Breathing while trotting

Breathing is most easily studied while an animal is resting, and indeed we know most about breathing under these conditions. But breathing while resting is breathing at its least intense. To see breathing when the demands on it are greatest, we must study it in animals running fast.

Dennis Bramble and David Carrier did just that, in pioneering work at the University of Utah that was published in 1983 (1). They showed that galloping dogs and horses take one breath per stride, this breath always coming (at least in horses) at the same point in the stride. The ideas that this observation suggested have guided much subsequent research. Now Bramble, working this time with Farish Jenkins of Harvard University, is forcing us to think again in a report in this issue of Science (2). This time, the emphasis is neither on resting nor on galloping (the fastest gait) but on trotting, the most used gait of dogs and the one used for traveling at moderate speeds.

While research in breathing was confined to resting animals, the mechanism seemed straightforward. The thoracic cavity, which contains the heart and lungs, is separated from the abdomen by the diaphragm, a musculotendinous partition which is domed, concave toward the abdomen. Contraction of the muscle of the diaphragm flattens it, enlarging the thoracic cavity (so drawing air into the lungs) and displacing the abdominal viscera caudally. When the muscle of the diaphragm relaxes, the air is driven out of the lungs by their elastic recoil, aided if necessary by contraction of the abdominal muscles. The effect of contraction of the diaphragm is supplemented by enlargement of the rib cage, driven by the intercostal muscles.

In running, the observed synchrony of breathing with leg movements suggests a much more passive role for the diaphragm, with breathing being driven largely by the movements of locomotion. The synchronization is widespread: Quadrupeds ranging from gerbils to rhinoceros take one breath per galloping stride (3) and wallabies take one breath per hop (4). Bramble and Carrier (1) suggested several possible mechanisms, among which their visceral piston hypothesis seemed particularly attractive. The diaphragm has elastic properties (5), so the abdominal viscera can be...