82 GATES: THE VALIDITY OF HELIANTHUS ILLINOENSIS 
situations this species grows erect from a long running rootstock, 
by means of which propagation is usually effected from year to 
year. Reproduction by seeds does not seem to be customary in 
view of the scarcity of seedlings observed. Observations on the 
association of the individual seedlings, in the Beach region at least, 
seemed to point to the dissemination — not of the seeds separately 
—but by heads containing the full quota of seeds or nearly so. 
At all events vegetative propagation and reproduction soon lead to 
the formation of a patch of plants. The patch is usually a closed 
association and admits of but few interstitials (such as Polygonum 
tenue or Arabis lyrata), while on the other hand the patch may 
spread outwards for a meter or more. It was this assemblage in 
patches that furnished the key to the situation. It was very 
plainly evident that edaphic factors varied within the extent of the 
patches. Of these the most important were light and soil. An- 
other factor, water supply, due to varying amounts of precipitation, 
varied from season to season and also within a given season. 
These three factors, acting either singly or conjointly with the 
compound factor, wind, may produce the subspecific type, 77. occi- 
dentalis iwllinoensts on sandy soil. In so far as could be determined, 
the greater amount of pubescence which is characteristic of the 
subspecies, /7, occidentalis illinoensis, was to afford the plant ade- 
quate protection from excessive transpiration brought about by 
varying edaphic conditions. Light, acting singly while the other 
factors mentioned were constant, could produce the subspecific 
type, provided the physiological water-content of the soil was such 
as to make the transpiration ratio (7. ¢. the amount of water 
transpired divided by the amount of water taken up by the plant) 
at or below the critical point for the particular plants. Some 
patches were found which extended from the full sunlight up into 
the fairly dense shade of oak trees and in every observed case 
those plants which received full sunlight were more pubescent than 
those in the shade. 
It is, however, only in case the water supply is deficient that 
the difference in pubescence is so marked as to constitute the sub- 
specific type. Long lower internodes are characteristic of the sub- 
species but several examples of plants growing in full sunlight with 
a dense villous pubescence had internodes but very little longer 
