Abstract
In the context of diffraction of Maxwellian (thermal) atomic beams by solid surfaces, the usual assumption that the angular position of the maximum in a diffracted beam corresponds to the diffraction angle of atoms with the most probable de Broglie wavelength is examined, and compared with other possible criteria and with the correct result. It is concluded that, although this criterion may be the best simple one available, it is certainly bad in some situations; the reasons why, and the conditions under which, it is expected to be good are discussed. Also, it is shown that considerable care must be taken when shapes of diffracted beams and when angular positions of their maxima are calculated, because certain physical effects (which are always present) may change these shapes and positions in unexpected ways. The theory is compared with two sets of relatively modern experimental data, one set for which the fit is good, and another set for which a fit is impossible.