328 Proceedings of the Royal Physical Society. 



coined the word Gnomonogram. The methods of construc- 

 tion of gnomonograms lead by easy gradations up to the 

 more complex methods involved in making Stereograms, and 

 therefore its consideration may profitably be noticed here in 

 some detail. 



16. To begin, we will employ concrete methods as far as 

 possible, as before. Instead of a small, simple crystal, we 

 will now take a series of crystal models, including all the 

 simpler forms, such as can be got by using Jordan's Crystal 

 Nets already referred to. We may as well begin with the 

 Orthorhombic System, as before. It is as well to have each 

 crystal model mounted on some kind of stand — a cork coated 

 on the top with seccotine does w^ell for the purpose. Now 

 mark the position of the three axes, which can be done by 

 small pins stuck into the models. It is advisable also to 

 employ some kind of symbol for each kind of face, using, 

 preferably, the symbols accepted by Dana. 



17. We may now place each simple form in succession at 

 the centre of a hemispherical dome of glass as before, 

 project its poles on to the glass, marking those of each form 

 with its distinctive letter or symbol (not with the indices 

 at present), and doing this with the whole series, as before 

 advised. Then all the various zones connecting any three 

 of these along a great circle can be marked in upon the 

 hemisphere. 



18. When this has been completed, the hemisphere is to 

 be fixed by any suitable means, and then a large and flat 

 sheet of glass is fixed so as to lean against the hemisphere, 

 which it may touch at any point, but preferably at the pole 

 of the UNIT PYKAMID of Miller's notation (111). If, now, the 

 eye be moved to various positions in succession, so that it 

 and a point on the glass hemisphere are in a line with the 

 centre of the hemisphere (which henceforth will be denoted 

 by o), there will be found no difficulty in marking the places 

 on the sheet of glass — in other words, in projecting the 

 points in question from a spherical surface to a plane one. 

 It will soon be evident that the zones (which mark GREAT 

 CIRCLES on the sphere) project on the flat surface into straight 

 lines, so that the whole of the projection now under con- 



