Using NASA's Hubble and FUSE space observatories, Shull and Danforth examined light from 28 distant quasars scattered across the night sky.
As some of the quasars' light travels through space, it pierces filaments of dark matter and gas.
Atoms of neutral hydrogen and charged oxygen clustered around the filaments absorb portions of the quasar's ultraviolet light, creating dark bands in the spectrum that reaches Earth.
By analyzing this altered light, scientists can determine the position of a filament and the amount of normal matter gathered around it.
The work is described in the May 20 issue of The Astrophysical Journal.
Ken Sembach is an astrophysicist at the Space Telescope Science Institute in Maryland who was not involved in the study.
"This is a truly amazing achievement of the FUSE and Hubble observatories," Sembach said.
"Until those two observatories started looking, we had no way to test whether these [dark matter] models that people take for granted are right."
Half Still Missing
Scientists think dark matter filaments are part of a larger cosmic web connecting vast dark matter isles.
Together the filaments and isles form a hidden support structure for galaxies that could be likened to a dense cluster of brain cells connected by gangly appendages.
Normal matter is drawn toward this dark cosmic web by gravity and flows within and around it like electric impulses coursing through neurons.
Where normal matter concentrates within the web, galaxies and galaxy clusters form.
The new observations confirm that much of the universe's baryons flow through the filaments.
"What we're seeing is stuff that hasn't fallen into galaxies yet but is strewn out along the web," study co-author Shull told said.
But even with the new discovery, more than half of the universe's baryonic matter is still unaccounted for.
Scientists think the remaining missing matter most likely exists in the form of extremely hot gas that also floats between galaxies.
This gas is heated to millions of degrees and will require future x-ray telescopes to detect.
"That's where the theoretical models tell us it should be," Shull said.
"If we build a bigger telescope in the next decade for x-rays and we don't see it, that will be very difficult to understand."
SOURCES AND RELATED WEB SITES