The solar abundance of helium (more specifically the ratio of solar helium to hydrogen) is a basic quantity in understanding many astrophysical and space physical problems. However, in spite of a great deal of work on the subject, it has proved to be very difficult to accurately determine the helium abundance of the sun. We here review critically the four methods that have been used to estimate the ratio of helium to hydrogen; the solar neutrino flux, spectral intensity of helium lines in prominences and the chromosphere, elemental abundance of solar cosmic rays, and variations of solar wind He/H. At present the measured value for the neutrino flux cannot be explained by any of the theoretical models of the sun, and thus the neutrino measurements cannot yield an estimate of He/H. Helium line intensity measurements give the most accurate values for He/H now available, i.e., 5–8% helium by number. However, these results refer to the chromosphere and prominences only, and questions of spatial and temporal variations of the observed quantities and of the inferred helium abundances have not yet been dealt with adequately. The cosmic ray observations are usually interpreted as indicating 6% helium at the flare site. However, this interpretation is here called into question because of an apparent energy dependence of the observations. The variations of He/H in the solar wind yield information on the coronal distribution of helium, rather than the solar abundance. The solar wind observations are most easily interpreted in terms of a moderate helium enhancement (He/H ∼ 10–15%) in the deeper levels of the solar atmosphere, but a uniformly mixed corona cannot at present be completely ruled out. It is concluded that the solar abundance of helium remains uncertain to within a factor of 2