Supernova encounter of a third kind.

Supernovas are usually discovered by accident: Astronomers looking at a particular galaxy notice a bright object that was not there before. So it happened to Alexei V. Filippenko of the University of California at Berkeley and Wallace L. W. Sargent of Caltech in Pasadena. They were making a spectroscopic survey of nearby galaxies. On Feb. 28, as they observed galaxy NGC4618, their spectroscope recorded a bright object near the center of that galaxy. The object, now catalogued as SN1985f, has aroused a great deal of interest within the astronomical community. With properties quite different from those of either of the two classes of supernova that astronomers have heretofore known, it is being called the first of a third class of supernova. Moreover, its existence could answer some long-standing questions about mysterious objects called supernova remnants.

Supernovas are intense explosions of stars. The explosion causes such an increase in brightness that the supernova is conspicuous on photos of even fairly distant galaxies. (A supernova in our own galaxy would probably be obvious to everybody without need for telescopes, but we haven't seen one of those in more than 300 years). These explosions are characteristic of the last stages of a star's life cycle.

Astronomers believe that type I supernovas occur when a dying white dwarf star with a mass about that of the sun is overwhelmed by matter falling on it from a nearby companion star. The influx causes the white dwarf to explode. Type II supernovas are attributed to the death throes of much larger stars (more than eight times the sun's mass). In old age such a star transmutes lighter elements to heavier ones. It develops a core of iron surrounded by layers of elements lighter than iron ranging to the lightest, helium and hydrogen, at the outermost. As the star makes more and more heavy elements, the core gets overburdened and implodes, sending out a shock wave that explodes the rest of the star.

Type I and type II supernovas are distinguished from each other by differences in maximum brightness, by their light curves (the way their brightness varies over time) and by their spectra. Type III -- if SN1985f may be called that -- differs from both in all these characteristics, prompting Filippenko and Sargent to call it "peculiar" in the title of their paper on SN1985f, published in the Aug. 1 NATURE.

Because it was a spectroscope that first found SN1985f, the object's spectrum was the first peculiar characteristic to arrest astronomers' attention. The spectrum is dominated by emission lines, bright resonant emissions at wavelengths characteristic of oxygen, sodium and magnesium. The spectra of type II supernovas also show bright emission lines, but the dominant ones are those of hydrogen and helium, the outermost layers of the exploding star. Oxygen, sodium and magnesium should lie deep within the star. SN1985f is also much dimmer than the other two types of supernova. There is evidence that it never achieved the maximum brightness of either type.

Filippenko and Sargent point out that these characteristics seem to match those of a kind of "peeled" supernova posited some time ago by Roger A. Chevalier of the University of Virginia in Charlottesville. He suggested that a star that had somehow lost its outer layers would explode into a supernova much fainter than a usual type II. But Chevalier did not say -- nor a ppararently has anyone else -- how such a star might lose its outer layers.

Other astronomers have been rushing to observe SN1985f. One of them, Michael De Robertis of the Lick Observatory, at the University of California at Santa Cruz, points out another peculiarity: In spectra he has taken between March and July, the intensity of the emission lines does not change. Those of known types of supernova do change over such a 90-day period. This "calls into question" the designation of the object as supernova, he says. But if it is not a supernova, neither De Robertis nor apparently any other astronomer has suggested what else it might be.

If SN1985f is the representative of a new class of supernova, Filippenko and Sargent point out, it can explain certain strange objects that astronomers have classed as supernova remnants: Cassiopeia A, a very famous radio source; the object N132D in the Large Magellanic Cloud; and "the extraordinary object in NGC4449." Supernova remnants are the clouds of ejecta thrown out by the explosion, which continue to emit light and/or radio for centuries and millennia. From the present condition of Cas A, astronomers can calculate that its explosion should have happened in 1665, but there are no contemporary records of a sighting. It it was type III, it may have been too dim for the naked eye.