I had seen the reference to Ruiz in the MPEP, but had not followed up on it.
As it turns out, that case points to a prior proceeding where the the CAFC had remanded Ruiz with instructions to perform a (pre KSR) Graham analysis. (Ruiz, 234 F.3d at 660)
Following up on that case I found Dystar 464 F.3d 1356, which, although finding obviousness, provides a wealth of insight into recent rulings on the subject of motivation and suggestion. (Apparently in a vain attempt to school SCOTUS on TSM considerations prior to its KSR ruling)
The suggestion to combine may be found internally or externally with regard to the references. I'm still looking for that elusive magic bullet (like Ratti) that better defines the requirement for internal suggestion.
PS: I find it unfortunate that the court refers imprecisely to the "problem to be solved" where "function" might give better direction.