We report deep near-infrared (HK) imaging observations of the dark cloud as
sociated with IC 5146. With an order of magnitude greater sensitivity, we h
ave imaged roughly half the region of the cloud originally surveyed by Lada
and coworkers. Using measurements of similar to 2000 stars, we have employ
ed techniques previously developed by Lada and coworkers to construct order
ed, uniformly sampled maps of the extinction through this cloud. With the i
mproved sensitivity, we detected approximately 5 times as many heavily exti
ncted stars (i.e., A(v) similar to 20-50 mag) as found in the earlier surve
y of this same cloud area. Moreover, we were able to produce a Gaussian-smo
othed extinction map of the cloud with an angular resolution (30") somewhat
more than a factor of 2 higher than achieved in the earlier study. With th
e increased sensitivity and angular resolution we were also able to measure
the average radial column density profile orthogonal to the major axis of
this filamentary cloud. Assuming cylindrical symmetry, we modeled this colu
mn density gradient and determined that the corresponding volume density pr
ofile of the cloud must smoothly fall off as r(-2).
To investigate the structure of the cloud on size scales smaller than the e
ffective resolution of our maps, we constructed plots of the relation betwe
en the derived mean extinction and its measured dispersion for all the pixe
ls in a series of maps made with varying angular resolution. We find, simil
ar to Lada et al., that the dispersion increases linearly with mean A(v), i
ndependent of the angular resolution of our maps. However, although we quan
titatively reproduce the earlier results at the same angular resolution (90
"), we find the interesting result that the slope of the sigma(Av)-A(v) rel
ation decreases in a systematic fashion with increasing angular resolution.
We construct synthetic models of the cloud density distribution and use Mo
nte Carlo techniques to produce artificial extinction maps and investigate
the origin of the sigma(Av)-A(v) relations. These models show that both the
observed form of the sigma(Av)-A(v) relation and its variation with angula
r resolution are the natural consequences of a smooth, radially decreasing
volume density gradient in a cylindrically symmetric cloud. For a volume de
nsity gradient falling off as r(-2), the quantitative agreements between th
e model predictions and data are excellent. Apparently, these relations can
be understood without the need for random fluctuations in the structure of
the cloud on small spatial scales.