rROCEEDlNGS OF THE EOTAL SOCIETY. 



677 



the distance of tlie eatth from the sUn would be .35'' and 

 .21" or .irt" respectively ; and that their real diameters are 

 about 103 and 95 or 71 English miles. There is no pro- 

 bability that either of these stars can have a satellite. The 

 coloui of Ceres is more ruddy than that of Pallas. They 

 ha\e generally more or less of a haziness, or coma, 

 but sometimes, when the air is clear, this nebulosity 

 scaiccly exceeds the scattered light surrounding a very small 

 star. From a view of all these circumstances. Dr. Hcrs- 

 chel proceeds to consider tlic nature of the new stars. He 

 thinks that they differ from the general character of planets, 

 in their diminutive dimensions, in the great inclination of 

 their orbits, in the coma surrounding them, and in the 

 mutual proximity of their orbits: that they differ from co- 

 mets in the want of eccentricity, and of a considerable ne- 

 bulosity. Dr. Herschel, therefore, wishes to call them 

 asteroids, a term which he defines as a celestial body, 

 which moves round the sun in an orbit either little or con- 

 siderably eccentric, of which the plane may be inflined to 

 tiie ecliptic in any angle whatever, the motion being either 

 direct or retrograde, and the body being surrounded or not 

 by a considerable atmosphere, or a very small coma. This 

 definition is intended to include such other bodies of the 

 same kind ?.s, Dr. Herschel supposes, will, in all proba- 

 lity, be hereafter discovered. Some additional observations 

 show, that the apparent comas, surrounding Ceres and 

 Pallas, scarcely exceed those, which are caused by aberra- 

 tion, round the images of minute fixed stars. 



The ineetings of the 20th and 27th of May were occu- 

 pied by an analysis of corundum, and of some of the sub- 

 stances which accompany it ; with observations on the af- 

 finities which the earths have been supposed to have for 

 each other in the humid way. By Richard Chenevix, Esq. 

 F.R.S. and M.R.I. A. 



After several ineffectual attempts to procure a solution of 

 corundum, Mr. Chenevix succeeded by means of subbo- 

 rate of soda, or common borax. He took 100 grains of co- 

 rundum, and having pulverised it in a steel mortar, after 

 repeatedly plunging it when red hot into cold water, he 

 washed off by muriatic acid whatever iron might have ad- 

 hered to it, and then levigated it in a mortar of agate, not- 

 ing the augmentation of its weight in the operation. He 

 exposed the powder with 200 grains of calcined borax in a 

 crucible of platina to a violent heat ; it was ■then boiled in 

 the same vessel with muriatic acid, which in about 1 2 

 hours dissolved the glass. The earths were precipitated by 

 an alkaline carbonate : and being redissolvcd in muriatic 

 acid, the silica was separated by evaporation. The alu- 

 mina was precipitated and redissolvcd by an excess of pot- 



ash, and separated from it by muri.ite of ammoniac. The 

 process is particularly exemplified in the instance of the sap- 

 phire, in which Mr. Chenevix found about one twentieth 

 of its weight of silica, although Mr. Klaproth could 

 scarcely perceive the presence of any silica. The consti- 

 tuent parts of many difierent corundums are enumerated ; 

 they all agree in the great proportion of alumina. T'he ma- 

 trix from the peninsula of India contained silica, alumina, 

 lime, iron, and a small quantity of manganese ; the felt- 

 spar, found in it, consisted of nearly the same ingredients, 

 with a greater preponderance of silica ; but the fibrolite 

 vras remarkable for being composed almost wholly of alu- 

 mina and silica, in the proportion of 3 to 2 ; the thallite 

 contained, besides these two earths, considerable portions 

 of lime and iron. The fibrolite of the matrix from China 

 contained alumina, silica and iron. The feltspar from 

 Ceylon difTeied but little from the Indian specimens. Mr. 

 Chenevix observes that, in such analyses, crucibles of platina 

 or silver ought to be exclusively employed : but that for 

 boiling earths in potash, silver must be preferred, since 

 platina is copiously dissolved by potash, its affinity with 

 this alkali being such as to enable it to form triple salts 

 with it, a property which the Spanish government employs 

 for detecting platina in gold. Mr. Chenevix thinks, that 

 the reddish colour produced in a weak solution of platina 

 by muriate of tin, is a more delicate test of its presence. He 

 observes, that neither potash nor soda, is, properly speak- 

 ing, a fixed alkali, especially when a little water is present. 

 In the second part of the paper, Mr. Chenevix considers 

 the supposed affinities of the earths for each other. He had 

 himself maintained the existence of some of these affinitie* 

 Kirwan and Guyton had carried the opinion much further. 

 But Mr. Darrac has combated this extension of the doctrine 

 with considerable success, and Mr. Chenevix has repeated 

 most of his experiments with a similar result. Dr. G. M. 

 at Freyberg, has excited further doubts on the subject. Mr. 

 Chenevix here enumerates the experiments of Guyton, and 

 considers them all as inaccurate, except those which re- 

 lated to the solution of silica in potash, and which "were 

 not new ; and even these he thinks scarcely sufficient to 

 justify, without further examination, the conclusion of an 

 affinity between this earth and others : and he explains 

 Guyton's error from the impurity of his material.?, especi- 

 ally from the presence of sulfuric acid, which Mr. Chene- 

 vix detected in the precipitates whenever they occurred. 

 The solubility of silica in acids after the action of an alkali 

 is. he thinks, a circumstance which has given the greatest 

 superiority to all modem anal)'scs ; and the solution is in 

 some measure facilitated by the presence of aUimina, Alu- 



