Experimental evidence accumulated to date by several research groups indicates that antisense oligodeoxynucleotides targeted against messenger RNA (mRNA) sequences located downstream of the initiation codon fail to inhibit the translation of this mRNA unless the hybrid is cleaved by RNase H. It has previously been shown that exogenous RNase H has to be added to rabbit reticulocyte lysate to obtain translational arrest (unless freshly prepared lysates are used). In contrast there is no need of exogenous RNase H by using wheat germ extract for translation because the level of endogenous RNase H is high enough to ensure cleavage of the hybrid formed between the antisense oligodeoxyribonucleotide and its complementary sequence on the mRNA. Surprisingly, we found that these two cell-free translation systems display similar amounts of RNase H activities when tested under standard conditions (extract diluted 500 times in the RNase H reaction mix). The RNase H activity of the rabbit reticulocyte lysate has a divalent cation requirement and sensitivity to inhibitors similar to class I ribonuclease H, whereas the activity of the wheat germ extract shows similarities to class II ribonuclease H. However, when these activities were assayed under conditions similar to those used for translation experiments, only highly reduced levels of activity were found in comparison to the standard assays. This reduction is due in part to sub-optimal ionic conditions for the endogenous RNase H activities in these extracts, and, for the other part, likely due to interactions with other proteins present in the lysates. In these conditions, however, the remaining activity found in the wheat germ extract was three times higher than the activity found in the rabbit reticulocyte lysate. Whether this difference can by itself explain the indicated differences in the two systems observed in hybrid-arrest of translation experiments remains open to discussion.