Physiological Foundations of Behavior

Figs. 127, 128.—Diagrams of reduplicated amphibian legs, showing symmetry relations: (PR), primary leg; (P DU), posterior reduplication; (A DU), anterior reduplication; (MM, M2), planes of symmetry, ‘“‘mirror planes,” between the two reduplications and the primary leg; (M) is radial (R) since the radial borders of the two legs, (PR) and (P DU), which are symmetrical with respect to it face each other; (M2) is ulnar (U) since the ulnar borders of the two legs, (PR) and (A DU), face each other; (D), dorsal surface; (P), palmar surface. In°the sectional diagram, Fig. 128A the mirror planes are radial (/;) and ulnar (M2). In Fig. 128B the mirror planes are radiodorsal (M)) and ulno-palmar (M2). (From Harrison, ’21.)

tation of the various phenomena of regulatory reduplication than the speculations concerning a space lattice offered by Przibram. Even if the space lattice is present it is by no means clear how it can determine physiological polarity and symmetry and their modifications as we find them. _The determination of polarity by light in the eggs of certain algee has already been mentioned (pp. 58-61). In most other plants the polarity of the egg and embryo shows in general a very definite relation to its organismic environment, though practically nothing is known concerning the determining factors in particular cases.

In many animals a relation exists between the polarity of the egg and its ovarian environment in the gonad. For example in Chetopterus (I*. R. Lillie, 06), Sternaspis (Figs. 63, 66, pp. 98-9) and various other invertebrates in which the growing ovarian egg is attached to the parent body by one portion of its periphery, the apical pole arises on the exposed free surface, the basal pole at or near the region of attachment and the physiological gradient corresponding to this polarity appears. It seems reasonable to believe that the polarity of such eggs is determined by the differential in the ovarian environment. The nutritive substances which give rise to yolk probably, in some cases certainly (Fig. 63), enter the egg chiefly at the attached pole, while respiratory exchange must occur most readily through the free surface or some part of it. In the hydromedusa Phialidium the egg axis is apparently determined by a similar differential, the egg region most exposed to the sea water becoming apical, the opposite end basal (Child, ’21 a, p. 54, Fig. 1). Such differentials provide conditions for the origin of an axial gradient as a physiological reaction of the egg cell to its environment.

According to Boveri (’01 a, b) the attached pole of the sea urchin egg cell becomes apical, the free pole basal, but Jenkinson (’11) maintains that, the attached pole becomes basal, the free pole apical. In the absence of definite knowledge any discussion of possible factors _is useless. In some of the higher animals, also, the attached pole of the egg apparently becomes the apical pole, perhaps because in these forms the presence of a circulatory system or of impermeable egg membranes determines a more rapid respiratory exchange through the attached region than elsewhere. And finally, it is possible that in some eggs the gradient persists from earlier cell generations and so is inherited, as in many cases of fission. Such inheritance is, of

course, not Lamarckian but represents merely the persistence through cell division of a regional cytoplasmic differential. Concerning the origin of symmetry in eggs and embryos, we know even less than concerning the origin of polarity. In the frog the point of entrance or direction of movement of the spermatozcdn is commonly believed to be a factor in determining bilaterality (Roux, ’85. ’87, °95, Bd. II), but Jenkinson (’07, ’09) maintains that various other factors may also be concerned. Brachet (11) finds, however, that in eggs induced to develop parthenogenetically by puncture there is no relation between bilaterality and the meridian of puncture, and concludes from this fact that the egg possesses a primary bilateral symmetry which is labile and subject to alteration by entrance of the spermatozodn. Whatever its origin may be, the regional symmetry differential is physiologically distinguishable in a quantitative way in early developmental stages (Bellamy, 719). In the echinoderms there is no definite evidence of bilaterality until after development has begun, but it is distinguishable physiologically as a gradation before it is distinguishable structurally (Child, ’16 a, Jenkinson, 711). In the annelids and mollusks bilaterality is both morphologically and physiologically distinguishable in the first cleavage and in many insects it is indicated even before fertilization by the shape of the egg.

As regards the origin of bilateral symmetry in eggs and embryos various possibilities exist. It may in some cases be determined by ovarian conditions, in others by conditions connected with maturation or fertilization, in still others by conditions arising later in development. It is also possible that in some cases bilaterality, like polarity, may persist from earlier cell generations and so be inherited by the individual, but this again is not Lamarckian inheritance. According to Bartelmez (’12), for example, bilaterality is distinguishable in very early ovarian stages of the pigeon’s egg and is presumably hereditary.

It is probable that the theories of polarity and symmetry most widely current are responsible in part for the state of our knowledge concerning the origin of polarity and symmetry in embryonic development. As long as these characteristics of organisms are regarded as matters of molecular structure and orientation, the problem of their origin is in the same category as other problems of molecular structure and orientation. Experimental possibilities are limited to the modification of polarity and symmetry by modification of molecular structure and orientation, and since we are far from any definite

_ knowledge of the molecular structure concerned or of its changes, the whole matter has only a speculative interest. Actually, however, there is no real evidence in support of these theories of polarity and symmetry and a good deal against them. They are essentially speculative hypotheses which have been advanced, chiefly by morphologists, in the absence of experimental physiological data. When we take the facts as they stand, viz., that polarity and symmetry appear primarily as non-specific molar or regional physiological differences which can be determined, altered or obliterated by non-specific environmental action, and that we cannot find, by the methods of optics or otherwise, any positive indication of a molecular or static polarity or symmetry in protoplasm, it is evident at least that such hypotheses have no adequate basis at present. As a matter of fact, the conception of the physiological gradient as the primary factor in polarity and symmetry is the only theory which is based on actual evidence from experiment.

In an earlier chapter (pp. 57-61) it was noted that surface-interior pattern represents a more generalized relation to environment than axiate pattern. In this pattern the only fixed differential is between surface and interior, consequently the relation to environment and the behavior of surface-interior organisms are less specialized than in axiate organisms. So far as different regions of the surface are concerned, the surface-interior organism possesses the highest degree of versatility, 7. e., any reaction may be performed by any region, the environmental factors determining which region shall react in a particular case.

While it is probable that some cells and perhaps many micro6érganisms possess only surface-interior pattern, at least during some stages of their life history, we find surface-interior, together with axiate pattern in most forms. Surface-interior pattern does not disappear when axiate pattern arises, but exists in all organisms. Every exposed surface, every epithelium, every cell, whether in contact with inorganic environment or with other cells, possesses a surface-interior pattern. In epithelial cells this pattern usually appears as an axis in relation to the surface-interior differential (see Figs. 61, 62), 7. e., the pattern is axiate as regards the cell, but surface-interior as regards the epithelium as a whole.

That this kind of pattern is directly related in some way to a differential of some sort between surface and interior or between a more exterior and a more interior region, cannot be doubted. We often see the pattern in its simpler forms appear as the direct consequence of exposure of protoplasm. Except in the more highly specialized protozoa in which the ectoplasm has attained a more or less definite morphological form, isolated pieces of cells usually become more or less spherical in form and if they contain nuclei, behave in all respects like whole cells. It is a familiar fact that the differences between ectoplasm and entoplasm in Ameba are determined by difference of exposure. Regions exposed to the external environment acquire the structure and behavior of ectoplasm and ectoplasm carried to the interior acquires the structure and behavior of entoplasm. Plasmolyzed or otherwise isolated portions of plant cells containing nuclei secrete cellulose over the exposed surface. In some forms of cell division morphological cell membranes or boundaries, ““cell-plates,”’ appear in the cytoplasm without any relation to exposure to non-protoplasmic environment. Such cases do not conflict in any way with the views advanced here. No one doubts that the localization and development of the cell plate or membrane is determined by certain physico-chemical factors associated with nuclei or cytoplasmic regions or both. In short the environment in such cases is intra-protoplasmic or intercellular.

Multicellular masses develop, at least superficially, some degree of epithelial pattern and arrangement, when brought into a definite differential to environment, whether intra- or extra-organismic. Karly developmental stages of animals usually show this epithelial pattern in some form (Fig. 45) and in later stages it appears on both external and internal surfaces (Fig. 46). In plants also cell layers differentially exposed show a surface-interior pattern related to the differential exposure.

Most of us do not hesitate to conclude that in each individual case of surface-interior pattern the differential exposure is a factor directly concerned. The particular kind of surface-interior pattern which develops in any given case is of course determined first of all by the hereditary constitution of the protoplasm, but differential exposure is necessary in each case for the appearance or realization of the pattern. The rédle which we assign to the environmental factor in these cases differs of course according to our conception of organismic pattern, but it seems evident that surface-interior pattern does not arise de novo, independently of environmental factors.

The question whether the cell originated as a surface-interior pattern has been touched upon elsewhere (Child, ’ 21 a, pp. 23, 60-62). - According to this conception, the localization and differentiation of the nucleus as a definite organ has been determined by conditions characteristic of the interior of a mass of protoplasm. Such a conception of nuclear origin does not conflict in any way with the rdéle assigned to the nucleus in inheritance. The persistence of the nucleus . from one cell generation to another merely means that the pattern which has been established is persistent or hereditary. Nevertheless, the growth of nuclear substance, even now is apparently possible only in the interior of a mass of protoplasm. In the case of the spermatozoén for example, life is narrowly limited (Cohn, 718) and the intake of nutrition and synthesis of new nuclear substance apparently does not occur unless the sperm head reaches the interior of a mass of protoplasm. The change in form of the sperm head induced by Loeb and Bancroft (712) outside the egg cytoplasm seems to be largely or wholly a matter of imbibition and swelling and there is no evidence that any synthesis of new protoplasm occurs. These and various other facts indicate very clearly that in the persistence and growth of nuclear substance in cells environmental factors, 7. e., “interior” conditions, as well as hereditary constitution, are concerned.

Taking the data of experiment and observation as they stand, polarity and symmetry appear primarily as physiological gradients. When the gradients are obliterated, polarity and symmetry disappear ~ and when new gradients are determined by the differential action of environmental factors, new polarities or symmetries, or both, appear. In many cases gradients once established persist through agamic reproduction and perhaps also through gametic reproduction and so are inherited by the new individuals, but the only logical conclusion in the light of all the facts seems to be that even in such cases the gradients in the first instance must arise through differential action of external factors. And finally, if we accept the conclusions of the geneticists and cytologists, the mechanism of heredity does not afford any basis for such regional molar differences as the physiological gradients. To refer again to Morgan’s statement: “Each cell inherits the whole germ plasm” (Morgan, ’19, p. 241): granting that this statement is correct, it is evident that the factors which determine that different cells behave differently in development cannot be sought in the germ plasm alone, but must consist in some external differential which brings about the realization of certain of the hereditary potentialities in one cell or cell group and others in other cells or groups. Jn his earlier

statement Morgan suggests the existence of regional cytoplasmic differences, but makes no attempt to account for the origin of such differences (see pp. 22-23). The physiological gradients, originating in external differentials provide an adequate physiological basis for such differences, and if we accept Morgan’s statements, the differences must arise in the final analysis in relation to environmental factors. The hereditary constitution of each protoplasm reacting to the external differential gives first the physiological gradient and then, on the basis of this, development and differentiation.

In fact not only the data of experiment and observation in the field of developmental physiology, but current theories of heredity force us to the same conclusion, viz., that environmental factors are concerned in determining the order and the “wholeness” which constitute the individual. The hereditary potentialities may be in large measure or In many cases wholly independent of each other, as the particulate theory of heredity maintains, but the individual, in each cell of which only certain of the potentialities inherent in it are realized in development, is none the less physiologically a whole. There is in fact no conflict between the particulate conception of heredity and the conception of the organism as a whole, unless we attempt to derive the wholeness from the particulate character of the hereditary potentialities. Such an attempt can lead only to the dilemma in which Morgan apparently finds himself (Morgan, ’19, pp. 241-246). In the light of present biological knowledge the only possible conclusion seems to be that the individual organism represents in each case the behavior of the hereditary potentalities of the “germ plasm” in relation to certain environmental factors.

Even this conception, however, does not involve the conclusion that heredity has nothing to do with the origin and development of new gradients. In the first place, the hereditary potentialities of excitation and reaction must be present, and since particular protoplasms are so constituted that they are more sensitive to certain environmental factors than to others, the gradients of polarity and symmetry may be determined by different factors in different protoplasms.

Again, the hereditary mechanisms may play a part in determining the particular factors to which the reproductive element is exposed. Admitting for the sake of argument that the differential exposure of ovarian eggs of various species determines their polarity (see pp. 98— 100), it is evident that the complex of metabolic, morphological and physiological conditions which lead up to and determine the differential exposure and the factors to which the egg is exposed,

depend primarily upon the hereditary potentialities of the protoplasm from which the parent body, including the ovary, develops. In still another way heredity is unquestionably concerned in the origin of new gradients in protoplasms. The differential action of the external factor merely initiates the physiological gradient. The final results depend upon the hereditary constitution. Some protoplasms, e. g.. Ameba, are incapable of maintaining for any great length of time the differential externally determined. In many other protoplasms, when once established it persists throughout life or even through reproduction. Moreover, the development of the gradient beyond a certain stage is undoubtedly rather a matter of protoplasmic constitution than of the external factor. When the action of the external factor has once determined a certain degree of metabolic and structural differential, we see that in most protoplasms the gradient may persist and determine a definite course of development and differentiation quite independently of further differential action of the external factor. . In the alga Fucus, as we have seen, only a few hours’ exposure to light of a certain intensity is necessary to determine a gradient which becomes the axis of the plant and plays a large part in determining its form and the relations of its parts. In short, heredity is a factor in determining the special characteristics of each physiological gradient in each particular species, such, for example, as its length at each stage of development, its slope or steepness, 7. e., the curve of each of its component factors, oxidation, colloid dispersion, water-content, etc. It follows, of course, that heredity is concerned in processes of budding, fission, etc., although the gradient is the physiological factor directly concerned in determining the localization of the bud and the time of its appearance, and even though each new bud develops a new polarity, as in axiate plants, hydroids, ete.

This relation between heredity and the physiological gradients has always been regarded as a fundamental, though self-evident aspect of the gradient conception and it has been something of a surprise to find critics of the conception apparently believing that it was an attempt to interpret, not only individual development, but the differences between different species in terms of differences in metabolic rate. The only question which remains for consideration in this connection is whether protoplasm can autonomously give rise to a gradient and so autonomously determine its own polarity or symmetry or both. Is an autonomous molar or regional segregation or stratification of substances possible in such a way as to give rise to the graded quanti-

tative differences characteristic of physiological axes? In various publications Morgan has postulated a stratification of materials as a feature of polarity (e. g., Morgan, ’07, Chap. XVII), but though he has suggested that such stratification results in some way from tension, it is not entirely clear whether he regards polarity as primarily autonomous or as originating under the influence of external factors. Loeb (’16 b) seems at times to regard a regional segregation of materials as occurring autonomously, though he also conceives it as occurring under the influence of external factors such as gravity.

It has been stated by various authors that the visible structure of animal eggs altered by centrifuging shows more or less tendency to return to normal (e. g., Gurwitsch, ’05, Conklin, 710). This might perhaps be regarded as indicating a capacity for autonomous stratification or arrangement, but the fact that centrifuging does not alter the polarity of the egg (Lillie, 708, Conklin, ’10) indicates that a differential of some sort still exists in the original axis, and this differential, whether it be represented by an elastic structure as Conklin has suggested, or by a physiological gradient, is undoubtedly the factor which determines the restoration of the normal visible structure. In any case the granules which are displaced by centrifuging are not essential factors in polarity, for that may persist unchanged, whatever their location. It may be said that at present there is no evidence to indicate that a regional stratification of materials, a physiological gradient or any sort of physiological polarity or symmetry of organismic magnitude can originate autonomously in protoplasm. The protoplasm of an axiate organism, such as the hydroid, from which polarity and symmetry have been experimentally obliterated, may remain alive for a long time without showing any indication of a reappearance of polarity or symmetry (pp. 83, 123). In such cases the hereditary potentiality of axiate organization is still present, for if we expose the apolar mass to a new external differential, a new polarity, a new physiological gradient, arises and development once more goes on.

In every such case, and, I believe, in all cases, the external differential determines that an actual physiological gradient shall arise, but the specific constitution of the protoplasm, in other words, heredity, is always a factor in determining the particular characteristics of the gradient in that protoplasm. If we admit that polarity and symmetry are fundamentally such physiological gradients, this distinction between the realizing factor and the hereditary factor is inevitable and the necessity for it self-evident. Nevertheless, certain critics

of the gradient conception have maintained that features so definite and constant as polarity and symmetry in organisms could not be of external origin. Such a position results simply from misapprehension. Even though a gradient may originate in a differential excitation, the constitution of the protoplasm in which it arises makes it as definite and constant a thing as that constitution itself. That is to Say, given certain environmental conditions, a physiological gradient in a particular protoplasm will possess certain definite and constant characteristics.

The existence of a relation of dominance and subordination between parts in plants has long been recognized by botanists, though the physiological basis of this relation has been obscure. While under ordinary conditions the dominance of the growing tip over other regions and parts of the axis is the most conspicuous and characteristic relation of this sort in axiate plants, the growing tip is not the only region which dominates or controls other parts. Apparently any region of sufficiently rapid respiratory exchange may dominate a less active region. This relation of dominance and subordination and the physiological isolation of parts from the influence of the dominant region (see pp. 62, 157) are the primary factors in determining the order and arrangement of branches and other parts in axiate plants.

Attention has been called elsewhere (Child, 15 b, Chaps. IX, 2.6. "15 c, Chaps. IV, V, ’21 a, Chap. V) to the fact that this relation of dominance and subordination is characteristic not only of axiate plants, but also of axiate animals, and not only in development. but in the special functional relations resulting from development. I have endeavored to show further (Child, ’21 a) that this relation of dominance and subordination is primarily a dynamic feature of the — physiological gradient and so of the physiological axis, that it is fundamentally excitatory and transmissive in character, and that it represents the physiological basis of the relation of dominance and subordination which we find in the reflex are as the physiological unit of the nervous system (see Herrick, ’24, Chaps. AVI VEL

It was pointed out in the earlier discussions of the subject that the high end of a physiological gradient differs from other levels as a region of excitation differs from an unexcited or a less excited region and the control or dominance of the high end over other levels depends primarily upon this fact. From this viewpoint there is physiological continuity between the simple physiological gradient and the reflex arc. It was mentioned in Chapter IV above that the relation of dominance and subordination is a general characteristic of organ-

ismic pattern and we have now to consider the actual working of the relation in nature and experiment. It has also been pointed out (Child, ’15 c, Chaps. IV, V) that physiological isolation of parts may occur, 7. e., a part may be isolated from the controlling or determining action of a dominant region, while still in physical continuity with that region. Some data have _ been presented to show that physiological isolation can be experimentally controlled and modified, that it plays a fundamental réle in regressive development and physiological rejuvenescence of parts and in fission, budding and many other reproductive processes. These earlier considerations make it possible to review the chief facts and conclusions rather briefly in the present chapter, but some space is devoted to discussion of certain aspects of the subject not previously considered.

The existence of this relation of dominance and subordination in axiate plants and its significance for the development of buds and branches and so for the form of the plant as a whole, has long been familiar to botanists. In general, the apical or chief vegetative tip of an axis dominates the axis and inhibits, retards or determines the course of development of other buds or growing tips along the axis. In the absence of the chief vegetative tip the tip of the most nearly apical bud or buds dominates all lower levels, in the absence of this the next lower bud, and so on toward the basal end of the axis.