Amazingly though, the last one being withdrawn wasn't the end of the story.  See this thread for  the second life of one (or part of one) of the class. https://www.rmweb.co.uk/community/index.php?/topic/156057-identification-of-a-mystery-outside-framed-engine/

Also https://www.flickr.com/photos/14581588@N05/4735888892 and http://www.historywebsite.co.uk/articles/Railway Gazette/OWWR.htm

Something that's struck me today, looking at these and trying to find pictures of them in their later careers - can anyone help - is what an incredible rag-bag of 0-6-0 freight engines the early GWR possessed. I haven't looked at numbers, but there seem to be many more than I imagines, quite outnumbering the GWR constructed early 0-6-0s. They were pretty much all rebuilt or renewed at Wolverhampton, tending too merge towards a common style, especially with GWR boilers fitted, but all different, all different. No wonder that Armstrong's class were called Standard Goods: there was so many that were anything but. I imagine the same is true of 2-4-0s and 2-2-2s, although I haven't really explored, but the 0-6-0s really struck me today.

I think a desire to standardise parts was widespread. Gooch rejecting locomotives that didn't come up to specification is a often noted early example, and there are lots of examples of 19thC Locomotive Superintendants attempting to standardise parts across two or three new classes. I think the credit due to Churchward is not that he founded the concept, but that he was the best at it, and by commitment and vision did a far better job than anyone else.

Another part that we perhaps tend to overlook is how important competent management was. Railway workshops - perhaps all workshops - had a definite tendency to revert to doing one job at a time with little regard to the big picture when it seemed convenient. Perhaps Churchward's biggest achievement was not that he introduced a number of classes with standardised parts, but that the (G)WR was still operating to those principles thirty years after he retired. I don't have too many non GWR books - only so many bookshelves even in a house where we have thousands of books, but one is Cox' 'Chronicles of Steam'. Around p48 Cox is talking about standard boilers, and reading between the lines some of us may judge that the LMS draughtsmen pushed back mightily on Stanier's ideas to refit LMS locomotives with a range of standard boilers by claiming it was just too difficult and not worth it. We might note their contemporaries at Swindon managed, but doubtless Stanier judged which battles were worth winning.

-----------------

Incidentally, am now going through RCTS trying to produce a rational list of pre Armstrong 0-6-0s. Damn its going to be a struggle. Did I mention management - locomotive policy in the early days was (predictably I suppose - it was early days) chaotic and loans, rebuilds, renewals, reconstructions, make it a very challenging task. Makes me recall why I tended to fight shy of that era in the book, and also increases my admiration for the RCTS author(s) who had to try and make some kind of order out of what must have been a spectacular mess of source documents from all over the place. In the past I've wished the author of RCTS Part three had used a more tabular and less narrative style describing classes, but I'm starting to realise that, especially in a pre word processing age, doing so would have been an almost insufferable burden. What a filing system he must have had to create!

There's another angle to this, which is the blurring (probably not the right word) of Swindon and Wolverhampton product. RCTS does a valiant job in trying to describe the complexity of the 645 class boiler situation, but suffice to say the impression I came away with was one of boilers being swapped between the different works. I wonder if there were trains going between the two places with boilers on Crocodiles?

Coincidentally, I was sorting through some Armstrong Goods pics, and this one jumped out at me: for what is ostensibly a Swindon-built loco, everything above the footplate is distinctly Wolverhampton.

http://www.gwr.org.uk/armstrong-goods/391-worcester-small.jpg

5 hours ago, Miss Prism said:

 the impression I came away with was one of boilers being swapped between the different works. I wonder if there were trains going between the two places with boilers on Crocodiles?

In RCTS (pC24) they mention No 245 having had at different times a Swindon boiler put on at Wolverhampton, and a Wolverhampton boiler put on at Swindon. I suppose one might imagine a Swindon boiler swapped off a locomotive at Wolverhampton, repaired, and put on the next repair, but I get the impression that in the 19thC boilers were not swapped regularly in the way they were after Churchward. A train of various boilers being taken from one factory to the other would be something different for modellers! More likely, I suppose, that they would go one at a time on normal service trains, but its a fun thought.

I agree that 19th century boiler swapping was not regular, because the dimensional interchangeability had not been so well established at that stage, and besides which, trains of boilers to-ing and fro-ing between Swindon Wolverhampton smacks of bad planning, so I guess it was a comparatively rare singular event.