NECESSARY CONDITIONS FOR ACCURATE RESULTS 201 



atmospheric air whose percentage of carbon dioxide was 

 accurately determined, and, further, employed ordinary 

 daylight. Brown also used in his bell- jars only single leaves, 

 which, however, remained attached to the plant, so more 

 easily guarding against withering than was possible in the 

 severed branches.' 1 In their later researches Brown and 

 Escombe (1905) estimated the formation of carbohydrate 

 by multiplying the weight of the absorbed carbon dioxide 

 by a 'carbohydrate-factor' of 0-640. The carbohydrate 

 formed under light, per sq. m. per hour, for Helianthus 

 annuus was thus estimated to be from 0-4 to 0-5 grm. 



The carbohydrate-factor of 0-640 given above cannot 

 be taken as a constant, for Krasheninikov (1901) found it to 

 vary in different plants from o-6o in the cherry-laurel to 

 -67 in the sugar-cane. An explanation must, therefore, be 

 sought for these differences, and further inquiry made as to 

 whether the carbohydrate-factor of a given plant remains 

 constant or undergoes variation under changed external 

 conditions. 



Necessary Conditions for Accurate Results 



The difficulties in the older methods of estimation of 

 the carbohydrate formed in photosynthesis suggest that a 

 perfect method should fulfil the following conditions : 



1. The complete experiment should be carried out with 

 one and the same specimen, since the physiological activi- 

 ties of two different specimens are not the same. 



2. The increase of weight measured should be entirely 

 due to C0 2 -assimilation, and not to the production of other 

 substances. 



3. The experiment should, as far as possible, be carried 

 out under the normal conditions to which the plant had been 

 accustomed in nature. 



4. There should be no shrinkage of the specimen under 

 light in consequence of increased transpiration. 



1 Jost, Plant Physiology, translated by Gibson (1907), p. 116. 



