To address the problems related to the transfer of helium and argon isotopes from rocks into related aquifers, the concentrations of U, Th, Li, and K and helium and argon isotopes were measured in a sedimentary sequence from Tertiary to Permo-Carboniferous and in crystalline rocks in northern Switzerland. In addition to whole-rock samples, mineral separates have also been investigated. The observed concentrations of 3He, 4He, and 40Ar in rocks, minerals and groundwaters are compared to calculated values which would result from in situ radiogenic production in a closed system. This comparison shows that the rocks and minerals have almost completely lost radiogenic in situ produced helium. Loss of 3He and 4He has been controlled by different retention capacities of the rock- forming and accessory minerals and fillings. Therefore, measured 3He/ 4He ratios in these rocks can differ from the calculated production ratios. The deviation is generally less than a factor of 3. However, in certain chemical sediments, e.g., anhydrites, elevated ratios of 3He/ 4He up to 15 × 10 −8, which is 20 times the production ratio, were observed. It is postulated that 3H, the precursor of 3He, might be chemically bound in such sediments during the 12.3 years half-life of 3H and that the beta decay energy of 3H is too small to liberate 3He. The calculated closed-system concentrations of helium in groundwater exceed the measured ones by up to three orders of magnitude, implying that the water/rock system has not been a closed system since the time of sedimentation and that movable waters have removed helium to a discharge area and to the atmosphere. Stagnant old waters, however, could contain high concentrations of helium and supply it to younger movable waters. Aquitards with low permeability (Permian shales) appear to be a more important source of helium isotopes than water-bearing rocks (Permian sandstones). High abundances of all parent elements in the shales, almost complete loss of radiogenic helium, and the enhanced production ratio of 3He/ 4He = 7.2 × 10 −8 in these rocks, similar to the ratio observed in the neighbour Permian aquifer, 9.4 × 10 −8, suggest the shale as a major source of helium isotopes for this aquifer and diffusion as a principal process for helium transfer from the stagnant porewaters in shales into the movable waters in sandstones. Some sedimentary rocks, e.g., Permian sandstones and shales, contain more helium than could have been produced since the time of sedimentation. While Permian sandstones mainly have retained this excess rare gas component, they have released radiogenic helium. The difference in the retention of these two components most probably demonstrates that migration of radiogenic helium from damage tracks along crystal imperfections to grain boundaries is an important process controlling its loss. The abundances of excess 3He and 4He in rocks cannot be predicted by in situ production calculations; studies of rocks and minerals are required. The average ratio of 40Ar/ 36Ar = 385 in groundwaters collected from Permian sediments from the Weiach borehole is much higher than that in the aquifers from the crystalline basement below the sediments. Radiogenic 40Ar* in Permian sediments is mainly liberated by diagenetic water-rock interactions. The low contribution of radiogenic 40Ar* in deep aquifers of the crystalline basement and the fact that chlorine concentrations in groundwaters from the basement are also much lower than in the overlaying Permian aquifer eliminates a hypothetical upwards fluid flux. From this study we envisage intrabasin sources for helium and argon isotopes in groundwaters with mainly lateral migration of waters through sedimentary layers at different rates and diffusion of helium from “slower-velocity” older waters in more stagnant zones (shales) into flow-passes with “highervelocity” younger waters in sections with higher permeability (sandstones). An external source for noble gases is not required in such a model.