[en] Experiments to measure the mass of the electron antineutrino from tritium β-decay are reviewed. The spectrum shape and various distributions, needed for the analysis of a measured β-spectrum, are discussed. Their relative importance is pointed out and possible systematic errors are discussed. Recently published measurements gave no indication for a non-zero mass and upper limits in the range 9.3 to 15.4 eV were claimed. These experiments are critically reviewed. It is concluded that the only previous experiment suggesting a mass of about 30 eV, must be wrong. Possible reasons for this are discussed. A brief survey of the masses of the other neutrino types and of neutrino mixing is given and some recent results are summarized. Also the experimental situation of Simpson's 17 keV neutrino is discussed. (author)

[en] A search for the 17 KeV neutrino has been performed by measuring the β-spectrum of ⁶³Ni with a high resolution magnetic spectrometer. No indication for a nonzero mixing probability |U_ex|² of a heavy neutrino with a mass m_ν in the range 12 to 22 keV was found. The data show a high degree of internal consistency and, for m_ν=17 keV, the result, |U_ex|²=(0.37±0.55)*10^-3, is compatible with zero within statistics. We conclude that the 17 keV neutrino does not exist with the parameters claimed by several other groups, and thus most likely, does not exist at all. (authors). 10 refs., 4 figs., 2 tabs

[en] The usual two-frequency method to determine the muonium hyperfine frequency γ₀ suffers from lack of precision. It has not been possible with this method to measure the dependence of γ₀ on external parameters such as temperature and pressure. A μSR-apparatus of high time resolution has recently been built which allows the direct measurement of γ₀ for muonium in Si and Ge. (Auth.)

[en] The direct measurements have so far given no indication for a nonzero (positive) mass of any of the three known neutrinos. The experiments measuring the tau and the muon neutrino are good shape. The tritium experiments are in an unfortunate situation. It is unclear to me whether the problems are experimental or theoretical or a combination of both. The electronic final states distribution have been calculated, but the results have never been tested experimentally. The most important question to be answered is about the validity of the sudden approximation. (author) 9 figs., 2 tabs., 16 refs

[en] A detailed description of a Tret'yakov-type magnetic β-spectrometer with electrostatic acceleration is presented. The instrument is being used for high precision measurements of the β-spectra of various isotopes and to search for neutrino mixing and heavy neutrinos with masses up to 100 keV/c². The relative momentum resolution, obtained with an extended source (3x2 cm²) and a large silicon strip detector, is δp/p=6.9x10^-4 (FWHM). The width of the energy band, which can simultaneously be measured with this resolution, is typically 50 times larger than the corresponding width of the energy resolution. The detector delivers pulse-height and one-dimensional position information. The read-out electronics provides a separate channel for each detector strip, allowing measurements to be made with high count rate and small dead-time losses. The detection of scattered electrons was minimized by the arrangement of baffles and by the use of materials with small atomic number. Experimentally, scattering in the spectrometer was investigated by accelerating photoelectrons up to 60 keV energy. A system for measuring the corresponding high voltage is described. Extensive Monte Carlo simulations of scattering in the spectrometer have been performed. Good agreement with the experimental data was obtained and there was no need to adjust any scattering cross section to the particular experimental situation. We conclude that calculated scattering probabilities can reliably be used in the analysis of experimental β-spectra, which is important as otherwise scattering would constitute a major uncertainty

[en] A new method for the measurement of hyperfine frequencies of muonium is described and applied to muonium in quartz. With an apparatus of high time resolution (FWHM=150 ps), measurements were performed in the temperature range 30-300 K. The results are compared with EPR-studies on atomic H in quartz. (Auth.)

[en] The hyperfine frequency ß⁰ of normal muonium Mu has been measured as a function of temperature in zero magnetic field with an apparatus of high time resolution. Compared with the usual two-frequency method, the direct measurement of ß⁰ typically gives more accurate results by a factor of 10. For Mu in Ge, ß⁰ shows a monotonic decrease with increasing temperature which is well described by a Debye model. For Mu in Si, ß⁰ exhibits a nonmonotonic temperature variation with a maximum at 80 K. By a measurement with external pressure up to 1.5 kbar it is shown that the nonmonotonic lattice expansion of Si cannot cause the observed effect. Rapid relaxation of the direct transition was observed in Si below 20 K. A similar anomaly of the triplet transition signals in a magnetic field could not be found. A model is presented which assumes diffusion of Mu over slightly different sites and which explains all observed effects consistently

[en] A high time resolution μSR apparatus is described. The set up is mainly used for precision measurements of the muonium hyperfine interaction. (Auth.)

[en] The hyperfine frequency of muoium in Si in zero magnetic field has been measured by the muon spin rotation technique using an apparatus of high time resolution. The measured frequency ν₀ = (2011.8 +- 0.5) MHz is compared with the value extracted from triplet muonium precession in an external magnetic field. (orig.)

[en] Measurements of the tritium β-spectrum are reported. Thin sources consisting of a monolayer of tritiated hydrocarbon molecules were used. No indication of a nonzero mass m_ν of the electron antineutrino was found. The result is m_ν²=-24±48±61 eV² (1σ). An upper limit m_ν<11 eV (95% confidence level) is derived. (orig.)