> > In the classic model of sex ratio evolution, Fisher (1958) predicted that population-wide investment in male and female offspring (as measured by biomass) should approach equality in panmictic populations. In most bees, males are smaller than females, meaning that populations are expected to be numerically male-biased, but to test the hypothesis of equal investment, one needs to consider differences between males and females in adult body weight, or some other measure of offspring "cost". To calculate investment sex ratio in the emerging population of Andrena regularis, we combined numerical sex ratio data from emergence traps with body weight data for male and female A. regularis. Average weights of the male and female A. regularis were calculated by weighing 24 dried specimens of each sex

Why do we want to measure this in dry weight? Water is also a resource, one that takes a good amount of work to supply to a beehive.

Is this more of a situation where...

- We believe that differences in water allocation are significant to the question, but we also believe that all bees receive allocations proportional to their dry weight;

- We believe that differences in water allocation are not significant to the question, because there is effectively unlimited water available and every bee can have as much as they want without affecting any other bees; or

- We believe that differences in water allocation are significant to the question, but we're measuring something else because we don't know how to measure the water allocation?

Not my area of expertise but I expect that water is disproportionately heavy and ephemeral while protein and fat is extremely resource intensive in comparison. I also expect its presence to greatly complicate specimen handling since humidity, airflow, and time since collection would then matter whereas dry weight shouldn't vary over time.

So my guess is that including it would increase variance and error without offering any benefit.