Of the three forms of human Apolipoprotein E (ApoE2, ApoE3 and ApoE4), the ?4 allele coding for ApoE4 is associated with a higher risk of developing Alzheimer's disease (AD) and an earlier age of onset, whereas ?2 may be protective. The mechanisms underlying such influences are still unclear. This thesis has investigated the influence of ApoE polymorphism on structural synaptic parameters in the middle molecular layer (MML) of the dentate gyrus of wild type (WT), ApoE knockout (KO) and human ApoE (hApoE) transgenic mice, from 6 to 24 months of age using unbiased stereological methods at the EM level. In hApoE4 mice, there was a 34% decrease in the synapse per neuron ratio (Syn/nrn) accompanied by a 22% increase in the mean apposition zone area (mAZA) during aging. This pattern resembles closely the synaptic changes occurring early in AD, which have been correlated to the first symptoms of memory loss in humans. In contrast. there was no such synaptic loss in hApoE2, ApoE KO and WT mice. At old age, hApoE4 mice had the lowest syn/nrn but their mAZA was comparable to that of other groups. These data appear to be consistent with the view that possession of ?4 is deleterious to cognitive functions in the elderly and AD patients. Notably, aged ApoE KO mice did not show any sign of synaptic degeneration, suggesting the involvement of other proteins to compensate for the lack of ApoE. At 18 months old, hApoE2 mice have a greater hippocampal volume and display the highest syn/nrn and glutamate immunogold labelling in presynaptic boutons and dendrites of the MML, compared to hApoE3 and hApoE4 mice. In AD patients, such effects of the ?2 allele may act as a synaptic 'reservoir' and delay the onset of AD symptoms. Thus these lines of hApoE transgenic mice could provide a good basis for the future production of multiple transgenic mice in which to model AD pathogenesis.