Because most of the small computer and tv/video projectors have UHP lamps, i think that i saw many times the spectra of these lamps. To me the spectrum of the UHP look like a continuous spectrum: bearly seeing the mercury lines because they are so wide and the rest of the molecular continuum is so intense that they merges together. My main question is why there is no any dark re-absorption lines on each mercury line like in the main lines of sodium, thallium and indium?

Because visible mercury lines are from higher excitation order:
The first transition from totally unexcited state correspond to UV line.
Such exited electron might be excited further, this time by lower energy increment, what correspond to longer wavelength.
So only already excited atom might absorb the photon of the longer (e.g. blue or green) wavelength. But they are only few such excited atoms and these concentrate to discharge core, the majority of the volume is not excited at all, so it does not absorb visible photons, so yu do not see selfabsorbtion in the visible part of the spectra.
This is different from e.g. sodium lamp, where the yellow correspond to already first transition from unexcited state, so majority of the sodium volume (unexcited) does absorb the main sodium line, hence the observed selfabsorbtion in HPS lamps.
Similar situation exist for mercury as well, but affect only UV lines, as these correspond to first transitions from an unexcited atom. That's why mercury pressure in fluorescent should be kept low and tube narrow - with too much mercury in the tube it would absorb UV photons, so they would not reach the phosphor layer.

But i saw in 24x enlarged spectroscopic photos of areas of the high pressure mercury (MB) lamp spectrum at the site http://ioannis.virtualcomposer2000.com/spectroscope/ both in B&W and color that there is a self absorption at both the blue line (Color and B&W photos) and the yellow doublet (B&W photo only) of the mercury spectrum at high pressure.

Really? my projector (DLP 800x600 from HP) has more of a regular metal halide spectrum. Only three colors are modulated through the color wheel, so there is no need for more than a red, green, and blue. So I would assume as long as it's a high color temperature, and as long as RGB wavelengths were present, CRI would be irrelevant. Laser TVs have three and only three wavelengths and they are generally heralded as being the most saturated and brilliant. What would be the advantage of a continuous spectrum?

Xytrell: All of the projectors that i had managed to check the spectra of their lamps contained UHP lamps which are most popular in small projectors because it is a much smaller point light source then MH and halogen and less expensive then MH. The spectrum that i saw from them contains extremely wide mercury atomic bands (Lines in regular MVs) and a very intense continuum background and they almost merged together. If the projectors would contains metal halide lamps, their spectra would look like the spectrum of the Osram HQI-BT daylight rare-earth (But much more intense and without the green line of thallium) which is not continuous.

DLP color wheel has three colors only on cheapest projectors. Three colors are not sufficient to provide saturated red or blue and at the same time keep reasonable efficacy: In order to allow displaying color appearance of deep red, the "R" has to be of the long wavelength tone, what mean you will nee huge amount of power to provide sufficient lumens of e.g. orange (as the eye is not as sensitive to deep red), while shorter wavelength red would stimulate the "green" eye cells as well, not allowing to display the deep saturated colors well.
So 4 color (R, G, B, W; mainly intended for office use, as there is no color improvement, they give highest lumens for given lamp)
Highest quality projectors use 5-color wheel (deep red, orange, green, greenish blue, deep blue) are used: Deep colors are used, when there is high signal only in "R" or "B" and zero in "G" (from deep color tone scene) and the "orange" and "greenish blue" are used for high brightness objects (as eye is highly sensitive to them, so they provide high lumens with quite low radiation power)

@dor123: The dip in the spectrum is so small, it might be even measurement error of the instrument. But even if it is reabsorbtion, it is very weak, corresponding to very low amount of mercury atoms in the state ready to absorb this wavelength. And as these atoms have to be already excited, their concentration would not grow with the arc load, so the effect would stay negligible.

You're right, I just checked it again and it is as you describe. I guess the first time I checked it, it wasn't completely warmed up yet.