of the Crystallography Laic of Valency Volumes. 525 



OAGrB by sliding, and therefore the lattice would not be 

 hexagonal. To make the system hexagonal, the repre- 

 sentative points on the plane KLMN would have to be 

 at the middle points of the sides and the centre of the 

 rhomb. The middle plane must, however, be different in 

 some way from the end planes. This might, of course, be 

 brought about by the oxygen atoms. 



Without entering further into a detailed analysis of the 

 conditions which will satisfy the symmetry and explain 

 the intensities of the various orders of the spectra, we at this 

 stage merely wish to indicate the probability of the sulphur 

 atoms being located at the corners and centres of the faces 

 of the unit rhomb. This would give a pseudo-hexagonal 

 structure for the sulphur atoms in planes parallel to the 

 (001) face. 



The metals are also probably arranged in hexagons. 



Summary. 



An account is given of an X-ray investigation into the 

 crystalline structure of the Isomorphous Alkali Sulphates. It 

 is shown that for each of these crystals there are four molecules 

 in the elementary cell. It follows from this that the dimen- 

 sions of the crystal units of ammonium sulphate and rubidium 

 sulphate are almost identical. This furnishes a critical test 

 of the Valency Volume Theory of Pope and Barlow, which 

 is shown to break down in this case. Further evidence in 

 support of this contention is provided by the agreement 

 between the observed and calculated values of the angles of re- 

 flexion from these crystals. Although the complete structure 

 of the series is not solved, the probability of the hexagonal 

 arrangement of: the metal and sulphur atoms is pointed out. 



Most of the crystals used in this investigation were lent b/ 

 Dr. A. E. H. Tutton, F.R.S. They had been prepared with 

 optically worked faces and used by him for his classical work 

 on the thermal expansion of crystals. We have to thank him 

 for his very great kindness in this matter, and also for the 

 keen interest which he took in the work. 



It is also a pleasure to record our thanks to Professor 

 W. H. Bragg for the valuable assistance and advice which 

 he was ever ready to give during the progress of the 

 measurements. 



University College, London, 

 April 1916. 



