I have regularly had fish enter overflows, ending up in sumps, but not the other way around.
After an @ 40% water change yesterday, I slowly started refilling the tank, and as one sometimes does, became momentarily distracted.
By the time I remembered I was refilling, the 125 was filled to the rim, but was caught before overflowing.
View attachment 1451199
The 180, and 125 are recirculating the same @ 300 gallons of water, and connected with a 1.5" PVC pipe from the bulkheads on the 180, and pumped back from the 125.
Water flow thru the PVC is @ 1200 GPH, and as you can see above, a strong flow normally enter above the water line of the 125.
View attachment 1451202
The 1.5" line is about a 6 ft run between the two tanks, there's almost 2 feet of space (gap) between the two tanks and the 180 sits another foot higher than the 125
The effluent end of the pipe is usually above the water line of the 125, but yesterday during the overfill became submerged.
About an hour later, I noticed the mullet was now in the 180 (not where I had put it, in the 125).
View attachment 1451203
You can (sort of) see the silhouette, and an Andinoacara cichlid in the lower left above (in the 180).
I toyed with the idea that it jumped, up and over between the tanks to get there, but.....
My theory is, the mullet (being rheophillic) and instinctually drawn to heading upstream (dubbed by some a Mountain Mullet), was drawn to and entered the temporarily submerged PVC effluent, to swim against the 1200 GPH concentrated flow, upstream from 125 to end up in the 180.
View attachment 1451206
another shot above, of the mullet now, in the 180.