Patrick,
There is a good drawing of the grip in the Museum of London at this link:
http://www.romanarmytalk.com/rat/archive...378-3.html
Unfortunately, despite having several good pictures of the Heddernheim one on paper, I cannot find an image of it online to link to right now. However, it is very similar to the one in the link above, excepting that the pommel expansion lacks the ridges seen on the Wallbrook one. As you can see, the Wallbrook handle finishes in a clean line at the bottom, meaning that the guard must have been a separate piece which was perhaps glued to the extant part of the handle, as it does not appear to be broken. The Heddernheim handle was broken below the central expansion and although Obmann's picture of it shows it complete, the lower part of the grip and the guard seem cleaner and as I said above, I am not sure if this is a restoration or simply that the missing portion had been preserved differently before being reunited with the main grip part. If the the present guard is the original, then unlike the Wallbrook example, the Heddernheim handle would have been in one piece. I am fairly confident the Wallbrook example is the way it is because the manufacturer was limited by the size of the pieces of ivory he had available and ivory was not something which was to be wasted. I am fairly sure a wooden version of the same thing would have been the full handle including the guard.
You can see though from the picture in the link how the rod tang would have passed up though the handle to emerge at the top to be peened over. You can also see the recessed area on the front side, which appears to be designed to accommodate a normal inlaid iron grip plate (normally in the usual construction of a pugio handle with two iron grip plates sandwiching organic plates that themselves sandwich the tang, the front grip plate is inlaid, while the back one is not, although in a few cases both sides are inlaid). The two holes seen in the pommel part of the recessed area are presumably intended for nails which also pass through the two holes normally seen in the pommel expansion of a grip plate, to secure the plate to the ivory handle. The Heddernheim handle features two holes in the same position which pass right though, suggesting rivets may have been used in that case.
The guard now seen on the Heddernheim handle does not feature any holes, but then, if it is a restoration rather than the original part, holes may have been deliberately omitted so as not to place them wrongly and misleadingly. However, on many pugio handles the only actual constructional rivets to be seen are those which pass through the guard, with all other visible 'rivets' actually being decorative bosses or small nails (although there are often other constructional rivets holding the organic layers in place which are hidden by the grip plates which overlay them), so I would tentatively suggest that the Walbrook handle's guard might possibly have had a hole at either end of the guard part of the recessed area, which I would guess would probably not have had a lower border so that both iron and ivory sat flush against the shoulders of the pugio blade. Certainly two holes in these positions would make for a far more secure fitting for the iron grip plate, with it thus being secured both at top and bottom.
As you rightly recognise, there are two distinctive types of pugio tang, normally referred to as 'frame tang' and 'rod tang' (the latter type being what you are referring to as a 'rat tail tang'). It seems pretty certain that the frame type of tang was the earlier type and in fact it can be seen in two forms, with many being an outline shape of the whole handle but others only extending a little further than the central expansion and thus requiring a solid piece of bone or wood in the pommel expansion between the tops of the grip plates, as seen on a pugio found in Colchester.
The rod tang seems to have been a later introduction, but existed side by side with the frame tang for much of the first half of the first century AD. By the middle of the century it may have been the dominant type. These tangs were not made from frame type tangs and the pugios are not cut down swords - they are simply a different style of tang (and Mainz sword blades, despite resembling some pugio blades, are far bigger overall).
As we have already seen, the method of hilt construction seen on pugios with frame tangs must be modified or replaced with something else for daggers with rod tangs. At least two methods were in use, one being the type of handle exemplified by the Wallbrook and Heddernheim examples, and the other being the method used on the Usk 1 pugio's handle, which featured a pair of iron grip plates which appear to be normal in every respect, suggesting that unless viewed close up, the Usk 1 pugio would hardly have looked any different to a pugio with a frame tang. Unfortunately, as the organic matter of the handle has completely perished, we cannot know whether the wood, horn or bone portion was in the form of a single thicker piece with a hole drilled longitudinally for the rod tang, as on the Wallbrook/Hedderheim type, or two separate pieces which sandwiched the tang tightly between them, being held in place themselves by the grip plates. Rod tangs are not pierced for rivets to pass through, as any rivets or nails would pass around them.
As you rightly state (and I said myself in my earlier post), pugios did sometimes lose their handles (the Wallbrook and Vindonissa examples being the incontravertable proof of this) and it certainly does seen as though, if no proper pugio handle could be used to replace the lost handle, then some other handle which would do the job was employed to bring the pugio back into a state in which it could be used. Thus we see that the pugio from Vindonissa was repaired by jamming an old gladius grip onto the tang, and the Wallbrook pugio was repaired in exactly the same way but with a more general turned tool handle.
Now to your final question about the three rivets in the tops of some pugio hilts. We do not know the reason for these, but the simplest probability is that they were simply a decorative feature which became part of the grammar of pugio decoration. In fact, these should probably be referred to as decoratively headed nails rather than rivets, as they were hammered into the top of the pommel expansion, often (although not always) through a top plate. This top plate might be metal or it could be of bone, horn or antler, as the Velsen example was. Some grip plates of the thin embossed iron type were make in one piece which folded over the top to be secured on both sides by the rivets passing through the guard. In such circumstances there would be a ready made metal layer to act as a top plate if nails were to be inserted into the top. Otherwise, the grip plates end at the top of the pommel expansion and either leave the tang and organic layers open, or a top plate was added to act as a washer for the nails.
If the tang was of the shortened frame type, then the nails would simply be hammered though into the solid organic core of the pommel expansion. If the tang was of the full frame type, then the nails would have to be carefully hammered into the organic layers at angles which avoided both the tang and the grip plates.
If the tang was a rod tang, the obvious thing would appear to be that the tang, on emerging from the top of the hilt, was peened over into a more or less domed shape, and then two dome headed nails might be (but then again, might not be) hammered in either side of the peened over tang.
I hope this makes it a little clearer and helps a bit. Sorry for making it such a long post. Feel free to keep asking any questions which come to you and I will do my best to answer them for you. There are no stupid questions and if I have said something confusing or left anything out which you need to know, then you should have no hesitation in coming back to me for more.
Crispvs
Patrick,
I realised I do have digital images of both the Wallbrook and Heddernheim handles. The Wallbrook image is the same as the one in the RAT thread I linked to above and the drawing of the Hedderheim handle is from Jurgen Obmann's 1992 JRMES article.
I hope you find the pictures useful.
Crispvs
