It was reported in [1] that a bulk diffuse dielectric barrier discharge (DBD) was implemented in atmo spheric air with natural humidity of 40-60% in a 2 mm discharge gap (DG) without any additional sources of external preionization when rectangular voltage pulses with nanosecond rise times (formed by alternating closing of high voltage solid state switches [2,3]) are fed to electrodes and alumina ceramics (AC) is used as a barrier. It was noted that the rate of rise of the electric field strength in the DG and the external circuit resistance play an important role in the formation of diffuse DBD. The influence of the barrier material on the degree of homogeneity of this discharge was not analyzed. In this paper, we report the results of studying a diffuse DBD in atmospheric air for two barrier materials at the voltage pulse repe tition frequency from 30 to 3000 Hz and estimating the degree of homogeneity of the discharges obtained.The experimental setup for igniting DBD was sim ilar to that described in [1]. For the DBD initiation, 60 μs rectangular voltage pulses with an amplitude of 0-16 kV and rate of rise of 0.4 kV/ns were fed to the DG. The pulse repetition frequency was varied in the range from 30 Hz to 3 kHz. The electrodes formed a plane parallel system. Each electrode had a circular cross section with a diameter of 30 mm. Dielectric plates made of AC with 95% Al 2 O 3 content and poly methylmethacrylate (PMMA) were used as a barrier. The plates were 50 × 50 × 2 mm in size.The air gap was 2.3 mm for ceramics and 1.75 mm for Plexiglas, which provided an identical equivalent capacitance of the system of electrodes (1) where C a is the air gap capacitance and C b is the bar rier capacitance; these values were found from the C eq C a C b C a C b + 2.6 pF, = = known formula for a flat capacitor. The barrier capac itances were 27 and 11 pF for ceramics and PMMA, respectively. Discharge characteristics were estimated from oscillograms of the discharge current and photographs of the glow from the DG obtained by a camera. This technique is widely used for studying discharges [4-9]. The exposure time was chosen so that the image of glow in the DG would correspond to a single voltage pulse applied to the electrodes. The degree of homoge neity was estimated proceeding from analysis of the image brightness distribution in the cross section of the DG middle (the selection layer height was 70 μm).The distribution of the DBD glow brightness in the DG was determined by digitizing the discharge photo graphs. To analyze the degree of homogeneity of the discharge, this distribution was compared with the ideal curve of glow brightness distribution obtained on the assumption of homogeneous and isotropic dis charge glow with absorption neglected. For cylindrical electrodes, this curve is described by the dependence of the perpendicular chord length on the horizontal coordinate in the cross section (from one electrode edge to the other)where R is the electrode radius and x is the horizontal coordinate. Figure 1 shows the typical...