Generally accepted scheme distinguishes two main classes of supernovae (SNe): Ia resulting from the old stellar population (thermonuclear explosion of a white dwarf, or a merger of two white dwarfs in a close binary system), and SNe of type II and Ib/c whose ancestors are young massive stars (died in a core-collapse explosion). Historically, classification of SNe, according to their optical spectra, began by recognizing SNe I, with no hydrogen lines, and SNe II which do show hydrogen in their spectra. In addition, SNe II were shown to exhibit much wider photometric behavior than SNe I, which seemed to be a rather homogeneous class of objects. Nevertheless, it was shown later that there are actually two spectroscopically and photometrically distinct subclasses of SNe I: Ia located only in ellipticals, and Ib found in HII regions and spiral arms, which strongly suggested that their progenitors were massive young stars with their hydrogen envelopes stripped. The third subclass, SNe Ic, discovered later, show no helium lines either, and thus correspond to massive stars stripped of their H and He envelopes.

SN Ia are known in astronomy as the "standard candles". Better understanding of some type II events (e.g. SNe II-P) and stripped-envelope SNe (Ib/c) could lead to their potential use as distance indicators as well. The problem of extinction is the most important issue to be dealt with in the process of obtaining true SN luminosities (absolute magnitudes). The plane-parallel model which gives absorption dependent on galaxy inclination (A ~ sec i), widely used in the past, was shown not to describe extinction adequately. There is some hope in an alternative model which introduces radial dependence of extinction, but further investigation is necessary.

During this phase the shock wave decelerates according to the law v = dR/dt = 2R/(5t), i.e. R = 1.17 (E_o/ρ_o)^1/5t^2/5, which is the well known Sedov's solution for a point-like explosion in uniform medium with density ρ_o. When the temperature behind the shock drops the energy loses due to radiation become significant and a remnant enters the radiative phase of evolution. It is expected in this epoch that SNR undergos a phase of dense shell formation. The remnant continues to expand as pressure-driven, (R ~ t^2/7) or momentum-conserving snowplow (R ~ t^1/4). Finally, when the expansion velocity becomes comparable with the sound speed in the surroundings, v² ~ c_s² = γp_o/ρ_o, supernova remnant will merge with the interstellar medium and the evolution ends.

An interesting issue concerning SNRs is the question of their radio evolution or the so called Σ–D (surface brightness to diameter) relation which can be written in the form Σ = A D^-β. Since parameters of the explosion (E_o, M_o) and the expansion (ρ_o) may substantially differ from remnant to remnant, it has been generally accepted that no single Σ–D relation can be constructed for all SNRs. However, it still might be possible to construct the relations for some classes of SNRs. This would be of great practical importance for estimation of distances to SNRs when no other methods are available.

Although they are known as relatively strong synchrotron radio sources, SNRs can also be observed in optical. In optical search for supernova remnants we use the fact that the optical spectra of SNRs have elevated [S II] : Hα emission-line ratios, as compared to the spectra of normal HII regions. This emission ratio has proven to be an accurate means of differentiating between shock-heated SNRs (ratios > 0.4, but often considerably higher) and photoionized nebulae (0.4, but typically < 0.2). The physical basis for this is as follows: in typical H II regions, sulfur exists mainly in the form of S++, yielding low [S II] λλ 6717,6731 to Hα emission ratios. After the shock wave from an SN explosion has propagated through the surrounding medium and the material has cooled sufficiently, a variety of ionization states are present, including S+. This accounts for the increased [S II] : Hα observed in SNRs.