METEORIC AND ARTIFICIAL NICKEL-IRON ALLOYS. 
25 
The true angles between each of the planes 00, 11, &c., and the plane faces of the 
core were determined by means of a hand goniometer. The numbers on the cylindrical 
surface in the diagram represent these angles. 
Within the limits of observation the plane 00 was at right angles to the plane 
faces of the core. This accidental circumstance has the effect of simplifying slightly 
the calculation of the angles between the various planes. For, taking the plane of 
the circular face of the core as the plane of xy, the plane 00 may be taken as the 
plane of yz % Attention being paid to signs, the intercepts of each of the other three 
planes on the axes of x, y and 2 are then in the ratio 
tan 0:1: sin 6 tan (p, 
where 6 is the angle between the trace of 00 and of one of the other planes on the 
circular face and <j> is the real angle between the latter plane and the circular face. 
Hence, using the data given on the diagram, the direction-cosines of the planes 11, 
22 and 33 can be calculated. 
In this way the angles between the planes of the layers were found to be :— 
001 = 70°, 0 18 = 77°, 023 = 71°, 0 3O = 71°, 0 O2 = 73°, 0 13 = 61°. 
Hence it is seen that, with the exception of 11, the planes are inclined to one 
another in practically the same way as the faces of a regular octahedron, the interfacial 
angle for the latter being 70° 32' approximately. 
The markings on the plane faces of the cylindrical core, from which the measure¬ 
ments were taken, are shown in the accompanying photographs, figs. 2 and 3 (Plate 1), 
which are slightly less than natural size (actual diameter of core about 37 centims.). 
The relative position of the markings in space would be reproduced by rotating either 
of the photographs through 180° about the line 00 and then placing it behind, and 
about 1 ’8 centims. away from, the other photograph. The markings on the cylindrical 
part of the surface of the core are very distinct, but difficult to reproduce photo¬ 
graphically owing to the curvature. On this part of the surface, the layers parallel 
to 11, which are not very clearly seen on the flat ends (see photographs), are very 
numerous, owing to the acuteness of the angle (29°) which these layers make with 
the flat ends, and are as conspicuous as the layers running in the other directions. 
Since the layers parallel to 00 run practically perpendicular to the plane ends of 
the core, the true thicknesses of these layers are represented in the photographs. On 
the other hand, the thicknesses of the layers parallel to 11 are about one-half 
(= sin 29°) of the apparent thicknesses shown in the photographs. The thickness of 
a layer parallel to 22 is about three-quarters (= sin 50°) of the apparent thickness, 
whilst the layers parallel to 33 are about 5 per cent, thinner than they appear in 
the photographs. From these data it will be seen that the distribution of the layers 
VOL. CCVIII.—A. 
E 
