Flight evokes awe among us largely earth-bound creatures and the anatomical and physiological adaptations underlying aerial locomotion remain of great interest to scientists and the general public alike. Within extant vertebrates, birds and bats exemplify the notion that there is more than one way to take flight. For example, avian fliers have fused hand bones and use feathers to generate lift while bats construct their wings out of skin spanning extremely elongated fingers. Nevertheless, the physical demands of flight are great enough that vertebrate fliers have converged on various traits, and one that has received much attention is the genome size in these animals. Flying vertebrates have relatively small genomes; recent work suggests that even pterosaurs had a relatively small genome. Why do fliers have small genomes? Less DNA generally means smaller cells, which have a large surface area to volume ratio, facilitating oxygen diffusion to intracellular mitochondria. Flight requires high levels of energy to be sustained and thus improved oxygen transport in flying animals makes good sense.

Chandler Andrews and Stuart Mackenzie, along with T. Ryan Gregory, from the University of Guelph, who has a strong interest in genome diversity, wanted to investigate more closely the relationships between DNA, cell size and flight ability in birds while carefully controlling for phylogeny. To do so, they collected data on genome, nucleus, cell and body size from animals spanning 74 species, 51 genera and 18 families within the largest avian order,Passeriformes. In addition, from these same species they also made measurements of wing shape (aspect ratio) and wing loading (ratio of wing area to body mass) to look for connections between cellular features and macroscopic flight morphology.

Genome size ranged between 1.15 and 1.62 pg (mean, 1.32 pg) among the 74 species. To give you some sense of scale, most non-flying mammals have genomes at least twice this size, and salamanders, known for their large genomes, can reach values of over 100 pg. Even within this relatively small range among passerines, however, genome size was found to be positively related to nucleus and cell size in phylogenetically independent contrasts. Moreover, nucleus size increased disproportionately such that cells with larger genomes also had less cytoplasm. As bird erythrocytes have retained their nuclei, having a relatively large genome reduces the volume of cytoplasm in the cell and may consequently reduce their hemoglobin levels and oxygen carrying capacity. However, whether erythrocytes with larger genomes carry relatively lower levels of hemoglobin remains to be tested.

With respect to wing attributes, aspect ratio and genome size were largely unrelated. This is intriguing because it suggests that flight `style', at least among the limited range sampled in Passerines, doesn't impact on genome size, i.e. a bird with wings more suited to gliding will have a similarly small genome compared with an animal with wings designed for maneuverability. However, wing-loading index was positively related to genome size. The authors point out that as there was no correlation between body size and genome size,this relationship occurs because birds with larger genomes also tend to have relatively small wings. Although they make it clear that this is not due to some causal effect, they emphasize that even at a relatively fine scale,animals likely to be considered good fliers (i.e. those with low wing-loading index) also have small genomes and cell sizes.

As the database of genome sizes gets ever larger and more phylogenetically diverse, the window into evolutionary patterns and physiological consequences of a cell's DNA content will open wider. But even today it is becoming ever clearer that genome size and flight ability are both tightly and inversely linked.

Andrews, C. B., Mackenzie, S. A. and Gregory, T. R.(
). Genome size and wing parameters in passerine birds.
Proc. R. Soc. B