When Vincent Reddish was appointed by the United Kingdom Science Research Council (SRC) to plan and build the 48-inch Schmidt telescope in Australia, it was clear that a major opportunity would be created to discover new astronomical objects and phenomena in the relatively little-explored southern hemisphere. One of the first programmes to be undertaken on the telescope would be a systematic survey of the whole southern sky. This project was part of the ESO/SRC Southern Sky Survey and was one of the responsibilities of the United Kingdom Schmidt Telescope Unit. The prototype telescope for the UK Schmidt was the 48-inch Schmidt on Palomar Mountain which, in the 1950's, had produced the famous and widely used sky survey of the northern hemisphere. The Schmidt in Australia was to utilize the new Eastman Kodak IIIa-J emulsion, which, together with the fast (f/2.5) optics and the dark skies at Siding Spring, would penetrate to fainter magnitudes than had ever before been systematically observed in either hemisphere.

Early in 1973 Reddish decided that the many new and interesting objects in the southern sky survey should be systematically recorded and published. To that end he invited Arp to Edinburgh to consider the production of a catalogue of the objects then appearing on the first of these survey plates. Arp (1966) had previously published the Atlas of Peculiar Galaxies: 333 unusual objects in the northern hemisphere discovered on the Palomar Sky Survey by astronomers such as himself, Vorontsov-Velyaminov, A. Wilson, F. Zwicky and others. Arp realized that the deep U.K. Schmidt Survey would detect low-surface-brightness features better than any previous survey and would therefore register much larger extensions around the outer edges of galaxies. He felt that these faint outer regions would reveal pecularities of a kind not ordinarily visible and demonstrate associations between objects not normally suspected of being associated. Therefore he enthusiastically began the project on his own in 1973.

After the first session of examining plates it became clear that thoroughly searching all the plates in the southern hemisphere survey was a formidable task for one person. In the fall of 1974, to alleviate the situation Reddish sought participation from some of the other astronomy centres in Britain. As a result of this invitation Donald Lynden-Bell proposed to Barry Madore, then in Cambridge, that he join the project. Madore accepted, and in November 1974 Arp and Madore searched the first set of plates together in Edinburgh. This collaboration was to last more than ten years.

During the first seven years Arp and Madore made visits about once a year to Edinburgh to search each batch of new plates which had been obtained in Australia and shipped to Scotland. At first, each plate took a considerable time to scan; inspecting every galaxy down to a size limit of about 0.2 millimeter (=10 arcsec) is by no means a trivial task. But gradually, as facility was developed, the plate-scanning speed rose from an average of two per day to about six. Still, even with this improvement the decision was made to limit the Catalogue to zones only as far north as a declination centre of -25 degrees. Therefore the Catalogue is a complete listing of peculiar objects from the south celestial pole to a declination of -22 degrees. There are 537 plates, nominally 6.5 degrees by 6.5 degrees square, comprising the SRC Sky Survey coverage of this area in the southern sky.

During all these years the whole staff of the U.K. Schmidt Telescope Unit were enormously helpful. Russell Cannon and Tim Hawarden took the responsibility of setting up plate-scanning and measuring facilities. Russell Eberst and Elizabeth Sim helped to locate the best available plates on which to carry out the survey. When the last plate was scanned in the fall of 1980 the work of collating the tabular descriptions and positions of some seven thousand objects, and producing the photographs of about a sixth of these, was started. It was decided to produce two volumes. The first, containing positions and descriptions for all peculiar objects, would be the quantitative basis for future studies. The second, containing selected photographs, would illustrate the various categories of peculiarity and bring attention to the outstanding members of each category.

Because some of the survey plate centres did not coincide exactly with the nominal centres, some of the early positions of particular objects were inaccurate. It was therefore decided that the positions of all objects should be checked by overlays. In 1980 Petrusia Bojetchko and Ed Anderson spent part of a summer in Toronto correcting positions; and in 1981 Matthew Bates extended the work in a one-month stay at Edinburgh. It later became clear, however, that some of the simple procedures adopted for correcting the co-ordinates were inadequate and so during 1982/83 Madore and Robert Freedman repeated the overlays and off-sets for the entire Catalogue.

This re-analysis was almost as time consuming as the original search and delayed the publication considerably. But it was essential. The pointing accuracy of most large optical telescopes is now better than a few tenths of an arcmin, and accordingly the TV acquisition finder fields are usually only a minute or two in diameter. If the Catalogue was to serve the practical needs of optical astronomers, the co-ordinates had to be determined to better than the one arc minute accuracy obtained for the first rough reductions. Other catalogues were to be cross-referenced with our Catalogue, thus good positions were essential in eliminating confusion at a later date. Finally, since the names of the objects were to consist of the truncated co-ordinates (the so-called Parkes nomenclature) the derived co-ordinates had to be correct to the limits of the truncation.

The co-ordinate revision also provided unexpected benefits. Because the off-sets were generally made using copies of plates that were not those originally surveyed, ambiguities caused by plate defects were resolved (for instance, ``dwarf/defect?'' classifications were usually decided by the second plate). Every classification was in this way verified for at least a third time after a considerable period. Very occasionally, a new object was included in the sample if it were of sufficient interest. Although this type of inclusion was not to be desired for reasons of uniformity, it was usually only made for dwarf and compact galaxies which have an intrinsically uncertain discovery rate.

It was decided that in the case of well-separated associations of objects, every effort would be made to assign individual names and positions to the prominent members, while preserving the nature of the association in the description and code. Often it is by no means clear what single co-ordinate should apply to groupings: frequently no single member is dominant, and in the case of pairs, for instance, the midpoint contains neither member. This latter situation was seen as potentially leading to problems in TV acquisition and cross-referencing with other catalogues. Still, because of the numbers involved in such groupings, some practical limit to this process had to be imposed. Thus, for systems less than a few minutes in total extent, or for larger systems in which there is an obvious and dominant centre of symmetry, only one co-ordinate is listed. For larger and for more irregular systems with less than half a dozen dominant members, individual co-ordinates were obtained. We trust that the effort required in making these modifications and extensions will make the Catalogue both a practical and long-lived tool despite the resulting delays.

By the end of 1980, the volume of photographic reproductions was also ready to be prepared. The thousand photographs which were scheduled for publication represented a major undertaking for the ROE Photographic Laboratory. Careful consultation with the head of that laboratory, Brian Hadley, produced a plan for the production of standard sizes and formats for the photographs. Two test runs were made in order to set the optimal contrast and density for the prints. For the final production, Russell Eberst of the Schmidt Unit identified each object on the best available plate or film and marked the area to be photographed. The successful completion of this phase owes much to the good judgement in identification and framing which Eberst made on this long list of objects. Lastly, Bill Roberton of the ROE Photographic Laboratory produced all the final prints. The superb quality of the final prints is a tribute to Roberton's skill and diligence.

During this period, Malcolm Longair assumed the Directorship of ROE. In spite of his many new duties in the vigorous research activities at ROE he found time to see that the grant given by the SRC for the completion of the Catalogue was maintained. He consulted and advised in the publication of the Catalogue, and the whole of the Observatory staff helped with suggestions and criticism of the final edition.