In order to take advantage of the good pointing of the WHT to speed up
acquisition, it is essential that object positions be measured as
accurately as possible, preferably to better than 1 arcsec for slit
spectroscopy. In such cases, acquisition to within 3 arcsec is usually
possible and even objects close to the limit of the integrating TV can be
identified unambiguously. Obviously, less accurate positions will suffice
for imaging observations and for bright objects which are immediately
recognisable, but errors comparable with the size of the TV field (1.6 
1.1 arcmin) will inevitably cause delays (recall that there is no finder
telescope).
Given accurate positions, it is often possible to do without a finding
chart, even for slit spectroscopy, but it is vital to ensure that either the
target object or a suitable offset star will be visible on the integrating
TV system. The approximate limit for a stellar object, under perfect
conditions, at new moon and with maximum integration is roughly m
= 21.5,
but under many circumstances, this is a gross overestimate. If
the object itself is fainter than m
= 17 -- 18 (and especially if it is
diffuse), then a suitable offset star brighter than this limit should be
found. The recommended procedure for blind offsetting is to
slew to the reference star,
centre it on a defined position using the handset and then to use the BLIND
command to move to the target object. This requires the RA and Dec
of both objects, rather than one position and an offset.
A finding chart is obviously essential for very crowded fields or if positions are poorly known. The chart should cover somewhat more than the TV field (say 3 -- 5 arcmin square) and should reach to the limit of the Palomar or Southern IIIaJ survey. Directions and scale should be clearly marked.
The telescope computer accepts coordinates in three systems: mean pre-IAU76 (FK4; default B1950), mean post-IAU76 (FK5; default J2000) and geocentric apparent of the date and time of observation. Neglect of the distinction between the two mean coordinate systems can lead to errors of up to 1 arcsec, comparable with the absolute pointing error of the telescope. For most purposes, it is enough to know that published coordinates with equinoxes earlier than 1975 are usually in the pre-IAU76 system (most are B1950) and that modern astrometric measurements use the post-IAU76 system (J2000). Optional parameters are proper motions (any epoch), parallax and radial velocity. For solar-system objects, differential tracking rates in RA and Dec may be specified. A list of conventions and units is given in the section on catalogues (see below). Very few observers will need to bother with parallax or radial velocity, and most will also be able to ignore proper motions. There are cases, however, where neglect of the proper motions will lead to severe difficulties, the most common trap being set by white dwarfs, which are often used as spectrophotometric standards, and which have large proper motions.