Seybert’s Analysis of Pyroxenes 145 
carries over the vapor of the tar, and mixes it with that of 
the water; when the proportions are right, it is then 
curious to observe the very white, silver-like appearance of 
the flame. When the snapping of the tar would be incon- 
venient, it may be prevented by boiling it a few minutes be- 
fore it is used, to get rid of the coarser particles of water. 
But when it is desired to give a steady, durable light of any 
size with tar, the best mode I have yet tried is to raise the tar 
by machinery to the upper part of the tar vessel, passing it off 
through a small tube into a small cup, say to hold half a gill. 
Another small tube, from near the bottom of the tar vessel, 
passes upward through this little cup, about two thirds the 
way tothe top. The machinery for raising the tar may be 
kept in motion by connecting it with any moving power. 
The tar, when raised, flows out into the small cup, is there 
inflamed, and then continues to burn, without apy further 
attention, so long as the supply is kept up. It is furnished 
faster than it is consumed, but the overplus flows out before 
the small cup is full, down the small tube into the tar ves- 
sel at bottom, where, by its high temperature, it preserves 
the body of the tar at a temperature that makes it flow free- 
ly if ever so cold, As the tar in this little cup is all that is, 
or can be inflamed at atime, it renders it perfectly safe, 
and the time of burning in a sense unlimited. 
ours very respectful 
SAMUEL MOREY. 
Pror. Sruuiman. 
Art. XIX.—Analysis of the Pyroxene, found at the Frank- 
lin Iron Works, near Sparta, Sussex Co. New-Jersey, by 
Heyny Seyserr. 
Aw account of this mineral, by Messrs. Wm. H. Keating. 
and Lardner Vanuxem, was published, in the Journal of the 
Academy of Natural Sciences in this city.* After an at- 
tentive examination, they decided, that it was a mew species, 
and called it Jeffersonite. M1. Keating analysed it, and this 
they observe “ fally confirmed the conclusions drawn from 
its mineralogical characters.” His analysis gave the fol- 
* Journal of the Academy, Vol. 2, p. 194 
Vou. VIL.—No. 1 ° 19 
