3 8o SCIENCE PROGRESS 



K. Ges. Wiss. Gottingen, Math. Phys. Klasse, 1916, Phys. Zeit. 

 18, 291, 191 7 ; cf. Engineering, 104, 17, 594, 191 7), consists in 

 allowing a narrow beam of X-rays to traverse an aggregate 

 of small crystals, the resulting diffraction pattern being photo- 

 graphed. The disposition of the crystals is assumed to be 

 thoroughly irregular so that at any instant a certain number 

 of crystals are giving reflections from, say, a (100) face, others 

 from a (in) face, and so on. The structure can then be eluci- 

 dated by analysis of the diffraction pattern, due attention being 

 paid to the intensities. This method, therefore, does away 

 with the necessity of using well-developed crystals, as it is 

 applicable to any crystalline aggregate provided the number 

 of crystals is sufficiently great for some of them to be so 

 orientated as to give suitable reflections. The authors have 

 investigated the structures of graphite and amorphous 

 carbon, and conclude that the latter is merely very finely 

 divided graphite. The elementary parallelopiped in the case 

 of graphite is a rhomboid, the carbon atoms being arranged 

 so that they occupy the alternate corners of regular hexagons. 

 It is suggested that graphite is the basis of the aromatic com- 

 pounds and diamond of the aliphatic, while the apparent 

 trivalency of carbon which occurs in such compounds as tri- 

 phenylmethyl is ingeniously explained on the basis of the 

 graphite structure. 



In a further paper (Phys. Zeit. 18, 481, 191 7, Jour. Chem. 

 Soc. 112, ii. 574, 191 7) the same authors give the structure of 

 tungsten, which is found to be the centred cube, while Scherrer 

 (Phys. Zeit. 19, 23, 191 8; Jour. Chem. Soc. 114, ii. 113, 191 8) 

 ascribes the face-centred cube to the structural unit of 

 aluminium. It is pointed out that although the edge of the 

 elementary cube has the same dimension (4*07 x io~ 8 cm.) for 

 aluminium and gold, these metals do not form a continuous 

 series of solid solutions. By the same method, using material 

 of diameter 0*5— ro/t, S. Kyropoulos (Zeit. anorg. Chem. 99, 

 197, 191 7) &as investigated the structure of quartz, cristobalite, 

 and amorphous silica. The last shows no evidence of regular 

 internal structure, but after calcination the interference bands 

 characteristic of cristobalite appear. This method is also used 

 by J. Olie and A. J. Byl (Proc. Akad. Wetensch. Amsterdam, 

 19, 920, 191 7) in an investigation of the structures of diamond 

 and graphite. The mathematical basis of the method is dis- 



