Lynn Margulis was one of those scientists who's theory was mocked and derided for a long time until it was proven. It's a fascinating look at a theory of convergent evolution as opposed to traditional Darwinian divergent evolution. Vindication aside, the bitterness of the author is very, very, apparent in this intruiging book. (did i mention bitter?) However, this is one of the few popular evolution books that deals with microbiology,(also one of my favourites) and I agree with the author's critique that, like in physics, starting at the bottom with the microbes & viruses & DNA is the best place to look for the mechanisms of evolution. ( )

This is an immensely fascinating, and immensely frustrating book.

It expands substantially on the standard Margulis/Sagan spiel, that eukaryote organelles (mitochondria and chloroplasts) derive from bacteria that were ingested, but not digested, by other bacteria, to make the much bolder claim that speciation is driven by this process of one genome acquiring another genome whole.

The fascinating part of the book is the many examples of absolutely bizarre symbioses and so on that I'd never have dreamed of, including things like
* worms that incorporate algae into their skin and live via photosynthesis,
* tubeworms that incorporate archae and live off their byproducts, * echinoderms can fertilize sea squirt (chordate!) eggs to give rise to viable offspring, some of which develop along the male parent pathways, some along the female parent pathways.

So what's frustrating?

Firstly the book is an angry polemic, repeatedly striking out at a parody of neodarwinism that looks grossly simplified, even to me. Half its agenda seems to be to tell us about some very interesting biology, while the other half is to tell us that all other biologists are fools. For example it repeatedly states the (obvious) point that SNPs are probably not powerful enough to generate the diversity we see around us, but never discusses the facts that most diversity is actually believed by most biologists to be generated through crossing over of chromosomes along with diploidy, ie through sex; neither does it deal with the fact that while end-protein SNPs are limited in what they can do, SNPs in regulatory proteins can have profound effects on development.

Secondly the book leaves too much unsaid. Just as we get to really fascinating details, like how exactly the nucleus of eukaryotes could have evolved from the fusion of a eubacterium with an archae, the discussion kinda peters out. The really interesting and unexplained stuff (at least to my mind) like quite how the absorbed archae lost all its cytoplasm (but became the nucleus), and what happened to the DNA of the host eubacterium.

Thirdly, I have to wonder the extent to which the authors overstate their case. The examples they give are quite fascinating; the idea, that the eukaryotic nucleus, mitochondria and choloroplasts are ingested genomes is full of promise. Even some of the other ideas they give, that larval forms of marine life derive from a different genome from the adult form seem promising. But to leap from this to the uneqivocal statement that genome acquisition is the ONLY driver of speciation seems an awful leap, and a more satisfactory book would list and discuss the qualms of those who disagree with them, rather than simply describing them all as unscientific know-nothings, (a description that seems a bit rich coming from authors who have discussed the existence of free-will in bacteria). ( )