The smallest of the four detectors which claim to have observed neutrinos from SN 1987a registered the events more than 4 h earlier than the other three ones. This claim is not usually accepted because it is difficult to understand that the other (an
d larger) detectors did not register any events at the same time. It is shown that microlensing of the neutrinos by a star in-between the supernova (SN) and Earth can enhance the neutrino intensity at the position of one detector by more than an order of magnitude with respect to the other detectors. Such a configuration is improbable but not impossible. Essential for this enhancement is the small source diameter, of order 100 km. So if two bursts of neutrinos were emitted by SN 1987a at a separation of about 4 h it could be explained easily that the smallest detector observed the first burst while the other ones missed it and vice versa.
