When I was teaching introductory astronomy laboratories, one of the fun exercises we did was to give the students some sheets of paper and safety scissors (yay for regression therapy!). The sheets of paper had a map of the Galaxy on
it which you could cut out and fold into a proper 3-D map of the Milky Way. The lab exercise wasn't just to cut and fold, but they had to also color in the various types of objects visible in our sky as they appear to us. What is the distribution of star-forming regions? Of globular clusters? Of the "spiral nebulae"? When they got done coloring and cutting and folding and whatnot, they'd hopefully be able to see what the distribution of the different objects in space was like. In so doing, you might learn something about their physical properties and their history.
The open clusters and bright nebulae tend to lie close to
the plane of the Milky Way. They are part of the Galactic disk, so it isn't too surprising that they might be found more abundantly there. The globular clusters are distributed throughout the Milky Way, even outside it a bit, though they tend to cluster most strongly near the Galactic center. It turns out that the "spiral nebulae" -- what we used to call galaxies external to the Milky Way -- are least abundant near the plane. Since they're external to the Milky Way, and (as far as we know) distributed isotropically throughout the universe, they shouldn't have any correlation at all with the plane of the Milky Way. So why do they?
The zone of avoidance is the observed deficit of external galaxies along the plane of the Milky Way. The light of the external galaxies is obscured by dust within the plane of our
galaxy. The galaxies really are there, we just can't see them.
Assume we sit in the middle of a thick, circular disk of uniformly distributed dust particles. If we look directly up (or down) perpendicular to the disk, our view is only modestly obscured. Light gets dimmed a little bit (defined by convention as an amount τ) but not much. Now as we lower our gaze toward the plane of the disk, we have to look through more and more dust, until finally we have
to look through the entire disk. The more dust between us and the source, the less light we see.
Now assume that our disk is surrounded by external galaxies,
equally distributed in space around us. What do we then
see? In quantitative terms, if an external galaxy has a magnitude m, its light will be dimmed by an amount equal to tau times the cosecant of the angle b away from the plane (parallel to the plane,
b = 0; perpendicular to it, b = 90). Since dimmer objects have larger magnitudes, this
is written quantitatively as
mobserved = mactual +
(1.086 τ csc b)
(The 1.086 is a numerical constant which comes from the conversion of measured light flux from linear units into logarithmic magnitudes -- it is 2.5 × log10e.)
As your line of sight gets closer and closer to the plane, b goes to zero, which makes the cosecant of b infinite. The external source is essentially invisible.
As a result, we see fewer external galaxies along the plane of the Milky Way than we do away from the plane.
Edwin P. Hubble was the first to correctly explain the zone of avoidance as being caused by dust distributed throughout the plane of the the Milky Way. The zone of avoidance was first noted by Frederick H. Seares of the Mount Wilson Observatory in
a 1925 paper entitled "Note on the Distribution and Number of Nebulae" (Astrophysical Journal, v62, 168). He found a deficit of spiral and elliptical nebulae
at low galactic latitudes. Hubble published several works on the distances and distribution of "extra-galactic nebulae" in the decade to follow. He fully developed and expanded upon Searle's ideas in the paper "The Distribution of Extra-Galactic Nebulae" in 1934 (Astrophysical Journal, v79, 8) in which he uses the term zone of avoidance for the first time.
The zone of avoidance gave rise to an important question in
the early part of the twentieth century, when astronomers were trying to figure out exactly what
spiral nebulae were. In
the Shapley-Curtis Debate of 1920,
Shapley argued (incorrectly) that the zone of avoidance was evidence that the spiral nebulae were a part of our own
Galaxy. Why else would their distribution depend upon
their location relative to the plane? But Curtis countered
that if our Galaxy were filled with obscuring material (or
as Curtis believed, perhaps surrounded by a ring of
obscuring material), the light from external, distant
spiral nebulae lying near the plane of our galaxy would
be dimmed. The latter turned out to be correct.
References: Galactic Astronomy by D. Mihalas and J.
Binney, Astronomy of the 20th Century by O. Struve and V. Zebergs, Edwin Hubble 1889-1953 by
Allan Sandage (Journal of the Royal Astronomical
Society of Canada v83, 351), and