The Ironbridge Gorge Museum's collection of photographs now contains a considerable number which depict boat types formerly in use in the Gorge, few of which have been published. The more important items are cited at the end of this note, which also draws upon some of the better line drawings. They reveal almost without exception a type of barge unnoticed by historians. The better known version of the Severn trow (as in "crow", c.f. Shropshire trow as in "cow")

is conspicuous only by its absence. While this is not unexpected, granted the probable dating of the collection (1860-1900 say) and the economics of transport at that period, it is remarkable how quickly the old up-river boats disappeared from the collective memory of maritime writers.

This occurred to such an extent that a renowned scholar could write in questioning terms of, for example, trow topmasts being set up behind the main masthead - as clearly seen in many eighteenth century drawings. This is indeed an unusual feature, but a photograph now vindicates the accuracy of some drawings, at least. This feature also serves to indicate the great differences which occurred between the up-river boats of the late nineteenth century (themselves remarkably like, but more heavily built than, the boats of the lower river in the eighteenth century and the popularly received connotation of trow as an estuarial and coastal vessel, fore and aft rigged, "carvel" built, often "boxed", and of considrable draught and solidity. Since most features of the up-river trows are different from their estuarial descendants, I shall not attempt to compare them directly in this note. (The classic source for more general information is Grahame Farr's "The Severn Trow", in Mariner's Mirror, Vol.32, 1946, pp66-95. Earlier items in Mariner's Mirror are regrettably unreliable - describing, for example, the great traffic coming down the Severn from the coalmines of Worcestershire in the century before the opening of the canals, and locating the Bower Yard at Shrewsbury; or if you prefer, Bristol !).

The photographs show a class of boat which from other evidence may be "barges" rather than "trows" (there is no satisfactory definition to distinguish trow): many vessels registered at Gloucester after 1850 under local owners are " barges", and these were arguably among the larger craft in use. Those photographed are demonstrably of the order 65-70 feet long, 14-16 feet wide and all very shallow (Registered lengths were up to 80 feet, and "depths" were frequently less than three feet). They are clench built, and the individual strakes are exceptionally wide - the width after cutting to shape is as much as 16 inches, with bevels of about 30 degrees on the plank edges at the bilge, also unusual. The hulls are clearly round-bilged, and parallel-sided for most of their length. Their stems are perhaps straighter and more vertical than drawings suggest; their transoms are upright and a very shallow "D" in elevation. The sterns are formed without a concave tuck, but with a skeg formed in the deadwood. There are mouldings as rubbing strakes; capping rails; and small washstrakes at stem and stern. All have completely open holds (with in one case side cloths rigged); there are the usual short working decks fore and aft, perhaps fifteen feet long. The bulkhead at the forward end of the hold, at least, is not complete; there are small hatches on each deck; and in several cases a suggestion of a stove chimney.

The shape of the hull leads to conjecture about the run of the planking: the Hartshorne drawing in the collection and some folklore from down-river boats suggest that the floors could have been laid with carvel planking, merged into clinker planking on a different alignment at the bilge, to get round the problem of the planking runs at the very full, shallow bows. It would certainly be reasonable to employ flush planking on the flat floors in such a shallow, rocky stretch of river.

Rudders are massively constructed, and because the barges were operated primarily by towing from the banks, at low speed and often shallow draft, they are very long - possibly as much as eight feet. The tillers are correspondingly long to reduce the effort and were evidently cut from carefully selected timber to elegant curves. The top edge of the rudder blade was also curved, in a "S" form. The sheer size of the rudder makes it unwieldy, and problems must have arisen in eighteenth century locks.

The single masts, usually with one square-sail yard, are set up by simple rigging: two shrouds and a relieving tack (set up for alternative use as cargo gear ?) each side is typical, all necessarily led aft of the tabernacle (of which little can be seen). No ratlines can be seen - largely superfluous if the whole mast can be lowered (at least if there is no topsail fitted), but nonetheless are seen in several drawings. There is no evidence that the heels of the trows' masts were counterbalanced (as for instance in wherries) and the load in the stayfall tackle is consequently of the order of half a ton when the mast is lowered - heavy work, and the forestay has massive sister blocks leading to a simple timber-framed winch on the foredeck. (There is also a heavy timber windlass in the eyes of the vessel for handling an anchor or warps). Backstays and running rigging to the yard appear from what evidence there is to run to timber-heads set on the quarters: but it must be said that the rigging is not generally clear - only two photographs actually show sails, which are not set. The standing rigging in at least two photographs is seen to be formed of a long-link chain, which Stuart Smith considers to be colliery winding chain - doubtless near the end of its life !

One photograph and several excellent drawings show topmasts, set up behind the lowermast, probably to gain a couple of feet clearance under old stone arch bridges, which would often have to be negotiated in high water conditions; but possibly for other reasons also. For example, to facilitate the lead of the mast-head tow-ropes, led high to allow the rope to clear obstacles and scrub on the river bank (the heavy tree growth in the banks along the old towing paths cannot be a long-standing feature !), or to be able to lower the topmast without first lowering the main yard. The topmasts could be struck independently of the lowermasts, perhaps to reduce the load on the stayfall tackle when raising the masts, or for clearance again.

The square sail yard is slung centrally and is generally seen stowed fore and aft between the shrouds, rather than crossed. This might indicate several things: habit formed of long necessity when lying against warehouses or other vessels, clearance on bridges, general disuse of sails, or normal use of the spar as cargo gear. Again, there is too little evidence. Strangely the topsail yard is stowed in a similar fashion, but below the main yard, which would have called for some intricate handling to get it across the forestay: even the main yard would require some direct handling to stow it between the shrouds.

Little can be seen of internal structures: the floors would be ceiled (certainly, except that one drawing shows two barges both with unceiled holds !), and a heavy sheer-clamp can be distinguished. There is no ceiling between the bilge and the clamp and the frames are seen in the gap. These are single timbers regularly spaced in the hold.

All else is conjecture: a light keel-plank, massive keelson, and framing on a largely ad-hoc basis, probably. Some at least of the later down-river trows were also lightly built; although the planking was flush-laid in these latter boats, it is quite clear that the frames were not pre-erected as in strict "carvel" fashion: much of the framing was fastened only to planking which must have preceded it. The frames are single, and (away from the crutch and cant timbers at bow and stern) comprised floors cut from straight timber with only slightly up-turned ends, futtocks to form the bilge, and top-timbers from there to the capping rail, with timber-heads and rail stanchions added separately. As the size and range of trows increased the frames and fixings became correspondingly heavier, up to double frames in the style of large ships. Examples of various styles can be seen in hulks in the Lydney-Sharpness reach.

The following is a note of the more important items in the collection. A word of caution is necessary: many photographs are clearly different views of the same barge; the numbers of different barges represented is unknown. Few can be dated with any confidence.

Science Museum 33/39. Barge and Iron Bridge. Bow view with considerable detail. (Seen from a different angle in A1476 and probably in A902).

A9. "The last Severn trow" (William, of Broseley), with a punt. A fine study from the port quarter, seen opposite Coalport. The punt is of the old Severn style, having deep sides, marked sheer, and undecked - the type can still be seen in Gloucestershire. The barge is also seen broadside-on in A623.

A11. Coalport ferry. Appears as a small flush-decked barge, and in A116 with curious repairs to the clinker hull. Also seen in A569, and A627, and in a poignant distant view in A1481.

A282. View of a barge at Ludcroft Wharf, from Benthall Edge.

A352. The classic view of the Severn Wharehouse with three large vessels and a punt. The nearest is of much larger burthen than the other vessels seen in the collection; one of the barges is completely unrigged.

A696. Fine view of three barges at Ludcroft Wharf, with considerable detail.

A706. Intriguing view of a barge with a bulky cargo contained by hurdles supported on spare sweeps.

A598. Barge at Ironbridge. An eceptionally clear view of a barge loading a cargo of bricks.

A2312. Three barges aground at the Bower Yard, including the only topmast in a photograph.

There are also several valuable drawings in the collection, including:-

A1911. A sketch showing Ludcroft Wharf, Bower Yard and two barges.

Harsthorne drawing, c.1858, from Northants CRO, of barges and boats at Bower Yard. Finely drawn, showing repair work. A small boat is shown fitted with a large winch, for no obvious reason.

A56. River scene, 1804. The finest of all trow drawings, it is from an original in Tewkesbury, where the scene is set. It epitomises the river prior to canalisation.

J.Fidlor's "Ironbridge", c.1850. Apart from the interesting details of the barge in this view, there is an intriguing variation of the "traditional" Ironbridge coracle.

We are clearly left with many unanswered questions and a long way from producing a reliable, comprehensive, drawing of a barge or trow of this period. Internal structure, underwater shape and details of rigging are matters largely of conjecture. The latter is profusely rperesented in countless drawings in the Museum's collection and elsewhere, but is generally of doubtful reliability and presents more questions than answers. If we accept the ratlines of Fidlor's "Ironbridge", for example - which certainly shows the topmast abaft the lowermast, should we not also accept the tumble-homed, circular coracle ! We must re-write the history of the Ironbridge coracle, too.

There is a distinct gap between the light and lively eighteenth and early nineteenth century vessels with their distinctively spoon-shaped bows and the barges of these photographs; just as there is between the clinker barges and the later trows of the canalised river. There is also a marked lack of rigging in these tethered Shropshire barges compared with those of the Tewkesbury drawing, indicating a gross difference in usage. The conclusion has to be, in fact, that these barges are the last few dinosaurs from a past age, effectively stranded by the unimproved shallows above Stourport, and finally displaced by the warm-blooded railway.

One final thought: just how significant is it that the known photographs of these large clinker barges are virtually all confined to the Gorge ?