It was hypothesized that the long lengths of sampling catheters required when a mass spectrometer is multiplexed to more than one operating room limit the upper frequency at which a gas concentration may be determined accurately. This possibility has not been investigated. Known step changes of CO2were generated by a solenoid valve driven by an electronic timer that was adjustable from 0.1 to 10 Hz. The valve alternated between 100% O2and 7% CO2in 50% O2and balance N2. CO2concentration was monitored by a mass spectrometer after the gas passed through a 3.7 m Teflon® catheter or through 30 m Teflon®, nylon, polyethylene (PE), or polyvinylchloride (PVC) catheters. Gas flow for all catheters was adjusted to 1.1 ml/s. The peak-to-peak output of the mass spectrometer was read from a storage oscilloscope. The 3.7 m catheter caused a 10% error at 5.5 Hz (330/min). In sharp contrast, 30 m catheters made from Teflon®, PVC, and PE caused errors greater than 10% at only 0.6 Hz (36/min). The 30 m nylon catheter passed 1.1 Hz (66/min) with a 10% error. Teflon®, PVC, and PE are not suitable materials from which to make long catheters sampling CO2. Because the frequency response of the nylon catheter appeared similar to that of a low-pass filter, an electronic circuit was designed and tuned to extend the high-frequency response of the catheter. With the circuit in place, the frequency at which a 10% error occurred in the measurement of CO2improved from 1.1 Hz (66/min) to 2.2 Hz (132/min). This simple circuit, which may be retrofitted to existing units, is an effective method to extend the frequency response and accuracy of long catheters sampling CO2.