You are here:
"A unique and priceless record
About The ArchiveThe Frith archive is recognised as being the only photographic archive of British national importance still in private hands.
- 365,000 photographs, all catalogued by town;
- depicting 7,000 cities, towns and villages throughout Britain;
- contains many towns that have been photographed continuously in the period between 1860 and 1970, thus illustrating the physical and social changes that have taken place over that period
- over 120,000 available now as high resolution digital scans;
Catalogue systemFrom 1860 to 1940: All photographs were numbered in numeric sequence and the titles written in hand-written ledgers. Since these were generally entered in the ledger shortly after the photograph was taken the negative number also provides a dating guide as all photographs taken in each year are in one number sequence.
From 1945: When the business recommenced normal operations after World War II, it changed to a new numbering system based on an alpha code for each town, followed by a number e.g.. WKG23 = Wokingham, photo number 23. Unfortunately, this number provides no clue as to the date of the photograph, which was written on the photographic print - sometimes! To further complicate matters no central record was maintained of the titles for the 250,000 photographs taken in this period. Until a photograph is scanned and data entered, the only way to see what is available has been to manually search through the individual photographs.
Although not perfect, this level of cataloguing was unique in a collection of this size and is one of the main attractions of the Frith archive from a commercial and historical point of view. However, like all human endeavours it is subject to human failings!
Some of the issues affecting the cataloguing of subjects include:(i) when a photographer went to a town in 1890, 1910 or 1920 he could, and sometimes did, make mistakes in obtaining the name of the street he was photographing;
(ii) street names have and do change over a 110 year period;
(iii) the titles given by the photographer were recorded in the hand-written ledgers which over the years has been written by a number of people, and it is possible to misread words and allocate an incorrect title to one of our newly published views;
(iv) some views have no title in the ledger - just a town name. Where we cannot be certain we usually leave the title in this form but in some cases our staff will have added a street name based on their personal knowledge of a town or by comparison with other, named, views;
(v) in a small percentage of cases, the original company F. Frith & Co Ltd. reused some views and "moved" them from an earlier catalogue number to later number thus throwing the date coding out completely e.g.. a photograph taken in 1890 might be reallocated a 1923 number and we have no way of knowing - unless there is some obvious visual dating - or until a customer tells us that a building in the picture was not erected until 1920 etc. Even then, customers can also get it wrong!
(vi) the hand-written ledgers contain no reference to the County in which the town is located, however, with the aid of a gazetteer, and reference to the towns listed immediately before and after, it is usually possible to identify the County quite easily. Unfortunately, in some cases this process can lead to the incorrect conclusion. Many town names appear in several counties and it is possible that a few photographs are allocated to the wrong county - Newport is a good example, it appears in Devon; Dyfed; Essex; Gloucestershire; Gwent; Highlands; Humberside; Isle of Wight; Norfolk; Salop and Somerset; Just to add to the difficulty, County boundaries were extensively changed in 1974 and continue to change, almost constantly.
(vii) the post 1940 material is a lot harder to date because a high percentage of the photographs do not have any date on the print and we have to guess the date from the content. Where a specific date is available we use that, without a "c" for circa, but where no date is available we use either "c1955"; "c1960" or "c1955" to indicate that these may not be accurate. Where we subsequently obtain more dating information we correct these, where required;
(viii) another feature of the 1940 to 1970 material is that the staff of F. Frith & Co Ltd were not too careful (or too concerned in the days of manual processing) to ensure that the alpha codes they allocated to photographs were unique. We know that it is possible for two towns with similar names to have been allocated the same alpha codes e.g.. WKG for Wokingham and also for Woking. Until the last few years this was not too problematic because the number of views we had published from the Post 1940 section was minimal. However, with the introduction of digital technology it is possible to publish many more views but in the process of doing so it has become apparent that there are more subjects than we originally thought where this is a problem. The fact that F. Frith & Co Ltd did not keep a central register of these alpha codes has obviously led to the errors, and made it impossible for us to check until going to use an image. Unfortunately, the computer is even less able to detect errors of this kind than the human and will happily overwrite one view onto the same negative number if asked to do so;
(ix) The other difficulty that is not always appreciated by people living in a locality depicted by a photograph is that, even though they can identify that view 71623 is not Bradford Tyrell Street but Bradford Manningham Lane, we cannot! The Frith staff do not personally know every town and street in the country and neither would those who went before us. For this reason it is not always possible to do anything but accept many photographs and their titles at face value and assume they are correct. When we acquired the archive in 1977 we decided to utilise the original cataloguing systems - there being no real alternative. Between 1977 and 1992 we published approximately 11,000 subjects out of the 330,000 available. Clearly, for most of these we found and have used the correct titles and dates, however, from time to time mistakes have been discovered and corrected. As we have now published over 120,000 subjects using digital technology, it is inevitable that more will come to light, but we correct at least 98% before publication. Any that do slip through are covered by our replacement or refund guarantee.