Jim, thanks for your mentoring.  I am going to follow up and report back with the specific data you requested.  What I am trying to do here is explore the electromagnetic behavior of random wires AND try to optimize performance of the simplest and most efficient designs possible.  So my goals are both academic and pragmatic.  My usually practice is to use a 9:1 unun to feed my random wires.  In combination with the KX2 ATU this makes almost all things possible.  But I wanted to reduce/simplify/streamline as much as possible the electronics involved so I am now going back and experimenting with random wires fed directly by the transceiver without the unun.  The idea, perhaps an erroneous one, is maximize power efficiency in the output circuit by providing a direct connection between the XCVR ATU and the antenna - no unun and no coax.  Trying to get as much oomph from my QRP output as possible.  I might be on the right track if. and that is a big if, I can get the antenna and the counterpoise perfectly optimized for a specific frequency.  Is there any validity to my thinking? Or am I just spinning my wheels and worrying too much about a little power loss.

So, in answer to some of your cogent questions; 1) no feedline - antenna and counterpoise connect directly to XCVR ATU, 2) far end of antenna is at about 35 feet, 3) counterpoise is single conductor , 4) counterpoise runs down through floor and runs along top of basement foundation wall so about 18 inches above ground, and 5) counterpoise is only 17 feet long.  Okay, that seems rather short for a 40m band counterpoise so I will try lengthening that for 40m or better yet add a second counterpoise of 33 feet.
Thanks,
Dave W0DCX