Archiver > GENEALOGY-DNA > 2010-03 > 1268611757

From: Steven Bird <>
Subject: [DNA] Percentages of males and females who reproduce - compare tothe "one third rule"
Date: Sun, 14 Mar 2010 20:09:17 -0400

Dear listers,

I posted to the list some time ago concerning the "one-third rule" of Cavalli-Sforza concerning the effective population size compared to the census size for a given population. In short, C-S stated that about 1/3 of any given census population [N(c)] is of child bearing age at any given time. This rule was appropriated (incorrectly, in my opinion) by Zhivotovsky, Underhill and Feldman in 2004 and 2006 to justify a reduction in the germ line mutation rate from .002 per generation to .0069 (the so called "evolutionarily effective mutation rate". I stated at the time that this appeared to be a misapplication of the rule, since any reduction of genetic diversity [the "effective population size", N(e), compared to N(c)] would have to be a direct consequence of the overall reproductive rate, not simply the number of people reproducing in a given census year.

I found an interesting report by the U.S. Department of Health and Human Services which stated that 84% of men and 86% of women in the United States have children eventually at some point in their lives. This does not reflect the number of children had by each person, but rather whether any children were had at all. The average number of children per family at present in the U.S. is 1.86, according to the Census Bureau in 2004.

This has really started me thinking about the basis for the 1/3 rule. Since reproduction in humans is not synchronous (obviously, we don't all have children at the same time each generation), the number of people reproducing at a given time is largely irrelevant. The total number who leave offspring, and the average number of offspring per pair of parents, OTOH, would appear to be directly relevant to any reduction in genetic diversity within a given population or haplogroup. More people having offspring and fewer children per couple would apparently lead to higher genetic diversity; fewer people having offspring and/or more children per couple would lead to reduced genetic diversity. (I am speaking only of the non-recombinant DNA here; Y and mtDNA.)

It's apparent that the number of children per couple was higher in the past and is a function of population (demographic) history. I wonder, however, if the 84% and 86% levels for men and women might not be pretty close to the mark overall. One thing is for sure, these numbers don't seem to represent anything close to a 1/3 rule for past populations.

I'm trying to think through the consequences of this information and would appreciate any suggestions from the list. Comments are welcome.

Steve Bird

The New Busy is not the old busy. Search, chat and e-mail from your inbox.

This thread: