First they said that fats were bad. Then they said it was carbs. Now, in mice at least, it's protein:

Eating too much protein can prevent an embryo attaching to the wall of the womb or hinder its early development, the findings suggest.

Although the research was done on mice, scientists believe there are implications for humans, especially people on Atkins-style diets.

"It’s conceivable that people who have protein intakes greater than 30 per cent may have problems conceiving," said Dr David Gardner, who led the American study.

The upper limit for protein consumption under Atkins guidelines is 35 per cent of total calories, but devout disciples of the diet might consume a greater proportion.

The research stems from the observation that protein in the diet affects levels of ammonium in the female reproductive tract. In herbivorous animals such as cows this has been known to cause reproductive problems.

Previous studies have shown that ammonium harms mouse embryos grown in the laboratory, causing genetic effects and retarding development.

Dr Gardner, the scientific director of the Colorado Centre for Reproductive Medicine in Englewood, set out to discover whether the same effects occurred in living mice.

The scientists found that ammonium levels in the oviduct, where the early embryo forms, were four times higher in mice given the high protein diet.

Four-day-old embryos in these animals had fewer cell numbers and a higher rate of cell death.

This is a crucial stage in embryonic development, since it is just before the embryo attaches to the inside of the womb. Without implantation, pregnancy cannot occur.

A total of 174 young embryos were transferred from both groups of mice to surrogate mothers fed a normal diet.

Despite the switch, only 65 per cent of those taken from high protein mice developed into a foetus. In contrast, embryos from mice on the lower protein diet had an 81 per cent success rate.

As the Atkins people point out, what happens in herbivore mice and cattle may not apply to humans. I certainly hope not. I look to fiber and protein as being the almost unquestionable goods in dietary habits. While I have no intentions of ever getting pregnant, there may be different effects for men. I drink a lot of milk. Milk has a lot of protein.

What I'm curious about - and what the article doesn't say - is whether or not protein has the same nutritional effects in herbivores as it does in carnies. If so then there's reason to believe that this problem could transfer over. If not, then trying to draw a parallel seems like a pretty pointless task.