the pipet and expelled back into the test tube. This maneuver must be repeated three times to insure perfect mixing. Exactly 1 cc. of this saliva mixture of tube 2 is now transferred by means of the same pipet to tube 3. The same maneuver of mixing and transferring the diluted saliva as contained in tube 3 is repeated, and so on until finally tube 10 is reached. Again, 1 cc. of the mixed saliva of this last tube is withdrawn and discarded. Each one of the ten tubes should now contain exactly 1 cc. of fluid; the amount of saliva present in each tube differs exactly by 100 per cent. from its right or left fellow, as may be easily ascertained by reference to Table No. 1.

PREPARATION OF THE STARCH SOLUTION.

One gram of so-called soluble starch (Lintner) is carefully weighed off, transferred to a porcelain capsule or casserole, and throughly mixed with 100 cc. of cold distilled water. The mixture is then heated with constant stirring until a perfectly clear solution is obtained. After cooling, the solution, if not perfectly clear, is filtered into a graduated cylinder, and the lost portion of the water is replaced by fresh distilled water. The completed solution is transferred to a clean glass-stoppered bottle. This starch solution, if kept at a cool temperature, will remain unchanged for about a week. The addition, to aid preservation, of chloroform, toluol, etc., is not to be recommended, as there is always a liability of precipitating the starch.

PREPARATION OF DECINORMAL IODIN

SOLUTION.

Decinormal iodin solution should preferably be obtained from chemical supply houses, or it may be prepared as follows: Dissolve 12.653 grams of dry iodin U.S.P. in a solution of 18 grams of potassium iodid U.S.P. in 300 cc. of distilled water. Then add enough water to make the solution measure at ordin

ary temperature exactly 1000 cc. Transfer the solution to an amber-colored glass-stoppered bottle and keep in a dark place. For testing the starch solution under consideration, a conveniently small quantity of the decinormal iodin solution should be diluted with nine times its volume of distilled water. This N/100 iodin solution preferably is kept in an amber-colored dropping bottle.

TECHNIQUE IN DETERMINING THE AMYLOLYTIC INDEX OF SALIVA.

Solutions required.

(1) Ferment solution.

(2) 1 per cent. soluble starch solution. (3) Distilled water.

(4) N/100 iodin solution.

Ten test tubes are placed in a testtube rack and numbered in rotation with a wax pencil. The first tube receives 1 cc. of the saliva to be tested; the other tubes receive 1 cc. each of distilled water. The second tube now receives in addition 1 cc. of the same saliva. In accordance with the directions given above, the saliva solution of the remaining tubes is now geometrically diluted so that each tube will contain 100 per cent. less saliva than the preceding tube.

With a suitable 5 cc. pipet, each tube will now receive 5 cc. of the 1 per cent. starch solution, starting with the tube which contains the smallest amount of saliva, namely, tube 10.

The ten tubes are at once transferred to a suitable water-bath having a constant temperature of 38° C. This water-bath consists of an ordinary enameled metal dish, about 8 in. wide and 4 in. deep, placed upon a suitable tripod. To hold the tubes in position, a wire basket such as is used in bacteriologic work is placed in the dish together with a centigrade thermometer. The dish is filled with tap-water to within an inch from the top, and the water is heated to 38° C. By lowering the flame sufficiently, the water is kept constantly at the above temperature. The test tubes must be immersed in the warm water

of the N/100 iodin solution, starting with tube 10. A definite color reaction can be observed at once. The highernumbered tubes, say 8 to 10, will usually exhibit a deep blue color, tube 7 a bluish-red and tube 6 a Bordeaux-red color, tube 5 a brownish yellow color, while tubes 4 to 1 will show a very slight yellow color. The colorimetric manifestations indicate the reaction between the iodin and the starch solution. Completely reduced starch (maltose) gives no color with iodin. This is equally true of achroö-dextrin, while erythro-dextrin produces a Bordeaux-red color. The unaffected starch shows a blue color. (See colored plate.) Occasionally it

limes, that is, the border line between reduced and unreduced starch, and the next tube lower in the scale (brownishyellow color) furnishes the basis for obtaining the amylolytic index.

ESTIMATING THE AMYLOLYTIC INDEX.

To express the amylolytic index as deducted from the results of the above tabulated experiment, it is necessary to divide the quantity (5 cc.) of the starch solution employed by the absolute quantity (0.016 cc.) of the saliva employed. The resultant quotient represents the number of cubic centimeters of a 1 per cent. starch solution which are reduced