Two contrasting experimental descriptions of PI-plasmid replication during the cell cycle of Escherichia coli have been described. One set of results led to the proposal that replication of PI-plasmid occurs at a specific time during the cell cycle over a wide range of growth rates and follows rules similar to that governing bacterial chromosome replication. Experiments supporting this proposal utilized membrane-elution experiments, radioactive double-labeling of DNA and scintillation counting of purified plasmids. An alternative experimental description of P1-plasmid replication during the cell cycle, also based on membrane-elution methodology but measuring radioactivity incorporated into plasmid DNA by autoradiography and scanning of films, proposed that P 1-plasmid replicates at all stages of the cell cycle in rapidly growing cells, but with a slight periodicity or increase in P1 replication probability within the cell cycle of slower growing cells. These discordant experimental results are analyzed. It is concluded that the direct double-label counting approach is to be preferred, as the results are consistent with a large number of experiments, are supported by theoretical considerations and yield a unified view of plasmid replication over a wide range of growth rates. Theoretical ramifications of each view of P 1-plasmid replication-cycle-dependent and cycle-independent are compared. An analysis of PI-plasmid segregation is also presented.