion at which we have already arrived regarding this instinct is that it forms part of the inherited nature of most species; that its functioning is suppressed when a bird is actually in occupation of a territory; and that it is serviceable in promoting the welfare of the individual. We cannot of course observe the instinct. What we observe, when reproduction is ended, is a change in the relations of different individuals; instead of arousing mutual hostility, they attract one another, from which we infer the existence of something which determines their conduct, and this "something" we speak of as an instinct.

To what does this change lead? Let us suppose that there is an area inhabited by one species; that the number of inhabitants has reached the maximum that the means of sustenance will allow; and that the season of reproduction is drawing to a close. The position will then be as follows. All the available breeding ground is divided into territories; each territory is occupied by one unit, the family, and each individual is able to fend for itself; changes both internal and external begin to take place, the gregarious instinct comes into functional operation, and the supply of food diminishes—that roughly is the position. The internal factor operates so that the sight of this individual or the call of that, instead of evoking hostility as heretofore and keeping different units apart, proves now an irresistible attraction; so that in place of a number of individuals evenly dispersed over the whole of this area, a small number of flocks of various dimensions are stationed at certain points, which points are determined partly by experience, partly by the supply of food, and partly by accident. This implies for each individual some movement in some direction. But since the population of this imaginary area has reached the maximum, and the supply of food, though limited in distribution, is nevertheless plentiful, such movements will be irregular and will proceed in no definite direction.