SEPTEMBER 2004

    Along with writing and producing a new book over the past few months, I have devoted a fair bit of thought to the 1:1 rendition of the north half of the CNR Palmerston yard which I am constructing in HO scale. The focus of this operating diorama will be the 650-foot l.c.l. transfer shed, which forwarded an average of 28 boxcars on a working day in the early 1950s (these in addition to the "straight cars" of merchandise which bypassed the shed). Specifically, I have been determining the nature of the boxcar roster I will need to replicate the less-that-carload-lot activities which occurred six days a week at this branchline hub.

    That the majority of the boxcars handling l.c.l. at Palmerston during the late summer of 1952 will be home road CNR is not in question. I plan on using a scaled-down cross section of the actual roster numbers from the January 1953 Official Railway Equipment Register (which actually describes the situation about three months prior to its publication). Rather, it is the foreign road boxcars--their proportions to the home road and their consitutional makeup--which have been the focus of my attention.

    Originally, I decided to amass a 100-boxcar roster, half CN, the other half all foreign roads (including CP). Based upon data from wheel reports, switch lists, photographs, etc., I have determined that a proportion of 3:1, CN cars to foreign cars, is appropriate for the Palmerston layout. That would translate to acquiring 75 home road and 25 foreign road cars for the initial 100 boxcar roster, if all cars were to show up equally in operations. However, that would not be an efficient use of the 100 cars. Disregarding all operations save l.c.l., of the 28 cars which will appear at the shed on an average day, there will be roughly 21 CN cars to 7 foreign road cars. With a foreign road boxcar roster of only 25 cars, that would allow for fewer than four operating days before cycling through the cars again. Taking 50 foreign cars allows for seven days of operation. Meanwhile, I have reasoned that the other 50 (CN) cars will be in sufficient number throughout their various types (40' steel, 40' single-sheathed, 36' single-sheathed) as to be inconspicuous. Even though the home road cars are recycled after approximately 2-1/2 days, they will appear to represent a vast fleet. While the CN cars would be in continuous service, the foreign road cars will be called at 1/3 the frequency.

    Well and good. So what of those 50 foreign cars? The concept of distribution of foreign cars by road name in North America of the 1940s and 1950s has been debated at length in freight car and model railway operating forums. When all the smoke has cleared, the data suggests that the distribution of foreign boxcars in general service on a given railway is roughly in proportion to ownership. That is, we can take the boxcar totals from the ORER and project them onto a model roster of foreign cars--50 of them, for instance. Applying the numbers from the Jan. 1953 Register to a CN layout, I came up with the following breakdown by road:

5 cars--NYC
4 cars--PRR, CP
3 cars--ATSF
2 cars--SP, MILW, B&O, UP, SOU and C&NW
1 car---GN, CB&Q, IC, RI, NP, MP, C&O, SLSF, L&N, WAB, ACL, ERIE, SAL, NKP, N&W, SOO, RDG, T&NO, DL&W, P&LE, GTW and GM&O

    Prior to attending last year's Prototype Modellers Seminar sponsored by Sunshine Models at Naperville, Illinois, I generated a list of a plausible 50 cars to acquire (some of which I already had), using the above breakdown and an analysis of the dominant car types within the respective railroads. This list complemented a similar list of CN cars, distributed among types. The end result was a blueprint for amassing a representive 100-boxcar roster for the late summer of 1952 on my CN layout.

    Now, I readily admit a preference for resin kits over plastic kits, generally because of superior quality of detail, not to mention the joy of constructing, painting and lettering my equipment. When one performs the necessary research before choosing a car, then purchases the kit and sets out to build it, the whole experience is a lesson in freight car history. Despite my preference for the CNR, I have spent many joyful hours studying the boxcars of other roads for my specific time period. I should also mention that due to my fondness for wooden equipment, I have been tempted to model an earlier time period such as sometime in the late 1940s. Armed with an appropriate ORER, that would be as easy to accomplish as the late summer of 1952. In time I may do that.

    For the present, however, I determined that just to amass the 100 designated boxcars in finished form, building a mix of resin and plastic kits, I would spend a minimum of 2-1/2 years. With no attention to anything else in the hobby--track, locomotives, structures, operation! The prospect of doing things sequentially did not appeal to me. I did not want to wait until building the 100 boxcars before beginning construction of the transfer dock, not to mention reconditioning several Ten Wheelers, laying the remainder of the track and many other aspects of this hobby. For that matter, the 100-boxcar figure was too rigid. I wanted to be able to pursue operations and such before achieving that total, and continue to expand the roster beyond it. It occurred to me that a better approach would be to gradually increase the roster of operating cars in the original proportions, by railroad and car type--to build this roster one car at a time, home road and foreign road in a 1:1 proportion. Should I wish to build more cars than required at a given time (batch-build), I could put them in service gradually as other car types caught up.

    Back to the ORER numbers. If you are contemplating such an approach to your own CNR layout set in the early 1950s, this blueprint may work for you. The sequential order of building CN boxcars is shown below. The leftmost column describes the car type; the second column indicates the respective numbers present for the time period in question. The rightmost column describes the car series, offers more specific breakdowns, and identifies appropriate HO scale models. The intermediate columns describe a pattern for continually building a representative cross section of CNR boxcars; simply work down each column in "waves". For example, the first four cars I would acquire, in order, would be a 40' steel car of 10'-0" inside height (Red Caboose or InterMountain), a 36' single-sheathed with a 5' door (Westerfield, Kaslo or the upcoming Life-Like), a 40' steel car with 10'-6" inside height (Branchline) and a 40' single-sheathed 1929 car (Sylvan, Funaro & Camerlengo or Kaslo). I would then proceed to the top of the next column and work downward, adding cars to the roster appropriately. And so on.

    All the while, I will be building the foreign car roster to match the CN roster. The "waves" of construction are described below. As we are dealing with non-CN cars, disregard the roster total for the home road at the top of the second column. The third column is an empirical sample of 700 actual boxcars in and around Palmerston in the mid-1940s to early 1950s time period. It is instructive to note that the proportions from the field study are reasonably well represented by the columns to the right, describing successive waves of construction to build a roster of 10, then 20, then 32, then 46 foreign road boxcars. Work your way down the columns as with the CN cars. The beauty of this approach is that one is not overwhelmed by grand expectations. A roster of boxcars can be built one car at a time to any size, all the while remaining representative of the actual cars seen in the Southern Ontario locale for the era.