Oddities and anomalies
When road numbers were first allocated in 1922, the system was virtually perfect. The most important roads had the shortest numbers. Every classification reflected accurately the traffic use and local importance of its road. Road numbers were correctly allocated by zone and grouping. Each section of classified road had one number and one number only.
Such happiness cannot last forever, of course, and as the original spirit of the numbering system was forgotten, mistakes crept in. Priorities changed, so the A30 plays second fiddle to the A303. Roads run concurrently for the convenience of planners. These and many other curiosities have been exhaustively logged, categorised and discussed over the years by members of SABRE, and so this page is simply an introduction to a complex and intriguing world.
With the topic of road numbering, it's quite satisfying that many of these oddities can be categorised with mathematical terms.
Some roads stay where they are in admirably reliable fashion. If you went looking for the A676 at any time between 1922 and the present, you'd find it between Bolton and Edenfield, faithfully manning its post. Other roads sometimes go AWOL.
The A471, for example, used to run from Abergavenny to Usk. Not any more. It was reclassified as the B4598 at some point between 1980 and 1988. The A398, on the other hand, was eaten by another route, the A361, as it went steaming through North Devon in the 1970s.
Now you see it, now you don't: the A648 in 1967, 1972 and 1988
Some routes just go into hiding and don't want to be found again, to the extent that they'll have several changes of postcode. In 1922, the A648 was assigned the route between Brighouse and Denholme in West Yorkshire. But it had tired of this vocation before long, and by 1953 the A644 had been extended along this route, obliterating the A648. It reappeared in 1968 as the new link road between M1 junction 24 and Nottingham. There must have been another change of mind, though, because it had given up again by 1988, the Nottingham link became part of the A453, and it hasn't been seen since.
If you have a road that is something of a success, you might like to capitalise on its popularity by making it go even further. In most situations this isn't a problem, though it might create an unexpectedly protracted route if you're not careful.
Roads from zone 3 like to travel a lot, mostly thanks to the process of addition. The A34 is a good one; it started life reasonably enough as the Southampton to Oxford road, but then got a bit carried away, ploughing onward to Birmingham and then Manchester. Its northern terminus is now in Salford. The A38 is the UK's second longest classified road (the A1 is longest): initially it was the ambitious enough Plymouth to Derby, but had new bits stuck on each end and is now the positively astonishing Bodmin to Mansfield road. The A361 pulled off a similar trick to become the scenic route between Ilfracombe and Kilsby, Northamptonshire.
The problem with attempting something like this is that your addition might well carry the road across a zone boundary. By the middle of the twentieth century this obviously wasn't being seen as a problem, and greater emphasis was placed on retaining a suitably important number. The A66 is a good example. Initially it ran from the A6 at Penrith to Scotch Corner and then Hull. Its eastern terminus became Middlesbrough before too long; then in the early seventies a coast-to-coast trunk route was created by extending it west, across the A6, and into the Lake District. It should now have a zone 5 number, but A66 sounds better and so A66 it remains.
A similar trick was pulled off with the M62, which was meant to start at Manchester and strike out over the Pennines. Late in the day the M52 Liverpool to Manchester motorway was absorbed to form a coast-to-coast motorway (near enough), with the result that the M62 is out-of-zone at its Liverpool end.
Let's go to Leicester. You never know, you might like it; and even if you don't, we won't be staying there for too long anyway.
The A594 Waterloo Way, Leicester
This is the A594 Leicester Central Ring Road, a route built in the sixties and seventies to carry traffic around the city centre. I know what you're thinking, though: it's very urban, isn't it? Very built up? Not to worry: the A594 is one of those roads that are so good they actually appear twice. If you tire of circling Leicester and instead want fresh air and scenery, then you should choose the other A594, which runs from Maryport to Cockermouth in Cumbria. It's been there since 1922 (and was once much longer, before the A66 made its westward lurch) but that didn't stop it spawning a twin in the East Midlands.
This little incident is far from isolated: look hard enough and you'll find that as well as the B198 Lieutenant Ellis Way in Cheshunt, there's also a B198 running all the way through Wisbech. The B5444 crops up in both Swansea (in the wrong zone) and Mold (in the right zone), but not in between. There's plenty more.
Perhaps the most bizarre is the A601, which reasonably enough is part of the Derby Inner Ring Road. Its partner is the A601(M), which is just north of Carnforth in Lancashire. A more strange number for this motorway is hard to imagine. You have to wonder whether it really ought to be in Derby and there are people there wondering where it has gone.
You can save yourself the trouble of having to duplicate a road by simply taking an existing road and chopping it in two. This is best done with long routes where people are unlikely to notice. It's tempting to call it "long division".
The usual culprit for this sort of thing is a new motorway. The old trunk road alongside the motorway will be renumbered or downgraded completely, but only for the duration of its replacement, and sometimes not even for that. The M5 relieves the A38 all the way from Birmingham to Exeter, but only south of junction 27 has the A38 been disguised as the B3181. It now lies in two halves, Bodmin to Exeter and Sampford Peverell onward.
A much more thorough effort was made when the M11 replaced the A11. Those responsible clearly enjoyed their work and made a very complete job of it. The old road has become variously the A104, B1393, A1184 and B1383. It leaves an important A11 running from the M11 at Great Chesterford to Norwich, and a slightly more obscure leftover running from London to Wanstead. Much the same was done to the A47 with the arrival of M6 and M69 across the Midlands, with a gap in the route between Birmingham and Leicester.
The haste to finish the M40 meant that renumbering across Warwickshire was considerably more slapdash, with the old A34 becoming A3400 and the old A41 now A4100. It's the road numbering equivalent of wearing dark glasses and a rather unconvincing false moustache. All the same, it does leave the A34 and A41 in two distinct parts each.
The Severn mysteries
It's not a mathematical term, I admit, but cramming "Severn" into the list is still in keeping with a numerical theme in my book.
The issue here is at Bristol and concerns the motorway zones. Until 1996, all was well, with the M4 and M5 crossing each other, and the Severn continuing the boundary between zones 3 and 4. The opening of the Second Severn Crossing saw the M4 diverted a little way to the south, leaving a section of motorway across the original Severn Bridge and a new connection to the M5 at Avonmouth. So far so good.
One of these routes was in zone 3 and one was in zone 5. So what were they called? Why, M48 and M49, of course. And this is where it gets interesting (or particularly dull - if the latter, you should probably stop here). You could argue that the M48 and M49 are both out of zone and therefore incorrectly numbered. That's certainly the most obvious conclusion to draw. But you can make a case for them being correct. It's up to you.
The first trick is to get the M48 back in zone. Because there are no other motorways between M4 and M48, one can claim that when the M4 moved south, the boundary between zones 4 and 5 didn't. It can't be proven either way, but supposing the zone boundary remained on the original line* (so that the M48 now forms the boundary) then the M48 is correctly numbered.
One down, one to go. If the M48 is the zone boundary, we now have the M49 and, oh dear, the M4 Second Severn Crossing both in zone 3. This won't do. The solution is this: we know that the Severn Estuary forms a zone boundary. We know that there are no other motorways on the south side of this. The only two pieces of evidence that exist for motorway numbering south of the Severn and west of the M5 are both zone 4 routes. So, according to the only available evidence, zone 4 must be hourglass shaped, squeezed to nothing at Almondsbury Interchange and widening back out again**. If it is, this puts M49 back in zone. Perfect.
On the other hand, M48 and M49 could both have been selected because they connect to the M4, and numbers beginning 3 and 5 would have looked odd. But that's no fun.
The rest of the alphabet
The government, all manner of cartographers and highway professionals are part of a conspiracy. They will tell you that road numbers are always prefixed by A, B or M. As readers of this site will probably already know, that isn't the truth.
The fact is that numbers that are part of the national numbering scheme are always prefixed by A, B or M. But virtually every road in the UK - and many of the bridlepaths, footpaths, green lanes, byways and other public rights of way are also numbered. It's not just C's either - there are plenty of D- and U-roads in the list (that's U for unclassified, which is rather oxymoronic, considering that all U-roads are by their nature classified).
The extent to which it occurs is quite phenomenal. One report suggests that Devon County Council keeps such a close watch on the rights of way under its control that it uses virtually every letter in the alphabet, including X, to number the different classes of track, back lane, housing estate street and footway that it maintains.
These odd numbers shouldn't appear on road signs, but occasionally they do by mistake. CBRD has a photo gallery of locations where this has happened.
* And this does happen - sometimes. On the two occasions that the A1 has changed its alignment around Newcastle, all the surrounding roads have been renumbered so we can assume the zone boundary moved. But on other occasions, such as when the A3 was rerouted around the Guildford Bypass, the zone boundary has stayed put and resides on the old road through town.
** This happens too. If you go right back to the original numbering scheme, the A5 and A6 bounce off each other in St Albans town centre, meaning that zone 5 begins in London but is squeezed down to nothing before widening out again. These roads are now renumbered but the zones still do this.