Recently one of us (STENFi_O, 1969) described a method of measuring longitudinal solar magnetic fields by means of fringes introduced in spectroheliograms by alternately using a+ and -2/4 plate in front of the entrance slit. A similar result can be obtained by replacing the varying 2/4 plate with a stationary Babinet cornpcnsator whose fringes are parallel to the spectrograph dispersion. Polarization of the incident light will result in a fringe pattern in the spectroheliogram. A magnetic field on the sun will, therefore, show up by the presence of fringes. We have tried this technique at Sacramento Peak with good success. There are signilicant advantages in using a Babinet compensator:
(1) The width of the spectroheiiograph exit slit is not restricted by the fringe separation.
(2) The fringes are at right angles to the spectroheliograph slit so that fringes introduced by irregularities in the spectrohcliograph motions have less influence.
(3) Thc Babinet compensator method can also be used with a narrow band filter.
(4) The position of the fringes contains additional information on the linear polarization of the light. This, of course, necessitates the careful elimination of any linear polarization originating in the instrument. In the modulating ;./4 plate technique this is not necessary.
Many modifications of the technique using a Babinet compensator are possible. For example, two compensators in series, with an angle betwcen the birefringent axes r or 90 ~ (e.g. 45:'), will give a two-dimensional fringe pattern on the sun from which all the Stokes parameters and, hence, the vector magnetic field can be determined. This technique will be of special interest when using birefringent or Fabry-P6rot liltcrs. Babinet compensators and other devices behaving similarly have been used in the past OnmAtq, 1950) in experiments related to the one described * Presently Guest Investigator at the Mt. Wilson and Palomar Observatories.