In a recent article in this journal, Mitra (1983) extended work by Espenshade, et al. (1982) who examined long-run implications for population size and composition if below-replacement fertility rates remain fixed in face of a constant influx of immigrants. By differentiating fundamental renewal equation for annual births, Mitra developed a unified framework for studying existence of long-term equilibrium solutions when constant annual immigration is combined with fertility rates that are below, at, or above replacement. When fertility is below replacement, Mitra confirmed conclusion reached by Espenshade et al. showing that a stationary population is long-run outcome. He showed that annual births, and therefore total population, grow linearly if fertility is at replacement. For fertility above replacement, Mitra found that population growth rates tend asymptotically to intrinsic rate they would exhibit in absence of immigration. In case where fertility is below replacement, it would be more accurate to say that Espenshade et al. have generalized Mitra's results. Mitra assumes that immediately upon their arrival, immigrant women adopt fertility behavior of native born women. Espenshade et al. have shown that a stationary population will materialize in long run even if immigrant women and successive generations of their descendants have fertility above replacement. All that is needed for a stationary population to result is that, at some point in chain of immigrant descendants, one generation and all those following it must adopt belowreplacement fertility. Moreover, in case where fertility is at replacement and a constant number of immigrants is assumed to enter population each year, Ansley Coale (1972) preceded Mitra in showing that population growth will be linear in equilibrium. In last part of his paper, Mitra studies a population with two homogeneous groups, one having fertility rates above replacement and one with fertility rates below replacement. He then examines the mechanisms by which growing population may adopt reproductive norms of declining population to result in eventual stationarity of both groups (113). It is not entirely clear how this section of Mitra's paper relates to his previous results on immigration. To link two, it is useful to introduce migration explicitly into analysis. To do so, imagine a closed population of females divided into two separate geographic regions. Suppose that women with below-replacement fertility reside in north and those with fertility above replacement live in south. Let mH(a) and mL(a) be age-specific fertility schedules of high and low fertility women, respectively. Assume that both regions are characterized by same force of mortality schedule, yd(a). Assume further that women from south can migrate to north (but not vice versa), and that they migrate at a rate which is constant through time, but not necessarily across age. Once in north, these migrant females take on fertility schedule mL(a). The rate at which women move from high fertility area to low fertility one is formally equivalent to Mitra's concept of rate at which high fertility women adopt reproductive norms of other group. Suppose we represent rate of out-