DECEMBER 15, 1899. ] 
that the plant was Clematis orientalis L. ; from 
Kew the further information was sent that it was 
a variety of the species, exactly agreeing with 
specimens from the N. W. Himalayas. 
As the plant is apparently with us to stay, it 
may be worth while to give the description of 
it, based on Las Vegas material. 
Clematis orientalis, variety—Low straggling 
climber; stems slender, purplish at the nodes; 
leaves, including petioles, 7 to 12cm. long, with 
five leaflets, which are rather thick, perfectly 
glabrous, a somewhat glaucous green, more or 
less lanceolate in outline, the terminal one often 
linear-lanceolate, the lateral ones sometimes 
ovate-lanceolate, all more or less coarsely and 
irregularly serrate towards the base, or even 
lobed, the upper leaves especially having nar- 
row leaflets, distinctly lobed at the base, the 
lobes pointed and often notched ; in a well devel- 
oped leaf the terminal lobe is about 4 cm. long. 
Buds pale greenish-yellow, obpyriform, nodding, 
4-angled ; flowers at first nodding, ultimately 
erect; sepals four, pale sulphur-yellow with a 
greenish tint, rather thick, recurved at tips, 7- 
nerved, nearly glabrous, perfectly so below 
except edges, but above with scanty white 
woolly hairs, and the lateral margins, which are 
bent inwards, quite conspicuously white-woolly 
towards the tip; apex of sepal truncate in 
lateral view, with a linear green process, 2 mm. 
long, at the lower corner of the truncation. 
Length of sepal about 23 mm., breadth 10 mm. 
Stamens 32, anthers 4; mm. long, filaments 
about 6 mm., broad and flattened, especially the 
inner ones, glabrous with only a few hairs on 
the margins. Outer filaments tinged with pur- 
plish. No staminodes. Fruit a globular head 
with the usual long plumose tails, about 4 cm. 
long, the carpels also hairy, borne upon a 
honeycombed hairy receptacle. The persistent 
styles in the fruit are reddish, and the other 
long hairs silvery-white. 
The naturalization of a Himalayan Clematis 
in the mountains of New Mexico suggests the 
possibility that other plants from the same 
region might do well if introduced here, some 
of them being perhaps of economic value. 
T. D. A. COCKERELL. 
MESILLA PARK, NEw MEXxIco, 
November 25, 1899. 
_ SCIENCE. 
899: 
NOTES ON INORGANIC CHEMISTRY. 
Tue problem of the structure of the carbon 
molecule has attracted the attention of not a 
few chemists, though little progress has been 
made toward its solution, owing to the difficulty 
of obtaining soluble bodies of definite compo- 
sition by the action of reagents upon any form 
of carbon. Sometime since, L. Staudenmaier 
discovered a rapid method of oxidizing graphite 
to graphitic acid, and a continuation of this 
work is described in the current Berichte. 
Graphitic acid appears not to be a true acid, but 
a substance of a quinone nature. By heating 
it is converted into a simpler compound which 
the author calls pyrographitic acid, from which 
other derivatives may be formed. Among the 
oxidation products is mellitic acid C,(COOH),. 
From the analogy furnished by the oxidation of 
naphthalene to phthalic acid, it would appear 
that graphitic acid and hence graphite contains 
three naphthalene groups united together into 
a benzene nucleus. 
In the study of non-aqueous solutions more 
work has been done on ammonia as a solvent 
than on any other liquid. The work of E. C. 
Franklin and others shows that many salts dis- 
solve readily in liquid anhydrous ammonia and 
are electrolytically dissociated. According to 
Franklin, liquid hydrogen sulfid appears not to 
act in this manner as a solvent, and I know of 
no experiments with liquid hydrochloric acid. 
Great interest attaches to a series of experi- 
ments described by P. Walden, of Riga, in the 
Berichte, on liquid sulfur dioxid as an inorganic 
ionizing solvent. It is the more remarkable, as 
the solvent contains no hydrogen. As far as 
Walden’s experiments have yet gone, the halid 
salts have been found to dissolve readily in 
liquid sulfur dioxid and metathetical reactions 
take place in the solution. Organic substances 
of very different compositions dissolve readily, 
and often though solvent and solute are color- 
less, the solution is colored. A number of sub- 
stances were used for determination of molec- 
ular weight by the boiling point method. The 
solutions appear tolbe quite different from the 
aqueous solutions, showing the molecular weight 
in several instances double what would be ex- 
pected. The article is an interesting contribu- 
tion to the chemistry of solutions. 
