Epilogue

The classification of galaxies that we have used so far, in terms of spirals and ellipticals, is based largely on how these galaxies appear in pictures. With several $10^5$ galaxies in modern surveys, it is not really possible to eye-ball all of them in order to classify them. In addition, it is not even necessarily the best way, and certainly not the only way, of classifying galaxies and identifying different types. With so many spectra and images available, new statistical techniques are presently being developed, where computer programs which do not suffer from human prejudices, such as principle component analysis and neural networks, decide how galaxies should be grouped and classified.