ig4. THE PLANT WORLD. 



Plants, " usually on rocky banks," and Webber's Flora of Nebraska, 

 *' rocky banks in woods, in shaded caiions." 



These instances might be considerably extended and may be taken 

 to represent the conditions where any effort to be specific is made. 

 Some authors are more general, giving it as a mountain plant like the 

 majority of Aqiiilegias in different parts of the world. It will be seen 

 that most authorities in the above citation connect the plant with 

 rocks, or give such topographical features as are commonly associated 

 with rocks, though not exclusively rock cuttings, such as ravines, 

 glens and caiions. Only two mention sand, Provencher and Paine, 

 the latter with phraseology that indicates something unusual. Some 

 other authorities also mention sand and sandhills in New Jersey, New 

 York and elsewhere. 



My experience with the plant, mainly in the region of the great 

 lakes and the St. Lawrence valley, covers most of the conditions cited. 

 In western New York I have met with it on ledges and cliffs of lime- 

 stone, slate and sandstone; in woods where the soil abounds in chert 

 from the disintegration of flinty limestone, shaley banks and slopes of 

 wooded hills, clayey from the decomposition of slate or shale. These 

 localities have been mainly on the borders of streams and lakes, though 

 not confined to them. In the vicinity of Chicago the Wild Columbine 

 is most abundant in the sand of the dune region which encompasses 

 the head of Lake Michigan. It occurs on the top and sides of the 

 high wooded dunes when protected from the fiercer winds, but more 

 commonly near the bases of dunes in more open reaches, where it 

 frequently makes large patches and in its season may be the prevail- 

 ing floral feature in such locations. In these places, not far from a 

 neighboring slough, or on the low sand ridges between the sloughs 

 they separate, where the soil is better provided with humus, it attains 

 a larger size, sometimes becoming three feet high. But it occurs 

 under other conditions, as on banks of streams which have cut their 

 way through low ridges of drift, or cut out banks along the bases of 

 the higher hills, usually in wooded areas. Here the soil may be of 

 clay, loam, or gravelly, but of a consistency to make a rather firm 

 and often steep bank. On limestone bluffs and ledges of the Desplaines 

 river and tributaries it roots in the crevices of rocks or on some slightly 

 projecting shelf where the soil is so meagre as to remind one of the 

 conditions given by Dr. Barton. On these rocks it is associated with 

 Arabis hirsiita and Icevigata, HeucJiera hispida, and Pellcsa atropurpurea 

 under conditions the dryest and sunniest As I have often seen it 

 about the upper lakes and in Canada on granite and other crystalline 

 rocks as well as on limestone, sandstone and slate rocks, the particular 

 character of the rock substratum and the soil of the open woods and 



