KNowLTON.] DESCRIPTIUIS OF SPECIES. 77 



The .scar where it was attached to its sister fruit is oblique and very 

 long'. 



I Ivnow nothing either living' or fossil to which this fruit can be 

 compared. 



Locality. — Gulch 1 mile northeast of Belshaw's ranch, Grant 

 County, Oregon. Collected bj^ Knowlton and Merriam, Juh^, 1901 

 (U. S. Nat. Mus., No. 8502). 



RuivAC CKAT.I'XIIFOLIUM U. sp. 

 PI. XVI, fig. 7. 



Leaf coriaceous, trifoliolate or very deeply three-lobed, lateral leaf- 

 let (or lol)e) roughly ovate-lanceolate in shape, irregularly, deeph^ 

 cut into toothed lobes or large teeth near the apex, wedge-shaped at 

 base, acuminate at apex; middle leaflet (or lobe) much smaller than 

 the lateral ones, narrowly ovate-lanceolate, long, wedge-shaped below, 

 very acuminate at apex, margin coarsely toothed; nervation of leaf- 

 lets (or lobes) consisting of a strong midrib and several pairs of alter- 

 nate rather thin secondaries which end in the lobes or teeth; finer 

 nervation not preserved. 



This species is represented only by the specimen figured, and this 

 unfortunatel}^ is not sufficiently well preserved fully to reveal its 

 character. It is presumably trifoliolate, but may be only deeply 

 three-lobed. The larger leaflet is assumed from its position to be a 

 lateral one, the middle one being much smaller. This lateral leaflet 

 is 5.5 cm. in length, and 2.75 cm. in width between the larger lobes. 

 The central leaflet is -1 cm. long and 1.25 cm. in width. Both are 

 sessile. 



Owing to the poor state of preservation it is not possible to fix the 

 position of this leaf with certaint3^ In most trifoliolate leaves, if the 

 leaflets differ in size, the middle one is usually the larger. In this 

 case the larger leaflet of our specimen is the lateral one, and it is 

 assumed to be lateral because it curves away from the smaller one and 

 has a curved midrib, whereas the smaller one is erect and has a straight 

 midrib. 



With the limitations set by the imperfections of the specimen it is 

 perhaps unwise to attempt comparisons between it and described forms, 

 yet a few of these may be suggested. Thus the larger leaflet is not 

 greatly unlike Crataegus (*6vr(/6'//« Lesq.," from Florissant, Colorado, 

 which itself is probably the same as a part at least of the leaves 

 described as Myrica diversifoUa Lesq.,* from the same locality. These 

 are all long petioled, showing them to be leaves and not leaflets, 

 whereas ours is sessile, showing that it is probably a leaflet. 



"Cret. and Tert. FL, p. 198, PI. XXXVI, fig. 10. Mdem, p. 146, PI. XXV, figs. 6-15. 



