The use of chemically selective laser ionization combined with b-delayed neutron counting at CERN/ISOLDE has permitted identification and half-life measurements for 623-ms 61 Mn up through 14-ms 69 Mn. The measured half-lives are found to be significantly longer near N 40 than the values calculated with a quasiparticle random-phase-approximation shell model. Gamma-ray singles and coincidence spectroscopy has been performed for 64,66 In addition to the clear nuclear-structure interest, the neutron-rich Fe-group nuclei may also play an important role as possible seed nuclei in the astrophysical r process [13]. In the present paper, we report new measurements for the half-lives of heavy Mn nuclides up to 69 Mn and for the level structure of 64,66 Fe populated in the decays of 64,66 Mn.Manganese isotopes were produced at CERN by 1-GeV proton-induced spallation of uranium in a thick UC 2 target at the ISOLDE facility. The ionization of the Mn atoms was accomplished using a chemically selective, three-step laser resonance excitation scheme as described in detail earlier [14].Beams of Mn nuclides with masses differing by DA $ 4 were transported separately to two different beam lines equipped with moving tape systems where b-delayed neutron (d.n.) multiscaling and g-ray singles and coincidence measurements could be performed independently. In both cases, counting took place directly at the point of deposit, and the tape systems were used to remove the daughter nuclides as well as unavoidable surfaceionized isobaric Ga activities. Because the Mn half-lives being sought are in the millisecond range, data acquisition in both systems was initiated by the proton pulses from the CERN proton-synchrotron booster (PSB), separated by a multiple of 1.2 s, and continued for 1.0 s for each cycle.Beta-delayed neutron data of high statistical quality were collected by multiscaling measurements using the Mainz 4p 3 He neutron counter. The time dependence of the counting rates for 65 69 Mn is shown in Fig. 1. The decay curves were fitted with a constant small d.n.-background component up through A 65. Because there exist no measured d.n.-emission probabilities (P n values) for the A . 65 daughter and granddaughter isobars, the fits of the heavier isotopes were performed using theoretical P n values [10] along with the known half-lives [5,7,8,15]. For A 66 68, the contributions from d.n. emission of the Fe and Co isobars are quite small and actually do not affect the Mn half-life fits. For A 69, however, a multicomponent fit was necessary to account for the significant Fe and Co d.n. branches. The resulting data are summarized in Table I 0031-9007͞99͞82(7)͞1391(4)$15.00