Here are some historical notes, in complement to Pradines' comments on multiplicative graphs and on quotient categories. Charles Ehresmann introduced multiplicative graphs as an algebraic support to the notion of a "category kernel", which corresponds to the structure induced on an open subset of a topological category (in the sense of Charles, i.e. a category internal to Top). Such category kernels generalize group kernels (that Charles used in the second part of his thesis to define locally homogeneous spaces); we had considered them in the early sixties both in his papers on differential geometry and in mine on optimization problems. Then multiplicative graphs appeared as a natural tool in his study on quotient categories. Charles came to quotient categories by his works on foliations and on prolongations of manifolds. In his long 1962 paper on foliations (1), he constructs several kinds of holonomy groupoids by a quotient process. In his paper on topological categories (2), he construct categories of jets (local jets and infinitesimal jets) as quotient categories of the category of local sections of a topological category. In fact he had exposed these constructions much earlier in his lectures and briefly alluded to in his papers on differential geometry in the late fifties. So he was naturally led to a more formal study of quotients, which he began in his paper "Structures quotient" (3), where in particular he introduces the "strict quotient"; this paper (abstracted in (4)) is more easily read than later papers. In a comment I have added to this paper in the "Oeuvres" (comment 170, p. 375) I mentioned other authors who have studied quotient categories about the same time, in particular Dedeckerand Mersch, Higgins, Hoenke, Pümplun. Later on, Charles tried to develop a non-abelian cohomology for which he wanted to define a natural notion of "short sequence" in Cat. For this he needed to define quotients of a category or of a groupoid by a sub-category. It is done in the paper (5). The main results of this paper are taken back in his book "Categories et Structures" (Dunod 1965), with some illustrative diagrams (it is to this book that Pradines refers). In several papers, he generalized the theory of quotient categories in the frame of internal categories (cf. "Oeuvres", Parts III and IV), . 1. "Structures feuilletees", Proc. 5th Canadian Math. Congress; reprinted in "Charles Ehresmann, Oeuvres completes et commentees" Part II-2, 563-626. 2. "Categories topologiques, III", Indig. Math. 28, 1966; reprinted in the "Oeuvres" Part II-2, 655-669. 3. "Structures quotient",.Comm. Math. Helv. 38, 1963; reprinted in "Oeuvres' III-1, 143-293. 4. "Structures quotient et cegories quotient", CRAS, 256, 1963, 5031-34, reprinted in "Oeuvres" III-1, 9-11. 5. "Cohomologie dans une categorie dominee", Proc. Coll. Topologie, CBRM 1964; reprinted in "Oeuvres" III-2, 531-590.