Hi all,
I've solved the conundrum I think
The plate already has a black backing, so that's why nothing happened when Zebbie put a bit of black backing behind it. The plate is definitely a negative, and the pattern of the emulsion coating (with the fingerprints at the corners), the thickness of the plate, the rainbow-coloured interference patterns and the women's clothing all points to what we thought initially - date of 1850s, mid to late decade.
Zebbie, you're a lucky duck to have this, so many of them have been broken and scratched to buggery. If the surface that isn't painted is very shiny, that will be the glass side and not the emulsion side. However, it was pretty unusual to paint the emulsion side, so I imagine that the side that isn't painted is the collodion emulsion side. It scratches easily as previously mentioned. Also, the old glass can cause image deterioration (that's what is causing the interference patterns) as salts make their way to the surface and push the emulsion off. The best way to keep the plate is (a) in the dark, (b) in the cool (not refrigerated, just not hot), (c) flat, emulsion side up, and (d) in a paper wallet. Your local records office conservation section should be able to advise you on all that.
Sorry we hijacked your thread - you really only wanted a date but I was getting confused!!
Cheers
Prue