Having spent several years studying various phases of the pyelitis problem, it has become very clear to me that a correlation of the clinical, pathologic and bacteriologic findings must be established. Strange as it may seem, this really has not been accomplished. A short survey of the difficulties in our way may prove profitable.
There is no correlation between the severity of symptoms and the pathologic findings. The former cannot be used at present as a means of localizing the infection. The same symptoms may be associated with cortical abscesses of the kidney, with infection of the pelvis, ureters or bladder, singly or combined. In spite of the severe clinical course there are, at best, only few postmortem examinations, in which the pelvis alone is found to be involved. Accordingly, it is impossible to make a diagnosis from the clinical picture alone, and if we had to rely on the