In section (f) of our original electronic supplementary material, we incorrectly divided by two when determining our relationship between stance phase duration and animal mass. Therefore, when we expressed nerve conduction delay as a fraction of stance duration in our original paper, the fraction was twice as large as it should have been. For example, the portion of stance phase at the trot–gallop transition taken up by nerve conduction delay should be 2% for the shrew and 20% for the elephant, rather than our reported 4% and 40%.
This error affects our reported percentages in the second-last paragraph of our discussion, our y-axis scale in figure 4, and two equations in section (f) of our electronic supplementary material. It does not affect the magnitudes of our reported delays or our finding that delays increase in larger animals, but does affect our expression of these delays relative to stance phase.
Although all animals face less of a challenge owing to nerve conduction delay than we originally reported, large animals are still faced with long and significant nerve conduction delays. Our general conclusion that nerve conduction delay presents a bigger challenge to sensorimotor control in larger animals is thus not affected by our mistake.
Corrected figure 4 is shown below:
- © 2014 The Author(s) Published by the Royal Society. All rights reserved.