The M1938 Carcano Bayonet

by Dennis D. Ottobre


Although most books identify only 4 variants the Italian M1938 Carcano bayonet it is actually found in 5 distinct variants.  The first 3 are purpose built with folding blades and 2 other variants exist with fixed blades.  One is purpose built and the other is assembled from components of folding blade variants which are at times mixed yielding numerous possible sub-variants.  Figures 1 and 1a below show the 5 variants in order from top to bottom.



                                                                                         Figure 1                                                                                                      Figure 1a

The first variant is often referred to by collectors as the "Latch Lock" due to it's unusual lug catch arrangement being in the form of a latching hook that engages a cutout on the right side of the bayonet lug.  The catch mechanism proved prone to breakage and seems to have been quickly discontinued and replaced by a more conventional cross button catch as found on most bayonets.

The second variant is the most commonly found folding variant identical to the first model except for the conventional cross button lug catch.  This cross button catch locks into a cutout on the left side of the bayonet lug.  This is why the lug itself has cutouts on both sides.  The catch itself also has a cutout on its bottom surface to allow the edge of the blade to nest into it when in the folded position.


Figure 2 - Bottom view of M1938 Short Rifle bayonet lug with latching cutouts on both sides to accommodate both types of bayonet lug catches.

Both the first and second variants have square fullered blades with a knurled end on the tang which is visible when the blade is in the folded position.  They also have a shoulder at the base of the blade which locks into the crossguard when in the extended and locked position for stability.  There is a notch in the shoulder 1/3 of the way back from the cutting edge.  There is a widened area in the channel running through the hilt where the blade rests when folded.  There are corresponding lugs inside the widened well for the shoulder that engage the notches on both sides of it when slid back into the folded and locked position.  There is a cutout on the surface of bayonet lug catch on the inside of the channel in the hilt.  It's purpose it to allow clearance for the edge of the blade when it is in the folded position.  The locking system for the blade is a push button actuated catch nearly identical to that found on most Moschetto type carbines with triform folding blades affixed to the firearms.  A button on the the right side of the hilt is depressed to unlock the blade.  With the button held down through the whole process the blade is pulled forward so the should can clear the crossguard, folded 180 degrees into the storage channel in the hilt then slid back so the shoulder ends up closer to the crossguard and locked into the internal lugs.  Then the button is released leaving the blade locked in the folded position.  To extend the blade the process is simply reversed.


Figure 3 - Locking notch in the shoulder of the blade.


Figure 4 - Internal locking lugs in the shoulder well of the hilt.  The knurled base of the blade tang is also visible.


Figure 5 - Cutout on the inside surface of the bayonet lug catch to allow clearance for the folded blade.

The third variant is the one most books have overlooked.  It is a simplified folding design employing a round fullered folding blade with a solid locking shoulder and no internal locking lugs in the shoulder well in the hilt.  The end of the blade tang is not knurled.  The cross button bayonet lug catch is also simplified by deleting the cutout for the folded blade.  Because of this, the blade's profile is actually shifted back 2mm upon it's entire length so the blade clears the lug catch when folded.  The earlier second model lug catch with cutout is often found on the third model bayonet and seems to have been used until supplies on hand were exhausted.  Folding and unfolding the blade is accomplished in the same way as on the first 2 variants and a very similar blade lock assembly is employed but using a substantially smaller reverse screw head.


Figure 6 - Comparison showing the early blade shoulder and locking button screw on the left compared to the third variant on the right.  Note the third variant blade is shifted back by 2mm.  The square and round ended fullers are also apparent.



Figure 7 - Non-knurled tang on third variant blade.


Figure 8 - Inside surface of bayonet lug catch not inletted.

To refer to the 2 fixed blade variants as being fourth and fifth variants seems logical but production may have been concurrent rather than sequential.  For purposes of this article I will refer to the Fixed Folder as the fourth variant.  The Fixed Folder is not strictly a purpose built production variant but more of a salvage item.  It was made by cobbling together parts of existing folding bayonets and unfinished folding bayonet components on hand into a fixed bladed bayonet.  In general a somewhat crude quillon was welded onto the crossguard and the blade was pinned and welded into place.  Blades, hilts and lug catches from the first three variants can be found mixed together on fixed folding bayonets as well as unfinished hilts that never had their shoulder wells and blade lock apertures cut.  For the most part all grips on the first four variants have been retained by screws in escutcheons but a few fixed folders have been observed made from recycled hilts with crossguards modified by removing the entire blade lock housing and grips being retained by dome headed rivets in washers.  I have possessed one such example that, upon close examination, proved to be a hand worked fraudulent attempt to produce an "unknown" variant.  John Kaszowski of possessed another such example made from a recycled Latch Lock bayonet the seems to be genuine and he has graciously allowed me to include his pictures in this article.


Courtesy John Kaszowski of

Figure 9 - Latchlock variant converted to Fixed folder with rivet retained grips and modified crossguard.


Courtesy John Kaszowski of

Figure 10 - Grip rivet visible in the hilt channel.

The fifth variant is purpose built with a fixed blade and solid tang.  Grips are retained by dome headed rivets in washers.  The blade has round ended fullers like those on the third variant bayonets but the profile is not set back like the third variant blade.

The three folding variants are generally paired with one of 3 types of belt loop scabbard.  The scabbards are steel with a separate throatpiece which is retained by a screw that also secures the top end of the steel belt loop.  The lower end of the belt loop is permanently affixed to the scabbard body with a rivet.  3 variants exist, a very rare fluted body based on the fluted M1891 Carcano scabbard, a smooth body with steel throatpiece and a smooth body with a pot metal throatpiece.  The pot metal throatpieces are fragile and often found cracked, crazed, shattered or otherwise broken.  I speculate these were a brief experimental measure that was abandoned because they are not found on later later non-belt loop scabbards. 


Figure 11 - M1938 scabbard variants from top to bottom are Belt Loop with steel throatpiece, Belt Loop with potmetal throatpiece, Conventional M91 style frogstud containing throatpiece screw and Conventional M91 style frogstud with screw in scabbard face.  No image of the Fluted belt loop scabbard could be secured.

There are 2 scabbard variants deemed to be later production variants with a conventional frogstud nearly identical to those found on the M1891 Carcano bayonet scabbard.  The first is nearly identical the M1891 frogstud and has a retaining screw extending into the scabbard to secure the steel throatpiece.  The second has no screw and the steel throatpiece is secured by a separate small screw offset from the center in the face of the scabbard body set closer to the rim of the scabbard body.

I would like to thank Carl Ziegler of , David Franchi and John Kaszowski of for their help with my research for this article.