Conclusions The examples discussed above show clearly that the development of vectors for use in eukaryotic cells is presently proceeding very rapidly. Knowledge gained from the use of one type of vector defines the design criteria for the next type and thus many systems become obsolete before they are fully developed. Some future work can be predicted with reasonable certainty. The vast majority of future vectors will incorporate dominant selectable markers and regulatable promoters and we can expect to see vectors containing several different origins of DNA replication so that their replication can be controlled in several ways. In parallel with these refinements will occur the development of eukaryotic vectors suitable for the primary cloning of genomic DNA or cDNA. This will be of the greatest importance as it will allow the immunological or phenotypic recognition of genes by virtue of their expression in eukaryotic cells, thus facilitating the isolation of eukaryotic genes which can not be detected by the techniques available for screening prokaryotic clones. The use of such vector systems will surely be a central feature of all future work in eukaryotic molecular biology.