Context: a personal overview of NG modelling.
Many of the ideas in this article were originally published in Narrow Lines 147 as 'A systems approach?' but this article has been extensively revised from the original version. All the photographs are new.
It would have been nice to illustrate this article with photographs from the layouts referred to but I do not doubt that copyright laws prevent this. Brief thumbnail sketches have been attached to the bottom of the article.
Many years ago an editorial in Narrow Lines posed the question as to whether there were, 'any big layouts left out there in narrow-gauge land’. The question struck a personal chord, as I had had similar thoughts myself. Looking through the mainstream British modelling magazines one could be forgiven for thinking that the majority of domestic NG layouts, especially in 7mm scale, are small, compact affairs, designed to take advantage of the reduced dimensions associated with the NG prototype. Whenever larger NG layouts are featured they tend to be found in clubrooms or to be aimed at the exhibition circuit, rather than finding use at home. When one widens one’s horizons, however, and ‘crosses the pond’, the situation is rather different. Given that most US states seem to require houses to be underpinned by a basement there is plenty of space for the average modeller to build a large NG empire. The Colorado three footers are a popular choice and many such layouts have featured in Model Railroader, although Ophir trestle has been rather overdone!
The editor originally cited, as his exemplars, the Craig & Mertonford, the Augher Valley and the Madder Valley, each a famous layout dating from the 1950s or 1960s, although, given that we are discussing the NG, it should be noted that the Madder Valley was nominally standard gauge. As time has moved on, a younger generation of modellers, and those from overseas, will probably not be familiar with these layouts. It would have been nice to include photos of each of these layouts in this article but this would, no doubt, breach copyright laws.  For the benefit of those unfamiliar with these layouts I have appended thumbnail sketches of each of them, along with the approximate periods of their publication in the modelling press. Two other layouts, which came slightly later but similarly inspired me, and a generation of NG modellers, were the Aire Valley and the Stronalacher Railway. Examples of the latter system’s rolling stock have been preserved and can occasionally be seen on the OO9 Society’s stand at major exhibitions. The Carrabasset & Dead River holds a similar spot in the American pantheon, having demonstrated that N gauge running gear with HO scale bodies was a more viable option than HOn2.
These model railways were very different from those prevailing today in that they were conceived as complete systems in their entirety. There were no fiddleyards to hold spare stock, and every location and every piece of track was modelled. This is not to say that these systems were complex or even of vast extent. The Craig and Mertonford, as far as I can determine, never really consisted of more than two stations and the Aire Valley comprised three main stations, three isolated sidings and a branch terminus squeezed into an “average sized room”, actually a 12’x10’ space in the loft, albeit at the expense of any significant run between stations. For those interested, the Madder Valley can be viewed at Pendon Museum, and as a piece of modelling history, I can personally testify that it is worth the admission fee alone. J.H.Ahern, its creator, similarly managed to squeeze the mainline and a branch into various sized rooms in three different houses. (It ended up in a space approximately 17' by 10'.)
It is easy to forget that these layouts were the forerunners of numerous other NG layouts, but somehow the later ones were never quite so indelibly etched upon the memory. A random trawl through my magazine collection reveals just how many other unsung NG systems proliferated in the late 60s and the 70s and, for that matter, the intervening years as well. Nostalgia probably lends a rose tinted glow to our memories, but the seminal layouts were significant because they were pioneering  NG layouts, built before the advent of commercial NG models. Whatever the reasons, I am still drawn to stop and re-acquaint myself with these 'classics’ whenever I stumble across the relevant issues in my haphazardly stored magazine collection. I suspect that I am not alone in this. Each of these layouts had a ’personality’ and, in the cases of the Aire Valley and Madder Valley, also included a significant slice of the communities that they served. The introduction of Eggerbahn allowed other, less gifted, modellers to imitate these railways, with varying degrees of success, but the fundamental philosophy of the portraying a railway in its entirety was an enduring theme. One unfortunate side effect of the availability of commercial NG models was the rise of the 'rabbit warren' layout, usually ill conceived and badly constructed, by people with little or no knowledge of the NG prototype, which gained narrow gauge modelling a bad name. To this day the suspicion lingers in some quarters that NG modelling is somehow not 'proper' modelling, even when the models are scratchbuilt, in contrast to a more lenient disposition towards SG modelling using proprietary equipment simply removed from its packaging and otherwise untouched. These NG models were of course built in 4mm or 3.5mm scale but, given the current widespread availability of Bemo, one might reasonably ask where can a large NG system can be found today? One might simultaneously wonder where was the 7mm system using Fleischmann’s Magic Train as a basis? (The Magic Train sadly met its demise in 2008.) The products of Eggerbahn, Jouef, Minitrains and Liliput were no easier to Anglicise.
An Eggerbahn steam tram, one of the more credible models in the range. The introduction of Eggerbahn unfortunately allowed a lot of ill conceived layouts to be built and helped to spawn the 'rabbit warren' layout.
The answer to this conundrum lies in the more discerning 'mindset’ that dominates the hobby today. It was once common to find descriptions of SG layouts that ran to three or four stations and were 'closed systems’ in that they had no off stage hidden areas. Such layouts necessitated considerable compromise. Today that same space would be more likely to contain one accurately modelled station, fed from a fiddleyard. This sea change is perhaps the inevitable price to pay for the movement towards prototype fidelity, as is exemplified by the success of Model Railway Journal. The magazine unashamedly eschews the mainstream approach of competitors such as Railway Modeller, and propagates a deliberately finescale, and some would argue, elitist, approach. I don’t doubt that there are many readers who, like myself, admire the modelling contained therein, without attempting to emulate their contributors’ exacting standards. The wider philosophies underscoring the magazine have, however, been adopted by many less skilled modellers using the new breed of proprietary equipment, resulting in more railway-like layouts, with fewer of the abominations that can be found in older magazines. Where they have the space, British NG modellers have followed a similar path, rejecting the development of a cramped system in favour of one or two stations modelled in detail, connected to the inevitable fiddleyard representing the ‘rest of the world’.
Taking these factors into consideration, one other aspect of the sixties NG systems is apparent. They were freelanced. By that I mean that they were genuinely freelanced, in an unapologetically imaginary setting, rather than based upon a ‘might have been’ in a real setting. The Augher Valley, the Aire Valley and the Madder Valley existed in total isolation, not even condescending to connect with another railway, SG or NG. Their creators often based their individual models upon widely disparate prototypes, adjusting their relative proportions as necessary. This ‘flexi-scale’ approach, coupled with the ‘heretical’ creations of their own imaginations, is obviously at odds with the modern attitudes dominating serious modelling. In some respects it could be argued that this was an artistic approach to modelling rather than the scientific one, based upon well-researched sources, that predominates today. It is easy enough to understand why this approach became discredited. In the hands of a gifted individual a highly idiosyncratic but credible layout could emerge. In the hands of a journeyman, or worse, a dullard, a (literally) incredible self-parody resulted. Whatever the reasons, these freelanced systems, warts and all, often gelled to convey much more of that intangible ‘atmosphere’ than many of today’s meticulously observed layouts. 
An example of AFK flexiscale. A 60cm gauge prototype locomotive pulls a 75cm gauge coach tailed by an Irish broad gauge birdcage brake. By adjusting the various proportions the more outlandish effects can be ameliorated.
Continuing in this vein, it becomes apparent that successful narrow gauge freelancing usually has to be underpinned by scratchbuilding. Whereas it is possible to concoct a SG layout by using kits representing the products of the contract builders, such as the Gloucester Carriage & Wagon Co., Beyer Peacock or whatever, this is not really feasible for a NG layout. The widely differing track gauges and loading gauges of the narrow gauge companies prevented the development of a generic prototype, in Britain at least, although there was more uniformity in Europe where countries often, semi-officially, recognised a 'standard’ narrow gauge such as the 760mm in Austria or the metre gauge in France and Switzerland. The US similarly embraced the three foot gauge, with the notable exception of the Maine two-footers. It is difficult, however, to imagine a convincing layout based upon the superficially compatible British combination of the Festiniog and Lynton & Barnstaple, which both used the 1’11½” gauge but had different loading gauges and very distinctive locomotives and stock. The result of these design singularities is that, where kits are used extensively, a layout adopts the atmosphere of a particular prototype and it becomes difficult to present it as anything other than that. By scratchbuilding, and adjusting their prototype’s proportions accordingly, the sixties layouts avoided this trap.
Unfortunately this approach creates difficulties beyond the obvious one of the time commitment. Most modellers would expect their skills to improve over time leading to glaring disparities between their earlier models and those built much later. This is obviously not a practicable proposition if the system is to present a coherent whole. Collaborating with like-minded modellers, each of whom specialises in a different area, as has become more prevalent in recent years, circumvents this problem. With a freelanced system, however, this solution is not quite so easy to implement because there is no clearly defined prototype. Tony Koester discussed this problem at length in the Model Railroader, with reference to his Allegheny Midland. More recently he has moved to a prototype based layout, sidestepping these difficulties, (although this is merely exchanging one set of problems for another!) To be convincing a freelanced system usually needs to be built by an individual, otherwise one ends up with a ‘two headed camel’, i.e. an unsatisfactory horse designed by a committee!
Where does 7mm NG modelling fit into this overall view? None of the foregoing arguments in themselves preclude the development of a 7mm equivalent of the systems referred to earlier.  The most obvious constraint upon their development is the space within which to put them, as a 7mm layout needs roughly three times the space of a corresponding OO9 layout, or more likely four times once access aisles are taken into account. Modern British houses, so the theory goes, have insufficient space in which to construct a large layout, but this begs the question as to whether all 7mm NG modellers live in small houses. In any case where are the large home based, as opposed to exhibition orientated, OO9 systems today? There are, of course, plenty of modellers who are happily constructing their empire in the loft, the garden shed or garage, but relatively few ‘room sized’ domestic layouts have been described in the press over the years. One has to ask, why should this be the case?
I suspect that the construction of large systems is currently out of fashion partly because modellers can nowadays easily diversify into other scale-gauge combinations. In the era of the classic layouts alluded to, everything had to be hand built and the time and effort involved resulted in an emotional attachment to the models that was not easily forsworn. The proliferation of commercial On30 equipment in recent years parallels the sixties boom heralded by the introduction of proprietary OO9 marques. It allows modellers to dip into a scale, buy a few key items, build a layout and then move on, butterfly fashion, to the next project. This doubtlessly reflects the modern trend towards building 'disposable’ layouts, usually indistinguishable from their peers, with a 'shelf life’ of a few years. The next layout may continue the theme of its predecessor but, as the interests of the builder change, it is not uncommon to find that the successor layout is built in another scale and focuses upon a totally different prototype, often financed by the sale of the items bought only a few years earlier.
Different interests, different scales. The author dabbles in live steam. A modified Roundhouse 'Lady Anne' stands in the back garden.
This obsolescence is, understandably, most apparent in layouts built for the exhibition circuit but the underlying philosophy appears to be quite common, even for home layouts, and makes some sense, given that skills develop, as noted above. One side effect of this trend is that such layouts rarely mature, and are often characterised by ill conceived or tenuous backgrounds. Why go to the trouble of fabricating a plausible storyline when the layout will be in a skip within a few years?! An alternative scenario is that the modeller simultaneously engages in a number of projects and does not exclusively focus upon one to the detriment of the others. Given the relative increase in living standards and disposable income in recent years this is a more feasible option than it was in the sixties. I myself dabble from time to time in 16mm live steam, continental and British N gauge and HO9 trams, a situation that I would not have thought was that unusual, yet if there is no overriding focus it is unlikely that a large layout will result.
The site of the first ever published photo of the AFK. In truth the original was printed in black and white and taken from a much lower angle than this shot, although the subject matter was the same. A Billard railcar crosses the swing bridge over the creek to enter Eromarbordo.
In conclusion, it would seem that the clock will not be turned back. Large NG systems, particularly in 7mm scale, are likely to remain thin upon the ground, partly because few people have the space, dedication or inclination to build this type of layout. In some respects the AFK is a throwback, a missing link, in that it was conceived and began building before the advent of commercial 7mm NG kits, let alone RTR, as evidenced by my Association number (23). I would be dreaming if I thought that the AFK would remotely engender the general affection that any of the five named iconic layouts did, but I occasionally delude myself that the AFK bears the feeble embers of the torch that they lit in my youth. Enough of the waxing lyrical! You might find these views idiosyncratic or even iconoclastic but I suspect that the great classics of the past were a product of their time, much as each generation of modellers creates anew its own masterpiece layout. 
The iconic layouts
The Madder Valley was a nominally SG layout built by J.H.Ahern during the 1940s. Widely acknowledged as one of the first attempts, if not the first, to create a credible scenic setting for a model railway the layout was an inspiration for a generation of modellers. Built in a 14'9" by 7'6" room the line ran from Madderport to Gammon Magna with a branch leaving the main line at Much Madder to run to Gammon End. Model Railway Journal issue 75 (1994)  carried a comprehensive account of the layout and a list of all known articles and published letters, stretching from 1939 to 1963.
The Craig & Mertonford was created by P.D.Hancock during the 1950s. The line was sitated in Craigshire, an imaginary county to be found off the Scottish coast between Edinburgh and Dunbar. A rebuilding took place in the early 1970s which, in my opinion, rather destroyed much of the original's charm. Narrow Gauge Adventure, a Peco Publication of 1975, gives a thorough history of the line.
The Augher Valley was the layout of D.Lloyd, who later became the editor of Railway Modeller. This was perhaps the least well known of the layouts cited as it made a single appearance in the April 1964 edition of RM, although it was referred to in a reprise of OOn12 modelling in the Modeller Book of the Narrow Gauge (Peco Publications, 1986). This Irish line consisted of five substantial stations with the stated intent of extending the route to a terminus in an adjacent room.
The Aire Valley was documented in the Railway Modeller during the 1960s. Built in his loft by D.Naylor, the railway ran down a Yorkshire valley, reaching the sea at Saltaire. Traffic was provide from mines at Nethertarn and a branch to Stony Ridge. The layout was extensively reviewed in the 1972 and 1973 RM, with additional articles being published in 1974.
The Stronalacher Railway, built by D.Manders, was the subject of a five part series in Railway Modeller during 1970 and 1971. At the time of publication the line was incomplete although many of the intermediate scenes had been finished. A connection with the SG was made at Stronalachar before running down a Scottish glen to the sea at Portanrich, this being achieved by circling the 12 ' by 20' room twice.
Back to Main Menu