Reed Knight’s Experimental 1903 Semiauto

I saw this rifle when I visited Reed Knight’s Institute of Military Technology a while back, but didn’t have a chance to examine it. Well, he took it off the wall for a video segment with Jerry Miculek, and it’s definitely worth watching:

As I recall from asking about it at the time, there is no definitive evidence of exactly who made this conversion or when it was done, although the quality of the work is excellent. Springfield Armory has documented the attempts at developing a semiauto 1903 conversion, although they may not have included designs that failed to gain enough official notice (you can read their history here, in PDF form). Of the three designs that were recorded, this could theoretically be either a Young or a Hammond (same Grant Hammond who designed an unsuccessful challenger to the M1911). The third, the Smith-Condit, is definitely not this rifle (I will be writing up an article on that design next week). I have been unable to find any description, drawing, or photo of the Young system, so I can’t comment on it. The Hammond design was patented (download the patent here), but it is difficult to say if that matches the rifle in Knight’s collection, because details often changed between patent drawing and working model, and because of the lack of mechanical detail shown in the video.

Around the turn of the century there was a tremendous amount of interest by many governments in converting bolt action rifles into semiautomatics as a way to get the new technology without having to scrap all their existing rifles and buy new ones (a similar interest existed 50 years or so earlier, to convert percussion guns to use metallic cartridges, and before that to convert flintlocks to percussion locks). Most of the attempts had about as much practicality as this one, but most did so with a rather lower standard of workmanship.

To the best of my knowledge none of these types of conversions were ever accepted and mass-produced by any government or military force as an infantry rifle (though the Australians did make a significant number of Charltons as light machine guns). Here are a few other examples of the idea that I have covered previously:

Snabb semiauto conversion

Steyr M1888/90 semiauto conversion

Howell automatic rifle

Charlton automatic rifle

Huot automatic rifle

P14 semiauto conversion



  1. There is simple reasons why designs like this fails.
    The straight-pull bolt-action are easier to convert (only rearward – forward movement) but the repeating rifle were designed to be operated by human, the mechanism use much more crude force in shorter time so the weapon will wear down more quick than repeater.

  2. When I see this video it makes me think: how much easier had it inventors in times back. Something like this would be absolute no-no based on liability grounds of today.

  3. While absolutely fascinating from an engineering and technical standpoint, these conversions appear either (A.) Totally impractical for field use, and/or (B.) such a radical re-design of the parent rifle that it would negate the cost saving aspect of the concept, for the most part, as opposed to developing an entirely new system (save, perhaps, the recycling of barrels, stock fittings, and other small parts). Obviously, the process isn’t nearly as simple as cutting and fitting a breech-block into a Springfield (or other) muzzle loading rifle. However, when examining the various steps taken to make this rifle “safe” to operate (I’ll pass, thank you), and the progression in which the modifications are required, one can see this as a good way to highlight potential problems in the development of a new design. (Maxim’s experiments with the Winchester lever action come to mind.)

    • Looks like there were always some people ahead of ‘official’ engineering. Who were those people? I believe they were from user’s ranks; technicians, armourers and such. Some of them became serious full time engineers later. Names like Aimo Lahti come to mind.

      • Yes, you have to start somewhere. Often, the first person to ask the question, “What if…?” doesn’t profit from their curiosity and tinkering, but they open the doors for those that would. Both of my grandfathers were inventors and mechanical engineers with several patents to their names. One was extremely successful, but died at a young age (plane crash). The other saw his ideas copied by others and was never as successful, financially, but he did live to a ripe old age.

  4. There is another real oddball conversion that very few people know about and that is the Italian “Scotti X” semi auto rifle that is based on the long model 91 rifle. A considerable amount of the original parts were kept along with quite a few changes. This came out in the mid 1930’s or so but it was not as successful and the total run of these was less than 300 pieces.

    I have held one before and it is a super neat weapon that was a case of to little to late!

    • A Scotti was sold at auction here in Mattoon IL last Sunday. Unfortunately, the people who knew what it was were not sleeping, so I did not get it. How that rifle ended up here would be a great story.

  5. It’s interesting that although the gun appears superficially to be faithful to the 1903, eben down to the skinny receiver ring and reinforce pad over the right raceway of the ring.

    The machining of the receiver bridge is completely different, with the concealed extraction cam becoming a long exposed bolt handle turn down cam (best views are at about 3:50 and 4:40).

    I wonder what butchery went on inside the gun and how many machining operations were added to the already ridiculous number of busy work machine operations that the 1903 required.

    With a gun which was so expensive in terms of machine set ups, operations and skilled machine operators, as the M1903 was, it’s a good job that the united state stayed true to George Washington and Thomas Jefferson’s warnings about the remaining well clear of the vicissitudes of European politics and entanglement in foreign alliances…

    • When you say about machining time, the Pedersen Device comes to my mind as it need minimal alteration of the M1903 rifle – only adding hole to ejecting spent cases.

      • True, but technically the Pedersen was more like an SMG, except that it was semi-auto only.

        I’ve often thought that instead of building new M1903A1s with the Pedersen mods, a better course would have been to convert the early “single heat treatment” receiver M1903s that were in store. They were more than strong enough for the .30 Pedersen round’s pressures, a permanent conversion of them to semi-auto would have eliminated the problem of parts being damaged or lost in a front-line “changeover”, and they could also have been cut down to the length of the early Springfield cavalry carbine, making them lighter, handier, and much more maneuverable in CQB. Say, in a trench raid.

        Add a fire-selector switch, and the result would have been very close to the later Beretta “moschettos” that led to the Model 38. And even closer to the post-WW2 Dominican Cristobal .30 carbine.

        One of the “Great War’s” missed opportunities on the Allied side.



  6. Obviously a contraption like this would be impractical for field use, but it’s remarkable to me that having developed something like this around 1906, it still took 30 years to get the M1 Garand adopted.

  7. Unpractical side of these kind of coonversions is creating space for automaticaly reciprocating breechbolt without harming the user. In case of refiguring the design into what it needs, happens another rifle we say “Autoloading”.

  8. As stated, most of these conversions consisted of an added gas-cylinder and piston assembly, usually with some sort of cam-path added to turn the bolt as it was forced backward. The Swiss “straight-pull” rifles (Or as I like to think of them, “phony straight-pulls”) already had such a cam-path built into their mechanisms to turn the bolt, for unlocking/locking and for good primary extraction.

    The American Lee and Austrian Mannlicher Model 1885, of course, were “true” straight-pulls, with cam-type locking. The Lee’s angled bolt movement would probably have been more awkward to accommodate, but the M1885’s straight bolt path plus under-pivoted cam would seem to have been a natural setup for gas-operation. However, both were a bit weak on primary extraction, which would probably have resulted in the sort of case-head separations described in the Springfield report.

    It’s interesting to note that in Germany and Austria, Mauser and Mannlicher never went this route on their self-loading rifle projects. They began with recoil operation, went to gas operation on principles not unlike the Mexican Mondragon or the later U.S. Garand, and by WW2 Mauser (after a stumble with the muzzle-gas-cup system on the Gew 41M) ended up developing what was to become the Heckler & Koch roller-locking recoil system (StG 45M).

    As the old saying goes, if the experts don’t do it that way, there’s usually a pretty good reason.



  9. I have handled a Grant Hammond version of a semi auto Springfield. It came from one of his investors. Didnt look like this. I also looked up his patents for the same thing. Totally different mechanism. Had the opportunity to handle a Grant Hammond 45 at the same time. Very unusual firearms.

  10. A less kludgy version of the Howell rifle! More scary with the 30.06 cartridge though. I’d love to have a Howell rifle. My birthday is in June

Leave a Reply

Your email address will not be published.