1 24 Analysis, t^-c. of Cordier^s Essay upon the 



than any heretofore sug'g'ested. These phenomena appear taus 

 a phiin and natural consequence of the cooling of the interior of 

 the globe — a mere thermometric effect. The internal fluid mass 

 is submitted to an increasing pressure produced by two forces ol 

 immense power; although the effects are slow, and scarcely sen- 

 sible in their gradual operation. On the one hand, the solid 

 crust of the earth contracts more and more as its temperature 

 diminishes," (from radiation on the outside, and molecular com- 

 munication internally.) " This contraction is necessarily great- 

 er than the central mass experiences at the same time. On the 

 other band, this envelope, in consequence of the insensible ac- 

 celeration of rotatory motion, loses its interior capacity in pro- 

 portion as it recedes from a spherical shape. 



" The interior fluid substances are compelled to flow outward 

 in the form of lavas by the process which we term volcanic, and 

 with a precedent production of gaseous matters produced inter- 

 nally during eruptions Let no one be surprised at this hypoth- 

 esis: I can render it probable by a very simple calculation. 



" At Teneriffe, in the year 1803, I took as near as possible the 

 cubic dimensions of the ejected matters of 1705 and 1798. I did 

 the same by two eruptions more perfectly insulated in the inte- 

 rior of France ; in 1 806 those of the volcano of Murol in Au- 

 vergne ; and in 1809 those of the volcano of Cherchemus near 

 Izarles at Mezin. I found each eruption to have ejected less 

 than a cubic kilometre. On these facts, and others of the same 

 kind which I collected elsewhere, I feel authorized to consider 

 the cubic kilometre as the extreme general term of bulk of vol- 

 canic ejections. This is very small, compared to the whole 

 globe. Spread over its surface, it would not be one hundredth 

 of a millimetre in thickness. In exact terms, suppose the crust 

 of the earth to have a thickness of twenty leagues of five thou- 

 sand metres each, (fifty-seven miles,) a contraction that would 

 shorten the mean radius of the central mass 1-494 of a millimetre 

 would suffice to produce an eruption."* 



" Proceeding from these data, if we suppose that eontractioE 

 alone will suffice to produce the phenomena, and that five erup- 

 tions annually take place over the whole surface of the earth, 

 we find the difference between the contraction of the consolida- 

 ted crust and that of the fluid mass to shorten the radius of that 

 mass but one millimetre in a century. If there be but two erup- 

 tions annually, the same shortening will take place in two centu- 



* Kilometre 109364 yards English in length. A cuhic kilometre is about 

 two and a quarter million of cubic feet English. Millimetre ,03937 cubk 

 English inches. 



