Chap. XXIV.] kareh constituents of oees. 
braunite, and soft oxide, from Maadri ; and 0-055% in a sample of 
psilomelane containing some pyrolusite from Sivarajpur. The average 
of the 59 determinations is 0-023%. This is fortunately too small 
an amount to affect the commercial value of the ores. Manganese-ores 
could probably stand the presence of a considerably larger amount of 
arsenic than the above without the market value being affected ; for 
such amounts of arsenic would probably be volatilized off in the blast 
furnace. 
Every one of 25 analyses by J. and H. 8. Pattinson shows a small 
amount of sulphur. There is remarkable uni- 
Sulphur in .nangane^^e- formity between the amounts present, the 
ores. . . '- 
largest being 0-060%, in a sample from Kajli- 
dongri, and the smallest 0-01-1% in a specimen from Garbham. The 
average of the whole 25 is 0-029%. The sulphur may be suspected to 
be present in the form of minute quantities of barytes, especially as the 
maximum figure, 0-060^o, applies to Kajlidongri, in the ores of which I 
have recognized barytes. Such small amounts of sulphur do no harm. 
Indeed manganese-ores could probablv stand the presence of a consider- 
able quantity of sulphur, without their commercial value being reduced, 
except in so far as the sulphur took the place of manganese-ore ; for 
manganese acts as a remover of sulphur in the furnace. 
There seems to be considerable difference between the results of tests 
made by different chemists on Indian manga- 
Cobalt and nickel in r .-l xi_ii. i-ii 
inancanese-ores. nese-ores for the presence of cobalt and mckel. 
In many cases the chemists attached to the 
mining companies have oeen unable to prove the presence of these con- 
stituents ; whilst in other cases they have detected them. In some 
analyses that I once made of some Indian ores I found 0-48% of CoO -f 
NiO in a picked specimen of mixed psilomelane-brauuite ore from 
Kodegaon ; 0-17 of CoO and 0 38 of NiO in a specimen of wad from 
Sontulai in the Hoshangabad district of the Central Provinces ; 
and 0 27 of CoO and 0 56 of NiO in a specimen of conglomerate 
cemented by psilomelane from Pola Khal in the Dhar Forest, 
Central India. Dr. T. L. Walker found 0-82% of CoO in a specimen 
of wad from Olatura in the Kalahandi State. In the samples sent to 
the Imperial Institute the oxides of cobalt, nickel, and copper, were all 
determined together. The amounts of these constituents in 19 
analyses vary from nil to 0-08, the average being 0-022%. These 
analyses all represented specimens and samples from the Nagpur-Balaghat 
