Ageing is associated with cerebral atrophy and accumulation of cerebral β-amyloid. We examined the relationships between β-amyloid load and brain volume in a large cohort of individuals born in the same week in 1946. We analysed cross-sectional data from Insight 46; a sub-study of the MRC National Survey of Health and Development(NSHD) which includes individuals born in mainland Britain in one week in 1946. Participants had 18F-Florbetapir PET and volumetric T1-weighted MRI on the same 3T-PET/MR unit. β-amyloid load was quantified as a cortical standard uptake value ratios (SUVR) using a cerebellar reference region. The relationship between automated measurements of whole brain, ventricle, hippocampal and lobar volumes, age at time of scan and β-amyloid load was examined using regression models and voxel based morphometry, incorporating total intracranial volume and gender as co-variates. Data from 210 individuals without dementia or significant neurological/psychiatric illness were included. Mean age at scanning was 70.1±0.39 years (range 69.3–70.8). 38 individuals were β-amyloid positive: defined as SUVr=>1.078. Comparison of positive and negative groups revealed no difference in volumetric measurements. However, within β-amyloid-positive individuals, SUVR was negatively associated with total brain volume(p=0.034). There was also evidence of an interaction between β-amyloid status(positive/negative) and SUVR in terms of their effects on total brain volume(p=0.006) across the whole sample. Age at scanning was associated with increased ventricular volume (p=0.014) and with lower total brain(p<0.001), hippocampal(p=0.003), frontal lobe(p=0.023) and temporal lobe(p=0.042) volumes. Whole-brain voxel based morphometry analysis(FWE corrected, p<0.05) revealed age to be associated with volume loss in fronto-temporal regions, thalamus, and cingulate gyri. These preliminary data suggest that there is no significant difference in brain volume between β-amyloid positive and negative individuals aged ∼70 years, but there is some evidence for volume loss once a β-amyloid threshold level is reached. Unexpectedly, there was also a relationship between age and total brain volume independent of β-amyloid load over a very narrow age range. Whether this reflects recruitment bias, non-AD pathologies or ageing effects is unclear. These findings will be examined in a larger sample size utilizing life course and genetic data in the future.