[en] The KARMEN (KArlsruhe/Rutherford Medium energy Neutrino) experiment at the UK Rutherford Laboratory's ISIS accelerator/neutron source began running in 1990, and among other contributions has provided the first examples of nuclear excitations via neutral current interactions. As neutrino statistics continue to pile up, intriguing new effects appear to emerge from the data. ISIS fires a dense 800 MeV proton beam into a thick target of heavy metal, producing neutrons by spallation. While most of the ISIS protons convert into neutrons, about one in twenty radiates a charged pion, which decays to produce a muon and a muon neutrino. In contrast to a normal neutrino experiment, where the neutrinos are produced by the decay of secondary pions (and kaons) in flight, KARMEN's pions get no further than the dense neutron source/target. This acts like a 'beam stop', absorbing most of the secondaries usually seen in high energy reactions, leaving only highly penetrating particles, like neutrinos, which pass through material relatively unscathed. Captured in the neutron production target, the secondary charged pions decay at rest, producing muons and monoenergetic (30 MeV) muon-type neutrinos. The former subsequently decay too, producing electrons and both muon- and electron-type neutrinos. With three emerging particles rather than two, the energy of the neutrinos from muon decay has a continuous spectrum (as in beta decay), cutting off at 53 MeV. KARMEN's 56-tonne liquid scintillation calorimeter, 17.5 metres downstream from the target assembly and swathed in its 6000 tonne iron mantle to screen off background effects, monitors the emerging particles. During normal running, about 1 or 2 neutrinos per day are intercepted, with some 1500 neutrino counts having been amassed so far. By comparing the arrival time of the neutrino counts with the proton clock, neutrinos from the two different sources can be separated. The electron neutrinos from muon decay broadly reflect the 2.2 microsecond muon lifetime, but the experiment sees a slight shoulder in the decay spectrum, broadened towards 2.7 microseconds. With neutrinos reluctant to interact with matter, neutrino data are notoriously difficult to amass, so that results are continually at the mercy of statistical fluctuations. However the delayed shoulder in the KARMEN spectrum has now lasted for several years. It could mean that something else, a massive neutrino-like particle produced in a rare form of pion decay beyond any conventional theory, is making its presence felt. Only several years more neutrino running will tell