Been looking at the original 1922 lists and the 1935 (and later) revisions on the wiki, and it got me thinking, apart from obvious bloopers have the powers that be generally speaking got the numbering and routing right? Have they meddled too much and messed things up? What did they get right on the original lists that should never have been changed? Would any of the 1922 numberings that were later altered, make better sense today than the current numbering?

Finally the biggest question of all, does our road network need a major nunbering overhaul, particualy with respect to the F99s?

One case that got me thinking was the A30, given a major rerouting in the 1930's. Looking at the map the original route provides a much more direct route between basingstoke and Andover and even today the B3400 is a pretty good road, almost A road standard in places, despite 70+ years as a B road. There were I know good reasons for moving the road to its current route but i cant see why it couldnt have stayed as it was, and would be a usefull route today. Would avoid 2 multiplexes (A303 and A33) as well.

The whole A30-A303 is a blunder IMHO.

As it was known that A303 was going to the main route, why was the A30 number not transplanted?

That's the most obvious one. The A417/A419 between the M4 and M5 also needs to have one number throughout regardless of what it will do to the unimportant S2 sections that branch off it.

The 1935 renumbering of the A416 south of Amersham to A404 - while somewhat understandable at the time (though not fully, given the crazy routing and that it only gave the Knowl Hill (A4) to Rickmansworth (A412) section of the Reading - Hatfield turnpike one number, and the A412 N Orbital had been upgraded recently, and extended by Bucks CC in the same year to the A4, so it didn't really aid navigation).

In fact, that would be my 1922 blunder - the A32 not going to Hatfield via High Wycombe, Amersham, Rickmansworth, Watford and St Albans. Even if in '35 they renumbered it A33, and diverted it via a merger with the A4 to go via Knowl Hill, rather than Henley... Of course, I don't have traffic counts that they used, so I can't say.

The A417/A419 was the way we would want it in 1922. Ditto the A37 to Weymouth.
Because it was pre-35, and though the A30 was moved that year with the Basingstoke bypass, it was an almost unprecidented move (A1 Scotch Corner reroute, Kingston and A2 bypasses being the exceptions that weren't in the original plan) to divert an A road onto roads which already had A numbers. That it was numbered A303, rather than an extension of the A3036 (which was moved to Wandsworth) was also unprecidented pre-35.

Also, I'd imagine that most people would want the larger settlements on the A30, rather than tiny ones on the A303. Traffic counts again.

It was the 50s before the A41 and A1 used the Watford and Barnet bypasses respectively - they were even renumbered (as 3-digit roads) in 1935, but seemingly weren't seen worthy of the f99 status.

To understand how road numbers ended up the way they did, you need to understand the context in which they were introduced. Originally, A and B roads were going to be just like C roads are today - the council needs to know where they are so it can pinpoint indicents on them, but you don't see them on the signs. It was fairly late in the day that the decision was taken to publish the numbers on MOT Maps and put them sporadically on fingerposts. I'd be prepared to bet a Brynwich that no signpost for the A687 (Scotch Corner - Darlington) was ever manufactured. The flagship OS map of the time, the Popular Edition, was still concentrating on showing which roads were the best quality, as opposed to where all the traffic went.

That's how you ended up with all these really short and unnoticeable 4 digit roads that some of us delight in unearthing, like the A3039 (Axminster), or the A4 ending almost randomly in the middle of Bath by the old Post Office.

Even then, it wasn't really until the early 1930s that the Pre-Warboys direction signs appeared in earnest, at which point, particularly after the A30 got downgraded to the B3400, that local divisional engineers actually noticed where all the road numbers really went, decided a significant amount of it was bonkers (particular scorn was placed on 4 digit A roads like the A3039), got the 1935 Road numbering revision done.

In fact, if you read through the archive files, you really get a sense of the political situation involving some of the numberings - they weren't just fiddling things around on a map! Particularly regarding the very early rerouting of the A1 via Scotch Corner, this wasn't just "getting the route wrong", there was real anger and contempt to the MOT in London from local folk in North Yorkshire about sending the A1 via Northallerton, and a massive storm in a teacup ensued to get it changed.

Regarding the A354 taking over the A37 to Weymouth, I can quote directly from the MOT


I'd imagine the route of the A417 was chosen in a similar way, at the time, years before the M4 or any bypasses, it probably indicates that in 1935, the best route from Reading to Gloucester was via Pangbourne and Wantage, rather than right through the middle of Newbury, Hungerford and Swindon. Perfect sense when you look at it like that!

tl;dr - Numbering made sense for the circumstances of the time. But circumstances change.

I still find the numbering around Weymouth odd - I can understand why it was changed at the number, but I still think it would be better giving the main road into the town a 2-digit number (the A37).

I agree that the original numbering of the A417 and A419 made sense at the time, but now I feel it would be better if the numbers were switched as the main route has two numbers and neither really functions as a complete route any more.

Its a long and complex story. The A303 in it's modern form was built as the "New Direct Road" from Exeter between Andover and Honiton. It left the old "A30" what is now the B3400 at Andover and rejoined the old "A30" at Honiton.

It was built to speed up stagecoach traffic to Exeter and Plymouth (although much used existing and in some cases very ancient routes) as the A30 was not the fastest of routes then let alone now.

There was rivalry between the A303 and A30 turnpike owners for traffic to Exeter in much the same way as South West Trains and First Great Western now compete for Exeter rail traffic

The spiritual end of the A303 is actually the east end of the Honiton bypass where the original "A30" over Stockland Hill rejoined. This route was so awful that once the "A303" hads been built a new turnpike route from Chard to the "A303" was built to try and get traffic back onto the "A30" as places like Chard were suffering a loss of trade as a result of the "A303" That is why the A30/A303 junction is a Totso. It always has been.

Once the railways came the stagecoach traffic went and the A303, bypassing all the settlements fell into disrepair. However by the 1920's it was well known locally as a faster route to Exeter.

When the roads were numbered they made the principal routes the old Coaching routes so the A30 used the original coaching road through Salisbury and the A303 was basically a collection of B roads. A similar problem happened with parts of the A1.

After much protest the A303 was created but the MoT refused to make it a trunk road which did not finally happen until the 1950's when they finally gave up on the idea of improving the A30

The A30 was diverted via Popham as the B3400 was [I speculate] wholly unsuitable for trunk traffic passing through numerous villages and towns with narrow streets as can be seen in this picture of Whitchurch and the large town of Andover with a notorious Level Crossing.

Wheras the current A30 route via Popham (including the now unclassified bit between Micheldever and Sutton Scotney (the unclassified road that is important enough to justify a route confirmation sign to destinations nearly 100 miles away ) passed only through the settlements of Stockbridge and Sutton Scotney and was little longer mileage wise

So bypass Whitchurch, Andover etc and not have to build any road plus you make the Basingstoke Bypass cheaper by not having to build the western end of it between the A33 and B3400 (the economy in 1932 was in much the same state as it is now).

As to why they persisted with the A30 for so long as the trunk route. I suspect it was a mixture of "The Gentleman at Whitehall Knows best, What?" and that level crossing which although the rerouted A30 bypassed the A303 did not. It seems to have been the equivbalent of the current Stonehenge Bottleneck in it's day, raising concerns in Parliament as this extract from Hansard shows

A very interesting story indeed, but the question is now:

When the MoT gave up on the idea of improving the A30, why did they not switch A30 to the A303 route?

It was still common(ish!) back then for numbers to be moved onto a more sensible routing.

Is it possible at all that the ministry wanted to build an M30 to Exeter, and so did not want to waste time and money on removing a 3 from lots of signs?

Still don't fully understand the logic for redirecting the A30. The B3400 isn't that bad a road and is probably better in places than parts of the current non primary A30 further west (Salisbury to Shaftesbury and Chard to A303 spring to mind). Going through towns and villages seemed to matter less back then or the A30 would have taken the A303 route!

If the road had remained A30 or was returned to being A30 today then bypasses and link roads would have been built later. There is now an U/C road in Basingstoke between the modern a30/a33 and the B3400 and at the other end the B3400 terminates on the Andover eastern bypass which then free-flows onto the a303. Whitchurch and the intersection with the A34 would be the biggest problem on the route.

The B3400 is also a more direct route between Basingstoke and Andover. With an A30/B3400 you can still have the A303 (starting at first on the A33 then M3), and the M3 would still follow its actual route as a bypass for Basingstoke and the A33 (A30 using ring road to keep on course).

As for the A37/A354 to Weymouth issue, yes i understand the reasons why the Dorchester Weymouth road was changed in number, pre motorways Salisbury to Weymouth was a logical route, but today it is less so and the emphasis is more Yeovil - Weymouth i.e. A37 as this is a logical north south route for traffic heading for the M5 and beyond. So A37 to Weymouth would make far more sense and remove the A35/A354 multiplex. A354 would terminate on the Puddletown bypass. Not sure if I would restore A37 to Weymouth - Portland road. I'd say a353 but post relief road changes this would create a big multiplex along the WRR. An A3xxx perhaps?

What other original 1922 numberings would make sense again today when they didn't in 1935? What can be done with some of the F99 roads that have lost their importance? Should the numbers be given to more important routes?

As mentioned upthread, the A419.

Don't forget that very long distance traffic was a fraction of what it is now and London to Salisbury or Salisbury to Exeter would be about most peoples limit for both cars and lorrys by road. In the '30s the vast majority of London to Exeter traffic both passengers and Freight of all kinds would have gone by rail. There were vast goods yards at both Exmouth Junction in Exeter and Nine Elms in London to handle this.

If you look at it as the London to Salisbury Road and the Salisbury to Exeter Road it makes far more sense [in the 1930's] for the Main Trunk road [A30] to go through Salisbury but to provide a secondary, non trunk, Axx route bypassing Salisbury for the (relatively small amount) of London to Exeter traffic.

Plus - as well as that level crossing, the A303 went across the top of Salisbury Plain. Since stagecoach days it was it's achillies heel in winter and it had Snake Pass type snow gates as recently as the 1980's ***. In the 1930s it would have been a brave soul who took his, unheated in those days, car in mid winter across Salisbury Plain when he could take the safer route via Salisbury, for exactly that reason today people Drive from Manchester to Sheffield via the M62 in Winter rather than the A628 or A57. More likely they would have gone by train. As recently as the 1950s there were motorail services between Surbiton and Okehampton.

*** If Mr Corbyn is right about Solar Activity being the main driver of climate they may need to reinstate them shortly.

I think that's roughly what I meant. Today we would look at a route and avoid smaller towns and villages to allow for a better quality long distance route, back then as you say people would want to go either through or to those towns so yes it would make more sense to route a main road through them. When they rerouted the A30 they seemed to do the opposite. The old route went through places like Whitchurch to get from Basingstoke to Andover and was more direct, the 'new' route of A33 and pre-A303, seems to take the more modern and, for the day, a more anachronistic option to avoid built up areas (apart from Sutton Scotney) and take a longer route to get from A-B. So really the B3400 route makes more sense in that context so why change it at that time? Doing it later would perhaps make more sense and would be perfectly understandable.

Thinking about it there must be other places where a particular numbered road runs through built up areas and would have been the main route originaly but has been superceeded by a road that avoids most of the major towns. A natural bypass is created, but like the A30/A303 business, the numbering has stayed the same, thus still reflecting the route planning of the time that numbering took place, rather than modern route planning.

One i can think off immediately is the A35/A31 past Bournemouth ant Poole. A35 forms the continous route from Honiton to Southampton, via Bournemouth and Poole but the main route between the two destinations uses the A31 to bypass Bournemouth and Poole. If you were numbering today you would perhaps either have the A35 take the A31 route or have the main Honiton to Southampton route as A31.

Our local example is the A60 - you can go (if you include a bit of A6) from Leicester to Docaster on it, but it is full of villages, and also Loughborough, Nottingham, Mansfield and Worksop.

The alternative for Leicester to Nottingham is the A46/A606 (the A606 was improved to an S3 in the early 1930s. Similar thing happened to make the A614 a route between Nottingham and Doncaster. And finally the A46, the (now) A6097, and the A614 to link Leicester with Doncaster. All three of these roads were upgraded in the 1930s, and all of them are sparsely populated.

In a nutshell, I'd say that they did get it more or less right in 1922. But as noted above, circumstances change, and have changed many times over since then. It seems to me, at least, that our problem is that there doesn't seem to be much logic or consistency in the ways that those changes have been responded to.

The A303 and A417/A419 examples mentioned above, and discussed at length in other threads, bear this out.

A303 - an example of where they created a new major route out of existing roads, and they thought "let's give it one number throughout" - and quite a good one too, even if it has three digits.

A417/A419 - they kind of did the same thing a few decades later, but on that occasion no said "this is one road, so it ought to have one number". And a little piece of it near Cirencester actually appears to be a major D2 without any actual number at all!

No consistency, in either practice or logic.

Just out of interest, where else have they done aa "A303" and made one road out of many? Apart from the A14 I suppose. Speaking of which, it could quite fesably have been numbered A45 throughout as it almost mirrors the pre-existing A45 route - why A14?

And before the old A45 and current A120, the A604 was a composition of different routes.

I guess they used the A14 number for the whole D2 from the M6 to Felixstowe because using A45 would not quite have matched up with the existing Birmingham to Northampton portion of that route.

Also, the original A14 was absolutely pathetic, and a real waste of such an important number.

I just wish they'd put hard shoulders on the A168 from Dishforth to Thirsk and call it the M19 or A19(M) to get rid of that odd bit of A168 which many people refer to as "Turn off the A1 onto the A19 for Middlesbrough."

Or even make it A19, and reclassify the S2 bit down to York and Doncaster as something A1xx.

How about A168 for the York-Thirsk part, and also the A1079?