Petrol is a highly inflammable fluid which can be ignited easily and whose vapour forms with air a highly explosive mixture.

There are, however, possibilities of giving a certain amount of protection against such fire risks created by damage to petrol tanks.

The three main problems which have to be dealt with are:—

1. (i) To exclude an ignition which may be caused by bullets hitting a petrol tank;

2. (ii) To seal leaks caused by bullet holes;

3. (iii) To prevent fire after a crash.

These problems are by no means new. A large amount of research and experimental work has been devoted to their solution.

Before reading this paper, I should like to make a brief explanation of my reason for changing the published title.

The original intention was to give a survey of civil flying boats in general, but after due consideration the subject appeared to me to cover far too much ground, unless it was restricted to merely a collection of already published data of past and present aircraft of this type.

Whilst a paper of this nature would, undoubtedly, prove to be of considerable interest, yet, personally, I feel more competent to review the three types of flying boats designed purely for civil use, which have come directly under my supervision during the past eight years.

The Townend Ring has been the subject of a large amount of experimental research and its general characteristics are now well understood. The following paper is the first theoretical treatment of the ring which has appeared. It was originally presented as a thesis to the University of Delft by Capt. G. Otten and published in Holland in November, 1932. The author has made the following abbreviated translation of it into English in collaboration with Dr. H. C. H. Townend, who also suggested a few modifications.

The problem of keeping pilots and observers warm when flying at high altitudes was brought into prominence during the late war. At a height of 20,000 feet the normal temperature is approximately − 25°C. or “45° of frost,” Whilst in winter at an altitude of 30,000 feet temperatures of the order of − 5o°C. or “90° of frost” are frequently encountered. When flying at probably 100 miles per hour in temperatures of this order, all ordinary means of keeping the body warm are quite inadequate and no padding or lagging capable of being worn as a suit will keep the body warm for more than a few minutes.

The work described in this paper was done during the war, at the suggestion of the aircraft engine firms, as a result of difficulties with lubricating oil systems at high altitudes. Failure of oil pressure, seizure of bearings, or excessive loss of oil sometimes occurred and these defects were attributed to air in the lubricating oil, this air being greatly increased in volume at high altitudes because of reduced atmospheric pressure.

The behaviour of air-oil mixtures therefore was studied to obtain fundamental information which could be applied to engine lubrication at high altitudes. The present paper deals with that part of the research which was done on a complete lubrication system under simulated altitude conditions.