Dry matter (DM) degradation of Glycrrhiza glabra L, Arbutus andrachne, Juniperus communis, and Pistica lentiscus was determined using two different techniques: (i) the in vitro gas production and (ii) the in situ nylon bag degradability technique. Samples were incubated in situ and in vitro for 3, 6, 12, 24, 48, 72 and 96 h. In situ and in vitro DM degradation kinetics were described using the equation y = a + b(1 - e ct). At all incubation times except 3 and 72 h the cumulative gas production of J. communis was significantly lower than that of G. glabra, A. andrachne and P. lentiscus. At 3, 6 and 12 h incubation times the DM disappearance of J. communis was only significantly lower than that of P. lentiscus. At 24 and 48 h incubation times DM disappearance of J. communis was significantly lower than that of A. andrachne and P. lentiscus. There were significant relationships between in vitro gas production and in situ DM disappearance at 24 h and 96 h incubation times. The gas productions at 24 and 96 h incubation explained 51.2 and 52.4% of variation of DM disappearance, respectively. Gas production from the insoluble fraction (b) alone explained 66.4% of the variation of effective DM degradability (EDMD). The inclusion of gas production from quickly soluble fraction (a) and rate constant (c) of gas production in the regression equation did not improve the accuracy of predicting EDMD. It was concluded that in situ DM disappearance parameters of tannin-containing tree leaves such as used in this present study may be predicted from in vitro gas production parameters. © South African Society for Animal Science.