The fact that a multicolour image can be produced by a hologram recorded with three suitably chosen wavelengths was first pointed out by Leith and Upatnieks [1964].

The resulting recording can be considered as made up of three incoherently superposed holograms. When it is illuminated once again with the three wavelengths used to make it, each of these wavelengths is diffracted by the hologram recorded with it to give a reconstructed image in the corresponding colour. The superposition of these three images yields a multicolour reconstruction.

However, while multicolour holography was demonstrated at quite an early stage, its further development was held up initially by several practical problems. These problems, as well as later advances that have made multicolour holography practical, are described in this chapter (see also the review by Hariharan [1983]).

Light sources for colour holography

The most commonly used lasers for colour holography are the He-Ne laser (λ = 633 nm) and the Ar+ laser, which has two strong output lines (λ, = 514 nm and 488 nm; see Table 5.1). The range of colours that can be reconstructed with these three wavelengths as primaries can be determined by means of the C.I.E. chromaticity diagram [Optical Society of America, 1953]. In this diagram, as shown in fig. 9.1, points representing monochromatic light of different wavelengths constitute the horseshoe-shaped curve known as the spectrum locus; all other colours lie within this boundary.