20 LEAVES OF LIRIODENDRON HOLM. 



the shore of tbePotomac aud iii the Smithsonian Park, we must wonder 

 not a littlethat systematic authors have mentioned so few of these forms. 

 We see here the most ''insensible gradations" from the entire to the 

 deeply-lobed leaves. We see the orbicular form in Fig. 5, the obcordate 

 in Figs. 6, 7, and 8, the approximately bilobed in Figs. 16, 31, aud 41, and 

 the large variation in the upper margin of these small leaves, more or 

 less notched, as for instance in Figs. 11, 13, 16, etc., or almost horizon- 

 tal, as in Fig. 17, and, farther, how there is a weak, but quite distinct, 

 attenuation of the blade into the petiole, as in Figs. 6 and 7. And if 

 we examine the other forms with the lobes more or less developed, we 

 shall find quite a large number of variations from Figs. 11, 15, and 17, 

 to the more distinctly lobed leaves, as Figs. 18, 19, and 20, and finally 

 the large leaves. Figs. 2, 3, and 4. We find here another kind of varia- 

 tion, in regard to the depth of the sinuses, merely shallow in Figs. 24 

 and 25, or very deep as in Figs. 23, 26, 27, 28, 29, and 30. And farther, 

 as stated in the description of the small leaves, the upper margin varies 

 very much in regard to its sinns, deep as in Figs. 22, 24, 39, and 40, or 

 in a much smaller degree as in Figs. 3, 23, 25, 27, and 30. Finally, it 

 must be remarked that the sinus at the base of the blade shows quite 

 a large variation in its depth, but I have not found a case in the larger 

 leaves where the blade tapered into the petiole. 



As to the varieties, I have obtained rich material from a single tree 

 cultivated in the United States Botanical Garden under the name of 

 " integrifolia,^^ and I suppose this corresponds to Michaux's variety 

 " obtnsiloba.^^ Leaves from this tree are figured on Plate viii, Mgs. 32 

 to 37, and these also differ among themselves, ranging from the small, 

 obcordate leaf (Fig. 37) to the more obtuse one, Fig. 36, with the upper 

 margin almost horizontal in contrast with the leaf. Fig. 37, where a deep 

 depression is to be observed. The api)roximately roundish form. Fig. 

 33, was the most common on this tree, of which, however, several leaves, 

 as Fig. 32, showed two distinct teeth at the upper margin, corresponding 

 to the upper lobes of the normally developed leaves ; and moreover I 

 observed not a few with an incipient dentation at the base and at the 

 middle part of the blade, as shown in Fig. 34. The base showed here 

 the same kind of variation as mentioned above in the typical forms. 

 The other variety, ^^acntiloba^^ Michaux, is unknown to me, but it is pos- 

 sible that the small leaf. Fig. 23, and the larger ones. Figs. 26 to 30, show 

 a similar form. They represent, however, not a variety, but merely a 

 young state of the typical tree. The small leaf, figured on Plate vi, 

 Fig. 23, has been taken from a plant, 1 foot high, of which the age was 

 almost two years, and all the leaves on this specimen had this same 

 shape. The larger leaves. Figs. 26 to 30, belonged to a mere shrub, no 

 more than 5 feet high, of which the leaves had either the same deeply 

 lobed form, or, as shown in Fig. 31, the almost obcordate outline. It is 

 to be supposed that this form may not be constant, and as the shrub 

 gets older and attains a larger size the leaves will gradually acquire the 



