I am planting hot magenta goo kernels on strong magenta gleam and 90% (approximately) of the time this results in a mutation to strong violet however occasionally I am getting a mutation to dark turquoise which seems highly improbable (impossible) so it could possibly be a bug.
This has been occurring periodically on rock also, however I can’t pinpoint exactly which shade of red/purple is doing this on rock as they are planted quite randomly I’m afraid.
I believe the darker colors are clumped closer together in the color space allowing for more possible mutations for dark colors 
Specifically in the little ‘tail’ of colors at the bottom
I’ve seen similar unlikely results following the same pattern where there is about a 1-2% yield in a very unexpected color. I want to say it was a very bright magenta from a very light blue or green.
My theory is that a kernel can mutate to the closest color to it and any other colors at equal distance to that. I think that some groupings in the color space are just closer than others, causing more possible mutations.
Yep fair point, but from my calculations there are at least 20 colours between dark turquoise and either strong violet, strong magenta or hot magenta and it seems to be occurring quite frequently as opposed to closer colours which haven’t appeared at all.
However from what you’re saying it tells me that this isn’t an impossible result so most likely not a bug then.
Thanks for the reply.
np at all! I’m super curious about it too if you find out anything else.
Yeah I wish I could actually keep track of all my results but my farm is just way too big and takes up far too much of my time as it is without keeping spreadsheets full of all my mutation results, haha.
Will feedback if I find anymore out though for sure.
I’d say working as intended from the topics I read about goo mutations when it was first introduced. Here’s a quote to back up my theory as well.
I started doing this and it lasted about 10 generations. It was way too much work.
There’s also a lot of good info on that thread I posted in my comment above 
True, I don’t even know if the graphs on the CIELAB wiki page are the same ones being used by the devs
Here is the table that has been confirmed by devs that is correct when it comes to goo mutations. Thanks to @Firehazurd for their hard work on putting it together
Yep that is a phenomenal resource. I use it extensively to guide my mutation and mixing efforts.
Big thanks to @Firehazurd for creating it.
Yeah there is 20 % chance to get a random mutation in any direction in the color graph.
