Among vertebrates, hummingbirds (family Trochilidae) represent an extreme outcome in vertebrate physiological design, and are unique in their capacity for sustained hovering flight. Because hovering flight is one of the most energetically demanding forms of locomotion requiring high levels of metabolic power input and mechanical power output, hummingbirds offer a useful system to study their energetics and flight mechanics. There are about 330 species of hummingbirds, of which 70% of the species weigh less than six grams and about 30% are between six and ten grams. Surprisingly, just one species, Patagona gigas(the Giant Hummingbird), weighs on average 20 g, making it an outlier in the hummingbird body size distribution.

Because power requirements increase with (body mass)^1.17, whereas maximum aerobic capacity of volant animals scales negatively with body mass, large body size presents a dual mechanical/metabolic challenge to hovering flight in the Giant Hummingbird. In spite of the challenges of supporting its large body weight, P. gigas inhabits a broad altitudinal range, from sea level to 4500 m.a.s.l. At higher altitudes, where the combination of low air density, low partial pressure of oxygen and low temperatures makes both lift generation and flight energetics especially costly to achieve, P. gigas, on account of its large body size, is likely to face more extreme energetic restrictions.

The overarching goal of my research program has been to determine the specific physical and environmental constraints that limit upper body size in hummingbirds, and to elucidate those aspects of the Giant Hummingbird lifestyle, physiology, biochemistry and biomechanics that have liberated it from these constraints. First, through measurements of daily, basal, and hovering rates of oxygen consumption in the Giant Hummingbird, I have shown that this species, although a clear outlier in terms of body size, does not deviate significantly from metabolic relationships derived from smaller hummingbirds. During this research I also measured enzyme flux capacities (V_max values) of key enzymes in pathways of glucose and fatty acid oxidation in the flight muscles of the Giant Hummingbird as well as smaller species. Because flight muscles account for most of the VO₂ during hovering flight, I was able to accurately estimate metabolic flux rates from respirometric data obtained during hovering flight. My results reveal that hummingbirds share a highly conserved set of pathways for glucose and fatty acid oxidation. In addition, there was a lack of quantitative, mass-dependent interspecific variation in V_max values for most of the enzymes involved in glucose and fat oxidation, in spite of the mass-dependent interspecific variation in hovering metabolic rate found in my first study. These results suggest a "one size fits all" hypothesis, i.e., qualitative as well as quantitative evolutionary conservation of pathways of energy metabolism. The lack of correlation between V_max values and flux rates at most steps in energy metabolism, suggests that the interspecific variation in flux through pathways of glucose and fatty acid oxidation during hovering is achieved through modulation of enzyme activities, rather than adjustments in enzyme concentration.

Overall, the allometric analyses and the enzyme flux capacities indicate that the Giant Hummingbird is just a "big" hummingbird and not an allometric outlier: estimates of metabolic parameters fall close to the allometric projections from smaller hummingbirds. However, my previous experiments were done at low elevations where the energetic requirements are lower. Consequently, the occurrence of P. gigas at a wide range of altitudes provides an excellent "natural experiment," with which to assess the behavioural, biomechanical and energetic responses of this bird to natural hypobaric variation. Accordingly, I explored the mechanisms used by this species to cope with the enhanced energetic and aerodynamic demands of living at high elevations. I measured flight kinematics (wingbeat frequency and stroke amplitude) and energetics (oxygen consumption) during hovering flight at two elevations spanning a 3700 m elevational gradient along the Andes Mountains of Peru. Contrary to my predictions, the Giant Hummingbird increased wing stroke amplitudes and wingbeat frequencies equally, and thus increased mechanical power output at high elevations relative to sea level. Moreover, oxygen consumption during hovering increased significantly (~33%) at the high elevation site. However, based on the observed increase in mechanical power, the 33% increase in metabolic power was larger than expected. It is unclear why metabolic costs would increase at a substantially greater rate than mechanical power output. Perhaps, exploring how the Giant Hummingbird modulates detailed wingbeat kinematics (e.g. wing rotation, angle of attack, torsion along the wing) might help to explain the unexpected lack of correlation between mechanical and metabolic power. Also, Patagona gigas gigas is a subspecies that inhabits mainly low elevations up to 2500 m. Studying this group in comparison to P. gigas peruviana (high elevation subspecies) could elucidate the mechanical and physiological mechanisms that have enabled the Giant Hummingbird to adapt to extremely different environmental conditions in spite of the high cost. Moreover, these comparisons may shed some light on the evolutionary trade-offs and constraints that have shaped the history and the evolutionary potential of hummingbird species.

In summary, my dissertation shows that in spite of the large size of the Giant Hummingbird, it is not an outlier in terms of energetics or flight performance when compared to smaller hummingbirds. These findings may suggest that hummingbirds body size distribution and upper body size limit might have an ecological explanation rather than a physiological, biochemical or biomechanical constraint.