Exclude significant results from betadisper from adonis2?

[e_flat_minor (Jonas)]

Good morning,

I am trying to figure out the best way to use adonis2 on my dataset. I have three distances matrices (Bray Curtis, Jaccard and weighted Unifrac). I want to test four variables (A, B, C, D). Before feeding these variables into adonis2, I ran betadisper test to check the dispersion.
It turns out that the betadisper test for variable C gives a significant result.
My question now is: Is it forbidden to include variable see in my adonis2 formula? Or should variable C be excluded from adonis2 altogether?
Or should I switch vom adonis2 to anosim?

(I am using the adonis2 with "by margin" not "by terms" btw) Of course, adding or leaving out variable C changes the results from adonis2 (regardless if done "by margin" or "by terms"). Looking at my PCoA Plots I don't have a hard time leaving out variable C due to biological reasons/ extreme dispersions which is already visible to the naked eye in the PCoA plot.

This post here goes into the direction, but does not answer my question fully.

adonis, betadiver, and betadisp

An earlier post by @CarlyRae first introduced me to the fact that when examining beta diversity I need to consider not only the results of a multi or univariate anova, but also examining if the within-group distances to a centroid were equivalent among my groups. I'm wondering how to best interpret the situation where everything turns out to be significant. That is, if the results of both PERMANOVA and PERMDISP suggest significant differences, where do most people go next? I've run these tests b…

Thank you in advance!