Based on observations carried out at the European Southern Observatory (ESO, La Silla, Chile) and with the 1-m Swiss telescope at the Haute-Provence Observatory.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

Table 7 is only available in electronic form at http://www.edpsciences.org

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

...1,

Senior Research Associate, FNRS, Belgium.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... star

Interestingly, Pinsonneault et al. (1984) quoting a private communication from E. Luck, remark that Luck "is completing an abundance analysis of HD 139409, which will show that the light s-process elements do have ( $\sim$ +1.5 dex) enhancements, while heavier s-process element enhancements are much smaller ( $\sim$ +0.3 dex)''. Luck's study seems to have never been published.

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.

... matrix

Since it may be shown that $\chi^2_{\rm S} - \chi^2_{\rm T} = \chi^2_{ABFG}$ , the Pr₂ and Pr₃ tests are in fact equivalent provided that $\chi^2_{\rm T} \sim N-9$ . Thus, if Pr₂ and Pr₃ are significantly different, it means either that the Thiele-Innes orbital model does not fit the data very well ( $\chi^2_{\rm T} \gg N-9$ ), or that it fits much better than could be expected ( $\chi^2_{\rm T} \ll N-9$ ). We are indebted to L. Lindegren for this clarification (see also Jancart et al. 2005).

.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.