In birds, the gender of the off-spring is determined by the female (contrary to for instance mammals, where it is determined by the male), and in a normal situation the chance of the embryo to be male or female is 50/50. That means that when we have a large number of chickens hatching and we sex them, we expect that approximately 50% of them will be females and 50% will be males.
However, when we sex the birds for instance for parent stock production, we sometimes see that the distribution of males and females is not completely 50/50 but somewhat skew, for instance 47/53. When we produce broiler parents, we sometimes see that the percentage of females is slightly higher than the percentage of males. In layer parents, it seems to be the other way around, relative more males are hatching than females. This can be due to mistakes in the sexing accuracy, but also when the sexing is done very accurate we sometimes can observe this phenomena
This raises the question if perhaps the distribution of the male and female chromosomes as produced by the hen is not completely randomly 50/50, but somehow skew. In other words, do the follicles of the hen contain relatively more chromosomes from one gender than from the other gender?
This is biologically almost impossible, and the real explanation is more simple. It seems that in broilers the male embryos are producing slightly more heat than the female embryos. We can see this if we determine the gender during different moments in the hatching process, male broilers (or broiler breeders) tend to hatch approximately 6-8 hours earlier than females. As embryos are sensitive for overheating, it also means that when the eggs are too warm during incubation, males will suffer from this phenomena more than females, as they already have a slightly higher temperature in the egg by themselves. As a result, more male embryos will die in the last stage of incubation, and more females will be found after hatch. This is easy to check when we open the unhatched eggs and determine the gender of the dead in shells. What we can observe (in broilers) is that there are more male late deads than female late deads, and when we add up the dead shells with the live chicks, we see that the gender numbers are back to the expected 50/50 ratio again.
For layers, the effect seems to be the other way around. In layers the female seems to hatch earlier and when we overheat the eggs we see more female embryos that didn't hatch. This suggests that in layers, the females produce more heat than the males. Why that difference between layers and broilers exist is not known, but it perhaps the genetic selection for either eggs or meat production changes the characteristics of the embryo already in a somewhat unexpected way.