103 



phytic as the oak dunes of Lake Michigan, are not of an extreme 

 mesophytic character by any means. I have noticed that in 

 these IlHnois plants the culmination of the flowering period is not 

 reachied until the latter part of August or early September, but 

 the flowers do not come so late as in plants of the sand dunes. 

 That the habitat, in DeWitt Co., at least, is more mesophytic, 

 is indicated by the distribution of the xerophytic fern (Pteris 

 aquilina) in that county. Here in the dunes this fern is associated 

 with the oaks, but in the oak woods of DeWitt Co. it is entirely 

 absent, occurring only on barren, exposed hills, the most xero- 

 phytic areas in the region. The situation in southern Indiana 

 may be the same. 



The late blooming, September 15, is also noteworthy. Coulter 

 (i) gives the blooming period as extending from June to August, 

 which is the exact time given in Gray's Manual. ' It is interesting 

 that xerophitism, often so potent a factor in early blooming, 

 seems to have a reverse effect on the flowering of this species. 



To account for the occurrence of this mesophyte in a xerophytic 

 area, there seems to me two theories which should be considered. 

 One theory would be that some local condition, com.parable to the 

 usual mesophytic habitat of this species, exists. A second view 

 would be that the plant is able to endure xerophytic conditions 

 on account of its short aerial life, provided there is sufficient 

 stable humus to enable the underground portion to persist from 

 year to year. An alternative, of course, would be to consider 

 the plant as a xerophyte, but there is no evidence to support 

 such an assumption. 



Of the two theories the second seems to be the more valid one. 

 About the only evidence to support the first view is that appar- 

 ently the plants are here subjected to rather severe conditions, 

 as is shown by the scarcity and semi -dwarf ness of the stems, 

 these being also much blackened by the sun. But such evidence 

 may be more apparent than real, for the plant is able to persist 

 and ripen its seed as well here as elsewhere. To support the 

 second view we have the evidence of late blooming, which seems 

 to me very important. If a local condition existed, the plant 

 would be expected to flower in its usual season. Cowles (2) 



