The parts of an object put together in an assembly line are typically fully Euclidean, having rigid metric properties such as sizes, shapes and positions, a fact that limits the procedures that may be followed for their assembly. these procedures must include a rigidly channelled transport system (using conveyor belts or pipes to transport raw materials, and wires to transport energy and information) as well as sequences of rigid motions to correctly position the parts relative to one another. By contrast, the component parts used in biological assembly are defined less by rigid metric properties than by their topological connectivity: the specific shape of a cell's membrane is less important that its continuity and closure,and the specific length of a muscle less important than its attachment points. This allows component parts to be not inert but adaptive, so that muscle lengths can change to fit longer bones, and skin can grow and fold adaptively to cover both. It also permits transport processes not to be rigidly channeled, using simple diffusion through a fluid medium to bring the different parts together. Components may float around and randomly collide, using a lock-and-key mechanism to find matching patterns without the need for exact positioning.

All of this has consequences for the capacity to evolve through mutation and selection which each of these two assembly processes may have. If putting together organisms followed an assembly-line pattern, random mutations would have to occur simultaneously in matching parts, channels and procedures, in order to yield a viable entity on which natural selection could operate. the occurence of such a large number of simultaneous mutations is of course, a highly improbable event. In biological assembly on the other hand, mutations do not have to be so coordinated and this greatly enhances the possibilities for evolutionary experimentation.

(Manuel De Landa, from Intensive Science and Virtual Philosophy)