20 SCIENCE PROGRESS 



PHYSICS. By James Rice, M.A., University, Liverpool. 

 The Phys. Rev. of December 191 7 contains a paper by A. W. 

 Hull on a new method of X-ray crystal analysis. A brief 

 description had been published in the Phys. Rev. for January 

 191 7. The methods already developed by Prof. Laue, and 

 by Prof. Bragg in co-operation with his son are applicable 

 only to individual crystals of appreciable size reasonably 

 free from twinning and distortion and sufficiently developed 

 to allow the determination of the direction of their axes. 

 Prof. Hull's method permits investigation in cases where such 

 crystals cannot be found in nature or in ordinary technical 

 products and can only be grown with difficulty and much 

 expenditure of time. The amount of crystal material re- 

 quired is no more than 0*005 c - c -> an d admixture with foreign 

 material, provided it is amorphous or of known crystalline 

 structure, is not fatal to the success of the method. It con- 

 sists in sending a narrow beam of monochromatic X-rays 

 through a disordered mass of small crystals reduced to a 

 finely divided form and packed in a thin-walled tube of glass 

 which, where great precision is required, can be maintained 

 in continuous rotation. The X-rays diffracted by the various 

 crystal planes fall on a photographic plate, producing a definite 

 diffraction pattern. Owing to the disordered arrangement of 

 the crystals, assisted by frequent stirring (and rotation where 

 necessary), the average orientation of the little crystals during 

 a long exposure is a random one. A certain little group of the 

 crystals will at any given instant have their 100 planes (say) 

 at the proper angle with the X-ray stream to reflect the par- 

 ticular wave-length used ; and similarly there will be small 

 groups with other planes suitably oriented. In fact for a long 

 exposure there will be on the average an equal chance for 

 each plane of the crystal making its presence felt in the diffrac- 

 tion pattern. This statement is, of course, subject to the 

 limitation that crystal planes whose distance apart is less 

 than half the wave-length used, can only produce a diffracted 

 wave with very small amplitude. The scattered energy is 

 divided among a finite number of planes, each of which pro- 

 duces on the plate a linear image of the source. From the 

 positions of these images the crystal structure can be deduced 

 by a method similar to that employed by the Braggs, with 

 this difference. In the Bragg method reflections from three 



