STUDIES ON EXPERIMENTAL AND CLINICALHYPOCHLOREMIA IN MAN

Joseph B. Kirsner, … , Walter Lincoln Palmer, KathrynKnowlton

J Clin Invest. 1943;22(1):95-102. https://doi.org/10.1172/JCI101373.

Research Article

Find the latest version:

http://jci.me/101373-pdf

STUDIES ON EXPERIMENTALAND CLINICAL HYPOCHLOREMIAIN MAN1

By JOSEPHB. KIRSNER, WALTERLINCOLN PALMER,AND KATHRYNKNOWLTON

(From the Frank Billings Clinic, Department of Medicine, University of Chicago, Chicago)

(Received for publication July 21, 1942)

INTRODUCTION

In a recent study (1), it was demonstrated thatsevere hypochloremia and alkalosis could be in-duced gradually in the dog, by means of continuedloss of chloride, without the development of amarked azotemia or detectable renal injury.Kerpel-Fronius (2), Kerpel-Fronius and Butler(3), Ambard (4), and Hiatt (5) had previouslymade similar observations in the dog and rabbit.The purpose of the present investigation was todetermine whether or not the same situation couldbe induced in man.

METHODOF STUDY

I. Experimental hypochloremiaChloride deprivation was produced rather slowly by

Wangensteen aspiration of the gastric contents, for vary-ing lengths of time in 2 adult men (Cages M. S. andW. L.) with pyloric obstruction secondary to duodenalulcer.

The patients received no food or fluid by mouth duringthe period of chloride withdrawal; 3000 to 4500 cc. of 5per cent glucose in distilled water were administeredintravenously each day. The serum chloride, carbondioxide, pH, and blood urea nitrogen were measured fre-quently (6). Renal function in case M. S. was deter-mined by the urea clearance test (7). Measurementswere made also of the serum calcium (8), phosphorus(9), and potassium (10). The cell volume was deter-mined frequently by hematocrit readings. The plasmaproteins were determined by the Hanna-Campbell method(11), a modified macro-Kjeldahl technique being em-ployed for the nitrogen analysis; 4 per cent boric acidwas used to absorb the ammonia released. The bloodsodium was measured by a modification of the Butler-Tuthill method (12); the phosphate of the serum wasremoved by ignition with ferric sulfate and the sodiumdetermined on an aliquot from which the insoluble ironsalt had been removed by centrifugation. The urinarysodium was determined in case M. S. by the same tech-nique as that used for the serum sodium, modified as tothe volume used; ferric salt was added in amounts suf-ficient to prevent interference by phosphates. The total

'This study was supported in part by a grant fromJohn Wyeth and Bro., Inc., Philadelphia, Pennsylvania.

base was measured by a modification of the Stadie-Rosstechnique (13); the serum was ashed as for the sodiumanalysis; the ash was taken up in 45 cc. of water, thebases precipitated with benzidine hydrochloride, and thedifference between the titration of the acidity of thisfiltrate and that of the original reagents used as a meas-ure of the total base.

The patients' body weights were checked frequently.The fluid intake and output were recorded daily. ThepH of the urine was determined with a Beckman pHitheter; the chloride content of the urine was measuredby the Van Slyke-Sendroy method (14); and the am-monia plus ammonium salts determined by the techniqueof Henriques and S6rensen (15). The volume, pH, andchloride content of the aspirated gastric secretions weremeasured similarly.

Aspiration of the stomach was discontinued after asevere but safely obtainable state of chloride depletionhad been achieved. The body chloride was then replacedgradually in case M. S., over a period of 26 days, by theaddition of increasing amounts of salt. The body chlo-ride was replaced more rapidly in case W. L. (within 6days) by the parenteral administration of saline solution.

II. Hypochloremia incidental to ulcer therapyThe problem was studied further in 10 patients with

alkalosis and chloropenia complicating the treatment ofpeptic ulcer. While two factors were operative in thislatter group, i.e., chloride loss and alkali ingestion, thesecases are included because of the absence of nitrogenretention and because the evidence indicated that themajor, indeed, if not the only, factor in the acid-basedisturbance was the hypochloremia.

RESULTS

I. Experimental hypochloremiaCase M. S. (Unit no. 204878 (Figures 1 and

2). A 38-year-old salesman had taken alkalisintermittently for 5 years for the relief of symp-toms from a stenosing duodenal ulcer. He hadbeen vomiting for 5 days prior to hospitalization.Five per cent glucose in both normal saline anddistilled water was administered intravenouslyduring the first 4 days. Continuous Wangensteenaspiration of the gastric contents was begun onthe 5th day and continued for 96 hours. The

95

JOSEPH B. KIRSNER, WALTERLINCOLN PALMER, AND KATHRYNKNOWLTON

0

09

NORMALRANGETOTAL BASE1/Ifffffff/ff/f/it/ff

0 NORMALRAN

0

0i0

0

2 ,/,,,,/,//,//1-A,,///

0

_R

0

a

0

'0

so

. 0

n o

V.to

O4

IGE BLOOD'/rn/mi/m,iX//// zSODIUM/// , ,, ,X

TOTASBASE (u.Ee./L.) SERUMK(40Mt.13.4-4.9 Eo/L|1SERUMCA9. " -1.6- M.GP.C.BsLOOD..SODIUm (u.EQ/L.)

SERUMPROTEINS (G4S.P

HEMATOCRIT

~ BLOOD UREA NITROGEN (MG PER CENT)FE ~~~~~~~~~~~~~~~~~~~~~~~~~~~.S._.-...**._..|tI/,////g

SERUMC0,

EXPERIMENT1. THE EFFECT OF A GRADUALLYINDUCED HYPOCHLOREMIAON

THE SERUMELECTROLYTESIN MAN

gastric contents removed each day were acid, thepH ranging from 1.84 to 2.70; the amounts varieddaily from 1100 to 3090 cc. Wangensteen suctionwas then limited to a period of 10 hours daily,the amounts withdrawn ranging from 440 to 1200cc. The intravenous intake of fluid during theperiod of gastric aspiration consisted of 5 per centglucose in distilled water, except for 1500 cc. nor-

mal saline given on the 12th day. This repre-sented the only chloride (8.1 grams) taken by thepatient during the interval from the 4th to the16th hospital days. The salt intake was then in-creased very gradually until discharge from the

hospital on the 39th day, at which time the pa-tient was receiving 3.7 grams of chloride daily.This amount was increased to 5.3 grams daily one

week later, and the chloride intake was main-tained at this level until the experiment had beenconcluded.

The lowest serum chloride of 53.8 mM./L. was

obtained on the 11th day. The maximum bloodurea nitrogen of 30 mgm. per cent was obtainedat a time when the serum chloride measured 63.6mM./L.; the values ranged usually from 14.2 to19.6 mgm. per cent. The blood sodium decreasedto 115.2 m.eq./L. and the total base to 133.3

96

FIG. 1.

A/J/z//z//z////A/zzzzzzy//X/zzzA>/zAX//S

Y//////j.W,/T,NORMAL RANG wbh..* ,

EXPERIMENTALAND CLINICAL HYPOCHLOREMIA

m.eq./L. The serum calcium and potassium werenormal. The urea clearance was not lowered sig-nificantly. The patient lost 23 pounds in weight;the presence of dehydration was further indicatedby the increased hematocrit readings and the highplasma protein values.

The serum electrolytes, hematocrit, and plasmaproteins gradually returned to normal during theperiod of chloride replacement. The blood ureanitrogen ranged from 8.1 to 14.3 mgm. per cent.At the conclusion of the experiment, the patient'sbody weight had returned to its original level.The daily volume of urine varied from 775 to 1920cc., usually exceeding 1000 cc.; the pH rangedfrom 5.9 to 6.85. Its chloride content was mark-edly reduced; no chloride was found in the urine

on 5 days. Sodium excretion paralleled the valuesfound for chloride ion. Ammonia excretion inthe urine, as was to be expected, was diminished.

A histamine test performed on the 19th day(serum chloride between 65.5 and 68.8 mM./L.)revealed normal gastric secretion; the pH of thevarious samples of gastric juice ranged from 1.31to 1.71. An examination of the spinal fluid onthe 25th day (serum chloride between 68.5 and78.8 mM./L.) revealed a clear fluid under normalpressure. The chloride measured 100.4 mM./L.and the total base 141.5 mM./L. (both values mod-erately reduced). The calcium was 4.8 mgm. percent and the urea nitrogen 16.7 mgm. per cent(both values normal).

The patient was very weak and listless during

M S UNfl mO. 2MALE AGE38

DUODENALULCERWITH OBSTRUCTION

0

in 0-.*0 BODY WEIGHT (KG)* -

--

NORMALRANGE /7y IULE//////CAO

UREA CLEARANCE (PER CENTAV NORMAL)

DILY FLUID INTAKE AND OUTPUT(URINE I GASTRtIC I0) ( CC)ASPIRATIONS

in~~~~~~~~~~~~~~~~~~iE

DAILY CHLORIDE INTAKE I AND CHLORIDE LOSS IN GASTRIC AND URINE (GMS.)CONTENTSI a

I a''

-t d6- m6 Lql ot1I"X-COMPLCTZAMECMCOC@CLORIKIN URIE

II ISI 2IS SDEC.

FIG. 2. EXPERMENT1. ADDITIONAL DATA. HYPOCHLOREMIAIN MAN

B0

V

0

AA a

III10Nin

0

in

0

.0

26 31NOOCT NOV

0

97

JOSEPH B. KIRSNER, WALTERLINCOLN PALMER, AND KATHRYNKNOWLTON

° \ ,/BLOOD SODIUM (u.EaJL.)

f, SERUMI CALCIUM (MG. PER CENT)SERUMPHOSPHORUS

NORMALRANGE(m Eo/L.) /SERUMsPOTASSIUM

BLOODUREANITROGEN (MG. PER CENT)0

o ,///////// ///////////////// g///. NORMALRANGE Y/////////////////////////,

o M//////// ,////j</// / NORMALRANGE /

00

SERUMCHLORIDE

(\AMIL)

DU1WITI

0%D

. L UNIT NO a5418

MALE AGE 39

)ODENAL ULCER'H OBSTRUCTION

SERUMpH, s

S Mu_e~~~~~~~~~~~~~I I . I I I . I f

26 31DEC. JAN

4ORMALRANGESERUMCOa AND p Y / /

5 10 15 20 25 30

FIG. 3. EXPERIMENT2. THE EFFECT OF A GRADUALLYINDUCED HYPOCHLOREMIAON THE SERUMELECTROLYTESIN MAN

the period of chloride depletion. He had no appe-

tite and complained of a loss of the sense of taste.Mild muscular twitchings appeared on the 12th day(serum chloride 53.8 mM./L.) at which time thepatient received the 1500 cc. normal saline intra-venously, as noted above. The blood pressure,originally 130/95, decreased during the period ofhypochloremia; the systolic values ranged from90 to 110 and the diastolic from 60 to .72. Theblood pressure at the completion of the experimentwas 126/84. The patient's clinical course duringthe period of gradual chloride replacement was

characterized by a steady gain in strength andimprovement in appetite.

Case W. L. (Unit no. 254018) (Figures 3 and4). A 39-year-old male had experienced ulcersymptoms for several years. He entered the hos-pital because of a recent massive hemorrhage and

marked pyloric obstruction. Therapy consisted ofWangensteen aspiration of the gastric contents,3 blood transfusions, and, for the first week, 1500to 3000 cc. of 5 per cent glucose in normal salineeach day. On the 8th day and for the next 5days, the patient received 5 per cent glucose indistilled water. An average of 1860 cc. of gastriccontents were aspirated daily; the pH of the as-pirates ranged from 1.94 to 2.58. Wangensteensuction then was discontinued; the patient, how-ever, vomited large amounts of gastric contents onseveral occasions. Saline therapy was resumed onthe 14th hospital day and the patient was givenfrom 3000 to 4500 cc. of 5 per cent glucose innormal saline and distilled water daily. Aspira-tion of the stomach was resumed on the 20th daybut the electrolyte loss now was replaced by theuse of parenteral fluids.

/11, /// 1// ".

98

a

,............,,,,,,.,,,,,,

EXPERIMENTALAND CLINICAL HYPOCHLOREMIA

The serum chloride decreased to 58.9 mM./L.The maximum blood urea nitrogen was 23.1 mgm.

per cent, with the values ranging usually frqm13.1 to 21.2 mgm. per cent. The lowest bloodsodium was 129.2 m.eq./L., obtained on the dayof the lowest serum chloride. The serum cal-cium, phosphorus, and potassium were normal.Adequate urea clearance studies were not obtainedin this case. The patient lost 8 pounds in weight.The plasma protein and hematocrit values were

elevated. The volume of urine varied from 660to 4000 cc. daily, usually approximating 3000 cc.;

the pH ranged from 6.41 to 7.48. The chloridecontent of the urine was markedly reduced.

The patient gradually became very weak andlistless during the period of chloride depletion.His blood pressure, originally 125/92, decreasedslightly; the systolic ranged from 90 to 116 andthe diastolic, from 68 to 90. These manifesta-tions disappeared rapidly, however, as the acid-base balance was restored.

Comment. The maximum blood urea nitrogenvalues of 30.0 and 23.3 mgm. per cent noted inthese 2 cases are in sharp contrast to the severe

c - A, I I I ,I I SDEC. JAN.

FIG. 4. EXPERIMENT2. ADDITIONAL DATA. HYPOCHLOREMIAIN MAN

99

DAY OF LOWESTSERUMCHLORIDE

e (.e1"'> pH OF URINE

N

N; @__pH GASTRIC ASPIRATIONS

O*-PXBODY WEIGHT (KG)

t., PLASMA PROTEINS (GM PER CENT) W.LN.rNo.21*MALE AGE-39

0 9 < _ DUODENALULCERHEMATOCRIT WITH OBSTRUCTION

**.n \ ~~~~~~HEM@ATOCRIT_

8 DAILY FLUID INTAKE1 ANDOU PUT (URINE I GASTRIC 0) (C.C.)ASPIRATIONS

040O 600 CC. BLOODTRANSFUSIONS00

ODALYHLOIDE NTAE BANDCHLORIDELOSS IN GASTRIC |AND URINE (GhMS.)0

I- l 181L d' aa,

0~~~~~w ****{ .

JOSEPH B. KISNER, WALTERLINCOLN PALMER, AND KATHRYNKNOWLTON

nitrogen retention observed by other workers inpatients with hypochloremia and alkalosis sec-

ondary to pyloric obstruction, or during experi-mental salt deficiency produced by diet and sweat-ing (16), but are in agreement with the findings

of Kerpel-Fronius (2) and Kerpel-Fronius andButler (3). Dehydration was present in both pa-

tients as evidenced by the significant weight lossand the high hematocrit and plasma protein values.

Since large quantities of glucose in distilled

TABLE I

HypochJoremia withot asotemia during alkalosis(10 Patients)

Add-base balance Gaspiration

Cage | Semm CO pH | BuN Urea Day of treatment TherapySex-Age Serm Se pHmclearance Average Num-ci cot p daily ber of

amount days

mM./L. mM./L. e man.c c.J. S. 65.9 38.0 7.52 18.4 90 2nd 622 4 Calcium carbonate; parenteral

264531 92.1 32.0 7.47 3rd fluids; gastroenterostomy on 8thM-26 92.2 35.1 7.56 16.9 110 5th day.

J. C. 79.0 38.6 7.62 21.2 113 4th 393 11 Sippy powders; NaCI (4 to 6185870 67.6 38.2 7.63 14.6 200 7th grams) started on 6th day.M-29 85.2 32.4 7.50 8.8 91

D. H. 98.4 26.3 7.47 3rd 217 16 Calcium carbonate; NaCI (5112315 80.4 36.9 7.62 10A 85 11th grams) started on 14th day.M-45 92.8 35.2 7.47 7.9 15th

J. K. 90.0 29.9 7.39 15.0 41 4th 207 20 Calciukn carbonate; no NaCl.198489 86.2 35.0 7.46 9.1 67 12thF-55 89.2 34.9 7.50 13.7 53 16th

T. S. 94.8 32.8 7.48 14.5 6th 148 16 Calcium carbonate; no NaCI.212524 81.8 7.54 19.6 12thM-49 84.8 36.4 7.49 15.2 18th

97.6 31.5 7.47 10.4 29th99.0 28.8 7.48 8.1 33rd

F. W. 96.7 31.7 7.56 12.9 143 2nd 177 17 Sippy powders NH4CI (4 to 6194370 82.8 37.3 7.55 12.4 9th grams) 15th to 22nd days.M-43 82.6 37.5 7.53 12.2 93 14th

98.8 29.5 7.38 11.4 200 18th96.6 31.2 7.44 22nd

W. B. 31.8 7.48 18.7 96 7th 354 43 Calcium carbonate; NaCI (10211777 80.6 47.4 7.73 20.6 13th grams) 8th to 22nd days; vari-M-51 85.6 41.1 7.62 18.7 16th ous alkalis from 25th to 38th

98.1 28.0 7.49 8.8 128 28th days; calcium carbonate then96.8 32.4 7.49 16.9 54 40th resumed.97.2 34.2 7.45 18.3 69 47th

15.3 148 1 year later

H. B. 90.4 30.6 7.50 23.2 4th 356 17 Calcium carbonate; NaCI started225751 83.0 39.4 7.62 18.7 9th on 10th day; 200 cc. 5 per centM-49 97.0 35.0 7.47 15.7 85 13th saline intravenously on 10th

101.8 31.0 7.47 11.7 76 17th day.12.0 102 4 months later

S. S. 84.3 42.9 7.71 17.3 83 15th 85 20 Sippy powders; NH4CI (2 to 6101209 30.9 7.51 15.0 18th grams) from 16th to 20th days.M-37 95.2 36.0 7.53 22nd

100.2 32.9 7.50 24th11.1 112 1 week later

A. G. 96.0 31.1 7.54 11.7 2nd 100 15 Calcium carbonate; NaCl (3232838 89.4 34.4 7.47 4th grams) started on 12th day.M-52 87.8 35.6 7.48 22.4 8th

87.4 36.7 7.49 14.2 10th91.9 35.1 7.52 11.7 80 13th

100

EXPERIMENTALAND CLINICAL HYPOCHLOREMIA

water were administered daily, it would appear

that, as in the dog experiments, dehydration didnot exist in the sense of deprivation of water butwas present in the sense that fluid could not beretained in the body due to the marked depletionof electrolytes. The volume of urine excreted byboth patients during the period of hypochloremiawas large; the daily output averaged 1340 cc. incase M. S., and 1317 cc. in case W. L. The urea

clearance in case M. S. was not altered signifi-cantly, and the absence of a marked elevation ofthe blood urea nitrogen is further evidence ofsatisfactory renal function. Despite the severe

chloropenia, the pH and chloride content of thegastric aspirations were reduced only slightly; andthe pH values of the histamine-stimulated gastricjuice in case M. S. were as low as one finds in thehistamine-stimulated gastric secretion of indi-viduals with a normal acid-base balance. Thisfinding is in accord with the observations made indogs (1). The decrease in the spinal fluid chlo-ride and total base noted is, of course, to be ex-

pected; McCance (17) and Agar and MacPherson(18) have reported similar findings. It is ofinterest that these values were not lower thanthey were, in view of the severe hypochloremia andhyponatremia. The explanation for this discrep-ancy is not apparent.

II. Hypochloremia incidental to ulcer therapy

In Table I are recorded the acid-base balanceand other data in 10 patients with alkalosis andhypochloremia, complicating the antacid treatmentof peptic ulcer. Large amounts of gastric contenthad been aspirated in 8 of this group. In no in-stance was there any significant azotemia. Hypo-chloremia apparently had occurred gradually in all10 patients.

GENERALDISCUSSION

This study confirms the results of the animalexperiments referred to earlier, as well as the workof Kerpel-Fronius, and Kerpel-Fronius and But-ler, and demonstrates that, in man also, a severe

hypochloremia is not necessarily accompanied byazotemia. The important factors which determinethe extent of urea nitrogen elevation during hypo-chloremia, apparently, are the speed with which

the chloride loss is induced, and the degree andrapidity of the dehydration associated with thechloropenia. An abrupt depletion of chloride isassociated with an equally abrupt and marked lossof sodium and, therefore, of body water, as shownby Gamble and others (19). Loss of fluid de-creases the circulating blood volume which, byreducing the venous return to the heart, pre-sumably lowers the cardiac output. This courseof events may actually lead to peripheral circula-tory failure. The blood flow through the kidneyand the effective glomerular filtration pressure areconsequently lowered. The severe nitrogen reten-tion and decreased renal function, noted underthese conditions by other workers, evidently repre-sent a decreased renal circulation rather than in-trinsic renal disease. The absence of azotemiaduring hypochloremia in the present experimentsmay be attributed to two factors: (a) the gradualdeprivation of chloride, allowing the patients tomake fairly satisfactory adjustments to the severeelectrolyte changes, and (b) the daily administra-tion of large quantities of water which tended to"wash out" the urea nitrogen and apparently main-tained an adequate blood flow through the kidney.

CONCLUSIONS

1. Severe alkalosis without marked nitrogen re-tention may be induced in man by the gradualwithdrawal of gastric secretion.

2. Such hypochloremia and alkalosis are notassociated with decreased renal function, as meas-ured by the urea clearance test, when adequatequantities of fluid are administered daily.

3. Gastric secretion in man is not altered signifi-cantly by severe hypochloremia.

The authors wish to acknowledge the valuable technicalassistance of Miss Jacqueline Front.

BIBLIOGRAPHY

1. Kirsner, J. B., and Knowlton, K., Acid base balance,renal function, and gastric secretion during hypo-chloremia in the dog. J. Clin. Invest., 1941, 20,303.

2. a. Kerpel-Fronius, E., Zur pathogenese der "hypo-chloramischen" Azotamie. Ztschr. f. d. ges exper.Med., 1936, 97, 733.

b. Idem, Salzmangelzustande und chloroprive Azo-tamie. Ergebn. d. inn. Med. u. Kinderh., 1936, 51,623.

101

JOSEPH B. KIRSNER, WALTERLINCOLN PALMER, AND KATHRYNKNOWLTON

3. Kerpel-Fronius, E., and Butler, A. M., Salt and waterlosses in diuretin diuresis and their relation toserum non-protein nitrogen and electrolyte concen-trations. J. Exper. Med., 1935, 61, 157.

4. Ambard, L., Stahl, J., and Kuhlmann, D., Azotemieet chloropenie. Arch. d. mal. d. reins, 1933, 7, 465.

5. Hiatt, E. P., Extreme hypochloremia in dogs inducedby nitrate administration. Am. J. Physiol., 1940,129, 597.

6. Peters, J. P., and Van Slyke, D. D., QuantitativeClinical Chemistry, Vol. 2, Methods. Williams andWilkins Company, Baltimore, 1931. Serum CO2,p. 283. pH (Colorimetric), p. 796. Chloride, p.835. Blood urea nitrogen, p. 554.

7. Van Slyke, D. D., and others, Observations on thecourses of different types of Bright's disease andon resultant changes in renal anatomy. Medicine,1930, 9, 257.

8. Clark, E. P., and Collip, J. B., A study of the Tisdallmethod for the determination of blood serum cal-cium with a suggested modification. J. Biol. Chem.,1925, 63, 461.

9. Fiske, C. H., and Subbarow, Y., The colorimetricdetermination of phosphorus. J. Biol. Chem., 1925,66, 375.

10. Shohl, A. T., and Bennett, H. B., A micro method forthe determination of potassium as iodoplatinate. J.Biol. Chem., 1928, 78, 643.

11. Campbell, W. R., and Hanna, M. I., The albumin,globulins, and fibrinogen of serum and plasma.J. Biol. Chem., 1937, 119, 15.

12. Butler, A. M., and Tuthill, E., An application of theuranyl zinc acetate method for the determinationof sodium in biological material. J. Biol. Chem.,1931, 93, 171.

13. Stadie, W. C., and Ross, E. C., A micro method forthe determination of base in blood, serum, and otherbiological materials. J. Biol. Chem., 1925, 65, 735.

14. Peters, J. P., and Van Slyke, D. D., QuantitativeClinical Chemistry, Vol. 2, Methods, Williams andWilkins Company, Baltimore, 1931, page 367.

15. Henriques, V., and S6rensen, S. P. L., Ueber diequantitative Bestimmung der Aminosauren, Poly-peptide und der Hippursaure im Harne durchFormoltitration. Ztschr. f. physiol. Chem., 1909,63, 27.

16. McCance, R. A., and Widdowson, E.. M., The secre-tion of urine in man during experimental salt de-ficiency. J. Physiol., 1937, 91, 222.

17. McCance, R. A., The effect of salt deficiency in manon the volume of extra-cellular fluids and on thecomposition of sweat, saliva, gastric juice, andcerebrospinal fluid. J. Physiol., 1938, 92, 208.

18. Agar, H., and MacPherson, I., The cerebrospinal fluidin alkalosis. Lancet, 1940, 1, 171.

19. Gamble, J. L., and Ross, S. G., The factors in thedehydration following pyloric obstruction. J. Clin.Invest., 1925, 1, 403.

and McIver, M. A., Acid base composition of gastricsecretions. J. Exp. Med., 1928, 48, 837.

, Extracellular fluid. Bull Johns Hopkins Hosp.,1937, 61, 151.

102