This leveling effect of evolution resolves one problem and raises another.  If all species are forced into I, barring a few which are evolving rapidly in that direction, then it would be reasonable to expect that within an order of magnitude, all population sizes should be identical, although some forms might follow a complex-low population size pattern and some a simple-high population size pattern.  3,000 species across a variety of taxa were studied by Bazin et al (Population Size Does Not Influence Mitochondrial Genetic Diversity in Animals.  Eris Bazin, Sylvain Glémin, Nicolas Galtier SCIENCE VOL: 312 570 28 APRIL 2006.)  Although they reckoned that real population sizes varied over several orders of magnitude, overall they found little (factor of about 10 and simpler seeming forms with larger populations) variation in nuclear DNA over multiple taxa (bottom line), and none at all in mitochondrial DNA (upper line).  They suggest multiple selective sweeps; we suspect multiple local extinctions.

Population Size Does Not Influence Mitochondrial Genetic Diversity in Animals.  Eris Bazin, Sylvain Glémin, Nicolas Galtier SCIENCE VOL: 312 570 28 APRIL 2006.

Unstable Population

The question arises whether a population permitted to expand without limit will always grow indefinitely, run the risk of extinction or stabilize at some level.  We examine here 10 simulations with a maximum mating population of 200 and see if they survive for 1,000 generations, recording the number of offspring in the final generation compared with 10 simulations that are identical except that the maximum number is permitted to rise to 20,000.  If the population dies, the number of generations it lasted is shown.  The other parameters are as on page IV.

It would appear that the larger population regularly goes extinct and after a characteristic number of generations, in this case near some low multiple of slightly over 80.

Graph of Birth Curves at 200 and 20,000 Maximum

This was the rising and falling of the birth rate in the population of maximum 20,000 that lasted 170 generations.  There does seem to be a periodicity to the track. 

This for comparison is the history of the population with maximum 200 that had 288 offspring in generation 1,000.  It is more stable than the larger population, but it is also noisier.  In our experience this size of population would rarely go extinct.  One would expect a very small population, because of increased stochastic chatter, to go extinct more frequently. 

Strategies for Limiting Effective Population Size

If, indeed, most species have genomes that are competitive when the effective population size is small, but go extinct when the effective population size is too large, then we should expect to see enormous energy committed to limiting gene pool size.  And we do. 

• Gregarious animals such as deer typically have a rutting season in which a small minority of the males sire the offspring. 
• Social insects confine reproduction to a few individuals. 
• Salmon return to their natal stream to mate and spawn.
• Migrating birds stay in their flocks. 
• Coral reproduce only locally.
• Animals tend to be territorial. 
• Some bats mate by having sisters all choose the same male (Same Bat Family.  NATURE: 437 xi 15 September 2005). 
• Some populations – notably lemmings – tend to surge and collapse. 
• Even diatoms floating in the sea have identifiable local populations (Dispersal Limitations Matter for Microbial Morphospecies.  Richard J. Telford, Vigdis Vandvik and H. J. B. Birks SCIENCE: 312 1015 19 May 2006). 
• There are increasing numbers of cryptic species discovered (Body Doubles.  Alberto G. Sáez and Encarnación Lozano NATURE: 433 111 13 January 2005), species that have not undergone detectable evolution, but have undergone speciation and remain that way, even though with no selective advantage between them, one should eventually wipe the other out. 
• Sympatric speciation occurs (Evolution Standing in Place.  Elizabeth Pennisi SCIENCE: 322: 1372 10 March 2006). 
• It may be that tropical heat increases the mutation rate in tropical trees, accelerating evolution (A Heated Debate.  ECONOMIST: 379 no 8476 81 6 May 2006), but the fact remains that old tropical forests have a greater variety of trees than young northern ones, so the number of species increases with time, although in theory they should diminish (Paradox of the Clumps.  Sean Nee and Nick Colegrave NATURE: 441 417 25 May 2006).

All of these phenomena have the result of reducing the size of the mating pool. 

Wet Lab
It should be possible to put the principle to a direct test.  Choose guppies or fruit flies or a suitable animal and put them in a circle of fishbowls numbered, say, 1 through 32, a male and a female in each.  In each generation, the parents and most of the offspring are removed, one female offspring stays in the same bowl, one male is moved.  The male offspring in bowl 1 moves to bowl 2 in the next generation, 4 in the next, 16, 32 and 1 again to start over.  A male from bowl 2 moves to 3, 5, 17, 1 and 2 again.  It should be possible to continue this indefinitely, each successive parent separated from its mate by 4 generations.  Using circles of 1 bowel, 2, 4 and so forth to perhaps 4096, it should be possible to demonstrate fertility depression both from too small a population and from too large a population. 

There have been 310 visitors so far.

Home page.