Dogs (canis lupus familiaris) are dichromats, meaning that their eyes contain two types of cone, responding maximally to two different wavelengths of light, namely blue (429 nanometres) and yellowish green (555 nm). (Source: Jay Neitz, Timothy Geist, and Gerald H Jacobs, "Color Vision in the Dog", Visual Neuroscience 3 (1989), pp.119-25.) (Humans are trichromats.)
Visible light for humans has a wavelength between 380-400 nm and 740-780 nm. (Different sources give different figures.)
What I would like to know is what are some recent estimates for the shortest and longest wavelengths visible by dogs.
Extrapolating from the graph in the above-mentioned 1989 source gives a band from around 360 nm to 680 nm, which would mean that at the blue end dogs can see wavelengths of ultraviolet that humans cannot. Indeed they may be able to see wavelengths that are considerably shorter than 360 nm. I have found an article from 2014 (R H Douglas and G Jeffery "The spectral transmission of ocular media suggests ultraviolet sensitivity is widespread among mammals", Proceedings of the Royal Society B 281 (1780)(April 2014)) in which the authors report their finding that dogs' lenses allow some ultraviolet through (50% of it at 335 nm) and note that mammals' cones have a secondary maximum in the ultraviolet band, the inference being that mammals such as dogs whose lenses do allow ultraviolet through can therefore see ultraviolet well into what for humans is the invisible short-wavelength part of the electromagnetic spectrum. But unfortunately they do not give an estimate for the shortest visible wavelength.