We postulate that zinc(II) is a keystone in the structure of physiological mouse copper metallothionein 1 (Cu-MT 1). Only when Zn(II) is coordinated does the structure of the in vivo- and in vitro-conformed Cu-MT species consist of two additive domains. Therefore, the functionally active forms of the mammalian Cu-MT may rely upon a two-domain structure. The in vitro behaviour of the whole protein is deduced from the Cu titration of the apo and Zn-containing forms and compared with that of the independent fragments using CD, UV-vis, ESI-MS and ICP-AES. We propose the formation of the following Cu, Zn-MT species during Zn/Cu replacement in Zn7-MT: (Zn4)alpha(Cu4Zn1)beta-MT, (Cu3Zn2)alpha(Cu4Zn1)beta-MT and (Cu4Zn1)alpha(Cu6)beta-MT. The cooperative formation of (Cu3Zn2)alpha(Cu4Zn1)beta-MT from (Zn4)alpha(Cu4Zn1)beta-MT indicates that the preference of Cu(I) for binding to the beta domain is only partial and not absolute, as otherwise accepted. Homometallic Cu-MT species have been obtained either from the apoform of MT or from Zn7-MT after total replacement of zinc. In these species, copper distribution cannot be inferred from the sum of the independent alpha and beta fragments. The in vivo synthesis of the entire MT in Cu-supplemented media has afforded Cu7Zn3-MT [(Cu3Zn2)alpha(Cu4Zn1)beta-MT], while that of alpha MT has rendered a mixture of Cu4Zn1-alpha MT (40%), Cu5Zn1-alpha MT (20%) and Cu7-alpha MT (40%). In the case of beta MT, a mixture of Cu6-beta MT (25%) and Cu7-beta MT (75%) was recovered [1]. These species correspond to some of those conformed in vitro and confirm that Zn(II) is essential for the in vivo folding of Cu-MT in a Cu-rich environment. A final significant issue is that common procedures used to obtain mammalian Cu6-beta MT from native sources may not be adequate.