Thanks for all the kind comments
The previously posted pic (the middle of a 3 shot bracket) was the result of good luck as can be seen from the two additional shots posted here,resulting from a 3shot bracket, A fixed at F8, ISO fixed at 200, Bracket High speed 0, -1, +1 resulting in 1/1000, 1/2000, 1/500 SS.
Hand held, D800, Nikon 800-400 G VR at 400mm
The overexposed shot is definitely fuzzy due to relatively long exposure time and simultaneous panning to follow bird

Prior pic

(Download)

After pic

OK I give up --change my vote to Brewers Blackbird based on eye color, the Rusty blackbird has a darker head and a slightly curved bill-- the head in the pic sure looks brown to me but lack of yellow eye eliminates cowbird

Great Blue from Boise Idaho at the Rookery On the Greenbelt

Great Blue Heron Barrel roll

(Download)

How about a brown headed Cowbird
definitely not Redwing Blackbird or Brewes

As other posters have said your combination of lens and TC may or may not work physically
The real question is whether or not a TC with a particular combination of camera and lens does any good, A recent thread challenged anyone to produce a pair of pictures with/without a TC in which the TC improved the real resolution.
What a TC does is to magnify the optical image projected on the matrix of cells in the sensor (cropping the resultant picture and making it darker)
If your camera already has sufficient pixels to fully resolve the sharpest portion of your image, then optically magnifying it does not help- you would get the same result by digitally cropping/zooming

On the other hand with a low (but beautiful) pixel camera such as the D4 with a good prime lens the opposite may happen and the apparent resolution would improve with a TC

Conclusion: Optical magnification using a TC may or may not improve your ability to read that last line on the eye chart

Along the same line, switching your lens from a FX body to a DX
does not improve the "reach" it only crops the image, exactly as if you had cropped the FX image in photoshop

A 300 mm FX lens and body has exactly the same reach as a 300 mm DX lens/ body. The FX is only bigger, heavier and more expensive with a wider FOV. above assumes both cameras have same technology and pixels per inch of sensor

"P" is "Programmed Auto" IE Auto
Within "P" one has the option to favor shutter speed or aperture
If you want really auto auto turn on "Auto ISO"
I agree with the lens comment, you cannot take advantage of those extra pixels with lower quality lenses

Hi Gene51
You got me thinking about pixels and lens resolution
So I printed up a lens test target, much enlarged so that printer resolution is not affecting results very much
Pasted it on my back yard wall and moved back about 30 ft
Set up my D800 on tripod, remote trigger, VR off, ISO 200 , F 7.0 and took pictures at 200 , 300, 350 and 400 mm
Opened in PSCC and zoomed way in so that individual camera pixels are easily seen
Examine the test chart where there are a bunch of gently tapered black lines with equal white spaces
The sweet spot seems to be around 350mm with slight degredation at 300 and 400 mm- pretty much per your experience

The result I found interesting with my setup and a perfect 400 mm lens (ignore that this is not possible) my picture would have had 4 black pixels and 4 white pixels alternating, my best result provided 2 black (maybe 2.5) , 2 grey, 2 white 2 grey, 2 black etc

Conclusion is "not bad" the camera is somewhat better than the lens but not by much, as evidenced by the fact that I had at least 2 pixels per blur region, and that with this setup a teleconverter would not help much if any
Now with one of those $10,000 primes and a D4 that everyone says to get, (better lens resolution and fewer pixels) the result would probably be different

Our local Heronry has equal numbers of DC Cormorants and Great Blue Herons 30 to 40 of each

Herons n Cormorants in Boise

(Download)

Good Post
This is an issue that I have been thinking about recently

The answer I think relates to the relative quality of your camera (primarily pixel count) and the sharpness and VR capability of your lens
Take your picture and digitally zoom in /crop

If your image becomes visibly pixilated (subjective evaluation) you are pixel-poor as melismus puts it and an extender might help

Otherwise just digitally zoom and crop for the same or better effective resolution and better low light performance
This is the situation I find myself in with a D800 and Nikon 80-400 G

Nikon makes it clear on their web page that the 28-300mm f/3.5-5.6G ED VR cannot accommodate their TC.
I have that lens and have recently ordered the new 80-400 G which I believe does work with TC
This has been a very active discussion item on UHH recently

What I refer to is what Nikon calls Auto HDR which is different from taking 3 shots and post processing

Per the manual
Actually the D800 takes only two shots and averages them but with only one mirror flip so for reasonably fast shutter speeds that is close to "one exposure" it seems like one exposure
is only for JPEGS

It is good talking with someone who understands things
Everything is a compromise in our case how big a kit you can lug around and how dark is it when you pack up and go home

To clarify IMHO
Yes heat can be a factor but not a significant one for normal cameras (astronomers use liquid helium cooled sensors)
The noise all comes from the amplifiers not the actual sensor
Changing ISO is done by changing the amplifier gain and does not affect power dissipation / temperature
All other factors being equal, a larger sensor cell size will provide a stronger signal and hence better SNR/Higher ISO
Sensor cell size depends on sensor array dimensions (IE APS-C vs FF) and the number of sensor cells in the array IE Pixels

300 vs 400 is not the issue, but with a sharper lens motion blur previously hidden by lens softness will be more evident