from "Bulletin of the Univ. of Wisconsin"
(Science Series, Vol. 3, No. 1, Pp. 1-31)
Louis Kahlenberg, PhD (July 1898)
Assistant Professor of Physical Chemistry

The Action of Solutions on the Sense of Taste

Introduction.

In order to be capable of affecting the sense of taste a substance must be soluble to a certain extent in the saliva, which is, in most cases at least, the same as saying that it must be soluble in water. Substances that are practically insoluble can when introduced into tho mouth produce only sensations of temperature and of touch. The fact that a substance is soluble in water is not of itself sufficient, however, to enable it to cause sensations of taste, for there are many substances that are quite soluble in water and yet their solutions possess veiy littie or no taste. It is evident that the effect of a solution on the sense of taste depends upon the concentration of the solution, the chemical nature of the dissolved substances, and the conditions in which the latter exist in the solution.

Investigations on the subject of solutions have been vigorously pushed during the last ten years by workers in the field of physical chemistry, and as a result of their labors we have today a far better understanding of the condition in which a substance exists when dissolved than ever before. Indeed, van't Hoff's [1] theory of solutions and Arrhenius' [2] theory of electrolytic dissociation have practically solved the mystery that has heretofore engrossed the whole subject of solutions.

Since substances must be dissolved in order to be tasted, and since the taste of a solution depends upon the nature of the dissolved substance and its molecular condition when in solution, it was thought that a systematic study of the effect of solutions on the sense of taste, pursued in the light of the knowledge that has in recent years been gathered concerning the subject of solution, would more clearly define the nature of the substances that produce certain tastes and possibly also indicate the mode of action by means of which these substances cause the sensations of taste. The present investigation is an attempt to study the effect of solutions on the sense of taste in the light of the modern theories of the nature of solutions.

The Nature of Aqueous Solutions

Aqueous solutions may be divided into two classes, those that are practically non-conductors of electricity and those that conduct electricity readily. The former are generally termed non-electrolytes and the latter electrolytes. The aqueous solutions that practicall do not conduct electricity are those of substances that possess neither acid, basic, nor salt-like character, or at least they poesess these characteristics only to a slight degree. Into this category belong, for example, solutions of the mono- and polyatomic alcohols, the sugars, the esters, very weak acids and bases, colloidal substances, and other compounds generally spoken of as neutral substances (the term 'neutral substances' as used here does not include salts). The solutions that conduct electricity readily are those of compounds of pronounced acid, basic, or salt-like character. This class includes solutions of all the acids, bases, and salts except some very weak acids and bases and the salts formed by their combination.

Measurements of the osmotic pressure, the lowering of the freezing point, and the elevation of the boiling point of solutions of non-electrolytes have shown that generally the dissolved substances contained in these solutions exist in the simple molecular condition that is expressed by their chemical formulae as usually written; in other words, the molecular weight of these substances when in the dissolved condition is that which is generally ascribed to them. When it is possible to find the molecular weight of a non-electrolyte by a vapor density determination as well as from the osmotic pressure, the freezing point, or the boiling point of its solution, the results are generally identical.

Instances where double or even more complex molecules are formed when substances are dissolved in water are known, and yet these are perhaps not much more common than cases of polymerization in the gaseous state. The analogy that exists between gases and solutions of non-electrolytes is almost perfect. This appears clearly when we regard the statement of Avogadro - equal volumes of all gases under the same conditions of temperature and pressure contain an equal number of molecules — in conjunction with this statement modified through the work of van't Hoff so as to apply to solutions, — equal volumes of solutions having the same temperature and the same osmotic pressure contain an equal number of molecules, which number is identical with that contained in a gas having tha same volume and temperature as the solution and a pressure equal to the osmotic pressure of the latter.

The solutions that conduct electricity on the other hand are different in character. From what has been said, it is clear that these solutions are such as contain the ordinary salts, acids, and bases. These substances do not exist in aqueous solutions in the molecular condition represented by their chemical formulae as ordinarily written. The osmotic pressure, the lowering of the freezing-point, and the elevation of the boiling-point of these solutions are all greater than they would be if the substances when in solution possessed the molecular weight indicated by the usual formulae. According to the theory of Arrhenius these substances exist in solution in a partially dissociated state; this explains the peculiar behavior of these solutions as compared with those of non-electrolytes. The degree or extent of this dissociation depends on the nature of the dissolved substance and the concentration of the solution. Theoretically the dissociation is complete only at infinite dilution; many substances, however, are so strongly dissociated when dissolved in water that dissociation is practically complete at finite dilutions. The part-molecules into which the dissolved molecules are dissociated are termed ions. These ions migrate through the solution under the influenoe of the electric current, whence their name. The conduction of electricity by the solution depends upon the presence and the movements of the ions. Those ions that travel toward the negative pole are considered as charged with positive electricity and are termed cathions (now, cations), while those that move toward the positive pole are called anions and are charged with negative electricity. The number of cathions in a solution is always equivalent to the number of anions present, so that electrical neutrality of the solution is preserved. Thus in the case of sodium chloride the ions are Na and Cl, and any solution of this salt contains these ions together with a certain amount of undissociated Na Cl, the relative quantities of dissociated and undissociated salt depending on the concentration of the solution. Similarly solutions of hydrochloric acid contain the ions H and Cl and undissociated molecules of HCl.

It would seem at first that matters become more complicated by thus considering the solutions just mentioned; while this is true in some cases, namely when concentrated solutions are under consideration, yet in dilute solutions where dissociation is nearly complete things appear more simple. Thus in very dilute solutions hydrochloric acid and sodium chloride are practically completely dissociated, and the dissolved substances that these solutions contain are consequently the ions H⁺ and Cl^- in the former and Na⁺ and Cl^- in the latter. The ions are to be regarded as distinct and separate substances subject, however, to the law that the solution contains as many positive as negative ions. The ions are furthermore not identical with ordinary hydrogen, chlorine and metallic sodium respectively, for they differ from these in the amount of energy they contain. It is necessary to supply these ions with additional energy in order to change them from the ionic condition to the ordinary state. This can be done, for example, by electrolyzing hydrochloric acid or sodium chloride.

Two dilute solutions containing chemically equivalent quantities of hydrochloric acid and sodium chloride are alike in that they contain chlorine ions in the same conoentration, while they differ in that the former contains hydrogen ions and the latter sodium ions. It has always been regarded as axiomatic that the properties of a solution are determined by what the solution contains. Hence the differences in the properties of the dilute solutions of hydrochloric acid and common salt are simply to be ascribed to the different effects of the H ions and the Na ions, the Cl ions being common to both solutions. Now it happens that a solution of hydrochloric acid still has a very pronounced taste at a degree of dilution at which a common salt solution containing a chemically equivalent quantity is perfectly tasteless. It follows, therefore, that the taste of such a dilute solution of hydrochloric acid is simply due to the hydrogen ions it contains. The dilute solution of hydrochloric acid gives one the sensation of sour, and consequently hydrogen ions cause sour taste. The latter statement will be more fully established below.

From what has been stated concerning the nature of aqueous solutions, it ollows that in the case of solutions of non-electrolytes we have simply to investigate the taste of the dissolved molecules and to seek a relation between the chemical nature of these and the sensations that they cause; while in the case of solutions of electrolytes, we shall have to consider the taste of both the undissociated molecules and the ions present, trying to discover if possible a connection between the gustatory sensations that these create and the other properties that they possess. It is clear that the study of the second class of solutions may be much simplified by working with dilute solutions. By so doing the taste of individual ions may even be determined as indicated in the preceding paragraph.

The Sense of Taste

The organs of the sense of taste are located in the epithelium of the upper surface of the tongue and possibly also in the lining of other parts of the mouth cavity. The nerves of the tongue run into papillae, of which the so-called circumvallate, on the posterior part of the tongue's surface have an especially rich, nerve supply. On these are the terminal organs of taste consisting of peculiar bodies, the so-called taste-bulbs or taste-buds, discovered by Sehwalbe and Loven in 1867. These taste-bulbs, which are minute bodies, oval in shape, are lodged in the epithelium covering the side of the papilla. Each consists of twa sets of cells. On the outside are a number of flat, fusiform, nucleated cells known as supporting or protective cells; these are bent like the staves of a barrel and arranged side by side so as to form a bulb-shaped body, having an aperture at the apex known as the gustatory pore. The inside of the taste-bulb contains five to ten so-called taste-cells, which are pointed at the end next to the gustatory pore and branched at the other end where they are probably connected with nerve fibres. According to Ranvier, supporting cells are also found in the interior of the taste-bulbs, the taste-cells being found interspersed between these. Taste-bulbs occur also on other papillae of the tongue, and it is possible that simpler structures consisting of fewer or even single taste-cells exist where no taste-bulbs are located, for it is known that taste does exist on parts of the tongue where no taste-bulbs have been found. The tip, edges, and back of the tongue are sensitive to taste, while the middle is devoid of taste. The organs of taste are similar in construction to those of smell.

The above is a brief statement of what is generally given concerning the organs of taste in standard works [2] on physiology histology, and physiological psychology. Retzius p3], who has made an extensive study of the nerve endings of the various organs of the special senses, states that he has been unable to find nerve-fibres connecting the so-called taste-cells with the nerve underneath, that the lower terminus of the taste-cells is in many cases foot-shaped, and that it is highly improbable that they are connected directly with underlying nerves. He finds that nerve-fibres ramify between the cells throughout the entire interior of the taste-bulbs. These nerve-fibres, which he terms the intrabulbular nerves, are a continuation of underlying nerves; they traverse the taste-bulb in a general vertical direction, running out into free ends that are located in many cases near the gustatory pore, in other cases more remote from that opening. Retzius thinks the so-called taste-cells are true epithelial cells; he regards them ba "secondary sense cells" similar to the "hair cells" of the sense of hearing. Retzius then finds more analogy between the organs of taste and hearing than between those of taste and smell. Speaking of the sense of taste, he says:

"Es sind meiner Ansicht nach im Geschmacksorgan keine wahren Sinnesnervenzellen vorhanden. Die Nervenzellen des Geschmacksorgans haben sich ebenfalls, wie im Tastorgan, aus dem Epithel zuriickgezogen und liegen in den Ganglien des Geschmacksnerven. Das Geschmacksorgan steht also in morphologisch-phylogenetischer Beziehung auf etwa demselben Standpunkt wie das Tastorgan und gewissermassen das Gehororgan. Die weit gegen das Centralorgan zur¨at;ckgetretenen Nervenzellen senden in das peripherische Organ ihren peripherischen Fortsatz, welcher unter starker Verastelung mit frei auslaufenden Spitzen frei und interzellular im Epithel endigt; in dem Epithel der Geschmackszwiebeln sind indessen eigenth¨at;mliche Zellen vorhanden, welche ungef ahr, wie die Haarzellen des Gehororgans, als eine Art secundarer Sinneszellen aufgefasst werden konnen." According to this view it is of course not difficult to see why certain portions of the tongue are sensitive to taste and yet possess no taste-bulbs.

It is generally accepted that sensations of taste are caused by certain irritations of the nerve terminals, — whether we are to regard the "taste-cells" or the "intracellular nerves" as representing these end-organs is perhaps still an open question, though Retzius appears to have excellent grounds for his opinion. Authorities apparently agree that this irritation of the nerves is due to chemical action upon them. Now in order to get into contact with the end-organs a substance must be in solution. This, however, is not of itself sufficient. To get into the taste-bulb, the mouth of which is always covered with mucous, and to get at the nerve, the dissolved substance must diffuse with a fair degree of readiness; and finally, when it has come into contact with the nerve terminus, it must be capable of acting chemically on the protoplasm of the same, thus causing the irritation that is interpreted as taste. Many substances are tasteless simply because they are insoluble; others, although sufficiently soluble, do not diffuse readily enough to come into contact with the nerve terminus; and still others, which though soluble and sufficiently diffusible, are devoid of taste because they fail to react chemically with the protoplasm of the nerve.

It is evident that for each substance there is a certain minimum amount that must be present in order to cause sufficient irritation at the nerve. This amount will naturally be relatively less in the case of those substances that react more intensely with the protoplasm of the nerve terminus. Again, in the case of those substances that because of very slow diffusion possess but little taste, the solutions must be relatively much stronger in order that sufficient substance may come into contact with the nerve, for the speed of diffusion of a substance is proportional to the difference in concentration that exists between the two layers in contact. No doubt the mechanical action of rubbing the tongue against the palate as we do in tasting aids in bringing the substance to be tasted into contact with the taste-organs. We should, other things being equal, expect a sub.stance that diffuses readily to exert an effect on the end-organs in less concentrated solution than a substance that diffuses more slowly. The electrical conductivity of solutions of electrolytes is dependent upon the number of ions present and the speed with which they move through the solution. From this it is clear that the conductivity of electrolytes and their speed of free diffusion are closely connected, and we should consequently expect to find a relation between the electrical conductivity of solutions and their effect on the end-organs of taste.

When volatile substances are introduced into the mouth, the volatilized portions fill the mouth cavity and also the nasal passages; in the latter thej frequently act on the organs of smell, and we are very apt to confound the smell of such substances with their taste. Indeed, it is well known that many volatile substances, which we commonly regard as having a strong taste, in reality have no taste at all, for when the nasal passages are obstructed these substances are without taste. It is clear from this that experiments on the sense of taste are best conducted with substances that are non-volatile.

The sensations of taste are commonly classified as those of sweet, sour, salty, and bitter, to which Wundt [1] adds alkaline and metallic. There can be no doubt, however, that there are very many kinds and shades of taste that are quite distinct and not to be referred to sensations of touch, and that the above classification can claim at best to be only a very rough one. The investigator of this subject is soon struck by the fact that we have so few names to describe the various tastes. It is often veiy difficult for the subject experimented upon to report in words what taste the substance under consideration has, in spite of the fact that a veiy definite impression is experienced.

The sense of taste is frequently regarded as rather vague, indefinite and uncertain, probably in part because it is not more definitely localized, and yet, experiments show that it is exceedingly sensitive toward many things, and the fact that it may be cultivated to distinguish very small differences is beyond dispute.

The Method of Experimentation

Fifteen persons served as subjects to be experimented upon. Thirteen of these were between twenty and thirty years of age; of this number three were ladies. The other two, a lady and a gentleman, were about sixty and sixty-three years old respec- tively. All were in excellent health and were practically total abstainers from the use of intoxicating liquors and tobacco.

The solutions, which were prepared with distilled water that was practically tasteless, contained chemically equivalent quantities, i. e., they were so-called normal solutions. I chose the solutions of such strength that they would give me distinct impressions of taste, not sufficiently strong, however, to produce in any case lasting disagreeable or painful sensations. A portion of each of these solutions was then diluted with water to one-half its former strength; a portion of each of the solutions, thus obtained was again diluted with water to one-half its strength, and so on until a solution was obtained, the taste of which I could not distinguish from distilled water. About 200 cc of each dilution was prepared. The solutions were kept in flasks thoroughly cleaned and steamed; they were labeled in cipher known only to me. This was done because many of the persons tested were conversant with chemical symbols, and it was my purpose to have them entirely ignorant of the contents, of the solutions they were tasting, so that they would report the sensations they received without being biased by thoughts as to how the solution ought to taste. The chemicals used were of the chemically pure kind of reliable makers. The distilled water was always used as a check.

The subject was first given an opportunity to thoroughly rinse the mouth with distilled water so as to remove any excess of mucous. Seated with his back toward the table on which were the flasks containing the solutions, the subject took from a porcelain spoon about four cubic centimeters of the solution to be tasted. The individual held this in the mouth for a few moments, being permitted to move the tongue and lips at will so as to spread the liquid over the entire cavity of the mouth, and bring the liquid into more immediate contact with the membranes by friction. The solution was then ejected, the report given, and the mouth generally rinsed with a little distilled water before aiiother solution was tasted. I had the persom ait with bis back toward the table on which the flasks stood, for I found it necessary that, in order to get from him an unbiased report, he should not know from whioh flask I was giving him. In testing as to the relative strength of the taste of several solutions, I could thus give the same solution twice in succession or give simply distilled water without the subject's knowledge. I found that this procedure was quite necessary in many cases in order to obtain reliable results. The distilled water and the solutions were of the same temperature, about 23 °C. As my purpose was rather to compare the tastes of different solutions than to find out in each case as accurately as possible the most dilute solution that could still be tasted, I did not deem it necessary to raise the temperature of the liquids to that of the body.

Each person experimented upon was not detained more than half an hour at a time, and the solutions were always given beginning with the weaker and proceeding to the stronger. This was very essential, for preliminary experiments indicated that when a strong solution is first given the effect of it is apt to remain in the mouth and make other tests difficult for the time being or perhaps even impossible. For the same reason, too, all strong solutions that would be apt to leave a prolonged taste in the mouth were either entirely avoided, or given the subject as the last solution to be tested at that sitting. The solutions used were in all cases perfectly odorless unless otherwise stated.

The Taste of Solutions of Electrolytes

The taste of solutions of electrolytes received attention, first, because these solutions have in many cases very pronounced tastes that render work with them relatively easy; furthermore, when I first began the experiments, it was simply my purpose to investigate the taste of the ions; as the work progressed it took a somewhat wider scope, nevertheless most of the experiments were conducted with solutions of electrolytes.

Sour Taste. As pointed out above the sour taste of acids is due to the hydrogen ions present [1]. In order to firmly establish this experimentally a n/200 hydrochloric acid solution was prepared ("n/100" means 1 gram of a substance 'n' in 100,000 grams of water); this I found to have a very decided acid taste. From this solution, n/200, n/400 and n/800 solutions were prepared as before described. On testing the fifteen individuals mentioned, it was found that four of them could detect a difference between the n/800 solution and the distilled water, while all of them distinctly tasted the n/400 solution. The n/800 solution was reported not as sour, but as slightly astringent ; the n/400 was reported as astringent by the men but by the ladies uniformly as astringent and slightly sour. [2] As at these dilutions the dissociation of the hydrochloric acid is practically complete, and as it requires a much stronger solution of Na Cl than n/400 to cause taste, as will be shown below, it is clear that the sour taste is simply due to the effect of the hydrogen ions. I have no doubt that with cultivation of the taste for hydrogen ions, and previous elevation of the temperature of the solutions to that of the body, even more dilute solutions than n/800 could be detected by the sense of taste. Indeed, the experiments of Richards [3] confirm this. He shows clearly that fairly accurate titrations of hydrochloric acid can be made using the taste of the solutions to indicate the end of the reaction.

The n/200 solution of hydrochloric acid was uniformly reported as sour, as was of course also the n/100. Solutions of sulfuric, hydrobromic, and nitric acids equivalent to those of hydrochloric were also prepared, and the subjects were tested with these. The results were the same as with the corresponding solutions of hydrochloric acid. The n/800 solutions were reported as astringent by those that could distinguish them from distilled water, n/400 were reported as astringent by the men and as astringent and slightly sour by the women, while all found the n/400 and n/100 solutions distinctly sour. No difference either qualitative or quantitative could be distinguished by these individuals between the solutions of these various acids of equivalent strengths. The electrical conductivity [1] of solutions of these acids shows that in their 4/400 solutions the compounds are practically completely dissociated, the number of undissociated molecules present at this concentration is then practically nil. It will be shown below that the sodium salts of these acids in n/400 solutions, or even much greater concentrations, are tasteless.

In testing solutions of acetic acid it was found that n/200 could be tasted as astringent while n/100 was reported as sour, though much less so than n/100 solution of the other acids mentioned. The electrical conductivity of solutions of acetic acid shows that in n/200 and n/100 solutions the degrees of dissociation are about 6 percent, and four percent, respectively. One would then expect acetic acid solutions to be less sour than equivalent solutions of the strong mineral acids; at the same time, it is apparent that if hydrogen ions can be tasted as astringent in n/800 solutions, a n/200 solution of acetic acid, which is dissociated only about 6 per cent, and hence with respect to hydrogen ions is 6/20000 normal, ought to be tasteless. To be n/800 with respect to its content in hydrogen ions, a n/200 acetic acid solution ought to be dissociated 25 percent. It is clear then that the n/200 solution of acetic acid, being dissociated only about 6 per cent, has a sour taste about four times as strong as it ought to have, assuming that the taste is due simply to the hydrogen ions momentarily present. Richards [2] obtained a similar result; he found that the acetic acid was about one-third as strong as an equivalent solution of hydrochloric acid, though the acetic acid was only dissociated to the extent of one-fourteenth. Richards gives no explanation of this phenomenon, and at the present time I also have none to offer. The further investigation of this point, together with that of the interesting question of the taste of acid sodium salts of polybasic acids, is contemplated.

The foregoing results and those of Richards show conclusively that hydrogen ions have a sour taste; furthermore, it is clear that in very dilute solutions they produce simply an astringent sensation. The question arises, is sour taste always due to the presence of hydrogen ions? I am inclined to answer this question in the affirmative, for I know of no substance that has a sour taste which on going into solution in water does not yield hydrogen ions.

From wikipedia entry on taste: "Sour taste is detected by a small subset of cells that are distributed across all taste buds in the tongue. Sour taste cells can be identified by expression of the protein PKD2L1,[24] although this gene is not required for sour responses. There is evidence that the protons that are abundant in sour substances can directly enter the sour taste cells. This transfer of positive charge into the cell can itself trigger an electrical response. It has also been proposed that weak acids such as acetic acid, which are not fully dissociated at physiological pH values, can penetrate taste cells and thereby elicit an electrical response. According to this mechanism, intracellular hydrogen ions inhibit potassium channels, which normally function to hyperpolarize the cell. By a combination of direct intake of hydrogen ions (which itself depolarizes the cell) and the inhibition of the hyperpolarizing channel, sourness causes the taste cell to fire action potentials and release neurotransmitter. The mechanism by which animals detect sour is still not completely understood."

With regard to the astringent effects, it seems that in many if not in all cases these can be ascribed to the presence of hydrogen ions in about n/400 solution. I was deeply impressed with the fact that many of the subjects tested said that the n/400 solutions of the mineral acids tasted like alum. It has been shown [1] that alum solutions contain hydrogen ions in small quantities due to the so-called hydrolytic dissociation of the aluminum sulfate, i.e., to a reaction of this salt with water forming a small amount of sulfuric acid from which in turn by electrolytic dissociation hydrogen ions form. In presence of the sulfate of the alkali metal an acid salt would no doubt form from which, according to previous investigations [2], hydrogen ions split off rather difficultly. The acid reaction of alum solutions toward indicators is of course further proof of the presence of hydrogen ions. The astringent taste of solutions of ferric salts and their acid reaction toward indicators are well known; these solutions like those of aluminum salts have long been used in medicine because of their astringent properties, which I am inclined to ascribe to the hydrogen ions present due to hydrolytic dissociation. This statement is of course not to be construed as meaning that the other ions and the undissociated molecules present in these solutions do not exert an effect, for no doubt they do especially in strong solutions, and to these effects the differences of the individual solutions are due.

It is well known that solutions of most of the salts of the heavy metals have add reactions and that they have an astringent effect upon the membranes of the month, besides creating in some cases the so-called metallic taste. The solutions of all of these salts contain hydrogen ions whose presence is caused by hydrolytic dissociation as in the case of the salts of iron and aluminum. To these hydrogen ions the astringent effect of the solutions is very likely to be ascribed in many cases. Salts of stronger acids with the alkalies are not decomposed hydrolytically and do not possess astringent properties. Long [1] used the method of sugar inversion in investigating the hydrolytic decomposition of salt solutions. He employed a polariscope in his work and consequently could test only colorless solutions. The freezings and boiling-point methods, however, can be used quite as well as the polariscope in this work; although by means of them the observations are perhaps not quite as accurately and readily made. They possess the advantage, however, that they can be used with colored solutions. Experiments along this line have for some time been in progress in this laboratory and the results will soon be ready for publication.

As to the nature of the chemical action of the hydrogen ions on the nerve nothing definite is known, indeed the same must be said of the action of any ingredient on the nerves of taste. It is significant, however, that hydrogen ions can be detected by the sense of taste in very dilute solutions, the limit being in the neighborhood of n/800 or one gram of hydrogen ions in 800,000 grams of water. The speed of migration of the hydrogen ion exceeds that of any other ion; it is about one and three-fourths times that of the next fastest ion, the hydroxyl ion. Intimately connected with this is the fact that solutions of strong acids diffuse more rapidly than those of their salts. By virtue of their great mobility, it is clear that hydrogen ions can easily get at the end organs of taste. It is well known that hydrogen ions in many cases accelerate chemical action, i.e., they have a so-called catalytic effect. Whether it will be found that they accelerate the action that goes on in the nerves or unite with their protoplasm, can not be stated now; it does seem suggestive, however, that the ion which has the least relative mass and by far the greatest mobility can be tasted in more dilute solutions than other substances and causes that peculiar sharp sensation.

Alkaline Taste. — Dilute solutions of caustic alkalies have the characteristic alkaline taste. The effect upon the tongue is veiy different from that produced by hydrogen ions. In aqueous solutions the caustic alkalies are dissociated into hydroxyl ions and the ions of the metal or basic radical; thus in the case of NaOH the ions are JNa and OH, in the case of KOH, K and OH, etc Solutions of the hydroxides of sodium, potassium and lithium of the strengths ^, ^, ^, and ^ were prepared, and their taste was investigated as in the investigation of the acids. It was f oimd that ^ NaOH could not be distinguished from dis- tilled water; the ^ solution could be tasted very faintly; — the taste was difficult to describe, emne calling it a rather stale taste. The ^ solution was reported as alkaline by all, at the same time some also received a slight sensation of bitter from the same. The solution of KOH and liOH yielded essentially the same results. The persons tested apparently were not able to dis- tinguish any difference either qualitative or quantitative between solutions of equivalent strength of these three alkalies. The al- kalinity of a A solution of NaOH is plainly recognized by the sense of taste. In this solution the dissociation of the NaOH is practically complete. As a solution of NaCl of Equivalent strength is tasteless, as will appear below, it follows that the alka- line taste of the NaOH solution is to be ascribed to the effect of the OH ions. Hydroxyl ions then have the so-called alkaline taste. In stronger solutions caustic alkalies are known to pro- duce nausea. It is probable that this is due to their large con- tent of OH ions, the caustic alkalies being even in fairly concen- trated solutions in a relatively highly dissociated state.

It is well known that solutions of salts of strong bases with very weak acids have an alkaline reaction toward indicators. This is due to the fact that these salts are to a certain extent hydro- lytioally decomposed by the water into free acid and caustic al- kali^ the latter yielding OH ions by electrolytic dissociation. As ^examples of salts whose salntMHia pomem alkaline reactions and tastes because of the OH ions due to hydrolytic dissociation may be mentioned the carbonates^ silicates, and borates of the alkalies, and the soaps. The fact that the latter In strong so- lutions produce vomiting is well known; probably this effect is due to the OH ions present in the solutions.

The taste that a dilute solution containing OH ions causes is difficult to describe. Some of the persons tested said it was a soft, smooth sensation quite unlike that produced by other substances. It would seem somewhat peculiar perhaps that the taste of hydroxyl ions, being not sharp like sour or salty tastes, should manifest itself in solutions containing only one gram ion (i. e., 17 grams of OH) in 400 liters. It must be remembered in this connection that hydroxyl ions, like hydrogen ions, are generally speaking very reactive. What the nature of the ac- tion of the OH ions on the protoplasm of the nerve is, is not known. The mobility of OH ions is very great (being second -only to that of H ions as already pointed out), consequently we should expect them to find little difficulty in reaching the nerve endings.

Water itself is slightly dissociated into hydrogen, and hydroxyl ions. The degree of this dissociation is, however, exceedingly small. Very pure water has been prepared by Kohlrausch, the •electrical conductivity of which showed that there were no more than 18 grams of dissociated water present in eleven million li- ters. It is clear that this is far beyond the limit at which H and OH ions can be detected by the sense of taste. It is not very difficult to obtain distilled water of a specific conductivity of 2x10^ and such water is devoid of taste. If we inquire as to the reason for this, we should agree that the undiasociated molecule of water is tasteless because it does not react chemically with the protoplasm of the nerve; the other alternative, that it does not reach the nerve becsanise of too slow diffusion is here excluded.

Salty Taste. — The taste of common salt is generally given as the type of salty taste. It was found that of the persons tested only three could distinguish a slightly salty taste in n/50 solution of sodium chloride, while all recognized n/25 as a trifle salty. It requires then a much stronger solution to produce the salty taste than either the sour or the alkaline. In n/50 solution Na Cl is dissociated to the extent of about 91 per cent, and in n/25 to about 88 percent. As equivalent solutions of sodium acetate are not salty, possessing hardly any taste, it is clear that the salty taste of the common salt solutions is due to the chlorine ions. The fact that borax solutions of equivalent sodium content are not salty and that many of the individuals tested found difficulty and others could not distinguish between n/25 solutions of KCl, LiCl, and NaCl, argues in favor of this view. The taste of the ions of the alkali metals will be discussed later.

Sodium bromide solutions of the strength n/25 were found to be distinctly salty by all; a few seemed to taste the n/50 solution, but very faintly. This salt is dissociated to about the same extent as the corresponding solutions of sodium chloride. The salty taste of the solutions of NaBr is due to the Br ion. The sense of taste is apparently able to detect this ion in solutions nearly as dilute as the chlorine ion. I found that several persons seemed to detect a qualitative difference between n/25 NaCl and NaBr solutions by the sense of taste, others again reported no perceptible difference. Those that did find a difference said that the chloride solution was a little sharper than that of the bromide.

Sodium iodide solutions could be tasted in n/50 solutions and more clearly in n/25. In neither of these cases was the taste, salty. It takes about a n/6.25 solution to cause the salty taste to appear. The same was observed in case of solutions of potassium iodide, only here the taste of n/50 solutions was more marked and rather bitter. The behavior of these solutions will again be considered below in connection with the taste of cathions. It is evident from the results obtained that iodine ions do not have as salty a taste as do bromine and chlorine ions. The salty taste of the chlorine, bromine, and iodine ions decreases as the atomic weight increases. The mobility of these ions as determined from the conductivity of the respective sodium salts is nearly the same.

In working with sodium nitrate, I found that all the individuals could taste a n/12.5 solution, though very faintly. They found it impossible to describe the taste. The n/6.25 solution was a trifle salty to eight, but they said it was quite a different taste from that of sodium chloride. Of the others, three could not describe the taste, four said it was a smooth taste, and one person said it was more like that of borax than common salt to him. In n/6.25 solutions sodium nitrate is dissociated over eighty percent, and in n/12.5 about 90 percent. It follows that neither the Na nor the KO₃ ions have a very pronounced effect on the sense of taste. The slightly salty taste of the n/6.25 solution is probably due to the NO₃ ion, for the sodium ion does not produce such an effect, as is evident from the taste of solutions of sodium acetate. The speed of migration of the sodium ion is about 1/7 of that of the hydrogen ion, and the speed of the NO₃ ion is about 1/5 that of the hydrogen ion.

Sodium sulfate was tasted in n/25 solution, though not as salty. It was found difficult to describe the taste. Even in n/12.5 solution this substance did not produce a salty taste. In n/6.25 solution a salty taste described by some as slightly bitter was recognized, though all agreed that the "salty taste" was different from that produced by common salt. In n/25 solutions Na₂SO₄ is dissociated about 75 percent, and in n/6.25 about 62 percent. The result shows that the SO₄ ion does not have a very pronounced taste. The mobility of the ion 1/2 SO₄ is about that of the halogen ions mentioned above.

Solutions of sodium acetate of the strengths n/25, n/12.5, and n/6.25 were distinctly tasted but in no case reported as salty. The taste was variously described as smooth, sweetish, faintly alkaline, etc. Even when this salt is taken into the mouth in very concentrated solution the taste is not salty. Indeed, the taste is not pronounced and it is most difficult to describe it in words. From the taste of sodium acetate solutions it follows then, that neither Na ions, CH₂COO ions, nor undissociated molecules of sodium acetate possess a strong taste. The taste of sodium ions is but slight, they seem to produce a smooth sensation that can not be detected in very dilute eolu- tions. This together with the fact that sodium salts are meet strongly dissociated admirably adapts the latter for investigating the taste of anions of various kinds. Since sodium ions migrate only 1/7 as fast as hydrogen ions their rate of diffusion is relatively slow and this may in part account for the fact that they do not affect the sense of taste more. There seems but little room for doubt that in the presence of an ion of pronounced taste, the taste of the sodium ion is completely masked.

The Taste of Cathions. — The tastes of hydrogen and sodium ions have already been discussed in connection with the taste of the anions. To find the taste of a cathion we shall choose a solution of a salt the anion of which has little or no taste at a dilution at which the salt is fairly nearly dissociated and at which the cathion can still be tasted.

The taste of potassium ions is rather bitter and disagreeable. Solutions of the nitrate, sulfate, and acetate of potassium that can be plainly tasted produce, especially on the back of the tongue, a bitter and rather disagreeable taste. The corresponding sodium salts do not produce this effect, and as potassium salts even in fairly strong solution are highly dissociated, this effect is to be ascribed to the action of the potasium ions. Thus the individuals tested could taste KNO₃ solutions that were n/25; the report was that the taste of the solution was smooth, alkaline (?). In n/12.5 solution the same taste together with a bitter sensation was reported, while n/6.25 solution was found to be decidedly disagreeable. In n/6.25 and n/12.5 solutions KNO₃, is dissociated about 77 percent, and 87 percent respectively; as the corresponding sodium salts at these concentrations do not produce this bitter taste, which is characteristic of all potassium salts, it is dearly to be ascribed to the potassium ions. The mobility of potassium ions is about one and one-half that of sodium ions so that their diffusion is more rapid. The taste of potassium ions being fairly pronounced, it is not as readily masked as that of sodium ions. Thus it is possible to distinguish solutions of KCl from NaCl and KI from NaI; the iodine ion not having as strong a sally taste as the chlorine ion, the difference in taste between the potassium and sodium ions comes out fairly distinctly in dilute solutions of the last named salts. Sulfate of potassium solutions have the characteristic bitter, disagreeable taste of the potassium ions; these solutions lack the sharp taste of the NO₃ ions, which we have in strong solutions of KNO₃ and other nitrates when these are applied to the tip of the tongue. Solutions of KClO₃ do not have the salty taste of the chlorine ions. The bitter taste of the potassium ion is somewhat masked by the effect of the ClO₃ ion, the latter acts especially on the tip and edges of the tongue creating its own characteristic sharp taste. Solutions of NaBrO₃ were incidentally tested in this connection; they have but a slight taste which is not salty. The Bro₃ ion has a taste similar to that of the ClO₃ ion, which would naturally be expected. The tip of the tongue is quite susceptible to irritation by various salts, many of them giving that peculiar sharp or burning sensation, which is different from the taste of hydrogen ions; thus, KI, NaI, NaCl, NH₄Cl, etc., besides the nitrates when applied in strong solutions on the very tip of the tongue cause a burning sensation, which they do not produce on other parts of the organ.

Lithium ions appear to have but little taste. A n/25 solutions of LiNO₃ was very difficult to detect by taste; n/12.5 gave a rather alkaline smooth impression, and n/6.25 sharp taste like a solution of NaNO₃ of equivalent concentration. This sharp taste is probably due to the action of the nitrate on the tip of the tongue and is to be ascribed to the action of the NO₃ ion. The mobility of the lithium ion is only a little over one half that of the potassium ion, hence its diffusion is slower, vhich would in part account for its leas pronounced taste. The taste of magnesium ions is probably bitter as appears from the taste of magnesium sulfate solutions. It takes about n/12.5 to n/6.25 solutions of this salt to produce a distinctly recognizable bitter taste. In the latter concentration the salt is dissociated only about 40 percent, so that it is an open question as to whether the undissociated molecules of MgSO₄ or the Mg ions cause the bitter taste. Magnesium ions very likely have a bitter taste as other solutions in which they occur have this taste; probably the undissociated molecules also have a similar effect. In solutions of MgCl₂ we have both the bitter taste of the Mg ions and the salty e£Fect of the Cl ions. The combined effect is such as to make the taste of the solutions of this salt most disagreeable. A solution of MgCl₂ that was n/25 was recognized as salty but not as bitter, while n/12.5 and n/6.25 solutions produced both the salty and bitter effects, which is what we should expect according to the dissociation theory and the results obtained in case of MgSO₄ and NaCl as given above. In n/6.25 solution the degree of dissociation of MgCl₂ is about 70 percent. The mobility of the ion 1/2 Mg is about the same as that of the lithium ion.

The taste of Ca(NO₃)₂ is a trifle sharp in n/12.5 solutions and distinctly bitter n/6.25. It is a different bitter from that of the solutions of magnesium salts. The bitter taste of the Ca(NO₃)₂ solutions is probably due to the Ca ions. The mobility of the Ca ions is about the same as that of the Mg ions.

Solutions of ammonium sulfate give scarcely any salty taste; their taste is rather to be described as bitter. This taste is caused by the NH₄ ions and the undissociated molecules (NH₄)₂.SO₄. The probability is that NH₄ ions have a bitter effect, since NH₄NO₃ solutions, besides creating a sharp, burning taste on the tip and edges of the tongue, also have a bitter taste.

To get substances whose solutions have a characteristic "metallic" taste, silver nitrate and mercuric chloride were selected. It was found that even in n/6000 solution silver nitrate could still be tasted, while in n/2000 its taste was very pronounced. From the limit of taste of sodium nitrate solutions as given above approximately, it follows that at the dilutions just mentioned the taste of silver nitrate solutions is simply that of the silver ions. Silver ions have a peculiar puckering effect on the membranes of the mouth which, if the impression has not been too weak, will remain for some time. The taste is frequently spoken of as "metallic". At first it would seem that silver ions can be tasted in more dUute solutions than hydrogen ions; this is true when we compare chemically equivalent quantities, but since the atomic weight of silver is about 108, a n/5000 solution of silver nitrate contains 1 gram of silver ions in 46300 cc, while an n/800 solution of hydrochloric acid contains 1 gram of hydrogen ions in 800000 cc.

Solutions of mercuric chloride can be very faintly tasted when n/2000, plainly when n/1000, the "metallic" taste being somewhat similar to that produced by silver ions. Mercuric chloride is not highly dissociated in solutions that are not very dilute. The concentrations just mentioned are such, however, that in them the salt is largely eleotrolytically dissociated. As Cl ions can not be tasted in n/1000 solutions, it is clear that we get here the taste of the mercury ions. [1]

The Relations of Overton's Work to the Taste of Solutions

Ernst Overton, in an interesting article, "Uber die osmotischen Eigenschaften der Zelle in ihrer Bedeutung fur die Toxikologie und Pharmakologie mit besonderer Beriicksichtigung der Ammoniake imd Alkaloide" (Zeitschr. f. physik. Chem. 22, p. 189, 1897), gives a list of organic groups arranged according to the degree of retardation that they exert in preventing the substance in which they occur from passing through vegetable and animal membranes. The list beginning with the group that retards most is as follows:

1. The amido-ccid group.
2. The carboxyl group.
8. The acid-amido group.
4. The alcoholic hydroxyl group.
5. The aldehyde group.

When several of these groups occur in one and the same compound, the retarding action increases with the number of groups in a rapid geometrical progression; for example, while substances that contain but one alcoholic hjdrozyl group readily pass through membranes, those that contain two or more such groups find increasing difficulty in doing so as the number of groups grows larger.

In order to affect the nerves of taste substances must be readily diffusible, as was previously pointed out; it follows, therefore, from the above table of Overton, that compounds containing the amido-acid or acid-amido group should have little or no taste. Glycocoll (CH₂-NH₂.COOH) (now known as glycine), which is soluble in about four parts of cold water, has a weak sweetish taste. In order to taste the substance, concentrated solutions or even a crystal directly must be taken into the mouth. Acetanilid (acetanilide), C₆H₅.NH(C₂H₃O), though readily soluble has almost no taste even in saturated solutions. Asparagine, C₂H₃.NH₂.CONH₂.COOH, which dissolves easily in water, is almost perfectly tasteless even in its strongest solutions. Acetamide, CH₃.CO.NH₂; also readily soluble, does possess a characteristic taste, which, however, is not very strong. It is difficult to ascertain the true nature of this taste as it is not easy to eliminate the mouse odor that the solutions of this substance have. Overton emphasizes that the acid-amido group does not exert nearly as great a retarding influence as does the amido-acid group, hence the behavior of acetamid is perhaps such as would be expected. Urea, CO.(NH₂)₂, very soluble in water, has a slightly bitter taste reminding one of tha2)₂ readily soluble, is practically tasteless; its most concentrated solutions are very faintly bitter.

It is difficult to study the effect of the carboxyl group because acids dissociate yielding hydrogen ions the taste of which is so strong that it masks that of other molecules present in the solution. The fact that solutions of sodium acetate, of sodium oxalate and sodium tartrate have no pronounced taste is evidence that the anions of these acids have but a slight effect, if any, on the end organs of taste.

The alcohols having but one hydroxyl-group, possess taste and also a very strong odor. As it is extremely difficult to exclude the smell of these substances while tasting them, nothing was done with them experimentally.

Great interest attaches to the polyatomic alcohols. Of these ethylene-glycol having two hydroxyl groups and glycerine with three hydroxyl groups have a sweet taste that can readily be detected in strong solutions. Erythrite with four hydroxyl groups and mannite with six are practically tasteless; only in very strong solutions were these substances found to be sweet. The taste of a sample of dulcite was pronounced to be nil even in the strongest solutions, while isodulcite and sorbite were found to be slightly sweet.

Turning now to the sugars, arabinose, laevulose, d-glucose and galactose were reported to be sweet, as were also maltose (malt sugar) and saccharose (cane sugar), while lactose (milk sugar) and xylose were found to have little or no taste. The aldehyde groups occurring in sugars, seem to render them more capable of permeating membranes, and probably they also modify the compounds so that in their action on the nerve they increase the sweetish taste, which on the whole is characteristic of the alcohols containing several hydroxyl groups. The intensity of the tastes of the polyatomic alcohols and the sugars is then in general such as one would expect viewing the matter in the light of Overton's work.

The Taste of the Alkaloids

Overton found that coniine (poison found in hemlock), nicotine, sparteine (poisonous), etc., very readily pass through protoplasm; of the alkaloids that contain oxygen, codeine, thebaine, cocaine, atropine, strychnine and brucine diffuse very rapidly, whereas morphine diffuses more dowly and ecgonine very slowly. The burning taste and the characteristic odor of coniine and nicotine are well known, as is also the very bitter taste of sparteine. With the exception of thebaine, the other alkaloids mentioned also have pronounced bitter tastes.

The taste of ecgonine, which, though very soluble, passes through membranes slowly, has been described as bittersweet. Thebaine seems to behave in an exceptional manner; it is very soluble, very poisonous, passes through membranes with great readiness, and yet is tasteless according to some authorities [1]. I have not been able to verify this since a sample of the substance was not available. With the apparent exception of thebaine, however, it is important to note that the alkaloids which diffuse readily through membranes and which are known to exercise a strong physiological effect on the nerves are also able to get at the nerve endings of the sense of taste, reacting upon the same with vigor, producing disagreeable burning or bitter tastes.

Colloidal Solutions

As typical colloidal solutions may be mentioned solutions of dialyzed silicic acid, ferric hydroxide, aluminum hydroxide, also solutions of albumen, gelatine, gums, etc. These solutions are practically non-conductors of electricity, i. e., they contain few or no ions; they have boiling and freezing points that differ but very little from those of water. The rate of diffusion of colloids is very slow, and in general they are very inert in their chemical behavior. Besides those already mentioned, other characteristics of colloidal substances are that they have no definite solubility and that their solutions gelatinize when treated with certain reagents. These colloidal solutions are devoid of taste. This is very likely due to the fact that the molecular weight of colloidal substances is very large and their diffusion so very slow that they can not get at the nerve endings; though, because of their inert character, it is quite reasonable to suppose that, even if they were able to come into immediate contact with the protoplasm of the nerve, they would probably not react with it sufficiently to produce sensations of taste.

Retrospect

One may briefly summarize the salient points contained in the foregoing as follows:

1. In order that a substance may affect the sense of taste; it must be soluble in water; it must be readily diffusible; and it must be capable of reacting chemically with the protoplasm of the terminals of the nerves of taste.

2. The modern theories of solutions lead to the conclusion that the taste of a solution that conducts electricity ought in general to be that of the ions and the undissociated molecules that the solution contains; furthermore, the taste of a solution in which ionization is practically complete should be simply that of the ions. This is supported by the results of the investigation of the taste of solutions of electrolytes above given.

3. Sour taste is caused by hydrogen ions. The sense of taste is able to detect hydrogen ions even in n/800 solutions. In more dilute solutions than n/200, hydrogen ions may cause simply an astringent sensation. The sour taste of acetic acid solutions has been found to be more intense than it ought to be according to the degree of dissociation of the substance. No explanation of this phenomenon has thus far been attempted.

4. Hydroxyl ions produce an alkaline taste, which can be perceived even in n/400 solutions. In strong solutions their taste is exceedingly disagreeable. Pure water, being very slightly dissociated, is tasteless probably because its undissociated molecules do not act on the protoplasm cf the nerve.

5. Chlorine ions have a salty taste. The taste of common salt solutions is mainly that of chlorine ions. Chlorine ions can still be faintly tasted in n/50 solutions. Bromine ions also have a sally taste, which, however, is slightly different in quality from that of chlorine ions. The sense of taste appears to be able to detect chlorine ions at a slightly greater dilution than broinine ions. The ions ClO₃ and BrO₃ have a somewhat similar taste; which, however, is not sharp and salty like that of chlorine and bromine ions. Iodine ions have a salty taste but it is different in quality and less intense than that of either chlorine or bromine ions. It takes about a n/6 solution of iodine ions to produce a distinctly salty taste.

6. The taste of NO₃ ions is slight, probably a trifle salty; only in strong solutions do they produce a sharp burning sensation on the tip and edges of the tongue. The ions SO₄ and CH₃.COO have but very little taste; the effect of the latter seems to be a trifle sweet.

7. The taste of sodium ions is slight. It is difficult to describe, being a smooth effect on the tongue somewhat similar perhaps to that produced by a very dilute solution of hydroxyl ions. Potassium ions have a more pronounced taste than sodium ions. It is a peculiar, bitter, rather disagreeable taste that can be much more readily detected than can that of sodium ions. Lithium ions have no pronounced taste, their effect is somewhat like that of sodium ions, though less in degree. Magnesium ions are bitter. It takes about a n/6 solution to cause a distinctiy bitter taste. Calcium ions are bitter, but the taste is different in quality from that of magnesium ions. Ammonium ions also have a bitter taste. The taste of silver ions is "metallic"; they cause a peculiar puckering sensation on the membranes of the mouth cavity. Even a n/5000 solution of silver ions can still be tasted. Mercury ions can be faintly detected by the sense of taste in n/2000 solution. Their taste is "metallic" and their effect on the membranes of the mouth cavity reminds one of that of silver ions.

8. The intensity of the salty taste of the halogen ions decreases as the atomic weight increases. The investigation of the cathions also indicates that a relation exists between their taste and their atomic weights in the sense of the periodic law. When the taste of the ions is compared with their mobility as expressed by their speed of migration imder the influence of the electric current, a number of instances are found that would point to the conclusion that the greater the mobility the more intense is the taste; but there are many exceptions that show that the intensity of the taste produced by the ions is not simply determined by the readiness with which they can get at the nerve endings, but also by the reactions they undergo with the protoplasm, which of course are determined by the chemical character of the ion.

9. The intensity of the taste of solutions of substances containing amido-acid, acid-amido, alcoholic hydroxyl, and aldehyde groups was investigated, and it was found that the results obtained are in general such as one would expect viewing the mat- ter simply in the light of Overton's determinations of the relative readiness with which these substances permeate plant and animal membranes.

10. It was pointed out that in general the alkaloids have a pronounced bitter taste, that according to Overton nearly all permeate protoplasm readily and that furthermore, they are known to exert a strong physiological action on the nerves.

11. Colloidal solutions are tasteless because the substances they contain diffuse very slowly and are chemically very inert.

Laboratory for Physical Chemistry, Univ. of Wisconsin, Madison.