1908.| Hydrochloric Acid in the Gastric Contents in Cancer. 457 
carried out in this manner, and the hydrochloric acid estimated therein was 
found to vary between 0:036 and 0°1059 per cent. 3 
The only conclusion we could come to from the results of the analyses of 
test-meals where a known quantity of water had been used in the withdrawal, 
is that each and all of them represent a quantity of hydrochloric acid which 
must be considerably less than the actual quantity originally present, but how 
much less it is impossible to say. We therefore once more discussed the 
matter with Mr. Leaf, who fully appreciated the point, and we were able by 
the exertions of Mr. Allen, his house surgeon at the Cancer Hospital, to obtain 
nine further test-meals which had all been withdrawn without the addition of 
any water whatsoever. That the difficulty of obtaining undiluted test-meals 
is not exaggerated is indicated in a statement made to us by Mr. Allen, who 
says that on an average, where test-meals had been given, two out of three 
cases had to be abandoned owing to the impossibility of inducing the with- 
drawal of the fluid from the stomach of the patient. 
Taking now the whole of the 34 test-meals from cancerous patients, we 
found in five of these in which the amount of water used in the withdrawal 
was unknown, the average amount of hydrochloric acid was 0°0646 per cent., 
with a minimum of 0°0552, and a maximum of 0:0895 ; in 16 cases in which 
the amount of water used in the withdrawal was known, the average amount 
of hydrochloric acid was 0°0554 per cent., with a minimum of 0:0127 per cent. 
and a maximum of 0°1059 per cent. 
But in the 13 cases where the test-meals were withdrawn without the 
addition of any water whatever, in two instances only was the hydrochloric 
acid as low as 0:064 and 0:073 per cent.; in another exceptional instance it 
was as high as 0°35 per cent. (cancer of liver), while in the remainder 
it varied between 0°084 and 0°22, seven of these being over 0°16 per cent., 
with a total average in all 13 cases of 0°1626 per cent., as the following 
summary (Table G) will show. 
As showing the influence of the addition of water in the withdrawal 
of the test-meal on the results obtained, the following figures will be of 
interest :— | 
Case 8 (Cancer Hospital, October 21, 1907), no water used. Found 
0°2227 per cent. HCl. 
Case 11 (same patient, October 30, 1907), 10 oz. water used. Found 
00748 per cent. HCl. 
Case 24 (Middlesex Hospital, January 28, 1908), 20 oz. water used. 
Found 0:0127 per cent. HCl. 
Case 25 (same patient February 4, 1908), .5 oz. water used. Found 
0°0858 per cent. HCl. 
