380 B. N. PEACH AND J. HORNE ON THE 



There is one noteworthy feature connected with the sections prepared by us 

 from the Shetland lavas, and that is the absence of olivine in recognisable 

 forms. In this respect they differ from the diabase lava of the same age in 

 Shapinshay, Orkney, which we detected in 1879. One of the sections from 

 this locality shows that, in addition to the plagioclase felspar, there is much 

 olivine distributed in crystals and crystalline grains, which, for the most part, 

 has been converted into serpentine. Chlorite is also present, but the augite is 

 hardly represented at all. The magnetite, which is very abundant, frequently 

 envelopes the crystals of olivine either wholly or in part only. This type is 

 very different from any we have met with in Shetland. It might be termed a 

 felspar-diabase, rich in olivine and poor in augite. 



On referring to the table of analyses appended to this paper, it will be seen 

 that of the two types — the one taken from the porphyrites in Northmavine, and 

 the other from the diabase lava in Clouster Voe — the latter is the more basic. 

 The proportion of silica in the porphyrite lava from Rooeness Voe is 51*82, 

 while in the diabase from Clouster it is 48 "36 per cent. The alumina in the 

 former is 1414, while in the latter it is 19*73 per cent. Another difference of 

 some importance is in the relative quantities of magnesia : in the porphyrite 

 lava it is 1*76, while in the diabase lava it amounts to 5*97 per cent. It would 

 seem, therefore, that the diabase lavas from the altered Old Red area west of 

 Weisdale were not only more basic originally, but the greater proportion of 

 magnesia points to considerable alteration at a subsequent date. 



A section of volcanic ash from Dales Voe, Sandness, when examined micro- 

 scopically is found to consist of angular and subangular quartz grains, with a 

 few fragments of orthoclase and plagioclase felspar. It also contains fragments 

 of felsite, and a rock exhibiting an ill-defined micro-crystalline structure re- 

 sembling that of some of the devitrified rhyolites. A fine dusty felspathic 

 substance separates the individual grains. 



It may be more convenient to describe here the microscopic characters of 

 the intrusive diabase rocks, as they have close affinities both chemically and 

 microscopically with the interbedded lava just referred to. These affinities 

 are all the more interesting when we remember that the respective eruptions 

 of the two groups were separated by a considerable interval of time, during 

 which highly acidic rocks were ejected. The microscopic characters of the 

 diabase from the boss in Sandsting resemble those of the dykes in North- 

 mavine, with this exception, that the former are more coarsely crystalline. 

 The sections from the Skeld Hill show that the felspar crystals are much 

 decomposed, but in some cases it is possible to determine that they are plagio- 

 clase. Instead of presenting the usual clear faces, they have a dusty brown 

 appearance, which has almost obliterated their twin structure. It is observable 

 that the triclinic felspars, though larger than in the interbedded lavas, do not 



