Previous findings showed allocentric and egocentric learning deficits in rats after MDMA treatment from postnatal days (PD) 11-20 but not after treatment from PD 1-10. Shorter treatment periods (PD 1-5, 6-10, 11-15, or 16-20) resulted in allocentric learning deficits averaged across intervals but not for any interval individually and no egocentric learning deficits individually or collectively. Whether this difference was attributable to treatment length or age at the start of treatment was unclear. In the present experiment rat litters were treated on PD 1-10, 6-15, or 11-20 with 0, 10, or 15 mg/kg MDMA q.i.d. at 2-h intervals. Two male/female pairs/litter received each treatment. One pair/litter received acoustic startle with prepulse inhibition, straight channel swimming, Cincinnati water maze (CWM), and conditioned fear in a latent inhibition paradigm. The other pair/litter received locomotor activity, straight channel swimming, Morris water maze (MWM), and locomotor activity retest with MK-801 challenge. MDMA impaired CWM learning following PD 6-15 or 11-20 exposure. In MWM acquisition, all MDMA-treated groups showed impairment. During reversal and shift, the PD 6-15 and PD 11-20 MDMA-treated groups were significantly impaired. Reductions in locomotor activity were most evident after PD 6-15 treatment while increases in acoustic startle were most evident after PD 1-10 treatment. After MK-801 challenge, MDMA-treated offspring showed less locomotion compared to controls. Region-specific changes in brain monoamines were also observed but were not significantly correlated with behavioural changes. The results show that PD 11-20 exposure to MDMA caused the largest long-term cognitive deficits followed by PD 6-15 exposure with PD 1-10 exposure least affected. Other effects, such as those upon MK-801-stimulated locomotion showed greatest effects after PD 1-10 MDMA exposure. Hence, each effect has a different window of developmental susceptibility.