Covid Stats

A long silence will be breached as a result of finding a Guardian article this morning about how well the European Union’s vaccine programme is going:


The basis for the article – that lots of people on the Continent (and in independent Ireland) have been jabbed is Good, but the tone is very much of the “Brexit was rubbish, because in the end we would have been better off had the EU done our vaccines” school. Now I was in favour of Brexit as little as the next rabid pro-European, but let’s have a reasonable basis for our arguments and a reasonable evaluation of what isn’t working about a given method of work.

So, let’s take a comparison piece from Our World in Data, from which the Guardian claims to have got its data.

In General

We did it,” said Ursula von der Leyen in her annual state of the union address last week. With more than 70% of its adult population now fully vaccinated against the coronavirus, Europe is, “against all critics, among the world leaders”.

Jolly good for them. The UK figure is 81.7% of over 16s.

Progress in specific countries

Nine EU countries, including Portugal, Spain, Ireland, France, Belgium and Italy, have now administered one or both doses of a Covid-19 vaccine to a larger share of their populations than the UK, with a further five having overtaken the US.

We can all agree that the silliest decision ever made by a United States of America Government official was to go for self-government; the country would be much better off with direct rule from London. The Covid response has highlighted this, as has the whole business in Afghanistan. And have you tried using American public ground transport? So the fact that the EU is doing better than the USA is no big surprise.

Four points on the statistics. One is the not-wholly-relevant one of averages between individual areas under a common vaccine board – some places will be above and some will be below. If you take any chunk of countries at a given moment and assess them on something then some will be doing better than others, and to an extent if they are all reasonably well-run then the precise order will be a bit random. If you took this statistic for the UK in mid-May then you’d find that England, Scotland and Northern Ireland had vaccinated about two-thirds of their populations and Wales had vaccinated three-quarters. This does not in itself mean that Wales is the best administered area.

A numerical point, partly explaining the discrepancy in Wales, is that Wales and Portugal have the bonus in this regard of relatively small populations compared to England and Germany, so an efficient vaccine programme making reasonable use of a sensible number of convenient facilities can have a big impact very fast. (But also the Welsh pulled their fingers out and did a good job at it. One suspects they were partly stung by the January stats where they were badly trailing the English and suggestions of rationing the Welsh share of the vaccine were getting replies from England of “if you’ve got some spare, can we have it back please?”.) An interesting question is “how did Portugal take so long to vaccinate ten million people?”. Back in February the UK had given one jab to marginally more people than the whole EU. Speaking of actual numbers, instead of percentages of population, Bolivia has vaccinated more people than Ireland – 4.2 million as against 3.7 million.

The accuracy-related point is that according to World in Data (at the time of scribbling) Italy’s percentage of total population jabbed is only a smidgen above the UK. Ireland, Spain, Belgium and Portugal are well ahead; Spain is the only one of these with a population comparable to England. France is a little below, having recently overhauled Germany. The EU average is just below that. The US has slowed up, albeit having jabbed over 180 million people, and percentage-wise is trailing other former British colonies like Malaysia – though is much better off than Australia. (This incidentally is why there is some concern in the US about easing other Covid restrictions. The vaccine situation is not directly comparable.)

The healthcare point is that the success is getting a vaccine into arms pronto. One does not win a race by turning up at the finishing line when all the other athletes are towelling down and racing past them to do a further lap. If you ranked this clump of countries for six weeks in June and July then the UK would be top by some margin, and on the one-jab percentage-of-population rating the UK was top from December 13th 2020 until July 27th 2021, when it was overtaken by Belgium.


One major point of the EU is to promote more economic and social equality across the continent, and help poorer nations catch up while smaller nations can punch their weight.

Success is far from even across the bloc: poorer states, such as Romania and Bulgaria, are struggling, with 27% and 16% of their populations vaccinated.

That’s not happening. The richer Western nations are still performing better than the poorer Eastern ones even under an EU-managed rollout. Poland’s even behind the USA… (The Bulgaria statistic, down below Covid-ridden Peru and consistently just better than the Lebanon, is really rather depressing for the European project.)

Pushing take-up

National vaccination drives have also been boosted, sometimes dramatically, in more than a dozen EU countries by domestic health passes, now needed to access anything from museums and gyms to cafes, shopping centres and trains.

In other words, in the EU you have to be jabbed or you can’t take a train to go shopping and have a cup of tea. In the UK this does not apply. How you enforce such a thing on people travelling between unstaffed halts or wandering into open publicly-accessible buildings is another matter, which is one reason why in this country we haven’t bothered. The fact that we haven’t bothered means that there isn’t a major direct boost to you personally to getting vaccinated, excepting being able to tell NHS Test-and-Trace callers to cordially consider calling someone else if contacted with a request to isolate. There is merely a boost to the entire country from reduced healthcare requirements and potentially to you at an undetermined later date from not falling ill. The massive easing of restrictions in England and the optimism of no more lockdowns is the bonus of most people getting jabbed. There is a certain fallback to national stereotypes within the UK – the English are optimistic and opposed to any limitations on life, while the Scots are dour and fatalistic.

Several EU nations evidently have ensured that you cannot live a normal life without getting vaccinated:

Austria, Belgium, Cyprus, Denmark, France, Germany, Greece, Hungary, Ireland, Italy, Luxembourg, the Netherlands, Portugal and Slovenia are among the EU countries to have introduced some form of Covid pass.

and thereby effectively given their people no choice but to do so.

If their primary aim was to increase the number of people getting vaccinated, some schemes – certainly in more vaccine-hesitant countries, such as France – do appear to have been successful.

This is on one level quite appropriate, if the aim is vaccination at all costs short of breaking into houses and vaccinating everyone inside, but the English (perhaps Anglo-Saxon?) approach tends to involve more caution about slippery slopes to future restrictions. Witness the unwillingness, even at the height of New Labour’s War on Terror and its enthused paranoia about whether Mrs Next-Door was building a home-made bomb, to carry confirmation that you are who you say you are and that you have a right to be in the country (or just to walk through a metal detector to enter a service station, though the last time I went to the National Exhibition Centre in Birmingham they were walking every visitor through such a detector and doing bag searches). The attitude being taken by the central Government in the UK would almost certainly be different if we had a Labour Government – see Wales – and if we correlate mask-wearing in public with the party make-up of Parliament then we may see reasons for a) Brexit and b) the Conservatives’ irritating tendency towards sufficient majorities and long-term polling leads.

Anyway, the main point is that you can achieve many things with a metaphorical cattle-prod. This one appears to be independent of the EU.

Youth vaccination

Further to increasing vaccination rates, many member states have – in contrast to the UK – already administered first jabs to as many as 80% of 12 to 17-year-olds.

Let us undermine all of the above blogpost, and its inferences about how great the UK vaccination programme is, by pointing out that the question for any vaccination programme is not “will this signal my eternal virtue?” nor “will this make my national statistics look better?” but “does it have any long-term health benefits?” Considering that the vaccine is perfectly capable of killing people if not properly managed then its use should be assessed sensibly.

Plus “Would those vaccines be better off in the arms of sub-Saharan Africans?” All the arguments at the beginning of the year were that they would. Of course, we must acknowledge:

Moreover, the Commission president said, the EU had exported half its vaccines: “We delivered more than 700 million doses to the European people, and we delivered more than 700 million doses to the rest of the world. We are the only region to achieve that.”

This is a major failing by Boris, unless he’s been doing this without telling anyone. (Assuming of course that the EU exported vaccines are actually donations and not that the major vaccine-producing factories are all in the EU, which has then been busy selling them to countries like the UK. That would be a bit like crowing that Oxford is the only net exporter of Mini cars.)

Government skills

With no relevant procurement experience, however, the commission approached talks with manufacturers like trade negotiations, prioritising price over delivery deadlines. Some contracts were signed months later than those of rival buyers.

We were always led to believe that the UK Government couldn’t procure beer in a brewery, so the idea that the EU was worse is quite remarkable.


Rather than being a marathon not a sprint, as EU officials like to say, vaccination campaigns should probably be both. But after a slow start that will certainly have cost lives, the EU’s collective approach may, finally, be paying off.

No. The countries that were always going to get in first got in first. The countries that left to themselves would struggle to vaccinate half their population have not vaccinated half their population. Apart from eight weeks between January 23rd and March 18th this year, the UK case fatality rate moving average has been below the EU’s. At the start of July a Covid-positive person was over ten times more likely to die of Covid in the EU than the UK – this improved to percentage errors, but has widened back to three times more likely. The case rate in the UK is higher, so the overall death rate is above the EU’s, but this is because the most stringent UK restrictions are amongst the slackest around. Not because of a vaccine programme.

Instead we must remember that Ursula von der Leyen is a politician making a political speech, and that the non-leading nations include several places like Egypt that have barely started vaccinating.

Which is a pity, because it’s nice to blame stuff on Boris.

(It’s also nice to have slack restrictions and freedom to go to places without being shouted at, of course, and I therefore particularly favour the UK vaccine programme because it got us to a reasonably safe place to ease restrictions before almost anywhere else.)

Exciting News

Health Secretary Matt Hancock finally admits that his Covid rules are too stringent and not humanly possible to follow.

Fifteen months too late, of course, and fourteen months after his lead advisor realised the same thing. (There seems to be a trend here.)

His resignation would be nice, but more value would be him ditching the rules earlier in light of this realisation. (Come on, Mr Hancock. You know it makes sense. If social distancing goes then you can spend more time with her, subject to the usual moral and legal objections raised by your wife and her husband.)

In such an event I think it would be possible to correlate this little Gilbertian number to it somehow, though some slight rewording to possibly involve Hancock, his aid and Boris would be useful:

(Why someone decided to shift one of Ruddigore‘s more notable numbers into Pirates of Penzance is another matter entirely.)


In the midst of all the current trouble and strife and horror and death and suffering I cannot quite explain why it is that the thing which has knocked me this year (even exceeding Prince Philip’s demise) was going onto my pet railway news website ( a couple of months ago and reading:

91132 (SAXL) has been released for disposal, being despatched by road transfer, noted on the M1 near Chesterfield this afternoon, from Wabtec at Doncaster to Sims Metals at Beeston.

She has since been cut up.

91132, for those not familiar with this creature, was one of the thirty-one Class 91 electric locomotives built in the late 1980s for use on the newly-electrified East Coast Mainline on services linking London with Leeds, York, Newcastle, Edinburgh and Glasgow. The project was born out of British Rail’s need for more and better electric locomotives in view of the cancellation of the Advanced Passenger Train programme, the authorisation of electrification of the mainlines from London to Edinburgh and London to Norwich and the tendency of the remaining examples of the fleets built at the end of the 1950s to provide free and frequent training opportunities to the Cumbrian fire brigades.

There were already seven types of four-axle electric locomotive in existence, designated Classes 81 to 87 – mostly 100mph locos (on paper, subject to ride quality) except the 87s and a handful of 86s that were signed off for 110. A modular freight loco design was proposed to be Class 88, but this was cancelled before it got very far along the drawing board. Brush Traction of Loughborough were employed to design a new 125mph mixed-traffic electric loco on six axles, to be the Class 89, but while the prototype for this was being built British Rail had a re-think and changed tack again. The new tack involved a fleet of improved Class 87s for general 110mph mixed traffic work, which became Class 90, and a smaller pool of locos for running at 140mph and with some basis in the APT, which became Class 91. Brush were duly told that BR was not really interested in the 89, scaled back development work and were hit for late delivery charges.

The official brand name for the 91s was the Electra, which was abandoned before the doyenne was completed. Thirty-one were built, originally numbered 91001 to 91031 and given a series of serious but traditional names like Terence Cuneo, Reverend W Awdry and Queen Elizabeth II (as opposed to the serious and commercially-driven names applied to the Class 90s, like Financial Times). The whole train of locomotive, coaching stock and driving trailer with luggage van was called the Intercity 225. Features carried on from the Advanced Passenger Train, after a highly-advisable rebranding exercise, included theoretical tilting capability (but the kit to actually tilt was never fitted).

They were the apex of British locomotive design – 140mph locomotives capable of continuously producing 6,480hp. And they looked rather impressive.

They have never operated in my necks of the woods – they were an exclusive East Coast Mainline thing, like Sir Nigel Gresley’s A4s and the later diesel Deltics. The East Coast Mainline always feels at its best when it has its own small fleet of locomotives to itself.

Once in traffic the fleet scored various achievements. The main one from an operating perspective is the consistent hammering of the UK’s fastest scheduled long-distance domestic services, posting average speeds of over 100mph for the run between London and York. Alongside the fleet also holds the roughly-joint record of the fastest train over the classic UK rail network, courtesy of 91010’s speed run in 1989. It raced down Stoke Bank, actually propelling the train from the rear (the driver was sat in the driving trailer at the south end of the set) at about 36mph more than the Mallard did when setting the world speed record for steam traction on the same length of railway 51 years earlier. (The other holder is a prototype Advanced Passenger Train set, which did the same speed down Beattock bank between Glasgow and Carlisle.)

After ten years of being thrashed up and down the East Coast Mainline the locos were put through a heavy overhaul following the turn of the millennium. This was deemed large-scale enough to warrant renumbering them to indicate modified locos, so they became 91101-91122 and 91124 to 91132. The mysterious renumbering of the twenty-third 91 onto the end of the number series as 91132 was simply because she had been involved in both fatal accidents involving a Class 91, about four months apart. Between them the accidents at Hatfield and Great Heck wrote off the hopes and dreams of fourteen people, vehicles from three trains and Railtrack plc – but failed to scratch 91023’s paintwork.

Now deemed better than new, the fleet continued being thrashed. A day of racing up and down the East Coast Mainline on 4½-hour schedules between London and Edinburgh – not much shy of 400 miles – will happily notch up 1,200 miles in 15 hours, or an average of 80mph over the whole day. (This means, if you are the sort of person who takes statistical averages seriously, that this brace of locos sat at Kings Cross are averaging 80mph.)

91132 kept a low profile during the following years. Other 91s gained new names and fancy liveries as the East Coast operation went through various companies. The Government operator “East Coast” was good at fancy liveries and seemed to have a campaign to get each 91 out of their plain silver into something more fetching. The results varied. 91107 briefly reverted to being 91007 – at least on the cabside number – to take up the name Skyfall and a white advertising livery for the 2012 Bond film. (She is still Skyfall, but now in standard livery.) I quite liked 91125 in blue, even if I wasn’t that gone on the idea of a 250m high-speed Sky 1 HD billboard. Rather regrettable that the whole fleet didn’t end up in this cheery blue tone, which wasn’t too far off the shade of the old LNER’s A4 locos.

Equally, one can appreciate the thoughtful gesture behind 91111’s First World War livery without thinking that it necessarily wholly suits the locomotive’s 1980s lines:

(She is still wearing this livery, but now with a conventional yellow block between cab windows and light clusters.)

The doyenne went purple and gained the name Flying Scotsman to go with the relaunch of the historic “named train service” (albeit southbound only and at a silly hour of the morning); 91110 went black with Battle of Britain branding; 91114 was given a little viaduct silhouette to go with the name Durham Cathedral.

Virgin East Coast did most of the fleet up in scarlet with a white stripe and flash, in which they perhaps looked better than they had ever done before (though lacked the sleek whole-train coherence of GNER dark blue with matching doors and red stripe – one can see why they did that, even if it made the doors impossible to find). A smaller black flash around the cab side windows gave an air of a superhero mask. It was a very clean livery, suiting the largely plain and unadorned bodyshell. The same colour scheme applied to the HST powercars never quite seemed to go with their bodyside grill or cab front skiting. There was a small variation on 91112, which had a grubby white skirt instead of a grubby black one. 91110 and 111 retain their fancy liveries, 101 has a variation on the Scotsman colours and 114 now has some nice tracery to go with its name.

Virgin red…
… and GNER blue. It had a certain poise when clean and seen from the right angle, but side-on makes the loco look like a blue brick. This is emphasised by the square cab at the back, butting onto the coaches. Ironically the operator that made up a few hundred cast heraldic plates for the coach bodysides refused to splash out on proper nameplates for the locos – scraps of vinyl at a non-obvious level were used instead.

Eventually Virgin Trains gave 132 some brief status by slapping vinyls on her side announcing that they had signed the Employer Pledge to end mental health discrimination. One could say that she was the most appropriate loco for the purpose.

By the time Virgin came on the scene it had been decided to buy enough new Intercity Express Trains to replace the 91s – the new units are better accelerators and a consistent fleet makes it easier to maintain reliability. Availability of class 91s for service was not great anyway, and in 2019 two class 90 electric locomotives were being hired in to work trains that should have been handled by 91s. It gave the fleet an air of being on their way out, but Virgin pledged to retain a small batch on shortened pocket-rocket sets of coaches to run additional services. Virgin has now been replaced with LNER, which is looking to buy a new fleet of something to replace this small pool. The locos should therefore be gone by 2025. They aren’t in traffic at the moment. (But as the Azumas, as Virgin dubbed their new trains, are having small technical hitches the 91s will be back sooner than expected.)

91112 at Kings Cross, with white skirt, the remains of a bird in the horn grill and two hired-in 90s for company. The two designs are of a similar age, but the 90 is an explicitly mixed-traffic machine which was intended to do most of its passenger work on the then slightly slower West Coast Mainline. In view of this and various other minor considerations a 90 has a less raked-back front end than a 91. (But it has a raked-back rear too, whereas a 91 is really an electric powercar so its back is almost flat.)

With the end looming 91132 now had the odd position of being the loco which, if believed jinxed, was the obvious one to get rid of first. But equally the first member of a modern traction fleet out of traffic is expected to surrender parts to keep its classmates going. And you don’t want to start sticking bits out of a jinked loco in non-jinxed locos. (If that doesn’t make any sense – well, you probably don’t really buy jinxes either.)

In the event classmate 91108 was first out of traffic, on the basis that she was due an overhaul that wasn’t worth doing, and has duly been relieved of an array of parts, the most obvious of which is her front end.

91108, in the days when she had a front end, dozing at Edinburgh awaiting her next working to London.

91132, meanwhile, seemed to be in excellent working order. She was happily thrashing away at the end of 2019, now ratcheting up the miles on London to Leeds duties. There almost seemed to be an air floating around her that she had cleared her former humiliations and now merely had just enough notoriety to be a preservation candidate. Then she tiptoed out of traffic during the 2020 lockdowns and went into store. The end for her few remaining siblings was quite sudden; the remaining members of the fleet came out of traffic after Covid Lockdown 3 was implemented. There is a promise to bring them back this summer, which will hopefully mean that unlike some recently-withdrawn train designs this fleet will get a proper send-off.

Such a thing makes it a bit hard to believe that they are really going to go, even if two of them are now in bits all over a Doncaster yard.

Then suddenly 132 was on a low-loader and off to a scrapyard. Apparently a use has been found for an intact loco which nobody wants – to take it apart in a scrapyard and work out how much the component parts of an intact 91 are worth allowing for the cost of extracting them.

It seems from my photos of 91s that I never actually met her properly. Unless I’m racing off to Scotland, which is not exactly a weekly (or even annual) event, I tend not to hang around Class 91 stamping grounds. I must have seen her through a window at some point, glimpsed far away, the mysterious thirty-second member of a fleet of thirty-one.

I’m therefore not quite sure why her imminent demise threw me. This is not like the specific class 153 DMU which might be unexciting and badly-designed but aboard which I had many happy jaunts around Cornwall during my degree and with which I associate several happy memories (361, in case you’re interested – I forget when I noticed that I seemed to see a lot of it – it had a slightly different “face” to the rest of the First Great Western 153s so was a tiny bit distinctive). Maybe it’s just that the 91s were the great pinnacle of excellence and the Intercity flagships, so it seems implausible that anyone would consider scrapping one. I believe the Deltic watchers felt the same way when No. 55020 Nimbus was suddenly reduced to a very impressive nose sat in a pile of mud.

Thus I might feel like buying a 153 if I had a use for one, but there is a temptation regardless of actual practicality to murmur “blow the mortgage” and ask the owner of the 91s how much one would set me back for a large and nicely-designed garden ornament. (Happily for the neighbours, my garden is too small.)

But like the Deltics the 91s are a small, non-standard fleet of very powerful and expensive locomotives built with the express intention of being thrashed on fairly specific duties. (Actually the 91s were always purportedly intended to do other stuff, but they never did and one now wonders why they were ever expected to do so except to make the business case look better.)

That work is now gone. They could be pushed towards other things, but really the other things don’t warrant the expense of keeping them going.

In the end 91132 is not going to be all that hard done by. She was the first to go, but it will only be by a margin of a month or so. 91102, 113, 126 and 129 (all of which I have met; 102 heads this article) will be off to Newport for mangling tomorrow (Monday 10th).

So one will have to make do with watching the tribute video. (91132, in her days as 91023, makes a short appearance in the full programme at Newcastle.)

The only tiny problem, eventually revealed by the producer of this video, is that they filmed it from a HST, after two attempts from a 91, and dubbed the 91’s soundtrack over it.

Apparently 91s don’t ride very well, and you need a good ride for a smooth video.

But then this is a problem with all British high-speed electric locos.

Another reason why we stopped using them perhaps.


The Campaign for the Protection of Rural England has a campaign on for more and better bus links:

It’s the sort of thing that I would like to expand on more, but I seem to have overworked my wrist carrying a load of stuff back from the local DIY store and can barely type this evening. So a nice short post.

Of course, two questions. One is where these more and better bus links go – one trusts that they start at a useful sort of place and then they scurry off in the direction of a place where people want to be, instead of scouting around looking for more places were people are but might not want to be at this moment. (If there happen to be more places to serve en route then so much the better, but let’s not have buses doing daft things like Hay-on-Wye to Hereford via Kington and Presteigne – three sides of a very large square instead of straight down the Wye valley in case the geography is unfamiliar – in a rural equivalent of some urban bus routes. If Hay needs an hourly bus to Hereford and an hourly bus to Kington then run two buses, possibly close together to nicely connect with the hourly bus from Brecon.)

The other is whether the Government will try to use these buses as an excuse for not opening much-needed rail links instead of using the buses for what buses do well – trotting around small communities and suburban areas, and linking scanty ribbon communities to local hubs – and leaving the railways to cover the longer-distance runs between urban areas.

(Actually, the big question is whether the Government will ignore the petition. So sign it and pass it on. As may have been realised over the last twelve months, rural public transport is also jolly useful for getting out of town if you happen to already live in the town centre but would like to be somewhere else today.)

And now, Unhelpful Expert News

Our top bod in some branch of Public Health England has come out and Said Things.

Specifically, that we may have social distancing and mask restrictions “for several years” until Covid eases in other countries.

This at a time when half the adult population has had a Covid jab, many of the people at very serious risk have had two jabs, and a reasonably average small city of a couple of hundred thousand people will have maybe 150 positive cases.

Now, I may be the only one, but I am keeping to restrictions, wearing a mask and not visiting my relatives on the basis that the current rules will not be around for very much longer (a substantial easing week tomorrow, all gone by July).

If it is going to be around:

  1. for years;
  2. based on the activities of 7-8 billion other people whose behaviour I cannot influence;
  3. based on the views of other Governments, many of which deny that there is a problem;

then could someone kindly explain to me why I am bothering?

By the time this person thinks I am going to be allowed to visit 95-year-old relatives on a legal basis again they will most likely be dead anyway.

It’s not like we’re allowed to go to these other countries anyway, and we’re all being told not to go on holiday this summer to these other countries that have problems, so who are these people who are bringing back a new strain every half hour?

And why are we wasting all this money on a vaccine that won’t get rid of the core restrictions – that a six-hour train journey to see distant relatives is spent in an uncomfortable suffocating mask, and when you get there you can’t hug them, spend the night with them or sit round a lunch table with them anyway?

It is high time the experts acknowledged an unpalatable truth – that if what they have said and done has worked then by August we won’t be needing to listen to them on a daily basis any more. Being deliberately negative about the long-term social possibilities does not help a lot of us and risks a view of “might as well go back to normal then”. Boris understands this, much as this then means he keeps writing big positive flowery cheques about Christmas and Autumn and last May and so on that he then decides he can’t cash.

Meanwhile, here is a picture of a flower.

(Be a plant. Plants don’t get Covid.)

Just A Minute on Architecture

(This is probably more something for Twitter. But I don’t have a Twitter account. But it still amused me. So you’ll have to see it here.)

I’ve dug out an ancient episode of Just A Minute this evening (one of those where the host and the panellists are all dead and won’t be participating in tomorrow’s recording, which is a risk with 53-year-old shows) and found Clement Freud MP expounding on “The World’s Most Beautiful Building”:

It’s awfully difficult to talk about the world’s most beautiful building, unless you have been all over the world and seen every single building. But I think what I can say is that Number 28 High Street Ewell is not the world’s most beautiful building. I passed it the other day and it was quite singularly ugly. The two-storey building with an old turret…

At this point Clement was interrupted by Derek Nimmo on the basis that “He’s not talking about the world’s most beautiful building, he’s talking about a building that he finds particularly ugly.” This challenge was upheld (“You had established that you were talking about the world’s least beautiful building in your mind”), Clement made some muttered comments about decision-making and Derek began expounding on the Taj Mahal.

While I am vaguely familiar with the Taj Mahal (though I have never seen it in the marble), I have not spent much time in Ewell so have not seen Number 28, High Street.

Happily we have the Internet.

It is apparently a not exactly attractive (though hardly especially ugly) white-painted block shared with a denim store, and now occupied by an estate agent. The fact that it is occupied by an estate agent is a bit distracting because it means the search engine pulls up every house in Ewell being marketed by said estate agent, though with especial reference to those on the High Street.

The estate agent does not, for some reason, advertise themselves as the occupants of a building once slated on an internationally-followed radio show.

However they are, at time of writing, engaged in selling a house on the other side of the crossroads that No. 28 faces onto. Should you wish to have more time to consider Sir Clement’s tastes in architecture, have £475k to spare and not really care for garden space then No. 15 may be of interest.

(But you would have Chris Grayling for your MP. Might be worth knocking another £25k off for that.)

Thinking through rail expansion

There’s a lot of muttering at the moment about various uses for former railway trackbeds and how to provide a more effective transport network post-Covid. On one level is the argument, subscribed to by the Chancellor of the Exchequer, that the New Normal is permanent and nobody will be allowed to use public transport again – thus more roads are necessary. On another level is the argument that we need a safer, cleaner and greener way of living our lives – thus more public transport is necessary. A further level submits that we need to live local and shop local to be best for the environment, so new long-distance transport links are not necessary and we should convert the railways (usually dead, but sometimes alive) into cyclepaths.

And even if we do go for public transport, there are lots of forms of public transport so lots of ways to do it.

Let’s take an example.

The Cirencester Branch

The line to Cirencester was built as part of the Cheltenham & Great Western Union Railway in the 1840s, diverging from the mainline from Swindon at Kemble. In fact Cirencester was Cheltenham’s railhead for London for some years, owing to the C&GWUR not having the funds to build the mainline over the hills and down the Stroud Valley to get to Gloucester and Cheltenham.

Once the mainline was completed Cirencester became very much the branch destination, with a change at Kemble usually being necessary to get further afield. Longer-distance travel was also served (if one can call three trains per day a service) by the Midland & South Western Junction line between Andoversford and Andover, which passed through Cirencester on its way southwards to Swindon, Marlborough and Andover for connections to Southampton.

A Map, in order to slightly clarify things. The GWR line from Cirencester runs to Kemble while the M&SWJR runs through Cricklade and then does some tight curves in Swindon. Most stations and several communities omitted to aid clarity.

The Midland & South Western Junction was shut while still working as a traditional rustic cross-country railway. (Thinning out half the stations, upping the linespeed to as close to 90mph as possible and laying on an hourly service with air-conditioned diesel multiple units would probably have made it relevant, if not profitable. Sadly this wasn’t an option in 1960.)

The branch from Kemble to Cirencester was being closely watched from the early 1950s, but the initial decision was to convert the passenger service to four-wheeled diesel railbuses instead of closing it. Three railbuses were bought for the Kemble area – one for the Cirencester line, one for the Kemble to Tetbury branch (vanishing off from Kemble in the opposite direction) and one to be under maintenance. They could not run in passenger service from Kemble down the mainline to Swindon, a mere 13 miles away, because they were too lightweight to operate the electric track circuits that controlled the signalling equipment.

There was also some suggestion that one railbus was too small for the Cirencester line. Unfortunately the railbuses could not operate in pairs while introducing a larger diesel multiple unit would ruin the economics of the two adjacent branches using matching vehicles. Both lines were closed instead.

There has been little subsequent development. The formation of the 7-mile Tetbury line is untouched, except for the intermediate station at Culkerton having entered private hands. The bay platform at Cirencester now acts as a siding in which things are occasionally dumped; most of the 4-mile formation is intact to the outskirts of Cirencester, but there has been some redevelopment in the town centre.

The Cirencester bay platform at Kemble, looking towards the bufferstop in a tantalising manner.


Well, they are almost endless.


Obviously both lines make excellent candidates for cycle paths, although there is no real case for spending money on buying up the Cirencester branch formation, clearing vegetation, rebuilding bridges and surfacing it given that there are some perfectly good Cotswold back lanes with scenery and hedgerows and no traffic, and that the awkward bit to cycle is the bit through the Cirencester outskirts where the railway formation has gone. At 7 miles the Tetbury line is a bit less fun as a commuter route on a bike, although the first section out of Tetbury has been developed for that purpose already.

Trams/ Light Rail

The railway is fairly short so is an obvious candidate for the modern version of the old railbuses – a self-powered light-rail vehicle of low capacity connecting into the existing mainline service. This would be cheaper to provide than a full railway.


Converting old railways into guided busways became a bit of a thing ten or so years ago (St Ives in Huntingdonshire, Luton to Dunstable, Gosport in Hampshire) but in this case is not really worth the candle. The lines were both single track, so a twin-track busway is awkward; through running from Tetbury to Cirencester is inconvenienced by the fact that the branches come into Kemble station on opposite sides of the Swindon to Gloucester mainline; and in any case there is a perfectly good 50-60mph road between Tetbury and Cirencester, much of which was built courtesy of the Roman taxpayer as part of the Fosse Way intercity road. A reasonable bus service could be laid on between the two places, and possibly Kemble, and the nearby towns of Malmesbury, Stroud and Swindon, without any inconvenience to the slumber of the railways. In fact there already is a bus service between Kemble and Tetbury/ Cirencester, so all that is needed is someone to pay to upgrade it to be more useful. This would be even cheaper to organise than the trams, and potentially cheaper than the cyclepath.

The bus at Kemble station. The Cirencester bay platform lies beyond it on the other side of the fence. Note thriving car park, which does need some kind of solution but also represents a lot of potentially longer car journeys already using Kemble station as a park-and-ride hub.

Pause a moment…

… what exactly are we seeking to achieve here?

Wanting to do something with the railway formation is an ambition, but it does not get us very far. Similarly wanting to have a railway into Cirencester again is cute and laudable, but not necessarily actually important. The existing bus can get from Cirencester Forum to Kemble station in 12 minutes. What needs improving here?

First up, an hourly service connecting with the trains at Kemble would be jolly nice – much aided by the fact that the weekday train service now involves the trains crossing at Kemble, so one arrival and one departure can connect with both trains.

Through ticketing would be handy.

Cirencester’s bus stop is advertised on National Rail Enquiries. At the moment the bus times are not, but this is easily remedied when public transport timetabling stabilises. A customer could then book a ticket from Paddington to Cirencester in a journey taking maybe 95 minutes, which is competitive with the car (Bing reckons the car trip is 115 minutes). To an extent what is needed is promoting what already exists.

But trains!

A perfectly true header. Rail services penetrate markets that buses do not.

So, what markets have we got to play with around a railway from Cirencester to Kemble?

Cirencester to Kemble traffic

Kemble is a small Cotswold village which by an accident of history now has an hourly intercity service to London, and is consequently the railhead for a whole host of other places that are too small or inconvenient to have rail services. This convenient high-speed service to an accessible station has made it a good place for people who want to live in the Cotswolds while working in London to buy houses. The journey time is similar to that from Chipping Norton’s nearer-to-London local station, and the area is not full of high-powered celebrities.

This is a long way of saying that there is no fundamental reason for someone in Cirencester to go to Kemble except to catch a train.

Although people do keep building more houses in the area and easy rail access to Cirencester would lower the growing community’s carbon footprint…

Kemble to Cirencester traffic

Better, but Kemble is still a small village. Also the range of shops and nightlife is better in Swindon or Stroud than in Cirencester so a significant proportion of people who live in Kemble and don’t happily drive to Cirencester probably take the train to one of those places.

Cirencester to Swindon traffic

Ah, now we’re getting somewhere. Swindon is a half-hour drive from Cirencester, though getting into the town centre is an alarming experience and not necessarily worth it when you get there. Still, this is a flow to appeal to.

Note that some time ago Swindon Borough Council tried offering these people a park-and-ride facility on the north side of the town to save them driving right into Swindon. It is probably fair to say that anyone interested in this from Cirencester was already using Kemble for the purpose, and therefore any attempt to attract the remaining drivers to Swindon out of their cars needs to be as seamless as possible. If they are willing to change at Kemble, they are either driving there already or they are using the bus link.

Cirencester to London traffic

This is most likely not a very large flow; Kemble makes it look like a large flow, but Kemble also covers Malmesbury and Chalford and Tetbury and Poole Keynes and Fairford and Bibury and probably some of the outskirts of Gloucester and Cheltenham where it is quicker to drive to Kemble than to drive into the city centre and catch the train out again. Also post-Covid home-working.

Cirencester to Gloucester and Cheltenham

Unhelpfully some Roman Emperor built Ermine Street to link Cirencester with Gloucester (links it to Swindon too – good bit of transport design, that one) which doesn’t go five miles out of the way via Kemble and Stroud. Through trains are a challenge because the junction points in the wrong direction and the station is on the wrong side of the junction. (Through trains from Tetbury are no problem – straight into the Gloucester-bound platform at Kemble and reverse.)

Tourists to Cirencester

There are quite a few of these – apart from anything else Cirencester is a good place to just go to on a miscellaneous sort of Saturday for a mooch and a spot of admiring the church, though it lacks a second-hand bookshop – so there is some market here. But discretionary travel tends to follow the road (or railway) of least resistance, so the service has to be pretty seamless. The sort of seamless where you turn up at a station and buy a ticket to Cirencester and board a train direct to Cirencester.

While local traffic will get used to the foibles of any transport link, it is important here to imagine the leisure tripper who is not well-versed on public transport but views it as a means of getting to Cirencester for a Saturday afternoon mooching the shops, pondering chic interior design, looking in the church and taking in a coffee parlour. The public transport option can be in with a good chance of winning on stress and price if it does not involve parking in Cirencester. But it needs to avoid having to run around Kemble station (not a totally intuitive place), buy multiple tickets or worry about tight connections. Think parent and two children while other parent is at football. Think young couple who want a relaxing day together. Think older couple who want a relaxing day together. Think elderly relations who want to sit down and settle back into a light and slightly deaf doze. Think person who has accidentally bought a nice vase in Cirencester and would like to put it down somewhere safe until they get home. Think the Chinese tourists from Bath for whom changing at Swindon is already a language test.

(Actually, for a true story, think of the three generations of a Japanese family with only one English speaker who want to intersperse their Olympic holiday with a trip to Cirencester, discover that Kemble is the railhead for Cirencester, assume that as Kemble is an intercity station it must be in some major place with a taxi rank and obvious onward transport, and get to the end of the station access road to find themselves overlooking the water meadows of the upper Thames Valley in a state of some slight bemusement. Also imagine the one person out for a stroll who appears over a gate and has to explain to them how to cover the remaining four miles to Cirencester.)

Cirencester town centre. Note existing vehicles owned by some possible future rail users getting in the way of gardens, benches, more market stalls, etc.

Cirencester residents without a car

This market is a hidden one because getting around from Cirencester without a car is a messy activity restricted to various bus services. People in this category are therefore more likely to be relatively economically inactive, especially if they are not fond of buses. A train service would help them get further afield more easily, simply and quickly, if it were easy, simple and quick. Some of them will be people whose partners have cars; some will be children and teenagers who do not yet have access to a car and need giving a reason to not buy one; some will be older people who are no longer comfortable driving. There will be a small number of households who are car-free by choice, but the lack of fixed frequent public transport connections means that Cirencester is not necessarily a place that one moves to with an aim of living without a car.


Cirencester is home to the Royal Agricultural College. Live-in students are a traffic rather inclined to awkward peak flows at irregular if easily-identified times, but a train might make it easier to commute in on a daily basis. It may also discourage live-in students from wanting to keep cars (some still will, but some will say “oh look, railway” and not bother).

The College/ University, looking needlessly impressive.

So what does this mean for rail services?

It means that:

  • we shouldn’t expect demand to immediately soar to levels over those shown in the 1960s, as the market would happily leave the status quo where it is – there is not evidence of pent-up demand, and the railway would need to establish its own market;
  • if trains do resume they need to run to/ from Swindon in order to maximise convenience and avoid changes of vehicle – people who want to keep breaking off and changing vehicle then more of them would drive to Kemble or use the bus link.

Note that a railway is unlikely to achieve 100% market penetration in any sector of society, so it needs to target as many sectors as reasonably possible without becoming unattractive to all of them. Obvious point, but easily missed – we are not talking about all people in Swindon who want to go to Cirencester always going by train (but half of them making half their trips by train could still be a good chunk of people and a marked cut in pollution) and we are not merely targeting the small group of people who will go by train no matter how much faff it is, and call it part of the fun, and we are not merely after the existing bus passengers.

What is certain is that a railway going to Swindon will attract more people that other options, for the following reasons:

  • pretty good in all weathers with a largely dry journey (unlike cycling or making a connection at Kemble);
  • smooth consistent run (unlike the car);
  • perception of quality (unlike the bus);
  • perception of permanence that can be depended upon (unlike the bus);
  • once a journey has begun there are no connections/ breaks of journey to worry about (unlike connections at Kemble);
  • car-free (unlike driving).

In fact the through train is in a small way a better bet than driving to Kemble on the penultimate point, because there is no risk of the drive being caught out at the last moment by getting stuck behind a tractor.

A side bonus of running to/ from Swindon is that this allows the service level to be set at a level which meets existing demand and generates more, rather than running to connect in and out of an hourly service at Kemble that is timed to at least some degree based on pathing into Paddington.

It is worth noting that the following costs are similar for an hourly shuttle-car connecting Kemble with Cirencester and an hourly train linking Swindon with Cirencester:

  • Land purchase;
  • Bridges;
  • Maintaining legacy infrastructure (embankments, old bridges, etc. built for heavy rail);
  • Driver;
  • Procuring a unit (one-off new shuttle car against spare 2-car Sprinter).
One of the few structures on the line – a handsome overbridge crossing a small back lane and a canal. Requires repointing, clearing of vegetation and provision of parapet walls whatever you want to run over it. (Also the drain underneath it needs unblocking, or possibly providing – the road floods badly most winters due to the artificial dip to pass under the railway.)

The following can be done more cheaply or conveniently as part of a large operation:

  • Managing maintenance spares (for track and train);
  • Managing traincrew and relief turns;
  • Managing safety cases;
  • Managing ticketing, planning, advertising and back-office functions (although obviously a local support group willing to stick up their own posters and run a ticket office on the cheap will work out considerably cheaper than making a big corporation with unionised staff do this).

All-in-all, persuading the local train operator to run a diesel unit from Bristol or Westbury to Swindon each morning and shuttle it up and down between Swindon and Cirencester all day will probably be better overall value than connecting in and out of the London trains at Kemble. It will also be rather more resilient – a train from London has to cover some 77 miles before it gets to Swindon, during which time many and varied things can go wrong to upset a journey from Swindon to Cirencester. (This thought is why people don’t like changing trains, never mind changing to a separate transport operator who may or may not wait around with a vehicle full of grumpy passengers to see if anyone’s coming up 20 minutes late from Swindon.)

Doubling-up business cases

Of course a train running from Swindon to Cirencester calling at Kemble every hour costs about the same amount of traincrew and rolling stock as a train running from Swindon to Cirencester that also offers a Swindon suburban service.

Roughly inserting three intermediate stops on the run out of Swindon, for Rodbourne, Mouldon Hill and Purton, would serve local housing estates and villages to:

  • provide a rapid means of transport into the town centre;
  • open up more scope for commuting by train to Gloucester, London and Bristol, albeit with a change of train at Swindon or Kemble;
  • ease access to the rail network generally.

On the scope to commute to London and Bristol – changing into these routes at Swindon is a little different to doing so at Kemble as the service level is better, and therefore longer-distance commuting can be done more reliably. Missing a London train at Kemble (one train per hour) is a problem. Missing a London train at Swindon (five trains per hour) is a small piece of grit in the millstone of life.

The three intermediate stops would boost usage of the unit(s) employed between Cirencester and Swindon, and as suburban stops go some way to justifying running a half-hourly service between Cirencester and Swindon instead of an hourly one. It is easier to work one’s life around a half-hourly service than an hourly one, particularly for journeys of less than 3 miles where it is quicker to walk than wait for the next train – therefore making a success of such suburban stops needs a half-hourly service, which in turn makes getting to Cirencester much easier. This also benefits Kemble’s status as a park-and-ride for its wider market, by giving in three trains per hour to Swindon instead of the current one and therefore making driving there from Tetbury/ Malmesbury/ South Gloucester more attractive than driving all the way to a destination.

They do of course need to justify themselves for:

  • expense of construction;
  • expense of stopping;
  • loss of revenue through extended journey times from Cirencester cause by the extra stops.

Given the fact that the railway has to build its own market in Cirencester, not all of which will be discretionary usage (i.e. some users don’t currently have a means of making the journey so will accept what they are offered), the third point above is hard to calculate and can probably be ignored.

On the other two points, the following considerations apply around each station.


There was a station here until the 1960s; the building still stands, although the platforms have gone. Central Swindon is reasonably accessible by car, though if this point against bothering with the railway has been rejected for Cirencester then it can equally be rejected for Purton. The village bears hallmarks of being a dormitory town for Swindon – it is not really self-supporting in terms of industry or retail facilities – and is all within a half-hour walk of the former station. There is scope for developing incoming tourists/ day trippers as well as encouraging commuting into central Swindon and shopping trips by train – the village is sat in attractive countryside with many public footpaths, an unusual church and a small hillfort. The train would allow residents of Central Swindon to get in and out of the town centre with greater facility.

Purton’s old station, with surviving ticket office and waiting room. The original main building was demolished and replaced at the end of the 1950s under the railway Modernisation Plan. The modern structure has a flat concrete roof and ceiling-level windows. It lasted five years or so as a station and is now part of a car maintenance garage. It is an unusual example of a circa-1960 railway station and should really have listed building status.

Mouldon Hill

The railway passes beneath one of Swindon’s key suburban distributor roads – really the outer ring road, and a rather stop/start and polluting affair – between Mouldon Hill Country Park and various new-build housing estates. Primarily a station would give the housing estates access to the town centre and make it easier to use the Great Western Mainline as a superior alternative to the M4 for longer trips. It would also give rapid access to the country park for town centre residents (say an eight minute journey time). Once the Swindon & Cricklade Railway has finished its northern extension this would make an interesting, if slightly disjointed, car-free way of getting to the charming Saxon town of Cricklade.

Good spot for a station really – on straight, plain-line, close-to-level track, adjacent to an existing bridge (so no footbridge required), adjacent to housing, adjacent to a main road, open space for a car park (requirement debatable though), on edge of major conurbation. All it needs is some trains of a sort that want to stop here.

Rodbourne & Even Swindon

Getting from this part of Swindon to the town centre – a journey of rather less than a mile – is an awkward exercise for anyone not on foot or a bicycle (for whom the Western Flyer route is a slightly grotty but extremely direct option along the formation of the North Wilts Canal). In any case the town’s one-way system does not favour motorists (neither, alas, does it favour anyone else) so a railway station would give an option for short hops and make journeys from the main station smoother – allowing the rail network to be joined here instead of having to get into the town centre first. It is conveniently situated for the main road from the M4 and for the Western Flyer multi-use path from West Swindon, though accommodating a proper car park for such traffic flows is not really possible (but a pick-up/ drop-off point on Galton Way might be). The vicinity includes several industrial estates, the local B&Q and a hotel, so some commuting traffic may be in prospect. Its main functions would be giving Rodbourne access to Mouldon Hill, Purton and the Cotswolds while allowing incoming traffic an alternative access route (via Rodbourne High Street) to Swindon’s Outlet Centre in the former railway works.

Part of the complex of road junctions – and associated traffic jam – at a large roundabout built around two railway bridges between Rodbourne and Even Swindon. A station on the embankment to the right of the bridge would be reasonably accessible from most directions.

Cirencester station

There is the small matter that through some act of carelessness (actually several acts of carelessness) the railway alignment to Cirencester GWR has been blocked by industrial estates, housing, new roads and a supermarket. The building is now a rather unusual-looking set of offices for the bus station.

Cirencester Town station, with more than a passing resemblance to the still-operational station buildings at Kemble and Stroud.

The no-demolition solution is to terminate the line on the very edge of the town (these acts of carelessness are almost too thorough to be mere carelessness).

Taking over the edge of a patch of alloments, strimming the edge off a back garden, knocking down two industrial units and reorganising some car hire offices would allow a small station on the Chesterton Lane, which is a little under a mile from the church in the town centre (but very convenient for the various bypasses and expressways that carve up Cirencester, and so would make an excellent park-and-ride if only someone had left room for a car park).

If the line were a success a case could be made for clearing most of Meadow Road (Scottish rail reopenings have flattened more houses simply to provide a turning area) and taking a lane off the Bristol Road dual carriageway (reducing the northern half of the road to two northbound lanes, slewed eastwards with the central reservation removed, and one southbound lane). This would give rail access to a single-platform terminus adjacent to the hospital, the amphitheatre and Cirencester’s Waitrose. Such a station would be half a mile from the church. Timetable considerations would probably necessitate shutting the Chesterton Lane station.

The Amphitheatre. Not terribly impressive nowadays, being a large grassy area with enormous hummocks arranged in a circle. Behind the trees is a main road which rather cuts off this part of Cirencester from the town centre proper.

Back in the day there were also intermediate stops at Oaksey and Minety. Neither of these represent large centres of population and are therefore unlikely to generate loadings that would warrant extending the journey times to stop at these stations (as the half-hourly round trip would then require three units instead of two).

And here’s another copy of that not-very-to-scale map for people who don’t know the area to keep up:


Well, there are two points to this.

First, a suggestion for a nice heavy-rail link to Cirencester which ties it into the wider rail network and increases connectivity for people outside Cirencester too. Not that it’s terribly likely to happen, and especially not like this, but a thought nonetheless.

Second, a thought for other schemes – what is your project actually for, other than restoring a railway? Any railway restoration will do something, but one with some thought given to details of its application and what its user base can do with it will obviously do better than one which says “look, railway”. To be quite honest, if it’s a good scheme you’ll be able to get ordinary people, or possibly even people who own large off-road vehicles for doing the school run, to say “Actually, I could use that”.

Kemble station looking north, with Thames Turbo unit preparing for the climb out of the Thames Valley on its way to Gloucester. The Cirencester bay is behind the station buildings out of view to the right while the Tetbury bay platform was behind the containers to the left.

US Election 2020

Elections in the USA are odd things to an UK observer. The President of London (we call him the Mayor) is elected by a sort of form of proportional representation where each voter marks a first preference on the ballot paper (for who they want to win) followed by a second preference (for which of the Conservative and Labour candidates they would like to win). All the votes are totted up, the top two candidates (usually Conservative and Labour, excepting 2000 when it was Conservative and Red Ken) are identified and the others are eliminated (acid baths generally). The second preferences for the less popular candidates are redistributed amongst the top two (where relevant) and one of these two wins.

The equivalent to the US system would be that every London borough gets to vote individually for who the majority of people in that borough want to win, and then the boroughs are allocated a number of votes based on their population and importance at some point, and those votes are allocated to the candidate that the borough voted for, and the candidate who gets most of those votes wins.

The main function of this is that it stops people voting for irrelevant and special-interest parties. (It is also supposed to stop the big states bullying the small states, on the assumption that the President doesn’t bully the small states anyway once elected.)

In the UK almost all the votes are counted on the night, even for complex things like the Mayor of London, and the new Prime Minister can be kissing hands with the monarch within 12 hours of the polls closing. It is shockingly quick – though not as quick as monarchical succession, which is instantaneous. (The business of Government must continue.)

Of course in the UK Government has all been at a standstill for the last six weeks because there’s been an election on; the Government can’t do anything interesting during the election, and the MPs (and most of the Cabinet, and the Prime Minister) are technically not members of Parliament but merely candidates for seats. None of this thing where you can appoint judges during election season in the UK. Not that UK judges are political creatures anyway.

So the idea that there won’t be a result from the US election for weeks and any new President won’t be in office for months is a bit ridiculous.

Not that my personal betting is that there will be a new President.

The final polling was something like Biden 52% against Trump on 44%, but that’s a nationwide result. That could mean that everybody in California is going to vote for Biden, which would be very cute – but won’t be of any more legal use for him than if 51% of the people in California vote for him.

Instead matters rest on swing states, in the way that they rest (usually) on marginal seats in the UK. The darling Beeb identified 10 of them – if Biden wins the right ones, or indeed all of them, then he’s President.

Thing is:

  • Arizona and Florida are very close;
  • Pennsylvania and Nevada are not much better;
  • Voters for Trump are more likely to be uncertain about telling pollsters this (he is hardly popular in certain areas);
  • The general impression is that Trump-voting is a vote for the President whereas Biden-voting is at least partly against the President, and voters who don’t want something are a very uncertain basis on which to fight an election. (Ask the Remain team in the Brexit referendum.)

If we go with the Beeb’s current predictions, and give Biden Wisconsin, Arizona and Michigan on top of the Democrat-leaning states, then Trump loses and we are apparently all happy.

If however we assume Trump picks up an on-the-day gain of one-to-two percentage points from:

  • being the incumbent;
  • shy Trump voters;
  • soft Biden voters not turning up;
  • polling uncertainties;

then Trump can expect Arizona, North Carolina, Pennsylvania, Florida and Georgia, which on top of the Republican-leaning states is enough to put him back in the Oval Office for a second term.

I personally am no enormous lover of Trump, but have not seen great signs that I should be enamoured of Biden. I have some sympathies for those who don’t really buy career politicians, and I’m not that certain about a party that can’t attract successful younger candidates.

However, we will all find out what happened when the Supreme Court rules on who won.

Trails from the Rails 20: Betws-y-Coed to Dolwyddelan

  • Area: Conwy/ Caernarfonshire
  • Local Train Operators: Transport for Wales
  • Length: 12 miles
  • Points of Note: Llyn Elsi, some nice views, the Lledr Valley, trains on the Blaenau Ffestiniog line.
  • OS maps – OL17 and OL18 (1:25,000) (spans both maps); Landranger 115 (1:50,000)

At time of writing and posting there are in fact no advertised trains running on the line between Llandudno Junction and Blaenau Ffestiniog, and have not been since a major wash-out in February 2020. Prior to this the service had been fairly intermittent, with a lengthy closure in November 2019 for tunnel repairs, another five-month closure between March and July 2019 due to another wash-out, two months out of action due to rolling stock shortages following flooding just after Transport for Wales took over the franchise and a five-day closure over the handover period from Arriva Trains Wales in October 2018 because it had been raining. Readers intending to do this walk are encouraged to do so in the next few weeks.

The route is due to re-open on Monday (the 28th September 2020) and this post aims to commemorate this with the final walk for this second batch of Trails from the Rails.


Betws-y-Coed station has a rather handsome building from the days when it was a major station for a Snowdonia tourist centre. Now it is a minor and badly-served station (sometimes) for a Snowdonia tourist centre – though it has a website which makes no mention of the rail service.

Leave the station into the forecourt, looking out for the interesting commemorative plaques:

(Much as this is a rather daring statement – Blodwen and Olwen could have plighted their eternal troth amongst the sheep on the watermeadows, and simply failed to go down in history for this. Choosing September 7th for 30 years earlier would have been a rather better bet, and possibly funnier.)

Amble up Station Road to the main road, which examination will reveal to be the A5 on its way from Marble Arch to the Port of Holyhead. Turn right and follow the trunk road to Ireland for a few yards to Vicarage Road, which loops around the back of the church.

Follow Vicarage Road to the back of the church and then turn left up the road up the hill.

This slogs away in a reasonably easy-to-follow manner for half-a-mile or so, after which it levels out and starts to have awkward branches. Keep following the generally main route (veer right, straight on, veer left, straight on, veer left) until the path drops suddenly to a reservoir called Elsi.

The quick route is to turn left, but for scenic charms turn right and go round most of the reservoir, admiring the views along the way.

The small path following the reservoir bank – not marked on the map, but followable by dint of continually turning left – eventually comes out onto a larger forest drive. Follow this on round a final corner to the south (top) end of the reservoir and take the drive that turns off to the right, dropping back down the hill.

Again, some moderate navigation is required; left at the first fork, carry on at the convergence and note that the useful path shown on the map bearing off to the right shortly afterwards doesn’t exist. In all honesty the next turning to the right, straight down the hill next to the stream from what the map says is a crossroads, isn’t much either.

At the bottom of the hill the path comes out abruptly onto a track beside the railway, just east of Pont Gethin – the railway’s viaduct over the River Lledr. Turn right and drop down to the road. This is the A470 Cardiff to Llandudno trunk road, which squeezes round a vicious “S” bend under the easternmost visible arch of the viaduct.

Once across the road and under the railway there is (around the point from which this picture was taken) a footpath dropping down off the road into the woods and working to the right towards a footbridge over the Lledr. Tree growth means that the railway viaduct is barely visible.

Having crossed the bridge, turn left and pass through the trees and an overgrown field to the access track to a farm.

Here there is a bit of a choice, as turning right, heading up this track to the road, going straight on and climbing southwards up the steeply climbing road beside the wood will eventually lead to a basin in the hills near the top of the Wybrnant, where the National Trust have a house and the landscape is rather something:

If following this option, keep on up the road over the top of the hill and down the ridge the other side, and look out for a point where the road sharply kinks to the left while a byway turns off under a tree to the right.

Alternatively turn left along the track, sticking close to the river, and come out onto the road a little lower down the valley. Turn left and enjoy the views as the Lledr valley widens on the approach to the confluence with the Conwy:

At the junction swing off to the right and climb up the hillside as the road enters the Conwy valley:

After some steady climbing the road turns to the right again past the buildings of Pandy Mill and into the Machno Valley. Soon after the road begins to fall and a forest track diverges to the right.

The track soon splits; leave the falling path to the left to fall away and take the rising one to the right. This swoops along the hillside, gently rising, for almost a mile and then doubles back. Another nearly-half-a-mile of rising leads to a junction. Double-back to the left and continue south-westwards for another half-mile of gentle undulations. At the next junction fork right and stick in a further stint of rising. This eventually offers some rather tantilising views of craggy peaks that the walk will neatly bypass.

Drop down through the junctions in a fairly straight line to the road and turn right up the hill. The road curves gently left and then turns off to the right. Ahead a byway leads under a tree into the woods. Follow it.

The initially claustrophobic passage through the trees eventually opens out into heathland at Pigyn Esgob; the byway, so far as it is of use to motorised vehicles, ends at a step in the hillside. The path continues through the heather across the watershed, eventually passing through the remains of a wall by a good viewpoint. This looks back across the Wybrnant valley towards Betws-y-Coed.

The path returns to being a track with two wheelways carved across the rocks and through the periodic wet patches, though exactly how four-wheeled vehicles get up here is not wholly clear.

After wandering through the heather and conifers for a bit the byway opens out again into plain moorland and descends past some more splendid views to the north – primarily towards Moel Siabod.

The mystery about the road vehicles is because the byway now disappears entirely into the heather before ending at a barbed-wire fence. A short walk to the left beside the fence will lead to a stile. Pass over it and continue in a similar sort of direction across the bogs.

Various small hillocks and peaks give an opportunity to get off the bogs and admire a view from a little above the prevailing ground level. Below lies the isolated little tower of Dolwyddelan Castle. Beyond is Snowdon – high and pointed on a good day, but usually covered in cloud.

The path drops across the surprisingly damp hillside in an unhelpfully indistinct manner to eventually pick up a forest drive, which can be followed straight down the hill. Technically the byway stays on the right of the stream and waits for the drive to come across and join it, but the path on the ground favours wading the stream to join the drive.

The byway then drops down the hill in a fairly easy-to-follow manner, which gives more scope for admiring the peaks and the unfolding valley.

As Dolwyddelan gets nearer the hillside steepens, forcing the byway to take up a zig-zag approach to the first houses.

The main path keeps weaving, but a side path drops down the hill by the banks of the stream beneath a couple of houses.

A proper surfaced road is then joined, terraces are passed, a junction is reached and the main road swings right to pass over the railway. This is the Sarn Helen Roman road. Beneath lies Dolwyddelan station.

Long ago the station had an island platform and it was accessed by means of a flight of steps down from this bridge. The point where the stairs left the bridge is still obvious – a narrow patch of wall between two 4ft pillars. As the northbound line and platform have been removed a new access route has been established across the formation. To reach this, continue down the other side of the bridge and turn back to the right.

The main part of Dolwyddelan village, with its shops, church and (more distant) castle is across the river and set around the A470. The infrequency of the train service may mean that those not already exhausted might fancy ambling onwards down the lane past the station to the next station at Pont-y-Pant instead of sitting on the platform here for anything up to three hours. The navigation is not difficult, but see Trails from the Rails 8 for more details.

Trails from the Rails 19: Gloucester to Lydney

  • Area: Gloucestershire
  • Local Train Operators: Transport for Wales/ Cross Country/ Great Western Railway (Gloucester only)
  • Length: 29.5 miles
  • Points of Note: May Hill, Soudley Ponds
  • OS maps – Explorer 429 (1:25,000); Landranger 162 (1:50,000)

This is a Stupidly Long Walk suited to people who have nothing better to do than spend 12 hours having a walk between two stations that are quite a long way apart (an urge to see Newnham reopened may arise about halfway through). It is not suited to anyone whose fitness levels have unduly slipped in lockdown, or who may have been in a habit of walking into things and injuring their knee recently.


We begin at Gloucester station. Its main point of note is that platforms 1 and 2 represent the longest uninterrupted platform face in the UK. Otherwise we will leave its uninspiring main building behind as swiftly as possible.

Having passed out of the main building, cross the main road and walk past the bus station – which is newer, and actually called the Gloucester Transport Hub, and seems to have got in an architect:

Turn right along the road south of the bus station and admire the mixed architecture of Gloucester; this area makes a fair sampling. The brand new bus station sits opposite a bit of 1960s false marble, a 1950s bit of creamy faux stonework in concrete and two chunks of Victorian brickwork in non-matching scales. Having passed the Victorian brickwork, turn left down Clarence Street.

Clarence Street’s main point of interest is the line of bus stops that enable looming double-decker buses to park up and benefit the public by blocking the view of the shopping centre behind them.

At the end of Clarence Street turn right and follow Eastgate Street past the shopping centre entrances (and the site of the Eastgate) to the very centre of Gloucester. This is marked by St Michael’s Tower, which turns out to be just a tower rather than a church, and is where Northgate Street, Westgate Street, Southgate Street and Eastgate Street meet in a useful geography lesson for Gloucester’s primary-school pupils. Turn left into Southgate Street. This again has a glorious mix of architecture, plus a rather notable Clock:

Continue down Southgate Street, past the fork into Commercial Road, and into the square where the pedestrianised precinct meets the traffic. Fork right here across the road and down the steps into the Docks.

There’s a waterways museum in here, but there isn’t time to visit – instead continue along the north edge of the dock complex past the Victoria Warehouse to the North Warehouse and a second dock area. Admire the travelling crane, the mysteriously not-auctioned 2015 rugby ball sculpture and the view down this dock from the walkway over the lock gates.

Pass between the next set of warehouses to the road, turn right, go up towards the swing bridge, cross the road where safe to do so and turn left to go over the footbridge signposted to Over and Highnam. This crosses the East Channel of the River Severn onto Alney Island. Here we pick up the Wysis Way, which will be unimaginatively followed (with minor deviations) for the next eleven or so miles to Longhope.

The Island is a charming place that is almost entirely uninhabited except for a lock cottage for a derelict lock, a car park, some electricity facilities, a few rows of lost terraced houses, a load of mobile homes, a surviving scrap of railway and four major “A” roads. Turn left and almost immediately drop down the grassy path off the former railway headshunt onto the marshes on the riverbank, just before the embankment ends and becomes a low bridge. Follow this path around the edge of the island, avoiding the car park. Looking back will reveal that the bridge is struggling to stand out amongst the brambles and trees.

After a bit the nice grassy path is joined by a fresh gravel path from the car park, which passes under the main road and then begins to swing away northwards. A further grassy path splits off to the left to stick with the riverbank and passes under the decaying bridge that carried the GWR Docks branch into the docks. The main path/ cycleway swings away northwards and picks up the much-landscaped railway formation straight across the island. Either option will do, as they both end up in the same place.

The same place, about quarter of an hour later and after some pleasant views off to the left of Highnam Church, is next to the railway bridge carrying the Gloucester & Dean Forest Railway over the river Severn on its way out of Gloucester towards Lydney. For about 80 years the bridge was bookended by junctions; the Docks branch converged up to the right, and the Newent Railway to Ledbury diverged northwards on the other bank of the river to the left. Now the mainline passes through uninterrupted, observed by grazing cows.

The Docks branch is the first of six railways which branched off the mainline between Gloucester and Lydney, all of which have to be crossed (and occasionally followed) over the course of the walk.

Pass under the railway and follow the path as it winds its way towards the eastern end of Thomas Telford’s single-arch stone bridge for carrying the main road to Fishguard over the West Channel of the Severn. Disappointingly vegetation means that the best view is actually from the train. The main path winds a bit further northward to provide access for bicycles, but a steep cut-off has been created for pedestrians. Once on top, admire the newer bridges on each side and reflect that until 1966 this was the main road from London to South Wales.

The old road is higher because it was crossing a navigable watercourse used by sailing vessels, but the high level also came in handy when the Newent Railway had to fit under the road just beyond the west end. The railway and approach embankments have gone, and little trace remains of the railway bridge at the information board where the footpath now reaches the modern A40. Double back to the right and pass under the new bridge to get to the northern side of the A40 (which underpass is all one expects of a riverside path beneath a major modern road in an urban area). Continue westwards alongside the road to the Over Farm Market and then turn right up the gated lane towards Highnam.

This represents the first climb on the walk (apart from odd stretches out of subways) but is nothing very notable. Initially it offers some views across the plains of the bottom end of the Severn Valley.

These change into a more gentle and rolling look once over the ridge and dropping down the other side, where the lane briefly coverges with the Newent Railway’s trackbed.

The road then curves away from the railway formation to pass an airfield (which does look rather more like a field with air in it than some kind of airport, and could be mistaken for a spare field if it wasn’t for the sign advising against entry). It heads along the edge of this airfield to the corner of the woods at the left of the photo, where it divides. Fork right into the woods and follow the path diagonally up the hill. When it levels out at the top let it swing round to the right some more and follow it until it emerges into open fields on the edge of Highnam.

Staying on top of the ridge, cross the field to the diagonally opposite corner and briefly abandon the Wysis Way to drop down the direct path towards Lassington – a small hamlet easily distinguished by its church. Or rather, the lack of most of its church – the main body was felled in 1975 before it fell down. It was an attractive little Victorian rebuild of a Norman church.

On reaching the bottom of the valley the footpath crosses a stream and comes out in someone’s front garden (probably why the Wysis Way doesn’t come down here, and stays in Highnam instead before dropping down other fields). Leave the garden by the main gate and turn right up the access lane to the road. Turn right and follow the road through a dip to the large red-brick manor house and the remains of the church. The churchyard can be investigated. Having done so, pass around the back of the manor house, across the farmyard and through the gate at the end into the field beyond.

There is no defined path across this field, but the target is the spike of natural vegetation sticking out into it some 300 yards away. A set of right-of-way signs indicate a path that drops out of the field into the bushes, ponders dropping properly, thinks better of it and returns to the field boundary and then sneaks off down into the undergrowth when it hopes you aren’t concentrating. After a short walk through the trees beside the railway trackbed this comes out into the open again at a corner in the hills. The railway sweeps away northwards; the footpath follows the natural contours around to the right to come alongside the road shortly before reaching the hedge across the northern end of the field, turns sharp right to follow this hedge, passes through a gate and finds itself on the old railway line.

Turn left and follow the railway northwards for a field until the land on the left begins to rise away again. The path sneaks up the banking to the left when it hopes you aren’t concentrating, passes through a gate into another field and turns right to follow the railway boundary at a higher level.

Halfway across the second field it will be necessary for those using the 1:25,000 maps to swap Explorer 429 for the East Sheet of OL14.

At the end of the third field the path provides an opportunity to fall over a Great Western Railway cast-iron boundary post before climbing up to meet a small lane that passes over the old railway at this point. Cross the road and drop down the other side to join the railway trackbed.

The railway here was obviously intended to be double-track (though was always single) and also throws in the odd widening to highlight that the railway was built on a rather bendier (and drained) canal. After a bit of wandering it straightens out and heads for a distant road overbridge – plus a lot of traffic that confusingly is passing in front of the bridge.

This is because the road has been realigned to avoid the faff of going over a narrow bridge to cross an abandoned railway, so it is necessary to cross the road on the level. Off to the right, raised a bit above the road, is a memorial – in very lichen-coated carving – to the Welsh of Lord Herbert’s force who did something now illegible in 1543.

Cross the road, pass the allotments and climb up the bank to the next railway bridge. A slight diversion to the right onto the bridge is repaid with a view of the stationmaster’s house and Barber’s Bridge station, which has been very charmingly restored. The GWR’s pine trees still provide a shady feature that denotes the presence of a rural railway station.

Not quite opposite the point where the footpath joins the old road – it’s a little down the approach embankment to the left – is a track that forms the next few hundred yards of the walk. It swings down from the road, briefly comes alongside the station, thinks better of following the railway and strikes away towards the farm at Bovone. This is the last we will see of the former Newent line.

At the farm some signs point one way while the map shows the public right of way going straight on, avoiding the main farmyard and then swinging round the north-eastern corner of the main buildings to take up a westerly heading towards a small gate just down the slope from the farm. Pass through the gate and follow the path across two fields to the back of the village of Tibberton, where it runs along the backs of some gardens before diving between two houses into a suburban road.

Turn right and follow the road round and down to join the main road through the village (which is largely a pleasant backwater). Turn right onto the road, cross the river and turn left again into the lane towards Huntley. The gateposts for Tibberton Court lie off to the right of this lane. The path sticks – for now – to the road.

At the top of a slight hill the Wysis Way turns off to the right, just when idle strollers are distracted by the adjacent war memorial, and runs down a narrow track before bursting back out into fields. Away to the north are the shoulders of the Malverns, looming deep blue on the skyline. Off to the west can be seen the distinctive clump of trees atop the apparently nearby and not-too-high May Hill. Actually it’s a big hill and something of a step.

At the end of the track the path swings slightly to the left to cross a field, jars some more to the left at the next field boundary to cross a second field, and then goes straight on into a third field to discreetly dive off through a hole in the hedge and over a footbridge into a slightly more landscaped field behind the Grove manor house.

Cross this field towards the tree-lined drive most of the way across, ignore the public right of way arrows encouraging you to explore it and continue to the far boundary. Here the onward tantalising track will turn out not to be the public right of way after all. The public right of way goes through the hedge twenty yards to the right and then follows the damp and overgrown boundary of a wheat field up the hill. Other footpaths fork off up the hill towards a spinney in a more obvious and tempting manner.

A right-angled triangle of trees with the sharper corner cut off takes a bite out of the top end of the field; trying to hide in the corner beyond this is a gate leading through into the field at the top of this climb. Pass diagonally across this field to the top right-hand corner, letting the hedgerow climb up and converge from the right (crop permitting), go through the gate and then trot up the edge of the next field to Taynton.

Taynton is a small village with some big buildings.

Turn right up into the village and then take the sharp left back down the slope out of the village. The lane meanders beside a field briefly before a gate off to the right is signposted as the next bit of public right of way. Turn right and cross the field, where the farmer is known to kindly mark out the right of way through the crop.

At the end of the field turn left briefly and swing round the end of a clump of trees before resuming a westerly heading up the now definitely climbing fields towards May Hill. Keep the hedgerow and its stream to your right and rise up through the fields, orchards and sheep to the grand house at Byfords Farm. Turn slightly to the right to go up the access drive up to the next road.

Cross the road into further fields, where the woods are now closing in.

Two fields and a patch of scrub lead up to a dam, which holds back a charming reservoir. Turn left and follow the bank to the top. A footpath at a slightly higher level provides a means to avoid any fishermen.

At the top the signage goes cold and a certain amount of dead-reckoning is required. Two streams feed the reservoir, combining just before entering it; cross the first and then take the path to the left up the hill instead of crossing the second. This rises up through the trees for a bit and then starts offering glimpses of a woodland cottage. After a few more yards it offers some full-on views of the cottage and then joins another path. Turn right, pick up the cottage’s drive and run along the upper boundary of the garden to find that this is not an isolated cottage. It lies in the hamlet of Glasshouse, where people do not throw stones.

Turn right onto the road, pass the pub and follow the road round to the end of the hamlet, where a path leads steeply off to the left up the hill. Climb it, ignoring a couple of temptingly easier-looking unofficial turnings.

Eventually the path wearies of the straight climb and swings leftwards to rise up the hill more gradually before coming to a crosspaths with a rather more important-looking track. Turn left and go round the gate. Turn right, following the National Trust signs to May Hill, and resume climbing past the cottages.

The Gloucestershire Way trails in from the left and then the track comes out into the open.

Follow the track straight up the hill to the clump of trees, taking care to dodge the bogs, and then pause to admire the view, taking advantage of the benches as desired. The woods include a couple of explanatory plaques announcing that they were replenished in 1977 in tribute to the Queen’s Silver Jubilee.

If not stopping then follow the main path as it skirts the trees and then turn left around the northern end of the clump. The path down now lies more or less straight ahead and due west through the gorse and bracken; it drops almost sheer down the hill into the trees.

On reaching the road, stagger left and then resume the plunge, past the houses at May Hill Farm, leaving the drive to swing off to the left and down between the woods past several rather good views away to the north.

A small patch of scrub across the path is followed by a slight wobble to the right to pick up the old (very steep) lane between two rows of trees, which leads down to a gate and the road. Turn left and follow the road down into the valley.

The Gloucestershire Way forks off to the left to cut off a corner here, or follow the Wysis Way under the former Hereford, Ross & Gloucester Railway to meet the A40 again. Most of its traffic has gone now – dispersed amongst the Gloucestershire villages or diverged down the A48 towards Lydney – and it is a rather more amicable road than was met at Over.

Turn left and follow the road down into Longhope. The “hope” is a hill but could just as well be the village, which stretches for a little over a mile. Most of it is off the A40 and it is generally quite a pleasant place, complete with a fine old church.

It also has a small wayside shop near the bottom of the village, which can be useful for stocking up on essentials.

Beyond Station Lane the road joins the Old Monmouth Road – formerly the A4136 Monmouth to Huntley road (which cuts off a long loop in the A40 via Ross but goes over the hills and through Cinderford and Mitcheldean). Carry straight on southwards down the old A4136 until it meets the realigned formation at the south end of the village. Cross this with due care and attention, continue along the road at the other side, note the total disappearance of the railway formation to the left and follow Mill Lane straight on down to the river.

The lane crosses the river, weaves amongst some cottages and ends at an array of right-of-way arrows. Follow the blue arrow to the right down a more overgrown extension of the lane and then suddenly out into fields.

From here the field appears to be “L”-shaped, viewed from the bottom. Cut across the foot of the “L” and then once on the vertical (which here is laid out flat pointing down the valley) climb up the bank to a gate into the next field up the hill. Go through it and then work out a suitable way around the patch of woodland to the left (probably just follow the field edge) to get to a gate in the hedge at the bottom of this field. Pass through the gate and swing round to the right, taking a course that cuts off the irregular corners of the field to drop down into the dell ahead and enter the main woods just after crossing the stream.

After fording the stream under the shade of the trees it will turn out that the woods are protected by a barbed-wire fence. Fortunately this is a good spot to pause and admire the valley.

Work along the top of the field to the next corner, where a small climbable bit of timber fencing permits access to the wood. Scramble up through the undergrowth to the right to discover that the track marked on the map is a trifle overgrown.

Follow the track southwards as it eventually turns into a proper track, then rises to join another track and then swings around the bottom of the wood into the valley leading up to the hamlet/ farm of Abenhall. We are now getting into the Forest of Dean proper.

Eventually there comes a junction where a track turns off to the right but the main track appears to go straight on through a gate to join the road. It turns out that the road is merely a surfaced track, but following this round the edge of the wood soon comes to the road proper plus a rather good view of the remains of Gunn’s Mill. This is now, as the information board puts it, “Under wraps for safe keeping”.

Turn left into the road and then almost immediately turn right past the mill. Follow the track as it passes a couple more houses and then swings round to the right to pass a small pond.

Turn off at the pond (before or after, doesn’t matter much) and work up past it to the spring that feeds it. Turn left here and cut across the open land beneath the trees, roughly following the contours. Some minor satisfaction can be had if this turns out to meet the start of a footpath through the undergrowth in a less tree-covered patch.

Follow the path until it comes out at a set of unmistakable Victorian municipal buildings, which turn out to be a former pumping station (now holiday cottages).

Follow the track from the pumping station to the road and turn right. This is an irritatingly busy road (for its size), but there is a lack of alternative reasonably-graded routes through Littledean towards Soudley.

After rising up into Littledean, turn left onto the A4151 and then take the first available right, just before the next useful corner shop. Drop down the lane back out of Littledean towards the Forest of Dean proper.

The lane wends its way down the valley, eventually staggering right to cross over from the left side to the right. Shortly after this is the turning to Baynham Farm. Shortly after that a signposted footpath converges from the right. Opposite is a well-hidden track into the undergrowth. Take this track and follow it as it starts to climb the other side of the valley, not quite sticking to the forest boundary as the map suggests. Eventually it swings rightwards to point south and begins to fall through the trees down a ridge of land.

At the bottom of the ridge is a general confluence of valleys and forest tracks, plus a car park. Stagger right into the car park, turn left and follow the track onwards down the valley.

To the right are a series of ponds which can either be visited (by dropping down the hill slightly) or merely glimpsed while trotting onward. The last one lasts long enough that the path catches it up and runs along its shores to the road at the bottom, where a bench offers a view back up into the trees. These are the Soudley Ponds, which are well away from the Dean’s other ponds and something of a hidden gem.

Having walked across the dam at the bottom, leave the Ponds through the gate into the road and follow it round the corner to Soudley. There are many ways of getting through Soudley, but for general variety drop down the lane to the left at the start of the village. This briefly levels to cross the former Forest of Dean branch between Bullo Pill and Cinderford – our fourth dead railway, with its Bradley Hill Tunnel visible off to the left beyond the former crossing-keeper’s cottage – before plunging onwards to the bottom of the valley. At the end of the road, after crossing the stream, turn right and follow the track along the banks to the next bridge. Cross back over (by either bridge – the old one was retained when the new one was put up) and climb a little up the hill before turning left again a few yards short of the single bridge abutment for a demolished rail overbridge.

Follow this side road as it works back up the hill to join the main road and turn left to cross the river again.

Here a road sign points in various directions with many comments about “Scenic Route”. Motorists wishing to take non-scenic routes should avoid Soudley.

A footpath leads straight up the hill ahead; follow it as it scrambles up a steep valley, almost gets to the top, staggers right through some old workings and climbs some more to a plateau. Go straight on at the crosspaths and climb up again through the next band of trees. The following junction has done its best to hide the straight-on route amongst low-hanging trees that were supposed to be an orderly hedge. The path then emerges at the complex junction around Staple Edge Bungalows.

The Bungalows are one of those places that you end up at a lot if you walk a great deal in the eastern part of the Dean. Three paths (including ours) start off the junction by converging at the north end; a public footpath trails in a few yards later (actually the eastern end of the Spruce Ride that provides a similarly-useful feature of Dean walking on its two-mile course to Speech House) and then our onward track diverges to the right. The main track continues swinging round to the left and ends up heading back in a direction which is useful for returning to Soudley.

Having diverged to the right the new track drops for a little distance and then splits. Fork left here and run level for a short distance. Another junction follows; the left fork curves away to the left, while we take the right fork that drops down the hill. It makes an awkward crossing of a more important level track halfway down and then descends to a junction just above the noise and bustle of Mallard’s Pike.

If wishing to visit the lake here, turn right and take the footpath dropping to the left. Otherwise continue straight on and cross over a deep cutting. Here there used to be a bridge for our track (obviously gone and replaced by this embankment) and below passed the ill-fated Forest of Dean Central Railway. The section of track running through this cutting opened in 1868 and became a dead-end headshunt in 1872 that was largely abandoned in 1921.

The path continues dropping steeply to join the access road to Mallard’s Pike at a cattle grid before coming out onto the main road between Parkend and Blakeney.

Cross the road and follow the track up the hill the other side, which mostly rises fairly gently through the trees. At a cross-paths there are some good views across the prairie to the Rising Sun pub.

The track continues southwards to the Yorkley to Parkend road, where a path wending through the trees and bracken to the right (just before the gate) helpfully cuts off a corner and avoids walking along the immediate roadside.

Cross the road and dive into the trees again on another forestry track. Shortly after the gate (as opposed to at the gate) another footpath wends off to the right. Follow it along the ridge (ignoring a couple of smaller paths that branch off to the right) and down the hill into a small valley.

The small valley turns out to mostly contain a further ex-railway, in the form of the Severn & Wye & Severn Bridge Railway Mineral Loop – a peaceful backwater built to serve various central Forest collieries and bankrupt the Forest of Dean Central Railway, at which objectives it did rather well (even allowing for the Central Railway already being bankrupt before it really started).

Follow the railway down to Whitecroft and along the gravelled path in front of some houses that, thanks to the closure of the Mineral Loop, have front gardens and car parking. This path weaves out onto a side road, which promptly drops onto the main road. Turn left and follow the road as it curves down into the valley, climbs out again, swings round the end of a hill and drops into Pillowell.

At the bottom of the road’s plunge into the Pillowell valley and in the apex of the nasty corner at the bottom is a side turning to the right to serve some houses. Take this turning and fork left up the hill. Go straight on across the intermediate road and keep climbing to the recreation ground.

Follow the right-hand side of the very definitely not level football pitches, watch out for any wild boar digging up the pitches and drop back into the woods at the top right-hand corner. The path falls briefly before picking itself up and rising up the hillside.

At the crosspaths turn right and drop again to the next junction. Turn left and follow the track as it rises and falls along the hillside in an irritating and unhelpful manner. After the third climb it encounters a series of overhead pylon runs and turns left to drop gently down the hill into the Norchard valley. If not too late in the day there is also a rather good view over the right shoulder out to the Severn estuary.

The next turning is off to the right into the valley; the challenge is not so much to spot it as to not panic about not having spotted it. Right down at the end, just before the main track twists back into the trees and begins to climb again, a small track doing ditch impressions doubles-back into the undergrowth and young trees to find the bottom of the valley. Having done so, it rises briefly to a multi-way junction.

Head more or less straight across this junction to the track apparently heading into the hill on the other side. This promptly turns right. A small footpath climbs away to the left, very steeply. Take it (it avoids some bogs that may be encountered if staying on the forest track). This climbs sharply up, turns right, climbs some more and comes out at the bottom of a short cul-de-sac on the edge of Lydney. A carefully-placed pallet keeps some of the boar out.

This is really the end of the walk. Climb up the cul-de-sac to the main road and follow it down into Lydney; an initially easy and level walk becomes progressively steeper until it tumbles down into the town centre. Unless hunting chip shops or the Co-op, cross the main road, go past the bus station (not as fine as Gloucester’s) and fork right into a residents’ parking/ turning area. At the far end of this a path leads down beside the River Lyd (a busy stream) to a park and lake. Follow the path round two sides of the lake – left or right – to the sewage pumping facility at the opposite corner (easily spotted by anyone with a halfway-working nose) (there is also a sign to the station). Climb up to the bypass, cross with care and continue straight on down the road across the marsh.

This eventually crosses the preserved Dean Forest Railway at Lydney Junction station – formerly the Severn & Wye Railway – and completes the collection of six railways.

Beyond the road dips briefly and then rises to Lydney station.

Lydney station is generally well-used despite being so far out of town and so excessively utilitarian. Shelter on the Up platform consists of an airy plastic awning for bicycles and an unappealing small stone hut for passengers. On the Down platform the original 1850s shelter has been modernised by having its roof replaced and its windows sealed up. The signalbox was demolished in 2012. Seating outside the shelters is limited. Otherwise it is a welcome sight after nearly 30 miles, though not quite as well as the Turbostar that will eventually appear to provide transport home.