56 



MEMOIRS NATIONAL ACADEMY OF SCIENCES, VOL. X, NO. 3. 



the record has been made, the blood is again allowed to flow through the heart, and so with 

 alternate periods of feeding and starving the heart continues its movements as long as may be. 

 It is evident that the conditions under which the heart is doing its work are again unnecessarily 

 unnatural, and further, that it is not possible to make a continuous uniform application of the 

 drug under study to the heart. 



Fourth. A possible source of fallacy is found in the difficulty of closing the stopcocks F 

 and L at exactly completed diastole, for it is evident that if in one record the heart had been 

 thoroughly dilated when closure was made, and in the other record the heart had been lacking 

 even 10 per cent of full dilatation, the rise of the mercurial column would be different in the 

 two records, although the real power of the heart is the same during the two periods. 



The four considerations given above seem to us sufficient to prevent reliance on drug exper- 

 iments made with Kroneker apparatus, except when the influence of the drug is so overpowering 

 as to overcome the limit of error due to the conditions of the trial. 



Flu. 1. 



In the Williams apparatus a double cauula is placed in the ductus arteriosus and the whole 

 heart is employed, so that some of the difficulties of the cardiac study are avoided. Two forms 

 of the Williams apparatus have hitherto been in use, and have been supplied to the university 

 laboratories by the Harvard Apparatus Company. 



In the first of these forms of apparatus the blood flows through the heart during the whole 

 experiment, avoiding in this way one of the difficulties of the Kroneker apparatus, and the 

 graphic record is made of the movements of the salt solution in the closed receptacle K. in which 

 the heart is placed. Two methods of study have been practiced in regard to these movements. 

 In one method the movement of the salt solution is simply measured by the eye, as the column 

 pulsates backward and forward over a graded scale. In the other method the movements arc 

 imparted to a comparatively heavy column of mercury which records them upon a revolving 

 drum. In the Harvard apparatus, fig. B, with which we experimented, the graphic method was 

 employed. 



