This paper was written by Alyssa Goodman et
al. and can be found here.
First, a little background: Light from stars can appear polarized. The observed polarization is
caused by dust grains along the line of sight which have been oriented
by an interstellar magnetic field. Dust grains are most effective at
polarizing light whose wavelength is comparable in size to the grain.
For this paper, surveys were done using the National Optical Astronomy
Observatory (NOAO), near-infrared array cambera (IRIM), and the
Minnesota Infrared Polarimeter (MIRP). The original goal of the paper
was "to study the magnetic field structure associated with elongated
dark clouds in star-forming regions" (748).
It was found that the situation in L1755 (the Lynds 1755 dark
cloud in Ophiuchus) is "very similar" to the situation in B216-217 (B
being Barnard). Namely, that there is little to no difference in
polarization direction and dispersion when comparing regions 'inside'
and 'outside' of a dark cloud in projection. A plot of observed
percentage polarization at J band, as a function of observed J-K color
for background stars (see FIG. 6, 755) shows that polarization does not
increase 'significantly' with increasing extinction in L1755. This is
consistent with a slope of zero in the polarization-extinction
relation, and hence regions of higher extinction within L1755 do not
have higher polarization values than regions with lower extinction
lines of sight. Other tests had similar conclusions, namely that
polarization simply does not increase with extinction in either
B216-217 or L1755.
It was seen that in most cases where the direct comparison between 'expected ' and observed K-band polarization were made, the results were disimilar, meaning that there is a large range in grain properties. In fact, various properties determine the polarization of background starlight. Here is a list of some of the important properties:
Size -grains with size approximately of wavelength L/2*pi are
best at polarizing background starlight of wavelength L
Shape -only substantially elongated grains will contribute
polarization
Composition -polarizing grains are non-metallic
Magnetic Field Structure -very complex (tangled) fields along a
line of sight will reduce the observed polarization
Three hypothesis were given to explain the lack of observed
polarization. The first being that the projected direction of the
uniform magnetic field is not affected by the presence of a dark
cloud. The second being that the number of field correlation lengths
along each line of sight varies in exact proportion to the square of
the ratio of nonuniform to uniform field energy. The last
possibility, and the one which if favored, is that the polarization of
background starlight is not a good tracer of the magnetic field in the
dense interiors of dark clouds.
It should be clear that only a small subset of interstellar grains causes the polarization of background starlight. And finally, to answer the question which is also the title of the paper, the concise and accurate response is "No"!
