> The use of the word variants makes this inaccurate.

Yes, "over-clocked variants" makes no sense at all. Overclocking does not create variants of a chip. The use of the word "partly" is nonsensical as well.
But this all has nothing to do with the "MC68060FE133" as this designation means it's supposed to be a genuine 133 MHz part and the manufacturer guarantees that it runs at 133 MHz *without* overclocking. And while I suspect this chip to be an illegally rebadged 75 MHz Motorola/Freescale part I have doubts that it's able to run reliably at 133 MHz even if watercooled. If at all, the line in question supports the suspicion that the "MC68060FE133" is not a stock 133 MHz part and thus a relabelled slower part.

> I've never heard of an overclock above 100Mhz (but then Andreas might have).

Indeed, I have:

"The 060 Rev.6 is guaranted for 90 MHz and can run at 100 MHz in 90% of cases and 105 MHz in some cases."
http://www.powerphenix.com/CT60/english/overview63.htm

"The world record is 108 MHz without CTPCI . Mine is at 105 Mhz + CTPCI and with the standard CT63 low profile cooler..."
http://www.powerphenix.com/CTPCI/english/Historical.htm

Benchmark results at 105 MHz:
http://didierm.pagesperso-orange.fr/ct60/benchs4.htm
http://didierm.pagesperso-orange.fr/ct60/benchs5.htm