140 Trans. Acad. Sci. of St. Louis. 



The bases of the leaves range from acute to cordate. It is 

 an exceptional plant on which the truncate or cordate may 

 not be found. In regard to their outline the variation on the 

 same plant is often very surprising. The shapes of the leaves 

 may all be reduced to about five types : First, the broad ovate 

 or ovate-oblong, Figs. 1 and 7 of the plates ; second, the 

 obovate or rhombic, Figs. 2 and 12 (this the Missouriensis); 

 third, the elliptical, Figs. 3* and 6 (a common tree form); 

 fourth, the oblong-lanceolate with obtuse, truncate, or cordate 

 base, Figs. 4 and 9 (the old recognized type form) ; fifth, the 

 long narrow lanceolate, Figs. 5, 10, 11. 



There is one idea which runs, more or less, through all the 

 forms, viz., the ob idea, — the leaf broadening above the mid- 

 dle. Even in the long narrow leaves it may usually be noticed. 

 There is, besides, another, a universal character of the highest 

 importance, it, in conjunction with the ob idea, being probably 

 the chief mark by which we so readily recognize the species. 

 This is the lack of symmetry of the two sides, or in other 

 words, the leaf is not equilateral. It has a peculiar concave 

 curve on one side towards the apex, which I believe is dis- 

 tinctive. 



It is, indeed, a wonder that, in spite of all the variation, 

 there yet is not the least difficulty in the recognition of the 

 species. 



How do we come to know it? Surely, not by a remem- 

 brance of its technical characters as given in descriptions. I 

 conceive it must be from a very complex impression produced 

 by the plant as a whole, too spiritual, perhaps, to fully in- 

 terpret upon paper. No doubt, the shade color of the foliage, 

 the way the leaves are massed on the twigs and branches, 

 as well as the manner in which these spread, all have to do 

 with it. 



A second inference to be drawn from the table is, that 

 there is not sufficient ground to draw the specific line between 

 tree and shrub. Comparing the column of 38 trees with 

 the 78 shrubs, it may, in a general way, be affirmed 

 that the latter bear leaves having greater tendency to the 



* After plate was drawn, when too late to rectify, I discovered that Fig. 

 3 represents a hybrid. — N. M. G. 



