How the Motorways were Numbered

27th November 1959

T. R. Newman

I understood the basic principle was to be a radial system of numbering with those motorways which lie between the main radials numbered according to the radial from which they spring.

Can you please explain how M.45 which is a spur from M.1 fits in with this numbering?

Whilst this memo seems very odd, no decision has actually been made about how to deal with the minor motorways! M45 is assigned rather out of the blue - probably because it fits neatly into both the "sector" system, and the idea of numbering motorways identically to the A-roads replaced.

Proposals fleshed out - 18 December 1959

A. W. Lovett:

You will remember that certain numbers have already been agreed (minute 5), among them M1 for the London - Yorkshire Motorway, M2 for a possible Channel Ports Motorway, M3 for the Exeter Radial, M4 for the South Wales Radial, M5 for the Birmingham - Bristol Motorway (and no doubt its eventual extension to Exeter) and M6 for the Penrith - Birmingham Motorway (and its eventual extension to join the London - Yorkshire Motorway north of Crick). The first four of these radiating from London will quarter the country and will give their numbers to sectors, or quadrants, bounded by them (or, I suggest, the mouth of the Thames) - thus sector 1 will lie east of M1 and north of the Thames, sector 2 south of the Thames and east of M3, sector 3 between M3 and M4 and sector 4 between M4 and M1. Where a motorway is simply a bypass along an existing route the letter M would be added in brackets to the existing number - e.g. A1(M) for the Doncaster Bypass.

I now suggest that sector 5 should lie to the west of the line formed by the M5 and the M6 between Bristol and Carlisle (this is in line with the practice for numbering Class A roads, all the numbers in the Lake District, for example, being numbered in the 500 series even though A.5 is over 100 miles to the south), and sector 6 to the north and east of M6 and west of M1 and its projection northwards. (This means that sector 4 would be totally enclosed by M1, M6, M5 and M4.)

Turning now to the detailed proposals for numbering motorways, I deal first with the only motorway not already numbered on the northern section of the map, the Darlington Bypass and Durham Motorway. In accordance with the last sentence of para. 3 this should be A1(M) despite the fact that this is a considerable length of motorway totalling 32 miles.

On the southern section of the map those motorways lying to the west of M6 and M5 from Kendal to Bristol (sector 5) would be numbered in the M50 series. I suggest M59 for the Blackpool Link and M56 for the Helsby and Frodsham Link. The Ross Spur, as the most important (indeed the only) offshoot from M5, would be M50. The Port Talbot Bypass would be A48(M). If any other motorways are built in sector 5 north of M59 a number could be borrowed from sector 6, since if the proposals outlined in the following paragraph are adopted there would be several numbers vacant in the upper reaches of the M60 series.

Ironically, sector 6 contains two of the largest conurbations in England - Greater Manchester and West Yorkshire, as well as part of South Yorkshire and West Midlands! This shows that the Ministry never expected the local authorities, especially SELNEC, to get the motorway bug quite as much as they did.

In sector 6 (north and west of M6), Mr. Wykes has already agreed that the Stretford - Eccles Bypass (west of Manchester) and the Lancashire - Yorkshire Motorway as far as M1, which are continuous, should be M62. (The Lancashire - Yorkshire Motorway parallels the A.62 from Manchester to Leeds).

Although M2 has been reserved for a possible Channel Ports Motorway I understand that it is now unlikely that this road will ever materialise. The nearest approach to it will be the Medway Motorway which, according to the principles set out above should be numbered A2(M). If this is so, however, we may never have an M2 unless it should be the Brighton Radial, but as Mr. Adams points out in minute 13 M2 should lie close to the Thames. The Medway Motorway is 26 miles long, shorter than the Darlington Bypass and Durham Motorway which I have suggested should be A1(M). My recommendation would be nevertheless that the Medway Motorway should be M2 as you will see from Doc.26 that the Daily Telegraph assumes that it will be. I also recommend that the Maidstone Bypass be A20(M) and the Brighton Radial M23 (because it parallels A.23).

In a short article and editorial in the Daily Telegraph, 15th December 1959, regarding tenders for the Medway Motorway River Bridge had the heading "Tenders for M2 River Bridge" and used this reference throughout, therefore the current M2 was born.

In addition, of course, bypasses that are eventually linked to form a continuous motorway will preserve their A numbers (plus (M)) until they are so linked. Thus the Maidenhead Bypass will be A4(M) until it becomes part of M4. I would propose an exception to this in the case of the Lancaster Bypass which is due to be opened in the Spring of 1960. I think this should be M6 since the Preston bypass a few miles further south is also M6 and the two are due to be linked in about 3 years.

More sense from the Ministry. Allow yourselves a "get out clause" for the Lancaster Bypass. Mind you, then they manage to come up with the seemingly-confusing renumbering principles for when a motorway got built as a series of bypasses - like A48(M) Port Talbot Bypass becoming M4 later on.

It would be prudent I think to regard all these numbers as provisional until they are finally allocated just before the signs have to be ordered - or at least until we have committed ourselves by communicating them to the Ordnance Survey or other map manufacturers - so that any motorways to be proposed in the future can be fitted into the numbering system and any necessary modifications made to numbers not yet finally allocated.

Yeah! Let's not actually decide on anything until the last minute. Seems rather dangerous to me, given how fast Government departments move at the best of times...

Mind you, this does explain why (even today) motorways are almost never allocated a route number until the last minute.

Return of the tree system - 29 June 1960

C. E. Hollinghurst:

The proposals outlined in minute No. 18 on this file which you passed to Mr. McNeil with your minute No. 21 are discussed in some detail in the subsequent minutes. There are two points where we think the proposals can with advantage be improved upon.

1. It is suggested that the Slough/Maidenhead Bypass should be numbered M4 from the start, the practice adopted as Preston and now Lancaster is a useful precedent.

2. That in numbering the short spurs the principle of a tree should be followed rather than the zones which from the basis of the classification numbering. On the tree system spur roads would carry as the first figure of their two figure numbers the single figure of the parent motorway. The only exception which would arise would be M45 running from M1 but this can reasonably be justified in the light of its close parallel track to A.45.

The following numbers are proposed.
London Airport Link - M44
Blackpool Link - M66

As you know the South Wales Motor Road takes off from the Slough/Maidenhead Bypass some little way east of Maidenhead Thicket there will thus be eventually a link at this point as well as the Huntercombe Spur into Slough. It is suggested that these might be numbered -
Huntercombe Spur - M46
Maidenhead Thicket Spur - M47

More tree madness! They're not giving this fight up easily, are they? Mind you, they are proposing allocating numbers to utterly tiny spurs. The Huntercombe Spur, for example, is that little tiny spur at M4 J7. It's more like long sliproads than a spur!

As an aside, Mr. Hollinghurst was was an early (1942) advocate of a London-Birmingham motorway routing which would take traffic from sections of both the A5 and A6. In other words, pretty much what we have today.

The sector system supporters hit back - 3 August 1960 - Minute 34

A. W. Lovett:

In the first place, may I say that I agree entirely that the Slough/Maidenhead Bypass should be M.4 from the start. I have thought for some time that the idea of changing the numbers of motorways at any stage was a weakness in our original proposals.

The "tree" system of numbering in preference to the sector system is at first sight attractive, but on closer examination I think there are certain objections to it. The motorways for which detailed numbering proposals are now under discussion are those planned for the next ten to fifteen years; by the end of that time it is probable that more motorways will be planned, at least in the form of spurs from the main routes (M.1 to M.6). The longer the main route, the more spurs from it there are likely to be and the more likely we are to run out of appropriate two-figure numbers if spurs on both sides of the main route are all numbered in the same series. Under Mr. Hollinghurst's proposals, for example, we should already have used six of the M.10 series and four of the M.60 series, mostly for spurs from the two longest main routes, while few numbers in the M.50 series would ever be used. It is even possible that some spurs might be extended to link up with another main route and under the "tree" system the numbering of the extended spur would not be consistent at both ends. In fact, the sector system is generally capable of more consistent application than the "tree" system - neither the M.45 nor the M.62, to both of which we are committed, fit in with the "tree" system, but they do fit in with the sector system.

I still consider, on the grounds of consistency, that the London Airport Link should have a number in the M.30 series; to meet Mr. Newman's point about confusion with the A.30, I suggest M.32. The Huntercombe and Maidenhead Thicket spurs will both be comparatively short and could perhaps best be treated as long sliproads to the A.4 and left unnumbered (as the Pepperstock - Kidney Wood spur from M.1 is unnumbered).

Mr. Newman has raised the question of M.15 for the North and South Orbital Roads. I think almost any available number between M.1 and M.49 would be suitable, but if M.15 does not find favour, I suggest M.25 instead - the motorway runs parallel to A.25 for a greater distance than to any other all-purpose road.

I think these papers should now go to Mr. O'Neill to arbitrate between the sector system and the "tree" system.

As a trivial aside, this is the first mention of using M25 for a London Orbital Motorway.

He certainly has a point about motorways that meet two other "main" motorways. Examples today would include M62, M42 and M49.

Part 4 - Decision Time >>

How the Motorways were Numbered


Part 1

Part 2

Part 3

Part 4

Part 5


Thanks to Jonathan Winkler, Chris Marshall and Mark Dillon for their help towards the production of this feature.