Physiological Foundations of Behavior

Organ gradients — The existence of a physiological gradient or gradients in various axiate organs of many organisms has been demonstrated by one method or another, or by several different methods. Such gradients have been observed in various specialized reproductive axes and in the “hairs” of algee, in the larger flagellum of Noctiluca, in the tentacles of various coelenterates, in the plate rows of ctenophores, the growing arms of echinoderm larve, the branchise and sensory tentacles of various annelids, the growing tail of ascidian and amphibian tadpoles. Dr. Hyman has found that the embryonic heart of fish and chick represents a physiological gradient with high region at the sinus end. Tashiro has found a gradient in CO, production in various nerves, and MacArthur and Jones a gradient in respiration rate in the nervous system as a whole. The most extensive work on the physiological gradient of any organ is that of Alvarez and his coworkers on the alimentary tract. They have found in the small intestine, for example, corresponding gradients in irritability, latent period, tone, rhythm, conduction, suscepti-

bility to various drugs, etc., and their work on stomach and colon shows the existence of similar gradients in those organs. Penfield has suggested the existence of such a gradient in the ureter.! The different lines of evidence indicating or demonstrating the existence of physiological gradients are in remarkable agreement. They force us to the conclusion that physiological axes in the simplest terms thus far discovered, are gradients in physiological condition in the protoplasm, or some part of the protoplasm of a single cell or a cell mass, involving the fundamental metabolic reactions, as well as physical condition of the protoplasmic substratum. Moreover, the various methods indicate that the physiological differences at different levels of an axial gradient are primarily quantitative, rather than qualitative differences, 7. e., they are first of all differences in rate, amount, degree, rather than differences in kind, though , qualitative differences may appear secondarily. Except as regards the apparently specific susceptibilities of certain tissues or organs to certain agents in the later developmental stages and mature forms, particularly of the higher animals, the different methods agree in their evidence concerning position, direction and course of the physiological gradients. That is to say, the differences in structure, rate of development, susceptibility, rate of penetration, rate and amount of reduction or of oxidation, electric potential, and, in cases where they have been determined, oxygen consumption and CO, production, all indicate the same regions of the organism as high and low ends respectively of a particular gradient and the regions between as intermediate in condition.

1 Most of the published data on organ gradients may be found in the following references: on the hairs of alge, Child, 17 a; ccelenterate tentacles, Child, ’19b, 21d, Child and Hyman, ’19, Hyman, ’20 b; ctenophore plate row, Childs 7c; arms of echinoderm larvee, Child, ’16 d, Galigher ’21; nerve fibers, Tashiro, 714, 15 a, b, 717; nervous system, C. J. MacArthur and O. C. Jones, ’17; alimentary tract, Alvarez, ’14, ’15a, b, 16a, b, 17a, b, 18a, b, ¢, ’22; Alvarez and Mahony, 21, Alvarez and Starkweather, 718 a, b, ¢, 719, Alvarez and Taylor, 717 a, b, Taylor and Alvarez, ’17; ureter, Penfield, ’20 and references given. The physiological literature on the vertebrate heart indicates very clearly the existence of a gradient of some sort in this organ, with the sinus end as pacemaker. Hyman (’21) has demostrated a susceptibilty gradient in the embryonic fish heart and Mr. Jee. Gowanloch has been able to alter and obliterate the gradient experimentally through differential susceptibility and so to alter the course and direction of the beat (Gowanloch 723). Other data on branchie and sensory tentacles of polychetes, tails of ascidian and amphibian tadpoles and various other organs are

The relation of the gradients to the various organs and parts of the developing organism is also definite. The apical region or head arises from the high end of the major, or polar gradient, and the other organs at different levels of the gradient. In the axes of radiate parts the relation between gradient and localization is also definite and characteristic for the species. The tip of the ccelenterate tentacle, for example, represents the high end of the tentacle gradient. In most bilateral invertebrates, e. g., turbellaria, annelids, arthropods, the median ventral region, and in the vertebrates the median dorsal region arises from the high region of the bilateral gradients. In the various organs, appendages, etc., of bilateral forms, the symmetries or asymmetries are associated, at least during early development, with physiological gradients.

The evidence indicates further that the chief gradients appear primarily in the superficial regions of the cell or multicellular body. In many of the simpler organisms, e. g-, many protozoa (Child, ’14 b) and plant cells, they are present throughout life only in the ectoplasm or the superficial regions. On the other hand, in more highly specialized organisms with definitely localized, axiate internal organs, gradients are also present in these organs, so far as examined, though these gradients may differ in position and direction from those of superficial regions. As regards persistence or modification of gradients during the life history it has been found that the primary gradient or gradients may persist in many of the simpler organisms throughout life or even through various reproductive processes, but complications and modifications of various sorts may also occur (Child, ’15 b, pp. 54, 60, ’16, "17, 721 a; p.. 26).. The original polar gradient may be obliterated or reversed, as in the development of the hydroid from the planula (Figs. 22-24). In the process of budding in plants, hydroids and many other forms a new gradient appears in each new bud (Child, ’19 b). In segmented animals, so far as examined, the lower end of the polar gradient becomes secondarily a region of high metabolic rate and from this new segments arise (Hyman, ’16 a, ’21, Hyman and Galigher, ’21, Child, ’17, Bellamy, 19). From the stage at which this posterior high region appears to the completion of segment development, or in some forms throughout life, the axis of the segmented animal represents a double gradient, both ends of the body being high ends with the low region at an intermediate level. The data obtained by Morgan and Dimon (04) on electrical polarity in the earthworm show the double character of the axis, but their significance was not apparent when

they were published. Again the origin of new radial axes about a circumference, such as the tentacles of a ccelenterate, the arms of a starfish, is associated with the origin of new gradients. Similarly the appendages of bilateral forms represent new gradients in more or less specialized tissues. Except in organisms with indefinite or temporary axiation, such as Ameba, the changes in the gradients during development of the individual, whatever they may be, occur in an orderly and definite manner and sequence for each form. In many cases we are able to analyze to some extent the physiological conditions associated with the origin of new gradients and even to determine their origin experimentally, but in other cases we know as yet little or nothing of the conditions concerned.

The further investigation proceeds along these lines, the more certain it becomes that physiological axes in their simplest known terms represent quantitative gradations or differentials in physiological condition. This means merely that the simplest sort of physiological axis first becomes distinguishable as such a gradient. It does not necessarily mean that the axis always returns to its primary condition with each reproduction, nor does it mean that every individual axis is nothing more than such a gradient. It is possible that in some cases qualitative differences may exist along an axis from the beginning of that particular axis, but the facts indicate that the quantitative differences are the fundamental factors in axiate pattern.

In pure surface-interior pattern the only differential is from the surface inward, all parts of the surface being alike (Chap. IV, p. 57, also Child, ’21 a, pp. 23, 60-62). The cell itself is apparently a further development of such pattern, with nucleus and cytoplasm as the fundamental qualitative differéntiations. Obviously interior rather than surface conditions are necessary for the formation of nuclear substance. There is no evidence to indicate that the cell is physiologically anything more than this, though those who regard polarity as a universal property of protoplasm must of course hold that all cells are fundamentally axiate.

Whether surface-interior pattern consists primarily in a quantitative physiological gradient like that of axiate pattern, but extending from all points of the surface inward, it is impossible to determine. External agents necessarily act from the surface inward, so that we cannot determine whether a susceptibility gradient exists. Direct investigation of the interior of such an organism is possible only by exposing it, 7. e., by making some part of it a surface. But the existence of physiological gradients of some sort from the surface inward in surface-interior pattern can scarcely be doubted. Ex- citation occurs primarily at the surface and if-it is transmitted inward at all, is very probably transmitted with a steep decrement. Oxygen and nutritive substances enter through the surface and waste products must pass out through it, so the possibility of various concentration gradients between surface and interior exists. It seems probable that the surface-interior pattern is fundamentally like the axiate pattern in so far as it originates in quantitative differences in physiological condition, rather than in qualitative differences in constitution.

The presence of axiate pattern does not mean the disappearanc2 or obliteration of surface-interior pattern. Surface-interior pattern appears in every organism, not only with relation to the external surfaces, but also to the internal surfaces. In fact, surface-interior pattern exists wherever living protoplasm is exposed to a non-protoplasmic environment, 7. e., wherever a surface occurs, and in many cases also in relation to two protoplasmic surfaces in contact, e. q-, two cells. In general, organisms represent patterns ranging from pure surface-interior patterns through various combinations of surface-interior, polar, radiate and bilateral patterns.

According to the Roux-Weismann theory of qualitative nuclear division, development is itself a distributing or sorting process in which the various determinants are assigned to their proper places. The localization of parts or characters results from this process and the differentiation of the cells is a matter of the determinants which they contain. As already noted in Chapter III, the data of observation and experiment have forced the abandonment of this theory, and it is now generally admitted that no such distributive process occurs in development. Morgan’s assertion that each cell inherits the whole germ plasm expresses current views and indicates the change in our conceptions of heredity and development.

Experimental investigation has shown beyond question that physiological polarity and symmetry are factors in the localization of parts in individual development. If polarity and symmetry are primarily quantitative gradients in physiological condition, as maintained in the preceding chapter, it is necessary to show, first, that such gradients can serve as the basis for localization and differentiation of parts, and, second, that localization and differentiation can be experimentally controlled and altered by controlling or altering the gradients.

The general question how a quantitative physiological gradient can serve as the basis or starting point for the orderly and progressive complication and qualitative differentiation of parts which takes place in the development of the individual is of course important. The experimental data show conclusively that localization and differentiation do occur in relation to physiological gradients, but with our present fragmentary knowledge of protoplasm as a physico-chemical system we can do little more than point out probable or possible ways in which this may occur.

Localization in the physiological sense means the determination of the locus in protoplasm of some particular process, change or condition, commonly the specialization or differentiation of some organ or part. Differentiation, as the term is usually employed, is the appearance of directly perceptible, 7. e., morphological differences in structure and constitution in different parts or regions of an organism. In general differentiation may be described as the appearance in protoplasms of substances and structures of molar, 7. e., morphological magnitude, which were not originally present. Localization and physiological specialization of parts may occur without visible differentiation, 7. e., parts specialized in different directions are not necessarily different in gross morphological structure. Differentiation in the sense employed here is primarily a result or incident of certain degrees of specialization. It is not necessary to suppose that all specializations and differentiations involve qualitatively different chemical reactions or substances, but certainly many of them do and the essential question is how or whether such qualitative differences can arise in relation to the differences in condition at different levels of physiological gradient.

Differences in the velocity of a chemical reaction do not of themselves give rise to qualitatively different products, but in the protoplasmic system the different levels of a physiological gradient represent not merely differences in velocity of a single reaction, but differences in many factors, physical and chemical, e. g., colloid dispersion, ionization, active mass of enzymes, concentration of various substances, probably water-content, and velocities of many different chemical reactions. These differences and probably others are all more or less closely interrelated and it is idle to speculate as to the primary factor: probably it is sometimes one, sometimes another, according to conditions. From what we know of non-protoplasmie physico-chemical systems it is evident that in so complex a system as protoplasm the possibilities of the origin of qualitative differences from differences originally quantitative are as numerous as could be desired (see Child, ’21a, pp. 90-97). Moreover, the fundamental chemical reactions in the protoplasmic system do not in general proceed to completion, or attain equilibrium, but in conquence of the entrance of nutritive substance and oxygen, the exit of CO, and other waste products, and the deposition of still others in inert form, they continue at a greater or less velocity throughout life. The relation between the rate of entrance of nutritive substance or of oxygen and the velocity of the oxidative reactions may be a factor in determining structural differentiation. If, for example, nutritive materia] enters

-acell more rapidly thanit is oxidized, it may accumulate as unoxidized or partially oxidized “reserves,”’ e. g., fat, or yolk in animal eggs, starch in plant cells, ete. The parts which become specialized and differentiated as regions or organs bearing reserves are in general regions of low oxidative metabolism. In the plant, for example, the reserves do not in general accumulate where they are formed, but in the regions of low respiratory activity. As Véchting (’78, ’84, ’87, ’00, etc.) and other botanists have shown, it is possible to determine the localization and differentiation of such reserve organs in regions of the plant where they never appear normally and to determine their dedifferentiation and redifferentiation into ordinary axes by changes

Frias. 61, 62.— Pancreas cells of toad: Fig. 61, fully loaded and almost quiescent; Fig. 62, almost completely discharged after prolonged stimulation (from Child, ’15 b, after preparations loaned by R. R. Bensley). in conditions which are not specific but primarily quantitative. The plant responds to the quantitative change with qualitative effects. Similarly in the animal body a decrease in rate of functional oxidative metabolism may result in the accumulation of fat in various cells and increase in metabolic rate in its disappearance. There is no indication that the changes in metabolism in such cases are anything more than quantitative, but the results are qualitative.

Again, the loading of the gland cell accompanied by very definite and conspicuous structural differentiation (Fig. 61) occurs in the presence of nutritive substance when the cell is unstimulated and the oxidation rate is low but when the cell is stimulated and the rate of oxidation increases “discharge”’ occurs and the differentiation of the loaded stage disappears (Fig. 62). Many differentiations are undoubtedly very similar in character to these, though often more stable and less readily reversible.

In many animal eggs yolk accumulates during the ovarian growth period and a gradient in rate and time of deposition or in amount of yolk is often visible. This gradient corresponds to the physiological gradient in the egg, the yolk being deposited earlier, more rapidly, or in larger amount at the lower levels of the gradient. In eggs in which nutrition enters chiefly or wholly at one point, this may be Fic. 63.—Half grown ovarian egg of Sternaspis scutata, showing the peduncle with vascular loop extending into cytoplasm of egg. The attached pole becomes basal, the free pole, apical (from Child, ’15 b).

Fie. 64.—Section of ovarian egg of frog, showing yolk gradient; (p), pedicle by which egg is attached (from preparation loaned by A. W. Bellamy). the chief factor in determining that region of the egg as the low end of the gradient. The ovarian egg of Sternaspis, for example, is attached by one pole, and a vascular loop enters the egg through the pedicle of attachment (Fig. 63). Apparently nutritive substance enters chiefly through this pole, which becomes the basal entodermal pole of the

egg. On the other hand, the free end of the egg is more exposed to the fluid of the ovarian cavity and this exposure may conceivably be more effective than the vascular loop in providing for respiratory exchange and determining the pole of the egg as apical. But whether the gradient of the Sternaspis egg is determined chiefly by one or by both of these Fic. 65.—Semi-diagrammatic view of ovarian differentials, the primary egg of frog, showing a characteristic arrangement difference is quantitative of arteries and veins: (a), artery; (v), vein; (p), : ; % 5 pedicle (from Bellamy, ’19). rather than specific. Nu-

tritive substance undoubtedly reaches all parts and respiratory exchange undoubtedly occurs in all parts. In various other invertebrate eggs the relations between _ polarity and attachment are very similar to those in Sternaspis, the free pole becoming apical and the high end of the gradient, the attached pole basal and the low end. In the frog’s ege a yolk gradient also appears (Fig. 64) although the whole egg in later stages is inclosed in a capil- 66

lary network (Fig. Fic. 66.—Axial section of full grown egg - sani! ind 1 oscak pergamentaceus before maturation, indicating the struc ura 65). The visible differentiation of the cytoplasm in relation to the axis (from structure of the F, R. Lillie, ’06). unfertilized egg of the annelid Chetopterus (Fig. 66) as described by F. R. Lillie (06) suggests that here more or less morphological differentiation has occurred in relation to the physiological gradient which is present in this egg (Child, ’17). But Lillie (08) has shown that this visible differentiation is not the fundamental factor in the polarity of the egg, since by means of centrifugal force the various spherules can be carried into different regions of the egg without altering essentially the course of development.

The point which I wish to emphasize in this connection is that an apico-basal physiological gradient is present in all animal eggs examined, including various ccelenterates, echinoderms, annelids, mollusks, amphibia — whether distinct visible axial differentiation is present or not. Apparently different eggs develop very different degrees of specialization in relation to the gradient before fertilization. It is sufficiently evident, however, that the gradient may serve as the basis or starting point of differentiation. If the gradient is present before yolk formation begins, yolk may accumulate more rapidly or in greater amount at the lower levels. On the other hand, the accumulation of yolk about the region of entrance of nutrition may perhaps itself originate a gradient. The differences determined by such a gradient may lead not only to qualitative differences in reaction at different levels, but they may determine differences in rate of cell division, turgor, viscosity, etc., which become factors in determining further relations of the parts concerned. There is every reason to believe that gastrulation with all its consequences results from differences originally quantitative, rather than qualitative.

In fact, localization and differentiation during embryonic development apparently consist in large measure of the origin of specific qualitative from non-specific quantitative differences, although it is often a matter of viewpoint where we shall draw the line between these two categories. The data of developmental physiology and differentiation considered without theoretical] preconception afford the strongest support to this conclusion, and there seems to be no other which will account for all the facts in physico-chemical terms.

Both synthesis and breakdown of various substances are going on in protoplasm. If for any reason the rate of synthesis of certain molecules exceeds the rate of decomposition, or if the molecules once formed are relatively stable under the conditions in the cell, an accumulation of these molecules will occur in the cell and it will become qualitatively, and if the process goes far enough, morphologically _ different from other cells in which decomposition keeps pace with

Apparently just such differences as these occur at different levels of a physiological gradient. The different regions of a cell or the different cells along a gradient become qualitatively different by the appearance in their cytoplasm of different substances. It seems to be true, moreover, that cells which represent lower levels of a gradient tend in general to accumulate non-protoplasmic substances to a larger extent than those of higher levels. This is certainly true in the plant, in which cells of the growing tip and higher levels consist wholly or largely of granular “embryonic” protoplasm, while at lower levels fluid vacuoles and other inclosures appear in increasing volume. In animals also differences of this sort appear to some extent. The nerve cells, for example, which represent the high ends of the chief axial Eaticuts, seem to retain a “protoplasmic”’ structure, to a higher degree than many other cells of lower levels in the eradients: in these latter we find various sorts of inclosures and non-protoplasmic structures and in some the original protoplasmic structure has largely disappeared. Skeletal and supporting tissues generally develop from cells representing relatively low levels of a gradient.

As I have pointed out repeatedly, this conception of the significance of the gradient in localization and differentiation is not necessarily in conflict with current views, concerning the importance of chemical or transportative correlation, hormones, etc., but may merely supplement them. It is evident that as soon as qualitative regional differences do arise, or probably when the quantitative differences become sufficiently great, the basis for transportative or chemical correlation is established. In the case of the polar differentiation of the animal egg, for example, the yolk of the one hemisphere is largely used as nutrition in the growth of the other, 7. e., the polar differentiation becomes the basis of a transportative correlation between the two hemispheres of the embryo. This is a factor in determining further differences and so on until the final relations of ectoderm, mesoderm and entoderm and their products are established.

Since the physiological gradients are characteristic of physiological polarity and symmetry, at least in the earlier developmental stages and since localization and differentiation of parts occur along these gradients, a certain relation, real or apparent, must exist between the gradients and the developmental processes. As a matter of fact, we find the apical region or the head of the organism arising from the high end of the gradient and the other organs characteristic of each species arise in order at various levels of the gradient. These facts, however, are merely suggestive, not demonstrative. From them alone we might as readily conclude that the physiological gradients and the developmental phenomena are both determined by some more fundamental factor, as that localization and differentation are determined by and result from the gradients.

The experimental modification and obliteration of gradients, and the determination of new gradients afford more _ conclusive, though strictly speaking Fics. 67-69.—Lengths of hydranth primordia of stem pieces of Tubularia under various conditions: Fig. 67, under standard normal conditions; Fig. 68, under conditions which accelerate metabolic, e. g., high temperature, slight dilution of sea water rate; Fig. 69, under conditions which decrease metabolic rate, e. g., low temperature, inhibiting chemical agents. The primordia show four regions: (a), pretentacular or oral; (6), distal tentacular; (¢), intertentacular; (d), proximal tentacular (from Child, 15 ¢c). not absolutely demonstrative evidence. As I have pointed out elsewhere (Child, ’15b, pp. 128-142), altering the physiological condition either in the direction of depression or of excitation alters the length of the gradient. If the gradient is a factor in the localization of organs we may expect to alter their relative lengths or distances apart as we alter experimentally the length of the gradient, and this proves to be the case. In the regulatory development of the hydranth of Tubularia from pieces of stem the hydranth primordium is formed inside the perisare, the two sets of tentacles being represented by longitudinal ridges, as indicated in Figures 67-69. In pieces of like physiological condition, from a given level of the stem, and kept under the same environment the length of the primordium