80 



of molten matter still exist, concentric with the crust the common 

 source whence they all proceeded, and whose vents or outlets these 

 dikes once formed, in past stages of the earth's history, when the 

 various rocky materials were elaborating beneath the primeval ocean. 

 The pressure of a great depth of that ocean, or of other strata, amid 

 which they were poured out, gave these various igneous products 

 that density and compactness, which constitute almost the sole differ- 

 ences between them and the modern products of fire thrown out 

 under the pressure of the atmosphere only. They differ little in 

 their chemical composition mainly in that arrangement of com- 

 ponent parts which would be given by different rates of cooling from 

 a state of fusion, to which reference has been already made in previous 

 portions of these notices (Art. 31, 37, 40). 



56. The trap rocks of Arran may be ranged under three classes, 



The Felspathic, comprising porphyry, claystone, compact felspar, 

 and pitchstone. 



The Hornblendic, as basalt, greenstone, clinkstone, and amyg- 

 daloid. 



The Hornblendo-felspathic, as syenite and trap-porphyry. 



These are all intimately connected, one species often passing into 

 another by regular gradations, and they are all found in the same 

 relative positions with respect to the sedimentary strata. Among 

 themselves they do not preserve any order of succession, nor do they 

 occupy separate areas, so that their continuity cannot be reckoned 

 on through a considerable space. Nor can they be indicated on a 

 map by distinct colours, so intimately are they blended with one 

 another. Basalt, greenstone, and trap-porphyry are by far the most 

 abundant, as well in overlying masses as in dikes and interposed 

 beds. Porphyry is next in abundance, and occurs in all these posi- 

 tions ; pitchstone alone has not overflowed the surface, and occurs 

 only in dikes and beds. The overlying masses are limited to the 

 southern section of the island, while dikes occur everywhere, not 

 however with the same frequency in all parts. On this subject 

 Professor Phillips was, we believe, the first to offer a good gene- 

 ralization : " Dikes are most abundant at some distance from the 

 granitic centre. At Corrygills, at Lamlash and Tormore, they 

 are exceedingly abundant in the red sandstone, while in the north- 

 eastern face of the island, where that rock is nearer to the granite, 

 fewer dikes appear ; and about Loch Eanza the slate is still less 

 divided by them. Perhaps we may venture to add another gene- 

 ralization, viz., that these dikes are most abundant beyond the 

 line of violent flexure of the strata from their horizontal position. 



