n 



52 BOTANICAL GAZETTE IjcjI-Y 



Newb., and Z. Meekii Heer, to L.^semialatiim Lesq. at the 

 other. 



This more or less closely related group includes four types 

 of leaves. The first, a narrow elongated leaf, with an emargi- 

 nate apex [Liriodefidropsis angiistifoli a) ^ occurs in such numbers 

 in the Amboy clays and is so uniform in shape as to warrant its 

 consideration as at least a permanent variety, which developed 

 from the simple lanceolate ancestral form by a reduction of the 

 apex until it became emarginate in the Lower Cretaceous. The 

 second type includes certain leaves which have been referred to 

 L, simplex, and which resemble the preceding in outline, but are 

 somewhat broader. They would have been developed easily 

 from Liriode7idropsis angiistifolia by a slight shortening and broad- 

 ening of the blade of the latter; and like it, this form is found 

 in the Amboy clays. If we compare some of the figures of L.^ 

 simplex with Lesquereux's Myrsine crassa^""^ they seem to be 

 identical. 



The third type of leaf is connected with the preceding by 

 insensible gradations from the elongated emarginate forms, 

 showing a constantly increasing width of blade, together with a 

 shortening of its length, the emarginate apex becoming. less and 

 less so, until it is simply retuse. If these leaves were somewhat 

 shortened and broadened, they w^ould be identical with the 

 usual form of young leaves of Z. Tidipifcra, If Phyllites orbicu- 

 laris Newb. (Flora Amboy clays, 136.pl. 24. Jigs. y,8) be com- 

 pared with young leaves of Z. Tulipifera, they seem very similar 

 and possibly identical. The various fossil leaves referred to 

 Cohiteji primordialis Heer are also of this type, and the same may 

 be said of Sapotacites retiisiis Heer.'s Xhe resemblance of these 

 ancient Liriodeudron leaves to various leaflets of existing 

 Leguminosae has also been noted by several authors. We have 



found 



Tulipifi 



which shows conclusively that these ancient leaves are true Lirio- 

 deudron leaves. This type is represented by various fossil 



'^ Flora Dakota group, pL 33, figs. 2,3. 



^5 See Flora of Amboy clays, pL £3- figs, j, 6. 



