We find the phase and flavor symmetry breaking pattern of each N = 1 supersymmetric vacuum of SU (n c ) and U Sp(2n c ) gauge theories, constructed from the exactly solvable N = 2 theories by perturbing them with small adjoint and generic bare hypermultiplet (quark) masses. In SU (n c ) theories with n f ≤ n c the vacua are labelled by an integer r, in which the flavor U (n f ) symmetry is dynamically broken to U (r) × U (n f − r) in the limit of vanishing bare hyperquark masses. In the r = 1 vacua the dynamical symmetry breaking is caused by the condensation of magnetic monopoles in the n f representation. For general r, however, the monopoles in the n f C r representation, whose condensation could explain the flavor symmetry breaking but would produce too-many Nambu-Goldstone multiplets, actually "break up" into "magnetic quarks": the latter with nonabelian interactions condense and induce confinement and dynamical symmetry breaking. In U Sp(2n c ) theories with n f ≤ n c + 1, the flavor SO(2n f ) symmetry is dynamically broken to U (n f ), but with no description in terms of a weakly coupled local field theory. In both SU (n c ) and U Sp(2n c ) theories, with larger numbers of quark flavors, besides the vacua with these properties, there exist also vacua in free magnetic phase, with unbroken global symmetry.

We study dynamical flavor symmetry breaking in the context of a class of N = 1 supersymmetric SU (n c ) and U Sp(2n c ) gauge theories, constructed from the exactly solvable N = 2 theories by perturbing them with small adjoint and generic bare hypermultiplet (quark) masses. We find that the flavor U (n f ) symmetry in SU (n c ) theories is dynamically broken to U (r) × U (n f − r) groups for n f ≤ n c . In the r = 1 case the dynamical symmetry breaking is caused by the condensation of monopoles in the n f representation. For general r, however, the monopoles in the n f C r representation, whose condensation could explain the flavor symmetry breaking but would produce too-many Nambu-Goldstone multiplets, actually "break up" into "magnetic quarks" which condense and induce confinement and the symmetry breaking. In U Sp(2n c ) theories with n f ≤ n c +1, the flavor SO(2n f ) symmetry is dynamically broken to U (n f ), but with no description in terms of a weakly coupled local field theory. In both SU (n c ) and U Sp(2n c ) theories, with larger numbers of quark flavors, besides the vacua with these properties, there exist also vacua with no flavor symmetry breaking.

We present a high precision Monte Carlo study of the spectrum of the Z 2 gauge theory in 2 + 1 dimensions in the strong coupling phase. Using state of the art Monte Carlo techniques we are able to accurately determine up to three masses in a single channel. We compare our results with the strong coupling expansion for the lightest mass and with results for the universal ratio σ/m 2 determined for the φ 4 -theory. Finally the whole spectrum is compared with that obtained from the Isgur-Paton flux tube model and the spectrum of the 2 + 1 dimensional SU (2) gauge theory. A remarkable agreement between the Ising and SU(2) spectra (except for the lowest mass state) is found. *

We determine the vacuum structure and phases of N = 1 theories obtained via a mass µ for the adjoint chiral superfield in N = 2, SO(n c ) SQCD. For large number of flavors these theories have two groups of vacua. The first exhibits dynamical breaking of flavor symmetry USp(2n f ) → U (n f ) and arises as a relevant deformation of a nontrivial superconformal theory. These are in the confined phase. The second group, in an IR-free phase with unbroken flavor symmetry, is produced from a Coulomb branch singularity with Seiberg's dual gauge symmetry. In the large-µ regime both groups of vacua are well-described by dual quarks and mesons, and dynamical symmetry breaking in the first group occurs via meson condensation. We follow the description of these vacua from weak to strong coupling and demonstrate a nontrivial agreement between the phases and the number of vacua in the two regimes. We construct the semiclassical monopole flavor multiplets and argue that their multiplicity is consistent with the number of N = 1 vacua.