One of the persistent aspects of the McClellan was the difficulty the mounted services had fitting this hard-frame seat to a wide variety of horses. From light cavalry horses ( near ponies in some cases ) to large heavy horse artillery mounts. The main feature of the saddle that was capable of some adjustment was the rigging, and so it was experimented with. Movement of the rigging ring forwards and backwards was attempted with the Godfrey modification that first appeared in the early 1890s, and first issued with the M1893 saddle. After about ten years of that uni-dimensional adjustability, the desire to be able to move the ring vertically resulted in the Model of 1904.
M1904 – Pattern I
This particular classification was made entirely upon the rigging change. Every other aspect of the saddle was identical to the previous M1896 in russet leather. Titled the Model of 1904, this new saddle first appears in 1906, with the new rigging and sheepskin padding for the underside of the sidebars. It’s a bit of mystery that actually M1904 saddles don’t show up until a couple of years after the rigging design was adopted – other than perhaps it missed a ‘production funding window’ in 1904/05. Still looking for info on that one.
To obtain the vertical adjustment required, a few simple changes were made to the 1896 design. The long quarterstraps were terminated at the edges of the seat, with halter squares sewn and riveted into the ends. Long side quarterstraps with roller buckles and sliding loops were then arranged through the halter squares and rigging rings. By shortening the strap, you could move the ring up, or by lengthening the strap, move it downward. Adjusting the ring position with the strap could move it forwards and backwards, same as the M1896. The rigging ring was made in a slightly more efficient shape, of a round disc. These were usually made in a layer of two pieces of collar leather, and lined with sheepskin. All metal hardware was also finished in brown jappan for enlisted men, and bronze for officer equipments. This only lasted for a few years until it was replaced by a more durable dark bronze chemical finish.
All other details of the M1904 Pattern I are the same in regards to the previous russet M1896.
This particular pattern is quite rare, in that is was only made for about 18 months or so, before additional alterations were made to the M1904 pattern. Most of these Pattern I saddles date from 1906, and the first few months of 1907.
M1904 – Pattern II
In mid- to late-1907, a couple of key design changes were made to the M1904 pattern – changes that are visually apparent to justify a new pattern classification (at least for the collector and student).
First and foremost, the stirrup loop was moved forward on the tree by 1 1/4″, much closer to the pommel. There are a few examples out there that show some evidence that a small number of the nearly completed saddles were taken back and modified to the new standard. All appear to have been done in 1907. Looking at Ordnance drawings, you see that a slight reduction in the width of the gullet and sidebars of the tree was also made. This is observable when comparing earlier 1896/1904 Pattern 1 trees to the later Pattern II. The difference was about 1/2″, and remained consistent in the arsenal drawings for the next ten years.
Size Variations: However, while the drawings were consistent, a lot of collectors have seen and have examples of varied dimensions – some are exceedingly close to the specification, and some have been found to be upwards of an inch wider. While the exact reasons have yet to be found, it is known that Rock Island Arsenal was not adverse to acquiring saddletrees from commercial sources, and may have been lax in enforcing the specifications.
The entire undersurface of the saddle was covered with one-half inch thickness sheepskin. This was first sewn to the bottom covers, after which the outer lower edge was sewn into the cover seam when that was closed. This is by the regulation description of the saddle – actual specimens show that many appear to have been made without this sheepskin. The method by which the sheepskin was applied on these earlier saddles made it quite easy for them to be removed without leaving tell-tale stitching holes. The rigging ring safes were universally lined with sheepskin.
A key identifier for early M1904 Pattern I and II saddles was the use of jappaned iron tubular rivets and caps for certain very defined locations. On the seat, these were used for reinforcing the folded and stitched laps that held the rigging squares. The stirrup hoods were also attached with these same types of rivets and caps. The iron tubular rivets and caps were highly prone to rusting, and the limited locations where these were used are rather puzzling – they were clearly inferior to standard brass rivets and burrs. All jappanning of brass/bronze parts was replaced with chemical bronze/black finish – somewhere between June 1908 and February 1909.
Stirrup hanger loops were iron, and this characteristic will clearly id a M1904 Pattern I and II, made from late 1907 to early 1912.