Observed today a moldy film on the surface of the distilled water with 1 cap-full of hydrogen peroxide that began testing on Feb 3rd.  In looking back at the update on April 19, the film was present at that time, meaning the film was present about 10 or fewer weeks into the testing. However, the filament continues to work fine. This data strengthens our recommendation to use 3 to 5 cap-fulls of 3% hydrogen peroxide with each reservoir fill. 

The filament with 1 cap-full of peroxide in distilled water, initiated on Feb 3rd, appears to have increasing pinkish discoloration in the loop area (bottom left in photo), but still not severe after more than 75 days of testing. The filament with 2 cap-fulls of peroxide in distilled water (bottom right in photo), also in testing more than 75 days, continues to show no sign of discoloration, as do the samples with increasing amounts of peroxide that began testing earlier this month. 

Removed the tap water, distilled water and distilled water with rubbing alcohol filaments, and continuing testing with previous distilled water with cap-fulls of 3% peroxide. These each had about 1.5 to 2 cups of distilled water, which translates to the filament with 1 cap-full of peroxide having less than 1/2 cap-full of peroxide per cup, and the filament with 2 cap-fulls of peroxide having more than 1 cap-full of peroxide per cup. Two new filaments have  exactly 2 cups of distilled water with 3 cap-fulls (1.5 cap-fulls of peroxide per cup) and 4 cap-fulls (2 cap-fulls of peroxide per cup). For reference, the last filament has only 3% peroxide.  Summarized here:

1. <1/2 cap-full peroxide per cup 
2. >1 cap-full peroxide per cup 
3. 1.5 cap-fulls peroxide per cup
4. 2 cap-fulls peroxide per cup
5. 100% peroxide

 It's probably worth noting again that previous testing has been completed inside a closed container, in a 100% relative humidity environment, meaning all testing is under worst case conditions - except for lighting. In the container, no UV light is present to catalyze the reaction and decompose hydrogen peroxide. Going forward, samples have been inserted into a cigar humidor with a glass top open to room lighting. I'll monitor and report periodically. 

Upon closer examination in bright light, the filament in distilled water with 1 cap-full of peroxide at beginning of testing has a pinkish tint, while the filament in distilled water with 2 cap-fulls of peroxide at beginning of testing does not. This pinkish discoloration is faint, but likely a sign of mold formation. By contrast, the filament with 2 cap-fulls of peroxide appears as white as it did at the beginning of the testing. Next testing will likely be investigating the optimum amount of 3% peroxide to add to distilled water to eliminate or reduce mold formation and extend the life of filaments.