62 THE SIDEROLITES AND PALLASITES. 



In a large number of the siderolites, etching develops in the iron those 

 structural planes, well known as the Widmannstattian figures, parallel to 

 the planes of the isometric octahedron and cube. This structure is parallel- 

 ized by the cleavage observed in magnetite, and by the observed structure 

 produced in titaniferous iron during the process of its alteration to " leuco- 

 xene." Indeed, many of the altered titaniferous irons show a structure closely 

 resembling some of the Widmannstattian figures. These latter figures can 

 be taken no longer as proofs of the meteoric origin of any iron, since they 

 have been developed in the terrestrial iron of the Greenland basalts. Tlie 

 Widmannstattian figures are so well known that it is unnecessary to repro- 

 duce them in the plates of this work, especially since good examples can be 

 fouud in the following papers and works : — 



Schreibers's Beitrage zur Geschichte und Kenntniss meteorischer Stein- 

 und Metall-Massen, Wien, 1820 ; Clark on Metallic Meteorites, Gottingen, 

 1853; Biichner, Bericht Ober. Gesell. Giessen, 1869, xiii. 99-115; Haidinger, 

 Sitz. Wien. Akad., 1855, xv. 354-360; 1862, xlv. (2), 65-74; xlvi. (2), 286- 

 297; 1863, xlviii. (2), 301-308; Amer. Jour. Sci., 1846, (2), ii. 370-376, 

 1853, XV. 363 ; Brezina, Denks. Wien. Akad., 1881, xliii. 13-16 ; 1884, xliv. 

 121-158; Tschermak, Sitz. Wien. Akad., 1874, Ixx. (1), 443-458; Denks. 

 Wien. Akad., 1872, xxxi. 187-195; Gustav Rose, Abh. Berlin. Akad., 1863, 

 pp. 23-161, etc. 



It is possible that could chemical analyses be made that would yield the 

 constitution of the siderolite masses as a whole, and thus enable their min- 

 eralogical and chemical constitution to be coijrdinated with their structure, 

 that the species siderolite might be separated into several distinct, natural, 

 and well marked species. The most probable divisions would be into those 

 composed of metallic iron, and those formed from metallic iron and pyrrho- 

 tite. The first might possibly be separated into those bearing much nickel 

 and those of nearly pure iron. Of course, no such subdivision of siderolite 

 should be attempted unless the composition and structure both pointed 

 towards the separation. This would require the study of a large number of 

 siderolites, an opportunity for which the writer has not, even if satisfactory 

 chemical analyses existed. 



A list of chemical analyses has indeed been collected and will be 

 appended ; but it is far from being v/hat could be desired, since it is the 

 custom of chemists to select the purest portion of the iron for analysis. In 

 fact but few analyses exist of siderolites that can be regarded as complete 



