Sex ratios by ucblockhead (4.00 / 1) #2 Thu Apr 24, 2008 at 02:43:19 PM EST
Sex ratios of parents are meaningless as its already well known that the sex ratio at birth is not exactly 50%. There are other animals that have ratios other than the human 51% male and there are lots of good examples of sex ratios changing in response to environmental changes in the animal world. Certainly unless a species is monogamous and mates for life (and humans are not in this class) there is absolutely no reason the sex ratio has to be 50-50.

In regards to famines:

  • It is not clear to me that most famines happen in conditions where'd people would notice the difference between 43% boys and 51% boys. It is not clear to me that anyone has actually thought to check this.
  • Famines tend to suppress births in general.
  • Famines tend to mostly hit the poor, who tend not to tracked well by those in power.
  • Famines tend to last a year or two, therefore the effect would only effect a fraction of a woman's breeding cycle. (Especially important in conjunction with the second bullet.)

----
ウセーバラケダ


Elephant Seals and Dutchwomen by TheophileEscargot (2.00 / 0) #3 Thu Apr 24, 2008 at 03:13:33 PM EST
Regarding famines, a couple of diaries ago I mentioned the genetic consquences of the Dutch hunger winter of WW2. Children born then, and the children of those children, had thrifty metabolisms and were more prone to diabetes.

A 59% male ratio is a huge demographic shift. I really can't believe that they wouldn't have noticed in that incident alone. I suspect it would also have shown up in parish records from the Irish potato famine. Keeping track of births doesn't really take high tech databases: it takes a lot longer to produce a baby than to write out it's birth certificate in longhand.

Regarding sex ratios, the classic example is the elephant seal: a highly polygamous species where a male maintains a harem of up to 100 females. People often intuitively expect that therefore an elephant seal will produce more female offspring than male. In fact though, they produce a 50-50 ratio, even though most of the males will never reproduce.

The reason is that every offspring has exactly one male parent and one female parent. The reduced number of males who reproduce is therefore exactly compensated by the increased success of those that do manage it. It's the one-mummy one-daddy thing that determines sex ratios, and that means there's a strong evolutionary pressure to keep the numbers pretty even.

If things worked the way the study says and you had a hard few years: maybe due to famine, maybe the local wildebeest have died off, maybe another tribe's forced you off the good land, then you'd have a generation with a skewed sex balance, producing fewer offspring in the subsequent generation. (That's not a group selectionist argument: producing a kid the same sex as everyone else's kid hurts the individual as well as the group).

This supposed effect just doesn't make sense to me.
--
"Everything is vague to a degree you do not realize till you have tried to make it precise." -- Bertrand Russell
[ Parent ]

skewed sex balance by ucblockhead (4.00 / 1) #5 Thu Apr 24, 2008 at 03:25:54 PM EST
If you have a skewed sex balance in which females are overrepresented, you don't necessarily produce fewer offspring in the subsequent generation.

Remember that the evolutionary pressure you talk about is to produce an even sex ratio at breeding age. That's not the same as producing an even sex ratio at birth.

The other thing to consider is that the study you labeled bad science was looking at slight changes in diet, not starvation. It is very possible that the effect goes away when people are literally starving. There's a huge difference, biologically, between skipping breakfast and going five months without enough food.
----
ウセーバラケダ
[ Parent ]

Well by TheophileEscargot (2.00 / 0) #6 Thu Apr 24, 2008 at 03:42:24 PM EST
If you have a skewed sex balance in which females are overrepresented, you don't necessarily produce fewer offspring in the subsequent generation.
However, the study claims that some diets will produce 59% boys.

They claim that the reason behind is is that boys are more expensive to produce than girls. If so, the effect ought to be more pronounced the greater the degree of famine.

The study was of 740 women. Not a tiny sample, but a few hours with Excel can easily produce as many statistically significant correlations from random data as you can be bothered to generate. Meanwhile, famines like Dutch hunger winter, the Soviet famines, the Irish potato famine affect hundreds of thousands, maybe millions. I'm betting on the bigger samples.
--
"Everything is vague to a degree you do not realize till you have tried to make it precise." -- Bertrand Russell
[ Parent ]

"ought" by ucblockhead (4.00 / 1) #7 Thu Apr 24, 2008 at 03:44:40 PM EST
It is very possible that during small shortfalls, it is a better strategy to produce fewer boys while in real famines, it is a better strategy simply not to breed at all.
----
ウセーバラケダ
[ Parent ]

In a parallel universe, perhaps by TheophileEscargot (2.00 / 0) #8 Thu Apr 24, 2008 at 03:49:57 PM EST
But as the thrifty-metabolism offspring of the Dutch hunger winter show, in this universe the human strategy is to produce offspring who are better adapted to live on fewer calories.
--
"Everything is vague to a degree you do not realize till you have tried to make it precise." -- Bertrand Russell
[ Parent ]

but by ucblockhead (4.00 / 1) #9 Thu Apr 24, 2008 at 03:59:15 PM EST
The one does not preclude the other. The effect off famine on fertility is very well known.
----
ウセーバラケダ
[ Parent ]

Yes it does by TheophileEscargot (2.00 / 0) #10 Thu Apr 24, 2008 at 04:04:55 PM EST
A famine doesn't starve everyone to exactly the same extent. Some people have enough to eat. Some are a little short of food. Some are literally starving.

In a famine, at whatever level of starvation this effects shows up at, there would be some people at this level, which means the effect would show up demographically.
--
"Everything is vague to a degree you do not realize till you have tried to make it precise." -- Bertrand Russell
[ Parent ]

Starving to the same extent by ucblockhead (4.00 / 1) #11 Thu Apr 24, 2008 at 04:18:34 PM EST
If people aren't starving to the same extent, then the effect wouldn't be 59%...
----
ウセーバラケダ
[ Parent ]

No by TheophileEscargot (2.00 / 0) #12 Thu Apr 24, 2008 at 04:31:23 PM EST
But if you're narrowing the calorie range at which this effect shows, it's even more puzzling that it happened showed up so strongly in the small sample of the study.

Remember this effect allegedly exists to help women produce smaller girl offspring when short of calories. Yet somehow in a situation when people are genuinely short of calories, it's so small as to be statistically indetectable with a sample size in the hundreds of thousands. Yet it shows up in a sample of hundreds when considering prosperous, well-fed mothers in the developed world. And it doesn't affect the sex ratios born to the overweight and obese: that would surely have shown up by now too.

It doesn't make sense: an economy measure that only shows up when there's a miniscule deviation from a healthy food intake, disappearing completely whenever there's a significant calorie surplus or deficit.
--
"Everything is vague to a degree you do not realize till you have tried to make it precise." -- Bertrand Russell
[ Parent ]

Maybe by ucblockhead (4.00 / 1) #14 Thu Apr 24, 2008 at 05:59:13 PM EST
I still think you are far too quick to call this "bad science" based on historical data. It seems to me like calling the life-extension effect of calorie restriction a crock because people don't live longer after famines.
----
ウセーバラケダ
[ Parent ]

Well, that's probably wrong for similar reasons by TheophileEscargot (2.00 / 0) #15 Fri Apr 25, 2008 at 03:19:47 AM EST
A single famine might not make a difference, but if there was a significant life-extension effect it would show up in the demographics. You could look at the developing world, Marx-Leninist economies (there's some interesting data from Cuba), isolated and religious communities and so on.

If there is any effect in humans, it must be pretty tiny to have not shown up by now.

It works in rodents of course. Rodents are highly optimized to grow fast, pump out a load of babies and expire, so there's plenty of room for life-extension trade-offs. But humans take about 13 years to get to reproductive age, and 25 before the brain fully matures. Having invested all that time maturing, evolution needs us to take advantage of it. It's a pretty fair bet that we're already got all the easy life-extension genetic switches and mechanisms turned all already.
--
"Everything is vague to a degree you do not realize till you have tried to make it precise." -- Bertrand Russell
[ Parent ]

Not just rodents by ucblockhead (4.00 / 1) #17 Fri Apr 25, 2008 at 11:51:32 AM EST
The effect has shown up in every animal from tapeworms to rhesus monkeys that have been studied and is pretty much accepted by the biological community as most likely existing in humans.

See here.
----
ウセーバラケダ
[ Parent ]

I think that article by R Mutt (2.00 / 0) #18 Fri Apr 25, 2008 at 12:34:25 PM EST
Illustrates Wikipedia's tendency to be provide evenly balanced arguments when the actual evidence is heavily weighted one way.

The "Effects on humans" section basically just talks about the normal ill-effects of being overweight. When it does cite actual scientific papers about calorie-reduced life-extension in humans it's papers like this: "Why dietary restriction substantially increases longevity in animal models but won't in humans."

That paper bears out the earlier point: there's a trade-off between fast reproduction and longevity. Humans have already traded off for that: we reproduce very slowly and live very long lives. Any life-extension effects of this in humans are likely to be noexistent or very small.

[ Parent ]

When I'll have time by ucblockhead (4.00 / 1) #19 Fri Apr 25, 2008 at 12:55:09 PM EST
I'll dig up more concrete stuff. It really is pretty well accepted these days.
----
ウセーバラケダ
[ Parent ]

The mechanism I'd heard was different. by ambrosen (4.00 / 1) #20 Sun Apr 27, 2008 at 06:34:17 PM EST
The mechanism that I'd heard, and strikes me as plausible, is that the male offspring have a much greater variance in the number of children that they have, so if your situation is good relative to the social background, then male children are the best, cause they'll all go round getting lots random women pregnant, but if you're at a low social point, far better to have daughters, and they'll all have one or two children with the cads from the posh family, and all will be well.

But this story seems a bit different. I'd like to see a reconciliation of the one I'd heard (a relative wealth situation) and the one presented in those articles.

[ Parent ]

I have the same objection by TheophileEscargot (2.00 / 0) #21 Mon Apr 28, 2008 at 03:25:21 AM EST
If there's one thing that historians have an Imperial Fornication-Ton of data on, it's aristocratic genealogy. If this effect is real and significant, it's the easiest thing in the world to prove. Look at the records of posh familes (Kings of England is a good place to start) and just see how many male and female children there are.

If you look up what they call Pathological Science or Voodoo Science, not all of the features but a disturbing number of them seem to apply to sociobiology.

1. Discoverers make their claims directly to the popular media, rather than to fellow scientists.

2. Discoverers claim that a conspiracy has tried to suppress the discovery.

3. The claimed effect appears so weak that observers can hardly distinguish it from noise. No amount of further work increases the signal.

4. Anecdotal evidence is used to back up the claim.

3 is the critical one. For 2, it's "political correctness" rather than a conspiracy which is assumed to be covering things up.

Anthropologists, historians and social scientist have spent an awful lot of time and effort painstakingly eating grubs with remote tribes, spending lifetimes analysing gravestones and parish records and so on; and applying the incredibly sensitive pattern-recognition system of the human brain to the resulting data.

I wish biologists who venture into their realm would start looking at their data, instead of just assuming that as mere social scientists they're just too thick to have done anything worthwhile.
--
"Everything is vague to a degree you do not realize till you have tried to make it precise." -- Bertrand Russell
[ Parent ]

All good points. by ambrosen (4.00 / 1) #22 Mon Apr 28, 2008 at 07:50:25 AM EST
I concede all of them.

[ Parent ]

Login
Make a new account
Username:
Password: