Why so few 1896 rifles?

[Dick Hosmer]

Model 1896 rifles were produced from early 1896 to early 1898.

And, both rifles and carbines will be found with two different 1896 markings - "1896" and "Model 1896". The earliest "1896 style" carbines (and a very few rifles, likely including the elusive cadets) will be dated "1895".

For nearly 40 years I have owned the highest-known "1896" specimen, a rifle, 37045.

The "1896" stamp, found on both rifles and carbines, is the least common date mark, with only about 7,000 so struck.

[butlersrangers]

FWIW - "Model 1896" is stamped very deeply on many Krag receivers. This causes a lot of unknowledgeable 'Owners and Sellers' to misread 1896 as 1898. Of course, the two models, 1896 and 1898, have many differences. But, I have seen a fair number of model 1896 Krags mislabeled on auction sites as '1898 Krags'.

By rough estimate, I believe approximately 84,335 model 1896 Krag rifles and carbines were manufactured. By contrast, approximately 389,908 model 1898 rifles, model 1898 carbines and model 1899 carbines were manufactured. The difference in numbers, between model 1896 and model 1898 Krags, is over 300,000.

Less than 1/2 million U.S. Krags of all models were manufactured. Model 1898 actions (along with model 1899 carbine variant) certainly make up the vast majority.

[Rick the Librarian]

I remember looking for an 1896 rifle maybe 2-3 years ago and it was far from easy. And often the rear sight or other feature has been changed, when you do thing you've found one. But keep looking - they're out there!

[FCPDAVE]

My 1896 is serial no. 108743. I read that 1898 production started at 109,100 in 1898 - is that correct? thanks

[butlersrangers]

I believe Franklin Mallory found model 1896 Krag rifle, #109020, the highest model 1896 serial number to be reported in his research.

Model 1898 Krag rifle, #109128, was the lowest model 1898 rifle serial number, in Mallory's data. (Destroyed 5/17/1929, New Cumberland Guard).

I imagine it is very possible there could be some overlap in serial numbers, as 1896 actions were in parts bins going through manufacture and assembly into rifles and bins of 1898 actions began the process.

Krag #109100, as the model change, would fit pretty neatly between highest model 1896 and lowest model 1898, reported in Franklin Mallory's work.

But, there was a War going on when the transition was occurring, during June-July, 1898, and things don't always happen cleanly!

[Dick Hosmer]

I believe that Joe Farmer has lowered the breaking point into the high 108K range, and also - while overlap does apparently occur at other breaks, at this one there is a distinct physical change in the numbered part itself, a notable distinction. Of course, I guess they could have annealed the receivers, turned off the bolt shroud, and re-hardened them, but that is a lot of work compared to just paying attention to the numbering - and - I have never heard that such work was attempted.

[butlersrangers]

If a model 1898 Krag were found in the 108K range, that would prove some overlap occurred. Model 1896 Krags are recorded that are above 108K.

[03collector]

Both the 1896 and 1898 rifles were being made at the same time. In the beginning of the production run.

[5MadFarmers]

I believe Franklin Mallory found model 1896 Krag rifle, #109020, the highest model 1896 serial number to be reported in his research.

Model 1898 Krag rifle, #109128, was the lowest model 1898 rifle serial number, in Mallory's data. (Destroyed 5/17/1929, New Cumberland Guard).

I imagine it is very possible there could be some overlap in serial numbers, as 1896 actions were in parts bins going through manufacture and assembly into rifles and bins of 1898 actions began the process.

Krag #109100, as the model change, would fit pretty neatly between highest model 1896 and lowest model 1898, reported in Franklin Mallory's work.

But, there was a War going on when the transition was occurring, during June-July, 1898, and things don't always happen cleanly!

In a thread where the difficulty in telling "1896" from "1898" is discussed it's escaped you that that problem might possibly have existed at the time? Really?

"A sample size of one is no sample at all." Two or three are unlikely to be misread consistently but one?

The interesting thing about books is reading and understanding them is only the first two parts. They're designed to educate so you can make additional leaps of logic. Else-wise you're just a Parrot for whomever.

I believe that Joe Farmer has lowered the breaking point into the high 108K range, and also - while overlap does apparently occur at other breaks, at this one there is a distinct physical change in the numbered part itself, a notable distinction. Of course, I guess they could have annealed the receivers, turned off the bolt shroud, and re-hardened them, but that is a lot of work compared to just paying attention to the numbering - and - I have never heard that such work was attempted.

The overlap in the 1898/1899 receivers would tend to indicate that the serial was stamped before the model. Overlap in the 1896/1896 receivers would startle me not at all.

The SRS and such were gathered at the time. Written records. That makes them susceptible to two, not one, transcription errors - initial and the SRS people. I'll give you a very easy to find example of the first. In one of the Ordnance Notes the serials for trapdoors in a unit are given. With that document in hand I put them into a spreadsheet. The same number showed up in two different companies. I'd say that's likely a period error. One of the Marx Brothers films has Chico asking Harpo: "who are you going to believe? Me or your own eyes?" Today, as we speak, are sites on the Internet with guns for sale. Many showing clear images of serial numbers. Snap, file, snap, file, snap, file. No need to worry about transcription as you can go back on the odd ones and recheck. In no time at all you'll have a better idea of where and how the numbers lie than anyone before you. Just takes a bit of effort. In the olden days the guys had to go to gun shows. For years and years and years. The Internet makes more guns available for review in a year than all the gun shows in the 1970s combined. It's trivially easy today.

What will be found is that what those "ranges" are good for is "general usage." "Carbines of this model tend to be in this range." As soon as a specific number is mentioned all bets are off as "a sample size of one is...."

Might I mention again that, sitting on the floor near my computer, is a Krag with the receiver stamped "1894" and sitting right next to it are two with "1895" stamps. The "1895" stamped ones being lower in serial? Mis-strike? Boo boo? Doesn't matter. What it is is definitive establishment of an "1894" having a higher serial than an "1895." RIA bayonets from 1917-1919 sufferer the same problem so now we're up to the O.D. doing that twice.

Back to the OP's comment on 1896 rifles. If you mean "rifles" specifically I'd mention that, today, there tend to be more "carbines" than one would expect....

[5MadFarmers]

Both the 1896 and 1898 rifles were being made at the same time. In the beginning of the production run.

That's insightful. Take it further to men working receivers at benches and it goes to the next level. "After operation X the receiver goes to the serial stamping machine." How do we know that Bill, working an 1896 receiver hungover, finished that operation after Pete, at the next bench working one of the 1898s, was done? They'd hit the serial stamper out of order. I doubt there was a fine. I doubt somebody walked a queue reordering the dudes.

So, yes, it seems likely.