We appreciate the letter of Dr Andersen, and we wholeheartedly acknowledge the contributions of Mel Barclay for not only his scientific work but also as the true humanist he was. However, there are important differences between the physiological assumptions of organ-level contractility that we used and those used by Andersen and Barclay in 1995 and Barclay, Andersen, and Simon in 2010.

In both papers, the authors assumed the presence of a small number of functional pacemakers (perhaps only one, though) and that the only means of recruiting new tissue into each contraction is through direct cell-to-cell signaling as the action potential passes through tissue. This mechanism also implies that the laboring uterus does not contain any intermediate structures but rather is composed of a few pacemakers and a billion or so functionally similar myocytes.

In our recent paper, we presented a soccer crowd analogy based on the organ-level physiology detailed in 2 recently published mathematical simulations. Although we used cellular automaton methods technically similar to those used by Barclay et al in 2010, we specifically omitted the near-neighbor requirement for signaling over long distances.

We agree that action potential generation and propagation through tissue is critically important for labor but propose that this mechanism is effective over only short distances. For long-distance recruitment, we used rises of intrauterine pressure to mechanically stimulate tissue and coordinate contractions at the organ level. The general concept of this mechanotransduction model was first proposed by Csapo more than 4 decades ago.

By adding mechanotransduction, our dual mechanism does not require an electrical pacemaker, and indeed we question whether a pacemaker, as it is usually defined, actually exists. Our key assumption is that tissue-level action potentials are not capable of traveling organ-sized distances. This assumption is experimentally supported in rodent and in human multichannel electromyographical studies. The limitation on distance results in electrically coupled functional regions, or patches, of myometrium that have been directly observed in the human.

Our model of organ-level physiology is fundamentally different from the model of Barclay et al. Whereas space does not allow detailed explanation, many predictions of the pacemaker-action potential simulations have not been supported by the observations of experimental studies including the following: no pacemaker, no linearly propagating bands, uterine propagation speeds greater than neuronal speed, no long-distance electrical propagation, wide variation of propagation direction. However, each of these observations is a natural consequence of our combined, action potential-mechanotransduction model.

The analogy with an orchestra is intended to focus on the need for a conductor to provide the brains in the central to peripheral organization of the heart, and this analogy also applies to the model of Barclay et al.

In contrast, our model is like a soccer crowd, which demonstrates emergent properties of a semistable system ( http://www.ajog.org/cms/attachment/2035140982/2050479843/mmc1.mp4 ). Here coordinated events arise without an obvious initiating factor, and the mechanism clearly requires more than near-neighbor signaling. Although more detailed analogies are possible, we propose that the key to understanding human labor is to acknowledge the importance of action potential propagation at the cell and tissue level but also to consider the need for a second mechanism of tissue recruitment at the organ level.