It is generally recognized that stars of multiplicity higher than 2 are promising objects of study for several reasons. First, they often provide comprehensive data on stellar masses and luminosities that can be confronted with stellar evolution predictions. Secondly, they seem to offer an insight into the formation mechanisms of binary systems as first noted by Batten (1973). Indeed, the number of parameters that characterize a multiple system is greater than that of a simple binary, and in this multi-parameter space the information on the formation history may be obtained from period ratios, mass ratios, relative orientation of orbital planes, etc.
Surprisingly, no general catalogue of multiple systems has been published until now, although there are several "partial'' catalogues with various limitations. Fekel (1981) published a list of triple systems with periods less then 100 years. On the other hand, a catalogue of long-period visual triple stars where all 3 components are resolved was provided by Anosova (1988). It was later extended by Popovic (1991) to include some 700 visual triple stars within 200 pc from the Sun. Only a fraction of them can be considered as genuinely triple, i.e. with proved physical relation between all 3 components. The catalogue of nearby wide binary and multiple systems (Poveda et al. 1994) includes triple and quadruple systems which are physical, but, like the two previously cited catalogues, ignores all short-period sub-systems. Chambliss (1992) published a list of eclipsing binaries with visual companions.
The fragmentary nature of the existing information on
multiple stars is due not only to the personal preferences of
the authors, but also to the intrinsic difficulty of discovering
multiple stars. The effects of observational selection are
already very severe in case of binary stars and are still more
restrictive for the systems of higher multiplicity. Often
the discovery of such systems has resulted from
accidental combination of different observing techniques. The
distribution of orbital periods of binary stars is shown to be
continuous in the range from 1 to 
days (Duquennoy & Mayor
1991). If we are to understand the statistics and origin of
multiple stars, it is evident that the information throughout the full
period range is needed. Hence, there is a need of the general
catalogue of multiple stars that would combine the results
obtained by all observational techniques, without additional
restrictions other than those imposed by the techniques
themselves. The compilation of such multiple star
catalogue (hereafter MSC) is the aim of
this study.
In Sect. 2 the input sources of MSC are reviewed and general remarks on the definition and description of multiple stellar systems are given. The procedures used to estimate and code the parameters of sub-systems are given in Sect. 3, while the catalogue itself is described in Sect. 4. Such properties of the MSC as completeness, period and mass ratio distribution, orbit coplanarity are studied in Sect. 5. The conclusions and future research programs related to MSC are given in Sect. 6.