Many of the registers of seafarers pre-1861 use a code called the port rotation number to show which ship the seafarer sailed on.
The quickest way to find details of a ship is to use the CLIP search page. Ships by port rotation number
You enter the port rotation number and the port/port number of the ship. The results will identify the ship's name, with links to further data.
This page gives full details of how the system worked and how to use the CLIP index to trace the records of a seafarer.
The port rotation numbers were a code used by clerks at the Board of Trade in London during the 1840s and 1850s to compile registers of seafarers which are now held at The National Archives (TNA). They were used particularly in the registers for 1845 to 1855 (TNA, BT 113) and cover more than 500,000 records.
From 1835, masters and owners of ships were legally required to make a written crew agreement which showed a list of the crew. After each voyage or every six months, the documents were returned to the port officials and forwarded to the Board of Trade. They are now held at TNA in series BT 98. The data from the crew documents was used to make up the various registers of seafarers using a numerical code.
The code included the port number which showed the ship's port of registry and the port rotation number, which showed the ship the seafarer had been working on.
The registers of seafarers and the crew lists have survived. The key to the port numbers is known Port numbers but, unfortunately, the original key to the port rotation numbers has not survived. So there are extensive records of seafarers showing which ships they worked on, but in code. There are also the crew records for the ships but, without the code, there is no way to make the connection between them.
For example, a seafarer's record in BT 113 might show a code: M, 456-64-6/51.
Most of it is clear: M means Mate, 64 is the code for London and 6/51 shows that the voyage ended in June 1851.
456 is the port rotation number and that is the missing key. There is no way to work from that number to the ship's name and so to the crew documents.
They could be anywhere in the forty boxes of crew lists in BT 98 for London ships for 1851, each with a hundred or so documents. The boxes are arranged in alphabetic order of ships' names, but if you do not know the name, that is no help. Searching those boxes would take some time.
There is a further example, here: Example
The official position was that 'No index to the port rotation numbers has been found'. That was true, until now...
CLIP was approached by family historian Peter Hamersley, who has researched crew lists for the period using images of the documents which are available online on FamilySearch via FamilySearch Centres or FamilySearch Affiliate Libraries. This online access has been a vital part of the project and is also important for researchers. Details are here: FamilySearch images
Peter suggested that the port rotation numbers were assigned to individual ships and did not change. He had made up a list of port rotation numbers for his own use and suggested extracting data from the crew documents to compile a modern version of the key to the numbers. That would take some considerable time, but the idea was that the few hundred boxes of crew documents for one year would provide a key which would work for other years as well.
Peter provided CLIP with some of the data he had extracted for Whitby ships. Peter Owens compared that with other data which CLIP has from the shipping registers for Whitby, and made a significant discovery: the port rotation numbers went in step with the numbering in the shipping registers. So the port rotation number did indeed belong to a particular ship, though with the slight twist that a ship gained a new port rotation number if she was re-registered, even if the new registry was at the same port.
This link to the shipping registers is highly important. It provides the system that must have underlain the allocation of the port roatation numbers.
The link can readily be tested against data from other ports. It works!
That being so, it provides a way of checking data extracted from the crew lists. Even better, it makes it possible to extrapolate from the extracted data for a port to other ships and other years providing extra data for no more transcription effort. A relatively small amount of data extraction would provide the key to most of the seafarers' records.
We explain below exactly how lists of port rotation numbers were created and used by the Victorian clerks.
We also explain how we have worked with Peter and volunteers on the Port Rotation Numbers Project to make the modern equivalent of the clerks' lists, which is now available on this site.
Several reference numbers turn up in the records of ships and seafarers. However, different documents and authors sometimes use the same term to refer to different things, so it is good to be sure what is meant.
We'll need to refer to some of them below, so this list explains how we use the terms:
|Numbers used as a shorthand for ports of registry - for example, London was port 64.
|Register entries at a port of registry were numbered sequentially, starting at 1 each year. This was used as a way of identifying a ship - for example, Mary, Hull 45/1842.
|Port rotation number
|Number allocated to a ship and used as part of a code to identify the ship and her port of registry.
|Register ticket number
|Number allocated to a seafarer in the registers of seamen.
Figure 1 shows how the port rotation numbers were allocated. The data is fictional.
They started from the shipping registers at the port of registry.
When a ship was registered, a fresh entry was made in the local shipping register. The entries were numbered consecutively, with the register numbers starting again at 1 each year.
That register number was marked on a copy of the register entry which was given to the ship's owner. It formed part of the ship's papers and was proof that the vessel was a British registered ship.
A further copy (transcript) of the register was sent to the Board of Trade (aka Registry of Shipping and Seamen, RSS) in London.
The port rotation number list was probably compiled from these transcripts. As each new transcript arrived, the clerks added details of the ship (name and register number) to the list of ships for that port and the new entry was allocated the next port rotation number.
The port rotation numbers therefore go in step with the register entries back at the port. Though the register numbers started at 1 again each year, the port rotation numbers carried on in sequence.
If a ship was re-registered at another port, it gained a new port rotation number corresponding to the new register entry when the transcript of that arrived in London. That number would be the next available port rotation number for the new port. The same applied if the ship was re-registered at her existing port (registered de-novo). A new port rotation number was allocated, superseding the old one.
Ships could therefore have several port rotation numbers, one for each registration. The reason was that the ship was identified on the crew list by the register number and year (see Figure 1) which would have been the most recent registration details. Also, it was probably simplest for the clerks just to allocate a new port rotation number, rather than searching back (unreliably) or cross-referencing to the old details.
Similar reasoning was probably why numbers were not re-used when a ship was lost or sold abroad.
The ports of registry also made annual returns of ships registered and it is possible that these returns were used as well.
The remaining question is what was the source for the first lists of port rotation numbers?
The port rotation numbers go in exact sequence with the register numbers from 1846 onwards. Prior to that, they are still linked to the register numbers, but it appears that some ships do not appear in the list of port rotation numbers. That would indicate that the lists were originally compiled in 1845, starting with ships which were still extant at that date. It is possible that the ports annual returns of shipping for 1845 were used to start the list, but unfortunately, only the returns for 1850 have survived and are in BT 162/19.
As we have worked further into the documents more details and twists have emerged, but they have confirmed and clarified the initial idea.
Figure 2 shows how the port rotation numbers were used. The data is fictional.
The port rotation numbers were used to compile the registers of seamen's service. There were various registers - some for seamen and others for officers. These registers have survived and are at TNA in BT 112, BT 113, BT 116, BT 119, BT 120 - some of which are indexes to the other volumes. Records for masters are in BT 115 and BT 124.
The source of their data was the crew lists and agreements which were returned to the Board of Trade at the end of each voyage or at the end of a half-year. These documents were headed with the name of the ship, her port of registry and her register number and year, together with details of the owner and master and the period that the list covered. Below that was a list of the crew, with their age, birthplace, previous vessel and the capacity in which they were engaged.
The first task for the clerks was to identify the ship by looking up the port rotation number list for that port and then running down it to find the correct ship, register number and year. Against that, they found the port rotation number. That was then written on to the crew list as part of a code, which was often in this pattern:
Port rotation number-Port number-Date the voyage ended eg: 236-104-6/47Once that was done, the crew document was passed to the clerks whose job it was to compile the seafarers' registers. They had the difficult task of finding the seafarer in the registers - difficult because the seafarer's name was handwritten on the crew list, often badly. However, on most crew documents, the seafarer's ticket number was also shown, which must have been of great assistance to the clerks.
Having found the right seafarer, they then copied the code from the crew list to the register entry for that person.
They then moved on to the next member of the crew and repeated the operation.
Figure 3 shows how the port rotation numbers have been recreated by a joint project between CLIP and Peter Hamersley. The data is fictional.
Recreating the list of port rotation numbers starts from data extracted from the crew lists in BT 98. The original documents are available at TNA and also as images made by the LDS Church and available on FamilySearch (though only at FamilySearch Centres or FamilySearch Affiliate Libraries).
CLIP and Peter Hamersley have worked through the crew documents for 1851 using FamilySearch films at FamilySearch Affiliate Libraries. There are 458 boxes of them, from BT 98/2371 to BT 98/2808 with an additional seven boxes of colonial records. From the documents, we have produced a list of the port rotation numbers, names of the ships and their registration details.
In parallel, CLIP has transcribed the data for each port from the annual returns for 1851 in BT 162/19, images of which can be downloaded from TNA for free. We have recorded the ship's name, register number and year, and tonnage as shown in the returns. BT 162/19 covers all British ports except London, for which there appear to be no returns, unfortunately.
However, BT 111 includes a full index of London registered ships from 1833 to 1856, which we transcribed. We added data from earlier parts of BT 111 which extends the data back to 1820 for all British ports.
To allow us to deal with data in BT 98 for Colonial ports (almost all of which were Canadian registered ships), we added data from the folios for Canadian ports from the 1851 Colonial annual returns in BT 162/20.
We then added other sets of data, such as the early entries in the Appropriation Books, some of which show register numbers and years. This data set is useful in its own right and we have made it available on the CLIP site - see below.
By cross-correlating these data sets, we were able to check the data in the various sets, which is important as these are all hand-written documents and the writing is sometimes not easy to read. Then we combined the data sets into a composite index to the port rotation numbers which shows the ship's name, port rotation number, register number and year, and tonnage.
There is a bonus however. Where the port rotation numbers run in strict sequence with the shipping registers, port rotation numbers can be deduced for ships for which there is no BT 98 data, and similarly register numbers and years can be deduced for ships for which those are missing from the other data.
So the final result is that working from the crew lists for one year has produced data which applies to all years for that particular ship's registration. It is not all the port rotation numbers, but a substantial proportion of them - over 90% for many ports for the period 1845 to 1851.
By far the most important output from this project is the Port Rotation Number Index.
This provides a resource to deal with a long-standing problem, giving the key that has eluded researchers for decades and unlocking the data for more than 500,000 seafarers.
The new index is now available on this site by following the link. The index page shows the coverage by port.
There are three further outcomes from the project:
The data from BT 162/19 and BT 162/20 provides lists of ships extant in 1850 by port, joining the existing CLIP data taken from the Mercantile Navy List (MNL) for every tenth year from 1860 to 1940. The data from BT 111 provides lists of ships back to the 1820s by port, which has also been added to the existing CLIP data from MNL.
The combined data has provided the opportunity to create a full ships' name index for the period from 1830 to 1855, with an estimated 30,000+ entries. This includes many small vessels not included in the records of insurance companies such as Lloyds. As with all CLIP data, links are provided to other sources such as existing register indexes and transcripts.
The data extracted from BT 98 provides the beginnings of an index to those pieces (boxes) for TNA.
Our initial technical description of the project is here: BT 98 technical description .
Figure 4 shows the register entries in BT 113/208 for an apprentice, James Breckon of Goathland (which is just inland from Whitby). He was apprenticed in 1851.
The record shows a port number of 104 which means that the ship he sailed on was registered in Whitby. The port rotation number, 86, shows the ship. Her name is not recorded, but the entries show that James worked on the same ship from 1851 to 1853 because the same port rotation number appears in each year.The voyages in 1851 and 1852 were in the home trade (around the British Isles and the continent from the River Elbe to Brest) because the entries cover both columns. Those for 1851 cover two half years ending June 1851 and December 1851, but there is only one list for 1852.
The record for 1853 (Figure 4) shows a foreign trade voyage. The ship is the same - 86.104.1 is shown top-right. (it's not clear what the 1 refers to and it can be ignored). The voyage is shown on the row below. She sailed from Woodbridge (port number 107) on 17 September 1853 and is shown arriving back on 9 September (which is probably a clerical error).
There is nothing to indicate the ship's name, so finding her crew documents could be difficult. There are ten boxes of documents for Whitby for 1853, references BT 98/3601 to BT 98/3610. Each of them contains up to a hundred documents for different ships. Knowing the ship's name would narrow the search down to just one of those boxes.
Using the CLIP search form, the results (Figure 5) show that the port rotation number 86 refers to a ship named Concord registered at Whitby, 6/1837, 189 tons.The results show that the crew lists for 1851 will be found in BT 98/2778, covering Whitby B-C. We are sure about that, because that is where our data came from.
The results also show where crew documents will probably be found for other years, for example in piece BT 98/3603 which covers Whitby CI-D for 1853.
The crew lists and agreements would show some details of the voyages and an example is shown in Figure 6 - a foreign-going voyage to Cronstadt in July 1851. This is not listed in James's records, but he is shown as an apprentice (Figure 7).
Newspapers can also be used to fill in details - for example, there is a newspaper report of the Concord arriving in Gothenburg from Woodbridge on 25 September 1853.
The 1851 Port Rotation Numbers Project would not have happened without Peter Hamersley's initial contact. That provided the stimulus which led us to discover the link to the shipping registers - a significant breakthrough in solving the mystery of these numbers.
Peter has followed up by joining us in the hard work of digging the sample data out of the crew agreements.
Thank you, Peter.
We are most grateful to Bruno Pappalardo at TNA who has provided valuable support and assistance.
We thank our volunteer transcribers, Robert Holloway, Laura Hamersley and Janine Hamersley in Australia, Maxwell Crockett in NZ and Myfanwy Gate at the National Library of Wales (NLW).
The staff of NLW have been most helpful and Anglesey Archives have provided valuable assistance in obtaining access to the records online.