Refractive index as a function of concentration of sugar solution and wavelength of light We recently performed an experiment with the idea to find refractive index of medium (water) as a function of wavelength of light. We then added some sugar to see how the refractive index changes with concentration of sugar solution. We got the following graphs.


Are the relationships actually linear? Or are these just limiting cases? 
Can someone shed some intuition on why the graph is in the way it is?  
 A: For the dependence on the sugar concentration you do expect from first principles (at least at low saturations) that the dependence will be linear, since the electric susceptibility $\chi$ is essentially the molecular polarizability $\alpha$ of each contributing species times the number density of said species.

As far as the wavelength dependence goes, it's not fully linear but it's a good approximation over that range. (And, as I mentioned in the comments, your data looks vaguely linear but once you put in the error bars and do a full uncertainty analysis, you're likely to find that it's much more consistent with a variety of non-linear behaviours than it looks from your graph.) This website has some reasonable-looking data, and this is backed up in the literature:

Refractive Index of Water and Its Dependence on Wavelength, Temperature, and Density. I. Thormählen, J. Straub, and U. Grigull. J. Phys. Chem. Ref. Data 14, 933 (1985), NIST eprint.

In particular, if you look for wavelengths just a bit above and below that range, you get something much more illuminating,

i.e. the downward slant comes because the visible range is sandwiched between two resonances, which themselves can be well described using the Lorentz oscillator model.
