alone, and the swelling even in the greatest attenuation of the acid was much less than in water. The limiting effect of the calcium chloride was also very marked.

TABLE IV.

HYDRATION OF MIXTURES OF AGAR 3 PARTS, GELATINE 2 PARTS at 14° C. Plates 0.18 mm. in thickness; swelling of sections given in thickness and in volume.

HYDRATION OF MIXTURES OF GELATINE 3 PARTS, AGAR 2 PARTS AT 14° C. Plates 0.18 to 0.19 mm. in thickness; swellings given in thickness and in volume.

The gelatine-agar mixture being a dominantly albuminous mixture, swelling in acid increased with the concentration which was carried to a PH value of 2.01. On the other hand potassium chloride exerted an effect parallel to that shown by its action on agar, the greatest swelling taking place at the lowest concentration with a PH value of 5.7, the increase being much greater than in the acid at the higher concentration.

In general the living cell masses taken from growing organs are dominantly pentosan, but some material has been examined in which the hydration reaction is that of a dominantly albuminous biocolloid. No conception of living matter in plants not including some of the all-pervading common salts is possible, and any attempt to make a complete picture of the colloidal material of the cell must

take into account the compounds of the fatty acids with the common bases, the soaps which as McBain and Salmon have recently shown may exist as both electrolytes and colloids in colloidal masses."

These soaps are an almost inevitable component of protoplasm, and some studies of their possible action in the cell will be taken up in a paper now in preparation. Preliminary to any profitable consideration of the soaps it is necessary to have some definition of the parts which salts may play in the biocolloidal machine. The measurements of swellings given in earlier papers showed that the incorporation of nutritive salts in colloidal masses lessened the hydraiton capacity. It is now apparent from the results given on the following pages that such restrictive action was due to the high concentrations employed. This however cannot be said of the amino-compounds, which used as hydrating solutions accelerated swellings, but which incorporated in colloidal masses uniformly reduced hydration capacity in whatever concentration used.

In the tests which are to be described it was planned to include the salts which are of importance in nutrition, which induce accelerated swelling in agar and agar-gelatine mixtures in implied concentrations, the calcium and sodium furthermore being used in approximately balancing proportions. These salts were first used with purified agar and hydration values as in Table VI were obtained.

TABLE VI.

HYDRATION OF AGAR AND SALTS IN SALT SOLUTIONS.

5 g. agar, 100 c.c. KC1 0.001M, 60 c.c. NaCl at 0.0001M, and 10 c.c. CaCl2 at 0.0001 M. Sections 0.2 mm. to 0.27 mm, in thickness swelled at 14° C.

The total swelling in the present instance is one which is equivalent to that shown by many preparations including that of agar 7 Jour. Amer. Chem. Soc., 42: 426, 1920.

Plate A with which the salt tests described in the previous pages were made. With reference to the hydrogen ion concentration the solution of salts added to the agar had a PH value of about 5.8 to 6, the agar alone being about 6.5.

The salted agar showed a swelling in even the attenuated solutions of calcium and magnesium chloride less than in water. Such solutions accelerate swelling in pure agar. The swelling of the salted agar in sodium solution was as 102 with water at 100, while that in the potassium solution was 132. That this inequality is not simply a matter of hydrogen ion concentration is evinced by the fact that the swelling in the acid at PH 4.2 was practically equivalent to that of the potassium chloride at 5.7.

The incorporated solution may be regarded as approximately balanced and when the sections were swelled in a similar solution the maximum of the series was reached, the increase being as 150 to water as 100.

An identical solution was used in making up "Bacto-gelatine and the results of the hydration of sections of the dried plates are as in Table VII.

Hydration of gelatine plate made up gelatine 5 g., 10 c.c. KC1 at o.or M, 6 c.c. NaCl at 0.001 M, and CaCl2 10 c.c. at 0.0001 M, and water 25 c.c. at 14° C. Sections 0.24 to 0.26 mm. in thickness.

It is to be recalled that this gelatine had a Ph value of 5.2 and that the incorporated salts a Ph of about 5.8 to 6, so that when the dried sections were placed in a salt solution of this constitution the swelling was less than in water and that a similar decrease was noted in the calcium and potassium solutions in 0.0001 M and 0.001 M solutions and in the calcium solution at o.or M also. Increases

took place in the potassium solution at PH of 6.6 and in the acid at PH of 3 and 2. It is evident, without a detailed analysis of these results, that many features beside hydrogen ion concentration are involved.

All of the foregoing tests had as their chief purpose the determination of the reactions of plasmatic constituents and the experiments were extended to include the swellings of a biocolloid including both pentosans and gelatine with the addition of salts, but without the third colloidal component, the soaps, which we now have ample reason to believe play a very important part in the mechanism of living matter. A mixture of gelatine 3 parts and agar 2 parts was made in the usual manner with the addition of salts as in the preparations of agar and of gelatine separately. The increases when hydrated in various solutions were as noted in Table VIII.

The more prominent reactions are those of the gelatine element as would be expected in accordance with which the highest concentration is in the most concentrated solution of the acid. It is notable that as in the gelatine-agar salt-free plate depression occurred in acid at PH 3 and that at PH 4.2 the swelling was still below that of water. Practically applied to the living cell this would mean that hydration lessened with increasing acidity to the region of PH 3 beyond which the active cell would rarely go. Sodium and calcium exerted similar effects, but the swelling in potassium at 0.0001 M was equivalent to that in water while at 0.001 M it was greater. A review of the results presented in this paper would show many special results from the action of this salt.

The recently published results of experiments upon the coagu

lating action of neutral salts upon plasmatic colloids by Tadokoro furnish some important collateral conclusions. Tadokoro found that all chlorides of aluminium, barium, strontium, calcium and magnesium cause coagulation of plasmatic colloids in an increasing series in the order given in concentrations from N/200 to N/ 10, and that KCl exerted a stronger action than NaCl. Many profitable comparisons may be made between the results of his tests of the action of salts upon crushed galls and of the swelling reactions of biocolloids given in the present paper. Furthermore both afford many parallels with the swelling reactions which were obtained by the extension of my experiments to include the measurement of the action of the salts, particularly the chlorides, upon living and dead cell-masses.

The first material tested was taken from the large "prop" roots of corn plants a meter in height growing in the garden at Carmel. Root-hairs were only sparingly developed on a terminal section 3 to 5 cm. long, and sections 4 or 5 mm. in length including the tip were taken and freed as completely as possible from particles adherent from the loose sandy soil in which they grew. The average thickness of trios placed under the auxograph ranged from 2 to 2.5 cm. Such sections dried down to thickness of 0.3 to 0.4 mm. when placed between sheets of blotting paper.

Such sections of living material made practically all of the changes in volume indicated within eight to fifteen minutes, the speediest action taking place in the acid and the slowest recorded. being in the potassium solution, although this matter is partly a function of the size or length of the sections. In illustration of the swelling and shrinkage it may be cited that in the acid the sections return to normal size in about 2 hours. Shrinkage to original thickness may in some cases take place within a day, although twice this period elapses in other instances.

The results of the auxographic measurements are given as average percentages of original thickness in Table IX.

If the facts in this table are taken for comparison with those

Tadokoro, T., "Kolloidchemische Forschungen ueber das Pflanzenplasma," Jour. Coll. of Agric., Hokkaido Imp. Univ., Sapporo, Japan, 7: part 5, 144-182, 1919.