Free convection at the surface of the primary mirror, the cause of so-called mirror seeing, does remain an important issue for telescope designers. In earlier telescopes, the evidence of mirror seeing was possibly hidden in the background of larger dome seeing effects and this effect was first suspected by [Woolf 82], [Woolf 82a], to be the main seeing contribution in the telescopes of the latest generation.
More recently,
experimental measurements by [Racine 92] showed that at the CFHT,
mirror seeing is indeed the main cause of observatory-made seeing.
[Racine 92] analyzed 562 image frames from the HRCam instrument
(for high resolution imaging) during 25 nights over an 8-month
period for correlations with temperature differences.
He found that the main effect
came from the primary mirror and evaluated it at:
where 
is the seeing FWHM contribution and
is the temperature difference between the mirror
surface and the surrounding air.
A lower dependency was found with the temperature difference
between the dome inside and outside air:
arcsec 
.
In the next section we report the results of the analysis by the author of a much larger set of image quality data of the CFHT.
Then a more extensive analysis of the mirror seeing phenomenon is found
in section 
below.