Physiological Foundations of Behavior

direction of the axis, but the chief radii are equivalent and, as in the sea anemone, any one may become temporarily dominant over all other regions of the circumference. In the scyphomeduse Aure- Fic. 25.—Oral area of a critime being the pacemaker for all the noid, Antedon, with bases of other marginal organs and so determines arms: (m), mouth; (a), anus; : (c), ciliated grooves leading from the rhythm of contraction of the whole ae to mouth (from Hertwig, umbrella. In this physiological equiva-

ponent of ccelenterate pattern is somewhat similar to the Ameba pattern, in that different radii may be dominant at different times. bilateral pattern exist and with these are associated interesting differences in behavior. In the crinoid (Fig. 25), for example, the disc shows a certain degree of structural bilaterality, but there is complete or practically complete physiological equivalence of the different arm radii in reaction. In the free crinoids locomotion may occur in the direction of any radius. ‘

As regards the starfishes, both asteroids and ophiurids, most authors state that there is no preferential use of the rays related Fras. 26-30.—Diagrams showing different patterns with respect to use of arms in locomotion of ophiurid starfish, Ophiura brevispina. The arms most active in locomotion in each case are indicated by heavier lines and direction of locomotion by arrow (from Glaser, ’07). definitely to the structural bilaterality, but Cole (13) has found that in Asterias forbesti a certain ray is physiologically anterior. This ray corresponds to the anterior ambulacrum of the spatangoids in which structural bilaterality is more highly developed (see below). In general the preferential function of one arm as anterior appears to be absent or not very strongly marked in the starfishes and any one of the arms may precede in locomotion. The variability of behavior of

the arms in locomotion determines different results in different — individuals and in the same individual at different times.! The process of locomotion in the brittle star, as described by Glaser (’07), will serve as an illustration. The various ways in which the arms are used in locomotion are indicated in Figures 26-30 taken from Glaser’s account. Locomotion over a surface is accomplished in these animals by strokes of the arms in the direction opposite to that of advance. In the figures tie arms most actively concerned in these movements are distinguished by drawing as heavier lines. Various modifications of what Glaser calls Type I appear in Figures 26-28. In Figure 26 arm 2 is held somewhat stiffly forward, arms 1 and 3 make the active strokes and arms 4 and 5 are dragged behind. In Fig. 27 the tip of arm 2 bends from side to side and so assists in the movement of the body, and in Figure 28 arm 2 makes a stroke as effective as that of arms 1 and 3. In this type of locomotion the stroke of arms 1 and 3 may be synchronous or alternate and those of arm 2 may be always toward one side or may alternate from side to side. Figure 29 shows Glaser’s Type II in which two arms on each side deliver the stroke and one is dragged behind. In this method of locomotion the anterior pair may work synchronously and alternately with the posterior pair, or synchronous strokes of one member of each pair may alternate with strokes of the other two members. In short any combination of the strokes of the four arms appears feasible. In Type III (Fig. 30) all the arms deliver strokes, three on one side, two on the other. This may be regarded as a more extreme modification of Type I in the direction indicated in Figure

| NEN Sw tion may vary wt SANS — in details, such EQ AAS as the rate or Fic. 31.—A polar-bilateral echinoid, Brissopsis lyrifera, in locostrength of , motion (from Lankester’s Zodlogy, Part III, ’00, after Lovén). stroke of dif- : ferent arms or pairs or groups of arms. In the asteroid starfish locomotion is accomplished by means of the ambulacral feet, not by strokes of the 1 See Romanes, ’85; Preyer, ’86; Grave, ’00; Glaser, ’07; Jennings, ’07.

arms, but the versatility in the use of the arms is very similar to that in the ophiurid (Romanes, ’85; Jennings, ’07). The echinoids (sea urchins, sand-dollars, etc.) show various degrees of development of bilateral pattern. In the common sea urchins fixed bilaterality is limited to the structural features common to all echinoderms: in these forms the radii are equivalent so far as determined physiologically and bilaterality in behavior is temporary and shifting. _ The other groups of echinoids, the sand-dollars (Clypeastroidea) and the heart urchins (Spatangoidea) both represent secondary polar and bilateral modifications superimposed on the radiate pattern and involving in the latter group general body form, position of mouth and arms, development and function of different radii and direction of - locomotion (Fig. 31). In the holothurian the axiate-bilateral pattern is developed in a different way. The original apico-basal axis becomes the longitudinal axis of the animal with one end anterior, the other posterior and the five radii of the radial pattern are represented by the five longitudinal zones of the body wall, but more or less dorso-ventrality appears among these radii, one radius being

median ventral and an interradius oot ; or Be median dorsal (Fig. 32). In some Se? CK Cm, forms the ventral surface including nets 3 5 OF TS SOR three of the rows of ambulacral feet "Se Me = (the trivium) is distinctly flattened eR and the dorsal surface (bivium) is a ee without true ambulacral feet. Motion (en wee, in the holothurians is chiefly in the diie aa fo E rection of the longitudinal axis, but RY ae fe the definiteness and degree of fixity \< tye © of the dorsoventral differential differs Ve, pf, in different forms corresponding in Ra Ass general to the structural differential. \e" a This brief survey is sufficient to a fe show that the echinoderms present Se He ae

extremely interesting combinations a ect ne ES ; : . . - Fie. 32.—A holothurian, Pentacta and modifications of radiate and bifrondosa, ventral aspect, illustrating lateral patterns. While we know in another bilateral modification of pattern in echinoderms (from Kellogg, general something of the relations Oi atiee tert between these various modifications of pattern and the motor behavior, further comparative study The versatility of the starfish to which Jennings (’07) calls par- — ticular attention must result from the physiological equivalence of the rays ' and the fact that any one of two or three of them may bebecome dominant. With different physiological states of the different rays the reaction of the animal as a whole to the same external factor will differ, as regards the relations and functions of the different rays. This versatility is very evident in Glaser’s figures of the motor reactions of the brittle star (Figs. 26-30). All the radiate animals show more or less of this sort of versatility and the Ameba is perhaps the most versatile of all. With the increasing development of bilaterality in the echinoderms this versatility as regards the functional relations and combinations of different rays must more or less completely disappear and the processes of motor reaction become more definite and stereotyped in character.

Bilaterality involves physiological differentials in all three dimensions, 7. e., between median and lateral in opposite directions and between dorsal and ventral in one direction in addition to the longitudinal or polar differential (see pp. 36-40). These differentials are of course factors in increasing the definiteness of behavior patterns. In bilateral animals the only possiblity of physiological equivalence of motor organs is that of right and left, or in animals with reduplication of segments and appendages, that between the organs of neighboring segments. With respect to the use of such organs in reaction we often observe a versatility which, so far as it goes, is similar to that of radiate and surface-interior patterns. For example, in various animals organs of right and left sides may move alternately or synchronously according to conditions and in many arthropods movements of the appendages of successive segments may be combined in various ways in locomotion. But even the equivalence of right and left is by no means alway present. Many animals, bilateral in certain respects, show in other respects definite asymmetries. These may appear in behavior in the preferential use of organs of one side, as in righthandedness or lefthandedness, or general morphological asymmetry may be associated with special methods of locomotion.

In a bilateral pattern there is not only differentiation along the anterior and posterior axis, but differentiation of another sort in 1 According to Cole (’13) the rays are not completely equivalent in Asterias forbesti. _ two opposite directions between median and lateral, and still a third sort of differentiation dorsoventrally. Besides all this, differentiation between surface and interior occurs as in simpler organisms. Evidently the bilateral pattern makes possible more definite and more complex differentiation than the other patterns. In such a pattern each particular behavior mechanism is more definite and more fixed in structure and in relation than in the simpler patterns and must therefore possess a more definite and stereotyped function. This being the case it appears at first glance that the behavior of bilateral organisms must be less versatile and more rigid and stereotyped than that of other patterns. As regards the particular mechanism this is true, but as regards the behavior of the organism as a whole it is far from being true. The versatility of the Ameba results from the absence of fixed specialization of different regions. Similarly the versatility of the starfish with respect to the different rays depends on the fact that the rays are physiologically equivalent, although each may represent a complex system of more or less definite and specialized mechanisms. The behavior of each part of the surface of Ameba or each ray of the starfish varies from moment to moment according to external factors and physiological states.

The bilateral pattern, however, affords greater possibilities of specialization and differentiation of parts than the other patterns and in the higher animals the complexity of structure and function is very great. The more definitely structure and function become fixed the more does versatility in behavior consist in the combinations and relations of the different mechanisms, each of which is in itself highly specialized and relatively stable in character. This sort of versatility is particularly characteristic of the higher animals and man. The relation between a particular muscle of the arm or hand and the skeletal parts with which it is connected is definite and fixed. The versatility in function of the arm and hand results from the many possibilities of combination and relation of the different musculo-skeletal mechanisms. Moreover, versatility also appears in the processes of realization of these various possibilities. The specialization of different sense organs makes possible the combination of various sensory Impressions and different combinations and relations may determine different use of the behavior mechanisms. But the most important factor in the versatility and modifiability of behavior in the higher animals and man is the central nervous system. In certain parts of

the nervous system the path of an impulse is not fixedly predetermined but is determined by the combination of conditions, physiological states, excitations, etc., in the nervous system or in the parts concerned at the moment (Herrick, ’24, Chap. XIX). There is then in these regions of the nervous system a certain physiological equivalence of path which is similar in a general way to the physiological equivalence of the starfish arms, the medusa sense organs, or the different parts of the Ameba surface. The physiological equivalence of path in the nervous system results, not from simplicity and absence of specialization, but from the complexity of structural mechanism and physiological relation.

In the higher invertebrates, notably the insects, this physiological equivalence of path in the nervous system is apparently but little developed as compared with the higher vertebrates and the behavior patterns of the insects are relatively fixed and stereotyped, and largely “instinctive” in character. For the present, these differences are merely noted without discussion of possible reasons for them. There can be no doubt that they both represent in some way the working out of bilateral pattern in different protoplasms.

There can of course be no doubt that in general terms organismic pattern constitutes the basis on which the excito-motor behavior patterns of an organism are built up, in others words, organismic pattern is the general framework within which such behavior occurs. But behavior represents the results of action of external factors upon an organism and these may not only change from moment to moment, but some of their effects may persist for a time or permanently and modify further behavior. Because of this relation between behavior and external factors we find certain behavior patterns integrated on an organismic scale, but transcending the permanent organismic pattern of the individual in which they occur. In the ophiurid starfish, for example, the general pattern is radiate so far as the physiological equivalence of the rays is concerned, yet behavior patterns which are just as truly bilateral as the usual motor patterns of an insect appear in the correlated function of the arms (Figs. 26, 29). To my mind this bilateral behavior pattern is just as truly on organismic pattern as that of any bilateral animal. It differs from such pattern merely in that it is temporary and functional, and does not possess a permanent morphological basis. In such cases as this we can only conclude that the temporary bilateral pattern has been

_ superimposed on the permanent. radial pattern by the action of external factors. Because of their relation to present or preceding Fra. 33.—Spiral path of Paramecium (from Jennings, ’06 © Columbia University Press). external factors two pairs of arms are in such physiological state that they react similarly. Again certain Amebe may temporarily assume the limax form, 7. e., an axlate pattern. In the case of the plant Fucus the reaction to light differential determines a new axiate organismic pattern which is permanent for the individual plant (pp. 5861). In bilateral animals, on the other hand, we often see asymmetric behavior patterns in the different use of bilateral organs. <A_ bird may stand on one leg and draw the other up beneath it, use it to scratch its head, etc. Many such asymmetric behavior patterns may bring about permanent modifications of the organismic pattern of the individual. Learning to write with one hand involves the establishment of permanent differences in pattern between the two sides of the body, chiefly or wholly in the nervous system. On the other hand, the muscular and skeletal asymmetry of the blacksmith involve the

gross morphological features. In still other forms we find an excito-motor behavior pattern compensating asymmetry or other structure as in the spiral course of Fire. 34.—S piral path of rotifer, Diurella tigris, showing that animal continually swerves toward dorso-dextral side (from Jennings, ’03). Paramecium (Fig. 33) and of certain rotifers (Fig. 34) with revoution of the body about.a longitudinal axis (Jennings, ’06 and earlier papers).

While it is perfectly true that the existing organismic pattern constitutes the general physiological basis for the behavior of the individual, it is also true that such behavior modifies the original pattern, determines new patterns within it or superimposes new patterns upon it. These changes may be either temporary or may persist. So far at least as the individual is concerned, behavior is the originator of new organismic patterns, new integrations, which are only potentially given in the original pattern. In other words, it is behavior, 7. e., the individual experience, which determines the realization in the individual of the potentialities of the protoplasmic constitution and the organismic pattern already present. In the light of these facts the statement in an earlier chapter that the organism represents a behavior pattern in protoplasm acquires a still more definite significance. At any rate it is clear from what has been said in the present chapter that any consideration of the origin of organismic behavior must deal sooner or later with the questions of the nature and orgin of organismic pattern itself.

Physiological polarity and symmetry, either radial or bilateral, are commonly thought of as conditions of some sort underlying and determining the localization and arrangement of parts and organs of the individual, 7. e., its spatial, morphological pattern. Actually, however, polarity and symmetry have to do not only with morphological pattern but with physiological relations of parts. The relation of dominance and subordination to which attention has already been called (pp. 34-36) is very intimately associated with physiological polarity and symmetry. Again, differences in rate of cell-divison, growth and differentiation are associated with polarity and symmetry, and the relations between the behavior patterns of the individual and polarity and symmetry have been indicated in Chapter VI.

Polarity and symmetry are really only terms to designate the directions, the axes or radii, in which a spatial morphological and physiological order is perceptible. As already noted, the simplest conceivable organisms are spherically symmetrical forms with surfaceinterior pattern. In these the order appears along the radii of a sphere. Whether such organisms actually exist, we do not certainly know, though some of the microérganisms appear to be of this sort and the simple cell or protoplast is probably little or nothing more than a surface-interior organism.

Most organisms, however, are axiate, the axes constituting a sort of physiological codrdinate system to which the order is referable, and the questions of the nature and origin of physiological axes are obviously of fundamental importance for any physiological interpretation of organismic pattern. When we examine a physiological axis, or more strictly speaking an axial direction in an organism we find that it is indicated in many of the simpler organisms throughout life and in the earlier developmental stages of the more complex forms by gradations of various sorts, rather than by clear-cut qualitative differences. We find, for example, axial gradations in protoplasmic structure, in rate of growth and differentiation, in rate of, or capacity for, regeneration, and in various other physiological char-

acteristics. Some of these gradations in condition, for example, the apico-basal gradation in yolk in many animal eggs, the apico-basal or antero-posterior gradation in rate of cell division, growth and differentiation in many embryos, have long been familiar to biologists. In various publications Morgan (e. g., 05) has advanced the hypothesis of a gradation of materials as a factor in physiological polarity. The results of experimental alteration of cleavage of the Ascaris egg led Boveri (’10) to the conclusion that polarity must be a gradation of some sort.

During recent years investigation along various lines has accumulated a large body of evidence in support of the conclusion that physiological axes are primarily quantitative dynamic gradients in living protoplasms, that theyfrepresent primarily differences in physiological state, rather than in molecular structure. Such gradients involve differences in rate of the fundamental metabolic reactions as well as differences in physical state of the protoplasmic substratum. They have been called for convenience, axial, metabolic, or physiological gradients. So far as the facts go, they are the primary indications of the existence of axiate pattern and there is at present no evidence to indicate that axiate pattern can arise in any other way than as a gradient in physiological state. It is not necessary to assume that these gradients consist primarily in metabolic differences alone. Protoplasm is a system in which the chemical reactions of metabolism are so intimately associated with other factors, e. g., colloid dispersion, active mass of enzymes, permeability of limiting surfaces, electrolyte content and dissociation, water content, etc., that to distinguish one particular factor rather than another as primary is at present impossible,

The evidence demonstrating or indicating the existence of physiological axial gradients in organisms is varied and extensive and has been more or less fully presented in earlier publications. The evidence at hand eight years ago was briefly discussed in “ Individuality in Organisms,” 1915, but since that time further investigation, the results of which have been published in part, has added much to the evidence. More recent reviews of this evidence (Child, ’20 ¢, 21 a, Chap. IT) have presented the data at hand up to the beginning of 1920 with a practically complete bibliography of the subject. These earlier publications make it possible to limit the present section to a brief review of the’ evidence.

Structure and development.— In the first place the axial gradients, particularly the apico-basal or antero-posterior gradient, are often visible in the protoplasmic structure, density, vacuolation, or accumulation of yolk. Such structural gradations are very characteristic of eggs and early embryos, but in axiate plants they occur in relation to the growing tips of both stem and root at all stages of growth.! A gradation in rate of cell division, growth, morphogenesis and differentiation is also a characteristic feature of physiological axes in both animals and plants. In axiate animals the rate of these processes is primarily highest at the apical or anterior end and de-

creases basipetally or in the posterior direction. In bilateral invertebrates it also decreases from the median ventral region laterally and dorsally, in bilateral vertebrates from the median dorsal region laterally and ventrally. In axiate plants the growing tip continues its growth and division as long as it remains active as a growing tip. In relation to the growing tips of plant axes, gradations in cell size, rate of cell division, rate of growth of single cells and differentiation appear.®

Differences in rate of reconstitution at different levels of an axis are also of very general occurrence, the rate of reconstitution of apical or anterior parts decreasing basipetally or posteriorly. In pieces of Tubularia stem, for example, the time from section to emergence of the new hydranth at the distal cut end of the piece increases basipetally (Driesch, ’99; Morgan, ’01 b, ’05, ’06, ’08; Child, ’07 ¢; Hyman, ’20 b). At the proximal end of the piece, however, conditions are different because the proximal end is subordinate to the distal end if not too far distant from it and the development of the proximal hydranth may ,be delayed or inhibited in short pieces, while in long pieces it may occur much more rapidly. Lund (723) 4

2 For figures of developmental gradients in animals see Child, ’15¢, Figs. 10-17. 3 For figures of developmental gradients in plants see Child, ’15c, Figs. 18-22, 4In this paper Lund maintains that in Obelia there is no difference in rate of regeneration at different levels of the stem, but rather a difference in length of time between section and the initiation of regeneration. He regards regeneration as beginning only when the visible development of the hydranth begins and takes no account of the extensive cell activity which has long been known to begin soon after section in most cases. Larlier authors have preferred to regard this activity as part of the regenerative process since it is a normal and constant feature of formation of a new hydranth from a piece of stem. But whether we -accept Lund’s definition of regeneration or that of other authors, the data which

has found an axial difference in reconstitution period in Obelia similar to that in Tubularia. In my experiments on Planaria doroto- ~ cephala I have found similar, though slight differences in rate of development of the head at different levels, and Sivickis (23, p. 145) has noted a similar difference in P. lata. The relation between the character of reconstitution and body level has been extensively studied in Planaria and it is a well-established fact that the frequency of normal heads decreases from anterior levels posteriorly in each zooid (Behre, ’18; Child, ’11 a, b, 16 b, ’20 a; Sivickis, ’23). Susceptibility. — Differences in susceptibility to a very large number of external agents, probably to such agents in general, are characteristic features of the different levels of the physiological gradients.! In general the physiological gradients have been demonstrated as susceptibility gradients in various ways and by many different agents, physical and chemical. The investigation of susceptibility in its relation to the physiological gradients has brought to light certain general relations between susceptibility and physiological condition, or more specifically between susceptibility and rate of fundamental metabolic reaction.2 These relations are briefly as

he presents constitute very definite evidence for the existence of a physiological gradient in the stem of Obelia. It is to be expected that the more active cells of more apical stem levels will make the physiological changes (dedifferentiation, activation, or whatever we may call them) which precede the development of the hydranth more rapidly than the less active cells of lower levels. After they have attained the condition in which, hydranth development becomes possible, they are probably much alike and morphogenesis may go on at the same rate at all levels, though Lund’s curves indicate a difference in rate.

‘Papers concerned with the demonstration of the susceptibility gradients as gradients in survival time under the action of various agents are as follows: Child, 13 b, ’14a, b, 15a, ’16 a, b, ¢, 19 b, e, ’23.a, 23 d; Child and Hyman 19; Galigher, ’21; Hyman, AGS 717,206,791 ; Hyman and Galigher, ’21; MacArthur, ‘21. Further data on various plants, protozoa, ctenophores, hydrozoa, flatworms, echinoderms, annelids, fishes and amphibia, and the chick embryo are not yet published. In the other literature of the action of external agents on protoplasm numerous data indicating differential susceptibility along physiological axes appear, and it is of interest to note that various authors have observed axial