Having such genes on plasmids make horizontal gene transfer possible coercing the cheaters to make the extracellular products. Although such systems can be robust to bacterial cheaters, they may not be robust to plasmid cheaters with the gene for the extracellular products deleted. Such plasmids will have an equal chance of horizontal transfer but the shorter plasmids could replicate faster and replace the wild type. Therefore although many plasmids carry useful genes, that does not appear to be necessary and sufficient cause of plasmid stability. Furthermore, it has been suggested that in the absence of selection for any plasmid borne gene a plasmid free cell should have a selective Hederacolchiside-A1 advantage and the stability of the plasmid is difficult to explain. Factors such as spread of the plasmid by conjugal transfer to compensate for the loss in host fitness, compensatory mutations, co-evolution of host and plasmid to reduce the cost or plasmid addiction are some of the possible mechanisms responsible for maintenance of a plasmid. Many mathematical models have focused on the problem of persistence of plasmid in the host cells. More perplexing is the problem of stability of copy number of a plasmid in the host cell. This is one of the many paradoxes in the evolution and stability of plasmids. Most of the plasmids are considerably smaller than the chromosome in terms of length of DNA. Therefore if the same machinery is being used for replication, there can be many replication cycles of the plasmid per single replication of the chromosome. This can result into rapid escalation of plasmid numbers in a cell leading to increased metabolic burden on the host cell and eventually cell death. This can be prevented only by tight regulation of plasmid replication. Ironically, all critical genes involved in the regulation of plasmid replication are on the plasmid. This raises a potential evolutionary problem. A low copy number is optimal for long term Harpgide survival of the plasmid in a host cell lineage. However, any mutation that loosens the control and thereby increases the copy number has a short term advantage over the wild type although it may affect long term stability of the system.