Mammalian metallothioneins bind up to seven Zn(2+) ions in two distinct domains: an N-terminal β-domain that binds three Zn(2+) ions and a C-terminal α-domain that binds four Zn(2+) ions. Domain specificity has been invoked in the metalation mechanism with cluster formation and bridging of the 20 Cys residues taking place prior to saturation with seven Zn(2+) ions. We report a novel experiment that examines Zn(2+) metalation by exploiting the expected decrease in K(F) at the onset of clustering using electrospray ionization mass spectrometry (ESI-MS). During the titration with Zn(2+), the ESI-MS data show that several metalated species coexist until the fully saturated proteins are formed. The relative Zn binding affinities of the seven total sites in the α- and β-fragments were determined through direct competition for added Zn(2+). The K(F) values for each Zn(2+) are expected to decrease as a function of the remaining available sites and the onset of clustering. Analysis shows that Zn(2+) binds to β-rhMT with a greater affinity than α-rhMT. The incremental distribution of Zn(2+) between the competing fragments and apo-βα-rhMT (essentially three and four sites competing with seven sites) identifies the exact point at which clustering begins in the full protein. Analysis of the speciation data shows that Zn(5)-MT forms before clustering begins. This means that all 20 Cys residues of apo-βα-rhMT are bound terminally to Zn(2+) as [Zn(Cys)(4)](2-) units before clustering begins; there is no domain preference in this first metalation stage. Preferential binding of Zn(2+) to β- and α-rhMT at the point where βα-rhMT must form clusters is caused by a significant decrease in the affinity of βα-rhMT for further Zn(2+). The single-domain Zn(5)-rhMT, in which there are no exposed cysteine sulfurs, is a key component of the metalation pathway because the lower affinities of the two clustered Zn(2+) ions allow donation to apoenzymes.