558 Messrs. G. A. Buckmaster and J. A. Gardner. [Mar. 7 T 



junctions should be avoided as far as possible. In other experiments- 

 a Woulff's bottle of the following description was used instead of bags. 

 Capacity of the bottle 580 c.c. ; volume of chloroform 170 c.c. ; surface- 

 area 43 sq. cm. ; volume of air-space 410 c.c. ; diameter of inlet-tube, which 

 was surrounded by a water jacket of the same temperature as the main 

 bath, 2*5 cm. ; diameter of outlet-tube 1*5 cm. The bottle was placed in a 

 large bath so that it could be kept at any desired temperature. The 

 apparatus was so arranged that the distance of the lower end of the inlet-tube 

 from the surface of the chloroform could be varied at will. The outlet-tube 

 was connected with the Chauveau-valve apparatus by means of a densimeter 

 bulb 250 c.c. in capacity. By this means the percentage of chloroform in the 

 inspired air could be ascertained. This apparatus proved in many respects 

 satisfactory ; not the slightest drag on the respiratory movements was noticed, 

 a matter of great importance in any experiment of this kind, and particularly 

 so when the duration of an experiment is prolonged. We have found that 

 when the normal movements of respiration were mimicked by means of a 

 bellows apparatus, which aspirated a mixture of air and chloroform vapour 

 from the Woulff's bottle, considerable variations, amounting to as much as 

 50 per cent, of the higher values, were noticed when the rate or amplitude or 

 both were varied. Actual experiments on cats showed that the variations 

 which occur during anaesthesia were much less marked. Thus, during one 

 experiment lasting 20 minutes, the following percentages were obtained : 1*8, 

 2*1, 2, 1*6, 1'7. In another experiment of 24 minutes' duration, the values at 

 intervals were as follows : 4*1, 3'6, 4'5, 3*1, 2 - 95, 3*8, 3*75. For the first 

 experiment the average was l - 8 and for the second 3'8. 



In the experiment (No. I), asphyxial convulsions began within a minute 

 of the cessation of respiration, and with the continued inhalation of chloro- 

 form the convulsions recurred at intervals until 4.23. The heart ceased to 

 beat at 4.28. 



The data obtained from this table are given in a graphic form in Curve I r 

 while in Column 7 of the table the type and frequency of the respirations, at 

 the time each sample was taken, are given. 



In the experiment (No. II), chloroform was commenced at 2.20 p.m., but 

 before the end of the first minute the animal ceased to breathe. The inhala- 

 tion was at once stopped, the animal recovered and the experiment restarted 

 at 2.45, by which time the chloroform would have been eliminated. This early 

 and sudden cessation of respiration we have repeatedly noticed with cats, and 

 it is not a mere holding of the breath, as in other experiments it was often 

 found necessary to use artificial respiration in order that the animal should 

 recover. After this has happened, under apparently precisely similar 



