Genetic diversity and the effect of lamb and dam inbreeding on multiple traits were analyzed in an 11-yr closed population of sheep established in 1983 and remained closed after 1987, with 50% Dorset, 25% Rambouillet, and 25% Finnsheep breeding to determine selection response for spring fertility. The population had been randomly divided in 1987 into a fall-lambing selection line (S) of 125 ewes and 10 rams, fall-lambing environmental control line (E) of 55 ewes and 5 rams, and a spring-lambing genetic control line (G) of 45 ewes and 5 rams. Inbreeding effects were estimated from 2678 lambs and 556 dams present after the creation of the respective lines. The traits assessed were ewe spring-fertility, lambing date, lamb birth, 60 d, and 120 d weight, and lamb survival to 1, 3, and 14 d. Genetic diversity was assessed by estimating change in inbreeding per generation (ΔF) and effective number of breeding animals (N_e), and parameters derived from gene drop simulations and an iterative procedure developed by Boichard et al. (1997); effective number of founders (f_e), effective number of ancestors (f_a), founder genome equivalents (f_g), and two additional measures of genetic diversity (GD₁, GD2). In order to estimate the diversity available in S and G, three sets of animals from the end of the study and one set of animals at line formation were considered in each line: all lambs born (including dead lambs), all matings (including potential offspring, even if a lamb was not born), and all rams and ewes available at the end of the study and at line formation.

At the time of line formation, most of the loss in diversity was due to unequal founder representation. The smaller population of G, as compared to S, caused a greater decrease in diversity due to bottlenecks at line formation. Very little diversity was lost due to additional drift by the time of line formation because selection had not occurred and a random mixing of founders was the goal. Allelic diversity decreased moderately; of the 322 founder alleles, there were 71% in S and 58% in G of rams and ewes (RE) that appeared in at least 50 runs of gene drop. By the end of the study in 1998, the amount of allelic diversity had decreased substantially. Of the alleles possible in RE at the end of the study in S and G, only 6 and 8 %, respectively, appeared in greater than 50 simulations of gene drop. The measures of f_e, f_a, and f_g revealed there was not much additional loss in diversity from the line founders to the end of the study due to unequal founder representation, but there was a larger amount of loss due to bottlenecks and additional drift. The diversity loss was similar, which was the goal of the selection study, when values for RE were compared in S and G.

The effects of lamb and dam inbreeding were estimated from REML analysis. Effects of lamb or dam inbreeding were negative but not significant for lambing date or survival to 1, 3, or 14 d. Spring fertility was estimated to decrease by 0.70 ± 0.30 %/% inbreeding of the ewe (P < 0.01), which seems even greater since the average spring fertility was only 47.5 %. Effects of lamb inbreeding on birth, 60-d, and 120-d weights were -0.012 ± 0.006 (P < 0.05), -0.045 ± 0.020 (P < 0.05), and -0.130 ± 0.034 kg/% (P < 0.01), respectively. Dam inbreeding had smaller effects on birth, 60-d, and 120-d weights of -0.008 ± 0.010 (ns), -.033 ± 0.034 (ns), and -0.087 ± 0.056 (P < 0.1) kg/%, respectively.