On this occasion of marking 50 years to the inception of hypernuclear physics, a brief overview of strangeness nuclear physics is given, focussing on the recent determination of spin-dependent effects in Λ hypernuclei, on the recently established repulsive nature of the Σ nuclear potential and particularly on new developments in double-strangeness physics. Special emphasis is placed on the extrapolation to strange hadronic matter. §1. Introduction and summaryThe properties of hypernuclei reflect the nature of the underlying baryon-baryon interactions and, thus, can provide rather stringent tests of models for the freespace hyperon-nucleon (Y N) and hyperon-hyperon (Y Y ) interactions. Over the years, the Nijmegen group has constructed a number of one-boson-exchange (OBE) models for the baryon-baryon interaction (NN , ΛN − ΣN , and ΞN − ΛΛ − ΣΣ − ΛΣ) using SU(3) f symmetry to relate baryon-baryon-meson coupling constants and phenomenological hard or soft cores at short distances (for a recent review, see Ref . 1)). In addition, the Jülich group has constructed meson-exchange models for the Y N interaction along the lines of the Bonn Model for the NN interaction using SU(6) symmetry to relate coupling constants and short-range form factors. 2) Baryon-baryon SU(6) quark model interactions have also been used within a unified framework of a (3q) − (3q) resonating group method, augmented by a few effective meson exchange potentials of scalar and pseudoscalar meson nonets directly coupled to quarks (see Ref.3), and references therein).To test these models against the considerable body of information on Λ hypernuclei, effective interactions appropriate for use in limited spaces of shell-model orbits must be calculated. This has been done by calculating nuclear-matter G-matrices 4) for the coupled ΛN and ΣN channels. For most models, the resulting Λ well depth is in reasonable agreement with the empirical value 28 MeV deduced from fitting the binding energies of Λ single-particle states 5), 6) observed via the (π + , K + ) reaction at BNL 7) and KEK, 8) but the partial-wave contributions vary widely for the different models. In fact, the spin-dependence of the central interaction varies considerably in the different models. 4) While empirical ΛN effective interactions fitted to the properties of light Λ hypernuclei are more attractive in the singlet than the triplet state, 9), 10) a conclusive choice between the models cannot be made since the effective ΛN N interaction resulting from ΛN − ΣN coupling, which contributes strongly to the spin-dependence for A = 4, 11) has yet to be evaluated in a consistent way. Nevertheless, the spin dependence of the effective ΛN interaction can be extracted by observing excited states in light Λ hypernuclei, as proposed long ago by Dalitz and Gal. 12) This program has been accomplished experimentally in a beautiful set of γ-ray measurements in BNL and KEK, as reported in this meeting by Tamura.