Nuclear state and level densities as a function of excitation energy, angular momentum, and parity have been calculated by a combinatorial method for 56Fe, 59Co, 60Ni, 61Cu, 62Ni, 63Cu, and 65Cu. Single-particle states for both Woods-Saxon and Nilsson potentials were used. These calculations were done with zero and nonzero pairing energy. State densities as a function of excitation energy have been calculated by an approximate inversion of exact partition functions; they agree well with state densities calculated by the combinatorial method. Average excitation energy as a function of temperature has been calculated from the partition function for each of the nuclei. Level densities as a function of energy, calculated by the combinatorial method, are compared with measured level densities. The agreement is either good or very good for most, but not all, of the nuclei. No evidence was found that must be interpreted as indicating a failure of the independent-particle model at higher excitation energies. For level density calculations with zero pairing energy, there is a suggestion, but no clear indication, that Woods-Saxon single-particle states are better than Nilsson single-particle states. Calculated and measured spin cutoff parameters are compared for 56Fe and 61Cu. Single-particle states for Nilsson-type potentials tend to give higher state and level densities than single-particle states for Woods-Saxon-type potentials. This tendency is not due to the larger number of single-particle states for Nilsson-type potentials, and it can be compensated for by using a nonzero pairing energy. The calculated fraction of negative-parity states is about one-half as expected, but this fraction varies much more than expected from one energy interval to another. The calculated M-value distribution is approximately Gaussian as expected.