Mineral Sulphides of Iron. 227 



A comparison of the different values of p Q with the average 

 value, c, for any crystal shows that the difference is greater 

 than 2 per cent only for crystal A, and in this case the angle 

 from which p Q was computed was so unreliable that it was not 

 considered in computing the average. In all cases where the 

 angle from which any p Q was computed seemed to be reliable 

 the difference between p and c is less than 1 per cent of c. 

 An error as large as 1 per cent in c may, therefore, be regarded 

 as exceptional. One possible exception is crystal I, as the 

 measurements on it were unsatisfactory and it has, therefore, 

 been discarded in the discussion to follow. If, however, we 

 compare the value of c for the different crystals we find a maxi- 

 mum difference of 0*0660 between crystals M and T or a dif- 

 ference of nearly 7 per cent of their mean value. Moreover, 

 the two extremes are neither exceptionally high nor excep- 

 tionally low but the other crystals are pretty evenly distributed 

 between them. It seems certain, therefore, that the crystal 

 constants of a-pyrrhotite vary considerably in accord with the 

 variable composition of the mineral. 



A further study of Table II shows that while the crystals of 

 a given preparation show a considerable difference in the con- 

 stant, c, yet those of preparation 3 are uniformly high, those 

 of preparation 4 generally low, and those of preparation 1 and 

 2 intermediate. Unfortunately it was not found possible to 

 determine the relation between the value of c and the chemi- 

 cal composition, as, even were any of the crystals large enough 

 for an analysis, it is probable that they are built up of succes- 

 sive shells of different composition. 



No consistent variation in the angle (100) (110) was recog- 

 nized but it probably varies with g. The measurements were 

 usually not good but the average value of 59° 55' is probably 

 within 15' of the true value. Computing the axial ratios of 

 a-pyrrhotite for the limiting values given by crystals M and T 

 we have : a : I : o 0*5793 : 1 : 0*9267 to 0*5793 : 1 : 0*9927. It 

 is probable that the limits are considerably greater than this. 

 Table XI lists the observed faces and the interfacial angles as 

 computed from the above values of the axial ratios. 



The symmetry and crystal constants of a-pyrrhotite. — 

 The evidence that a-pyrrhotite is orthorhombic in symmetry 

 is good, although it may not be conclusive. The angles in the 

 prism zone were not proven to differ from sixty degrees, but 

 the habit of the crystals consistently indicates orthorhombic 

 rather than hexagonal symmetry. A very few crystals 

 resembled hexagonal tablets. Nearly all of the crystals of the 

 first two preparations were developed much like the crystal 

 shown in fig. 17, although small domes and pyramids were 

 found on many of the measured crystals. The habit and 



