Editor’s Note: E. S. Russell was a Scottish marine biologist with a strong interest in the history and philosophy of biology. His 1916 history of morphological thought, Form and Function, remains a standard reference work today. We present here an abridged and slightly edited version of chapter 5 from his 1945 book, The Directiveness of Organic Activities. The chapter is entitled “Characteristics of Goal-Directed Activity” in the original.
Perhaps nothing about the life of organisms is more obvious to biologists than the future-oriented and end-directed character of development, physiology, and behavior. In Russell’s terminology, we observe a “goal- directed,” “purposive,” and “directive” aspect in all organic activity. At every level of their functioning, organisms carry out tasks, satisfy needs, and pursue their own interests. Biologists unavoidably define their subdisciplines and research projects in task-oriented terms: How is DNA replicated? How and why do cells divide? By what means is food converted into useful energy? What strategy enables the predator to capture its prey?
The future-and-task-oriented activity of living beings radically distinguishes them from rocks, waterfalls, and clouds. The odd thing is that, despite their unavoidable recognition of this fact, biologists commonly exclude it from their theoretical understanding of organisms. They talk about organisms (and their evolution) as if they were merely physical systems, even though such systems are incapable of directing their activity toward the future fulfillment of their own ends. It is, for example, more or less anathema to speak of evolution as an end-directed process. (On this latter point, see “Whole Organisms and Their Evolutionary Intentions” (Talbott 2018).
It was a major part of Russell’s life-long work to try to keep the directiveness of organic activity in mind, whether he was discussing heredity and genetics, animal behavior, or evolution. And therefore he brought disciplined and systematic attention to his observation of this directiveness. In today’s environment, that kind of attention may usefully remind us of the missing element in so much of our own biological science.
British spellings and usage have been retained here. Editorial interpolations are in brackets. We have prefaced the chapter with a brief excerpt from the concluding chapter of The Directiveness of Organic Activities. (For more about Russell, see In Context #22. And for further writing by Russell, see “From Mechanistic to Organismal Biology” in In Context #30.)
Preface: An Introduction to Some Key Terms
[Russell here refers to the overall content of the book:] We have reached the conclusion that directiveness and creativeness are fundamental characteristics of life, shared by no inorganic system; that they are not to be explained in terms of mechanism or of [human] purpose; that human directiveness and purposiveness in thought and action are a specialised development of the directiveness and creativeness inherent in life.
We recognise the fact that organic activities, as manifested by organised unities such as cells and organisms, show characteristics, especially their directiveness, persistency and adaptability, which are shown also in the instinctive and intelligent behaviour of ourselves and other animals. But we shall not fall into the error of interpreting organic activities in general in terms of conscious purpose or ‘mind’.
The organism strives to persist in its own being, and to reach its normal completion or actualisation. This striving is not as a rule a conscious one, nor is there often any foresight of the end, but it exists all the same, as the very core of the organism’s being. Confusion is apt to arise because to characterise [this drive] we have no other words than those which carry a psychological meaning, words such as ‘effort’, ‘perseverance’ and ‘urge’, which are primarily applicable to human behaviour. [Botanist Agnes] Arber has called attention to this difficulty, but holds, quite rightly, that the use of such words is unavoidable, ‘because we have no other set of terms in which to express that compulsiveness by which the plant works towards a certain end, which, if we were concerned with a self-conscious organism knowingly pursuing a train of activity, would be recognized as a goal’ (1941, p. 87). We really require new terms to characterise the goal-directed and biologically purposive activities of living organisms, of which only some reach the level of conscious purposiveness.
Characteristics of Goal-Directed Activity
There are certain general or normal characteristics of all goal-directed activity (whatever its biological end) which may be summarised as follows:
1. When the goal is reached, action ceases; the goal is normally a terminus of action.
2. If the goal is not reached, action usually persists.
3. Such action may be varied:
(a) if the goal is not reached by one method, other methods may be employed;
(b) where the goal is normally reached by a combination of methods, deficiency of one method may be compensated for by increased use of other methods.
4. The same goal may be reached in different ways, and from different beginnings; the end-state is more constant than the method of reaching it.
5. Goal-directed activity is limited by conditions, but is not determined by them.
I shall now proceed to illustrate and exemplify these rules, which are normally, though not invariably, valid for directive activity, whether behavioural, physiological or morphogenetic.
The Goal is a Terminus of Action
When a rat has satisfied its appetite for a specific food substance it takes no more of it, until the need arises again; a full-fed animal normally ceases to feed. When a wound is healed over and the normal density of cells restored, the cellular activities directive towards these goals come to an end. When Molanna [a genus of insects known as “hood casemakers”] has repaired its case to something like normal, it ceases its efforts at restoration. When an animal’s normal body temperature is restored by physiological or behavioural action, these regulatory activities are reduced to the minimal level required for the continued maintenance of normality.
When a motor nerve is severed, the fibres that run from the cut to the muscle degenerate and die, but the cut-end at once starts growing out to re-establish connection with the muscle.
The fibre, so to say, tries to grow out to reach to its old far-distant muscle. There are difficulties in its way. A multitude of non-nervous repair cells growing in the wound spin scar tissue across the new fibre’s path. Between these alien cells the new nerve-fibre threads a tortuous way, avoiding and never joining any of them. This obstruction it may take many days to traverse. Then it reaches a region where the sheath-cells of the old dead nerve-fibres lie altered beyond ordinary recognition. But the growing fibre recognises them. Tunnelling through endless chains of them, it arrives finally, after weeks or months, at the wasted muscle-fibres which seem to have been its goal, for it connects with them at once. It pierces their covering membranes and re-forms with their substance junctions [interface between distinct tissues] of characteristic pattern resembling the original that had died weeks or months before. Then its growth ceases, abruptly, as it began, and the wasted muscle recovers and the lost function is restored (Sherrington, 1922, p. 6).
This is an excellent case of persistent directive action, surmounting difficulties, and coming to an end when the goal is reached. It is, of course, typical of regenerative processes that they stop when normal structure is restored. If a newt’s leg or arm is truncated, the wound is healed and a regeneration bud formed under the new skin. This grows and differentiates, reforming the missing part of the limb, in normal shape. ‘When we are dealing with the regeneration of an adult organ such as a newt’s limb’, writes Waddington (1934, p. 339), ‘the equilibrium towards which the regeneration takes place is a stable one; once the whole limb is regenerated, the change stops, except perhaps for growth changes.’ Growth and development in general move towards a definite or end-state, and cease when this is reached. A waterlily leaf grows up towards the surface of the water, and there expands to its full shape; growth persists till this end-state is achieved, and then stops.
Waterlily
In animal behaviour cessation of action when the goal is reached is so common a phenomenon as to require no elaborate demonstration. When a starfish or a beetle is turned on its back, the righting behaviour which follows ceases as soon as the normal position is regained.
It is interesting and significant to observe that if the goal or end-state is supplied by external agency, and not by the animal’s own efforts, the restorative action will likewise come to an end. If, when the beetle is struggling to right itself, you present it with something to which it can cling, it will cease its efforts, which have now become unnecessary. The same is true of the starfish (see, for instance, Russell, 1919).
Here is a more complex but very instructive example, which I take from Boycott (1929). The number of erythrocytes in the mammalian blood — which Boycott calls collectively ‘the erythron’ — is kept at a normal level (which is relative to the pressure of the atmospheric oxygen). If the erythron is reduced by severe haemorrhage, the loss is ultimately made good by the production of new erythrocytes from the bone-marrow, a process which goes on till the goal or end-state of a normal concentration is reached. The erythron can also be increased by transfusing an extra quantity of blood into the circulation; when this happens the excess quantity is actively destroyed by phagocytosis and normality again restored. The end-state or goal of these processes is quite definite — the restoration of the normal amount of red cells. Both processes, by the way, are accelerated by practice. Now if, in a bled rabbit, the amount of blood removed is quickly restored by transfusion, and the normal end-state thus artificially restored, then neither the process of destruction of the introduced erythrocytes takes place, nor the production of new erythrocytes from the marrow, for the goal aimed at has already been reached.
Editorial Comment: Regarding the “bled rabbit” just referred to, and, even more, a number of experiments discussed below, we find ourselves experiencing an all-too-familiar discomfort. The assaultive animal research we occasionally find ourselves reporting on has at times been difficult to stomach, and certainly contravenes our own values. On the other hand, (1) we feel that E. S. Russell’s understanding of organisms holds great importance for biologists today, and we’re not sure anything would be gained by using his references to contemporary research as a reason for discouraging access to his writing; and (2) given the fact that the massive abuse of animals in research today is largely hidden from public view (and therefore from the public consciousness), the “jolting” feeling one has at reading some of the descriptions about this research may at least have the beneficial effect of bringing the reality to public attention. Beyond that, we still never feel completely at ease when presenting the results of this sort of research, and have never found a way of handling the issue that is wholly comfortable for us.
Persistency of Action
It is characteristic of animal behaviour that if the goal is not reached at once, directive action continues with or without variation of effort. The salmon encountering a waterfall on its upstream migration tries time and again to surmount it, until its leaps are successful or it falls back exhausted; there is active, persistent, long-continued effort. Here there is an element of active striving, closely similar to what we ourselves experience when trying to do a difficult job. But persistence in action need not imply conscious effort, and it may be quite stereotyped and unintelligent — that is, inadaptable to circumstances. A pair of sparrows one summer built a nest in the roof-gutter of my house, and when this was cleared away built another in the same place; several times the nest was swept away and rebuilt by the persistent birds. Nor is persistence of action limited to behaviour; it is shown also in physiological and morphogenetic activities. A good example of this is afforded by persistent egg-laying in birds, when they are prevented from accumulating a clutch of normal size in the nest.
According to Bickerton (1927, p. 21), a great tit (Parus major) has been known to lay 25 eggs, instead of the normal 6-11, when she was left each day with only one egg in the nest; a starling (Sturnus vulgaris) in similar circumstances produced 40 eggs consecutively, and a moorhen (Gallinula chloropus) 49 eggs in 57 days. According to Herrick (1935, pp. 256-7), ‘The northern flicker (Colaptes auratus luteus) commonly lays from five to nine eggs, but when it is systematically robbed from the time it lays its second egg, one being taken each day and one left as a nest-egg, the number it will sometimes produce is surprisingly great. The record of seventy-one eggs in seventy-three days made by a bird at Taunton, Massachusetts, beginning May 6, 1883, and reported by Charles L. Phillips, I believe has never been surpassed’.
Persistence of growth activity until a goal is reached is, of course, a common phenomenon in plants; the potato sprouting in a dark cellar sends out long white shoots, as it were in search of light; the prairie plant grows enormously long roots in its search for deep-lying water. The wood-rot fungi extend their rhizomorphs over brick, stone or metal in their apparent search for distant woodwork.
Variation of Action
(a) Persistency with varied effort. It often happens that the animal has several instinctive modes of action available for dealing with a particular situation, and if one fails it brings the others successively into play until the end-state or goal is attained. Take, for instance, the responses of the protozoon Stentor to nocuous stimuli (Jennings, 1906, pp. 170-9). If a stream of fine particles, say of Indian ink or carmine, is directed upon the disc of an actively functioning Stentor, no avoiding reaction is at first obtained; the Stentor ingests some of the particles. But soon it bends away, thus avoiding to some extent the irritating stream. If this reaction is not successful at first, it is repeated.
If the repeated turning toward one side does not relieve the animal, so that the particles of carmine continue to come in a dense cloud, another reaction is tried. The ciliary movement is suddenly reversed in direction, so that the particles against the disc and in the pouch are thrown off. The water current is driven away from the disc instead of toward it. This lasts but an instant, then the current is continued in the usual way. If the particles continue to come, the reversal is repeated two or three times in rapid succession. If this fails to relieve the organism, the next reaction — contraction — usually supervenes (p. 174).
It is important to note that the order of events is not stereotyped, for sometimes the reversal of current may be tried before the turning-away response.
The third method of avoiding the nocuous stream, contraction into the tube, may last about half a minute, when the Stentor expands again. It does not then repeat the previous reactions, of turning away and reversing the current, but if the nocuous stimulus is still present it contracts again, repeating this response many times, during a period of 10-15 minutes, staying in the tube a little longer each time. Finally, it ceases to expand, contracts violently and repeatedly in the tube, and breaks away its attachment to the substratum. It then leaves the tube and swims away, to form a new tube elsewhere. If, on coming out of its tube forwards, it encounters the cloud of particles, it may swim backwards and force a passage through the substance of the tube.
That is a typical example of what Jennings calls ‘trial and error’ behaviour. It is directive, in that it aims at a definite end-state or goal, relief from the irritating stimulus; it shows persistency with varied effort, for if one response fails to give relief others are tried until success is achieved. It does not imply learning, in the sense of bettering performance through experience and repetition, but it does imply the power of varying behaviour according to the result of previous action, a power which is essentially a psychological one, a power of relating events, and acting in accordance with the situation; what one may call a practical judgment seems to be involved.
Many investigators, following Bethe, have called attention to the fact that if an animal is deprived of one or more of its legs, it immediately regulates its locomotory movements so as to compensate for the loss. The ordinary rhythm of progression is radically altered, but unity of action is preserved. Thus the shore crab (Carcinus maenas) with all legs intact moves these in a regular order when crawling forward; amputate one or more, and progression is still carried out effectively, though the order of movement of the remaining legs is changed. Progression is a function of the neuro-muscular system as a whole, not a summation of separate limb reflexes. So, too, ‘an insect which has lost a leg will at once change its style of walking to make up for the loss. This may involve a complete alteration of the normal method, limbs which were advanced alternately being now advanced simultaneously. The activities of the nervous system are directed to a definite end, [namely] the forward movement of the animal — it uses whatever means are at its disposal and is not limited to particular pathways’ (Adrian, 1933, p. 468).
In these cases we have to do with an immediate physiological regulation of movements, and not, strictly speaking, with an adaptative regulation of directive behaviour, for locomotion is not by itself behaviour.
But we find a similar principle exemplified in behaviour. If the organ normally employed in a particular job is missing, [an adjustment] is often made with a substitute organ, or organs, so that the normal end is attained by unusual means. The long hind-legs of the dung beetle are well adapted for impelling the ball backwards over the ground by alternating strokes, and they play the major role in this activity. But if one or both are amputated, the beetle can still push the ball along; if one only has been removed, the ball tends to move towards that side and this obliquity has to be corrected by extra efforts on the part of the stump and the middle leg of the same side; if both hind-legs are cut off, the beetle manages quite well with its middle legs and the stumps of the hind ones. With the loss of both hind-legs and both middle legs it still persists in its efforts to roll the ball along, using the end of the abdomen as well as the stumps of the legs. It even succeeds in digging the usual burrow in which to bury the ball. But if the fore legs only are removed, the beetle is helpless; it cannot replace the powerful thrusting movements of these limbs by any other organic means.
Dung beetle
Behaviour is essentially a striving towards an end or end-state; the effort is persistent and, usually, varied. It may be blindly persistent along one line, without material variation of effort, especially in the case of highly specialised and stereotyped instinctive behaviour; but usually the effort is varied if success is not achieved at once.
(b) Compensatory activity. Where there are two or more functionally equivalent methods normally concerned in reaching a particular goal, if one is put out of action or deficient, the goal is often attained through enhanced activity of the remaining method or methods. We have seen several examples of this [under “Persistency of Action” above], and we may here recall and add to them.
In the healing of wounds in the higher vertebrates three processes are involved: (1) the active contraction of the tissues underlying the wound, whereby the exposed surface is reduced in size; (2) the migration of epithelial cells over the wound; and (3) their multiplication to supply the necessary number for epithelisation. In large wounds, where the need for cells is great, multiplication is more marked and goes on simultaneously with migration. Now if either contraction or epithelisation is prevented from taking place, the wound will nevertheless heal in normal time. Here is what Carrel tells us on this point. The wound scar
is due to the collaboration of two types of tissue, the connective tissue filling the wound, and the epithelial cells, which advance over its surface from the borders. Connective tissue is responsible for the contraction of the wound, epithelial tissue for the membrane that ultimately covers it. The progressive decrease of the wounded area in the course of repair is expressed by an exponential curve. However, if one prevents either the epithelial tissue or the connective tissue from accomplishing its respective task, the curve does not change. It does not change because the deficiency of one of the factors of repair is compensated by the acceleration of the other. Obviously, the progress of the phenomenon depends on the end to be attained. If one of the regenerating mechanisms fails, it is replaced by the other. The result alone is invariable (1936, p. 202).
To compensate for loss of blood in severe haemorrhage, the body disposes of various converging methods, which are thus described by Carrel:
First, all the vessels contract. The relative volume of the remaining blood automatically increases. Thus, arterial pressure is sufficiently restored for blood circulation to continue, the fluids of the tissues and the muscles pass through the wall of the capillary vessels and invade the circulatory system. The patient feels intense thirst. The blood immediately absorbs the fluids that enter the stomach and re-establishes its normal volume. The reserves of red cells escape from the organs where they were stored. Finally, the bone marrow begins manufacturing red corpuscles, which will complete the regeneration of the blood (ibid. p. 198).
For bringing about the first step in the process, the restoration of arterial pressure and blood volume, there are two converging mechanisms — contraction of the vessels and the taking up of water from the tissues and the alimentary canal. According to Carrel ‘each of these mechanisms is capable of compensating the failure of the other’ (ibid. p. 203).
In both these cases, wound-healing and the replacement of lost blood, it is the attainment of the normal end-state that matters; if one contributory means towards this end fails or is deficient, the others make up for it by persisting till the goal is reached, if reached it can be.
The work of Richter and his collaborators also points to the same conclusion. When the maintenance of body temperature by physiological means is upset in the rat by removal of the hypophysis [pituitary gland] and the consequent reduction in thyroid activity, the animal reacts by greatly increasing its nest-building activity; behavioural action compensates for deficient physiological regulation. We may recall here the observations of Gannon on the cat deprived of its main physiological defences against cold; it compensates for this loss by keeping in the warm, and by increased shivering.
Alternative Ways of Reaching the Same Goal
In regeneration it often happens that a structure is re-formed in a different way and from different cellular material than is the case in embryonic development; the result is the same, though the beginnings are radically different.
[Regarding the classical case of the regeneration of the lens in Urodeles, an order of amphibians that includes newts and salamanders:] In 1891 Colucci observed that if the eye was extirpated in the newt the new lens developed from the margin of the bulb. This was confirmed by G. Wolff (1895), who showed that the new lens was developed from the upper edge of the iris, a structure which in ontogeny has nothing to do with the formation of the lens. The process of lens formation from the iris is a very remarkable one. The iris consists of a double layer of epithelial cells, the inner layer of which is deeply pigmented. Cells proliferate from the top edge of the iris, forming a spheroidal mass hanging down in the position of the lens. The pigment disappears, and the cells become transparent; those in front become cubical epithelium, those behind lengthen and form lens fibres, and so a normal lens is produced in this unprecedented way. The later stages of the process are similar to those occurring in ontogeny, but the beginnings of the process are entirely different; specialised cells having no connection with the original lens dedifferentiate, multiply and redifferentiate in such a manner and in such an arrangement as to produce a typical and functional lens. It seems that what is lacking, a lens, is formed by the means nearest to hand. The organism, as Wolff says (1894, p. 620), ‘chooses the simplest way’. Thus, the same end-result is achieved as in ontogeny, but from a totally different starting point.
The reconstitution of a complete organism from dissociated tissue cells is an extraordinary phenomenon which has been demonstrated to occur in certain sponges by H. V. Wilson, Galtsoff and others. In [the sponges] Microciona and Cliona, for example, cells isolated by squeezing a part of the sponge through fine bolting silk come together and form aggregates, which later organise themselves into little sponges.
Under normal conditions, the essential features of sponge body, i.e. flagellated chambers, canals and spicules, are formed [from dissociated cells] within five or six days. Spongin [the collagen protein forming the fibrous skeleton of sponges] appears about the tenth day. The osculum [an excretory structure] appears by the end of the third week . . . The rebuilding of a new sponge is due to the activities and properties of individual cells forming an aggregate. The different types of cells forming a common mass find each other and then develop flagellated chambers, skeleton, mesenchyme, and other tissues’ (Galtsoff, 1925, p. 248).
What greater contrast can one imagine with the normal process followed in development from the egg? A sponge is of course more an organised colony of essentially separate cells than a true metazoan organism, as Bidder (1937) has clearly shown, and more independence of action might be expected from its cells. But the phenomenon of reconstitution from dissociated tissue cells takes place also in the hydroid Antennularia, though here the reconstitution of the whole is not so complete (Morgan and Drew, 1913-15).
Goal-Directed Activity May be Dominant over Conditions
In the examples described above it is clear that ends or goals are more important than beginnings; the same goal can be reached from quite different starting points, and by quite different routes. It is difficult then to think of goal-directed activity as being completely determined by its beginnings. Somehow, though we do not know how, the end or goal enters into the determination of these activities; they are, that is to say, directive.
The same conclusion imposes itself when we consider the fact that goal-directed activity is determined only in part by the conditions obtaining during its execution. When these conditions are unfavourable, the organism may yet achieve its biological ends, either completely or in a modified form. In dry or impoverished soil, for example, a plant may still achieve development and reproduction in spite of the adverse conditions, appearing in a stunted and depauperate form, which is the best that can be done in the circumstances.
A seed plant is normally dependent for its initial growth and development upon the stores of nourishment contained in the seed. If these are removed by operation or much reduced, a common result is the production of a stunted seedling. But this is not always the case. It has been shown by Raymond Pearl and his collaborators (Gould et al. 1934) that seedlings of the canteloup (Cucumis melo) from which parts of the cotyledons (containing the food reserves) have been removed show a more vigorous growth than unoperated seedlings; they are stimulated to utilise more fully and effectively the food material still remaining. Growth is not, as one might expect, proportional to the amount of food available, but considerably in excess of expectation where the amount is reduced. In fact, in the conditions of these experiments ‘the larger the amount of cotyledonary tissue removed by operation the more rapid was the inherent growth rate of the seedling per unit of time’ (p. 598). The seedling reacts to deficiency of normal food supply by more efficient utilisation of what remains.
Regeneration in planarians [flatworms], as Morgan has shown (1901, 1902), is dependent only to a slight extent upon the food conditions. The size of the regenerated whole naturally depends upon the amount of food and energy available; if the regenerating piece is well supplied with food it may reconstitute a full-sized whole; if food from outside is lacking, it will reconstitute a whole of much reduced size, drawing the necessary energy and material from its own substance. Even a planarian that has been greatly reduced in size by long-continued starvation will, if cut in two, reconstitute two new though tiny wholes. This, as Morgan points out (1901, p. 28), is a very remarkable phenomenon, and well illustrates the dominance of ‘drive’ over conditions, for the tissues of the regenerating halves, which are already ‘slowly starving to death’, are depleted still further to supply material and energy for the growth and differentiation of the new tissue formed in the reconstitution of the new wholes. We see very clearly from this example that the directive activities of regeneration are not determined, though they may be influenced, by food conditions.
Looking back over the characteristics of goal-directed activity, we see that such activity does not fall into line with that shown in the inorganic world, but is clearly separate and distinct. Coming to a definite end or terminus is not per se distinctive of directive activity, for inorganic processes also move towards a natural terminus; the moving stone rolls down the hill till it reaches the bottom, or is stopped by some obstacle; the unstable system moves towards a stable equilibrium; the same stable equilibrium may even be reached from different starting points. What is distinctive is the active persistence of directive activity towards its goal, the use of alternative means towards the same end, the achievement of results in the face of difficulties. Goal-directed activity is no mere resultant of material conditions, as is the case with inorganic systems; there is in it an element of effort or striving, which sometimes, as in our own purposive behaviour, becomes conscious of itself and its aims, but is more often unconscious and blind.
It is not dominated by conditions, but strives to surmount or utilise them in its movement towards its goal. One drive may dominate another.
This element of drive, effort or striving (which we experience in its highly developed form as conation [conscious human willing]) is one factor in all vital activity, behavioural, physiological and morphogenetic, which essentially distinguishes it from inorganic action. ‘Living things are not completely at the mercy of their environment, whereas non-living matter has a totalitarian subjection to external surroundings. Thus a non-living mass of protein always rolls down a slope with unquestioning obedience to the law of gravity; living protein in certain forms can move up the slope, following its internal direction . . . Motile bacteria can move against a slight stream of liquid’ (Grainger, 1940, p. 539).
Behaviour, as we have seen, is just one of the means or methods of action through which the living organism achieves its biological ends; physiological and morphogenetic activities are also means or methods, functionally equivalent to behavioural action. It is not astonishing then that they should have the same characteristics as behavioural action, for all three share the fundamental character of directiveness.
References
Adrian, E. D. (1933). Nature, London, 23 Sept. Arber, A. (1941). Biol. Rev. xvi, 81-105.
Bickerton, W. (1927). The Baby Bird and its Problems. London. Bidder, G. P. (1937). Proc. Linn. Soc. Lond. 1936-7, pp. 119-45. Boycott, A. E. (1929). Proc. Roy. Soc. Med. xxiii (Section of Pathology, pp. i-ii).
Carrel, A. (1936). Man the Unknown. London. Galtsoff, P. S. (1925). J. Exp. Zool. xlii, 183-256.
Gould, S. A., Pearl, R., Edwards, T.I. and Miner, J.R. (1934). Ann. Bot. London, xlviii, 575-99.
Grainger, J. (1940). Nature. Lond., cxlvi, 539-41.
Herrick, F. H. (1935). Wild Birds at Home. New York and London. Jennings, H. S. (1906). Behavior of the Lower Organisms. New York. Lecomte du Noüy (1936). Le Temps et la Vie. Paris.
Morgan, T. H. (1901). Regeneration. New York.
Morgan, T. H. (1902). Arch. Entw Mech. Org. xii, 179-211.
Morgan, W. de and Drew, G.H. (1913-15). J. Mar. Biol. Ass. U.K. x, 440-63.
Richter, C. P. (1941). Psychosom. Med. iii, 105-10.
Russell, E. S. (1919). Proc. Zool. Soc. London, pp. 423-39. Russell, E. S. (1930). The Interpretation of Development and Heredity. Oxford.
Sherrington, G. S. (1922). Rep. Brit. Ass. 1922, pp. 1-15. Waddington, C.H. (1934). Sci. Progr. 1934, pp. 336-46. Wolff, G. (1894). Biol. Zbl. xiv, 609-20.
Wolff, G. (1895). Arch. Entw Mech. Org. i, 380-90.
References from Editor’s Note
Russell, E. S. (1916). Form and Function: A Contribution to the History of Animal Morphology. Chicago: Chicago University Press.
Russell, E. S. (1945). The Directiveness of Organic Activities. Cambridge: Cambridge University Press.
Talbott, Stephen L. (2018). “Whole Organisms and Their Evolutionary Intentions: An Overview.” https://bwo.life/bk/thesis_34.htm