316 MEMOIRS NATIONAL ACADEMY OF SCIENCES, VOL. XIII. 



being insoluble in hydrochloric acid. The other alternative is to assume a complex molecule which is split into two 

 sections by the action of the acid. 



The meteorite is now known as the Safford meteorite. 



The results of Eakins were referred to by Merrill, 2 and further examination of the meteorite 

 reported by him as follows: 



Three specific gravity determinations, made on fragments from 15 to 25 grams in weight, including both iron and 

 stony portions, yielded the present writer an average of 4.32. 



In thin sections the structure of the stony portions is found to be holocrystalline granular, sometimes strongly cata- 

 clastic. This latter structure is particularly conspicuous in those portions rich in metallic iron, where the feldspars are 

 often inclosed in the form of sharply angular fragments in the iron or in its numerous embayments. The appearance is 

 not, however, that of a clastic rock, but rather that of a crystalline variety which has been subjected to dynamic agen- 

 cies. The structure as a whole is quite irregular and , as above noted , porphyritic through the presence of large pyroxenes 

 which at times are 5 to 8 mm. in diameter. 



The groundmass of the stone is composed mainly of granules of pyroxenes and plagioclase of such size as to render 

 their determination by the microscope a matter of considerable ease, but which are interspersed with innumerable 

 rounded and irregular granular forms so minute and so lacking in crystal outlines as to render their true mineralogical 

 nature a matter of conjecture only. 



The feldspars as a rule show polysynthetic twinning. Sections without twin striae (and which are assumed to be 

 parallel approximately to oo P oo, show the emergence of an optic axis just outside the field, and give extinction angles 

 as high as 38, suggestive of anorthite. By means of a specific gravity solution a small quantity of the feldspar 

 (0.19 gram) was separated out and analyzed with the results given below, and which confirm the result of the optical 

 determination. They usually contain a large number of cavities and inclosures in this respect and with respect to 

 their shattered condition as well, resembling those of the Sierra de Chaca stone. 



The analysis of the feldspar is as follows: 



SiO 2 A1 2 0, FeO CaO MgO Na/) 

 42.02 37.77 trace 16.41 0.96 Undt. =97.16 



Two pyroxenic minerals are present. The one of a gray color, owing to minute cavities and dustlike inclosures, 

 and giving in all cases extinctions parallel and at right angles with the evident cleavage; microchemical tests show 

 the presence of magnesia but not of lime or alumina. It is hence assumed to be enstatite. The second is very clear and 

 pellucid, of a faint greenish tinge, though without evident pleochroism. It shows an imperfect prismatic cleavage, a 



pronounced parting parallel to oo P i has at times somewhat fibrous or platy structure, and gives extinction on P, 



measured against cleavage lines, running as high as 30. Granules of this mineral isolated for microchemical testa 

 were insoluble in hydrochloric acid. With hydrofluoric acid on a slide first covered with hard balsam they dissolved, 

 yielding abundant rhombs of magnesiau fluosilicate. The solution treated with a drop of dilute sulphuric acid yielded 

 gypsum needles, and with caesium chloride and sulphuric acid abundant minute, more or less modified octahedra of 

 caesium alum. The mineral is, therefore, assumed to be diallage, though the angle of extinction is small. Olivine is 

 quite inconspicuous, and were it not for the magnesia in the soluble portion of the stone would be quite overlooked. It 

 seems to exist intergrown with the enstatite and can not be isolated. The powdered rock after being passed repeatedly 

 through a solution of sufficient density to separate the feldspar still yields a small amount of gelatinous silica, the acid 

 solution reacting for both magnesia and lime, suggesting the presence of monticellite. It is possible, however, that 

 the lime may have come from inclosures of anorthite too small to be recognizable. 



Repeated attempts were made at separating the two pyroxenes for complete analysis, but the variation in density 

 was too slight to permit this, even when the silver-thallium nitrate solution was employed. 



Inasmuch as the presence of the minerals above noted, as determined microscopically, did not satisfy all the 

 requirements of the analyses of the soluble portions, further qualitative and microchemical tests were resorted to. It 

 was found that merely boiling the pulverized stone for a few minutes in distilled water was sufficient to give a solution 

 reacting for chlorine, sulphuric acid, lime, and iron. These reactions, considered in connection with the minerals 

 known to occur in meteorites, are sufficient to suggest, if not prove, the presence of gypsum as an oxidation product of 

 oldhamite and of lawrencite. The phosphoric acid suggests schreibersite, and the odor of sulphuretted hydrogen given 

 off by the boiling solution, troilite. Instead of, then, attempting to account for the result of the analyses on the assumed 

 presence of two minerals having practically the same molecular ratios, as was done in the paper above quoted, the 

 present writer would suggest the following as the probable mineral nature of the stone, including the metallic portion: 



(1) Nickeliferous iron. (6) Oldhamite (or secondary gypsum). 



(2) Enstatite. (7) Lawrencite. 



(3) Diallage. (8) Troilite. 



(4) Anorthite. (9) Schreibersite. 



(5) Olivine (or monticellite). 



According to the prevailing system of classification the stone must be called a mesosiderite; viewed from the stand- 

 point of terrestrial petrography, it would be classed as a gabbro with gradations toward pyroxenite. 



A few words more may well be written concerning the structure of the stone. This, as above noted, is crystalline 

 granular throughout, no glass whatever being detected. In the finer-grained siliceous portions the constituents have 



