[Reprinted from the Boston Medical and Surgical Journal of 
May 23, 1889.] 
THE REACTION OF/THE BLOOD.1 
BY JOHN A. JEFFRIES, M.D. 
While much attention has been paid to the 
formed elements of the blood, but little compara- 
tively has been given to its chemistry. The hrst 
question to suggest itself from this side has re- 
ceived but a sparing amount of attention in Europe 
and none at all in America. I refer to the reac- 
tion of the blood. We read that the blood is alka- 
line and that is about all. To be sure, we hear 
much about an acid diathesis in rheumatism and 
gout, unsupported in the first case by any evidence. 
In gout Garrod has clearly showed the part played 
by uric acid, that is, its prevalence in the system 
and especially in the affected parts. 
While working in the Pathological Institute of 
Vienna from the winter of 1884 to 1886, I was 
struck by the peculiar general distribution of the 
tubercle bacillus, as represented by tubercular pro- 
ducts found at autopsies. 
One or another of the products of tuberculosis 
were to be found in the majority of the subjects 
which came to the table, and this in spite of the 
fact that those known to be dead of phthisis mostly 
went to the anatomical department. Yet tuber- 
culosis of the muscles is relatively rare ; it is by 
no means easy to procure examples of tuberculosis 
in muscular tissue for microscopic preparations. 
1 Read before Boston Society of Medical Sciences, March 19,1889. 2 
Pathological changes are to be found, due, as E. 
Frankel points out, to the general effects of the 
disease, but the bacillus is not so easily found. 
The same is true of the stomach and.to a less but 
distinct degree of the genito-urinary tract. In any 
case of acute miliary tuberculosis the little glisten- 
ing, transparent tubercles are to be found scattered 
along the blood-vessels throughout the rest of the 
body. 
Why this difference ? Why are organs of such 
diverse structure as the mucosa of the stomach, 
the muscles, and the genito-urinary organs compara- 
tively exempt ? It might be said that the histol- 
ogy of the muscles was very distinct from that of 
other parts, and so not adapted to the development 
of tuberculosis ; but the same cannot be said of 
the stomach, while the mucosa of the intestines is 
so often radically affected. 
These parts have, nevertheless, one character in 
common which separates them broadly from the 
rest of the body. They are exposed to an acid* 
reaction. During a good part of the time the 
mucosa of the stomach is bathed in an acid juice 
secreted by itself; the same is true of the urinary 
tract and of the muscles. The urine is mostly 
acid in reaction, the muscles change during action 
from an alkaline to an acid reaction. This is 
v-asily shown by inserting bits of litmus paper in 
any relaxed muscle; the red bits become blue. 
Now cause sharp contractions by applying a faradic 
current to the nerve above, and the blue paper will 
shortly turn back to red. 
As opposed to this, all know that as a general 
law bacteria grow poorly or not at all on media 
of animal origin with an acid reaction, and any 
one who has grown the tubercle bacillus knows 
that it forms no exception to the rule. The bacteri- 3 
ologist who neglects the reaction of his media 
will soon have cause to rue. 
Putting these two together the question, Is the 
reaction of the tissues one of the leading causes of 
the distribution of tuberculosis, naturally presents 
itself and may in a general way be extended to the 
whole group of infectious disease. 
To investigate this, naturally brought up the 
question as to the reaction of the blood: could it be 
changed and does it vary in different kinds of 
animals ? 
On returning from Europe during the latter part 
of 1886 and first part of 1887 I took up the subject 
in co-operation with Dr. James J. Putnam. Later, 
by pressure of more urgent work and the difficulties 
inherent to all scientific work in this country, we 
were both of us driven from the subject. At this 
time we stood alone in the question, and Dr. Put- 
nam was inclined to believe,in view of Meyer’s work, 
that our methods could not give any reliable results. 
As, however, the subject has been since taken up 
by Jaksch,2 it seems to me that a report of our 
methods and results on testing the reaction of blood 
may not be without interest. 
Of the chemical composition of the blood little if 
anything is known. We have analyses of the ash 
of the corpuscles and serum, and the elements in 
them, but how these are combined is a matter of 
speculation. We are therefore dealing in reality 
with an unknown mixture. 
To test the reaction of the blood two different 
methods have been pursued, one by the direct use 
of litmus or other indicator, the other by inference 
from the amount of carbon dioxide that can be ex- 
tracted by the air-pump, either alone or with the 
* Jaksch. Ueber die Alkalescenz des Blutes bei Krankheiten. 
Zeitschrf. f. klin. Med., 1887, p. 350. 4 
addition of a certain amount of acid, usually phos- 
phoric. We pursued the direct method. 
A moment must be given to what this method 
gives us. All reagents as to reaction are merely 
indices, and do not run, as many seem to suppose, 
parallel to each other and chemical composition. 
The parallelism is only general, not absolute. Of 
these indices litmus seemed to be the best adapted 
to our purposes. It shows both actions, acid and 
alkaline, distinctly, can be used on paper, and is 
fairly sensitive. 
Alkaline bases turn the color blue, and acids red, 
if they act. Some, as uric acid, act as neutral bodies 
—have no effect on the color. Salts follow more or 
less closely after their composition: thus most acid 
salts turn the paper red, alkaline ones blue, while 
still other are neutral, have no effect. But this is 
not invariably the case: bicarbonate of soda is an 
acid salt, that is, the acid element is not satisfied, 
is capable of joining with more base, yet its reac- 
tion is that of an alkali to litmus, which it blues 
strongly. 
In this paper, therefore, in common with all others 
who have written on the subject, by alkaline is 
meant the property to blue litmus. 
Owing to the haemoglobin and the coagulation of 
the blood when drawn from the body, testing the 
reaction offers peculiar difficulties, and various ex- 
pedients have been resorted to. Thus Kuhne8 tried 
to separate the fluid by means of dialyzation; Lie- 
breich 4 and others used a very thin plate of pure 
gypsum, through which the serum of the blood 
passed, and could be tested either by litmus paper 
3 Kuhne. Die einfaches Verfahren die Reaction hamaglobinhalti- 
ger Flusgigkeiten su priiten. Virchow’s Archiv, xxxiii. p. 95, 1865. 
< Liebreich. Eine Method sur Priifung der Reaction thierische 
Gewebe. Berichte der deutsch. chim. Gesellsch. zu Berlin, 1868, 
p. 48. 5 
or directly by previously soaking the tablets 
in a neutral solution of litmus. Lassar,5 Canard,6 
Mya and Tassinari,7 Landois,8 and von Jaksch all 
used litmus paper,some noting reaction from the part 
soaked beyond the clot, others washing the blood 
off. At times the paper was previously moistened 
with a salt solution to keep the haemoglobin off. 
To prevent coagulation the blood has by most 
recent observers been mixed with a neutral 10% 
solution of sulphate of soda. 
As an acid for titration phosphoric,9 tartaric, and 
oxalic have been used in very weak solution. Phos- 
phoric is ill adapted, since in small quantities it 
gives but a poor reaction with litmus; the salts are 
often amphoteric. Oxalic I have distrusted on ac- 
count of its strong affinity for lime. 
The blood has been drawn in various ways, either 
in considerable quantities from the arteries of ani- 
mals, or a few cubic centimetres from the finger in 
man by means of a lancet, or a drop from a needle 
prick, as practised by Landois. 
Landois and von Jaksch used a large number of 
solutions of sulphate of soda plus a given amount of 
acid, and added a definite amount of blood by means 
of a pipette, and then noted, after stirring, which 
was neutral; read off the acid strength. The other 
authors slowly added the acid solution and tested 
from time to time. This is inadmissible, since the 
blood diminishes in alkalinity very rapidly after 
being drawn. All have used a small amount of 
5 Lassar. Zur Alkalescenz des Blutes. Pfliiger’s. Archiv, ix. p. 44, 
1874. 
6 Canard. Essaie sur Falcalinite du Sang dans l’etat du sante efc 
dans quelques maladies. Paris, 1878. 
7 Mya, Tassinari. Sulle variazioni della reazione alcalina del sangue 
venoso. Archiv per le scienze mediche, ix. p. 379, 1886. 
8 Landois, Ileal-Eucyclopadie der gesammten Heilkunde, 2 Aufl. 
iii. p. 161, 1885. 
9 Zuntz. Zur Kenntniss des Stoffwechsels im Blute. Centralb. f. 
d. med. Wiss., 1867, p. 801. 6 
blood, but agree that the results are the same as 
with larger quantities. 
The method used by us, worked out in the rough 
by Dr. Putnam before I joined him, is as follows :10 
A small drop, .02 c.cm., of blood taken from a drop 
produced by a prick of the back of the finger is 
mixed with a given quantity of a standard dilute 
solution of tartaric acid and all then placed on a 
piece of very smooth-painted, dry litmus paper. 
This is then washed off in neutral water, and the 
color, where the drop has been, noted. As a rule 
both colors of litmus papers were used. The whole 
process takes from fifty to ninety seconds. If the 
first test does not come out right, a second, with 
more or less acid as indicated, is to be made. 
As our experience increased the method was 
perfected and various sources of error eliminated. It 
may therefore be well to give the whole in detail. 
Our pipettes are made from apiece of thermometer 
tubing of rather coarse bore, ground to a point at 
one end, from which marks at definite intervals, 
each equal to about .01 c.c., run, up to twenty. On 
the other end is slipped a piece of rubber tubing 
about three inches long and stopped by a solid, glass 
rod. By means of this rubber tube any desired 
amount of fluid can be drawn up into the tube and 
the end then dried with a bit of absorbent cotton. 
The acid solution is made by taking a sxfav solu- 
tion of corrosive sublimate and adding sufficient 
pure crystalline tartaric acid, so that fifteen parts of 
the acid solution exactly neutralized eight parts of 
the volumetric solution of soda of the United States 
Pharmacopoeia diluted with thirty-nine times its bulk 
of water. The acid salt of mercury is added to 
prevent the growth of the lower plants, which other- 
wise quickly occurs. 
10 This method is similar yet distinct from that of Landois, and 
more portable. 7 
From six to seven measures of this acitl solution 
are drawn into the tube and well up, the lower six 
being left empty, and the tip of the pipette dried. 
Around the second joint of the linger, carefully 
cleaned, a light rubber band is passed, and a prick 
with a surgical needle made in the back of the last 
phalanx behind the quick of the nail. From the 
drop of blood which quickly comes two measures are 
drawn into the pipette, and the tip wiped with cotton. 
Then blood and acid are squirted into a small vial 
and mixed by rapid alternate sucking and squirting 
of the pipette. Lastly, the mixture is drawn up and 
deposited on bits of the litmus paper previously 
pinned onto corks. At the end of five seconds the 
corks and paper are plunged into water and the color 
noted. 
The whole is to be done in much less time than 
it takes to describe it, and is quite accurate. The 
acid can be measured off at leisure: haste only comes 
after the blood is drawn. Any slight surplus of 
blood can be quickly removed by a pledget of ab- 
sorbent cotton. Some sources of confusion are to 
be noted. The blood is bright red; therefore if the 
observer gazes intently at it while on the paper, he 
will after washing it off see the complementary color, 
irrespective of the color of the paper. This is 
really nothing but an experiment on negative after- 
images which is out of place. Again the observer 
must be careful, especially if near-sighted, not to 
breathe upon the paper, as it will quickly become 
affected by the carbonic acid expired by the lungs. 
The test paper must be handled with clean forceps, 
not touched by the fingers, which are acid and will 
affect it; the same is true of almost any moist vege- 
table product, such as wood. 
The color of the paper must be promptly noted, 
since it fades on standing. 8 
This method is accurate, and I believe gives us 
about all that is to be gathered concerning the direct 
reaction of the blood. Over the older methods it 
has the advantage of being applicable time and time 
again to anybody, as the writer knows by personal 
experience, and portable. 
It gives about fifteen different degrees of alka- 
linity between a neutral reaction and the higher 
alkaline reactions noted. From the known ratio of 
blood and acid used the amount of acid required to 
neutralize 100 c.cm. of blood can be quickly counted. 
Thus, if two parts of blood and seven of acid are 
used we have 100 c.cm. of blood = 350 c.cm. of 
acid = °f volumetric solution of soda = 
4.0 = .186 gramme of pure NaOH. As the blood 
can be made a fixed quantity, two parts, a working 
table as given below can be made, where the first 
number is the number of acid parts taken and the 
last the amount of alkalinity measured by NaOH. 
in 100 c.cm. of blood. 
/ar/s 
&/. /iStf 
3&?-3s 
&t. jfsS 
tvp, 
3> //tOff-m 
Sj/ ASsaJ 
/oaA 
A*/i 
/?a.. am 
3 /////** 
/73 /3sS 
7 
■S3 
26 
> 
3.33 
3U 
2. 
3.06 
33 
3 
*26 
3/3 
3 
/60 
>?■ 
9 
7.30 
2 90 
. 9 
2.33 
306 
30 
3T33 
266 
3 66 
333 
33 
.rti 
293 
6. 
330 
360 
-AS* 
.226 
It is to be noted here that Meyer,11 a strong advo- 
cate of the C0o method of determining the alkalinity, 
has strongly attacked all efforts to determine the 
11 Meyer. Studien liber die Alkaiescenz des Blutes. Arcliiv f. 
exper. Path. u. Pliarm., xvii. p. 304, 1883. 9 
reaction by direct testing. He points out that the 
blood is of unknown composition, hence we do not 
know what we are dealing with; and, second, that 
many complex substances do not act on litmus as 
indicated by their chemical affinities. 
While not claiming that any direct method gives 
an absolute measure of the bases, it seems to the 
writer, in common with von Jaksch and others, that 
the measure of alkalinity as directly observed is of 
value from both clinical and physiological stand- 
points. 
We do not know just what the blood is composed 
of, but we do know that it is pretty much the same 
thing all through air-breathing vertebrates, and is 
therefore fairly comparable. 
The other objection, as pointed out by Mya and 
Tassinari, is a play on words. Nobody for an in- 
stant supposes that in measuring the reaction of a 
complex mixture he is obtaining an absolute measure 
of the alkaline or acid bases, as defined by chemical 
philosophy, which remain unsatisfied, but knows he 
is making an empirical observation of the degree of 
reaction as indicated by the index used. 
The true objection rests in the fact that the blood 
begins to lose its alkaline reaction very shortly 
after being drawn. Empirically, however, it can 
also be shown that but little diminution occurs in 
the first ninety seconds, when mixed with the acid 
mixture used. The corrosive sublimate tends to 
hold the blood unaltered. 
As for the C02 method, it is, for general applica- 
tion, utterly out of the question. It does not give 
any idea of the quantity of unsatisfied bases, and is 
based on a whole train of suppositions. Reducible 
to the fact that no other explanation of the reten- 
tion of the CO 2 in the blood has been given than a 
grip exerted by the unsatisfied bases, and the fact 10 
that acids fed to animals reduce the C02 in the 
blood, presumably by satisfying the bases. 
In defects the method is rich, among them the 
fact that the CO., in the blood is much influenced 
by the respiration, entirely independent of the alka- 
line bases present, as shown by Minkowski.'12 
In spite of this the method is of value, and to the 
writer’s thinking both methods are increased in 
value by the fact that the results obtained by care- 
ful observers by the two methods closely coincide. 
In studying the alkalinity of the blood we have' 
but varying degrees of reaction to compare,— possibly 
at times a change to acidity just before death, — that 
is, changes in quantity only. Qualitative changes 
are only to be got in changes of condition of the 
animal observed. Thus poisons, complete change 
from natural diet, and the like, may be made. Again 
the urine requires careful attention as being the 
chief scape-vent for the blood, and therefore influ- 
enced by the varying conditions of. the blood. But 
the first step is to acquire an idea of the normal 
conditions of the blood, and its relations to the 
urine and food, time of day, age, and sex. 
In determining normal conditions the CO? method 
is of very little value: continuous observations can- 
not be made, and the operation of bleeding throws 
the animal at once out of normal conditions. This 
is not true of our method: a simple prick does no 
harm, and the experiments can be conducted upon 
man. 
The figures of Meyer, Feitelberg,18 and Minkowski 
run, for each kind of animal tested, within a 30% 
limit of variation, and making due allowances for 
differences in the complex technique, justify their 
12 Minkowski. Ueberden Kohlensaueregehalt des arteriellen Blutes 
beim Fieber. Archiv f. exper. Path. u. Pharm. 
*3 Feitelberg. Ueber den Eintlnss einige Gil'te auf die Alkalescenz 
des Blutes. Diss. Inaug. Dorpat, 1883. 11 
assumption of a normal standard of CO._, in the 
blood. By the direct method Lassar, Mya and 
Tassinari, and Canard have endeavored to find 
standards, — the first in rabbits and cats, the last 
two authors in man. Reduced to a common stan- 
dard Lassar found — 
German rabbits, 100 c.cm. of blood = .; 188 NaOH. 
French “ “ “ 212 “ 
Cats, “ “ 241 “ 
the variations being within 15 per cent, from the 
mean. Mya and Tassinari place the average in 
man at .4, a figure much higher than those of any 
other observer. Canard gives .228 as the average, 
.272 as the maximum, and .203 as the minimum 
of fourteen healthy adults. My own figures run 
constantly much lower: thus rabbits at .160 ; frogs 
at .70; hens at .200; man, from 5 healthy indi- 
viduals and one hundred observations, at about .200. 
max. .250; min. .160. Von Jaksch figures in disease 
from .036 to .350. 
These figures, it will be seen, are very different 
from one another, and preclude all deductions as to 
quantity. They, however, in the case of each ob- 
server, give relatively similar results. These dif- 
ferences are doubtless due to the different acids 
used, the nature of the blood, arterial or capillary, 
the time the blood had been drawn, and the qual- 
ity of the litmus paper. Ordinary litmus paper, 
such as most of the authors used, contains a good 
deal of acid or alkali, which tends, when using small 
quantities of blood, to greatly increase the reading. 
My own observations have so far been directed 
to ascertaining the normal standard, and especially 
any variations, diurnal or from day to day. Below 
is given a set of figures, taken from one person, re- 
duced to milligrammes of NaOH to 100 c.cm. of blood. 
B, L, and D denote the last meal. Breakfast at 12 
?.3£>a4r. f/f U & 7/f 
A, />$,. ff itO 
(, />& & 
V t2?h 7/r 6 7?f 
//30 £ //f 
3 Jo />$> ff 
2 30 />& /c/f ff **r 
3, 30 />& "/r J t/o 
3 fffc /3/f d tvs' 
/2 //atm, ff 2fO 
3 /% w * m 
z /% **A w 
# 30 at> tyr 0asr 7to 
? jo V? ’> 7oo 
2, />& <2 770 
S />* ty £ *'* 
2 St fy XT /U 
* />**} £ *73 
//3e //?4t, "7? 4 000 
/ />X 'fy 7? M* 
3 />* "/f <? **» 
t3*¥-A "/? **° 
f.jo ** '¥? ff 
/ffS* /V? * *70 
// 6 #>. 4to ff t/3 
fJO 0*t %o iff too 
2 />& V/o ff *73 
9 02 ft Mo /ff 733 
/t » #0 tfi * '*+ 
//.30#4i 7//o £ /ft 
2. f0>. % iff *** 
7 a.m., lunch at 
p.m. when taken. 
The'regular dinner 
hour was 7 p.m.,too 
late for observa- 
tions to be made 
after it. On the 
two occasions 
where D is put 
down it was taken 
earlier in the after- 
noon. 
The above fig- 
ures look rather 
formidable, but 
with a knowledge 
of the life of the 
observer seem to 
clearly show one 
general rule, name- 
lo, a gradual rise 
in the reaction 
from rising to be- 
fore dinner. Thus 
the average on ris- 
ing is .155; be- 
tween 9 and 11 a.m. 
.189; between 1 
and 3.30 p.m., with 
lunch, .191 : with- 
out lunch, .212. 
These averages in- 
clude all the figures 
where a regular life 
was being led, in- 
cluding about five 
miles of walking 
and the ordinary 13 
46 fa 4 Jtf* 
XJd fa 4 2/3 
f/t f?6 fa 4 /ft 
0 /"&, t//o 0 /ft 
44?> foo 0 2/3 
Z />£ #0 4 **a 
/0 fa 4 /U 
ifo 0 z/j 
//. 06 fa /? 2/3 
£ 06 fa 4 /ft 
f 06 fa 4 /ft 
/a 46 fa 4 /ft 
a./r 0Z> 'fa 0 /tt 
zso 06 tfa 0/ //0 
f./r 06 fa 0 /ft 
2 06 fat 4 2<X> 
7,3t 06 /fa 0 /ft 
Z/o 06 'fa JT /ft 
//, 46 /fa A /ft 
2 06 /fa /? /ft 
/2t 06 /fa 0 2/3 
2 06 /fa 0 2/3 
// /fa 0 /ft 
2/f~ 'fa 0 2/3 
?/f 06 /jz> 0 //* 
2. 06 /#£> /? /J0 
9 46 fa 4 2/3 
2 06 fa 0 &3 
f 4A, fa 0 
0 46 *fa /46 
? jo 46 j /;j 
work of an out- 
patient service. 
The fact that the 
reaction was 
higher in tjie early 
afternoon of days 
when no lunch was 
taken than when 
lunch was taken is 
of interest; also 
the two measure- 
ments taken after- 
dinner,which show 
a lower reaction 
than before dinner. 
These two facts 
seem to stand in 
direct opposition 
to Canard’s and 
Sticker and Heu- 
ber’s14 observa- 
tions, who found 
that the alkalinity 
rose after meals. 
The rise they at- 
tribute to the loss 
of the hydrochloric 
acid poured into 
the stomach. 
Canard’s figures 
I cannot explain; 
those of Sticker 
and Heuber were 
done by Zuntz’ 
method with phos- 
phoric acid, which 
is unreliable, and 
Sulphate of quinine was taken in ten-grain dose on 10-10 at 9.15 
a.m., and 17-10 at 10 a.m. 
h Sticker u. Heuber. Ueber Wechselteziehungen zwiechen Secre- 
ten unu Excreten des Organismus. Zeit. f. klin. Med., Bd. 12, p. 136 
foot-note. 14 
4 3, *fyo A /rt 
3. /r /•#, a*rf0 zao 
}.2o /?**/■ /tO 
/o. fy/ z/3 
4 #/ SX6 
/a a a, J /u 
-t />& W/ 2/3 
2. 30 />#, £ /U 
/o.yy 3a// fy £ /u 
A, <6 /yt 
/ 3 a /°#> /yt 
A 44, % /;j 
0.30 $z yS /ft 
0./a />%, '//x. £ /ft 
2. />% "/z /yt 
y />& "/* £ /4o 
A yd 220 
/. d /?<> 
/.yr />% *V*. y$ //3 
z.yr £ /J3 
3, * yd a *yx /rt 
tfs- 44 /# f/ 4 At 
//ZO <£f y$ /ft 
/./’%'% 0 • /ft 
& X /tO 
\ 
// 43, /to 
A-36 *4 4, *¥z /to 
//Aoi/ta, /jj 
s*o*/?% *y<i /ao 
the time of day is 
not given; they 
may therefore have 
been dealing with 
the early morning 
rise, which is clear- 
ly due to the more 
active life. As the 
reaction of the 
chyme in the in- 
testines is alkaline, 
it is just as reason- 
able to assume that 
the blood should 
be reduced in re- 
action during di- 
gestion, especially 
as absorption is 
rapid and the 
result of ordinary 
food on the acid 
side. 
The reaction of 
the blood the writer 
believes to be of 
considerable i m - 
portance to the 
economy, and 
therefore not very 
variable in perfect 
health. 
Noting the fact 
that the alkalinity 
was very apt to be 
high at times when 
headache over one 
or both eyes, with 
tenderness of the 
Antifebrin was taken in four-grain doses 31-1, 87,10.47 a.m. ; 5-2, 
10-10 a.m., : 11-2,12.30, and 2 p.m. ; 22.2,10-10 a.m. and 1 p.m. ; 13.3, 
8-45,11.20, and 2.20. 15 
nerves, existed, tests were made with two of the 
drugs found to relieve the pain. Sulphate of 
quinine in ten-grain doses was found in three 
trials to depress the reaction as noted. The same 
was true of antifebrin in five trials. As the normal 
tendency of the reaction was to rise during the 
time of these experiments, it seems highly prob- 
able that the observation is correct. Perhaps the 
headaches were gouty in nature. 
The reaction of the blood in poisoned animals, as 
indicated by the amount of C0.2, has been studied 
by several observers, and found to be diminished in 
septic fever, iodin, mercury, nitrite of soda, toluy- 
lendiamin,oxalate of soda, and strychnine poisoning. 
Alcohol and salicylate of soda apparently increase 
it. 
In disease the writer regrets that‘he has been 
unable to procure sufficient material. Several 
writers, however, have turned their attention in 
this direction. There are many old records of the 
blood being acid in cholera and leukaemia, but 
most have been taken from the corpse and so are 
worthless. Cantani claims to have found the blood 
acid in advanced cholera cases during life, and Mya 
and Tassinari state that this was found to be the 
case in Naples during the last epidemic. The 
method of determining the reaction is not given in 
either case. 
Canard found the reaction increased in acute and 
subacute articular rheumatism (it is not clear if 
salicylate of soda was given), and diminished in 
arthritis deformans. Diabetes mellitus and cancer 
of the stomach also showed a diminution. 
Mya and Tassinari found the reaction increased 
in pneumonia, treated by baths and stimulants, 
diminished in typhoid, cancer, phthisis, Bright’s, 
diabetes, oligaemia, and chlorosis.  17 
Jaksch has made by far the most extensive study 
of the subject, and found a distinct diminution in 
the reaction during fever, quickly recovering after 
it. The most marked changes were found in uraemic 
poisoning, destructive diseases of the liver, and 
leukaemia, where the blood becomes almost neutral. 
The other blood diseases offer a less marked dimin- 
ution. 
How the standard is maintained in the blood is 
not thoroughly known. But in normal conditions 
animals maintain an alkaline reaction against the 
final result of the ingesta, which result in an acid 
surplus. The acid is thrown off by the lungs (C02) 
and in the urine. The weight of evidence seems 
to be that the acid does not go through the blood 
as such, but combines and is separated in the kid- 
neys in part, in part goes out as a whole. There- 
fore by feeding with sulphuric acid it is possible to 
materially reduce the alkali in the blood; that is, 
rob the system, of course with disastrous results. 
Under these conditions much ammonia appears in 
the urine of carnivora but not in the urine of 
herbivora. Reasoning from this, Minkowski con- 
cluded that the blood should diminish.in alkalinity 
in the latter, and has' found this to be true of sep- 
tic fever. 
Scattered as the above is, it seems to the writer 
to offer a field for productive investigation. We 
know the reaction does change, has a normal value, 
cannot be without effect on the system, and have a 
ready means of determining the reaction.