Basically the cone/rod distribution over the retina isn't constant but, thanks to selection, has a ratio in the center of the retina, and hence your vision, that mixes color perception and magnitude perception. Around the edges, the ratio changes to favor magnitude perception, so you can detect movement in peripheral vision. One of the essentials of observing N-rays was to not quite look at them directly, and Collins or his source supposed that most likely, the same light might be perceived as brighter as it falls off-center in the eye. Innnnteresting.
Re: To digress a step further...