Does the surface tension make the lung inherently unstable?

Many authors regard the human lung as a collection of 300 million bubbles independently connected by cylindrical tubes. Under surface tension such a model is inherently unstable in the sense that the small alveoli would empty into the large ones so that the lung would consist only of collapses and hyperinflated alveoli. It has been demonstrated that this basic model is wrong. My observation is based on the well-known fact that both sides of each interalveolar septum are exposed to ventilated air. When the topological relationship between the alveolar septa is properly taken into account, it can be shown that each interalveolar septum is a minimal surface and that there is no problem of inherent instability in the sense mentioned earlier. However, the lung structure is flimsy and can become unstable in the same sense that an airplane structure or an Atlas rocket can become unstable. The clarification of lung inflation and atelectasis can proceed in a rational manner when the confusion of an erroneous model is removed.