26 
Ingvar Jorgensen and Waller Stiles. 
carbohydrate factor of 0-640. This factor they obtain from the 
analyses in regard to carbohydrate content of leaves of Tropceolum 
viajus by Brown and Morris as described in the last chapter. Of 
course Brown and Escombe assume that the ratio between the 
various carbohydrates remains constant and further that carbon 
dioxide is used only in the production of carbohydrates. It is true 
that variations in the ratio of the various carbohydrates will make 
little difference in the carbohydrate factor, and similarly, the error 
introduced owing to the probable incorrectness of Brown and 
Morris’s analysis (cf. Chapter V) is likely to be small. On the other 
hand the error introduced by the assumption that the whole of the 
carbon dioxide is used in carbohydrate formation is likely to be 
larger, but on this subject our information is very incomplete. It 
may be interesting to compare this carbohydrate factor with 
values obtained by experiment for the ratio between increase in 
dry weight and carbon dioxide absorbed. The following table is 
due to Krasheninikov (1901) and although the values are probably 
not of a very high order of accuracy, they may give some idea of the 
variations likely to occur. 
Table XXXVI. 
Increase in Dry Weight of Leaves per gram of Carbon dioxide 
absorbed. 
Bamboo 
• • • • • • 
0-60 
Cherry Laurel 
• • • • • • 
0-60 
SugarCane 
• • • • • • 
0-67 
Lime 
• • • • • • 
0-74 
Tobacco 
• • » • • • 
0-68 
Thoday (1909) compared the 
increase in 
dry weight with the 
increase in carbon content of the 
leaf in the 
cases of Helianthus 
tuberosus and Cherry Laurel. His results indicate a considerable 
variation in the ratio of carbon increase to increase of dry weight 
in leaves of the same species, but in Thoday’s experiments a good 
many factors are not controlled, and it is impossible to say what 
causes the variation. We should like to emphasize that in all such 
cases in plant physiological researches where it is sought to determine 
the absolute value of a quantity, it is absolutely imperative to 
determine the probable error of the experiment. This, as far as we 
know, has not been done in any single instance in work on carbon 
assimilation. 
In regard to the direct determination of the products of 
assimilation, the principle of this method was first exposed by Sachs 
