[en] Full text: The CDF muon scintillator counter system consists of 4 main subsystems: Central Muon Scintillators (Top and Bottom CSP, North Wall and South Wall CSP); Central Scintillator Extension (CSX); Barrel Scintillator Upgrade (BSU); Toroid Scintillator Upgrade (TSU). All Top and Bottom CSP and all BSU counters, which were newly constructed at JINR (Dubna) for Run 2, using polystyrene scintillator plates manufactured at Kharkov seem to be performing as expected. However, one of these subsystems is still equipped with 'old' counters, which were used in the CDF Run 1. Amongst these are the CSP wall (North and South walls) counters which were found to suffer from premature aging and were refurbished during the CDF upgrade. These scintillators are 320 cm long. The 180 cm long CSX (270 counters) counters were constructed from the same material (N/A) and therefore suffer from the same premature aging problems. New plateau voltages were determined for all of them but it was noted that, on the basis of projections of the observed rate of degradation, they were expected to become inefficient at the end of 2005. Measurements of parameters of the CSX and Wall CSP counters were performed using cosmic muons, which were selected by a trigger telescope. The coincidence of these scintillators was used to trigger the data acquisition (DAQ) and to start the ADCs, which registered pulses from the counter outputs. The photomultipliers (PMT) were calibrated by illuminating them with weak light pulses from a fast blue LED ('MICHIA' NSPB310A). This procedure allows a calibration constant k (in channels/photoelectron) to be extracted for each PMT and ADC combination. On the basis of this calibration, the number of photoelectrons npe, emitted by the photocathode for cosmic muons traversing the counters normally, was measured for all counters. Measurements of a sample 24 wall counters were made in September-October, 2003 have shown that the average number of photoelectrons from the far end of these scintillation bars is ∼6 and that the average efficiency of this sample ≅97%. The ratio of the average responses (N1999/N2003=15.3/5.9≅2.60) and it corresponds to a degradation rate of ∼19%/year. On the basis of this new estimate of the rate of degradation, the average light output from the far end of the CSP counters is expected to be ∼5 ph.e./MIP by the end of 2004. Assuming Poisson statistics and discriminator thresholds corresponding to 1 ph.e., this corresponds to an efficiency of 98%. The responses of the CSX counters were also sampled during the same period. A total of 16 CSX counters was removed from the arches and tested on the bench following the same procedure as was used for the CSP counters and described above. The average number of ph.e./MIP is 8.1 which means that the average counter efficiency is ≅99%. On the basis of the mean of measurements reported in table 2. (8 ph.e./MIP) and of the average measured in 1998 (12 ph.e./MIP), we expect a deterioration rate of ∼8%/year. On the basis of this estimate on would therefore expect the average light output from the far end of the CSX counters to be ∼7.5 ph.e. by the end of 2004, corresponding to an average efficiency of 99%. (author)