Just a couple of comments on CHT's that haven't been mention:

I've read about "design intent" and proper temperatures for best "efficiency" and they might be misleading.  Our engines won't be much more or less efficient as temperatures change, but:

Higher temperatures will, in theory, produce higher efficiency (lower BSFC) because less heat is rejected into the cylinder, saving more for useful work.  The change will be very small, maybe not even measurable.
Higher temperatures will, in theory and practice, reduce power output.  This is because more heat will be transferred from the cylinder into the intake charge, reducing the density and therefore the power output at a given manifold pressure.  This is probably more than a trivial change, but not a big one either in the temperature ranges under discussion.
Higher temperatures will result in higher localized oil temperature in the piston ring area.  This probably is a matter of some concern.
Will small changes in temperature result in non-optimum piston/cylinder clearances?  I doubt it, but I have no supporting data.

These are all small effects and not as important as the one previously mention by others - high temperatures result in lower metal strength and reduced detonation margin.

Neglecting changing engine control parameters (like running LOP) lowering engine temperatures takes air flow because we are not likely going to change the design of the engine fins (oil cooler design is the exception).  Unless you get that by eliminating leakage flow someplace that airflow costs drag.  It takes more power to push air through the engine compartment than it does to push it around the outside of the plane.  Therefore, once the internal airflow has been optimized and all the leakage paths eliminated, further reducing the temperatures will cost.  I wonder if, once the CHT's are reduced to the 350 range in cruise, why one would keep going in the quest of lower temperatures.  If I had temperatures in the lower 300's I would start looking for ways to reduce airflow and therefore make the plane go faster.  I think this is a significant concern.  As the cylinder temperature is reduced via higher air flow the amount of heat rejected INTO the cylinders goes up and the amount of heat rejected OUT of the cylinders to the air (temperature rise) goes down, meaning that it takes a disproportionate increase in air flow to get a drop in temperature.  To restate, if it takes an increase of 1 pound per second to (just made that up) to reduce the temperature from 420 to 400 it might take a 2 pound per second increase to drop the temperature from 320 to 300.

In summary, the "desired" engine operating temperature is a balance of many factors, primarily between engine life and total airframe drag.

Gary Casey