It was observed by Pinch that zinc in the films formed by condensation of vapour on a polished copper surface, diffused rapidly into the surface amorphous layer of the substrate at room temperature. On the other hand, Crater reported that such a diffusion phenomenon was not observed. In order to ascertain whether such diffusion takes place or not, we set two copper pieces, one with polished surface and the other with etched one, each parallel to the (110) plane of a single crystal, in the vacuum of an . electron diffraction apparatus and condensed zinc vapour on both surfaces simultaneously and observed the patterns reflected from them. The thieknesa of 'the films varied from 20 to 100A°. We could not observe any diffusion phenomenon, such as reported by Finch, in every' case and on' both surfaces. When these zinc films were heated in the vacuum of the apparatds, i. e. about 10-3 mm Hg, at 200-600°C, they were oxidized and became to zinc oxide. In these processes there formed a thin layer of brass at the boundary of the film and the substrate. This was ascertained by etching off the surface layer of zinc oxide. When zinc vapour was condensed on copper substrates which were maintained at 150-200°C, diffusion took place rapidly and twins of the spinel type of brass Were formed on the surface of copper single crystals. While zinc oxide, formed by oxidation of thin film gave normal pattern of ZnO, the one formed by oxidation of thick film gave somewhat different one which we call“ZnO pattern”for convenience. This pattern can be explained as being due to a ring fibers arrangement of zinc oxide crystals in, which fiber axes  are parallel to, and distributed uniformly in the surface of the substrate. Ψ-ZnO, reported by Finch, can also jbe explained as a ring fibre of zinc oxide crystals in which fibre axes  are distributed uniformly in the plane of the surface.
The lattice constants of two artificially prepared jewels were determined using both the ordinary Debye-Scherrer and the back reflection methods and the following values were obtained : α (kx) c (kx) temperature white corundum 4.75320 12.95198 31° ruby 4.75683 12.98843 31° The latter contains several percent of chromic oxide. The direction of crystal growth and the defects in crystals of a number of these jewels were investigated by the divergent-beam back reflection method. The specimens are of the form of an elongated drop and split into two parts by light hammering, the fracture surfaces being parallel to the longer axis of the drops (i. e, the direction of crystal growth) . The hexagonal axes of individual domains within a drop lie in the plane of fracture. The distribution of these axes within the plane, is somewhat scattered and in most cases it lies nearly perpendicular to the direction of the longer axes in the case of white corundum whereas it prefers the direction nearly parallel to the direction of growth in the case of ruby. White corundum usually exhibits perfect structure but frequently consists of minute crystals, several millimeteres in diameter intersecting in few minutes, thus forms a mosaic structure. Ruby always shows imperfect structure, and its mosaic structure is usually built up by rows of crystals having. stringlet forms.
A general theory of the propagation of orders, an extension of Zernike's theory on the same problem, is developed to obtain the intensity formula of scattered x-ray by a partially ordered crystal. The intensity of the diffuse scattering, which was denoted by J (b) in Part i of this paper, can now be expresed in the form : J (b) =Σ _??_ where λi are the eigenvalues of a matrix B whose elements are determined from the molecular interaction between a lattice point and its first neighbours and from the path differences of the x-ray wave scattered at them. mii are the diagonal elements of another matrix M. As an illustration of the method, the diffuse scattering from a face centered lattice consisting of diatomic molecules, in which molecular axes are order-disorderly arranged among four directions (  and its equivalents), is considered. A theory of the phase transition of this lattice is also given. This example is presumably applicable to the low temperature form of N2 (ordered state) and the high temperature form of NaCNN or KCN (disordered state) . In the last section a general discussion of the relation between the diffuse scattering and the phase transition is developed; it is pointed out that the diffuse scattering should show a remarkable temperature dependence near the transition point.
For the x-ray diffuse scattering from cubic tetranitromethane crystals (α=7.08 A, Z=2, T3α-I43m), in which the molecules are in a statistical arrangement, here is given a tentative explanation' taking into account the short-range order. While actually the molecules possess the symmetry of S4-4, it was assumed for the sake of simplicity that the molecules had the symmetry of D2α-42m and took their molecular axes, the four fold rotation reflection ones, in equal proportions, to each of the three directions of the principalX, YandZaxes, the C-N bonds being parallel to the body-diagonals. It was observed experimentally that the diffuse maxima appear around the reciprocal lattice points whose indices areh+k+l=2n. This fact could be explained by an intensity formula for the diffuse scattering with a certain value of the parameter α, which measures the probability that for a given orientation of one molecule, its neighbour takes one of the three orientations.