Editorial Coinmcnt. I2I 
put into accord with the theory of a closed reservoir (vase 
clos), such as required by the apphcation of the principle of 
Soret: — that is to say of variations of temperature, in a solu- 
tion more or less saturated. Here the dissolvent would nec- 
essarily be a portion of the alkaline magma, the elements dis- 
solved and susceptible of diffusion, of course, would be the 
ferro-magnesian magma and silica, one replacing the other. 
Such is observably the method of segregation, of differentia- 
tion, in a well arranged "family." But, as already remarked, 
this method will never reach the creation of two distinct fam- 
ilies having very different proportions of the alkalies, such as 
Etna and Vesuvius. 
4. Local modification of granitic magmas by endomor- 
phous metamorphism explains why certain volcanic series do 
not present the air of family, or consanguinity required by the 
theory of a universal magma separated by successive segmen- 
tations or kerns. In general, further, it is necessary to admit 
that in the geosynclinals where accvimulate the thickest de- 
posits, at the foot of the mountain ranges of the earth, the 
isotherms should undergo a rise toward the surface, such as 
would permit the fusion more or less aqueous not only of 
magmas already consolidated but even of clastic strata thus 
brought into the region of the central heat. Now it is pre- 
cisely in these places of greatest accumulation of sediment- 
ary formations that successive foldings will be produced, such 
as to permit the ascension and intrusion of granitoid rocks, 
and later the subsidence of extensive basins like the Mediter- 
ranean or the Pacific a])out the borders of which are located 
lines of volcanic rocks. The formation of such magmas 
again, can scarcely correspond to the uniformity which is de- 
manded by the theory of "closed reservoir." and of segrega- 
tions and differentiation of magmas already differentiated. 
5. But the greatest error in the theory of successive dif- 
ferentiation consists in not taking any account of the circula- 
tion of volatile agents, and of the ceaseless movement which 
they must impress on those elements which arc susceptible of 
being carried by them. Geologists are just beginning to be- 
lieve in the modifying power of surface water, simply charged 
with oxygen, carbonic acid, or with traces of the chlorides 
and fluorides. There is no doubt that they cause the develop- 
