>the code does require the use of a metal underground water pipe as a grounding electrode if it exists. 250-50

I see what you are saying. I don't count it as a primary electrode for a number of reasons.

Please see http://www.nfpa.org/Codes/National_Electrical_CodeR__NEC/Proposals/necpdf/A250.PDF proposal 5- 152 - (250-50(a)) which was rejected. My thoughts follow along those of CMP member Schiff with respect to ground rods being primary and then bonding to the water line as opposed to utilizing the water line as a primary electrode.

I realize that some may think that the water line has to be a primary because of its low resistance, traditional use, etc.

However, because I can't prevent it from being removed and I neither check its resistance nor can I verify that it has at least 10 feet of metal in earth contact, I simply assume that I need two ground rods for my primary (permanent) electrode system.

Like CMPM Schiff, I am in the minority. But I am not just making an argument.

I am not 100% clear that Mr. Fletcher knew what he was saying. But in principle I agree with trying to reduce the current on the water line by bonding it to an alternative earth path.