18 PHYSIOLOGY OF THE NEW-BORN INFANT. 



digestion, etc., Scherer and Babak draw the following quite unjusti- 

 fiable conclusion: "Die Kohlensaureproduction und der Sauerstoff- 

 verbrauch sinken beim Neugeborenen etwas in den ersten Stunden 

 nach der Geburt bis ung. zur neunten Stunde," etc. If a new-born 

 infant kicks and cries soon after birth, is afterwards quiet and sleeps 

 almost to the ninth hour, then wakes and is laid at the breast, the results 

 obtained would nearly agree with the quotation given above, but other- 

 wise not. Speck, 1 a well-trained subject for physiological experiments, 

 tripled his respiratory exchange by doing considerable work with one 

 arm. So great is the influence of muscular contractions upon metabo- 

 lism that the arrangement of a schedule like the one indicated above is 

 rather a waste of time. 



Scherer and Babdk have worked with the same respiration apparatus. 

 That their results agree quite well, therefore, is but natural. If their 

 results are incorrect, as I must assume after carrying out my own experi- 

 ments, this can not be due to the inaccuracy of their methods (although 

 the limit of error of 6 per cent for carbon dioxide is quite large, the res- 

 piratory quotient can in a given instance vary from 0.65 to 0.60), but 

 must have its reason in a systematic error. I refer to Scherer's descrip- 

 tion of his experiments, from which it is clear that in the closed respiration 

 apparatus after the experiment there may be found an increase of 541 

 c.c. beyond the calculated nitrogen quantity; the oxygen consumed is 

 replaced from an oxygen bomb; the oxygen introduced into the chamber 

 is not analyzed for its purity, but is assumed from " Hoppe-Seyler's 

 analysis" of the same manufacture to contain 4 per cent nitrogen; if, 

 in reality, it contains over four times as much (which is perhaps 

 unreasonable but always possible), this would explain the rinding of 

 541 c.c. of nitrogen in excess. If on this basis a correction is made 

 in the above case, the quotient rises from 0.63 to 0.73. 



I do not dare insist that this explanation is very probable; from the 

 account of the experiment, however, I find no other. At any rate, I 

 may say that a method with such large sources of error is too crude to be 

 compared with mine, which has given especially accurate results in the 

 hands of numerous investigators. My method of experimentation is 

 given herewith: 



The atmospheric air from outside is drawn through a water gas-meter, and from there 

 through a large spiral lead pipe, where it is warmed to the experimental temperature . The 

 infant lies naked in a 15-liter respiration chamber, which is placed within a couveuse. The air 

 is sent into one end of the respiration chamber at the top, passes out below at the opposite 

 end, and through a flask where the moisture of the expired air and the perspiration is con- 

 densed. After this the air-current is passed into a simple sampling apparatus, consisting of 

 a mercury receptacle closed with three-way taps, from which the mercury can run out in a 

 thin stream during an experiment. By this means almost continuous samples of the air- 

 current, each with a volume of 70 c.c., may be obtained; 50 c.c. of the sample are used immedi- 

 ately for an analysis in a Pettersson apparatus (with potassium pyrogallate) . The air is 

 sucked over the child by means of a water-suction pump; the speed, which is sufficiently even 

 for our purpose, with variations never over 3 per cent, is regulated according to the percent- 

 age of carbon dioxide one desires in the respiration chamber. The samples of air are not 

 taken until the infant has remained in the air-current so long that we can assume a uniform 

 composition, i. e., when three times its own volume of air has been passed out of the respira- 



'Speck, Physiologic des menschlichen Athmens, Leipsic, 1892, p. 82. 



