Based on observations collected at the European Southern Observatory, Chile in runs 073.C-0164, 075.C-0710C, 076.C-0339B, 077.C-0264B, 078.C-0552B, and 078.C-0535A.

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

In the Feb. 2007 service mode observation, HIP 73357 B was inside the field-of-view only in two out of five frames, so we could not use it properly. Instead, we used HIP 73357 observations done with the identical set-up a few nights later on 1 March 2007 (run 078.C-0535A), where 43 useful high-S/N frames were taken ( $80 \times 0.347$ s each) by some of us. We used them to calibrate the 19 Feb. 2007 observations of GQ Lupi, because there was no intervention into NACO in the meantime, as confirmed with ESO.

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

... $\mu _{\delta} = -23.7 \pm 2.7$ mar/yr

These two proper motions are not only consistent with each other, but also consistent with the published proper motion of GQ Lupi A being $\mu _{\alpha} = -19.15 \pm 1.67$ mas/yr and $\mu _{\delta} = -21.06 \pm 1.69$ mas/yr (Mugrauer & Neuhäuser 2005).

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

... edges

We are monitoring these field distortions in a different (astrometric) observing program, where we observe the 47 Tuc cluster center several times per year with the NACO S13 FoV and always the same set-up (also 2005 to 2007 as GQ Lupi). We achieved sub-mas relative precision; first results are published in Neuhäuser et al. (2007) and Seifahrt et al. (2007a), more details will be given in Roell et al. (in preparation). From these observations, we already know that the field distortions are very small except at the field edges.

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

... $e \simeq 0.2$ )

We confirm a-posteriori, see below, that these assumptions are probably correct. In reality, we have started with slightly other assumptions and have then iterated; in the iterations, the values for pixel scale and detector orientations stayed constant, just the (maximal) errors due to (maximal) orbital motion changed.

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