LESQUEKEux.] ENUMERATION OF CBETACEOUS PLANTS. 345 



tion which is a character of this so-called variety. I will further remark 

 that specimens of both the entire and trilobate leaves were found at 

 the same locality. 

 Sassafras (Aealiopsis) mirabile, Lesqx., p. 80, PI. XII, fig. 1. 



Leaves thicJc, coriaceous^ large ; lobes broad and short, the lateral ones on 

 a broad angle of divergence, tvith borders dentate or deeply, undulately lobed ; 

 secondary veins mostly craspedodrome. 



A remarkable modification of the character of this species is observed 

 upon a fine leaf still much larger than the one figured in the Cretaceous 

 Flora. It is twenty centimeters broad between the points of the lower 

 lateral lobes ; about fifteen centimeters long from the top of the petiole 

 (lower part of the leaf destroyed), divided by the forking of the lateral 

 primary veins in five equal and equally sinuate dentate lobes, with the 

 same character of nervation as in the normal form. The lobes also are 

 of equal length and width ; the middle one fifty-seven millimeters broad 

 and scarcely sixty millimeters long, with the lower secondary veins 

 camptodrome and the upper ones craspedodrome, and entering the teeth. 

 As 1 have remarked it already, this subdivision in four lobes tends to 

 show the reference of those large leaves to Aralia rather than to 

 Sassafras. But it may be also an exposition of that disposition to poly- 

 morphy so remarkably evident in the leaves of our living Sassafras ofji- 

 cinale. As these generally tripalmate leaves pass to an entire oval sbape 

 sometimes, they subdivide more or less in the same way. One of the 

 largest specimens of 8. mirabile, lately received from Mr. Towner, of 

 Clay Center, Kansas, has one of the lateral primary nerves twice as 

 thick as the other, and its base is higher. One of its lobes is, therefore, 

 much longer, fifteen and one-half centimeters, jvhile the other is only ten. 



Sassafras, (Aealiopsis,) eecurvata, 



Platanus recurvata, Lesqx., Cret. Flora, p. 71, PI. X, fijrs. 3-5. 



Leaves three to five palm ately lobed ; lobes nearly equal in length, the mid- 

 dle one broader, lateral nerves curving doicnward, simple or forlcing above 

 the base; borders of the lobes entire or sparingly coarsely dentate. 



This form is still more uncertain, and, so to say, transient in its char- 

 acters. By the decurrent base of the leaves descending to the petiole, 

 lower than the point of union of the primary veins and also by the trilo- 

 bate division, it is a Sassafras. The irregularity, however, of the lobes, 

 the nervation and the double divisions in lobes or teeth refer it to Fla- 

 tanus; the tendency to become five-lobate by the forking of the lateral 

 nerves is a character of the Araiiaceo}. This disposition to a subdivis- 

 ion or multiplication of lobes is seen in fig. 3, where the lower branches, 

 though thick, do not diverge widely enough for modifying the borders 

 of the leat, but are curved inward and join the secondary veins at the 

 base of the lobes ; but in fig. 4, which represents a fragment only, the 

 subdivision in five lobes is evident. It is still more marked upon a leaf 

 recently found and figured by Mr. H. C. Towner. In this one the cuneate 

 base of the leaf descends far down, two and one-half centimeters lower 

 than the point of union of the primary nerve ; the lateral ones divide 

 in two branches from below the middle, and curving backward thej^ form 

 well defined ovate pointed short lobes, and thus a palmately five-lobed 

 leaf of the Aralia type. This leaf, without the petiole, which is broken, 

 is twelve centimi^ters long, and as wide between the points of the lowest 

 lateral lobes, which are only one and.oue-half centimeters long ; the oth- 

 ers, as also the middle one, are three centimeters long and about as 

 wide, being half round, cuspidate, and separated by narrow obtuse si- 

 uuses. 



