S. AUport — On the Fitchstones of Arran. 9 



ture, and the formation of tlieir mineral constituents, that the facts 

 discussed in the two papers are by no means the same ; and I trust 

 it will Tbe found that the preceding observations have not been 

 superseded. There is one point of some interest to which it will 

 be necessary to advert. 



At page 43 the author calls the green pyroxenic crystals liorn- 

 blencle, but gives no reason for the statement. In my paper they are 

 described as migite, for the reason assigned, viz., that none of the 

 crystals, large or small, exhibit the faintest trace of dichroism. Now, 

 Tschermak has shown that hornblende is dichroic, while augite is 

 not ; and that we have thus, at last, a satisfactory method of dis- 

 tinguishing these minerals from each other. I also made the same 

 observation, quite independently, towards the end of 1870, and have 

 since repeatedly verified it, not only in diorites and syenites, but also 

 in hornblende schists ; in all cases this mineral is distinctly dichroic. 

 On the other hand, all the augite examined either remains quite 

 unchanged or, at most, exhibits a very slight variation in shade. It 

 is true that some specimens of hornblende are more strongly dichroic 

 than others, and I have observed that the pale green varieties possess 

 this property in a less degree than those of a brown colour ; still, 

 the palest green crystals exhibit in the dichroscope a far greater 

 difference between the two rays than I have seen in any specimen 

 of augite. Unless, therefore, it can be shown that some hornblende 

 is not in the least dichroic, the mineral in question must, on this 

 ground alone, be regarded as augite. Having very great deference 

 for the opinion of Prof. Zirkel, I have availed myself of a delay in 

 the publication of the paper to make a re- examination of all my 

 sections, and have found one well-formed crystal which happens 

 to lie in such a position as to exhibit a transverse section of the 

 prism. A careful measurement of the angles with the goniometer 

 gives the following results : 132° ; 137° ; 87° ; the true angles being 

 133° 30^ ; 136° 30^ ; 87°. Now we have here as close an agree- 

 ment as can be expected under the conditions ; and as the corre- 

 sponding angles in hornblende are so widely different, there can be 

 no doubt, I think, that the pyroxenic constituent of the Arran pitch- 

 stone is augite. 



EXPLANATION OF PLATE I. 



Fig. 1. — Section of pitclistone from the East coast, containing fern-like groups 

 of green belonites and part of a spherolite. Described on p. 2. 



Fig. 2. — Striped pitchstone from a dyke south of Tormore. See p. 6. 



Fig. 3.— Pitchstone fi-om a vein in the cliffs near Tormore, showing stellate groups 

 of crystals in a brown glassy base. The latter should have been represented as in 

 Fig. 1, i.e., without a sharp outline round the stars. See p. 6. 



Fig. 4.- — Group of felspar crystals, as seen in polarized light ; « is a twin crystal 

 of orthoclase, showing two colours sharply divided by the plane of junction ; b, a tri- 

 clinic crystal with coloured bands. Several fern-like groups of belonites are attached 

 to the sides of the felspar. Magnified 20 diameters. 



Fig. 5. — Felspar crystals in polarized light x 20 diameters. The lower group 

 consists of several imperfect triclinic crystals, with an angular space between them 

 filled with the granular matrix in which they are imbedded. The dark part extending 

 from top to bottom is of one uniform coloui', and is evidently orthoclase. The upper 

 crystal is probably an imperfect twin of orthoclase, although one half appears to con- 



