380 . MESOZOIC FLORAS OF UNITED STATES. 



was excavated chiefly in a heavy bed of Rappahannock sand in which 

 much silicified wood occurs. Several of the specimens treated by Doctor 

 Knowlton were obtained from this locahty, and one immense log was 

 uncovered on the west side near the shaft, which has never been taken 

 out and is now walled in. At this date the bottom of the reservoir was 

 mostly dry, except for the stream from what was called^ the Capitol 

 Spring, which coursed through it. Below this bed of sand is one of 

 lignite, and this could then be traced all the way round the reservoir. 

 It was seen to thicken somewhat on the west side. About midway of 

 the reservoir from north to south, but on the east side, a little north of 

 opposite the shaft and as much south of opposite the Capitol Spring 

 tower, therefore only a short distance from the extreme southwest corner 

 of the Soldier's Home inclosure, a few feet above the lignite bed, I found 

 in a stratum of sandy clay quite abundant remains of plants, chiefly 

 ferns, and, as it proved, nearly all belonging to one species (see p. 

 516). The vegetable matter was black, but tended to peel off. 



Crossing to the west side, near the shaft I found another plant- 

 bearing vein consisting of buff-colored clay with sand partings, and 

 very frail. It held remains of conifers, cycads, and Gingkoaceae (see 

 p. 516). 



About two weeks later, on October 2, 1892, I took Doctor Ai'thur 

 HoUick to this locality and we made a small collection near the spot first 

 described, but we must have worked in a different vein, for nearly all 

 the specimens consisted of tubers of Equisetum. 



In October, 1892, I spent six days in reexamining the Potomac beds 

 on the James and Appomattox rivers. I was speciaUy interested to see 

 if any basal clays could be found, as aU the plants collected by Professor 

 Fontaine had come from clay lenses in the coarse sands. That these 

 lenses must have l^een parts of an underlying clay stratum was self-evident, 

 but the latter seemed to have been entirely eroded away before the depo- 

 sition of the sands, and only a few remnants left in the form of lenses, the 

 sand underlying as well as overlying them. These clay lenses become 

 numerous in descending the James after passing Warvsdcks Park, some 

 7 or 8 miles below Richmond. At Driuys Bluff they are prolonged into 

 strata A^ar^dng in color, but clearly included in the coarse sands. They 

 here rise 15 or 20 feet. Below Kingsland, Proctor Creek, and Red Water 



