One of O'Dwyer's assistants at this time at the Foundling Asylum tells of the amount of time the doctor gave to the {334} study of the problem involved in these difficulties and of his ultimate success therein. Putty was moulded in various ways on tubes, which were inserted in specimen larynxes, and plaster casts were taken, with the idea of determining just the form of tube which would so exactly fit the average normal larynx as to be retained without undue pressure, yet at the same time keep the false membrane from occluding the respiratory passages and furnish as much breathing space as possible. Finally Dr. O'Dwyer decided that the best form of tube for all purposes would be one with a collar, or sort of flaring lip at the top, which was to rest on the vocal cord, with, moreover, a spindle-shaped enlargement of the middle portion of the tube, which lay below the vocal cords, fitting more or less closely to the shape of the trachea. To avoid the pressure and ulceration at the base of the epiglottis--a very sensitive and tender portion of the laryngeal tissues--a backward curve was given to the upper portion of the tube. On the other hand, the lower end, which rests within the cricoid ring and which was likely to be forced against the mucous membrane of the trachea occasionally, was somewhat thickened to avoid the friction and leverage that might be exerted if there were any free-play allowed. At the same time the lower end of the tube was thoroughly rounded off. Thus Dr. O'Dwyer, realizing all the difficulties of this new method of treatment, solved them, as experience proved that the tubes could be made of still smaller calibre than had been hitherto supposed and yet be efficient in relieving respiratory dyspnoea. Experience also proved that the metal tubes at first used had a number of serious disadvantages. They were heavier than those which could be made of hard rubber in the same size and shape, while the metal tubes besides had a tendency to encourage the deposition and {335} incrustation on their surfaces of calcium salts. These incrustations, roughening the surface of the tube, increased its tendency to produce pressure ulceration, as well as added to the difficulty of its removal, and consequently to the liability of producing laceration of tissues after convalescence had been established. Accordingly tubes were made of hard rubber, which could be allowed to remain in the larynx almost for an indefinite period without any inconvenience. While at first intubation was looked upon as a merely temporary expedient, clinical experience showed that sometimes in neurotic patients it was necessary to let the tube remain in the throat for several weeks or even months.