Calculating egg volume when shape differs: when are equations appropriate?

Egg volume is often calculated from length and breadth, assuming little variation in egg shape within species. However, egg shape may vary among females, over time, or within clutches. If this variation is not predictable, volume should be estimated using another method. For example, in some species, egg shape changes consistently with laying order and volume can be estimated using laying-order specific equations. We measured the length, breadth, volume, and mass of 249 Magellanic Penguin (Spheniscus magellanicus) eggs and regressed volume on length and breadth for first and second eggs separately. Differences between measured and calculated volume averaged <1% for both eggs. Using linear measurements, we calculated egg volume for 7085 clutches from 1983 to 2006. First and second eggs had similar calculated volumes (P > 0.05) in 19 of 24 yr; when they differed, the second egg was larger. In contrast, using single equations typically employed to compute egg volume, we erroneously concluded that first eggs were usually (20 of 24 yr) significantly larger than second eggs. Our regression equations should be applicable to other Spheniscus penguins because ratios of elongation coefficients (length divided by breadth) of first and second eggs are similar among these species. In other species of birds where egg shape changes with laying order, the volume of a sample of eggs should be measured to develop regression equations specific to the species and laying order. Slight variation in egg volume can have important evolutionary and ecological implications that would be erroneously interpreted without field tests to measure volume.