THE TRAP DYKES OF THE ORKNEYS. 883 



crystals. I find no evidence of their primary origin. In that case the period of their 

 crystallisation would have been utterly indeterminate. It must have been one of the 

 earliest, one of the intermediate, and probably also the very last mineral to crystallise. 

 It has never traces of idiomorphism, or of that expansive crystalline force of which 

 Cross finds evidence (XIII. p. 686). Its constant association with calcite is practically 

 decisive as regards its origin in this particular rock. 



Another suggestion offers itself, viz., that the apparent phenocrysts may have 

 resulted from the decomposition of some such mineral as nepheline. In fact, out of 

 over thirty sections, two show each one crystal partly enclosed in porphyritic hornblende, 

 and corroded where surrounded by the groundmass. The only mineral these resemble 

 is nepheline, but in view, of the chemical composition of the rock it is difficult to see 

 how it could have been produced. It may be recalled, also, that the very early 

 phenocrysts of augite resemble aegirine augite very closely. What this mineral is I 

 have been unable to decide with certainty. 



General Features of the other Camptonite Dykes. 



Of the greater number of the camptonite dykes scattered so abundantly over the 

 West Mainland and Eousay, little remains to be said. They resemble in most respects 

 the groundmass of those already described. Most of them show in section phenocrysts 

 of olivine, less commonly of augite, and of hornblende only very seldom. The 

 groundmass is panidiomorphic and noncrystalline, consisting of augite and hornblende 

 in very variable relative proportions, and usually intergrown, the hornblende surround- 

 ing the augite or forming the terminations of the crystals. Felspar is never absent, 

 and may form quite one-half of the rock. It is mostly plagioclase, which stands as a 

 whole between andesine and labradorite, though orthoclase would seem to be present in 

 small amount in many dykes. It surrounds the plagioclase, or forms a sort of 

 interstitial substance without crystalline form. Often the long plagioclase felspars are 

 grouped in radiating fan-like aggregates, spreading out from a common axis. Except 

 in rocks which approach the monchiquites, traces of a glassy base are absent. The 

 felspars are usually filled with hair-like apatites. They are often greenish, and may be 

 in part actinolite, as suggested by Brogger (VIIT. p. 50). 



Most of the rocks contain pale rounded felspathic ocelli. These consist of radiating 

 plagioclase crystals, mixed with long narrow hornblendes. Even when augite occurs in 

 the groundmass elsewhere, it is absent from the ocelli. In the Hoxa dykes the felspars 

 are separated by much glassy material. Orthoclase may be comparatively abundant, 

 though not obvious in other parts of the rock, and is found chiefly in the interior. In 

 the centre there is always an irregular mass of calcite, filling what appears to have been 

 a miarolitic cavity. These ocelli consist, in fact, of the last minerals to crystallise, and 

 are due to the accumulation of globules of the still liquid magma at a late period in 

 consolidation. Crystallisation has proceeded from margin to centre, leaving a cavity to 



