The paper discusses the challenges and limitations of using indirect measures of gene flow and migration, particularly based on $F_{ST}$, a standardized measure of genetic variance among populations. The authors argue that while $F_{ST}$ is a valuable measure of genetic structure, it is often used to estimate $Nm$ (the number of migrants per generation) under unrealistic assumptions. These assumptions, such as infinite populations, constant migration rates, and no selection or mutation, are rarely met in natural populations, leading to significant deviations from the true values of $Nm$. The paper highlights several issues, including the impact of selection, spatial variation in population size and migration rates, and the non-equilibrium state of many populations. It also emphasizes the statistical uncertainty associated with estimating $Nm$ from $F_{ST}$, which can result in large confidence intervals. The authors conclude that while $F_{ST}$ can be useful for studying genetic differentiation, it should not be directly translated into estimates of gene flow or dispersal without caution. They suggest that direct methods of measuring migration, such as mark and recapture, are more reliable but labor-intensive, and that a combination of both approaches may be necessary for a comprehensive understanding of gene flow and dispersal in natural populations.The paper discusses the challenges and limitations of using indirect measures of gene flow and migration, particularly based on $F_{ST}$, a standardized measure of genetic variance among populations. The authors argue that while $F_{ST}$ is a valuable measure of genetic structure, it is often used to estimate $Nm$ (the number of migrants per generation) under unrealistic assumptions. These assumptions, such as infinite populations, constant migration rates, and no selection or mutation, are rarely met in natural populations, leading to significant deviations from the true values of $Nm$. The paper highlights several issues, including the impact of selection, spatial variation in population size and migration rates, and the non-equilibrium state of many populations. It also emphasizes the statistical uncertainty associated with estimating $Nm$ from $F_{ST}$, which can result in large confidence intervals. The authors conclude that while $F_{ST}$ can be useful for studying genetic differentiation, it should not be directly translated into estimates of gene flow or dispersal without caution. They suggest that direct methods of measuring migration, such as mark and recapture, are more reliable but labor-intensive, and that a combination of both approaches may be necessary for a comprehensive understanding of gene flow and dispersal in natural populations.