Artificial selection for 18-day pupa weight and opposing simulated naturel selection in Tribolium castaneum

Abstract

The effect of simulated opposing natural selection on the response to mass selection for 18-day pupa weight of Tribolium castaneum was studied for 10 generations of selection. Natural selection was simulated in replicated treatment lines by imposing a negative relationship between mid-parent genetic value for pupa weight and fertility. Responses to selection and realized heritabilities were smaller (P < 0.05 and P < 0.10, respectively) for the treatment lines than for control lines under selection for pupa weight only. One treatment, line E3G1, reached an intermediate selection plateau by generation 10, and responded linearly to 4 generations of artificial selection after natural selection had been discontinued. Possible explanations for the different behaviors of the replicate lines E3G1 and E3G2 were also discussed.

The effects of alternated cycles of full-sib and random mating on selection for pupa weight in Tribolium castaneum

“…The differentials of selection per generation were similar to those achieved in other experiments (e.g., Minvielle and Gall 1980). However, no significant difference was observed between the two types of lines.…”

“…Average response per generation, estimated by linear regression, and its standard error are listed in Table 1. The averaged observed gain per generation was higher than in several experiments of about the same duration (Gall 197 1 ;Enfield 1977;Minvielle and Gall 1980). It is likely that genetic gain was faster than in these experiments because gene frequencies in the base population, a cross between two very different lines, were certainly close to 0.5.…”

Artificial selection on larval growth curves in Tribolium: correlated responses and constraints

“…We suspect that this decline in survivorship is mediated through development rate, which genetically covaries negatively with body size in this population and which shows a distribution pattern that suggests that development times below about fifteen days are not viable. These results differ from most published artificial selection results in insects, which tend to require many generations of selection to elicit a response, followed by a prolonged response as selection continues (Yamada & Bell, 1969;Kaufman et al, 1977;Minvielle & Gall, 1980;Partridge & Fowler, 1993;Partridge et al, 1999;Teuschl et al, 2007). We think these differences are explainable by the type of selection imposed and by the correlated responses to selection.…”

“…These results differ from most published artificial selection results in insects, which tend to require many generations of selection to elicit a response, followed by a prolonged response as selection continues (Yamada & Bell, ; Kaufman et al ., ; Minvielle & Gall, ; Partridge & Fowler, ; Partridge et al ., ; Teuschl et al ., ). We think these differences are explainable by the type of selection imposed and by the correlated responses to selection.…”

“…In insects, the former scenario of limited selection response has been demonstrated in single traits for which there is insufficient additive genetic variance and in sets of two traits that genetically trade-off and are both correlated with fitness (Bell & Burris, 1973;Conner & Via, 1992;Davidowitz et al, 2005). The latter scenario of reduced fitness during selection has been measured in several of the aforementioned artificial selection studies, either through an increase in mortality or through an increase in sterility when selecting for larger body size (Bell & Moore, 1971;Kress et al, 1971;Enfield, 1979;Minvielle & Gall, 1980). Although past results have elucidated important lifehistory trade-offs and constraints on the evolution of body size, some shortcomings persist.…”