Demonstrating the accuracy of scaling parameters. The thick line shows the mean copy number from computer simulations run over 50,000 generations for a population size

*N* = 10

^{4}, with initial copy number

*x* = 10, rate of transposition per element

*u* = 10

^{−4}, selection coefficient

*a* = 10

^{−5}, and synergism coefficient

*b*. The equilibrium is solved for a given rate of excision per element according to . (A) An excision rate two orders of magnitude lower than transposition (

*v* = 10

^{−6};

*b* = 6.0 × 10

^{−6}; number of simulation runs,

*r* = 25). (B) An excision rate one order of magnitude lower than transposition (

*v* = 10

^{−5};

*b* = 5.4 × 10

^{−6};

). The thin lines denote the standard error of the mean between runs of the simulation. The triangles show the mean copy number from simulations scaled 10-fold (

*r* = 100). The circles show the mean copy number from simulations scaled 100-fold (

*r* = 1000). The generation time is also scaled upward to compare with the unadjusted simulation runs.

## PubMed Commons