1902] NOTES ON SASSAFRAS 445 



formed to the margin, but so small as to have been effaced in 

 the specimen ; the right one is stronger and runs almost to the 

 sinus where it makes a sharp turn upward, continuing until it 

 joins the next secondary. This feature is analogous to those in 



■ the modern leaf, which indicates the mode of origin of this 

 peculiar character. This leaf seems to form a central figure 

 from which a series of forms grade in several directions, culmi- 

 nating in quite dissimilar leaves. Lesquereux's 5. crctaceiim is a 

 more Platanoid leaf, with more acute tips, a tendency to become 



' dentate, and with the primaries inserted nearer the base. 

 Closely allied to the preceding is his 5. {^Araliopsis^ 7piirabile, 

 which serves as a connecting link with hxs Plafaniis rea^rvaia. 

 From the aforementioned S. cretacetim of Lesquereux it is but a 

 step to such a leaf as the one shown on//. 8. fig. 2, Later Ext. 

 Fl., and to the trilobed forms referred to Cissites harkeriaims, 

 and these in turn grade into the more Cissoid forms of this 

 species, such as those shown on //. //. fig, J, Cret. Fl. The 

 primaries are basal and of not much greater caliber than the 

 regularly succeeding straight secondaries. It is but a step from 

 this leaf to that of Cissites Hcerii on the one hand, with its pal- 

 mately five-pointed blade; and to such forms of Cissites aewni- 



P^' 5' fig- 



the other ; which in 



turn, by the elimination of the decreasing dentate points, gives 



In the second 

 etacenm, pL 8. 



us the leaf figured at j, //. 5, Cret. and Tert. Fl. 

 series of leaves diverging from the typical, 5. o 

 fig^ /, Later Ext. FL, is removed a slight distance by the short- 

 ening ot the blade, the thickening of the primaries and second- 

 aries, and the shortening and rounding of the lobes (5. obtnswn)\ 

 while a smaller leaf would be its logical descendant ; and from 

 these leaves to those referred to the typical Cissites salisburiae- 

 foliiis is but a step. In the third series of leaves diverging from 

 the typical 5. cretaceurn, we note that the leaf has its lobes much 

 produced, narrow and running to a sharp point, as in the beau- 



M 7^ fig 



however is still 



referred to 5. cretacetim. Lesquereux's S. acutilobwn does not 

 differ from the preceding except in the direction of the lobes, 

 which is scarcely a specific character. From this leaf it is no 



