By a whisker: the sensory role of vibrissae in hovering flight in nectarivorous bats.

Whiskers are important tactile structures widely used across mammals for a variety of sensory functions, but it is not known how bats-representing about a fifth of all extant mammal species-use them. Nectar-eating bats typically have long vibrissae (long, stiff hairs) arranged in a forward-facing brush-like formation that is not present in most non-nectarivorous bats. They also commonly use a unique flight strategy to access their food-hovering flight. Here we investigated whether these species use their vibrissae to optimize their feeding by assisting fine flight control. We used behavioural experiments to test if bats' flight trajectory into the flower changed after vibrissa removal, and phylogenetic comparative methods to test whether vibrissa length is related to nectarivory. We found that bat flight trajectory was altered after vibrissae removal and that nectarivorous bats possess longer vibrissae than non-nectivorous species, providing evidence of an additional source of information in bats' diverse sensory toolkit.

Specialized landing maneuvers in Spix's disk-winged bats (Thyroptera tricolor) reveal linkage between roosting ecology and landing biomechanics.

Disk-winged bats (Thyroptera spp.) are the only mammals that use suction to cling to smooth surfaces, having evolved suction cups at the bases of the thumbs and feet that facilitate attachment to specialized roosts: the protective funnels of ephemeral furled leaves. We predicted that this combination of specialized morphology and roosting ecology is coupled with concomitantly specialized landing maneuvers. We tested this by investigating landings in Thyroptera tricolor using high-speed videography and a force-measuring landing pad disguised within a furled leaf analogue. We found that their landing maneuvers are distinct among all bats observed to date. Landings comprised three phases: (1) approach, (2) ballistic descent and (3) adhesion. During approach, bats adjusted trajectory until centered in front of and above the landing site, typically the leaf's protruding apex. Bats initiated ballistic descent by arresting the wingbeat cycle and tucking their wings to descend toward the leaf, simultaneously extending the thumb disks cranially. Adhesion commenced when the thumb disks contacted the landing site. Significant body reorientation occurred only during adhesion, and only after contact, when the thumb disks acted as fulcra about which the bats pitched 75.02±26.17 deg (mean±s.d.) to swing the foot disks into contact. Landings imposed 6.98±1.89 bodyweights of peak impact force. These landing mechanics are likely to be influenced by the orientation, spatial constraints and compliance of furled leaf roosts. Roosting ecology influences critical aspects of bat biology, and taken as a case study, this work suggests that roosting habits and landing mechanics could be functionally linked across bats.

Wings as inertial appendages: how bats recover from aerial stumbles.

For many animals, movement through complex natural environments necessitates the evolution of mechanisms that enable recovery from unexpected perturbations. Knowledge of how flying animals contend with disruptive forces is limited, however, and is nearly nonexistent for bats, the only mammals capable of powered flight. We investigated perturbation recovery in Carollia perspicillata by administering a well-defined jet of compressed air, equal to 2.5 times bodyweight, which induced two types of disturbances, termed aerial stumbles: pitch-inducing body perturbations and roll-inducing wing perturbations. In both cases, bats responded primarily by adjusting extension of wing joints, and recovered pre-disturbance body orientation and left-right symmetry of wing motions over the course of only one wingbeat cycle. Bats recovered from body perturbations by symmetrically extending their wings cranially and dorsally during upstroke, and from wing perturbations by asymmetrically extending their wings throughout the recovery wingbeat. We used a simplified dynamical model to test the hypothesis that wing extension asymmetry during recovery from roll-inducing perturbations can generate inertial torques that alone are sufficient to produce the observed body reorientation. Results supported the hypothesis, and also suggested that subsequent restoration of symmetrical wing extension help to decelerate recovery rotation via passive aerodynamic mechanisms. During recovery, humeral elevation/depression remained largely unchanged while bats adjusted wing extension at the elbow and wrist, suggesting a proximo-distal gradient in the neuromechanical control of the wing.

The dynamics of hovering flight in hummingbirds, insects and bats with implications for aerial robotics.

We analyze the effects of morphology and wing kinematics on the performance of hovering flight. We present a simplified dynamical model with body translational and rotational degrees of freedom that incorporates the flapping, long-axis wing rotation and folding of the wing. To validate our simulation, we compare our results with direct measurements from hovering insects, hummingbirds and bats. Results show that long-axis wing rotation angle (a proxy for pronation) has a significant effect on energy efficiency. For a given wing rotation amplitude, the hovering system has a power-optimal flapping frequency for each stroke-plane orientation, and that frequency closely corresponds to the wingbeat frequencies observed in a diverse range of hummingbird species. We find that larger animals (with larger total mass and wing size), such as bats, require more power to maintain a stable hovering orbit and that hovering with a constant wingspan becomes increasingly impractical with increasing body size. We show, as an exemplar, that for a system of the size of a hovering bat, e.g. Glossophaga soricina, hovering with constant wingspan is dynamically possible, but is implausible and inefficient. For these conditions, hovering with varying wingspan, retracting the wing on the upstroke, is a more realistic hovering modality.