TAYLOR: RAUNKIAER'S GROWTH-FORMS 491 



figure for nearly the whole world. This is the chief point made here. 

 For, if the percentages of woody and herbaceous species, as shown by 

 Sinnott and Bailey, and percentages based on an approximation to a 

 plant census as shown by the figures from Long Island are not in 

 substantial agreement with Raunkiaer's system, then it follows that 

 that system does not, as yet, show what has been claimed for it. Just 

 what reshifting of the percentages in the normal spectrum is necessary 

 lies outside the scope of this paper. It seems evident, however, that 

 they need more study and over large areas, preferably based on plant 

 censuses such as has been attempted on Long Island. With the com- 

 pletion of such studies we might have in the revised percentages more 

 accurate data as to climatic response than is possible at present. 



