Re: SCI:BIO: maestro: more on organism complexity

Hal Finney (
Wed, 15 Jan 1997 15:51:42 -0800

From: Eugene Leitl <>
> Stuart A. Kauffman, "The Origins of Order", Oxford University Press
> (1993), p 34, 461.
> [...]
> Among the most obvious features of cell differentiation is that less
> complex organisms possess fewer cell types than more complex organisms.
> For example, yeast has three cell types, the coelenterate hydra has 15 to
> 17, annelid worms have about 60, and humans about 250. A plot of the
> number of cell types in an organism as a function of the estimated number
> of genes in that organism shows a profound and simple relation: The
> number of cell types increases as about a square-root function of the
> number of the genes [...].

That's a very interesting result. Later quotes suggest though that it
could be up to a linear function of the genes, which would be less
surprising. A priori there is nothing too astonishing if the number of
genes is proportional to the number of cell types.

I've been reading some Stephen Jay Gould since our discussion here
about his theories. I lost the original reference, so I have chosen
to start with "Wonderful Life", a book about early Cambrian organisms.
The main theme of the book is that these early animals were in some ways
more diverse than in the present day, with many body plans which don't
exist any more. I'm not sure though how that fits in with what was said
here about complexity expanding like a random walk.

I'm about halfway through the book and I've learned a whole lot more
invertebrate anatomy than I ever wanted to. It seems that Gould's
measure of diversity mostly relates to how the antennas and legs attach
to the various body segments in these creatures. It turns out that among
invertebrates today (the vast majority of animals), only a few specific
patterns are observed. Back in the early Cambrian era (shortly after the
appearance of multicellular life) there were many more arrangements of
limbs. So in a sense we have lost diversity, although the specific
patterns which have been retained have still branched into a wide range of
animal forms.

The main lesson I have taken from the book so far is that, despite what
Gould implies, defining and measuring complexity and diversity is a
difficult problem. I don't find his measures particularly convincing,
especially given the experimental difficulty even in figuring out what
these animals originally looked like.

The big puzzle is why it is that evolution has not been able to
depart from the established body patterns once they were established.
Earlier animals were able to evolve into wildly different shapes. Later,
they seem to have lost that ability.

Is there any significance to the fact that the survivors are ones which
have fixed forms? Gould seems to think that it was random which body
shapes survived. Does he assume that all of these Cambrian animals
would have led to descendants who were equally fixed in shape? Or is
it possible that somehow shape constancy was correlated with survival,
and that creatures which retained complete structural mutability (over
millions of years) somehow were selected against?

Maybe the last part of the book will say. It looks though more like a
polemic against the attitudes of 19th century paleontologists. Gould
really comes off as arrogant when he gets into this philosophical stuff.
Is it true that he's a dyed-in-the-wool Marxist? I thought I read that