From a CERN press release today:
Geneva, 31 May 2010. Researchers on the OPERA experiment at the INFN's Gran Sasso laboratory in Italy today announced the first direct observation of a tau particle in a muon neutrino beam sent through the Earth from CERN, 730km away. This is a significant result, providing the final missing piece of a puzzle that has been challenging science since the 1960s, and giving tantalizing hints of new physics to come.
The neutrino puzzle began with a pioneering and ultimately Nobel Prize winning experiment conducted by US scientist Ray Davies beginning in the 1960s. He observed far fewer neutrinos arriving at the Earth from the Sun than solar models predicted: either solar models were wrong, or something was happening to the neutrinos on their way. A possible solution to the puzzle was provided in 1969 by the theorists Bruno Pontecorvo and Vladimir Gribov, who first suggested that chameleon-like oscillatory changes between different types of neutrinos could be responsible for the apparent neutrino deficit.
Several experiments since have observed the disappearance of muon-neutrinos, confirming the oscillation hypothesis, but until now no observations of the appearance of a tau-neutrino in a pure muon-neutrino beam have been observed: this is the first time that the neutrino chameleon has been caught in the act of changing from muon-type to tau-type.
Antonio Ereditato, Spokesperson of the OPERA collaboration described the development as: "an important result which rewards the entire OPERA collaboration for its years of commitment and which confirms that we have made sound experimental choices. We are confident that this first event will be followed by others that will fully demonstrate the appearance of neutrino oscillation".
"The OPERA experiment has reached its first goal: the detection of a tau neutrino obtained from the transformation of a muon neutrino, which occurred during the journey from Geneva to the Gran Sasso Laboratory," added Lucia Votano, Director Gran Sasso laboratories. "This important result comes after a decade of intense work performed by the Collaboration, with the support of the Laboratory, and it again confirms that LNGS is a leading laboratory in Astroparticle Physics".
This is good news, as it means they may be able in another year or so to collect enough data to measure the mixing between those two neutrino flavors.
You may wonder why one type of neutrino would change into another; and rightly so because this is certainly an odd sort of phenomenon.
First notice that, contrary to the accepted wisdom up until '69, neutrinos turn out to have a mass. If they had been truly massless particles, they would always move at the speed of light and never have the opportunity to turn into anything else.
Second: it isn't just neutrinos that can change type.
We find there are "families" of particles that have the same sort of arrangements. Within the "first generation" family you have the electron (charge=-1), electron neutrino (charge=0), up quark (charge=+2/3), and down quark (charge=-1/3); plus of course their antiparticles. The "second generation" family has the muon, muon neutrino, charm quark, and strange quark having the same relation; and the third has the tau, tau neutrino, top and bottom quarks respectively. These aren't the same particles, and nobody has succeeded in finding a structure to them that would make the electron a ground state, muon an excited state, and tau an even more excited state of some primal "preonic atom." People have tried (me too) and they seem to be fundamentally different.
So a meson made of a bottom quark and an anti-strange quark is different from one made of a strange quark and an anti-bottom quark, but it turns out that they can and do transform into each other at a measurable rate. The trick is that they can temporarily transform into an intermediate state, which can then transform into the other type. Notice that electric charge stays the same: 0=0. It also appears as though a particular type (flavor) of quark doesn't have a well-defined mass, and a quark with a well-defined mass doesn't have a clearly defined type. This "flavor mixing" is an area of active research, and you can easily imagine why.
Neutrinos can do the same kind of thing for apparently the same reason. The neutrino masses are small (though the tau neutrino mass isn't known well), and difference in masses between neutrino types are also very small, but the masses of their partner particles vary a lot. We don't have any neat way to make a beam of tau neutrinos, because tau's are so unstable and hard to work with. Beams of electron neutrinos from reactors (or the sun) are easy and intense and, unfortunately, of such low energy that you haven't a prayer of seeing one turn into a tau neutrino and generate a tau in a subsequent reaction. So you need beams of electron neutrinos from anti muon beams--which also give you anti muon neutrinos to contaminate the signal with. It takes some careful analysis to make sure you can tell one from the other, but muon neutrino and electron neutrino mixing has been seen.
You can pick up more details on how this is modeled, but it helps to know a little about matrices.