I like the way Figure 1 presets the data with galaxy level phi in one color and are level phi in another. Figure 2 should do the same.
Legend in Figure 1 top overlaps a line label, to avoid this (and for constancy) all legends should be moved to the lower right.
Might be useful to spell out what the null hypothesis is in the last sentence of the caption: "Moving rightwards on the x-axis implies greater confidence in rejecting the
null hypothesis that the compared values were drawn from the same parent distribution" or similar.
I know it is kind of stated in the first sentence of figure 1 caption but it can be helpful to say it multiple times as these figures can be a bit hard to wrap your head around the first time.