Do you have a source for this that isn't, um, from the Simpsons? Geneticists have an equation for estimating the response to selection, the change in IQ from parent to offspring, and it uses an average of both parents' IQ. See http://en.wikipedia.org/wiki/Heritability#Response_to_Select... for the equation.
IQ is somewhat more malleable initially, but it seems to converge on a point set by one's genetics, by adulthood. The APA estimated heritability in childhood at .45 and .75 by adulthood.
My Simpsons reference was meant to be a colorful analogy, and not a "source." Sheesh.
The source for my statement was journal article I read a few months back. I can't seem to track it down online, but here's a blog that summarizes the studies:
http://flatrock.org.nz/topics/men/thank_mom_for_brains.htm
Like I said, the jury is still out on this. This is a relatively new hypothesis, and not a long-standing theory.
Here is another link: http://discovermagazine.com/2005/oct/sex
There are more genes related to intelligence on the X chromosome than on just the X and Y chromosomes. There are still 22 other pairs of chromosomes, which in offspring, are a mishmash of the father and mother's chromosomes. Genes on these chromosomes have been experimentally linked with intelligence (eg, http://genepi.qimr.edu.au/contents/p/staff/CV453.pdf)
Plus, there are sex differences in intelligence; there are more male dullards and geniuses. Men score higher on visuospatial ability, and women on verbal ability. Given that these are distinctly sex related, their genesis lies in differences between the X and Y chromosomes.
So all those uber-rich guys who select stereotypically dumb, hot trophy wives are doing their offspring no favors in this department. (On the flipside, they may be doing their offspring plenty of favors w/r/t looks, and countless studies have shown that looks are also highly correlated with professional success).
Finally, IQ is, for lack of a better word, malleable. It's not locked in at birth. It can be trained, and it can also wither on the vine if not properly used and challenged. And it's highly susceptible to environment: types of attention received as a child, education, socialization, and even environmental pollutants all play a big role. And they continue to play a big role throughout life, up until roughly the mid-20s, when the brain stops undergoing its rapid changes and loses a lot of its plasticity (though not all of it).
[1] This is something I think science should call the "Lisa Simpson Effect," inasmuch as it explains how a dolt like Homer and a smartie like Marge could produce a genius child.