Generally speaking, we tend to think that bigger is better. This indeed seems to be the case when to comes to egg size. Bigger eggs tend to have more resources and produce larger chicks that have a higher chance of survival. Why then do all females not lay large eggs? Or, in other words, why do we find variation in egg size?

In actual fact, many traits like this, which affect the survival or number of offspring that an individual has (known as fitness-related traits), are very variable despite there being an apparent optimum for all individuals (for example laying a large egg). Why then does variation in these fitness-related traits persist?

One possible answer to this is that egg size does not have any genetic basis, in other words, no part of the variation we see is due to variation in genes between individuals. Without a genetic basis, selection acting on a population does not result in any genetic change over time (evolution), and no reduction in variation.

In a recent study, published in the journal Heredity, we investigated this using a method known as artificial selection in a captive population of Japanese quail (think small and expensive eggs used in gourmet cooking). Artificial selection is a way of experimentally simulating natural selection. We selected those females producing the largest or smallest eggs from our starting population to create two lines. In each subsequent generation, we then selected the most extreme individuals within these lines to breed. The success of creating such divergent selection lines depends on how much of the variation in egg size if due to genetic differences between individuals, which is measured by the ‘heritability’ of a trait. If a trait is very heritable (in other words a large proportion of variation in egg size between individuals is due to genetic variation) then we would expect to see our lines quickly diverge. This was indeed what we saw, indicating that it is not the lack of a genetic basis that is constraining the evolution of egg size.

But maybe laying a larger egg does not mean that you are giving more resources to your offspring? We compared the resources in the egg after several generations of this selection, and found that as well as differing in egg size, the lines also differed in egg resources.

Another possibility is that the variation in the genes that influence egg size, also influences other traits that are linked to fitness. Selection on these other traits may then affect how egg size responds to selection. By comparing how much the number of eggs produced and the body size of females changed as a result of our selection regime, we could show that neither trait has a strong genetic link to egg size, and so would not constrain egg size evolution.

So, a substantial part of variation in egg size is due to genetic variation and egg size is a good representation of maternal resource investment. But egg size is not genetically linked to egg production, or (to any great extent) to body size. What then constrains the evolution of egg size?

Well, although overall we did see eggs get both larger and smaller than when we started the experiment, the progression of these changes was different in the two lines. In the large egg line egg size increased with one generation of selection but then plateaued and did not change after three more generations of selection, whilst we saw a consistent decrease in egg size in the small egg line.

Significantly, such an asymmetric response to selection has been noted in several other selection experiments on fitness-related traits. This has been attributed to what is known as genetic asymmetry, whereby the genetic variants (or alleles) that lead to high fitness are already at high frequency (or show dominance) in the population, meaning that it is more difficult to select for high fitness (i.e. large eggs) than for low fitness (small eggs). This would explain why in natural populations it is difficult to select for higher fitness.

Without more detailed genetic information, we cannot conclusively say this is the mechanism maintaining variation in egg size in natural populations. However, it is the most likely reason for our results, and undoubtably contribues to the maintenance of variation in fitness related traits in the wild.