The Compilation and Reliability of London Directories

P. J. Atkins, Durham University London Journal 14 (1989), 1, 35-46 http://dx.doi.org/10.1179/030580389793078488 In the view of Professor Harold Carter directories provide, given due care by scholars, a resource of inestimable value for the consideration of urban character .1 In London they represent an extraordinarily large and varied database, with an annually updated edition or sometimes several titles and editions each year since the mid eighteenth century. At the very least this presents the opportunity of cross-checking and complementing other archival and printed sources such as taxation records, poll books, rate books and the population census. A significant problem in the use of London directories has been a dearth of information about what titles and editions were published and where they are now available for consultation. In 1932 Charles Goss produced a bibliography of 285 items, but he chose the unfortunately early terminal date of 1855.2 The present author has produced a revised and extended union list of 5,800 editions for the longer period of 1677-1977.3 Another problem has been that directories are generally an unknown quantity in terms of their compilation and reliability. Historians and geographers have very properly expressed doubts about using them in research where a representative sample of the population is required. This article seeks to clarify these issues by discussing the evidence we have of how the name lists in London directories were assembled. The true nature of the source s constraints may then begin to emerge. Compilation It is exceptionally difficult to know how the majority of London directories were compiled. We have a few fugitive scraps of information, hints rather than hard facts, which help us to piece together an impression of the process of collecting and collating data, but our knowledge of the motivations and editorial organisation of the compilers is frankly little more than guesswork. There appear to have been several approaches adopted to gathering lists of names and addresses in the eighteenth and nineteenth centuries. A popular method, especially in suburban directories covering several separate settlements, was to glean information from people with local knowledge, such as clergymen, clerks of the peace, registrars, postmasters, boards of guardians, and other gentlemen.4 This saved time in checking every house, but the quality of data must have varied with the conscientiousness of the informant. Such a method would have worked best in villages, but the sheer scale of population in London itself would have defeated even the most assiduous and well-informed local official. An alternative but much more expensive option was the employment of agents. William Bailey may have claimed to have visited every town and personally waited on every house in the kingdom to acquire a proper information, 5 but most editors were less extravagant in their boasts. The staff of Paynes London Directory in 1768 comprised the editor and one person who enquired at every reputable house in London to check details of removals.6 Paynes employee had only about 5,000 houses to visit. The more comprehensive directories

of the nineteenth century were major logistical undertakings by comparison: in 1840, for instance, about 250 letter carriers were used by Frederic Kelly of the Post Office London Directory to revise copy. An inherent drawback of this system was that the editor had to take on trust the information supplied by unsupervised collectors, and we have evidence that this trust was often abused. Unscrupulous agents were occasionally responsible for the incomplete or fraudulent revision of addresses,7 and it was apparently common for confidence tricksters, posing as directory staff, to demand money for the inclusion of an address in the next edition of a work. Most directory publishers made no such charge for ordinary entries: Eliza Boyle respectfully cautions the nobility and gentry against a man, whom for great and repeated misconduct she has discharged from her employ. He is about 5 feet, 3 or 4 inches high, has a bald head, thick lips, and talks very fast. Having heard that this man has in various places, made a demand of money for insertions in the court guide, E.B. begs to state that the persons employed by her have positive orders, under pain of dismissal, to make no charge for any insertion or alteration they may receive .8 Other methods of directory compilation also had their problems. Some editors hoped that the public would supply information by post or by personal visits to the offices of the publishers. Advertisements along these lines are frequently to be seen in newspapers and directories, but the results were disappointing. Mr Mortimer, for instance, in 1763 found it impossible to procure the names and directions of every individual, mechanic, artist, and manufacturer, notwithstanding my repeated advertisements for that purpose. 9 No more satisfactory was the distribution of blank forms, which John Gray naifly expected people voluntarily to complete and return to him.10 Even honest and efficient directory publishers and agents were faced with considerable difficulties in collecting information. The extraordinary scale of London must have made directory compilation a daunting task. Yet, as Penelope Corfield has pointed out, these very problems of complexity also stimulated demand from the confused town dweller for a simple guide to their city. 11 One might even argue that directories were an essential prerequisite for making urban life bearable, at least for that small section of the population which needed to communicate commercially or socially with a wide circle of contacts. A newcomer to the directory trade was at a considerable disadvantage in having to start from scratch. All compilers were inconvenienced by the rapid turnover of population in some parts of London12 and by the lack of, or inconsistent, numbering of houses. According to Ambrose Heal the first numbered buildings were in Prescot Street, Goodman s Fields, and also in the Inns of Court and Chancery in about 1708;13 but this innovation took 60 years to diffuse throughout the city,14 and probably 150 years to reach the less accessible suburbs. In 1851 R. H. Mason remarked in his Greenwich and Blackheath Shilling Directory: that in many of the streets I have been unable to give numbers to the houses - for the simple reason that none exist: where they are occasionally to be found, the affixing a number to a particular house seems to be result of some mere whim of the present or former occupiers. I could point to certain streets - where it appears there is a perfect mania for a particular number; thus in George Street, there are at least half-a-dozen no. 4s.15 The difficulties inherent in their trade made directory publishers sensitive to any perceived criticism. R. H. Mason again, reflecting on the compilation of his Greenwich directory, predicted that some self-constituted critic will make use of his Argus-eyes to detect its errors, I think highly probable, but he will not disturb my repose by any ill-natured comment. If he points out an error, I will note it down for correction next year, and give him my thanks.16

There was also a keen sense of competition among directory publishers and often a sense of grievance. This was R. H. Masons theme in his Court Guide and General Directory for Brentford in 1853: I have had the misfortune to have been preceded in the district by a class of directory publishers, whose characters I will not venture to describe, and the consequence has been that a large number of tradesmen whom they had victimized have reluctantly been induced to place confidence in the genuineness of the present publication.17 Jesse Ward postulated a Darwinian explanation of the failure of his predecessors in Croydon: There was a time when Croydon was periodically invaded by alien directory makers, whose greatest care was bestowed upon their advertisement pages, and who succeeded only in earning the hearty disgust of the public for these compilations. Second editions were never contemplated, nor indeed possible; and the inexorable law of the survival of the fittest has been justified in this as in other cases.18 Many a directory publisher must have wished for, and perhaps even worked for the failure of his rivals, but few were as explicit about their feelings as Mr Mortimer: It is ... hoped, as a recompense to the compiler for his care and attention in establishing this useful work, that the artists ... manufacturers, and all other persons ... will not give any account of their situation, & c. to any other person applying for the same.19 Occasionally a skirmish between rivals reached the courts. In 1862, for instance, J. Giddings threatened a competitor, Tomkies and Son, the publishers of Simpsons Street Directory of St Pancras, with legal proceedings because he had already entered his directory at Stationers Hall, thereby establishing his copyright in advance. 20 Piracy was also the subject of litigation. In short, some directory editors considered it expedient to copy the work of others and, after making a few modifications, present it to the public as if it were their own. The Kelly family were particularly sensitive to this issue. The following is a summary of their main court cases of relevance to London, 1840-1900: 1840 F. Kelly v Hooper: sale of a one sheet almanac taken from the Post Office London Directory; 1859 F. Kelly v W. R. and R. OByrne: an injunction to prevent the sale of the Pocket Post Office Directory;21 1860 F. Kelly v T. Blower: order restraining Thomas Blower from printing and publishing Blowers Architects Surveyors, Engravers and Builders London and Provincial Directory for 1860, which allegedly contained Kellys copyright material; 1861 F. Kelly v J. T. Pickburn: a dispute over Pickburns The Watchmakers and Jewellers Directory and Clerkenwell News Almanack for Watch and Clockmakers, Jewellers, Goldsmiths, Silversmiths and their Various Branches; 1863 F. Kelly v Straker: concerning Strakers Annual Mercantile, Ship and Insurance Register; 1866 F. Kelly v J. S. C. Morris: after Morris had sold The Imperial Directory of London for 1866 at the same price as the Post Office London Directory, with a similar format, and with several known mistakes reproduced; 1868 E. R. Kelly v C. W. H. and E. F. Wyman: accusing the Wymans of breach of copyright in their Architects, Engineers and Building Trades Directory; 1900 Kellys Directories Ltd v National Trades Directory Co.: infringement of copyright in the National Trades Directory of Great Britain and Ireland.

The Kelly family were aggressive in their attitude to competition, fair or unfair. They took every opportunity to squeeze their adversaries out of the market and were successful in seeing off such formidable opposition as Kents Directory in 1828, Pigot in 1841, Robson in 1843, and Watkins in 1855. After 1855 they had a pseudo-monopoly of London directories and extended their power base into specialist directories and coverage of suburban areas. Eventually the phrase Kellys Directories came to enter the nations popular vocabulary as the firm broadened their list to include a wide range of directories of all types. Kellys Post Office London Directory is the best documented London directory as far as compilation is concerned. From a number of sources we know the procedure for the early 1840s.22 First of all each name and address was cut out of the previous year s commercial and court directories and attached to a query paper, of which there were 120,000 in 1846. These were sorted by street, and, after checking against the street directory for inconsistencies, they were handed to the 250 or so letter carriers on their 156 walks in London: When the lettercarriers have received the query papers, it is their duty to submit each to the person whose name it bears, to ask if it is correct for the next year s publication, and, in the event of any alteration being required, to request the party to make the same in a plain handwriting.23 Frederic Kelly was criticized publicly in parliament24 for using public servants in the compilation of what in practice was a private publication,25 but his defence was that the letter carriers were only employed on this work on their light days, Tuesdays or Wednesdays, once a year, usually at the end of June. He estimated that each postman spent about 18 hours in a year on directory work, all of which was outside normal working time, and for which the letter carriers were paid commission on sales of the finished book, and on the collection of money for advertisements and extra lines.26 The checked slips were returned to the office within a week or ten days, and the letter carriers were then issued with a supply of blank printed forms to be filled in by people who had moved to the district since the previous year. Where someone had moved out, the query paper was forwarded to their new address for checking. All this information was then collated, no mean feat, and re-sorted by streets. The street directory was always the first section of the new edition to be compiled: next came the court and commercial directories, and finally the classified trades. This mammoth task of collation was performed by the sub-sorters at the post office headquarters, assisted by the directory staff at Boswell Court. The compilation process was completed generally by September or October, and the proofs were checked on the streets by the letter carriers, a source of complaint amongst those who resented this unpaid task. Corrections were possible until the first or second week in October,27 although the deadline of the addenda section was not until the end of November. The directories were then printed and sewn28 in time for publication in December, ahead of the Christmas and New Year peak demand. A supplement was published in February of the following year and was included in the Spring edition of the directory. The letter carriers were engaged on a further six hours or so of additional work checking amendments for this. In 1847 Frederic Kelly was prevented from further use of the letter carriers in the gathering of information.29 This cannot have come as a surprise because already in the 1844 edition he notes in the preface that he has begun to employ selected full-time agents. Ironically Kelly subsequently claimed that the excellence of the directory was the result of the editing of the information rather than the method of collection, whereas in 1840 he had different views:

The information from which this book has been compiled has been, by the employment of more than 250 agents possessing local knowledge, collected during the month of September, and very many changes which occurred much later are introduced. Were there but few collectors employed, so that division of the vast metropolis must be taken in rotation, it would be unavoidable in the majority of districts to apply much earlier. It is therefore clear that the system under which the Post Office Directory is compiled furnishes largely information which could not otherwise appear until the next annual edition of the work.30 This detailed account of the Post Office London Directory in the 1840s should not be taken as representative of other directory series, nor even of Kellys directories of London in the later period. It is interesting to note, however, that some suburban directories were compiled over a three-month period, a similar time-scale to that of the Post Office London Directory. One example, for instance, was Gray and Warrens 1861-2 edition of their Croydon directory, which had changed two-thirds of its detail since the previous edition.31 Grays 1867 directory of Wandsworth and district, on the other hand, was the result of fifteen months assiduous labour.32 One of the major issues of compilation was the editorial policy concerning those names and addresses to be included or excluded from the directory. We have no direct evidence of the criteria used, and in the case of some volumes one suspects that the decision was an ad hoc one made by the collector of information rather than the editor. We do, however, have a couple of quotations which appear to confirm the suspicion that the lists of names and addresses in directories were a biased sample, under-representing the disadvantaged sections of society. Thus Thomas Mitchell, in evidence before a committee of enquiry in 1846 reminisced that: When I first came into the office, I was told by the letter-carrier on the City Road Walk, that I need not be particular about putting in those petty little shops, and so forth, but to put respectable characters in, who were necessary to refer to, in the directory.33 A similar selective approach was adopted later by W. Tomkies for his directory of Croydon: The proprietor has been at great pains to weed the districts, so that the names of only those who are likely to be of interest to the tradesmen and gentlemen of the locality are inserted. Croydon, like other large and increasing towns, is inhabited by a mixed population; but we have been at great pains to classify them. The houses inhabited principally by the labouring classes we have almost entirely omitted, particularly where they are weekly tenants, and where more than one family resides, it being a known fact that this class of people are so frequently changing their residences that, to insert them, would only be to mislead.34 In south London Effingham Wilson used the selective criterion of a house s rateable value. No household in a property of less than 30 per annum value was included. 35 Nowadays Information Services Ltd retain a staff of between twenty and thirty as checkers in the London Postal region for Kellys Post Office London Business Directory.36 Until the mid 1970seach of these agents had their own area to cover and proceeded literally by knocking on doors. Their season was April to August, and some were part-time staff such as retired persons and students. In addition to checking the accuracy of address information, they were also encouraged, on a commission basis, to sell copies of Kellys directories.

One might imagine that the availability of rate books, electoral registers and other sources of listed address data would reduce the effort of compiling a directory. This would be an incorrect assumption, however, because electoral registers do not give telephone numbers, and there is no substitute for personal contact in the quest for sales and advertising revenue. 37 A recent trend that has affected directory compilation, especially since the Second World War, is the increased employment of women in the nation s workforce. This had made the canvassers task more difficult because houses are often empty when they call during the daytime, and a fully comprehensive check is therefore more time consuming and more expensive. Partly for this reason, the residential or court section of the Post Office London Directory was dropped in 1971. Representatives now call only on businesses38 in London, but even this operation has been curtailed. Where the whole of London used to be checked every year, now only the centre is subjected to an annual scrutiny, and the rest of London every other year. Until 1978 each entry for the Post Office London Directory was recorded on a slip of paper which was given to the agent for checking, much as Frederic Kelly had organized the compilation of his directory in the early 1840s. Now records are in machine-readable form in a data base system,39 with computer printed slips for checking in the field and visual display units for entering any alterations which may be notified. Specially written software has been designed to convert the raw records into the final format for publication. A bromide of each page can be made without the need for galley or page proofs. The blank sections deliberately left for advertising blocks can then be overprinted, and from 1985 the opportunity has been taken to use spot colour, creating a very effective visual presentation. This improved appearance has been assisted by the related development of computer typesetting, which has meant the abandonment of letterpress. Previously each name and address would be individually set so that, jf the address remained the same for a long period, there was a chance that the type might wear out, losing serifs and giving the typeface a rather rough look. Problems of Interpretation and Reliability The commercial, professional and classified directories of London facilitate a reconstruction of the evolving structure of the many trades which made up the London economy, but there are several problems which must first be faced by the researcher.40 Firstly, the description given of each individual s trade or profession is usually very brief. It is impossible to say in most cases whether this description is accurate or complete. Difficulties arise where a person may have engaged in more than one economic activity, but was recorded under only one heading. Studies over long periods of time are also faced with the changing nature of some trades. The apparently straightforward label of dairyman, for instance, hides a change from the cowkeeper of the eighteenth century to the fixed-shop urban retailer or deliveryman of the late nineteenth and twentieth centuries, who was divorced from milk production.41 The label of merchant is another problematical category, especially in the eighteenth century when it was used as a catch-all description, much to the distaste of a Mr Mortimer: After a strict and impartial examination of all the printed lists of merchants and traders of the City of London, that have been hitherto published, I find myself obliged to declare, that they are extremely imperfect; several persons being inserted as

merchants, who have not the least pretence to that very important and honourable title, in any shape whatever: others again are denominated merchants, who keep retailing millinery and haberdashery shops, merely because they have, perhaps once or twice, imported some trifling commodity, and made a few entries at the customhouse.42 Some directory editors reduced their number of trade headings to a minimum in order to save work in classifying their entries. This may create difficulties for the student of economic history because such headings will inevitably be very broad and subjective in their definition. Comparisons between directories with different systems of classification may be impossible. Secondly, one would like to know how complete a record classified directories were of the trades they listed. This is very difficult to establish, but it seems likely that there was a hierarchy of comprehensiveness, according to the perceived importance of the trade or profession in question. Checking is possible where a trader was licensed by the local authority for the purposes of taxation or regulation. Thus in 1880, for instance, we know that only 56% of the licensed milk producers in London were entered in the Post Office London Directory.43 This was presumably a direct result of the low esteem in which so-called noxious trades such as cowkeeping were held by contemporaries. On the other hand, a comparison of the law directory in the Post Office London Directory at the same date with The Law List44 shows that 94% of London solicitors from the official list were also to be found in the directory.45 Another consideration was location: small firms in poor districts and enterprises newly established at the suburban frontier both seem to be under-represented. Directories have also been used by historical geographers in work on urban residential patterns and processes. The turnover of householders in a neighbourhood and, by implication, the consequent residential mobility, have, for instance, been foci of interest. Technical problems have arisen, however, which throw doubt upon the reliability of directories as a source for such studies. A person whose name appears at an address in one directory may by the next edition no longer be recorded there. She or he may simply have moved to a new house or flat but there are other alternative possibilities: 1. They may have died, in which case cross-checking with the parish burial registers should be fruitful.46 2. People who cease to be household heads disappear from directory listings. A newly married woman, for instance, will have assumed a new surname. Migration was very common for brides and evidence from the marriage registers may therefore be difficult to acquire. 3. The individual may have moved to a street or court not covered by the directory. 4. Generally speaking directories record only one name per property, and people living in dwellings of multiple occupation will therefore be under-represented.47 5. Movers often acquired a new initial or a slight spelling variant of their surname. This problem of knowing whether one has precisely the same person in the different editions of a directory is compounded in London by the large numbers of people bearing common surnames. It is perhaps worth noting that these problems of identification are more often acute in provincial cities, many of which had new directory editions only once every few years, whereas in London there was a new edition or even two editions of most titles produced annually.

Although the boast of completeness was used frequently by editors hoping to encourage sales, the reality was very different. The name and address lists expanded substantially during the eighteenth and nineteenth centuries (Table 1) but they still struggled to keep pace with the phenomenally rapid growth of Londons population. Table 2 is an attempt to estimate the population in the catchment of the Post Office London Directory.48 It shows that 1891 was the peak year of coverage, although even then the directory listed only 16.1 per cent of the households which might have been recorded. This is a very small proportion by comparison with some provincial directories, which managed figures as high as 70 per cent.49 Several geographical problems are worthy of note. Firstly there is a difficulty in knowing the area covered. Even the Post Office London Directory, which came to have a broad zone of coverage (Figure 1) by the end of the nineteenth century, had started with just central London. There was also an array of suburban directories which were by definition restricted in their extent, but fortunately establishing their boundaries is relatively straightforward. Secondly, directory compilers were naturally most concerned to record the names and addresses of their clients, the people who bought directories. They were neglectful of people in the poorer areas of the city, and one therefore has an inbuilt areal as well as a social bias. This is demonstrated in Table 3, where samples of residents in Bethnal Green and Belgravia, drawn from the manuscript census enumerators Books, were checked against the street listings in the Post Office London Directory at two dates, 1851 and 1881.50 Clearly, the welloff area received more assiduous attention, and within that district Belgrave Square and Wilton Crescent were more fully covered than Belgrave Mews and Wilton Crescent Mews. This was because upstairs-downstairs corresponded to front street-back street. Thirdly, even where the area covered is known, such as Londons West End, there was sometimes an extraordinary variability in the overlap of names and addresses between different titles published in the same year. Table 4 shows that for the sample year 1860 there is an alarmingly low correspondence between, for instance, Websters Royal Red Book and Boyles Court Guide, both of which set themselves the task of listing the capital s elite. Summary It may be apposite at this juncture to reiterate some of the pitfalls of directory use: 1. Class bias, which leads to the under-recording of working people. 2. Areal bias, whereby poor neighbourhoods and some suburban districts are underrepresented. 3. Main road bias. Here residents, whatever their social standing or trade, were more likely to be listed than those living in the back streets and courts. 4. The differential listing of occupations according to their perceived importance. 5. The problem of multiple occupancy. Only one name per property was allowed in most directories. 6. Double-counting, where an individuals name may appear more than once. This has been exaggerated as a problem by some writers.51 7. Inconsistencies from year to year in details recorded. 8. The renaming and renumbering of streets.52

9. The constraints of producing a book which was easy to handle, and which stayed within the pocket of its readers, meant that no directory of London was ever able to record the names and addresses of more than a small fraction of its inhabitants. All of this sounds very negative, but these problems are listed merely to alert users of London directories. With a suitable degree of caution this is a source with rich potential for work in economic and business history, urban history and local topography, the spatial analysis of socio-economic structures, and of course genealogy. Directories have the advantage that they are published, easy to use, and readily accessible in reference libraries. Ultimately their contribution to the reconstruction of past communities may be strongest in record linkage, the pooling of information from complementary manuscript and published sources. This database approach, relying on computer technology, is not new, but it does provide exciting possibilities for the manipulation of large data sets. London directories have been perceived by some workers in the past as an oversized source of low grade information. The computer is the ideal medium for concentrating their value and adding to it by the high speed analysis of linked data.

NOTES 1 H. Carter, An 1ntroduction to Urban Historical Geography (1983), 88-9. 2 C. W. F. Goss, The London Directories 1677-1855: a Bibliography with Notes on Their Origin and Development (1932). 3 P. J. Atkins, The Directories of London, 1677-1977 (forthcoming). 4 E.g. R. Simpson & Co.s Bow, Stratford and Mile End Road Directory (1866); Deacons Mid-Surrey Court Guide (1878). 5 J. E. Norton, Guide to the Provincial Directories of England and Wales, Excluding London, Published Before 1856 (1950), 16. 6 T. Lowndes, The London Directory for the Year 1778, 2. 7 Kellys actually re-issued the 6th edn. of their Essex directory because of extensive errors in the Upper Clapton section due to the dishonesty of the agent employed : Post Office Directory of Essex, Herts, Kent, Middlesex, Surrey, and Sussex (1867). 8 E. Boyle, Boyles Fashionable Court and Country Guide and Town Visiting Directory (Jan. 1813), [iii]. 9 Mr Mortimer, The Universal Director; or, the Nobleman and Gentleman s True Guide ... (1763), part II, vii. 10 J. Gray, A New Commercial and General Directory of the Town of Croydon (1851), iii. 11 P. J. Corfield, Giving Directions to the Town: the Early Town Directories, Urban History Yearbook (1984), 22-35. 12 Wards Commercial and General Croydon Directory ... 1876, preface: Compiling a directory is very much like taking a photographic group of moving objects and figures ... Gentlemen will keep movin. For a commentary on residential mobility in Victorian cities see R. Dennis, English Industrial Cities of the Nineteenth Century: a Social Geography (Cambridge, 1984), ch. 8. 13 A. Heal, The Numbering of Houses in London Streets, Notes and Queries, 183 (1942), 100-1. 14 Enforced by Acts of Parliament in 1762 and 1765, but there were still complaints about the lack of systematic numbering in central London 80 years later: The Post Office London Directory (1846), iii. 15 R. H. Mason, Masons Greenwich and Blackheath Shilling Directory and Handbook ... for 1852 (1852), [iii]. 16 Ibid., 11. 17 R. H. Mason, Masons Court Guide and General Directory for Brentford ... &c. 1853 (1853), iii. 18 Wards Commercial and General Croydon Directory (1887), xi. 19 Mr Mortimer, The Universal Director (1763), 171. 20 J. Giddings, The St Pancras Directory for 1862, iii. 21 The only surviving copy of this is held by Brixton Public Library. It was an exhibit in the court case and has ms. annotations pointing out specific copyright infringements by page reference to the plaintiffs book, i.e. the Post Office London Directory. 22 Principally from the ms. records held in the archives of Business Press International, Sutton. See also the Postmaster Generals Minutes in the Post Office Archives and the prefaces published in the Post Office London Directory itself. 23 Memorandum in explanation of the system pursued in the compilation of the [Post Office London] Directory admitted as evidence in Parliamentary Papers, 1847 (744), LXII, Papers relative to the Dismissal of Robert Grapes, late a General Post Letter Carrier, 84. A note in this paper suggests that the 2,160 folio pages of evidence taken in the case were deposited in the office of the Journal of the House of Commons. Unfortunately none of these documents

have survived in the archives of parliament. I am grateful to D. J. Johnson, Deputy Clerk of the Records, House of Lords Record Office, for this information. 24 Hansard (1845), LXXXI, cols. 1319-30; (1846), LXXXV, cols. 807-40; (1846), LXXXVI, cols. 615-16; (1846), LXXXVIII, cols. 957-62. His chief critic was Thomas Duncombe, the radical member for Finsbury: Anon. 1799-1899: Centenary of the Post Office London Directory (1899), 26. 25 His predecessor B. Critchett had not issued slips of paper for correction, but had held a series of blue books at headquarters for alteration as appropriate. 26 The average letter carrier made 4.50 to 5 p.a. in commission. 27 In later years Kelly & Co. seem to have relied increasingly upon information supplied to them by the public about necessary alterations. In 1891, for instance, they received 1,000 letters per day in October and November; but no changes were inserted until they had been checked, in case of hoaxes: Post Office London Directory (1891), vii. 28 In 1855 the printing and sewing of 7,000 volumes took only ten days: Notes and Queries, 1st series, XI (1855), 83. 29 Hansard (1847), XCI, cols. 265-9. It was reported that the Chancellor of the Exchequer had forbidden Kelly from using the letter carriers and that no compensation would be paid. 30 Post Office London Directory (1840), iv. 31 Gray & Warren, The Commercial and General Directory of the Town and Parish of Croydon (1861), v. 32 J. W. Gray, The Metallurgicon Local Directory ... (1867), 3. 33 Parliamentary Papers, 1846 (607), XLV, A copy of the memorials, reports and communications, ... passed between Messrs Playle and Kelly, the Postmaster General, and Thomas Mitchell ... , 45. 34 W. Tomkies, Wilkins Street Directory and Court Guide of Croydon ... 1872-73, preface. 35 E. Wilson, The South London Suburban Directory for 1872, preface. 36 My thanks are due to David Lammin and Dennis Springett of Information Services Ltd for their detailed description of the compilation of the Post Office London Business Directory in recent years. 37 The London Gazette cannot be used as a source for information about newly registered companies because accommodation addresses are often the only ones recorded. 38 The definition of a business used for the Post Office London Business Directory is an increasingly narrow one. From 1983 onwards retailers have been excluded. 39 This enabled a 50% reduction in headquarters clerical staff. 40 A substantial literature exists on trades directories: J. L. Oliver, Directories and Their Use in Geographical Inquiry, Geography, 49 (1964), 400-9; E. P. Duggan, Industrialization and the Development of Urban Business Communities: Research Problems, Sources and Techniques, Local Historian, 11 (1974),457-65; C. R. Lewis, Trade Directories - a Data Source in Urban Analysis, National Library of Wales Journal, 19 (1975), 181-93; P. Wilde, The Use of Business Directories in Comparing the Industrial Structure of Towns , Local Historian, 12 (1976), 152-6; G. Shaw, The Content and Reliability of Nineteenth-century Trade Directories, Local Historian, 13 (1979), 205-9; B. Timmins, Measuring Industrial Growth from Trade Directories, Local Historian, 13 (1979), 349-52; P. J. Corfield, Giving Directions to the Town: the Early Town Directories; G. Shaw, Directories as Sources in Urban History: a Review of British and Canadian Material, Urban History Yearbook (1984), 36-44; P. J. Corfield, Computerising Urban Occupations, in P. Denley & D. Hopkins, eds. History and Computing (Manchester, 1987),68-71. 41 P. J. Atkins, The Milk Trade of London, c. 1794-1914, Ph.D. thesis, University of Cambridge (1977). 42 Mr Mortimer, Universal Director (1763), part III, 3.

43 Atkins, thesis, 276. 44 Stevens and Sons, The Law List ... 1880. 45 6% of these had address information which was different. 46 R. Dennis, English Industrial Cities. 47 C. G. Pooley, Residential Mobility in the Victorian City Transcriptions of Institute of British Georgraphers, New series, 4 (1979), 258-77. 48 The editors of the Post Office London Directory did not always respect administrative or census boundaries in deciding upon their outer limits, and estimates of population are therefore very difficult. 49 50-60% of the householders in Leicester were covered by directories: see R. M. Pritchard, Housing and the Spatial Structure of the City (Cambridge, 1976). The populations of some other towns were under-recorded, but these tended to be the smaller settlements. Gareth Shaw has suggested that larger cities such as Liverpool and Manchester enjoyed a coverage as high as 70%: G. Shaw, British Directories as Sources in Historical Geography, Historical Geography Research Series, 8 (1982), 31-2. 50 Because of the 100-year rule of confidentiality, no later date could be used for comparisons: see R. Lawton, ed. The Census and Social Structure: an Interpretive Guide to Nineteenth Century Censuses for England and Wales (1978). For an interesting commentary on bias in modern north American directories see R. Harris and B. Moffat, How reliable is the modern city directory? Canadian Geographer, 30 (1986), 154-8. 51 Particularly by W. K. D. Davies, J. A. Giggs and D. T. Herbert, Directories, Rate Books and the Commercial Structure of Towns, Geography, 53 (1969), 41-54. A rare chance for checking the double counting problem is provided in T. E. Whibley, Whibleys Court Directory and London Fashionable Guide (1864). The alphabetical section of this directory acts as an index, showing the number of times that an individual is recorded in the street section. Approximately 3.3% of names were duplicated. 52 This affected about 5,000 streets and terraces in the County of London, about one quarter of the total, between 1889 and 1912. See London County Council, List of the Streets and Places Within the Administrative County of London (1893 edn.).