In modern small optoelectronic devices like field emitter displays, miniature cathode ray tubes (CRTs), channel photo multipliers, etc., the vacuum requirements are much more stringent than in conventional electron beam devices. As there should be a pressure in the ultrahigh vacuum region and the volume is only a few cm(3), a direct measurement is not feasible and is often estimated on the basis of the expected pumping speed of the getter. The present study was arranged to investigate the pressure in small CRTs (25 cm(3)) during a period of several months, namely after the conventional pumping and bakeout procedure, immediately after the activation of Ba getters and after the accumulation of some months. All the CRTs were equipped by a spinning rotor gauge ball. Two barium getter sizes were studied: St15/AM/O/9.5 and St15/AM/O/5, both made by SAES. After the evaporation by the prescribed procedure the pressure did not drop, but increased from p similar to 1 x 10(-5) mbar up to p similar to 1 x 10(-3) mbar, showing that the pumping speed was completely suppressed by forming of a nongetterable gas. When the same experiments were repeated inside identical glass bulbs connected with a valve to a quadrupole mass spectrometer, a formation of methane was observed. The initial rate just after the activation was Qsimilar to10(-8) mbar 1 s(-1), but even after several hours it was still as high as Qsimilar to10(-9) mbar 1 s(-1). By switching-on the cathode heater, methane was pumped by the getter after a precedent cracking procedure. The virtual pumping speed was directly related to the heater power, but independent of the getter area. Therefore, within the cathode warm-up period, methane was "pumped" and did not represent a harmful gaseous contaminant in a small electron beam device with a thermionic cathode.