How many measurements do you need to make? If you just want to make the odd measurement then the Konica Minolta (a 'point and click' type product) that Dario Ambrosini mentions is a good choice. It looks like an old camcorder, and is extremely quick and easy to use. Used in many published papers. Solid bit of kit.
If you want to make lots of measurements (e.g., are trying to characterise a whole screen, or to see how your stimulus varies with pixel/command level) then you'll want a programmable USB device like the Spyder that Claudia Menzel mentions, or the CRS ColorCal II.
I like to use a combination of the CRS ColorCal II and the Konica Minolta myself, each of which cost in the region of 2000 €, as Dario says (though they can sometimes be found 2nd hand, or borrowed from people in your department). The spyder is a bit cheaper but I don't know whether or not it is well regarded.
Different equipment is required if you want to measure very small, very near, or very intense stimuli. Or if you want to measure ambient illuminance, or colour spectrum. For example, if you want to measure ambient illumination you can get a cheap lux meter online (50-500€), so it's worth investing in one (unlike cheap sound level meters, cheap lux meters are generally fine for psychology/neuroscience)
EDIT: Oh yes, and it may be of interest to note that you can now get some displays that have programmable photomters built in. For example, see:Article Plug and play perimetry: Evaluating the use of a self-calibr...