The ability to analyze null alleles of genes can be an important means of studying both a protein's function and its interactions with other proteins involved in a particular process. However, if the gene encodes a protein that is essential to the viability of the cell, such analysis becomes complicated because a complementing copy of the gene must be present in the cell. In order to study the effects caused by the null allele, the complementing copy must be inactivated or lost. We report the development of a system in Escherichia coli which facilitates the manipulation of null alleles of essential genes. It consists of a strain deleted chromosomally for the essential gene and complemented for its function by a wild-type (wt) copy expressed from a plasmid counter-selectable for two markers bracketing the gene. Using this system, we have (i) searched for bypass suppressors of a deletion of the essential secE gene, (ii) ascertained the ability of various mutant secE genes to complement a deletion of the wt copy and (iii) isolated intragenic pseudorevertants of a null missense allele of secE. This methodology should be widely applicable to other cases in which essential genes are to be studied genetically.