>On 24 Jan 2000 14:37:34 -0800, [EMAIL PROTECTED] (Drake R.
>Bradley) wrote:
>> It is not clear to me why the poster's question is nonsense. He seems to be
>> asking whether the outcome of successive births (boy/girl) are independent
>> or not. < ... >
In article <[EMAIL PROTECTED]>,
Rich Ulrich <[EMAIL PROTECTED]> wrote:
> - I suppose I was taking the question more literally than you are.
>He was not asking if the births are independent (which they are). He
>was asking if he *might* be remembering some claims, or if anyone had
>information, or if it was complete nonsense.
>
>I don't have references; but I remember classroom exercises of obscure
>tests to check on any *conceivable* dependency -- it is not like
>nobody has every collected this kind of data.
Failure to reject the null hypothesis is not the same thing as proving
that it is true. General biological knowledge is enough to pretty
much guarantee that some dependency exists, though it might well be so
small as to be undetectable.
Here's one possible scenario: Imagine that a man carries a highly
detrimental gene on his Y chromosome (which, however, must obviously
not be so detrimental as to be invariably fatal). This gene could
cause high mortality for male embryos that he fathers, leading to most
of his children being female.
Another possibility: The man's X chromosome could carry a "meiotic
drive" gene which (a) causes sperm carrying it produces a toxin, and
(b) conferes immunity to this toxin on the sperm carrying it. The
result is that the sperm carrying the X chromosome kill off the sperm
carrying Y chromosome, leading to all his children being female. Such
meiotic drive genes have been observed in various animals, though not
yet in humans, to my knowledge.
Radford Neal
===========================================================================
This list is open to everyone. Occasionally, people lacking respect
for other members of the list send messages that are inappropriate
or unrelated to the list's discussion topics. Please just delete the
offensive email.
For information concerning the list, please see the following web page:
http://jse.stat.ncsu.edu/
===========================================================================
