You're assuming that the set of genes that produce multicellular life are largely disjoint from those that produce unicellular life, and that is unlikely to be the case. Much more likely, multicellular life is an incremental set of changes layered on top of unicellular life, and so the only thing required to revert from multicellular back to unicellular is a breakdown in that higher-layer functionality.

If you're not already familiar with them, you might want to look up HeLa cells:

https://en.wikipedia.org/wiki/HeLa

> You're assuming that the set of genes that produce multicellular life are largely disjoint from those that produce unicellular life,

I assume no such thing.

I've read (but can't cite) that roughly 90% of our genes are related to construction and maintenance of the cell itself, and only about 10% are related to higher order organization. Just as it has been said that something like humans and tomatoes have 50% of our genomes in common, there is a significant but smaller percent that we have in common with single-celled life forms.

> HeLa

I read the book years ago.

> I assume no such thing.

You certainly do:

"This theory, from the way you've stated it, would require preservation of inactive genes for hundreds of millions of years, only to spring back to action in cancer cells."

This is only true if the genes for being single-celled are inactive in multi-celled creatures and would need to be re-actived to revert to being single-celled. But this is not the case. Milti-cellularity is a functional layer built on top of single-cellularity, so to revert to being single-celled you have to deactivate currently active genes, not activate inactive ones.