Experiments on the cough reflex (Larsell and Burget, 1924; Widdicombe, 1954a and b) have shown that two afferent nervous mechanisms are present. One has receptors lying in the trachea and is excited by mechanical stimuli, whereas the other extends throughout the tracheobronchial tree and is sensitive chiefly to irritant gases; the two types of cough stimulus give somewhat different motor responses. The central nervous pathways for coughing have not, however, been investigated, and little is known of the interactions between the respiratory centre, the cough reflexes, and other respiratory reflexes. The experiments described in this paper were carried out to see whether drugs that depress the cough reflex, and that are believed to act centrally, have quantitatively different effects on these reflexes. At the same time a new morphine derivative, pholcodine (2'-morpholinoethyl ether of morphine), became available, and, since it had not been directly tested against coughing, its action was compared with those of morphine and codeine. METHODS Tracheal cannulae were inserted into 28 cats, anaesthetized with pentobarbitone sodium, 32 mg./kg. intraperit oneally. The cats were enclosed in the body plethysmograph described by Dawes, Mott, and Widdicombe (1951), from which respiratory volume changes were recorded on a smoked drum, using a float recorder. The method permitted measurement of tidal volume and of changes in expiratory volume (functional residual volume). Blood pressure was measured by a Hg manometer from a carotid artery. Intratracheal pressure was measured by a tambour, and intraoesophageal pressure by an electrical condenser-manometer and displayed on a cathode-ray oscilloscope.Coughing.-Coughing was induced in two ways: firstly, by passing a polythene catheter down the trachea until it touched the carina, after which it was immediately withdrawn; and secondly, by blowing air containing SO2 into the trachea. The SO was generated by adding 50 ml. of N-HCl to 5 g. of sodium thiosulphate in a conical flask; to excite coughing, a measured quantity of air (0.5-1 ml.) was bubbled through the mixture into the tracheal cannula at the beginning of an inspiration. Both types of stimulus (mechanical and chemical) were applied once only to produce a cough, and they were repeated at intervals of 5-10 min. With the mechanical stimulus control experiments showed that refractoriness did not develop for at least an hour; with SO2 the respiratory responses sometimes decreased slightly during that period, but never by more than 25%.The latter reflex became refractory if large amounts of SO2 were administered.The cough in response to a mechanical stimulus was expressed as the ratio of the volume of the initial expiratory effort after the stimulus to the average of ten preceding tidal volumes. The percentage changes in this ratio after administration of drugs were calculated. SO2 did not always cause an initial expiratory effort, but one or more deep gasps were invariably seen; the depth of the first inspiratory effort was ther...