We all know that fluorescent lamps sort of fade on while starting up on a rapid-start ballast. Most of the time, the lamps illuminate dimly as soon as they are turned on, and then they get gradually brighter or snap to full brightness. However, I have noticed that on some ballasts, there can be a split-second delay between when you turn the fixture on and when the lamps start glowing dimly. It doesn’t seem to have any adverse effects on the functionality of the fixture, just a weird effect.

One example of a ballast that behaves this way is the Universal 446-LR-TC-P ballast from 1985 that I installed in my bathroom light recently. I have noticed that after it has been switched off for a while, when you go to turn it on again, there is a ~0.5-1 second delay before the lamps begin starting up. If the lamps are already warm when starting up, the delay goes away.

Does anyone know why this delay occurs? I have noticed it on a handful of ballasts over the years, but the vast majority don’t seem to have this effect. Is it something that caused by a specific pairing of lamps and ballast, usage conditions, or just something intrinsic to the ballast?

There is a S-like curve between lamp ignition voltage and cathodes temperature (which depends on preheat voltage and time). Lamp starts to glow when OCV is above this curve. Curve shape depends on a lamp type, exact lamp sample, grounded objects around and even glass surface resistance. And of course different ballasts will have different OCVs. Later ignition is beneficial on the lifetime as the lamp spends less time in glow discharge, but the risk of overall ignition fail increases.

Rapid start doesn't mean instant start though so keep that in mind. 

In other words it is the time needed for the filaments to warm up to a temperature enough to even initiate the ionization.
And as mentioned, it is not just about the ballasts alone, it is really a combination of the ballast OCV (so combination of ballast step up ratio and the actual mains voltage), the lamp properties, temperature, lamp surface contamination, the exact design and even tolerances of the fixture and the lamp position (the fixture metal work is acting as an external capacitively coupled electrode for the lamp), whether the fixturfe is properly grounded,...

One issue is definitely grounding... I've had this issue in the past where there was slight delay before the lamps even began attempting to start,
but after properly grounding the fixture, this slight delay went away. I've also seen this delay with properly grounded fixtures though, when the tubes
are exposed to high humidity... especially with full power F40T12's, unfortunately... I've never seen this occur with the F34T12's, and it makes me wonder
whether that may have something to do with the F34T12's containing conductive coating.

Like was said before, a slight delay with ignition is good, as the tubes spend less time with the discharge on cold filaments, so less sputtering. Rapid start circuits balance on quite tight tolerance between no ignition at all and cold starts.

Yeah I figured it had no negative impact on the life expectancy of the lamps.