PGPLOT device? (? for list)
> xw

At which point you should see a separate window with something like:

 
and in the command window:
 Expand plot if needed:
 Cursor ("e" or left mouse button to mark edges,
  "q" or right button when OK)
 Cursor ("e" edges, "q" OK)
 ? Typed in the graphics window to get a list of options
 Left mouse button, or "e", expands plot   Click on left of the region you want, then the right
 Center button, or "r" replots
 Right button, or "q" to exit   To leave this panning/zooming mode
 "." shift range up
 "," shift range down
 "a" plot whole array
 "d" demagnify by factor 2
 "y" max y from cursor
 "Q" abandon this region
 .. any lower case letter for command line prompt
 .  To shift the spectrum to see longer wavelengths

Then mouse pointer on left of region you want, left-click, move pointer to right, left-click,
until you get something like:
 
 

Now is the time to right-click, still with the pointer in the graphics window. This will get you out of search mode and allow you to mark the region you want to include in the fitting process. After you do this you get:

 Mark region in which data is to be fitted:
 left limit (space, or left button, when ready)
 right limit
 Region limits:  3849.279A (channel 15987) -   3851.903A ( 16039)
  Channels   15987  16039
  Vacuum wavelength for start of chunk is     3849.30
 FWHM : 0.085A
  Line   1 :ion, lamda0, N,b,z? <CR> to end
  .. in the graphics window
HI 1215                    Enter this with the cursor in the graphics window
 Wavelength used:    1215.67
 set cursor x- wavelength, y- base of feature
Put the cursor at the line center (for wavelength), and bottom (for optical depth at line center), and hit the left button. It jumps up to the half-optical depth point. 
  .. and now half width at half (optical) depth
Move the cursor to either left or right to roughly where the data reaches at roughly that height to provide a line width esitmiate.
  Estimated vac wavelength & z are    3850.49    2.16738
     13.9041    25.6666    2.16738
  Line   2 :ion, lamda0, N,b,z? <CR> to end
  .. in the graphics window
You could put another line in here, even of a different species if you want to. Let's not this time, so just hit the return key. The red + in the graphics window becomes a yellow *, so move back to the command window.
  Include other data? y, n (def),
   (or sf for more from the same file)
>  n
 
 fort.26 summary to: f26.t3851
 
  Continue (c), fit (f), or stop (s)? [f]
> f
 
  no. of ions for fitting is   1
 

   ion          N           z           b         bturb   temp

 iteration   :   0 ( 1 )
 chi-squared :        1.207 (    60.3736,   50 )

   H I       13.9041    2.167382      25.67       0.00   0.00E+00  0    !   1
 
 iteration   :   1 ( 1 )
 chi-squared :        0.841 (    42.0524,   50 )

   H I       13.9140    2.167377      25.53       0.00   0.00E+00  0    !   1
 
 iteration   :   2 ( 1 )
 chi-squared :        0.839 (    41.9583,   50 )

   H I       13.9144    2.167377      25.49       0.00   0.00E+00  0    !   1
 
 iteration   :   3 ( 1 )
 chi-squared :        0.839 (    41.9452,   50 )

   H I       13.9146    2.167377      25.48       0.00   0.00E+00  0    !   1
 
 parameter errors:
   H I        0.0024    0.000001       0.12                             !   1

 statistics for whole fit:
  Runs test  K-S test Chi-squared  Chans ndf   APr   Xp(.68) Xp(.95) Xp(.99)
   0.44586   0.79333     41.95      53   50   0.784   54.29   67.22   75.40

 Statistics for each region :
    Start      End    Chi-squared  Chans df?
   3848.80   3852.40     41.95      53   50   0.784 < Prob < 0.863 g= 0.784   1
   3848.75   maxdev       0.2881              1.000   0.956   1.358   1.627

  Plot? y,n, or c=change device [y]
>
  Line, system number: (* for internal guesses)   Use internal guesses generally
> *
 at co   co   3
 at er   co   2
 at ti   co   5
 plot parameter? (type he for options list)
> <Return key>

At this stage you should see something like:

which is the fit superposed on the data. Then:

 Plot? y,n, or c=change device [y]
> n
  Fit more lines? [n]
> n
[EXITS]

It is then looking to see what you have by way of output files. There will be:

fort.13: Initial guesses, in case you have to restart.

fort.18: Much of the fit history you saw flashing by on the screen, in case you wanted to know what happened.

fort.26, or, in this case, f26.t3851:  Summary of the fit parameters, goodness of fit, fit regions. By default this is given the name fort.26, but this can be over-ridden (in a setup file) to be some root (here f26.t) followed by the integer part of the central wavelength of the region  being fitted ( or the first in the list for multiple regions). It contains at least, with the important variables identified here:

 %% q1122   1    3849.25    3851.95       marker, filename, 1, fit region start wavelength, end wavelength    
  Ion        z         error     b        error logN      error          Normalized chi^2       probability
  H I     2.167377   0.000001   25.48     0.12  13.915    0.002  0    3     0.8389031   53   50  0.784  0