4 MEMOIRS NATIONAL ACADEMY OF SCIENCES. rvoi.xiv 



The mineral portion amounted to 5.94 giains and consisted of: 



Pel cent. 



Silica SiOj 41. 16 



Alumina AI2O3 3. 97 



Ferrous oxide FeO 18. 48 



Manganous oxide MnO 0. 39 



Chromic oxide Crfii 0. 20 



Phosphoric acid P^Oj 0. 30 



Sulphuric anhydride SO3 5. 56 



Lime CaO 1. 92 



Magnesia MgO ' 26. 88 



Soda NajO 1. 14 



Potash K2O 0. 06 



Total 100. 06 



A recalculation of these figures gives the following, representing the composition of the 



stone as a whole: 



Per cent. 



SiOa 38.50 



AlA 3.71 



CrA 0.18 



FeO 17. 28 



MnO 0.36 



MgO 25. 14 



CaO 1. 79 



NajO 1.06 



K2O a 05 



P2O, a 28 



SO3 5.20 



Fe 5.74 



Ni 0.66 



Co 0.06 



Total 100. 01 



These figures fall well within the range of chondritic stones. No barium strontium or 

 other alkaline earths than those mentioned could be detected. No calcium in a water solu- 

 tion, hence no oldhamite. The mineral composition is olivine, monoclinic and orthorhombic 

 pyroxene, calcium phosphate (merrUlite of Wherry), chromite, nickel-iron, and troUite. 



Hartford (Marion), Linn County, Iowa. — The first descriptions of this stone are by Shepard.' 

 His determination of its lithological nature is excusable only in consideration of the times and 

 the means at his command. He wrote: "It appears to contain but a single mineral species of 

 this (i. e., 'earthy') description, and this one which . . . has imtil now escaped a separate 

 recognition." For this mineral he proposed the name Tiowardite and gave the complete mineral 

 composition of the stone as howardite, 83 per cent; nickel-iron, 10.44 per cent; magnetic 

 pyrite, 5 per cent; olivinoid and anorthite, traces. Some twenty and odd years later Eam- 

 melsberg ^ reviewed Shepard's work and showed the stone to consist of 10.54 per cent nickel- 

 iron; 6.37 per cent troilite; 41.58 per cent soluble silicate, and 41.24 per cent insoluble, the 

 soluble portion being identified as olivine; the insoluble, which was analyzed, being "almost 

 exactly a bisilicate," but which he does not name. 



An examiaation of thin sections from fragments in the Museum collection shows the essen- 

 tial constituents to be olivine and enstatite, with the usual interstitial calcium phosphate, 

 nickel-iron, and troilite. The structure is not strongly chondritic. (Fig. 2, PI. III.) No poly- 

 syntheticaUy twinned pyroxenes were noted. The phosphatic mineral was evident to the naked 

 eye in two instances as small white spots, perhaps 2 mm. in diameter, on a broken surface of 

 the stone. These were so soft and friable as to fall down to almost dustlike particles when 

 touched with a needle point. It is doubtless this brittle property of the mineral, causing it to 

 break away in the process of grindiag the section, that has prevented its earlier detection. It 



• Amer. Joum. Sci., vol. 4, 1847, p. 288, and vol. 6, 1848, p. 403. > Mon.-Ber. Berlin Akad., 1870, pp. 457-459. 



