1. Both conventional whole cell and perforated-patch voltage-clamp recordings were made of high-voltage-activated (HVA) calcium (Ca2+) channel currents in acutely dissociated medical septum and nucleus of the diagonal band neurons from young (1-3.5 mo) and aged (19-26.5 mo) Fischer 344 rats. Barium (Ba2+) was used as the charge carrier to minimize secondary Ca(2+)-induced conductances and Ca(2+)-induced inactivation. 2. When HVA currents generated by voltage ramps from a holding potential (Vh) of-60 mV were recorded within minutes after whole cell formation, no change in peak current density was observed between young (-44.7 +/- 2.5 pA/pF, mean +/- SE, n = 93) and aged (-44.2 +/- 2.1 pA/pF, n = 86) cells. However, currents recorded later with voltage step protocols revealed a reduction in peak current amplitudes and a trend toward larger peak current densities in aged cells. From a Vh of -60 mV and with steps to -10 mV, current densities were -21.5 +/- 1.9 pA/pF in young cells (n = 55) and -25.0 +/- 2.0 pA/pF in aged cells (n = 44). The differences in current densities recorded by the two protocols were explained by nonspecific current rundown and the development of a slow (min) inactivation process. Slow inactivation was different from conventional rundown of HVA currents because it was reversible with the use of perforated-patch recordings. 3. Perforated-patch recordings were used to characterize slow inactivation. There was significantly less slow inactivation in aged cells. When voltage steps (200 ms in duration, from -80 to -10 mV) were delivered at 12-s intervals, slow inactivation reduced the current after 15 min to 63 +/- 7% of control in young cells and 86 +/- 4% in aged cells (P = 0.028). When voltage steps were delivered at 20-s intervals, the current at the 15th step decreased to 93.4 +/- 1.5% of control in aged cells, compared with 86.6 +/- 1.6% in young (P = 0.007). There was less slow inactivation with increased intervals between voltage steps and with shorter step durations. There was also less inactivation with reduced concentration of charge carrier, indicating a current-dependent component to slow inactivation. Additionally, a voltage-dependent component was evident, because slow inactivation was increased at depolarized VhS. 4. Perforated-patch recordings were used to study at least four pharmacologically distinct fractions of HVA currents in both young and aged cells. Nifedipine (10 microM) blocked 16.9 +/- 2.8% and 23.6 +/- 2.5% of the HVA currents in young and aged cells, respectively. omega-Conotoxin GVIA (500 nM) blocked 53.2 +/- 5.8% in young and 53.6 +/- 2.9% in aged cells. In young cells, omega-agatoxin IVA (200-400 nM) blocked 28.4 +/- 2.2% of the HVA current, and it blocked 29.9 +/- 2.8% in aged cells. A fraction of the current (young cells: 13.8 +/- 2.2%; aged cells: 11.4 +/- 1.6%) was resistant to a combination of all three antagonists. Cadmium (100 microM) completely blocked the remaining HVA current. No significant age-related differences in the HVA current fractions were observed. 5. The HVA current density, current-voltage relationship, and voltage-dependent activation were unchanged with age. However, slow inactivation of HVA currents was reduced in aged cells. The age-related difference in HVA Ca2+ currents reported here suggests a possible mechanism by which Ca2+ homeostasis may be altered in aged neurons.