A Contribution to the Study of the Diamond Made. 
161 
above is the spread to which the diamond made tends to conform, and 
therefore that the standard made is not rightly to be regarded as consisting 
of two halves of the standard octahedron. More than that, a made which is 
equivalent to two halves of an octahedron is as much abnormal as a made 
of four times its spread.* 
It is a curious circumstance that although dodecahedral stones are 
common at Bultfontein, where quite a half of the yield is prevailingly of 
this type, and that Wesselton is a mine of stones inclining to the octahedral, 
yet more glassy and octahedral macles are found at the former place than 
at the latter. In fact, of the whole 169 Bultfontein macies dealt with in 
Tables 1 and 2, 105 (= 62 per cent.) were of the octahedral type, whereas 
only about a half of the Wesselton macles were so. Bultfontein octahedral 
macles, however, average smaller than Wesselton ones do — at any rate 
among diamonds exceeding one-tenth of a carat each. 
Ta:6LE 4. — Number of Dodecahedral Macles of Given Spread. 
E/T under 1. 
1 to 1-99. 
2 to 2-99. 
3 to 3-99. 
4< to 4-99. 
5 upwards. 
Koffyfontein 
2 
11 
1 
Jagersfonteiti 
4 
15 
Dutoitspan . 
8 
7 
1 
Bultfontein 
2 
27 
26 
8 
1 
Wesselton . 
15 
41 
10 
Totals . 
2 
56 
100 
20 
1 
The greatest spread-ratio hitherto observed by me is 6'37 (E = 51 mm., 
T = 0*8 mm.) on a glassy made from Bultfontein. Spread-ratios of 5 on 
Bultfontein macles are not uncommon. 
A Note on the Internal Structure of Diamond. 
The grain which appears in herring-bone pattern on a broken made is 
sometimes shown in straight pattern on a broken simple diamond crystal. 
This will be when the fracture happens to lie at right angles to an octahedron 
edge, i. e. parallel to a dodecahedral plane of symmetry. This grain is 
parallel to the plane of a continuous line of edges of the hexakisoctahedron ; 
it is parallel to a plane joining any two opposite edges of the cube, or what 
is the same thing, to a plane joining any two opposite shorter diagonals of 
* I am unable to say how this result compares with the average twin of spinel. 
Lewis (p. 467) observes that the twin of spinel often "acquires a more or less strongly- 
marked tabular habit by the disproportionate development of the faces parallel to the 
combination plane." An excellent little twin of Burma spinel in my possession has a 
spread-ratio of 1"40. 
