An example of a parameter combination in which resistance develops more slowly when insecticides are used in sequence relative to when they are used in a mixture.

<p>The labels on the upper margin of the plot mark where the resistance threshold is reached for the second insecticide in a sequence ('seq') and the first and second insecticides in a mixture ('mix1' and 'mix2'). Note that resistance spreads faster for both insecticides when they are deployed on their own (dashed lines) than when they are deployed in a mixture (solid lines). The reason sequential use is favoured is because of the time delay that occurs before insecticide #2 is deployed in the sequence. In this example, the frequency of resistance alleles to both insecticides in the mixture exceeded 50% at generation 100. Resistance allele frequency to insecticide 1 deployed alone required 53 generations to exceed 50% (at which point it was replaced by insecticide 2 as indicated by the vertical dotted line). It then took 75 generations for resistance allele frequency to exceed 50% for insecticide 2 deployed alone. The total of these timescales, i.e. 53 + 75 = 128 generations, is greater than the 100 generations for resistance to evolve to the mixture, indicating that sequential deployment is favoured in this simulation. [parameter combinations are as used for Curtis in his Fig 2 in [<a href="http://www.ploscompbiol.org/article/info:doi/10.1371/journal.pcbi.1005327#pcbi.1005327.ref017" target="_blank">17</a>] (our <a href="http://www.ploscompbiol.org/article/info:doi/10.1371/journal.pcbi.1005327#pcbi.1005327.g002" target="_blank">Fig 2B</a>) with the exception that each effectiveness was reduced by 0.2 to 0.53 for insecticide1 and to 0.8 for insecticide2]</p>