Introduction:  Widespread anecdotal recommendations for visible-light (VL) curing through enamel (E) and dentin (D) presume facilitated transmission.  Objectives:  Test if VL transmission through human E and D is better than through four popular dental composites.  Methods: Composite disks (n=4/material; f=10mm; thickness=1.00±0.04mm) were molded (split molds; Mylar-covered faces), cured (40s; Demetron 501, output=700 mW/cm2), stored (DI-water, 25C, 24h), polished (600, 1200-grit SiC), and analyzed (Argon Ion and He-Cd lasers) for total transmission (T=I/Io) to test two conventional hybrids (Z100[3M]=Z; Herculite-XRV[Kerr]=H) and two mini-hybrids (Esthet-X[Dentsply]=X; Point-4[Kerr]=P).  Within composite disks, location-to-location variations  (n=4/disk) and sample-to-sample variations (n=3 disks) in transmission were extremely small (£0.03%) so all measurements were pooled (n=12).  Results were compared to 1mm thick facial (f) or sagital (s) sections of incisal E or D.  1mm versus 2mm thick H disks also were analyzed.  Transmission curves were smoothly changing, and so test-group results for transmission (%) were compared (ANOVA, t-test, p£0.05) at three representative VL wavelengths (442, 488, 514.5 nm).  Results: Transmission within different-thickness composites was less than an exponential decrease with increasing thickness (at 488 nm, H=37±2% for 1mm vs 25±2% for 2mm) indicating that multiple scattering (diffuse>>direct) was occurring.  Transmission always increased with wavelength (as expected). Statistical results were identical among groups at all wavelengths.  At 488 nm, transmission values (mean±s.d.) for composites (Z=40±2%; H=43±2%; P=40±2%; X=32±2%) were similar to tooth structure (fE=42±11%; sE=38±6%; fD=33±4%; sD=33±7%).  X was worst (p£0.001) and H was best (p£0.01) for transmission among composites.  VL transmission through fE was no different (p>0.05) than best composites (Z,H,P).  VL transmission through fD and sD was no different (p>0.05) than the most difficult composite (X).  Conclusion:  Enamel transmission of VL during curing was no higher than through conventional hybrid composite, and transmission through dentin appeared worse.