Identification of the source of energy for protein translocation across biological membranes is important in understanding the mechanism of this process. In eukaryotic cells, the tight coupling between translation and translocation and firm attachment of the secreting ribosomes to membranes, as well as theoretical calculations, have led to the suggestion that energy derived from protein synthesis is sufficient for protein translocation. On the other hand, in bacterial systems neither the attachment of ribosomes to membrane (other than nascent chains) nor tight coupling of translocation to translocation has been observed. Moreover, certain proteins can be translocated across membranes either at the time of, or after, translation. The separation of protein translocation from translation has made possible the demonstration that ATP hydrolysis is essential for post-translational protein translocation across bacterial membranes and, more recently, also across canine and yeast endoplasmic reticulum membranes. Here we report that certain ATP analogues inhibit co-translational protein translocation at concentrations that do not interfere with protein synthesis, suggesting that ATP is also required for co-translational protein translocation.