4.1. Methodology - different methods to obtain the magnetic field strength
Most of the galaxies studied here are located nearby, well within 20 Mpc of the Milky Way. Many claims have been made to show that the magnetic field strength in late-type galaxies is close to the "equipartition magnetic field" value Beq, but one recent claim (Chi and Wolfendale, 1993) suggested otherwise. Vallée (1995a) carried out a statistical study 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.
A first method to determine the strength of a galactic
magnetic field is through the
linear polarization (Stokes Q and U) observations of the radio
synchrotron emission from
electrons trapped in the magnetic fields, with observations at
several radio frequencies between 1 and 5 GHz. The observed Faraday rotation
(
PA) of the
position angle PA of the
electric (E) vector of the radiation varies as the square of the
wavelength 
of observation
(ie: PA ~
RM 2 ),
where RM is the rotation measure which is proportional to the uniform
component of the total magnetic field strength along the line of
sight BI , the thermal
electron density n, and the path length L (i.e., RM
~ nBIL).
The fractional linear polarization
of the radiation is a function of a beam depolarization (e.g.,
Equ. 6 and 7 in
Ehle & Beck 1993)
and of a wavelength-dependent Faraday dispersion measure
(e.g., Equ. 11 in
Ehle & Beck 1993),
both occasioned by a random component of the total magnetic field. The
total magnetic field strength
Btot is the quadratic sum of the
random component Bran and the
uniform or regular component Breg (often using
Breg 
1.6 BI). Here we call Bfa the results
obtained by this Faraday method. More details on the derivation
of Bfa can be found in
Ehle & Beck (1993),
Vallée (1980),
and depolarization in
Chi et al. (1997), etc.