460 Drs. 8. M. Copeman and H. W. Hake. [June 19, 
hydrochloric acid is of any diagnostic value, or, as far as we can see, of any 
practical importance, and it is a striking fact that in 20 hospital cases 
(non-malignant), Moore found an average of 0-063 per cent. of free 
hydrochloric acid, an amount which differs but slightly from the average 
of his second series of cancer cases (0:051 per cent.). Dr. Palmer, employing 
the same method, estimated the free hydrochloric acid in test-meals from 
14 cancer patients, and found an average of 0:0217 per cent. It may be noted 
that Palmer lays special stress on the necessity of obtaining the withdrawal — 
of the test-meal without any possibility of dilution with water. Moore 
makes no reference to this point, but in Case 9 states, that instead of 
1 pint of tea, 1 pint of gruel and 2 pints of water were administered and 
withdrawn one and a-half hours afterwards. It is, perhaps, hardly surprising 
that, under these circumstances, the free hydrochloric acid obtained was only 
0-001 per cent. 
It is of importance to note, also, that the interval between the administra- 
tion of the test-meals and its withdrawal, as recorded by these observers, 
varied between the wide ranges of one to two hours, and in a single case was 
as much as three hours. This absence of uniformity as regards time of 
withdrawal clearly detracts from the diagnostic value of the results, and it is 
sufficiently obvious that where a number of experiments are being made under 
varying conditions of time, whether by one or more observers, no proper 
comparison will be possible between the results or the deductions drawn from 
them. 
We may next discuss the results of the Morner Sjéqvist estimations. This 
method is intended to estimate both the free hydrochloric acid and that 
combined with proteid or organic bases. The underlying principle is that, on 
bringing the filtered gastric contents into contact with pure barium carbonate, 
these various forms of hydrochloric acid interact with it to produce barium 
chloride, and that, on subsequent evaporation, incineration, and extraction with 
water, the amount of barium chloride formed may be estimated as barium 
sulphate and expressed in terms of the original hydrochloric acid. In 16 out 
of 34 test-meals examined, the result obtained by this method agreed almost 
theoretically with the physiologically active hydrochloric acid found by the 
method of precipitation with silver; in the remainder it was less, in some 
cases considerably so. The failure to estimate the full amount of hydrochloric 
acid in these cases we account for by the presence of mucus-like substances 
in the fluid examined which, by preventing complete contact of the solid 
barium carbonate, causes loss of hydrochloric acid during evaporation. 
The fact, however, that in several cases the method gave results strictly 
comparable with our silver determinations, entirely disposes of the objection 
