= 
Analysis of Iron Meteorites, Siderolites, and Meteoric Stones. 707 
Lines oF THE ALKALI Merats 1x Mereoric Irons: Kayser 
AND RUNGE’S MEASUREMENTS. 
Sopium. Porassium. N RuBIDIUM. 
X 
5896-16 } { 4047-36 \ 4201:98 
5890°19 i 4044-29 4215°72 
3303°07 
Lanes or orHeR Merats 1n Mereoric Irons. 
GALLIUM. COPPER. SILVER. LEAD. 
Xr Xr Xr 
4171°8 3274:09 3382°98 4057-97* 
4032°7 3247°65 3280-80 3683-60 
36389°70 
A calcium line was observed, wave-length 422691. This was 
believed to be caused by dust. There is an iron line at 4227°60 
(K. & R.), but it does not appear in flame spectra. In the 
meteoric irons from Arva, Hungary, in which there is troilite, and 
in the specimen from Cafion Diablo, Arizona, there is a trace of 
manganese. The lines just visible are those with wave-lengths 
4033 and 4030 (Liveing and Dewar’s numbers are 4032-0 and 
4029°5). 
The nickel and iron lines are strong in all the specimens. The 
alkali metals are weak, potassium being found in traces only. 
These spectra disclose a marked difference between meteoric iron 
and telluric metal, not only in the presence of a large proportion 
of nickel in the former, but in the absence of manganese, an 
element which is invariably contained in the latter. 
The presence of gallium in variable proportions in the iron 
meteorites is remarkable. The only one in which its occurrence 
was at first doubtful was that from Thunda, Windorah, Queens- 
land. Cobalt occurs in all specimens except in one from Virginia, 
but it does not appear in the meteoric stones. 
The meteoric stones all contain chromium in variable propor- 
tions, and manganese in traces only. 
The meteorite from Atacama consists of a honeycombed mass 
* This line is shown as 43857 in the plate 7, Phil. Trans., vol. 185, 1894, by an 
error of the engraver. 
SCIEN. PROC. R.D.S., VOL. VIII., PART VI. 3F 
