Yes that is an anomaly. I've never seen that before, but I think there might be a logical explanation.
Firstly I had to do back through my archives to examine each and every Constantinople trademark in there so I could prove that this example is genuine, even though my gut already told me that it was, despite that flaw in the placement of the "J". Everything about your trademark checks out and with a high degree of accuracy so that I am 100% sure that its authentic.
After being convinced of that, I tried to see if I could find another of any type with a reversed "J" in it. I could not. But I did find one example of the exact type that you have. This type has two unique identifying features. It's one with an older type Arabic script but that isn't unique but only helps in dating. The unique parts are the tall lettering in "K. Zildjian & Cie." (in your case the "K" didn't print, but that's not too unusual), and the point in the "M" character in "Trade Mark", that doesn't reach all the way to the bottom of the lettering line, but terminates half way. Those 2 unique features only appear in yours and in others of that exact type such as this one that I present to you here on the left:
[img]http://photos.imageevent.com/drumaholic/lizziesstuff/12%20Const%20TM.jpg[/img]
I have these stitched together as a pair because these were originally from a pair of 12" hi-hats that were foundry matched. And these were real "dance band" style hi-hats and not some marching band cymbals that were marketed as being hi-hats like you usually see on eBay. The left one checks out in every detail as being the same as yours.
My idea as to how this reversal may have occurred is that the dies for these could have been in the form of individual letters place in a block template. So it could be that just one letter could have been inadvertently reversed when the block template was first set up. But this is the only anomaly of this type that I've seen. So with your permission I'd like to use this one for the anomaly section that I have planned as part of a future web-site.
As for the dating, I would have to place this one in the early to mid 1930's period.
Anomalies such as these are what keep cymbal trademark analysts like me awake at night. But yours is certainly a one of a kind example so unique that I'm going to have to present it to the esteemed South Korean vintage cymbal scholar, Dr. Bong. His official title:
"Trademark Analyst Emeritus".
