The problem of assessing the numerous new drugs offered to clinicians is continuously increasing. It is not always possible, even if desirable, to mount a substantial and definitive therapeutic trial for each new drug in a single department. In these circumstances close observation of a few patients in several centres may provide data adequate to justify more general distribution of a new drug, with minimal diversion of workers from their other tasks. The animal pharmacology of a new adrenergic beta-receptor blocking agent. pronethalol, has been reported by Black and Stephenson (1962), and some aspects of the human pharmacology by Dornhorst and Robinson (1962 Recording of Data.-Patients were asked to take glyceryl trinitrate as usual for the attacks of pain, but not prophylactically, and to keep a record of the attacks of pain and number of tablets taken. They were given a record sheet for each week, each day being divided into four periods starting at the time pronethalol (or placebo) was taken. They were asked to fill in their sheet at the end of each period and ,to record the time of onset of any attacks of angina, their duration; to assess severity as mild, moderate, or severe; to note the number of glyceryl trinitrate tablets consumed; and to record any comments they desired. Patients were supplied with a known number of glyceryl trinitrate tablets, and this number was recorded on their sheet and checked with the number remaining in the bottle when they attended at the next visit. Clinicians saw the same patients weekly and recorded their progress without reference to previous assessment sheets. Patients were asked standard questions, and their subjective impressions and any complaints of side-effects were noted.Assessment.-See results. Selection of Patients.-The case-notes of over 100 patients with a diagnosis of angina were examined and only 21 of these subjects were considered to be suitable for this study of a new drug-that is, those having at least two attacks of angina regularly each week, and without other disease or complications that might make the trial of a new drug inadvisable. The pain had to be of characteristic nature, site, and radiation, brought on by exertion, relieved by rest and/or glyceryl trinitrate, and lasting one to three minutes. Of these patients, nine had to be withdrawn during the early stage before the double-blind study began, for the following reasons: (1) hypersensitivity in two .patients (measles-like rash in 2nd week in one, and angioneurotic oedema, "collapse " 2nd day in the other); (2) diarrhoea in one patient (refused to continue)-(3) depression in one patient on a high dose-he then went abroad; (4) angina subsided, apparently spontaneously in one patient ; (5) fatal myocardial infarct in one patient; (6) one patient objected to having his tablets " mucked about "; and (7) two defaulted.