296 CHEMICAL DISCOVERY AND INVENTION 



in a heated state. During the upheaval of mountain chains, and 

 in times of superficial disturbance owing to contraction, water 

 may penetrate through the crust of the earth and thus come into 

 contact with these carbides, the hydrogen of the water uniting 

 with the carbon, while the oxygen of the water forms oxides with 

 the metals. From direct experiment on cast iron and on Spiegel- 

 eisen, both of which contain iron and manganese united with 

 carbon, it has long been known that this change is produced by 

 contact of the metal with water or acids. This hypothesis 

 harmonises also with what is known of the constitution of our 

 earth, which is probably a metallic mass, having a thin earthy 

 crust on the surface, like so many of the meteorites which fall 

 from the skies. The mean density of the earth as a whole is 

 about 5-5, that is to say it would weigh 5-5 times as much as an 

 equal bulk of water. But the crust of the earth accessible to us 

 consists of minerals which are generally not more than about 

 twice as heavy as water, and therefore the interior must contain 

 a large quantity of something much denser. 



An alternative theory has been proposed more recently, about 

 1904, by Professor Paul Sabatier of Toulouse, whose discovery of 

 the remarkable catalytic actions of certain metals has been referred 

 to elsewhere (p. 202). He assumes that in the depths of the 

 earth are found deposits of the alkaline metals, sodium, potassium, 

 etc., as well as compounds of these metals with carbon. This 

 assumption is less novel than might appear, if it were not re- 

 membered that Davy, a century earlier, had supposed that 

 deposits of this kind might occur in the earth's crust, and had 

 attributed volcanic explosions in certain cases to the entrance of 

 water, with which as is well known they react violently. 



Sabatier imagines that such deposits of metal may be the 

 cause of the production of hydrogen, while their carbides in 

 contact with water simultaneously produce acetylene. These 

 two gases in variable proportions are then assumed to come into 

 contact with one or other of the metals, nickel, cobalt, or iron, 

 in a finely divided state and at an elevated temperature, the 

 result being that by condensation of the acetylene alone or by 

 union with hydrogen under the influence of the heavy metal a 

 mixture of hydrocarbons results. 



These may consist of (1) saturated paraffins as in the Pennsyl- 

 va*nian petroleum, or (2) cyclic paraffins such as occur in the 

 Caucasian oil, or (3) a mixture containing also benzenoid hydro- 



