evolved on heating Mineral Substances, Meteorites, &c. 



Ferrous oxide in mineral after heating, cor- 

 rected for loss of weight of mineral 9'18 per cent. 



Difference in amount of ferrous oxide present 



in mineral before and after heating 1 -43 ,, 



The amount of ferrous oxide oxidised to ferric oxide can also be 

 calculated from the quantities of hydrogen and carbon monoxide 

 collected. 



2-180 c.c. per gram of hydrogen is 



equivalent to 1 '401 per cent, of FeO 



0*094 c.c. per gram of carbon mon- 

 oxide is equivalent to 0*057 ,, 



Total 1-458 



This number agrees very closely with that already obtained. 



The application of this method of investigation is very limited, as it 

 is difficult to obtain minerals which are quite free from sulphides and 

 carbonaceous matter, and which at the same time give on heating a 

 sufficient quantity of carbon monoxide and hydrogen. Other speci- 

 mens of chlorite gave large quantities of hydrogen on heating, but 

 they were found to contain sulphides. 



The minerals and rocks, which had been investigated up to this 

 point, had all been of igneous origin. At Professor Bonney's sugges- 

 tion, I next proceeded to determine whether minerals of aqueous origin r 

 containing ferrous oxide, water, and carbonates, would give hydrogen 

 and carbon monoxide on heating. Professor Bonney kindly obtained 

 for me a specimen of chalk marl, rich in glauconite, from the Cambridge 

 beds. 



The glauconite was freed from chalk as far as possible by washing 

 with dilute hydrochloric acid. The residue, seen under the microscope, 

 appeared to consist of foraminiferous casts, with adhering grains of 

 glauconite. The sample taken for the experiment contained 1*33 per 

 cent, of ferrous oxide, 15 per cent, of water, and 13'7 per cent, of 

 calcium phosphate. 



Heated in a hard glass tube to a dull red heat for some time, only 

 5-7 per cent, of the water was given off with a considerable quantity 

 of gas, which was found on analysis to have the following composi- 

 tion : 



CO., 4*518 c.c. per gram. 



H./. 3-128 



CO 0-083 



CH 4 , &c 0-204 



On heating to a bright red heat in the flame of a Bunsen burner, the 

 remaining water and more gas was given off : 



