is»; 
///' Ann mil II i riiiJiiilist . 
Dectlllbfr, 1H81 
s()(»n iiftiT tlu' ••fall aiitl idiltiislied in the Amitiu.'' a iiiaii"azine 
<'(lit»'<] liy the students of the Iowa Ajiriciiltural oolk'irc. 
l-'nr comparison tlii'sc analyses will lie placcil tojrcthi'r. 
rwKCOKDEl) ANAI.YSKS OK THE WINXKI5A(!0 METKOHITK. 
'Kiikiiis 
Fiitoii 
ToiTt'V ;i 
11(1 liiirbour 
(Jovt. Surv. 
matrix 
it'litirc. 
Xickelilerous iron 
19 40 
20.22 
45.00 
SiO„ 
35.71 
2;$.H8 
4T.O:^ 
25.H0 
FtO 
4^15.12 
2fi.Tl 
•20.4:5 
16 IH 
MiiO 
24 (M) 
25.1)1 
2.1l<i 
1.6:^ 
Al..(), 
2.(t8 
2.62 
2.94 
1.62 
CaO 
1.41 
Tr. 
17.5S 
9.67 
s 
2.1:5 
Tr. 
K,0 
.SO 
Tr. 
Ni 
.1:) 
Tr. 
Cr,03 
.OS 
Tr. 
10O.S« 
t»9.:;4 
99 94 
99.96 
In their last article professors Torrey and liarUour do not srive 
tlie pereentaije of metal. l>nt in a former artic-li' un<ler the same 
analysis of matrix it is staled to lie 457^.** In the fonrth eolnmn 
the analysis is ealenlatetl on this Itasis. 
It will l»e observeil that 'l'orre\' and Bai'lioiir's analysis contains 
over twice as much metal, almost seven tin)es as mnch lime and 
less than one-fourteenth as much ma<iMU'sia as that rejjorted l»y 
Kakins. Do the sections of the aerolite show such diversity in 
ohemieal comixisition? That it is not altotiethei' homoijeneous is 
evident from a microscopical examination. The nickeliferous 
iron occurs in nodules of varyin<i size while tliei'e is no reiju- 
laritv in the distrilmtion of olivine, triolite. fermus oxide 
or otlu'r constituents. .Mso in the fi'aiiiuents the surface is 
oxidized in an extremely thin layei' while on the larirer masses it 
is sonu'what thicker. That this variahility is not conspicuous in 
dealinu; with laruer masses is indicated In Torrey and Harlnmr 
olttaininir tlii' identic:il specilie >rnivity as that reported liy Kunz 
(;>.«»OS)tt while my own CJ.Ih) is nearly a nu'dium lietween Kunz 
and Kakins (:).Stl4 at liS. ")('.). J j My chemical analysis coire- 
sponds (juile closely with Kakins. 
~*Vol. XIX July, 1890, p. 62. tAm. J. Sc. Vol. 140, 1890, p. 312^ 
;J:Allowiiis 45 per cent, for metal. 
J^Someof the iron coniltine<l a.^< Fe8 heuce per cent. FoO too large. 
In oriL^inal article estiniated as Fe^O, (29.68 per cent.) * Ferric oxide. 
**Ani.'.Io. Sc. Vol. ]:i9. ]>. 521. 
t+Trans. N. V. Acad. Sc. NO), ix, No. 8, May-June, abo Ameuican 
•TIkoi.Ooist, A'ol 6, p. 249. 
t+Aui. J. Sc. Vol. 140. p. -.Wl. 
