Priority of cycle tracks across side roads

In summarising the results of their recently-conducted survey into the opinions of both members and non-members into varying types of cycling infrastructure, the CTC had this to say on the particular matter of cycle tracks –

Respondents were offered a set of conditions that CTC suggested be met before implementation of cycle tracks. These include cyclists having priority over turning traffic, adequate width and a high standard of maintenance. Support for these conditions was very high, with 88-95% of respondents saying that these were very or quite important.

Later in that report of the survey, the CTC go on to say –

Traffic regulations in the UK only permit cyclists to have priority over vehicles entering or leaving side turning[s] in very restricted circumstances. In nearly all cases cyclists must give way and are sometimes blocked from crossing with barriers, as in Figure 3, an image from Oxford. By contrast, rules in the Netherlands (Figure 4) permit priority for cyclists over side roads, without the need for anything other than road markings and some signs.

This is the vexed question of priority for cycle tracks across side roads, which is a particular bugbear for the CTC. Roger Geffen, their campaigns director, had this to say in the Guardian last year, in response to the findings of the Understanding Walking and Cycling report –

[An] important pre-requisite for segregation to work is legal priority for cyclists at junctions, given that this is where around 70% of cyclists’ injuries occur. In countries like Denmark and the Netherlands, if you’re on a cycle track and travelling straight ahead at a junction, the law says you have clear priority over drivers turning across your path

Obviously that’s the way things should work, but the CTC are concerned that we just can’t do that in this country.

This is the picture of a cycle track in Oxford that the CTC refer to in their report on infrastructure, used to illustrate the ‘lack of priority’ in the UK –

This is plainly awful design.

The cycle track runs along Park Street in Oxford, and the side road is a minor entrance into the University Science department. There should be no barriers, and the track should have priority across the side road.

There is, of course, no UK legal requirement for barriers like this at junctions. This is just sloppy, lazy engineering. The CTC suggest that barriers ‘sometimes’ block crossings, but they should also have made plain that such a tactic has no basis in regulation, and that they should be removed. Indeed the most idiotic section of barrier has disappeared since the CTC’s picture was taken. There was no legal requirement for it to be there in the first place.

Barely a hundred yards from this complete mess of a junction, the cycle track crosses another side road, in slightly smoother fashion.

No barriers, and no ‘give way’ markings. Cyclists would seem to have priority – albeit unclear, and potentially dangerous.

The CTC claim that a lack of priority is basically what we have to expect from cycle tracks under UK law, regulations of which mean that cyclists ‘have priority over vehicles entering or leaving side turning in very restricted circumstances.’

Is this true? A respondent to their survey – quoted in their report – doesn’t seem to think so.

It would be good to have new rules to support right of way at side roads as a default, but I don’t see them as essential since as far as I can see, high quality Dutch style segregated tracks could be built with currently available give way and priority markings.  

This stands in disagreement with the CTC’s claim about priority for cycle tracks only being available under ‘very restricted circumstances’. What’s the reality?

The relevant document is the latest edition of The Traffic Signs Regulations and General Directions, 2002, available as a huge 300MB pdf (hereafter TSRGD) – this 450 page beast sets out how signs and markings can and should be set out on British roads; it’s the bible of what is permitted.

The portion on ‘Give Way’ markings is found in Section 4, ‘Road Markings’, under Regulation 25. That regulation is online here. The subsection of this regulation we are concerned with is 25 (6), which states that

Where the transverse lines are placed in advance of a length of the carriageway of the road where a cycle track crosses the road along a route parallel to the transverse lines, then the requirement shall be that no vehicle shall proceed past such one of those lines as is nearer the cycle track, in a manner or at a time likely to endanger any cyclist proceeding along the cycle track or to cause such a cyclist to change speed or course in order to avoid an accident.

This is a lengthy way of saying that give way lines on a side road, placed before a cycle track crossing that side road, apply in the usual fashion – don’t cross that line if in doing so you make the cyclist change speed or course.  This subsection 25 (6) s a new addition to the TSRGD; it is not present in the Regulation 25 of the 1994 TSRGD, a document which seemingly has nothing to say about cycle track priority, or cycle tracks at all.

So cycle tracks can have priority over side roads, it seems.

It’s not quite that simple though; we also have to refer to the ‘General Directions’ section of the document, about how precisely to apply the signs. Regulation 34 (2) is found in the section covering ‘Signs to be placed only at specified sites or for specified purposes’, and states that

The marking shown in diagram 1003 may only be placed on the carriageway of a road in circumstances such that regulation 25(6) (transverse lines placed in advance of a cycle track crossing a road) applies, if the length of the road which is crossed by a cycle track consists of a road hump extending across the full width of the carriageway and constructed pursuant to—

(a) section 90A of the Highways Act 1980(a) and in accordance with the Highways (Road Humps) Regulations 1999(b); or

(b) section 36 of the Roads (Scotland) Act 1984(c)and in accordance with the Road Humps (Scotland) Regulations 1998(d).

This is, again, a lengthy way of saying that a cycle track having priority across a side road using give way markings must consist of a road hump; i.e. the cycle track must run along that hump. (‘Diagram 1003’ is simply the familiar double-dashed ‘Give Way’ line painted on a road). That hump must be constructed according to condition (a) if you’re in England, namely

So in plain English, we can have cycle tracks, with priority, across side road junctions, marked with give way markings, provided they run across a hump.

What else is needed? A useful document is the Traffic Signs Manual, which explains, as simply as possible, how highway engineers should implement the conditions found in the TSRGD. Chapter 5 (pdf) deals with Road Markings, and has this to say about the implementation of the above.


3.25  Regulation 25(6) [from TSRGD] enables the marking to diagram 1003 to be used to give priority to a cycle track crossing a road. The length of road crossed by the cycle track must consist of a road hump, which should be of the flat-topped type. The hump must extend across the full width of the carriageway, in accordance with direction 34(2). The marking to diagram 1023 should also be provided, together with a longitudinal warning line to diagram 1004 on each approach. The hump must be marked with diagram 1062 (see para 21.9). The Give Way marking should be placed on the carriageway of the road, not on any part of the hump.

Again, 1003 is the double-dashed line. The hump must be flat, and extend across the road (obviously enough). 1023 is the familiar triangular ‘Give Way’ marking. A ‘longitudinal warning line to diagram 1004’ is simply a dashed central dividing line of the kind usually found between carriageways. Finally, diagram 1062 are the solid warning triangles painted on the upslope of the hump itself.

The picture below (courtesy of Google Streetview) seems to me to be a textbook example of how this has been done, literally by the book. It’s the start of the Bath end of the Bath-Bristol Railway Path.

We have the give way markings either side of the track (1003), together with the triangles (1023). The continuous flat, hump across the road. The (more than slightly faded) solid triangles on the hump (1062). And finally the ‘longitudinal warning lines’ (a dashed central divider to you and me) either side of the crossing (1004).

I have been informed by Paul James, while discussing this issue of cycle track priority, that Transport for London are of the opinion that DfT regulations mean a second set of give way markings cannot be placed within 5m of a junction.

I certainly have not been able to find such a clause within the discussion of cycle track priority. The best I can do is a section contained within LTN 2/08, Cycle Infrastructure Design (pdf), which has this to say

On a bent­out crossing, the cycle track approaches are deflected away from the main carriageway to create a gap of one or two car­ lengths between the main road and the crossing. A gap of about 5 metres is required to accommodate one car. The arrangement allows drivers turning into the side road extra time to notice the crossing and provides somewhere for them to stop for crossing cyclists without obstructing traffic on the main road. It also allows a vehicle waiting to exit the side road to do so without blocking the crossing point. [my emphasis]

Note, however, that this is only a statement of the length required to accommodate a car, not a formal stipulation. In addition, it only applies to the design of ‘bent out’ crossings, not crossings in general – indeed the previous section, 10.3.6, says

Where the crossing is placed on a road hump, it may be better if  it is “bent out”

‘May be better’. Crossings don’t need to be bent out. And even if they are, the distance mentioned is only suggestive, not a stipulation. I think TfL might be getting the wrong end of the stick, if this is the source of their ‘5 metre’ claim.

There are crossings in the UK which quite obviously do not meet this requirement.

(picture courtesy of David Arditti)

The 5 metre suggestion is actually quite sensible; indeed it accords with best Dutch practice, which sets the cycle track crossing back from the main road, allowing drivers to come off the main road and pause before giving way to the cycle track. It also means vehicles queueing to get on to the main road needn’t block the cycle track while queuing. An example in Assen –

But in any case a UK cycle track need not be ‘bent out’. It can run in a straight line, remaining parallel to the road, like this example, again from Assen –

If there is not much queuing traffic coming out of the side road, this is a sensible design, provided that the geometry of the junction is sufficiently tight to force slow entry speeds.

Again we have UK examples of this type of design –

(again courtesy of David Arditti)

The design is obviously inferior, particularly in the width of the track – but we can do this in this country. David Arditti has documented plenty of examples, mostly of this form. This track does not meet the strict requirements of TSRGD, because it does not have give way markings on entry (only the triangle markings on the hump), but I see no reason why  this could not have been done properly.

The new cycle tracks on Old Shoreham Road have even fewer markings –

(picture courtesy of Jim Davis)

I think this is slightly underdone – the painting of triangle markings on the hump at the right hand side could be a useful addition here, along with a continuous line along the hump, to give an extra implication of priority.

So the big question is – given both the provision in law for cycle track priority, and the evidence of plenty of cycle tracks with priority in the UK that accord with law to varying degrees of strictness – does this statement from the CTC

Traffic regulations in the UK only permit cyclists to have priority over vehicles entering or leaving side turning in very restricted circumstances

reflect reality?


I am grateful to David Arditti for pointing out in the comments where TfL’s aforementioned 5 metre ‘requirement’ originates – the Appendix of Transport for London’s London Cycling Design StandardsIn Appendix C we find these two diagrams –

The first shows a cycle track without priority, continuing (more or less) without set back; the latter does have priority, being set back ‘5m or more’.

It is worth noting, as David points out, that these are only suggestions, not legal requirements. I have no idea where John Lee’s belief that the law is against cycle tracks having priority unless they are set back 5 metres originates.

This entry was posted in CTC, Department for Transport, Infrastructure, London, Priority, Road safety, Subjective safety, The Netherlands. Bookmark the permalink.

21 Responses to Priority of cycle tracks across side roads

  1. On the subject of priority for crossing traffic doesn’t the highway code already state that motorists have to yield to pedestrians already crossing a road when they are turning onto it? Of course even with that in place you still take a gamble. We’ve recently had a new zebra crossing installed in Croydon on what was already a fairly blind turn. The approach to it has been covered in an anti-skid covering and the crossing itself isn’t even 2m from the turn, I’ve already seen 1 driver come to a skidding halt (an they where coming from the much safer giveway to the left!) to avoid people crossing….

  2. Richard Mann says:

    Cyclists certainly take priority here: (the “obligatory” hump markings are missing). There are several like this along Marston Road. Whether the cyclist has formal priority over the turning vehicle is debatable, but almost all cyclists “read” those crossings as giving them priority, and almost all motorists accept that they have priority (the hump does most of the work).

    The barriers on Parks Road are on University-owned land that isn’t technically highway, so it was a pretty poor example for CTC to pick. There is a genuine visibility issue there, though it could have been dealt with less obstructively. Since the outer barrier has been removed, it’s no longer much of an obstruction (and in any event, most cyclists use the cycle lanes on the road).

  3. davidhembrow says:

    I also wondered about the insistence by the CTC that it couldn’t be done in the UK, because I’ve seen examples in Britain too.

    For example, these in Cambridge: 1, 2, 3, 4.

    • davidhembrow says:

      Oh, please scrub that comment. I just realised I was looking at the Dutch style “shark’s teeth” and not the actual British give-way symbols. I’d mis-remembered how it is.

    • Those are odd ones, because there’s a give way marking in the road, as well as on the cycle track. Both parties apparently have to give way to each other, which is pretty poor design. It wouldn’t take much to adjust them to look like David’s example in my post.

      • Rad Wagon says:

        The “idea” is that no-one has priority, so everyone should stop/slow. That way everyone is more considerate. In reality you get people driving pointing out the cyclepath give way lines to people cycling saying “why didn’t you stop?” whilst pulling out to improve their (car-based) restricted view. So it reverts to the law of the biggest bullying their way forward.

  4. Cyclestrian says:

    “The picture below (courtesy of Google Streetview) seems to me to be a textbook example of how this has been done, literally by the book. It’s the start of the Bath end of the Bath-Bristol Railway Path.”

    I can see how this could be improved. There are two white lines crossing the cycle path that detract from the continuity and priority of the cycle route.

    Really we should look at such junctions as a crossroads of a major road (the cycle path) and a minor road (vehicular) and mark it up accordingly in the conventional way. A dashed median line on the cyclepath, even if it doesn’t aid passing cyclists, would be another visual hint to drivers that they do not have priority. Give way signs might be overkill though!

  5. OK. I think one thing you have missed here is the London Cycle Design Standards, which, though not legal requirements, may have influenced the CTC and may influence designers across the country. These, in the Appendix C diagrams pp199–200 suggest that if the 5m setback is not present, cyclists should be made to give way when crossing a two-way side road.

    This is a reference to the most recent published edition, 2005, and in fact there is an improvement here from the previous edition (a reference to which I do not have to hand, but you might find it), where it was indicated that the only circumstance in which a track could have priority was if the side-road was one-way going into the main road. This influenced a lot of traffic engineers in London to refuse to construct cycle tracks with priority at junctions – I had direct experience of this.

    The diagrams and recommendations in both editions of the LCDS were drawn up by John Lee of TfL. I have discussed this issue with him. He is firmly of the opinion that the law is against cycle tracks generally having priority over minor roads unless set back 5m, and he believes the Camden cycle tracks (inc. Royal College Street) are legally challengeable. I think he is wrong and don’t know where he has got this idea from. Indeed the Camden tracks have been operating for 12 years without being legally challenged, and the engineers who originally put them in did their homework and did not believe there was anything illegal about them. However, the LCDS may be where a lot of this problem is coming from.

    • Thanks for this David – I had looked at the TfL Cycling Design Standards, but only Chapter 6 (pdf), ‘Signs and Markings’. This chapter does mention priority of cycle tracks over side roads, subject to the markings and hump stipulated in TSRGD. There is, however, no mention of the 5 metre ‘requirement’, which only features in the Appendix C you mention.

      Like you I have no idea where this opinion that priority should depend on the amount of setback has come from. It might be worth asking John Lee for the basis of his opinion.

  6. I believe that pedestrians in Holland and Germany can continue across a side road at a junction without waiting. That is, anyone going straight ahead on the main road, whether on foot, cycling or driving has the priority over people turning off. This continuity of priority means that drivers, and other road users, expect to give way when turning off in, whereas in the UK pedestrians should wait on the pavement, giving way to turning traffic. It would be worthwhile considering whether we should campaign for such a change here too – it will facilitate pedestrian journeys and long-term make for safer segregated routes. However it shouldn’t be allowed to be an excuse to avoid implementing cycle paths that continue over junctions in the meantime!

  7. Pedestrians crossing a side road do actually have priority over turning traffic in the UK – see – though this only applies once the pedestrian has started to cross, so maybe it’s different in other countries. The rule as it stands seems reasonable to me – if it was actually observed.

    The problem is that the real rules of the UK roads aren’t found in the Highway Code, but in the rulebook of the jungle: might wins. A growling motor wins every time, in fact most people assume that motor vehicles have priority at all times. In fact, I’ve asked many people if they knew about this pedestrian priority rule, and not one knew it existed!

    Perhaps a government ad campaign to remind drivers of their responsibilities is in order?

  8. PaulM says:

    Your first picture shows the eastern facade of Keble College (“this is a collage, but think of it as a Fair Isle sweater”) which brings back memories of my own 4 idyllic years there, 35 years ago, as a chemistry undergraduate.

    I used to cross that road from the porter’s lodge, visible just to the right of the right-hand tree in the picture, to the science park directly opposite, notably to the University Science Museum where many of my lectures were delivered. My memory of the road, and indeed of Oxford as a whole, was that traffic was pretty sparse, apart from the main drags such as St Giles, and Banbury and Woodstock Roads, and even there it wasn’t particularly intimidating.

    Like almost all students, and most dons, I got around on a bicycle, one of the many thousands around the city, all marked with a compulsory registration mark to identify the owner if he should lose it or leave it chained to the wrong railings. Mine was KC801. Even then car parking was just about impossible, in a landscape of 16-17th century buildings all cheek by jowl, all no doubt legally protected from unsympathetic “development”, such as car parking.

    I’m trying to imagine what can have changed so radically. I am sure that Parks Road still takes you into a warren of Cotswold stone ancient buildings, separated by narrow alleys and with virtually nil car parking anywhere. It doesn’t take you to a central destination, and it doesn’t function well as a through route either. In fact, it should have all the qualities to be a perfect example of where segregated infrastructure is NOT needed. So how can Parks Rd have become so hostile that an off-road cycle path would be needed?

    • Richard Mann says:

      In 2000, the northern half of Parks Road became part of the only east-west through route when the High was closed 0730-1830 every day. See (Keble Road was also blocked at the Banbury Road junction at the same time).

      So it’s pretty busy. It is 20mph with cycle lanes along most of its length, and the traffic is pretty slow-moving, so it’s OK for even fairly timid adult cyclists. The route on the pavement was done at the same time to link Norham Gardens to the southern half of Parks Road (part of NCN51). The junction of Parks Road with South Parks Road is a bit of a mess, but the shear number of pedestrians at certain times of day makes it difficult to improve.

  9. David says:

    What hope has cycling got when then this gets changed to this at the last minute because “Through these discussions it was concluded that the proposal to give cyclists priority over traffic did carry a high risk of road safety problems with particular concern over drivers turning left off Boroughbridge Road not having sufficient time and space to give way for cyclists crossing Ouseacres.”
    Really, what, this is just like a dutch design with a car lenth of space after the turn before the crossing – a good design, just like the TFL design.

  10. Pingback: Response to Portsmouth Road Consultation | Stuff Rich Writes

  11. Eric D says:

    It seems that cyclists often won’t get priority because motorists will kill them !
    There’s a lot of unspoken ‘Might is Right’ going on here.

    Incidentally, I came here to compare a recent incident in Reading, involving a car turning right across a physically-segregated Bike&Bus Lane

    I guess the driver would have seen a bus, but don’t know that the driver expected the cyclist to give way.

    Google StreetView shows the junction being remodelled May-Jul 2014 : a couple of feet narrower and flowerbeds ! It would be better half that width.

    Ironically the comments include
    “we need to find a way to separate cycles from main vehicular routes.. just as we use pavements to separate pedestrians from vehicular routes.”
    It _was_ a segregated lane – the only answer is overbridges or subways.

    • Eric D says:

      So the system rewards people with the greater potential to cause damage.
      Perhaps cyclists need to think up ways of causing more damage ?
      Joke !)

  12. UK laws are manmade laws, they are not the laws of physics. You can equally not be allowed to go faster than the speed of light nor can you go around building cycle priority crossings without a raised table like this, but one is possible to violate, and thus is changeable, and the other isn’t. When our laws don’t make sense, they must go.

  13. Pingback: Kingston Go-Cycle summer proposals, Part Two – Stuff Rich Writes

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s