An analogy would be when you're scanning the radio dial and you get the same station separated by a small blank space. The size of the blank space is directly proportional to the strength of the magnetic field at the location in space where the station is being broadcast.

This hydrogen gas emits radio waves with a wavelength of 21 centimeters, so they come in at about 1,420 megahertz on the radio dial. We were able to find that there is a magnetic field permeating this gas cloud and that its direction is pointing towards us above the cloud and then flips to pointing away from us on the bottom.

It took hundreds of hours of telescope time before we were confident that we could do reliable science.

You can think of this structure as a giant, magnetic Slinky wrapped around a long, finger-like interstellar cloud. The magnetic field lines are like stretched rubber bands; the tension squeezes the cloud into its filamentary shape.

The telescope was designed to be very sensitive and have a very clean response to signals from the sky.

These results were incredibly exciting to me for a number of reasons. There's the scientific result of a helical field structure. Then, there's the successful measurement: This type of observation is very difficult, and it took dozens of hours on the telescope just to understand how this enormous dish responds to the polarized radio waves that are the signature of a magnetic field.

First, we spent hundreds of hours since January 2003 trying to characterize how the telescope responds to radio waves.

We combined radio measurements we made of the Zeeman effect and optical measurements of magnetic fields from starlight.

What we see strongly suggests that the magnetic field is indeed helical. If it is so, then this tells us a lot about the role magnetic fields might play in molecular clouds and star formation. It would be a major piece of evidence in trying to solve this mystery of how stars form.

We wanted to make very high-resolution slices across the filament and see if the field really did reverse; if it did, then it's very likely that the field is somehow tied up in the filament.