Characteristic timescales for galaxy-galaxy interactions are much shorter than a Hubble time and therefore interactions between galaxies are of vital importance for their evolution. We are fortunate in having a pair of Magellanic galaxies so close to a large spiral like our Galaxy, since the LMC-SMC-Galaxy system provides us with an ideal chance to make a detailed study of galaxy interactions and system evolution. As the Magellanic Clouds are also the only galaxies near our own with significant amounts of gas, there will also be interactions with any outer halo gaseous component of our Galaxy (cf. possible origins and development of the Magellanic Stream, Murai & Fujimoto 1980; Wayte 1991). It is also quite feasible that the distribution and the origin of at least some of the Galactic dwarf spheroidal satellites and outer halo globular clusters are intimately linked with the development of the Magellanic System (e.g. Lynden-Bell 1976; Kunkel & Demers 1976)
However, before we can fully understand these outer halo stellar systems or even the star formation, chemical evolution and kinematics internal to the LMC and SMC, we need to have a thorough understanding of the global dynamics of the Magellanic System. An essential step towards achieving this is to investigate the kinematics of a representative sample of LMC-SMC members. Throughout the last twenty years, our two nearest neighbours have been surveyed to identify numerous types of stars, of which cool AGB carbon stars are just one example. These intermediate age, extremely red giants, are a few to several Gyr old and represent the age of the majority of the stellar population of the Magellanic Clouds, making them excellent probes of the dynamics of the bulk motion of the Cloud stellar populations. Furthermore, AGB carbon stars are found not only in the inner regions of the Magellanic Clouds, but also in their "disks'' and outer halos. Demers et al. (1993 - hereafter DIK) demonstrated that intermediate age AGB carbon stars can be found out to an angular distance of 10 degrees from the centre of the LMC and that they can be readily identified from sky survey photographic plates.
Previous large area surveys for Cloud carbon stars (e.g. Westerlund et al.
1978; Blanco et al. 1980; Blanco & McCarthy
1990; Rebeirot et al. 1993) have, with the exception of
the SMC survey of Morgan
& Hatzidimitriou (1995), searched areas close to the centre of the
these galaxies but have neglected the low density periphery of the Clouds.
These outer regions are vital to the study of the interaction dynamics of
the LMC-SMC-Galaxy system. The 
100 square degree survey by DIK in the
inter-Cloud bridge region was our first attempt to remedy this situation.
Feast & Whitelock (1994) obtained JHK infrared photometry of
the stars listed by DIK and concluded that they were similar in age (1 to a
few Gyr), colour and magnitude to stars in intermediate age clusters in the
Clouds, confirming that these field carbon stars do indeed represent the
intermediate age bulk population of the Clouds.
Our current survey is essentially an extension of the DIK work making full
use of available UK Schmidt Telescope (UKST) sky survey plates. An example
of a typical colour-magnitude diagram (CMD), that can be constructed from
UKST 
and R survey quality material, is shown in Fig. 4 of
Irwin et al. (1990). These passbands are particularly good
for an AGB carbon survey because the colour at the tip of the giant branch
, equivalent to 
, lies redward of
Galactic foreground contamination or any other possible population. Apart
from photometric errors throwing a small number, 
, of Galactic
stars into the sample, virtually all the selected candidates are either
late-M giants in the Clouds or cool AGB carbon stars. M-dwarfs in the solar
neighbourhood have colours barely bluer that the quoted limit above.
Dwarf stars made up about one third of the non-carbon stars found.
The carbon star candidates,
identified from their red colour, were then spectroscopically confirmed.
During the course of the survey nearly one thousand spectra were obtained.
They are of sufficiently good quality to derive accurate radial velocities
(
) for a large number (> 500) of carbon stars. Radial
velocities for these intermediate-age carbon stars (and for the Cloud late-type
M giants) define ideal kinematic test particles to investigate recent dynamical
interactions between the Galaxy-LMC-SMC system and in particular the origin
of the morphological disturbances seen in the SMC and parts of the LMC,
the origin of the Magellanic Stream and the total mass of the LMC. These
will be the subject of future papers; in the remainder of this paper we
briefly describe the survey rationale and present coordinates, finding charts
and radial velocities for the carbon stars observed.