Abstract
The analysis of Part I is repeated for the second boundary value problem in which the normal derivative of the wave function vanishes on the elements of the grating; the existence of anomalies is again inferred. The behavior of the spectral intensity near an anomaly for a grating of small elliptic cylinders is considered and found to be much different from the case studied in Part I.The infinite reflection grating is discussed. Weak anomalies are predicted in the spectra for the case in which the wave function vanishes on the grating; strong (Wood) anomalies are found in the spectra for the second problem.The finite transmission grating is treated very approximately. Anomalous behavior, dependent now on the number of elements, is inferred.

This publication has 6 references indexed in Scilit: