The forces producing neural closure in amphibia.

I T has been shown by Smith (1946), working with trout eggs, that when an embryo develops the heat lost to the surroundings is balanced by the loss in the calorific value of its energy-rich storage substances. Other workers (e.g. Shearer, 1922) have shown that the heat lost by eggs and developing embryos is approximately equal to the available energy deduced from the observed respiration. It would seem, therefore, that only a very small fraction of the energy known to be available to the embryo from its aerobic metabolism can be stored as potential energy in the visible pattern of tissue structures which the embryo progressively develops. It has sometimes been concluded from this that little energy can be required for morphogenesis. This conclusion is doubtless true, but the argument is unsound, because no knowledge of the energy balance between the total energy being released by the embryo and the energy being lost to the surroundings can tell us anything about the work which may be required to overcome a potential energy barrier. For instance, consider this example which is used here as a simple illustration of the argument rather than as a model of what actually happens. It is conceivable that when a newt embryo gastrulates, considerable work must be done in so deforming the cells against their own elasticity that the blastula tissue may pass through the narrow blastopore opening. To perform such work energy is required from the chemical energy source; it will be stored as potential energy by deformed cells as they pass through the blastopore; yet when the constriction is passed, the cells relax to their equilibrium shapes once more and the energy is dissipated as heat. In such a case a knowledge of the energy-balance could tell us nothing about the work done in gastrulation. On the other hand, if we knew about the forces required to deform the cells we should be able to estimate the work required for gastrulation, using this particular model for the process. Tuft (1953) working with Rhodnius made an indirect estimate of the proportion of the total energy available which could conceivably be used by an embryo in performing morphogenetic movements. As pointed out by Waddington (1956), one weakness of this estimate is the necessity of assuming a constant energy requirement for maintenance. Tyler (1942) found that two dwarf echino-