49 



for us by Mr. Ernest J. Lederle, of the School of Mines : 



SiHca ( Si O2) - - - - - 39.85 



Alumina and Oxide of Iron (AI2 O3 and Fcg O3) 18.58 



Lime (Ca O) - - - - - - 9.35 



Magnesia (Mg O _ _ _ . _ 19.79 



From this it is seen that the magnesia constitutes about one- 

 fifth of the entire ash of the plant, and is present in larger 

 quantity than any other constituent except the silica. 



It should be remarked that the specimens seen from about 

 Washington, D. C, and from Montgomery county, Md., are 

 larger than those from the serpentine areas. The same may 

 also be said of the specimens from Amherstburgh, Ontario, said 

 by Mr. Macoun to grow in "damp woods," and also of those 

 from the West. In some respects these approach the forms 

 referred by us to var. maximum. 



In Meehan's " Native Flowers and Ferns of the United 

 States," Vol. ii., plate 48, is an illustration of one of these large 

 forms, made from a specimen collected in Bergen Park, Colo., at 

 an altitude of 7,000 feet. If this drawing is correct, it very 

 nearly represents our var. maximum, but we have not seen any 

 specimens of this from the Rocky Mountain regions. 



Between Dr. Torrey's original description in the American 

 Journal of Science and Arts, in 1822, and his later description in 

 the Flora of the State of New York, published in 1843, there is 

 the following discrepancy : In the former the leaves are described 

 as acute, and the capsules as shorter than the calyx, while in the 

 latter the leaves are described as mostly obtuse, and the capsules 

 as about twice as long as the calyx. This is, perhaps, to be 

 accounted for by the original imperfect material. The latter 

 description agrees with the characters of our var. oblongifolium,. 



Var. VILLOSUM, n. var. Stem leaves lanceolate to ovate- 

 lanceolate ; capsules 2 to 2\ times the length of the calyx ; the 

 whole plant densely villous-pubescent. (Plate LXV., f 8.) 



C. villosiim, Muhl., Cat., 46; Darlingt., Flor. Cestr., 2d Ed., 279. 

 C. hirsutum, ? Darlingt., Florula Cestr., 54. ? 



C. oblongifolium, Darlingt., Flor. Cestr., 3d Ed., 33, in part; Torr. & Gray, Fl. 

 N. A., i., 188, in part. 



Habitat. On serpentine rocks, Lancaster Co., Penn. (Porter) ; 



