A divergent selection experiment with mice, using plasma concentrations of insulin-like growth factor-1 (IGF-1) at 42 days of age as the selection criterion, was undertaken for 7 generations. Lines were not replicated. To obtain sufficient plasma for the IGF-1 assay, blood from four individuals was volumetrically bulked to obtain a litter mean IGF-1 concentration. This necessitated the use of between family selection. Although inbreeding accumulated in a linear fashion in each of the high, control and low lines, the rates were different for each line (3·6, 1·6 and 5·3% per generation for the high, control and low lines, respectively). As a consequence, the effects of selection and inbreeding are confounded in this experiment. Divergence between the high and low lines in plasma concentrations of IGF-1 continued steadily until generation 5. In generations 6 and 7, there was a reduced degree of divergence and this contributed towards the low realized heritability value of 0.15 ± 0.12. Six-week liveweight showed a steady positive correlated response to selection for or against plasma concentrations of IGF-1 until generation 4 (high-low difference = 1·7 g = 12%). In generation 5, a substantial drop in 6-week liveweight in the low line relative to both the high and control lines occurred (high-low difference, 3·9; g, 25%). This difference was maintained until generation 7.
This experiment suggests that genetic variation exists at 6 weeks of age in plasma concentrations of IGF-1 in mice. Furthermore, genetic covariation between plasma IGF-1 concentrations and liveweight at 6 weeks of age is likely to be positive. Further experiments have been initiated to examine these theories.