Although important cell biological questions concerning the compartmental organization of the secretory pathway still remain, enough is now known to permit a meaningful dissection of the molecular machinery of individual segments. It is both fortunate and remarkable that these kinds of pathways, whose essential purpose is to propagate the three-dimensional organization of the cytoplasm, can nonetheless be faithfully reproduced in dispersed cell-free systems without the benefit (or constraint) of preexisting spatial arrangements. This has opened the door to biochemistry, and superficial outlines of the steps involved in vesicle budding and fusion are now emerging. Crucial points remain. Among them, how, in step-by-step fashion, do coats assemble on membranes to yield a vesicle? What is the targeting signal that triggers uncoating and attachment of a vesicle? How can a protein machine fuse two lipid layers? Answers to questions of molecular mechanism at this level have and will necessarily continue to emerge from cell-free systems, and thus need in one way or another to be confirmed in vivo. Since so little is known, or can be learned, at this level from studies in whole cells, how can this be done? The answer is that the molecules discovered with cell-free systems, putatively performing the same roles in living cells, will provide the very tools to make the assessment of authenticity. As genes encoding these purified transport components are manipulated, and antibodies microinjected, the predicted effects on cellular physiology can be scrutinized. This has already begun by synthesizing the results from animal cell-free biochemistry with those from yeast genetics, and the results are encouraging. We can now look forward to a rapid fleshing out of the secretory pathway so that, in the not too distant future, we will be able to discuss these complex events of macromolecular targeting with the same kind of sophistication with which we now describe the biosynthesis of macromolecules like proteins and nucleic acids, and will be able to do so in a common language, that of protein biochemistry.