PHYSICAL CONDITIONS AT BEGINNING OF LIFE 587 



searches), which is a second condition of the continuance of 

 terrestrial life ; is the chief source of the chlorination of the sea, 1 

 and also probably the chief source of phosphoric acid. 



It was long assumed, from the enormous cloud-formation 

 and the heavy downpour of rain that usually follows, that the 

 chief eruptive constituent was steam. This, however, has lately 

 been called into question by Brun, who has shown that in 

 some eruptions water vapour may be completely absent. It is, 

 however, well established that in an eruption there are often 

 quantities of hydrogen. This, with the other eruptive gases, 

 carbon dioxide, ammonia, hydrochlorides, etc., would be driven 

 into the air with explosive force, together with the usual 

 scoriae and volcanic dust. On this dust the water vapour of 

 the atmosphere condenses, while the hydrogen burns, uniting 

 with the atmospheric oxygen to form more water, so that, 

 whether steam is part of the eruption or not, heavy rainfall 

 is a familiar if not invariable characteristic. 



We should have, then, in the upper levels of an eruption, 

 a concentrated mixture of carbon dioxide, water vapour, 

 deoxidised air, free hydrogen, and various hydrogen, chlorine 

 and sulphur compounds, with, in all probability, large quantities 

 of ammonia. This latter, uniting with the hydrochloric acid, 

 gives rise to sal-ammoniac, which is a familiar volcanic product. 



Yet another characteristic of an eruption is an incessant 

 play of lightning. This may be due to the condensation of 

 the atmospheric water vapour, as in the ordinary formation of 

 thunder-clouds, or to gaseous friction from the explosive force 

 with which the gases are driven into the air, as in an Armstrong 

 hydro-electric engine. 2 In any event, the electrical potential is 

 there, and in the presence of water vapour it is probable that 

 it would be manifested in the form of the silent discharge, as 

 well as in the more violent bolts of lightning. 



In an eruption, therefore, conditions apparently exist prac- 

 tically identical with those under which Berthelot, Lob and 

 others have realised their syntheses. The volcanic formation 

 of formaldehyde and its congeners seems readily conceivable. 

 We know of ammonia in abundance, and the formation of 

 prussic acid, especially in a primeval atmosphere devoid of free 

 oxygen, seems more than probable. One or the other of these 



1 Cf. Suess, Geog. Jonrn. 20, 520, 1902. 

 3 Judd, Volcanoes, p. 28, 1881. 



