KNOWLTON.] DESCRIPTION OF SPECIES. 61 



Four more or less perfect examples of this exceeding-ly interesting- 

 species have been found. The original specimen is nearly 4 cm. by 

 3 cm. in size, the large sepals being 2 cm. long and 1.5 cm. broad. 

 The next discovered specimen was obtained b}- Dr. John C. Merriam in 

 1900. It is not quite perfect, but has the nervation very well pre- 

 served. The sepals are about 2 cm. long and nearly as broad. The 

 two remaining examples were obtained by myself in 1901. 



The original specimen, as may be noted under the S3^non_ym3% was 

 firsttentativ^ely referred to Marsilea, and was later placed under Porana 

 byLes(}uereux, who correctly recognized its nature. Lesquerex, how- 

 ever, was in error in supposing that the sepals were " connate to above 

 the middle." They are undoubtedly free and overlapping, as shown 

 by the present drawing-, (See PI. IX, fig. 6.) The example secured by 

 Dr. Merriam is, as alread}' stated, not perfectly preserved, but as nearly 

 as can be made out, it consists of only three sepals, although the fourth 

 may be present, concealed under the others. They are clearly free, 

 however. 



The fossil forms referred to Porana consist of the more or less 

 leather}^ sepals and are usually five in number. Two species have 

 been described from the United States by Lesquereux, both coming 

 from Florissant, Colorado. Porana Sjjeirii'^'' is a five-lobed connate 

 species, while P. tenuis^ has never been figured, but is described as 

 having the sepals separate to the base. The first is very distinct 

 from the form under consideration, while the latter can not be well 

 compared. 



On first examining these specimens, especially the one collected by 

 Dr. Merriam (PI. IX, fig. 7), they seem to be referable to a Cornus of 

 the t3'pe of C. florlda or C. Nuttallii Aud., but the absence of any 

 marked indication of the flowers in the center apparently precludes 

 their reference to this genus. 



At the suggestion of Mr. Charles Louis Pollard, of the United States 

 National Herbarium, these fossils were compared with various species 

 of the genus Hydrangea, and the resemblance between them and the 

 sterile flowers so characteristic of this genus was so striking that they 

 have been referred to Hydrangea. 



Eight fossil species of Hydrangea have been described from the 

 European Tertiary, but none of them approach closel}^ to our form. 



Locality. — Van Horn's ranch, South Forkof John Day River, 12 miles 

 west of Mount Vernon, Grant County, Oregon. Collected by Maj. 

 Charles E. Bcndire (U. S. Nat. Mus., No. 8515), and by Dr. John C. 

 Merriam (Pal. Col, Univ. Cal., No. 851). Also obtained from a gulch 

 half a mile northwest of the Belshaw ranch, the next ranch east of 

 Van Horn\s, by F. H. Knowlton, July, 1901 (U. S. Nat. Mus., Nos. 

 8991, 8995). 



aCret. and Tert. Fl., p. 172, PI. XX VIII, fig. 15, 1883. & Idem, p. 173. 



