Principles of General Physiology

Further details as to the properties of this " secretin." as we called it, being unable to think of a better name, will be found in Chapter XXIV. The name itself has now come into general use and, whatever objection may be made to it, it must lie admitted that it lias the advantage of making no assertion as to the chemical nature of the substance, as to which we have little positive knowledge. The juice formed under the action of secretin appears to be identical with that formed during natural digestion; perhaps it may be rather more dilute; it contains trypsin in the inactive, zymogen form, amylase and lipase, together with alkaline salts. The pancreas can be caused to secrete continuously for many hours by repeated doses and, although in the later stages a somewhat more dilute juice may be obtained, it is a matter of considerable difficulty to induce signs of fatigue in the cells, so far as microscopic observation can detect them.

Atropine has no effect on the action of secretin, contrary to its action on secretion produced by stimulation of nerves. Secretin must act, therefore, on the cells directly, or, at all events, on some part of the cells beyond nerve terminations. How far this natural form of chemical stimulation of glands applies in general remains as yet uncertain. It is comparatively unimportant in the case of the salivary and sweat glands, but the work of Pavlov (1901, Lecture VII.) and of Edkins (1906) shows that the gastric juice is partly produced by the agency of

a chemical substance produced in the stomach itself by certain constituents of the food, and that this substance acts through the intermediation of the blood current (see page 372 below), although the gastric glands are also powerfully excited by fibres in the vagus nerve. Secretion of Bile is produced by the same acid extract of duodenum which excites the pancreas, but whether the same " secretin " is at work we cannot state. The liver can also be excited to secretion by injection into the blood of bilesalts ; in such a case, the concentration in the blood of the constituents of the secretion plays a part in determining the activity of the cells in transferring them from the blood to the duct. The blood supply of the liver (Heidenhain, 1880, pp. 259-268), both in respect of rate of flow and of pressure, affects the rate of secretion to a large extent.

The Succus Entericus is secreted in a particular section of the small intestine when trypsin is present in a part preceding it in the normal direction of the passage of food. The work of Pavlov (1901, p. 161 of the English edition) tends to show that this is a chemical mechanism. The Nervous Mechanism of Secretion. — The majority of glands, including those with internal secretion, are supplied with nerves by which they can be excited to action by reflexes from the central nervous system. Most of our knowledge is derived from the study of the salivary glands, owing to the comparative ease with which experimental work can be conducted on these organs. We will consider, in the first place, the special case of the submaxillary gland of the dog and afterwards apply the results to other glands. This submaxillary gland is supplied by two sets of nerve fibres, both of which play a part in the secretory mechanism. The first set, contained in the chorda tympani nerve, arise from the brain in the small-fibred portion of the facial nerve, corresponding to the intermediate nerve of Wrisberg in man, which leaves the mid-brain between the facial and auditory nerves (see Gaskell, 1889, p. 172).

These fibres may be called the cerebral supply ; the other set comes from the sympathetic system, a special outflow of nerves to the viscera, blood vessels, and similar structures ; to this system of nerves attention will be directed in a latec chapter. It is found that excitation of the chorda tympani nerve produces a copious watery secretion, while that of the sympathetic nerve produces a small quantity of a very thick saliva. Heidenhain (1868, p. 113) propounded the view that there are two different kinds of fibres concerned, one set with the secretion of water, together with the diffusible salts present in the blood, and the other set with the formation of the specific solid constituents of the secretion. On this ground he called the former " secretory," the latter " trophic," using this latter word in a rather special sense (1868, pp. 101-104, and 1880, p. 51). The two kinds of fibres are supposed to be present in both nerves, but in different relative amount, which varies according to the kind of animal. In the cat, for instance, Langley (1878) showed that the chorda and sympathetic nerves both give very much the same kind of saliva. The sympathetic nerve contains fibres which cause great constriction of the arterioles of the gland, while the chorda contains fibres which dilate them, so that it has been held (see Langley, 1898, p. 529) that the restricted blood supply is responsible for the relatively concentrated saliva produced by the "trophic" nerve fibres, and that there is no necessity to assume the existence of two kinds of nerve fibres presiding over secretion. The question seems, however, to be definitely decided in the latter sense by the experiments of Babkin (1913), who investigated the properties of the saliva secreted reflexly by placing in the mouth, in the one case, meat powder, in another case, hydrochloric acid.

It was found that the blood flow through the gland was equally accelerated by both, but, while the content in inorganic salts was identical, the organic constituents of the saliva secreted under the stimulus of meat were in four to five times as great an amount as in that formed under hydrochloric acid. Since the removal of the superior cervical ganglion, by which the influence of the sympathetic fibres is removed, had no effect on the result, it is necessary to assume that the chorda tympani

nerve also contains " trophic fibres." There was no evidence that vaso constrictor fibres were excited in either case. Expressed on Heidenhain's view, we m.iv say that acid excites the " secretory " and vaso-dilator fibres, while the " trophic " fibres are very little affected. Meat excites both the secretory and the trophic fibres of both nerves, in addition to the vaso-dilator fibres of the chorda. Bahkin appears to favour the view that the same nerve fibre conveys different kinds of impulses to the cells, hut, as we shall see in Chapter XIII., there are reasons for doubling this.

Should it be true that there are two different kinds of fibres to the salivary glands, it seems probable that, whenever we have a liquid secretion containing constituents foreign to the blood, the secretion of these substances is under the control of special nerve fibres ; this statement, of course, only refers to those cases where the process is effected by nervous, not by chemical, agency. The Pancreas. — The chemical mechanism in this case appears to be so adequate and appropriate that its discoverers were inclined to doubt the existence of a nervous mechanism, although we were careful not to deny it (Bayliss and Starling, 1902, p. 343). At that time, the experimental evidence did not exclude the possibility of explanation on the lines of a chemical mechanism, but Pavlov has since brought forward evidence which amounts to a satisfactory proof that the vagus nerve contains fibres that cause the production of pancreatic juice, although there are several peculiar facts in connection with the phenomenon. G. von Anrep has demonstrated in England the method of experiment, and there is no doubt that secretion can be obtained by exciting the vagus under certain conditions, which have to be pretty closely adhered to. Some reflex inhibitory influence is exercised by the operative procedures, so that it is necessary to divide the spinal cord at the foramen magnum ; the secretion does not appear until after several successive periods of stimulation of the vagus nerve and, when it appears, it is much less copious than after secretin, and contains active trypsin. This last fact presents some difficulty in regarding the vagus effect as a normal mode of production of the juice, since Delezenne and Frouin (1902 and 1903) have shown that the juice which appears copiously from a permanent pancreatic fistula, when food is being digested, both in the dog and in the ox, is inactive until acted upon by enterokinase. The action of the vagus is paralysed by atropine, like other gland nerves. From the very concentrated character of the juice it would seem that the vagus contains chiefly " trophic " fibres. The question of inhibitory nerves to glands will be discussed later. The paper by Bylina (1912) on the two kinds of mechanism, chemical and nervous, should be consulted.

The view that there are distinct " trophic " fibres in gland nerves receives further support from the changes in microscopic appearances of the gland cells. Excitation of the sympathetic produces considerably more signs of fatigue in the submaxillary gland cells than that of the chorda does. Babkin, Rubashkin, and Savich (1909) have described similar facts with regard to the pancreas. As already stated, by the action of secretin it is difficult to produce signs of fatigue in the cells, while stimulation of the vagus nerve results in marked changes. According to the' observers named, the process of secretion in the case of the chemical excitant is as follows : — Water flows through the cell in quantity, and one sees in the cells what look like channels of fluid (see their Fig. 23). This current carries out the zymogen granules into the ducts, where they can sometimes be seen as granules, but they soon become dissolved. It is found, on staining with eosin and orange, that the secretion in the ducts takes the same red colour as the granules inside the cells, and appears to be of the same chemical nature. We know that the trypsin in it is still in the zymogen stage. No cell constituents staining with orange are to be found.

After nerve stimulation, which gives only a small quantity of a thick juice, we have a different picture. The granules inside the cells undergo a transformation ; they gradually lose the property of staining with eosin or iron haematoxylin and become stainable with orange, sometimes forming large "vacuoles" before passing into the duct. The secretion itself in the ducts stains with orange, not with eosin (see Figs. 16 and 18 of the paper). As we saw, it contains active trypsin. Little or nothing is to be seen of the intracellular channels of the more watery s<><-ivt inn

Fio. 92. STACKS OF ACTIVITY OF THE PANCREAS UNDER THE ACTION OF SECRETIN, AND OF a, 6, c, Stained with iron hfomatoxylin. a, From fasting dog. 6, After action of secretin (acid in duodenum). Comparatively slight disappearance of granules and little structural change. 1 c.c. of juice in the first fifty-three minutes ; 5'1 c.c. in the next hour and a half ; C'4 c.c. in the last d> «. /. ff, Stained with eosin-orange-toluidine blue by Dpminici's method. The parts stained with eosin are shown black ; those stained orange (drops of secretion within the cells and ducts) are shaded obliquely.

d, After vagus stimulation. Granules stain red with eosin, but the contents of most of the vacuoles in the cells and the secretion in the large duct stain orange. e, Vagus stimulation. Some vacuoles stain red ; others orange. /, Weak stimulation of vagus. Empty open vacuoles are seen, which communicate with intercellular spaces. g, Action of secretin. The juice in the duct stains red with eosin, like the granules. There are no vacuoles which stain with orange. (Babkin, Rubaschkin, and Ssawitsch, 1909. Figs. 1, 3, 5, 16, 18, 21, and 22 in order.)

with secretin. Some of these figures are reproduced in monochrome in Fig. 92 (see description of figure). If the juice secreted under natural conditions contained active trypsin, it is difficult to understand the use of the production of enterokinase in the intestine. No doubt, however, the flow of water through the cells might carry away zymogen material before it had been worked up by the cell and this would require activation. It is not only glands with visible secretion that are under the control of the nervous system, but also those of internal secretion. The fact has been shown especially in the case of the adrenals. When the splanchnic nerves are excited in any way, there is an output of adrenaline into the blood, which produces the various phenomena due to stimulation of the sympathetic, such as rise of blood pressure, etc. (see Asher, 1910; Elliott, 1912, etc.).

It seems evident that there are two kinds of processes, or rather two factors, at work ; one concerned with the transfer of water, together with certain solutes already present in blood, the other concerned with the elaboration of new chemical compounds. Whether either of these can be excited without the other, by means of specific nerve fibres or by chemical means, it is at present impossible to state. It is to be remembered that the passage of water in itself would wash out constituents of gland cells previously stored therein, but the results of vagus stimulation on the pancreas indicate that new chemical changes can also be set in action by nerve influence. In the idea of " trophic " nerves, Heidenhain appears to include the function of exciting the formation or replacement of the substances which had been given off from the cells in the process of secretion previously. The vagus effect, described above, suggests rather the setting into play of a chemical change in the products already stored in the cells. Secretin, on the other hand, apparently sets going a process by which water washes out stored substances without change. The prolongation of the period of increased oxygen consumption considerably beyond the actual period of secretion itself, induced by the stimulation of nerves, suggests that the restitution process, by which the cells are restored to a state ready for renewed activity, is an automatic process and controlled by mass action in a reversible system. In the moderated natural process of secretion, such as that of the pancreas induced by the introduction of acid into the duodenum, the fact that signs of fatigue appear in the cells only after very prolonged activity shows that the natural process of restitution keeps pace with the secretory activity of the cells.

On the whole, it appears that the usual process of secretion is somewhat as follows : — During the period of rest, the cells build up compounds which are preliminary stages of constituents of the secretion, which is afterwards set going by excitation, nervous or chemical. The formation of this material is probably a reversible reaction, so that, after a time, further production ceases, owing to accumulation of products. When the gland is excited to activity, a current of water is set flowing through the cell by some means, probably of an osmotic nature and effected by a combination of increased permeability of the outer end of the cell together with splitting up of some substance into smaller molecules. This current of water washes out into the duct the substances of the secretion already stored in the cell, sometimes after they have been further changed by a process which does not take place until the cells are excited to secretory activity. As the stored substances are lost from the cell, there will be a renewed formation to re-establish equilibrium ; so that, if the activity is not too violent, there will be a balance between the amount secreted and its new formation. Continuous secretion will thus be possible without fatigue. It will be seen that, on this view, the increased production in the cell of the substances which give rise afterwards to the actual products contained in the secretion is not to be supposed to be under the control of the nervous system or other excitatory influence, but that it is a spontaneous activity of the cell itself, controlled by chemical equilibrium. Thus

the trophic nerves of Heidenhain are not trophic in the sense of presiding over processes of growth of material, but control the changes in the cell which lead to the transformation of stored substance into the specific organic constituents of the secreted fluid. Under certain conditions, stimulation of the vagus nerve stops a pancreatic secretion in progress, owing to a previous effective excitation, or from injection of secretin. Yon Anrep (unpublished as yet) has investigated this effect and finds that the explanation lies in a contraction of the ducts. It is not surprising that this should be the case, since, as we shall see in the next chapter, the vagus nerve causes contraction of the intestinal muscle, and the pancreatic ducts are outgrowths from the intestine in development. Anrep placed a portion of the pancreas in a plethysmograph and found that, during the cessation of the outward flow of secretion, the gland increased in volume. This latter fact shows that the secretion continued to be formed, but was unable to escape. After a time, the pent-up juice forces its way out and, as the first drop appears, there is a diminution in the volume of the gland, which returns to its normal volume after the apparent inhibition has ceased. It is of interest, also, to note that there is no evidence in these experiments of the presence of vaso-dilator fibres in the vagus, nor of more than a minimal vascular dilatation in the gland when secretin was used, provided that the preparation was free from depressor substance (probably /3-iminazolylethylamine, see Chapter XXIV.).

Bradford (1888, p. 315) considers that the most satisfactory explanation of the curious phenomenon of the "paralytic secretion" of the submaxillary gland is to be found in the hypothesis of a special set of fibres in the chorda tympani nerve. Their function is to check or inhibit the spontaneous activity of the gland cells. After section of this nerve, u secretion of saliva commences in about four hours and lasts for some time, the gland undergoing atrophy at the same time. Further discussion of the action of inhibitory nerves will be found in Chapter XIII.

How far the chemical mechanism applies to all glands and whether there are any glands devoid of nervous control, it is not as yet possible to state definitely. Although the latter mode of excitation appears to be complete and adequate in the case of the salivary glands, some observations by Demoor (1911, 1912, 1913) show that, in the absence of certain chemical substances, stimulation of nerves is without effect. If the submaxillary gland is perfused with Ringer's solution, oxygenated, excitation of the chorda tympani nerve still brings about vaso-dilatation, but no secretion of saliva. Under the same conditions, the pancreas produces no juice when secretin is added to the perfusion fluid. At first sight, it might be thought that it is impossible to supply sufficient oxygen merely by solution in a saline solution, considering the large consumption of oxygen by the gland cells. That this is not the cause of the complete absence of secretion is shown, however, by the fact that if a certain amount of serum of the same animal (100 c.c. to 1,400 c.c. of the saline solution) is added, excitation of the chorda tympani nerve produces a flow of secretion, but only for thirty to sixty seconds. It seems probable that the presence of some constituent of the serum is necessary for the due change in permeability of the cell membrane associated with the process of secretion. The comparatively small amount obtained may arise from the previous store in the cells, and the oxygen supply may be insufficient to afford the energy necessary for the new formation of such substances, or only at a minimal rate. Further observations by Demoor are regarded by him as showing that the way in which a nerve acts in exciting secretion is by causing the production of a chemical substance, which itself acts on the cell processes in a way similar to that in which secretin acts on the pancreas. This exciting substance is perhaps of the nature of a hormone and is carried away in the saliva secreted. The evidence consists

in the fact that addition of saliva to the perfusion fluid causes the gland to secrete. The exciting substance is apparently of a compound nature, since, after heating to 65° C., saliva has lost its power of producing secretory activity from rest although it is still capable of accelerating the rate of flow when this has nearly stopped, subsequent to stimulation of the chorda tympaiii nerve. The work of Hustin (1912, 1913) on the pancreas is also of interest in this connection. Perfusion with oxygenated Ringer's solution, to which secretin has been added, does not result in secretion ; the addition of the blood or certain liquids derived from it, such as hydrocele fluid or lymph, is also necessary. The author concludes that secretin, oxygen, electrolytes, and some substance contained in blood must be simultaneously present. As far as oxygen is concerned, the experiments are conclusive. A mixture of blood, secretin, and saline solution, effective when oxygenated, becomes ineffective when the gases are pumped off. We can readily understand the necessity of electrolytes for maintaining the normal character of the cell processes, and Hustin's experiments show that blood dialysed against isotonic sodium chloride solution is much less effective than normal blood ; even dialysis against Ringer's solution seems to deprive it of some important diffusible constituents, since it is not as effective as non-dialysed blood, although greatly superior to that deprived of all its diffusible constituents except sodium chloride.

For example (1913, p. 89), the amount of juice obtained in sixteen minutes by the use of the latter was 0'05 c.c. ; if dialysed against Ringer's solution, 0'33 c.c. in fourteen minutes, rather more than seven times as much ; with normal blood, 0'70 c.c. in fourteen minutes, or twice as much as the preceding. The necessity of the presence of some substance contained in blood, other than haemoglobin, as carrier of oxygen, is not so satisfactorily shown. Washed red corpuscles were found to answer the purpose of the whole blood ; although one experiment was performed with a solution of haemoglobin, which was found ineffective, it must be noted that the material used was a dried preparation by Merck, which probably consisted of methaemoglobin and could not, if so, act as an oxygen carrier. The evidence that certain tissue extracts and lymphatic fluids do not owe their favouring property to their being better oxygen carriers than the saline solution is not sufficient. Moreover, it was found impossible to separate any constituent from these liquids which was able to take the place of blood. The explanation of the process, on the lines of the Bordet-Ehrlich theory of haemolysis, does not throw much light on its actual nature.

A fairly considerable amount of work has been done in connection with the difference of potential found, on stimulation, to occur between that end of a gland cell which is in relation with the duct, or free surface, and that end in relation to the blood supply. The cause of this phenomenon has not yet been made out, but there are one or two points in the process which have a bearing on the questions before us. Although it had been known for many years that the various glandular tissues of cold-blooded animals, and also the sweat glands of the mammal, gave rise to electrical changes on excitation, it was not until 1885 that it was possible to investigate the different effects in the salivary glands produced by different nerves from this point of view. In that year, in conjunction with Bradford, I was able to show that the potential difference between the hilus of the gland and the opposite surface, that is, between the duct and the surface of the cells turned towards the blood vessels, is of the opposite sign when the chorda tympani nerve of the dog is excited to that when the sympathetic nerve is excited. If the curves of Fig. 93 are consulted, it will be seen that the former is accompanied by a large secretion of saliva, which follows a course very nearly parallel to the electrical change, whereas the latter, of the opposite sign and much smaller, results only in the formation of one drop of saliva. The support which these two opposite effects give to the hypothesis of two different kinds of nerve fibres

FIG. 93. ELECTRICAL CHANGES IN THE SUBMAXILLARY GLAND OF THE DOG. UPPbLektli>neCUrVe8 (ab°Ve d°tted ]ine>— chorda stimulation. Period of stimulation marked by thick Bottom curve— rate of secretion, deduced from the intervals between the drops of saliva marked on the hue above the stimulation signal. Lower set of curves— sympathetic stimulation. Top curve— galvanometer deflection. Note that it has an opposite direction to that given on chorda stimulation. Bottom curve— rate of secretion (approximate). Only one drop obtained.

is obvious. Just as the sympathetic fibres are known to require a larger dose of atropine in order to paralyse them than the chorda fibres do, so the electrical change from the latter nerve was abolished by a much smaller dose than that from the sympathetic. But this refers only to that part of the electrical effect from the chorda which is of opposite sign to that of the sympathetic. After a dose of atropine sufficient to paralyse the "secretory" fibres of the chorda, excitation of this nerve gave a small electrical effect of the same siyu as tli.it from the sympathetic. This effect, normally, is swamped by the much larger opposite one and is, no doubt, due to fibres of the same kind as those which preponderate in the sympathetic. That vaso-motor effects are not concerned in the phenomena is shown by the fact that the electrical changes tire abolished by atropine, which does not affect the vascular ones. In the cat, as was shown by Langley (1878), both nerves produce a watery secretion and, accordingly, we find that the electrical change from both is of the same sign as that of the chorda in the dog, but is usually followed by one of the opposite sign.

We therefore drew the conclusion that the electrical change of the sign of the typical chorda effect in the dog is due to the flow of water (together with salts of the blood) and that the other one is connected with the elaboration of the specific organic constituents of the saliva. Further evidence of the same kind was given by the later experiments of Bradford (1887). Two experiments are of particular interest (pp. 92, 93). The sympathetic in the dog was being excited, giving the usual scanty viscid secretion, with the usual small electrical change. Suddenly a large electrical change of the opposite sign appeared and, coincidently, a copious secretion of watery saliva. In the second experiment the chorda had been stimulated at intervals for an hour and a half. After such treatment, as Langley showed (1889), and as would not be unexpected, since both nerves act on the same cells, stimulation of the sympathetic is apt to give a watery secretion for a time. This was the case in Bradford's experiment, but the watery secretion appeared only after a long latent period, during which the electrical effect was of the usual " sympathetic " sign. As soon as the watery secretion appeared, there was a change in the sign of the electrical effect. After a period of rest, the sympathetic failed to give the watery secretion and the usual " sympathetic " electrical effect reappeared.

The possible causes of these changes will be best appreciated after Chapter XXII. has been read. That the chorda effect is not due to mere flow of liquid along the ducts is shown by another experiment of Bradford's (p. 98) in which clamping of the duct had no effect on the electrical change. Removal of the clamp, after stimulation had been stopped, produced no electrical etfect, although a free flow of saliva took place along the ducts. The electrical change is therefore due to phenomena in the cells themselves.

The corresponding changes in the sweat glands (Hermann and Luchsinger, 1878, 1), in the frog's skin (Hermann, 1878) and tongue (Hermann and Luchsinger, 1878, 2) may be mentioned, since they are easily observed. This phenomenon, as due to increased osmotic pressure in the fluid of the lymph spaces, on account of the diffusion into them of the small molecules of the products of metabolism of the active organ, has been described above (page 165). The detailed observations of Bainbridge (1900) on the submaxillary gland should be consulted.

The possibility of increased production of an appropriate enzyme, in response to the stimulus of a particular article of food, has occurred to several investigators and positive results are said to have been obtained. Careful testing by subsequent observers, however, showed the presence of unsuspected sources of error. The only case in which any satisfactory evidence exists is that of the increased amyloclastic action of the saliva, as described by Lovatt Evans (1913, 1). Carbohydrate food only had this effect and mere chewing, without swallowing, is ineffective. The simplest explanation is that r

chemical substance of the nature of a hormone is produced by the action of the carbohydrate on the mucous membrane of the stomach, similar to the secretin of the pancreatic mechanism. Owing to the peculiar arrangements present, special description of the mechanism of this organ is necessary. We have seen that its activity is confined to the separation of substances which already exist in the blood, with the exception of hippuric acid, and even in this case the chemical change merely consists in the combination of glycine with a benzoyl group, both supplied by the blood.

We have also discussed the function of the glomeruli and come to the conclusion that the liquid leaving their capsules is a filtrate from the blood, having the same composition minus the colloids. The urine as it leaves the kidney, however, is much more concentrated and, as we have also seen, the concentration does not affect all the constituents equally. The problem now before us is the way in which this change is effected as the glomerular filtrate passes along the tubules, which consist of a series of tubes lined with cells of various structure.

To understand the evidence on the question, a knowledge of the structure of the kidney is necessary. This can be obtained from Starling's book (1912, pp. 1264-1268) or from the article by Metzner in Nagel's Handbuch (1907, 2) and it must be assumed in what follows here. Fig. 94 will serve to give a general idea of the arrangement of the tubules. In the higher animals the function of the kidneys may be said to be of two kinds. In the first place, non-volatile products of metabolism, which are useless or injurious, have to be removed. In the second place, the osmotic pressure of the blood has to be kept constant. This osmotic pressure is due chiefly to the salts, so that the excretion of salts must be increased or diminished, according to the amount taken in with the food, and that of water adjusted in accordance with that taken or lost in other ways. In the lower animals, where the osmotic pressure of the body fluids is that of the solution in which they live, the first function is the chief or only one, so that we will consider this to begin with.