Cuvier used the notion of organic structure to establish the idea
of an animal being built according to a blueprint, or archetype.
You knew, Cuvier thought, that if an animal possessed a certain kind of respiratory system, lungs, for example, the rest of its internal organisation must fit in with this: it would not work if the animal was otherwise adapted to live on the bottom of the sea. The various components which together made up the whole system which is the animal must be correlated together.
He achieved notoriety for being able to argue on the basis of a mere scrap of animal remains a fossil bone for example to conclusions about how the animal must have been as a whole.
"At the sight of a single bone, of a single piece of bone, I recognise and reconstruct the portion of the whole from which it would have been taken. The whole being to which this fragment belongs appears in my mind's eye" (quoted by Mayr: Mayr, Ernst, The Growth of Biological Knowledge, Cambridge, Mass 1982 Harvard University Press p. 460).
Some features would clearly have wider implications than others: the colour of an organism would not have the profound effect on its total organisation as would, say, its circulatory system. It is those with the larger implications which Cuvier, reasonably enough, is suggesting are the best guide to an animal's "fundamental organisation" its archetype.( Vernon Pratt, 'Biological Classification', British Journal for the Philosophy of Science 23 (1972), p 312) (e.g. whether a car runs on petrol or diesel has a bigger influence on its organisation than whether it has tweed or plain upholstery.)
Functional sub systems hang together: have one of a particular sort, and you need certain others if the whole is to work. Some functions are more important than others, in that (I think) that some necessitate others, occasioning the notion of a hierarchy of levels.
It follows, I think, that there must a rather limited number of basic blueprints for living things if they are to work.
Cuvier identified a handful, which he identified with the great divisions of the animal kingdom that had half been recognised already. The vertebrate was one; the mollusc the articulates and radiates were others. (Mayr op.cit., p. 182, 3)
Each archetype then can be regarded as a design for a working whole or system, which is the new way of looking at what an animal is.