We are given a two-dimensional array of boxes - each box defined by a single stellar spectrum, or spectral scan.
The classification of an unknown stellar spectrum makes use of all features (lines, bands, blends, patterns) within the specified wavelength interval. The classification act itself consists of comparisons with the series of standard spectra that define the boxes, with the question: ``Is the unknown spectrum (x) `like' or `not like' this particular standard spectrum?''
A ``like'' classification does not imply identity in the appearance of x and the comparison spectrum. On the other hand, the resemblance is close; after a certain amount of experience, the ``like'' judgment becomes stabilized.
If only ``not like'' judgments are obtained for all boxes in a system, it is necessary to go on to comparison with the boxes in other systems.
The procedure outlined above implies the availability of a number of classification arrays (compatible with the MK Process) that include all types of stellar spectra frequently encountered in our Galaxy and in other nearby galaxies. The preparation of the arrays is a pressing problem for the immediate future.