ok, so. I'll have to work this out as I go, so please bear with me.
If you have two geckos, male and female, both normal het for eclipse (or any recessive trait) then they'd have a genetic make up of eE.
Which, I should state, is just the way I signify recessive traits, with the lower case first, so I hope its not wrong.
If you've worked with punnet squares it makes everything easier.
But they'd be set up like this
... e E
e ee eE
E eE EE
which, if lowercase is not having the trait, and upper case is the trait,
then you have 25% which will be eclipse, 50% het eclipse, and 25% which don't carry the trait at all.
But you can't tell the 50% that are hets from the 25% of the offspring that won't be het by sight, you'd have to raise them and prove their het through breeding.
So you'd have to say that of the offspring 75% will not show the trait, only 2/3 of those would be het for it, the other 1/3 not being so.
So they'd be 66% chance of being het for Eclipse.
I hope that's correct, I'm sure someone will show me better if not.