YROC:
The reason that there was ATF in your cartridge is because the oil inside the damping cartridge and the oil in the air spring chamber will mix by design. When the fork bottoms out, the "head" of the cartridge has the "oil lock ring" attached to it and it goes into the "oil lock case" at the bottom of the fork. When the cartridge head enters the oil lock case, the "bottoming cone" feeling is a strong hydraulic cushion created by the oil lock ring entering the oil lock case, displacing oil up around it. Think of it like forcing your fist into a mug of beer: the beer will be forced out of the mug around your fist. The hydraulic resistance is the beer forcing it's way around your fist as it displaces area in the mug. This situation creates a very high pressure inside the oil lock case as the oil is being forced around the outside of the oil lock ring. THis high pressure actually will force oil INTO the cartridge past the oil seal and cartridge rod. The oil seal on the cartridge rod is directional: it keeps oil INSIDE the cartridge, but does not really keep oil from being forced IN by pressure. Everytime oil is forced INTO the cartridge, the floating piston position becomes higher, but as the fork is bottomed, it also purges oil out of the top like it does when you are bleeding the assembled cartridge. So oil "circulates" in and out of the cartridge with the oil outside of it. That is why your ATF made it's way into the cartridge (and S1 made it's way OUT of the cartridge). By the way, this same function occurs with SHOWA twin chamber forks.
I have a question for you, why did you use ATF in the outer chamber (OR in the inner chamber for that matter)? S1 oil is designed to provide consistant damping force in a slightly pressurized environment (inside the cartridge) and also provide high film strength lubricity for the structure parts (inner and outer tube bushings). I don't think ATF offers any advantages to either of those functions. I actually tried building a twin chamber fork a long time ago by putting two seals in the cartridge (one lip faced up and another faced down to keep oil from getting IN or OUT), then ran a high viscosity index SHOCK fluid in the cartridge forthe damping, then a thicker, higher film strength lubricity oil in the outer chamber for better smoothness. My thinking was that the shock oil would fade less inside the cartridge (not that there was a fade problem, but I thought I was being smart!) and then run a more lubricating oil where all the big friction parts were working. The project was a failure for a few reasons, but ultimately, I decided that I should not try to improve on a system that was working pretty good already and spend time thinking about things that were NOT working so smoothly....