Many thanks to bplus — the implementation was relatively painless.  Not quite as clean as I'd like it but it's a good start.  Rather than trying to clean up the map outlines I'll try for a better map and see how that goes.

Once again, thanks for all your help.

Here's the right one WITH the legend stuff — used only "Dos" available colors.

My bad on the homework thing - I mean, you know how this question was sounding. Can't blame us (me)! Anyway, welcome aboard.

Lemme make one question-comment... are you trying to get accurate distances on the globe using a Mercator projection? Cause... you can't right? My humble understanding of geometry (taking the earth as a sphere) dictates that you need to correct for the actual shape of the planet using a suitable coordinate system.

If you've done so, then great. If this sounds foreign, then imagine a 6 inch line in the map drawn just under the north pole. How many miles long is that? Next imagine the same-length line drawn horizontally near the equator. That one will represent way more miles.

So that's the comment. The question is: what's your distance formula?

Thanks STx.  I don't see how "homework" came into it — must be something between the lines that I can't see — but hey, no problem at all.

No, the distances are all calculated internally for a 'spherical' earth (using high-falutin' trig functions that I couldn't possibly understand!) but the RESULTS are plotted on a 360 x 180 map — hence "Mercator" projection.

I'll see if I can find a link to the original program: "MicroMUF".

Thanks for all your replies.  Homework?  Where did that come from?  Nope, not homework!  Here's the background. 

About 25 years ago I re-wrote an existing HF radio propagation program.  The original calculated only between two fixed points — a transmitter and a receiver.  I re-wrote it to calculate from a single point (receiver) to all other points on the earth's surface at fifteen degree intervals.  (Point of interest — that version took 5 days to run on a 2Mhz Motorola MC6809 running XBASIC.)  Several years ago (when machines started to get seriously fast) I changed that to calculate at one degree intervals, i.e., a 360 x 180 map.  (I don't know how to grab a screenshot of the result but will do so when I figure it out.)

A couple of months ago I came across this site and decided to try the program again — it ran just fine and super fast too — about 5 seconds!  I never did get around to overlaying the screen plot with a world map but I am intending to do that now.

Looks to me like bplus is on the mark and I will see what I can do with his text file as that seems the best way to go.  My map is centered on 180 degrees (roughly New Zealand) so I will need to rearrange.  Thanks again.

(Yes, NZ would be a great addition (and on a lesser note, Tasmania).  Ha!  (Kiwi)