Daguerreotypes had a metal base, typically copper, which had a silver coating on top, if I remember correctly. The silver was immersed in halide fumes, typically iodine, forming light-sensitive silver halide crystals on the surface of the photograph. This was next placed in the camera and exposed to the image produced by the lens. This exposure could be as long as 15 minutes.

To develop the daguerreotype the exposed plate was placed in a dark box over a dish of liquid mercury which was heated by an alcohol lamp or a candle. Heated mercury fumes would bombard the plate. Wherever the plate was exposed to light the silver halide was converted to pure silver metal. The mercury atoms would stick to the pure silver metal and form an amalgam, (this process was known since the Middle Ages). The amalgam was lighter in color than the unexposed silver halide, so wherever light struck the plate it would appear lighter in color than the unexposed parts. Therefore daguerreotypes were a direct positive process which usually produced a mirror image.

Considering that the image layer of a daguerreotype contains silver, silver halide, mercury, mercury silver amalgam and copper it's a wonder that more daguerreotypes do NOT tarnish. There are so many chemically reactive metals touching each other it is hard to see how the system could be stable.