82 THE SIDEEOLITES AND PALLASITES. 



becomes less, and the greenish serpentinous product more abvmdant, until 

 they pass into a greenish-gray serpentinous rock spotted with secondary 

 ferruginous products. In many of the intermediate forms, short, brilliant 

 crystals of actinolite are to be seen, while the rock assumes a more or less 

 perfect schistose structure. 



In the microscopic sections the series passes from a more or less spongi- 

 form mass of magnetite, holding olivine and more or less feldspar, into forms 

 that show in portions of the mass an alteration to a greenish serpentinous 

 aggregate. It then passes into a form destitute of the feldspar, in which 

 the partially altered smoky olivine grains are surrounded by a band of 

 secondary actinolite, while the greenish serpentinous product increases in 

 abundance. These changes go on with diminishing olivine and increasing 

 actinolite and greenish serpentinous material, until they have entirely 

 (especially the last) replaced the olivine. The structure I'emains the same, 

 but the magnetite sponge is more discontinuous and in part dissolved, while 

 talc appears. In others dolomite is seen. 



In some sections the greenish secondary products, with little or no actin- 

 olite, replace the olivine. The figures 5, 6 (Plate I.), and 1, 2, and 3 (Plate 

 II.), fairly represent the general structure of these rocks, and their resem- 

 blance to the meteoric pallasites. 



No satisfactory analyses of the pallasites exist except that of Professor C. 

 A. Joy of an Atacama meteorite, but the imperfect ones that have been 

 found have been tabulated, so as to give a I'ougli approximation to correct- 

 ness. The specific gravity determinations are also not satisfictory as a 

 whole, since they have not been made upon characteristic specimens, l)ut 

 upon selected ones. They run from a little above 7 to some below 4, but 

 it is probable that the majority of typical pallasites lie between 4 and 6 ; 

 although we must expect to find them graduating in specific gravity, as well 

 as in other characters, into both the siderolites and peridotites. The silica 

 ranges from 3 per cent up to 33 per cent, but the probable limits are 

 between 5 and 30 per cent, averaging about 20 per cent. The magnesia 

 ranges from 2 to over 30 per cent, but the average probably will be found, 

 by correct analyses, to lie somewhere betweeu 10 and 20 per cent. The 

 iron in various conditions is a variable quantity, but averages about 60 per 

 cent ; while the nickel, with one exception, is less than 10 per cent. In the 

 terrestrial forms (Ciuiiberlandite) nickel is wanting, l)ut from G to 15 per 

 cent of titanic oxide occurs. Some tin, copper, zinc, cobalt, phosphorus, sul- 



