184 STUDIES ON RESPIRATION. II 



material within the cells, as this decrease did not occur when a 3 

 per cent sugar solution was used in place of tap water. 



All the experiments were made at room temperature (18-20°C.). 



In regard to the accuracy of the results it may be stated that in no 

 case did the probable error exceed 2.4 per cent of the mean. 



The measurement of respiration at frequent intervals, as here 

 practised, has great advantages. Methods heretofore employed in 

 measuring respiration have usually necessitated the use of long periods. 

 It will be noticed readily from curves here presented that if periods 

 of an hour's duration had been used no increase would have been 

 found except in one or two cases, because the increase, though distinct 

 enough, lasts but a short time (in many experiments only from 5 

 to 10 minutes) . This increase is followed by a decrease, and if the 

 total amount of CO2 given off in 1 hour were measured, the large 

 output for a short time would be more than counterbalanced by the 

 longer period of small output, so that the total would be below nor- 

 mal, and we should record a decrease instead of an initial increase 

 followed by a decrease. In the writer's opinion there is no doubt 

 that this is what has happened in many cases, where only a falling off 

 of respiration has been reported. Schroeder^ got a decrease in respi- 

 ration of Aspergillus niger when he used 6 and 7 per cent ether. In 

 the present investigation 3.65 and 7.3 per cent ether both gave a de- 

 cided increase (except in a few cases), but this was followed by a drop, 

 and if 1 or 2 hour periods, as Schroeder used, had been employed, it 

 is very likely that only a decrease would have been noticed. Ko- 

 sinski^ as the result of experiments with Aspergillus niger reports an 

 increase in CO2 output with 0.5 per cent ether and a decrease with 

 5 and 7 per cent; with 0.2 per cent cocaine and 0.02 per cent strych- 

 nine nitrate an increase was observed. 



The first experiments were made with formaldehyde. A few ex- 

 periments were conducted at first with 0.2 per cent (by volume), but 

 as this concentration gave a very small change, 0.4 per cent was used. 

 This stronger concentration gave a very large increase followed by an 

 abrupt decrease. The results are shown in Fig. 1. Six experiments 



^ Schroeder, H., Jahrb. wiss. BoL, 1907, xliv, 409. 

 ^ Kosinski, I., Jahrb. wiss. Bot., 1902, xxxvii, 163. 



