It is known that the halogens have the following colours:
$\ce{F_2}$: Pale Yellow
$\ce{Cl_2}$: Greenish Yellow
$\ce{Br_2}$: Reddish Brown
$\ce{I_2}$: Violet
If we talk about predicting the colours based on their HOMO-LUMO transition energies, we may say that the $π^*_{np}$ - $σ^*_{np}$ gap decreases with increasing $n$, thus the wavelength of energy absorbed should increase and correspondingly the wavelength of emitted light should decrease as we go from $\ce{F_2}$ to $\ce{I_2}$.
However, while $\ce{F_2}$, $\ce{Cl_2}$ and $\ce{I_2}$ seem to adhere to this order, $\ce{Br_2}$ doesn't, showing up on the far end of the spectrum.
What is the reason behind this anomaly? Here's an illustration: