428 



THE AMERICAN NATURALIST. [Vol. XXXVIII. 



East of the first locality. Ground was much less moist. 4c 

 plants were collected at random and all the flowers counted as 

 in lot No. 2. Collected July 12th. Fig. 2. 



Lot. No. 4. — 173 heads. From the prairies about eight 

 miles southeast of the preceding localities and not far from the 

 lake shore. Ground moist from recent and continued rains. 

 Plants were all rather low and much shaded by tall grass and 

 tall weeds. Collected July i6th. Pig. 3. 



Lot No. 5. — 246 heads. From the same general region as 



in the case of Lot No. 4, but from portions of the prairie where 

 there were fewer weeds and from lanes in the vicinity. Collected 



In the case of both Lot 4 and Lot 5 plants were pulled up at 

 random and all the flowers of each plant included in the count. 

 The results of the various counts is shown in the following curves. 



Comparing the above curves we notice one rather interesting 

 fact. The curve of those heads which were growing under what 

 had been noticed to be especially unfavorable (curve 3) is entirely 

 different from the other three, curves i and 2 being especially 

 symmetrical. I am of the opinion that the reason that curve 4 

 is not so symmetrical as i and 2 is that the heads were taken 

 from plants not far enough removed from those represented in 



