24 BiCKNELL : Ferns and flowering plants of Nantucket 



leaves are even more deeply cut, with the sinuses wide and 

 deep, often much wider than the lobes, leaving the lamina 

 along the midrib narrowed to a width of only 0.8 to 1.5 mm. 

 These leaves were mostly wider than long, the median lobes 

 divergent and spreading, about twice the width of the narrowed 

 basal pair, widely dilated towards the end into divergent, triangular, 



mos 



I to 1.5 cm. wide. Many of these leaves as late in the season as 

 the middle of August, with their petioles and the branchlets, were 

 clothed with a fine stellate scurfiness ; others were glabrous except 

 in the axils of the primary veins. 



* QUERCUS PAGODAEFOLIA (Elliott) Ashe. 



NearQuaise Point, June 11, 1908 — a single tree in full flower 

 growing in a low thicket. The tree was about twelve feet high 

 and twenty-four inches around two feet above the base ; the lowest 

 branches, over six feet above the ground, were nearly horizontal 

 and their widest spread about ten paces. The tree stood in an 

 opening into the thicket and much of the bark of the trunk had at 

 one time been destroyed, apparently by cattle ; the wound had 

 healed but the trunk had been nearly girdled and the living bark 

 on one side reduced to a strip only three inches wide. 



Doctor N. L. Britton has joined me in astudyof the leaves of this 

 tree and we have reached the conclusion that it can be referred to 

 no other species than Querciis pagodaefolia. The leaves are not in 



all respects precisely identical with typical examples o{ pagodacfolia 

 from the southern states, their lobes being often less acuminate and 

 entire and the pubescence of the lower surface less firmly persistent 

 and of a more yellowish tinge ; but the form of many oi the leaves 

 is exactly that of the southern tree and quite unlike that of any 

 of our northern oaks, the lobes being triangular-lanceolate and 

 wholly entire. Other leaves approach in form those of the black 

 oak but differ In the lobes being more narrowed towards the ends 

 and more entire, the basal pair spreading as widely as the median 

 ones, or nearly so, instead of being conspicuously shorter. 



In Britton's Manual the range of Qtierciis digiiata (Marsh.) 

 Sudw. is given as extending north to Long Island. This northern 

 limit was based on a single tree found by me near Hempstead, 



