FLORA OF THIO WOOUBINF. SAXI) AT AHTHTRS BLl'FF, TKX. 



167 



Phi/llilis, aiul ill liis account of 1S74 ho sug- 

 gested coniparisoMs with Smilas, Paliiirun, 

 ami Populus. It has no <haractei-s which ally 

 it to Smilax or Paliurus. hut it is nuK'h like a 

 variety of species that iiave heen refcrretl to 

 Populus, the majority of which do not present 

 clear evidence to warrant such a reference. 



The sj)ecies may l)e descriheil as follows: 

 Loaves rhomhoidal to orbicular, with a rounded 

 or cimeate decurrent base, presumably varying 

 to more or less cordate. Apex broadly rounded. 

 Margins entire in their lower halves, undulate 

 toothed in their distal halves. Texture sub- 

 coriaceous. Length about 4.5 to 6 centi- 

 mctei-s; maximum width, midway between the 

 apex and the base, o to 6.5 centimeters. 

 Petiole stout, curved, 2 centimeters in length 

 in the smaller leaves, not preserved in the 

 larger. \'enation five-palmate from the base, 

 the midrib no stouter than the lateral {)rima- 

 ries; the}' diverge at acute angles and arch 

 near the upper margin to join branches from 

 the short camptodrome secondaries. Areola- 

 tion obsolete. 



In form and venation the fossil is much like 

 the modern leaves of both Tetracnitroii and 

 CercidophyUiim, especially seedling leaves; it 

 is less like Trocliodendron but is approached bj- 

 seedling leaves of that genus. If, as sonic 

 botanists believe, these somewhat anomalous 

 vesselless dicotyledons are primitive in their 

 features and not reduced, they should be 

 present among the earlier fossil angiosperms 

 and should show considerable abundance and 

 diversity. The abundance of comparable fossij 

 forms, as I have pointed out, masquerading 

 under various names and widespread and 

 abundant in the closing days of the Lower 

 Cretaceous, in a measure substantiates the 

 first of these assumptions. 



Family BANTTNCTJLACEAE ? 



Genus DEWALQUEA Saporta and Marion. 



Dewalquea insigniformis Berry. 



DewaJffuea insigniformis Berry, Torrey liot. Club Bull., 

 vol. 44, p. i79, pi. 7, figs. 6, 7, 1917. 

 Leaves digitate, of probably five leaflets. 

 Leaflets linear-acuminate, with prominently 

 serrate margins. Length about 12 centi- 

 meters; maximum width mostly 1 to 1.25 

 centimeters, at or slightly above the middle. 

 Base very gradually narrowed and with entire 

 margins for a distance of about 2 centimeters. 

 Midrib stout, prominent on tho under surface 

 68031°— 22 2 



of the loallots. S<M-ondarit>s numerous, diverg- 

 ing from the midrib at angles i>f .W to 40°, long 

 ascending and eventually camptodrome, send- 

 ing off small outwardly directed branches to 

 tho miuginal te<-th. Texture coriaceous. 



This characteristics species with its coriaceous 

 texture nmst have had i-atlu^r stiff, stri<'t loaves 

 in life. It adtls to our Mora anottuT form of tho 

 curious genus Ihwalqura, which is .so striking 

 an elcMuont in the UjipiT Cretaceous and lower 

 Eocene. Tho only known .Vmeriean species 

 that resembles this form in any respect is 

 Dtwdhjuia smithl Horry," of the Tuscaloosa 

 and lilack Creek formations. D. smithi is 

 much larger, with relatively broader leaflets, 

 which have less prominently serrate margins 

 and partly craspedodromo venation. 



Dewalquea insigniformis is, however, as its 

 name indicates, very much like Dewalquea 

 insir/nis Hosius and \'on der Marck," a promi- 

 nent species in the Campanian and Maostricht- 

 ian substages of Europe. D. Insidni.s has 

 r«'latively broader, less prominently toothed 

 leaflets, in some specimens as many as seven, 

 and the venation is said to be craspedorlrome. 

 The venation is, however, a character of slight 

 value, for entire and toothed leaflets gener- 

 ally occur together, and I imagine that De- 

 walquea insignis is merely a serrate form of 

 the associated Dewalquea haldemiana Saporta 

 and Marion. If the latter had prominent ser- 

 rate ttw?tli added it would be identical with 

 Dewalquea insigniformis. 



Deivalquea insigniformis was described from 

 a large number of fragmentary specimens ob- 

 tained in the upper part of the Bingen sand in 

 Pike County, Ark. .V characteristic leaflet is 

 present in the early collections from the Wood- 

 bine sand at Arthurs Bluff, Tex. 



Order ROSALES. 

 Family CAESALPINIACEAE. 

 Genus PALEOCASSIA Ettingshausen. 

 Paleocassia laurinea Lesquereux. 

 Plate XI., figures. 

 Paleocassia laurinea I.e.s(|uereux, The flora of the Dakota 

 group, p. 147, pi. 64, fig. 12, 1892. 

 Berry, Torrey Bot. flub Bull., vol. ;J9, p. 396, 1912; 

 U. S. Geol". Survey Prof. Paper 112, p. 100, pi. 23, 

 fig. ], 1919. 

 Leaflets ovate-lanceolate, subinequilateral, 

 with a pointed ape.x and a cuneatebase. Length 



« Berry, E. W., Torreya, vol. 10, pp. 34-3.*, flg. 1, 1910. 

 » Hosius, A., and Von der Marck, W., Palaoonlographica, vol. 20, p. 

 172, pi. 32, flgs. 111-113; pi. 33, fig. 109; pl..34, fig. 110: pl.35, fig. 123, 1S80. 



