Abstract: A reconstruction of Patrick Suppes' conception of physical theories and of the empirical theories in general (quantitative and non-stochastic) is intended. According to Suppes there are four elements within a theory: a set-theoretical structure; its possible realizations and models; measurements theories (fundamental as well as derivate) and data modela of experimentation.

In addition I present Suppes' proposal of axiomatizing a theory set-theoretically. I explain, then, Suppes views concerning the use of a logical concept of model in the factual interpretation of theories. I pretend to show that it is not enough to specify the próposed physical meaning of the primitives of an axiomatic theory in order to get the intended applications: within this specification the only things possible to eliminate -in an analogous way of the mathematicians- are its "non. standard" models.

Finally, I describe the role of measurement theories in Suppes' views of physical semantics: they behave as a bridge between the models proposed by the theory and certain data systematisations from experimental experience. They connect metrical concepts of theoretical physics with laboratory observations and operations.
[J.L.R.]
Keywords: