We point out that the usual experimental upper bounds on the ''tau-neu
trino mass'' do not apply if neutrino mixing is considered. The suppre
ssion of the population of the tau decay spectrum near the end point,
caused by mixing, may be compensated by an enhancement because of a re
sonant mechanism of hadronization. It is necessary therefore to analyz
e the whole spectrum to infer some limit to the '' tau-neutrino mass.'
' We argue that, consequently, neutrino mixing evades the objection to
interpret the KARMEN anomaly as a heavy sequential neutrino.