The _hardness_ or _resistance to abrasion_ is determined by shaking a 50-gram sample with steel ball bearings for 30 minutes on a Ro-tap shaking machine. The material is then screened and the hardness number is determined by multiplying the weight of absorbent remaining on the screen by two. The _efficiency_ of an absorbent against various gases depends upon a variety of factors. Because of this, it is necessary to select standard conditions for the test. These were chosen as follows: The absorbent under test is filled into a sample tube of specified diameter (2 cm.) to a depth of 10 cm. by the standard method for filling tubes, and a standard concentration (usually 1,000 or 10,000 p.p.m. by volume) of the gas in air of definite (50 per cent) humidity is passed through the absorbent at a rate of 500 cc. per sq. cm. per min. The concentration of the entering gas is determined by analysis. The length of time is noted from the instant the gas-air mixture is started through the absorbent to the time the gas or some toxic or irritating reaction product of the gas begins to come through the absorbent, as determined by some qualitative test. Quantitative samples of the outflowing gas are then taken at known intervals and from the amount of gas found in the sample the per cent efficiency of the absorbent at the corresponding time is calculated. Per cent efficiency = p.p.m. entering gas - p.p.m. effluent gas ----------------------------------------- × 100. p.p.m. entering gas These efficiencies are plotted against the minutes elapsed from the beginning of the test to the middle of the sampling period corresponding to that efficiency point. A smooth curve is drawn through these points and the efficiency of the absorbent is reported as so many minutes to the 100, 99, 95, 90, 80, etc., per cent efficiency points. The apparatus used in carrying out this test is shown in Fig. 74. Descriptive details may be found in the article by Fieldner in _The Journal of Industrial and Engineering Chemistry_ for June, 1919. With modifications for high and low boiling materials, the apparatus is adapted to such a variety of gases as chlorine, phosgene, carbon dioxide, sulfur dioxide, hydrocyanic acid, benzyl bromide, chloropicrin, superpalite, etc. As the quality of the charcoal increased, the so-called standard test required so long a period that an accelerated test was devised. In this the rate was increased to 1,000 cc. per minute, the relative humidity of the gas-air mixture was decreased to zero, and the concentration was about 7,000 p.p.m. The rate is obtained by using a tube with an internal diameter of 1.41 cm. instead of 2.0 cm. CANISTERS