About a week ago I bought an older Japanese fluorescent desk lamp (circa 1962 as that is the year the GE lamp was made) and I have a few questions.

First, I've read that vintage Japanese lamps have brittle plastic especially in the switches. I've been using this lamp for no problem and the switch seems solid and the reflector is not hardened or brittle. Some of the plastic is yellowed though (not the starting switch though) which I consider normal for something of that vintage. Should I have any concerns with the plastic as the lamp is primarily plastic?

Second the base has a sticker in primarily Japanese and it says 100 volts. I saw no signs of overvoltage during operation US is 120 volts nominal and my house is around 123 volts. Should I open the lamp and look for a voltage rating on the ballast?

Third, also on the sticker was a basic schematic (typical of Japanese electronics of that era anyway) and it shows a capacitor in parallel with the start switch. Has anyone had an issues with these capacitors failing? Should I follow traditional radio restoration procedures and determine what type of capacitor it is and replace if it is a wax-paper type?

Fourth, the cord is obviously original, I should replace the cord? I didn't notice any signs of the cord being brittle, should I replace as a preventative measure? I will use a polarized plug. should I wire normally (hot to the off switch) or is there a different method for fluorescent desk lamps.

Lastly and this one is a randacnam7321 specialty. I have this lamp connected to the battery side of the UPS and works better than an Incandescent lamp and virtually no noticeable buzzing on inverter power. Will this either be damaged by or cause damage to the inverter in my UPS?

Replace only, what is broken, If all parts you have seen are in good condition, you could keep them original.

For the rating: Isn't it rated for 100V/50Hz? Then the 120V/60Hz would be about the same current, so no problem...

Connector polarity: When the "OFF" is only a momentary disconnect switch (so the fixture remain energized all the time), the polarity does not matter at all, so you should in fact use the non-polarized plug (this is not a safety requirement, so the polarized plug won't mean any code violation).

For the UPS: These are primarily designed for a rectifier type load, with maybe a small transformer adapter in parallel. It depend on the exact concept, but better keep such load way below the UPS rating. But make sure there is no larger capacitor (mainly a PF correction) connected directly across the UPS AC output. The exception are the true sinewave output models, there it would be OK.

Yes, the plastic in the switch can break. So can the sockets. Just be very careful. 100 volts at 50 cycles was Japan's AC.  Don't worry about it. 

Re the UPS. On my experience fluorescents that are run on UPS tend to kep the starter glowing after the lamp started, which i dont like. In your case there is no starter so no such problem

Im wondering if this can cause closer to breakdown conditions in the capacitor across the switch. As i see it, the problem is possible only at the time when the lamp is not lighting, so the capacitor s getting straight square wave across it (so higher current), possibly along with some spikes coming from the UPS - esp if it is running 50Hz transformer and not HF (different types of UPSes). When the lamp is on i see less of an issue since the arc in the fluorescent lamp is clamping the voltage way down

The UPS can see a problem. Especially if it is a HF UPS. The H bridge elements in a UPS of this type may "not like" the back emf from the lantern's ballast at all....