Cadmium metal nanocrystallites (NCs), prepared on graphite surfaces by electrochemical deposition, are employed as precursors to synthesize core-shell nanoparticles consisting of a crystalline cadmium sulfide (CdS) core and a sulfur or polysulfide shell. Core-shell NCs having a large CdS core (radii R-Cds > 40 Angstrom) were prepared by exposing electrodeposited cadmium particles (R-Cd > 25 Angstrom) to H2S at 300 degrees C, whereas nanoparticles having a smaller CdS core (down to 17 Angstrom) were obtained from cadmium precursor particles via a Cd(OH)(2) intermediate. For both large-core and small-core CdS nanoparticles, the addition of the sulfur capping layer (ranging in thickness from 5 to 30 Angstrom) occurred during exposure to H2S st 300 degrees C. Transmission electron microscopy (TEM) and selected area electron diffraction (SAED) data show that the synthesis of CdS NCs proceeded on a particle-by-particle basis such that the particle size and monodispersity of the CdS core were directly related to those of the cadmium metal precursor particles electrodeposited in the first step of the synthesis. The CdS cores of these particles were found by electron diffraction to be epitaxially aligned with the hexagonal periodicity of the graphite surface and oriented with the c-axis of the wurtzite unit cell perpendicular to the surface. The low-temperature photoluminescence (PL) spectra for CdS nanocrystals without the sulfur capping layer were dominated by broad trap state emission peaks. In contrast, the PL spectra for sulfur-passivated CdS NCs were characterized by a prominent exciton emission band and much weaker trap state emission peaks. As the radius of the CdS core was reduced from 50 to 17 Angstrom, the energy of the exciton emission peak shifted from the macroscopic value of 2.56 to 3.1 eV in excellent agreement with the predictions of the Coulomb-corrected, effective mass model.