We show by examples that multichannel mixing can affect both the parameters extracted from neutrino oscillation experiments, and more general conclusions derived by fitting the experimental data under the assumption that only two channels are involved in the mixing. Implications for MiniBooNE are noted and an example based on maximal $CP$ violation displays profound implications for the two data sets (${\nu }_{\mu }$ and ${\overline{\nu }}_{\mu }$) of that experiment.

• Received 6 April 2007

DOI:https://doi.org/10.1103/PhysRevD.75.091301