Now here’s something you don’t see every day:

Color composite produced by Travis Rector, University of Alaska Anchorage.
Credit: Gemini Observatory/AURA Get the 3844×2444 version!
This is NGC 660 as seen by the Gemini North observatory at Mauna Kea, Hawaii and before we go any further, you definitely want to grab the full-resolution version. NGC 660 is an example of a rare polar-ring galaxy – that is, a galaxy surrounded by a ring of stars that rotates over the poles of the galaxy. First we have the main galaxy itself, seen edge-on to us in the middle. The host galaxy has a very thick central bulge, which classifies it as a lenticular galaxy.
Surrounding the central galaxy is a ring of stars and gas. And the ring itself is bursting with star formation! Take a look in that full-resolution version and you’ll see nebulae illuminated by hot young stars, and bubbles blown out by massive stars that went supernova.
Of course, neither the galaxy nor the ring are solid, and both are warped by their mutual tidal forces on each other. Just wow!
So how did this whole thing come to be? Polar-ring galaxies are thought to form in one of two ways: either in a head-on collision or when one galaxy rips apart a passer by, strewing the former galaxy into a ring.
In the merger scenario, one galaxy pierces the heart of another at a right angle, and the “pierced” galaxy ends up as a ring around the intruder galaxy. In such a scenario, you end up with a collapsed core and a burst of star formation, and NGC 660 certainly has both.
But there are some compelling reasons why the piercing scenario may not be the case here. For one, the galaxy and its ring aren’t at right angles to one another, but rather at roughly 45-degree angles. For comparison, here’s NGC 4650, another polar ring galaxy that was probably formed as a result of such a collision:

Another problem is that in a piercing scenario, the gas gets concentrated into the host galaxy’s central region, while the ring is left largely stripped of this gas. In NGC 660, not only is there a lot of gas and star formation taking place in the ring, but the host galaxy itself is “thick” with gas, and there’s quite a bit of star formation going on in there as well. In addition, ring itself is tilted about 45-degrees. Simulations of piercing mergers don’t get us gas-rich, 45-degree ring that we see.
That leaves us with with another possibility, called tidal accretion. In other words, the host galaxy “shredded” a lower-mass passer-by galaxy into the ring. There’s some compelling evidence for that here. In tidal accretion, you can have a wider distribution of gas both in the ring and in the host galaxy itself, allowing for the star formation we see in both. It is also possible for the ring to be oriented at any angle with respect to the host, since you don’t have to start out with a right-angle collision. Finally, in a right-angle collision, you’d expect the pierced galaxy’s nucleus to fall inward toward the host, giving you a “double nucleus” at the center. But NGC 660 only has its original nucleus.
That means that we’re likely looking at the aftermath of a lower-mass galaxy that passed close to NGC 660 long ago, and lost a lot of its stars and gas to the larger galaxy as it passed by. NGC pulled this “chunk” of the passer-by into a ring and both the ring and the galaxy tidally distorted themselves into the shapes we see today.
There’s one more very cool thing going on here, hidden from our view but bright at radio wavelengths is a compact , 32 light-year diameter source near the center of NGC 660. This source is most likely a super cluster of stars in a dense cloud of dust and gas, hidden from view but transparent at radio. This source contains perhaps a few thousand hot, blue youthful stars.
Whatever caused the ring, it led to a lot of star formation in NGC 660, making this already rare polar-ring galaxy even rarer – a polar-ring starburst galaxy! Very cool!
It’s truly amazing how much we can figure out about the origins and evolution of galaxies “just” by looking at them and critically examining their features. The answers don’t jump out at us, but we can take what we see and propose models to explain them. Those models that don’t explain the observations are discarded, while those that do are refined to come up with a picture that better matches what we see. This is how science – astronomy in particular – works!