Why measure astrophysical X-ray spectra, and reply by I.J.D. Craig and J.E. Brown

Reference is made to Craig and Brown (Nature 264:340 (1976)) who advised caution against some of the limitations of X-ray astronomy, and emphasised a need to take theoretical considerations into account before planning expensive experimental work. It is stated that whilst their remarks must be taken seriously by those involved in space experiments, their statements represent only one extreme viewpoint on the problem. Their main argument concerns the equations required in order to carry out an analysis based on emission measurements, but it is pointed out that emission measure analysis is only a small, although important, part of the full capability of spectroscopic diagnostics, and at present it is possible to combine it with spatial and time resolution of the source. It is basic to all interpretational physics to find the simplest model consistent with observations, but to imply, as Craig and Brown do, that all possible models, however complex, are equally valid and worthy of consideration, would seriously impede progress in any branch of physics to which it is applied. An example is given of the histogram of emission measured against temperature. Some remarks are made in defence of high wavelength resolution spectroscopy. Craig and Brown suggest that crude detectors are more than adequate but this is incorrect - when a future generation of cosmic X-ray spectrometers is available to resolve and separate lines a new dimension of diagnostics will become possible and dramatic advances in our understanding of recently discovered objects can be anticipated. It is emphasised that space experiments nowadays are not designed with regard to theory. A reply from Craig and Brown is appended. (U.K.)