ANALYSIS OF CRYSTAL-STRUCTURE BY X-RAYS 399 



2-90 A.U., and at 1,425° C. to 2*93 A.U., which agrees very 

 well with the known coefficient of expansion of iron. 



With regard to the space-lattice of iron in hardened steels, 

 it was found that the 7-iron lattice of austenite steels is en- 

 larged by the dissolved carbon, while the a-iron lattice in 

 martensite also seems to be influenced by the carbon present. 

 If these results are to be relied on (the evidence is not exception- 

 ally strong), they indicate that in martensite the carbon, as is 

 generally supposed, is so intimately mixed with the iron 

 that it may be considered to be a real atomic solid-solution. 

 The breadth of the three very faint and diffuse lines in the 

 martensite photographs seems to show that the individual 

 pieces of homogeneous lattice in the steel are, on an average, 

 only 20 A.U. or so across, which means that each of them 

 contains only a few hundred atoms. Martensite can thus be 

 considered to be on the verge of being totally amorphous. 



An investigation has also been conducted with a view to 

 determining the true crystalline form of cementite. Three 

 different powders gave identical photographs, from which fact 

 it may be concluded that the cementite in all three alloys 

 used is probably composed of a unit chemical individual. 

 With the aid of a Laue photograph, the crystals were assigned 

 to the orthorhombic system. In each unit cell (compare 

 acenaphthene above) there are four molecules of FcgC. 



While on the subject of the X-ray spectra of metals, 

 mention must be made of the graphical method devised by 

 Hull and Davey {Phys. Rev., vol. xvii, May 1 92 1 ) for determining 

 the crystal-structure of tetragonal and hexagonal crystals. 

 In such crystals there is only one unknown parameter which is 

 really essential to the symmetry, and that is the " axial ratio " 

 [c : a). It is the ratio of the length of the vertical axis, c 

 (the tetrad axis in the tetragonal system, the hexad axis in 

 the hexagonal system), to the length of the horizontal axis, a 

 (a dyad axis in both systems). The essence of the new method 

 is to plot the logarithms of the theoretical spacings of the 

 different planes as functions of the axial ratio, and to the 

 graph so obtained to fit the logarithms of the observed spacings. 

 The logarithms of the spacings are plotted all on the same 

 logarithmic scale, so that they may be compared directly 

 with experimental values without regard to the absolute 

 lengths of the axes. By such means Hull has arrived at the 

 crystalline structure of most of the common metals. His 

 paper on the " X-ray Analysis of Thirteen Common Metals " 

 {loc. cii.) is a very important contribution to the' science. 



In his work to date on organic crystals, Bragg, in order to 

 demonstrate the physical existence of the benzene and other 

 ring-formations, has confined himself to aromatic compounds; 



