256 S. Stanley Jevons — Scale of Rock Texture. 



between certain hi_^her and lower limits. In this latter case a state- 

 ment of the general average size will still give a very good idea of 

 the appearance of the rock, but for the sake of accuracy it would be 

 well also to state the limits. As examples I may quote the following 

 readings I have taken from various hand-specimens : Carrock Fell 

 gabbro, 8; minette from Sale Fell, 1'5 ; Shap granite, 3 to 4, with 

 porphyritic crystals, 25 ; Buttermere granophyre, 2 ; the tonalite of 

 Monte Tonale, 4; a granite from Ben Nevis, 2-5, with interstitial 

 matter about 0*3, and porphyritic crystals up to 7. 



The principle and method are equally applicable to microcrystalline 

 rocks if one uses a micrometer scale in the microscope and reads 

 the lengths of the average crystals in fractions of a millimetre. For 

 instance, a specimen of the nosean-phonolite of the Wolf Eock, 

 Plymouth, is in texture 0-084 ; with porphyritic crystals, 1-0. 

 Perhaps it may be interesting by way of illustration to give the 

 measurements by which the size of the porphyritic crystals was 

 determined in this case. They are 23, 32, 60, 26, 40, 80, 40, 40, 

 28, 90, 40 and 60, mean 47. Since twelve divisions of the scale equal 

 •254 mm., the average length of the crystals measured is -996, say 

 1 mm. That there is a considerable difference between the maximum 

 and minimum lengths must be admitted (a rather greater one here 

 than usual, I think), but still there is a general average size, which 

 the eye can get hold of and distinguish easily from one a little 

 higher or a little lower. As another example I may mention 

 a biotite-olivine-dolerite from Kintellan, Argyleshire, which is in 

 the groundmass -89 ; porphyritic constituents 2-03. 



The following measurements which I have made of some of the 

 figures in Teall's " British Petrography" may be of service as a kind 

 of criterion by which any person may see whether he is measuring 

 in the way I have explained. The results quoted are the direct 

 measurements, as I have not troubled to reduce for the magnification. 

 Plate XXV, fig. 2, and pi. xxvii, fig. 1, are both easy to measure, and 

 give 15 and 16 respectively. PI. viii, fig. 1 is not so easy, because 

 many of the crystals are very ragged, but if each large patch of 

 viridite be taken as representing one crystal we get the figure 14, 

 or taking the magnetite separately (at 8), we have for the remainder 

 16. As an instance of how to measure a ragged crystal I will take 

 the brown hornblende in the lower half of the figure. From the 

 bottom right-hand corner to the rounded top I make it 21, and 

 I neglect the ragged tail which projects upwards on the right. 

 This procedure may seem arbitrary, but it is really necessary, 

 because otherwise such crystals would receive a value out of all 

 proportion with their area, the quantity I have already shown that 

 we are really endeavouring to express. In the case of ragged 

 crystals, therefore, discretion must be used to give a number which 

 would probably represent the linear dimensions of a crystal of the 

 same area but of regular outline. 



In pi. xii, fig. 2, the constituents must be mentioned separately, 

 thus — porphyritic crystals (only one in the figure), 23 ; lath-shaped 

 felspars, 8; augites, 3-6; iron-ores, 1. One of the most difficult is 



