COBE PICKS OUT PRIMORDIAL SEEDS AT LAST , at a level of one part per 100,000. Prevailing theories of cosmology, such as the inflationary big bang model, predict that measurements of the cosmic microwave background (CMB) should reveal small fluctuations in the temperature of the universe across the sky. These
fluctuations---patches of sky with temperatures slightly higher or lower than surrounding areas---correspond to regions of the early universe with slightly greater or lesser concentrations of matter. This pattern of matter would serve as a sort of template for cosmic evolution. Small as they are, the observed patches of enhanced density are thought to have provided enough of a gravitational "valley" into which surrounding matter could collect. This process would presumably lead after billions of years to the clustering of galaxies we see today.
Until now, instruments measuring the CMB, including the Cosmic Background Explorer (COBE), were unable to see the template. This dearth of measurable structure in the early universe (about 300,000 years after the big bang) was beginning to make theorists nervous; how could the modern universe have produced galaxies without there being precursors, however faint they may have been?
This changed yesterday when, at the APS April Meeting in Washington, D.C., the first observation of temperature variations in the CMB was announced by COBE scientists. And the fluctuations were faint indeed: roughly 30 microkelvins on top of an average sky temperature of 2.73 kelvins. George Smoot of LBL, head of the team which operated the Differential Microwave Radiometer (DMR), one of three principal instruments on board COBE, reported that these results were based on the meticulous computer analysis of a year's worth of data. This analysis had to carefully subtract competing microwave emissions from such nearby objects as the Earth and the Milky Way galaxy, and to take into account the effect of the Earth's movement (and that of our galaxy) through the universe, a motion that imposes a dipole shift on the pattern of temperature measurements across the sky.
Edward Wright of UCLA asserted that the new results support some cosmologies, such as those that prescribe the existence of dark matter, and rules out others, such as those suggesting the existence of "textures," wrinkles in space-time. Without dark matter, for example, the temperature (or equivalently matter density) fluctuations should have been larger than those seen.
Smoot and his colleagues were not lacking for confidence in their measurements, but an enlarged data sample would greatly help to sharpen the primordial structure pattern. Another COBE scientist, John Mather of NASA Goddard, said that as analysis techniques became better understood, the presentation of data would speed up. He estimated that two or even three years' worth of data would be analyzed by next year.
