It has been suggested that the active site of acetylcholinesterase contains a hydrophobic binding site (HBS-1), which is closely adjacent to both the anionic and the esteratic sites. In this paper, we assumed that there exists another hydrophobic binding site (HBS-2), some distance removed from the anionic site. On this assumption, a new working hypothesis was proposed for the design of acetylcholinesterase inhibitors. A series of 2-[omega-[N-alkyl-N-(omega-phenyl-alkyl)amino]alkyl]-1H- isoindole-1,3(2H)-diones was designed based on this hypothesis and tested for its inhibitory activities on acetylcholinesterase. Some in this series were revealed to be more potent than physostigmine. Optimum activity was found to be associated with a five carbon chain length separating the benzylamino group from the 1H-isoindole-1,3(2H)-dione (phthalimide) moiety. Quantitative study of substitution effect on the phthalimide moiety revealed that hydrophilic and electron-withdrawing groups enhance the activity.