During general anesthesia, removal of exhaled carbon dioxide by rebreathing appliances is provided by chemical absorption and this carbon dioxide absorption during anesthesia would at first glance seem to be a simple problem. The problem, however, is far more complicated and much attention must be given to details of rebreathing appliances. Incomplete removal of carbon dioxide in circle absorber systems, which are commonly used today must be carefully monitored to prevent hypercapnea caused by accumulation of carbon dioxide in the breathing circuits. Although circle absorbers have been developed through clinical trials anesthesiologists have no reliable indication of the end point of useful life of the absorbent. This study was undertaken to investigate the useful life of the soda lime used in two type of canisters (type A: canister of anesthesia machine. Quantiflex, USA. 12.9cmX16.6cm, tbye B: canister of anesthesia machine, International 2. U.K., 13.1cmX16.9cm). the effect of flow rate on this useful life and the relations between indicator change and the endpoint of useful life of the soda lime. The control group used a as total flow rate of fresh gases at 2.0 liters per minute and the experimental group used a total flow rate of 4.0 liters per minute. The carbon dioxide concentration of 0.1% at the canister exit was employed as the end point of useful life of the absorbent and the exit carbon dioxide concentration were monitored continuously by intra-red carbon dioxide monitor apparatus(Datex Normocap CD-102, Pinland). Also the carbon dioxide concentration in the respiratory circuit of exhaling side was examined. The results were as follows: 1) In type A, the average useful life of soda lime was 1,885 minutes in the control group and 2,514 minutes in the experimental group. In type B, the average useful life of soda lime was 1,587 minutes in the control group and 1,980 minutes in the experimental group. 2) At the end point of useful life of the absorbent the level of the color change of the indicator of the absorbent in all types was above the three-fourth height of two chambered canister. 3) At the end point, end tidal CO2 concentration was 3.5 vol% in the control group 2.88 vol% in the experimental group. The data obtained from this observation indicated that the increase of total flow rate of gas and the type A canister prolonged the useful life of soda lime and that the double chambered canister could be used safely until the color change of the indicator reached the level of three fourth the height of the canister.