During his long activity in research, Angelo Favini has given several relevant contributions in many elds of mathematics. In our long collaboration, testied by more than thirty joint papers, I highly appreciated his deep competence, his brilliant intuitions, his extraordinary knowledge of the contemporary literature, in addition to a special acumen in nding possible critical points to be claried, or better solved, during the preparation of the papers.Here follows a short survey of the main problems studied in our joint papers over the last twenty years, often in collaboration with other outstanding mathematicians.(1) At the beginning of 1990s there was a long standing open problem concerning the existence of analytic semigroups generated by second order elliptic dierential operators degenerating at the boundary. It is well-known that the existence of a (C 0 ) semigroup guarantees the wellposedness of the abstract Cauchy problem associated with the evolution equation governed by the generator, provided that the initial datum is in its domain. The analyticity of the semigroup provides, in addition, the best possible regularity of the solution, even by starting from the initial datum in the ambient space. As already known in Probability, or in Approximation Theory, in the space of continuous functions C(Ω) (here Ω is a bounded open subset of R N with suciently smooth boundary ∂Ω), a possibly degenerate on the boundary, second order elliptic operator A can be naturally equipped with the so-called Wentzell boundary condition Au = 0, introduced in the one-dimensional case by W. Feller in his pioneer work [7] (see also [6]), and in the multidimensional case by A.D. Wentzell [38]. For Wentzell boundary conditions in dierent spaces see e.g. [33,37]. Observe that, in an evolution equation u t = Au, replacing Au by u t in Wentzell boundary condition reveals that, under suitable regularity assumptions on the elements of the domain of A, a solution u of the equation with Wentzell boundary condition remains constant with respect to t along the boundary. An easy example of degenerate elliptic second order dierential operator on the space C[0, 1] equipped with Wentzell boundary conditions is A j u = x j (1 − x) j u ′′ , j > 0, well-known also for governing some evolution problems in genetics. In [1] the existence of analytic semigroups generated by A j was proved for j ≥ 2 on C[0, 1] by using dierent types of boundary conditions, including Wentzell's ones. After that, much attention was paid to this problem and relevant contributions by A. Favini et al. appeared in dierent papers, where, in addition to the space of continuous functions, as in [27,30,31], also L p spaces, with or without weights, and Sobolev spaces were considered, see e.g. [2,20,28,29]. Results concerning the wave equation with Wentzell boundary conditions in C[0, 1] were also obtained in [16].(2) In the real-valued space C(Ω) other interesting problems concern the existence of Feller semigroups generated by second-order elliptic dierential operators degenerati...