Dissolved and particulate metal concentrations in the St. Lawrence estuary were determined during the high spring runoff and at a period of lower runoff in the fall. The distributions in the estuary are influenced by seasonally variable freshwater discharges and a turbidity maximum zone (TMZ). Dissolved Al, Fe, Mn, Co, Cu, Ni, and Zn concentrations generally decrease significantly with increasing salinity, but Cd shows no significant covarience with salinity. Most metals show complex distribution patterns, and only the distributions of Cu and Ni can be described by linear regression equations. Maximum concentrations of dissolved Al, Fe, Mn, and Co occur in the turbidity maximum and decrease seaward with increasing salinity. During the spring runoff period, total particulate mass was almost twice that in the lower fall runoff period. Chemical partition of the suspended particulate matter (SPM) shows that detrital particulate Fe, Mn, and Zn concentrations vary directly with SPM load and inversely with salinity. Nondetrital metal concentrations vary inversely with SPM (Mg, Pb, Cd) and salinity (Al, Fe, Mn, Zn, Cu). Such changes are not evident in the particulate composition during the low-runoff sampling period. Factor analyses imply that fine-grained aluminosilicates are the main carriers of detrital Si, Al, Ca, Cu, Fe, Mn, and Zn. Nondetrital Al, Cu, Fe, Mn, Pb, and Zn appear to be held in ion-exchange positions, and Fe–Mn grain coatings during the period of high spring runoff and for the fall data set (Cd, Fe, Al, and Pb only). Nondetrital Ca appears to carried by carbonates in the spring data set and along with mixed Ca–Mg carbonates in the fall data set. During high and low river-flow periods, riverine particulate inputs predominate and control the overall abundance, dispersal, and composition of the particulates. Data analyses indicate that, within the TMZ, internal particle cycling and resuspension combined with adsorption–desorption processes override some of the riverine effects and control or modify the particulate composition. Model calculations suggest that dissolved – nondetrital particulate metal exchanges can account for some but not all of the observed changes in the distribution of the dissolved and nondetrital phases.