gastric contents, and it has been erroneously held that the presence of the free acid is of more importance than the organically combined acid. But inasmuch as the latter may have been free a short time before its estimation, the exclusive determination of "free" hydrochloric acid. cannot but lead to fallacious conclusions.

The method which we adopted, after careful consideration, for the estimation of the physiologically active hydrochloric acid, is based on Volhard's volumetric estimation of chlorides by precipitation with excess of standard silver nitrate, in the presence of free nitric acid, and the subsequent determination of the excess of silver nitrate by standard ammonium thiocyanate, using iron-alum as an indicator.

By estimating (a) the total chlorides and (b) the inorganic chlorides in an aqueous extract of the gastric contents by the above method, we obtain the physiologically active hydrochloric acid = a-b. A further determination (c) of the total acidity with standard sodium hydrate, using phenol-phthalein as an indicator, gives us in addition the free organic acids = [c-(a−b)]; all of these constituents being expressed in terms of hydrochloric acid.

We did not, at first, see our way to determining these various constituents of the gastric contents in single stomachs (owing to their small weight, frequently not exceeding 0.5 gramme), and we therefore took batches of stomachs, varying in number from 6 to 60. Later, by using or solutions in place of solutions, and by slightly varying

N

N 10

N 50 100 the details of the method, we were able to make the estimations in single stomachs, with accuracy and comparative ease.

We made, in the first instance, five series of experiments in which no account was taken of the period of digestion, the mice being simply removed from their cages while feeding.

Under the above conditions, and somewhat to our surprise, an increase in the secretion of active hydrochloric acid was indicated in mice with transplanted tumours, as compared with normal mice. Taking the average of all five series, we found that 150 normal mice gave an average of 0.1121 per cent. hydrochloric acid, while in 178 mice with transplanted tumours the average was 01752 per cent. (see Tables I., VI., VII.).

A second series of five experiments was next instituted on similar lines except that a period of one hour's digestion was selected, the mice being put in empty cages overnight and fed next morning for one hour before removal. We found, again, an average increase of hydrochloric acid in the gastric contents of mice with transplanted tumours as compared with normal mice; thus 144 stomachs from normal mice.

gave an average of 0.1488 per cent. hydrochloric acid, 183 stomachs from mice with non-ulcerated tumours gave 01627 per cent., and 52 stomachs from mice with ulcerated tumours gave 0.2100 per cent. during one hour's digestion (see Taoles III., VIII., X.).

A brief preliminary note of these results was published in The Lancet' of November 10th, 1906, and the work was referred to by Dr. Bashford in the Fifth Annual Report of the Imperial Cancer Research Fund (June 1907).

We next made a number of experiments to determine, if possible, the usual period of digestion for normal mice; but the results obtained, after periods of half an hour, one hour, and one hour and a half respectively, showed that while, on the whole, more mice attained a maximum secretion of hydrochloric acid at one hour and a half, many attained a maximum at one hour and a few even at half an hour (see Tables II., IV., V.).

Similar experiments were carried out with mice with transplanted tumours for the same periods of digestion, and we found practically the same variations during these periods but with a general tendency to increase of hydrochloric acid (see Tables IX., XI.).

Summarising all the results obtained in the experiments relating to definite periods of digestion, we found on an average that 245 stomachs from normal mice gave 01456 per cent. hydrochloric acid and 290 stomachs from mice with transplanted tumours gave 0.1673 per cent. for periods of digestion of one hour and one hour and a half.

We also examined thirteen single rat stomachs (weighing, with contents, from 2.5 to 10 grammes). Six of these were from normal rats and seven from rats with transplanted tumours (the tumours varying in weight from 0.3 to 15 grammes). The former gave an average of 0.1427 per cent. hydrochloric acid after one hour's digestion, while the latter gave an average of 0.1837 per cent. after the same period (see Table XII.).

We also had the special opportunity of examining, between May 1906 and May 1908, 15 single stomachs of mice with spontaneous tumours. These showed an average of 0.1929 per cent. hydrochloric acid during different periods of digestion (see Table XIII.).

Hence, comparing the secretion of hydrochloric acid in the stomachs of normal mice and of mice with transplanted tumours during definite and indefinite periods of digestion, we found practically the same variations throughout, but with a general tendency towards increase of hydrochloric acid in the case of mice with transplanted tumours.

The total number of experiments made, from which the above averages have been deduced, was about 150, involving from four to six estimations in each experiment.

These results are interesting not only as showing that, chemically, the digestive process in mice is comparable with that in the human subject, but also as confirming, and to some extent explaining, the observations made by Dr. Bashford in 1905 as to the absence of cachexia in mice suffering from cancer. Now, inasmuch as recent extensive statistics, collected by the Imperial Cancer Research Fund from the various London Hospitals, have shown that cachexia is not a constant accompaniment of cancer in man, we might expect to find, in its absence, the same compensating influence as regards increased or undiminished secretion of hydrochloric acid in the stomachs of human beings afflicted with this disease.

As regards this point, however, so far as indicated by the investigations of various scientific workers, the opinion is prevalent that the reverse obtains; Moore, Palmer, and others having indeed asserted that there is a marked diminution or even an absence of hydrochloric acid in the gastric contents in malignant disease of organs other than the stomach.

These conclusions, however, have been mainly based on estimations of "free" hydrochloric acid only, in the gastric contents, as represented by the fluid withdrawn from the stomach after the administration of "Test Meals"; but as we have shown, such estimations may easily lead to fallacious conclusions.

We felt that inasmuch as our determinations of physiologically active hydrochloric acid in the gastric contents of mice seemed to be in contradiction to Moore's results as regards "free" hydrochloric acid in human gastric contents, it was obviously necessary to repeat Moore's experiments, and, at the same time, for purposes of comparison, to carry out a parallel series of estimations by the same method that we had employed in our work on mice.

We therefore examined the gastric contents as represented by the fluid withdrawn one hour after administration of test-meals to 34 patients suffering from cancer, which by the courtesy of the Surgeons to the Cancer, Middlesex, and Westminster Hospitals, we were able to obtain from time to time.

We estimated, among other constituents, the "free" hydrochloric acid by the inversion of methyl acetate, the method employed by Moore The Test-Meal in all cases consisted of one pint of tea and a large round of toast it was given in the morning, fasting, and withdrawn one hour afterwards.

and others; and the physiologically active hydrochloric acid by Volhard's method, as already described, which we had used in the experiments with mice-stomachs.

Summarising these experiments we found that our estimations of "free" hydrochloric acid by the methyl acetate method, more or less agreed with those of Moore in his later experiments published in the 'Biochemical Journal' (Vol. i. p. 274), where, as an average of 13 cases of cancer, he found 0.0515 per cent., our average, also for 13 cases, being 00407 per cent. Moore, however, had found in 13 previous cases (Roy. Soc. Proc. Ser. B, vol. 76, p. 138) an average of only 0.0039 per cent, and Palmer in 14 cases found an average of 0·0217 per cent. Such discrepancies are not re-assuring, and some explanation may be found in the varying periods of digestion selected by these workers, viz., from 1 to 2 hours, but in our opinion a more probable explanation is afforded by the evanescent character of the so-called "free" hydrochloric acid to which we have referred above.

On the other hand the determination of the physiologically active hydrochloric acid in our 13 cases averaged 0.1626 per cent., and in five of them was above 0.18 per cent. (see Table XIV.)

We also found that a considerable difficulty often arose in connection with the withdrawal of test-meals, and that it was a not altogether uncommon practice to add water to assist the withdrawal. Obviously the amounts of hydrochloric acid found in such diluted test-meals and the conclusions deduced from them can be of little or no value. We therefore refrain from publishing the details of these estimations (21 in number). We cannot, of course, from the comparatively few experiments made by us on the human subject, assert that the secretion of hydrochloric acid in cancer is normal, or above normal, for in a few cases it was decidedly below normal, but we think it has been sufficiently indicated that the whole question of amount of hydrochloric acid in human gastric contents in cancer would repay further investigation, and that it is not justifiable in the absence of more comparable experiments to conclude that it is always greatly diminished in the presence of this disease.

In conclusion, we desire to express our thanks to the Executive Committee of the Imperial Cancer Research Fund, from whom we received a pecuniary grant which rendered possible the completion of this research. To Drs. Bashford and Murray also, we are much indebted for the trouble they have taken in providing us with the necessary material, and for the interest they have shown in the progress in the work.