E. T. Allen — Native Iron in Missouri. 101 



scratching one another with difficulty. When filed the metal 

 exhibited a color almost silver white and a high luster which 

 remained permanent in a dessicator. While most of the pieces 

 weighed about *5 gr., there were a number which weighed 2 

 grms. or more, one weighed 8*4 grms. and the largest 45*4 

 grms. The last was only a little darkened by tarnish when we 

 received it. On the edge a layered structure was very notice- 

 able. In the other pieces this was not so apparent, but they 

 often separated along cleavage planes when hammered and by 

 the aid of a pair of pliers, one layer after another, sometimes 

 exceedingly thin, could be peeled off more or less perfectly. 

 The surface between these layers was sometimes blackened, but 

 often entirely fresh and metallic. Probably on account of this 

 structure and perhaps also on account of unequal hammering 

 by the drill, the specific gravity of different pieces varied con- 

 siderably. That of the largest was only 7*43, while that of 

 the smaller pieces varied (after the outer layer was removed) 

 from 7*63 to 7*73. The layered structure also made it difficult 

 to produce a continuous polished surface, the boundaries of 

 the layers showing in it as fine irregular lines at whatever 

 angle the plane was cut. The action of dilute nitric acid was 

 tried on such a surface, but it appeared to attack it evenly, 

 developing no semblance to Widmanstatten figures. In the 

 chemical examination of the iron we employed only the metal- 

 lic core prepared by filing away the outer layers. This dis- 

 solved in hydrochloric acid with the evolution of hydrogen 

 which possessed comparatively little odor, and left a very slight 

 residue of crystalline silica and a few minute fragments of 

 carbonaceous matter. A careful analysis of a solution pre- 

 pared from several grams of the iron, failed to reveal any 

 traces of copper, nickel, cobalt or other foreign metals. With 

 the exception of oxide of iron, silica, phosphorus and carbon 

 in small quantities were the only impurities detected. The 

 analyses made of different pieces gave nearly identical results. 

 Iron — 



1. Metal taken = -2910 gr. KMn0 4 required = 42*53 cc . l cc = 



•006785 gr. iron. Fe = 99*16 per cent. 



An examination of the iron taken, by means of a lens, 

 showed small dots of rust in a few places that seemed to pene- 

 trate deeply. Other determinations made on smaller pieces 

 where the rust appeared to have affected the iron still more 

 gave lower results. 



2. Metal taken = -2147. KMn0 4 required = 31*31 cc . Fe = 98-93. 



3. Metal taken = -4611. KMn0 4 required = 66*88 cc . Fe = 98*40. 



Silica — 



1. Metal taken =3*1603 gr. Si0 2 obtained = -0117 gr. =*37 per 



cent. 



2. Metal taken = 5*2953 gr. Si0 2 obtained = -019G gr, = *37 per 



cent. 

 Am. Jour. Sci. — Fourth Series, Yol. IV, No. 20. — August, 1897. 



