. 1. Under no condition, likely to occur in working, must it be possible for the pressure in any part of the apparatus to exceed that necessary to support a column of water 100 inches in height.

2. When the apparatus is first charged, in no case must the air in the generating chamber and receiver exceed one-fifth of the capacity of the apparatus.

3. On shutting off the outlet cock of the generator, the generation of the gas should be so speedily arrested that no large escape of gas may need to take place. But in any case there must be an arrangement by which any surplus gas can be delivered outside the building.

4. The apparatus should be so arranged that the decomposition of the carbide should not give rise to excessive heating.

Non-automatic Generators.

1. Under no condition, likely to occur in working, must it be possible for the pressure in any part of the apparatus to exceed that necessary to support a column of water 100 inches in height.

2. The air space in the generating chamber should be as small as possible, and the apparatus should be so arranged that the decomposition of the carbide should not give rise to excessive heating.

3. There must be some arrangement by which, if the ordinary pipe from generator to holder becomes choked, the gas can escape by blowing a seal or by driving back feed-water and escaping through the tank.

The said committee also expressed the opinion:

That many types of acetylene gas apparatus can be so constructed as, with ordinary precaution, to be absolutely safe.

Although it does not follow that the generator which yields the largest amount of gas is necessarily the best, yet this factor is a most important one in the choice of any apparatus. The generators which combine the largest yield of gas with strength of material and simplicity in charging the carbide and in emptying the residue are those which will recommend themselves to the public.

Where the public is most likely to be misled is by the exaggerated claims made by makers as to the number of lights which a given machine will supply, and herein may possibly be an element of danger due to excessive heating caused by too rapid generation. Even if there be no danger, the overheating will considerably lessen the quantity and lower the quality of the acetylene gas evolved from the carbide, as well as tend to cause smoking of the burners.

The committee recommend -

That every apparatus sold should be accompanied by a written guarantee that it will light a specified number of burners consuming a given quantity of gas per hour over a consecutive number of hours without increasing the temperature in any part of the carbide receptacle above 228° C, that is to say, the fusing point of tin.

That non-automatic generators with a holder capable of taking the gas generated from the largest charge of carbide the generator will hold are free from objections attending all automatic generators examined, and that every generator should be fitted with an arrangement by which all air can be rinsed out of the generating chamber by acetylene or some inert gas before action is allowed to commence between the water and carbide.

That every generator should be fitted with a purifying chamber or chambers in which the acetylene is purified from ammonia and sulphuretted and phosphuretted hydrogen, and from other impurities.

Another important point is the length of time over which generation of gas continues after the addition of water to the carbide has ceased.

The general idea which seems to exist among makers of automatic apparatus of this type is that all they have to do in order to stop the generation of acetylene is to stop the water supply; this, however, is an utter fallacy, as liberation of gas continues with ever-increasing slowness for sometimes an hour and three-quarters after the water supply has ceased, whilst the gas so evolved is very considerable in volume. The length of time over which the generation extends will of course depend to a certain extent upon the amount of water added, the percentage of carbide undecomposed, and the temperature at which the mass of carbide happens to be when the water supply ceases, whilst the generation will itself depend upon

The dehydration of the calcic hydrate first formed, and

The decomposition of water condensed from the gas present as the temperature of the generator falls.