bound electrons equal in theory and experiment. The same practice is used for x-ray and y-ray Compton-profile normalization but not in the highenergy electron-impact method. For the other two theories the theoretical curves have been arbitrarily placed on the logarithmic scale in the best way to emphasize shape discrepancies 0The discrepancy between theory and experiment in the earlier (e, 2e) measurements 5 can be explained in part. Firstly, the theory used was slightly in error, since the sum over vibrational states was not carried out although these states were summed over in the experiment. Secondly^, there was a systematic error of 6% in the calibration of q.The small discrepancy between theory and experiment in the Compton-profile measurements gives rise to a more serious disagreement between the calculated and experimentally derived momentum distributions. This discrepancy could quite possibly be due to failure to satisfy binaryencounter conditions in the experiments. If this is indeed the case then, as in (e,2e), the measured Compton profiles will depend on the overlap of the initial target and final ion wave functions rather than the single-particle momentum densities. Of course at sufficiently high energy, closure over final states reduces the theory to dependence only on initial states.We would like to acknowledge very helpful conversations with Dr. R. A" Bonham, Dr c W. A.The plasma-(Stark-) broadened Balmer-line profiles of hydrogen measured during recent experiments 1 * 2 exhibit much less structure in their central regions than predicted by the most comprehensive theories. 3 " 7 This difference is quite serious since the plasma-broadened Balmer lines have found many applications as a convenient pre-The plasma-broadened R a and Hg spectral-line profiles of hydrogen were calculated retaining the time ordering in the S matrices of the width-shift operator by diagonalizing with the 0(4) group. Significant changes occur: The H a peak is decreased 15%, the halfwidth is increased 25%, and the H 0 relative dip is decreased 23% without significant alteration of the maximum intensity of half-width. Agreement with experiment is improved appreciably.
785