1995A&A...296..819V

Recent cosmic-ray data have led to the suggestion that the magnetic field B in late-type galaxies may be quite large, i.e. B>3B_eq, where B_ eq_ is the magnetic field derived from the equipartition method (e.g., Chi & Wolfendale 1993IAUS..100..159B; Zweibel 1993Natur.362..591Z). Previous radio data had led to the claim that the magnetic field B in late-type galaxies was close to B_eq, i.e. 0.3B_eq<B<3B_eq (Duric 1990IAUS..140..235D) and 0.75B_eq<B<1.25B_eq (Hummel 1986A&A...160L...4H). It is important now to evaluate whether the real magnetic field B is far or close to B_eq. Here, a statistical study is carried out of the three methods often employed to give magnetic field strengths, namely (i) the Faraday rotation method, (ii) the Equipartition method, and (iii) the Cosmic-ray particle method. I use statistics with small observational samples to find that two independent methods (Faraday rotation B_fa and Equipartition B_eq) are converging on similar values of the magnetic field strength. For the available observational data, I find on average that 1.0B_eq<B_fa<1.2B_eq. This convergent result, by two independent methods, strongly suggests that the real magnetic field B≃B_fa≃B_eq. The third method (cosmic-ray particle B_cr) appears at times to predict magnetic field strengths different than those of the other two methods. Thus a caution is in order; possible reasons for that divergence are discussed.