138 Mr. A. E. Tutton [May 2, 



is an irreproachable one, the face reflecting it being truly plane. Its 

 narrow part is capable of accurate adjustment to a vertical cross- wire. 

 On rotating the crystal so as to bring the image from the next face, 

 of the particular zone which has been previously adjusted, into the 

 field, you observe that this also is a very good image, but much 

 weaker than the other. This is simply because it is derived from a 

 much smaller face, but the face is quite plane. The next image you 

 see is a bad one, being not only a multiple image, but distorted. 

 Such an image would be quite useless for our purpose. Thus, on 

 rotating round the whole zone — for it is one of the geometrical pro- 

 perties of crystals that the faces lie in zones — we find that some 

 images are good and some are bad. In the case of very small crys- 

 tals it generally happens that the greater number of images afforded 

 by the faces are either one thing or the other, and crystals must be 

 selected yieldingr the maximum of good images. 



The first resuic of importance that has been brought to light is 

 that, with regard to each series of salts, there is an interesting rela- 

 tionship between the general exterior character — or habit, as the 

 crystallographer terms it — of any triplet of salts containing potas- 

 sium, rubidium and cassium respectively. 



We will see on the screen, for instance, the configurations of the 

 three double salts, potassium zinc sulphate, rubidium zinc sulphate and 

 caesium zinc sulphate. The same planes are present in all, but very 

 differently developed. In the potassium salt the shape is determined 

 by the large development of the prism zone (p faces), and the large 

 flat end faces of the basal plane, c. The caesium salt, on the other 

 hand, exhibits a prismatic habit formed by the faces of the clino- 

 dome, 5, and the basal plane c is reduced to a strip. Intermediate 

 between these two types comes the rubidium salt, and this is the case 

 with the rubidium salt of every one of the sixteen triplets examined. 



Hence the habit clearly follows the order of the atomic weights 

 of the alkali metals. 



The next general result arrived at is, that small angular differ- 

 ences between the faces have been established, but they rarely amount 

 to a degree in magnitude. And what is of even more interest is, 

 that the faces of every rubidium salt are inclined to each other at 

 intermediate angles to those of the potassium and caesium salts. 



This important fact may be illustrated by the vacuum-tube model 

 on the table, which has been specially constructed to emphasise the 

 point. The three crystal outlines represent sections through one of 

 the principal planes of the alkaline sulphates or selenates. The 

 outer one represents the outline of caesium sulphate or selenate, 

 the middle one that of the rubidium salt, and the inner one a 

 section of the potassium salt. The intermediate inclination of the 

 domal faces will be clearly apparent. This is only one of some sixty 

 angles which have been examined, and all show the same beautiful 

 progression in the order of the atomic weights. 



