APPENDIX. 123 



doubtless a delicate fern of small size, the pinnules deeply cut into linear or spatulato lobes, those 

 of the fertile portions of the frond being specially slender, and bearing the sori at the extremity of 

 each lobe. A fruit-bearing fragment visible in one of the specimens before us calls to mind Lindley's 

 Tymfanophora racemosa, which is now regarded as the fertile portion of the frond of Goniopteris 

 Murrayana. 



This fossil also occurs at Sanyii, near Chaitang, with Sphen. orienlalis, thus linking together, 

 geologically, these two localities. 



Taxites spatulatus, Newh. 



Plate IX, Fig. 4. 



T. foliis coriaoeis Hneari-lanoeolatis vel spatulatis, curvatis, apice rotundatis, basi cuneatis, nervo medio valde 

 distiucto. 



In a yellow sandy schist, from near the Futau mine at Chaitang, with pinnas of PodozamUes, are 

 numerous linear or spatulate one-nerved leaves, evidently derived from some coniferous tree, appa- 

 rently of the family of Taxineis, though larger than the leaves of any of the known Yews. 



By their size, curved outline, cuneate base, and their variable width, these leaves bear some resem- 

 blance to some of those which have been referred to the genus Podooarpus, but with one exception 

 all the described fossil species have been found in Tertiary rocks. The exception referred to is 

 Podocarpiles acicularis, Andrse, from the Lias of Steierdorf, in which the leaves are very long and 

 narrow, having more the foi-m of those of a pine. 



Podooarpus Taxites, Unger (Flor. Poss. v. Sotzka), has almost precisely the form of some of the 

 leaves before us ; but it is very doubtful whether that was really a Podocaiyus. 



Brongniart has enumerated in bis Prodromus a Taxites podocarpoides, from the Oolite of Stones- 

 field, but no figure or description of it has yet been given. Possibly that species may have relations, 

 with the one under consideration, which would give the latter a value in determining the precise age 

 of the rocks which contain it. 



APPENDIX NO. 2. 



Analyses of Gidnese and Japanese Coals. 



Made for R. Pumpelly by Mr. James A. Macdonald, M. A., of the Sheffield Laboratory, Yale 



College. 



In the following analyses each determination is the mean of two closely agreeing ones. For the 

 water determination the coal was pulverized and heated in an air-bath at 110° C. until it gave a 

 constant weight. A portion was then ignited in fragments, in a closed crucible, to determine the 

 "volatile matter." The ash was estimated in the usual manner by incineration. 



I. Tatsau mine (43 feet seam) near Chaitang. 



Hard anthracite. Decrepitates very slightly, and yields a little HO in a closed tube. Spec, 

 grav. 1.5T. 



Carbon 89.81 



Volatile matter 3.08 



Water 2.67 



Ash 4.44 



100.00 



II. FuTAU mine. Chaitang (west of Peking). 



Bright, bituminous, coking coal, yielding a little HO in the closed tube. Spec. grav. L30. 



Carbon 85.77 



Volatile matter 11.94 



Water 0.35 



Ash 1.94 



100.00 



