KNOWLTON.) TRIASSIC WOOD FROM NORTH CAROLINA. 278 
only a slightly divergent variety of it, from the copper mines near 
Abiquiu, New Mexico.’ Since preparing this report for Professor 
Russell I again looked over the slides in connection with the study of 
a number of pieces of wood from the Richmond Basin, Virginia, a 
report of which is given in the Nineteenth Annual.? Among the Rich- 
mond Basin specimens I found one having the same structure as those 
from North Carolina, which had previously been referred to Arauca- 
rioaylon virginianum. Although very close to the species from New 
Mexico and Arizona, there seem to be slight, but thus far constant, 
differences, and I gave the name Araucarioxylon Woodworthi to the 
specimen from the Richmond Basin. A more complete study of the 
material from North Carolina confirms this view, and it is so referred 
here. 
In 1889 I described, under the name of Araucarioxylon virginianum, 
a piece of fossil wood that was supposed to have come from the Potomac 
formation at Taylorsville, Virginia. Subsequent investigation has 
shown that this specimen came from Triassic strata, the locality where 
it was found being almost the only known place where the Potomac 
formation rests on the Triassic. The specimen from North Carolina 
mentioned in my report to Professor Russell as doubtful appears to 
belong to this species, although not agreeing in every particular. The 
following is a brief discussion of the two species based on the North 
Carolina material: 
ARAUCARIOXYLON Woopwortui Knowlton. 
Pl. XXXVII, Figs. 7-9. 
1899. Araucarioxylon Woodworthi Knowlton: Nineteenth Ann. Rept. U. 8. Geol. Sur- 
vey, Pt. II, p. 517, pl. lii, figs. 1-6. 
As may be seen in comparing the figures here given with those 
accompanying the original description of A. arizonicum,* the agree- 
ment between the woods from North Carolina and those from New 
Mexico and Arizona is very close indeed. The annual ring is very 
faint and is detected with difficulty. It consists of only two or three 
rows of smaller, thicker-walled cells. The wood cells are seen to be 
equally thick-walled from both localities. The wood cells in the 
Richmond Basin specimen are also identical. 
The medullary rays in A. arizonicum are composed of 1 to 22 super- 
imposed cells, whereas in this species, both from the Richmond Basin 
and from North Carolina, the number ranges from 1 to 12, the usual 
number being perhaps 4 to 6. The rays are short-celled in all. 
The bordered pits as seen on the radial walls of the wood cells are 
1Proc. U. 8, Nat. Mus., Vol. XIII, 1890, p. 285. 
2Nineteenth Ann. Rept. U. §. Geol. Survey, Pt. II, 1899, pp. 516-519, pl. lii. 
3 Bull. U.S. Geol. Survey No. 56, p. 50, pl. vii, figs. 2-5. ; 
4Proc. U. 8. Nat. Mus., Vol. XI, 1888, p. 3, pl. i, figs. 1-5. 
20 GEOL, PT 2 18 
