124 Mr. R. Mallet on the Origin and Mechanism of 



goes, been hitherto done. He proposes to show that the play 

 of contractile forces alone is sufficient, in a large mass of ho- 

 mogeneous material, when cooling from a state of igneous 

 fusion, to account for all the observed phenomena of the 

 production, by successive Assuring, of prismatic columns and 

 of the cross joints or " cup-formed articulations " which se- 

 parate these into longer or shorter pieces — and to show that 

 concretionary or crystalline forces perform no necessary part 

 in the process, and merely tend, through a greater or less 

 want of homogeneity, more or less to modify the forms into 

 which parts of the mass become separated. Incidentally it will 

 be pointed out that the production of such jointing, or even of 

 prismatic masses at all, by the mutual compression of superposed 

 orbicular masses, in the way imagined by Gregory Watt, De la 

 Beche, Scrope, and some others, is impossible and inconsistent 

 alike with observed facts and with the physical conditions in- 

 volved. The distinguishing characteristic of masses of prismatic 

 basalt, when compared with the masses of lava-streams, are 

 that the basaltic masses are tolerably regular in their general 

 form, which is usually, and on the whole, more or less tabular, 

 that the material before splitting up was, on the whole, very 

 homogeneous throughout, and that there are few or no indica- 

 tions of joints or flssurings, whether curved or straight, except 

 those separating the prisms themselves ; whereas in great lava- 

 streams the general forms of the whole mass are not regular 

 nor tabular, and present great want of homogeneity. The masses, 

 when large, have usually filled ravines or valleys, so that the lava- 

 stream presents all the irregularities of contour of the valley 

 which forms its mould — its upper surface being also highly irre- 

 gular, and the interior of the mass, owing to the mechanism of its 

 flow, being extremely heterogeneous. These conditions involve 

 immense inequalities in the rate of cooling, and in the direc- 

 tions in which heat is principally lost at successive periods ; hence 

 internal strains through contraction and unequal subsidence 

 result, which crack and divide it (by planes which are very gene- 

 rally curved) into masses of very various sizes and devoid of all 

 regularity of form. The fractures, in fact, are often as irregular, 

 and as much the result of what we may call accident, as are those 

 into which a large lump of red-hot glass or red-hot flint becomes 

 shattered by being thrown into water. To speak of such irre- 

 gular breaking up as cuboidal, or as constituting one distinct 

 and analyzable class of jointed structure in lavas, is merely to 

 give rise to confusion. 



It must be remarked, however, that this absolutely capricious 

 fracturing may occasionally be seen passing into something like 

 regularity and approach to columnar structure, where previously 



