The hypothesis here is that it is not the father that determines the gender of the baby.
In the facts given in the article I found nothing convincing that the gender selectors are females i.e. they conclude that they are.
Now I haven't studied seriously biology but to change the meiosis of sperm cells to skew the odds seems more effective (and simple) than to have the egg cell select the sperm with the X or Y chromosome during the fertilization that is messy as it is. And if X and Y sperm cells were somehow different I think we may have found out.
That's not a meaningful sense. Obviously it's a question of whether a sperm cell carrying an X or Y chromosome gets into the egg, but both parties have a lot of biology involved between meiosis and fertilization that might impact that.
> Now I haven't studied seriously biology but to change the meiosis of sperm cells to skew the odds seems more effective (and simple) than to have the egg cell select the sperm with the X or Y chromosome during the fertilization that is messy as it is. And if X and Y sperm cells were somehow different I think we may have found out.
X and Y sperm cells are different - one is carrying an X and the other is carrying a Y and these will therefore have different masses and probably some other characteristics will differ. As female mammals tend to invest hugely more resources in the individual child, it's more likely to be worthwhile to them to sacrifice some chance of conception to favor a particular gender. Furthermore, if the child is more likely to stay with the female, the female is a better predictor of the environment the child will be raised in. Since meiosis produces an X and Y cell inherently (it splits a cell with an X+Y in half, with one chromosome going each way) and furthermore happens well in advance of mating when the partner may not even be determined, it seems less likely that the cause for any pattern in SR would be manipulation of the ratio of X vs Y cells during meiosis. Of course, biology surprises on occasion, I just think you're making entirely unwarranted assumptions.
Evolution doesn't work like an engineer. If you or I were going to sit down and say, "Hmm, how do we control the sex ratio from within the female's body?" we'd come up with things like a mechanism to differentiate the sperm and then try to apply that to a mechanism to block or help one group or the other, because modularity is good and such.
Evolution doesn't pay attention to the mechanism, just the result. Maybe it happens that one type of sperm is disproportionately affected by acidity levels or viscosity of cervical mucous or any of potentially hundreds of attributes of the environment the sperm will be encountering between meiosis and conception, and that's assuming any differentiation happens pre-conception. If there is such an effect, there's a good chance evolution will "spot it" and exploit it where the gain outweighs the cost; the female has (evolutionarily speaking) far more motivation to make use of it, if it involves any tradeoff with chance of conception.
Note that by his hypothesis, smaller litters or longer time between pregnancies probably is what you'd see if the decision was being made in the female's body, whether pre- or post-conception.
In particular, the statement "men who have higher testosterone levels are more likely to produce male offspring" is pretty much equivalent to the statement "women having sex with men who have higher testosterone levels are more likely to produce male offspring", unless it was measured by impregnating women in a double-blind setting. If it was based on conceptions from a sperm bank, it might be better evidence, though my understanding is they still generally report some information to the women about potential fathers and give her some choice in the matter (but I could easily be mistaken, there).