1884.] 699 (Ashburner, 
glucina, I am not aware that this method has ever been suggested to sep- 
arate glucina from alumina by boiling a solution, nearly neutralized with 
sodium carbonate, with sodium acetate. It is a known fact (see Graham- 
Otto’s Anorganische Chemie, by Michaelis, iii, 2 Hilfte, p. 694) that from 
a solution of beryllium chloride the glucina is precipitated on boiling with 
sodium acetate,* 
Dr. Winkler does not state that he has tested his so-called alumina for 
its purity, which is unfortunate, or he would have found thata slight trace 
of it might have been present, but that the precipitate was nearly pure 
glucina. There can be very little doubt that the Hhrenfriedersdorf and 
Stoneham mineral are identical in composition. There is also a larger 
percentage of ferric oxide in Dr. Winkler’s analysis than found by me. 
Might this not have come from the molybdic acid which he used? The 
ammonium molybdate—prepared from Merk’s molybdic acid—which I 
use contains in 100° 0.002 grms. ferric oxide. As I used measured 
quantities, a corresponding amount of ferric oxide was deducted. 
University oF Pennsyivanta, Philadelphia, October 17, 1884. 
Notes on the Natural Bridge of Virginia. By Charles A, Ashburner. 
(October 3, 1884; see page 690.) 
During a recent trip to Virginia (September 2 to 6), I visited the Natural 
Bridge, and although in possession of the tourist guide book of the lo- 
cality (edition of 1884) and the admirable articles published by Major Jed. 
Hotchkiss in ‘The Virginias,’’ I failed to obtain certain information relat- 
ing tothe bridge which would be of special interest to the topographer 
and geologist. Some of the observations which I made, although of a 
general character, may be of interest to members of the American Philo- 
sophical Society. 
The bridge is undoubtedly the remnant of the top of a cave which was 
*In order to show the value of the method used by Dr. Winkler for the 
separation of alumina and glucina, a quantity of beryllium carbonate was 
dissolved in hydrochloric acid, evaporated to dryness, diluted with about 
150° of water, nearly neutralized with sodium carbonate, then about 2 grams 
of sodium acetate were added and the solution heated. At about 80° it be- 
came turbid, and after two hours boiling a considerable precipitate had 
formed, which was filtered off and washed, It was dissolved in dilute hydro-, 
chloric acid, then sodium hydrate was added to redissolve the precipitate, then 
it was diluted with much water and gave on boiling beryllium hydrate, which 
after ignition weighed 0,0855 grams, The filtrate from the precipitate produced 
by sodium acetate was precipitated by ammonia and gave 0.2705 grm,. This 
shows that 24.8 per cent were precipitated by boiling, and that the method is 
worthless for the separation of glucina from alumina, By a greater dilution 
and a more strict neutralization perhaps all the glucina might have been 
precipitated, 
