PULSE DURING MENTAL AND PHYSICAL WORK. 211 



that is greatest at the end of the session. The evidence is less clear 

 after dose B. But in neither case does it appear clear that a real climax 

 of the acceleration effect has been reached in the 3-hour session. 



The question of cause can not be answered from our association-pulse 

 data. These data alone can not even answer the question whether the 

 relative acceleration is a general consequence of the ingestion of alcohol, 

 or a consequence that is peculiar to a special kind of moderate mental 

 activity after taking alcohol. Both of these questions need the addi- 

 tional data from the pulse-records during the other experimental 

 processes. 



THE EFFECT OF ALCOHOL ON THE PULSE-RATE DURING WORD- 

 REACTION AND FINGER-MOVEMENT EXPERIMENTS AND ALSO 

 DURING MODERATE MUSCULAR ACTIVITY AND REST. 



In accordance with our general program (Appendix I), pulse-records 

 were taken at a variety of homologous points in the course of every 

 experimental session. In the light of the results, it would doubtless be 

 desirable if these records, like those taken during the association experi- 

 ments, could have been more numerous, or perhaps even continuous. 

 That tliis was not arranged for was due partly to the enormous addi- 

 tional labor and expense that would have been entailed, partly to the 

 technical difficulties, and partly to the belief that shorter records 

 covering several respiration rhythms at homologous points in the experi- 

 mental session would contain sufficient pulse data to indicate clearly 

 the effect of alcohol on the pulse frequency during the experiments. 

 With respect to continuous records, it is obvious that they would be 

 useful only if homologous moments of the session were clearly indi- 

 cated, and only the records of such moments might be compared. Our 

 "sample" records, as we may call them, are theoretically as adequate 

 as any comparable phases of continuous records could be. The only 

 advantage of continuous records would be that the number of phases 

 could be indefinitely extended. 



It will be obvious, to all those who have struggled with the difficulties 

 of securing pulse-records during muscular activity, why sphygmo- 

 graphic devices which depend on air transmission were not even con- 

 sidered for the present group of records. The pulse-recording technique 

 first used in these experiments was the Dodge telephone-recorder from 

 the temporal artery, our method I. Later we took electro-cardiograms 

 from body leads through condensers, our method III. Both methods 

 call for the use of a string galvanometer as a recording instrument. 

 Both methods are equally accurate, but in simplicity of adjustment, 

 and dependability under all sorts of conditions, we believe that the 

 latter method has no equal for recording the pulse-rate. 



