The capability of Delftia acidovorans MC1010 (SdpA+, RdpA–) to degrade the herbicides 2,4‐dichlorophenoxyacetate (2,4‐D) and 4‐chloro‐2‐methylphenoxyacetate (MCPA) was investigated in batch degradation experiments and interpreted in the light of the catalytic activity of the initial enzyme SdpA, an α‐ketoglutarate‐dependent dioxygenase, catalyzing the cleavage of the ether bond. Dense inocula of this strain degraded high initial 2,4‐D concentrations quickly to below the detection limit. The specific degradation rate became slower and the degradation remained incomplete when low initial biomass concentrations and/or 2,4‐D concentrations were used. With low initial 2,4‐D concentrations, the degradation rate became nearly concentration‐independent. This effect was less pronounced with MCPA, a substrate that is more readily degraded by SdpA than 2,4‐D. (S)‐2‐(2,4‐dichlorophenoxy)propionate ((S)‐2,4‐DP), the favored substrate of SdpA used as comparison, was almost completely utilized under the various conditions. The observed differences were attributed to the easier regeneration of the essential co‐substrate α‐ketoglutarate with MCPA and above all (S)‐2,4‐DP as a substrate compared to 2,4‐D. Decreasing SdpA activity, which was observed during the starvation of MC1010, will contribute to the apparent recalcitrance of 2,4‐D. Degradation of 2,4‐D by Bradyrhizobium sp. RD5‐C2, a strain initiating 2,4‐D cleavage by a monooxygenase not requiring an organic intermediate as a co‐substrate, proceeded to completion despite the approximately 20‐fold lower degradation rates compared to those obtained with strain MC1010. However, lowering of biomass concentrations resulted in restricted degradation also with strain RD5‐C2. The results were discussed in terms of the kind of co‐substrate in the cleavage reaction, the availability of essential intracellular metabolites, and the cells' maintenance demands.