ON THE ANALYTICAL FORMS CALLED TREES. 



257 



mitted to this test. The spherules do not exceed 1 millim. in diameter. Iron 

 and iron sulphide are disseminated tlirough both the spherules and pulverized 

 ground mass, the latter compound sometimes forming grains and spherules 

 about which metallic iron has incrusted, and conversely ; and the general 

 character of the meteorites is that of the minutely chondritic class. 



"Water dissolves out from the stones 0-12 per cent, of sodium chloride, which 

 can also be sublimed from them in a hydrogen atmosphere at a red heat ; and 

 the presence of a trace of copper can be detected by the aid of a spectroscope. 

 Besides metallic grains of iron (7-81 per cent.) containing nickel and cobalt 

 and iron pyrites (14'28 per cent., no troilite or iron monosulphide, it appears, 

 occurring in this aerolite), and 1-36 per cent, of sodic chloride and hygroscopic 

 water, there was found 42-4-i per cent, of silicates soluble, and 33-3-lr per 

 cent, insoluble (probably including chromite) in hydrochloric acid ; total (with 

 loss 0-G6 per cent.) 100. Of the latter silicate or bronzite no analysis is given ; 

 but of the soluble silicate the following was the composition : — 



SiO,. 



17-20 13-80 



MgO. 



FeO. 

 11-33 



MnO. 



0-05 



Total. 

 42-44 



and no traces of calcium, barium, or strontium were detected in the stone. 

 A perfectly similar chemical analysis of the same meteorite by Dr. R. von 

 Drasche is added by Von Tschcrmak to the above account of its miiicralo- 

 gical examination and description. 



On the Analytical Forms called Trees, ivith Application to the Theory 

 of Chemical Combinations. By Professor Cayley, F.R.S. 



[Plate VIII.] 



I HAVE in two papers " On the Analytical forms called Trees," Phil. Mag. 

 vol. xiii. (1857) pp. 172-17G, and ditto, vol. xx. (ISGO) pp. 337-341, con- 

 sidered this theory, and in a paper " On the Mathematical Theory of Isomers," 

 ditto, vol. xlvii. (1874) p. 444, pointed out its connexion with modern che- 

 mical theory. In particular- as regards the paraffines C„IIo„^;, we have n atoms 

 of carbon connected bj^ ji—1 bands, imder the restriction that from each 

 carbon-atom there proceed at most 4 bauds (or, in the language of the papers 

 first referred to, we have n knots connected by » — 1 branches), in the form 

 of a tree ; for instance, n = 5, such forms (and the only such forms) are 



3o 



71 

 O 



3»- 



-•3 



is 



20 



J 



lS7i 



