1.2-8: Updated plot.lisa file. Fixed problem with xlim argument in plot.spline.correlog and plot.Sncf. Fixed error in treatment of 'quiet' argument

1.2-7: Fixed error in plot.lisa help file.

1.2-6: Fixed stupid coding mistake in print.mSynch

1.2-5: Final tune-up for complete compliance with github, Rstudio and devtools

1.2-4: Fixed ... in plot-function

1.2-3: Rewrote source to use devtools/roxygen (thanks to Jun Cai). Fixed a silly error in summary.Sncf. Changed name of spline.correlog.2D to spline.correlog2D for consistency (hope this wont cause trouble). Linked source to github.

1.2-2: DOIs added. Fixed calculations with latlon data.

1.2-1: New plot.cc.offset-function. Fixed a stupid double loop (and vectorized gcdist).
	Vectorized spherical correlation function (added circ.cor2) to greatly speed calculations.
	Prettied up the progress counters and turn default for quiet back to FALSE. Major under-the-hood re-code to ease maintenance.

1.1-8: Plotting functions for Spline.correlogs and Sncf changed 
       Default for 'quiet' switched to TRUE
       print.Sncf and print.spline.correlog added.
       Bit of code tidying and CRAN compatibility fixing.

1.1-7: Small changes to plot.lisa.nc, plot.lisa and spatial.plot.
       Bugs in correlog and correlog.nc when latlon=TRUE fixed

1.1-6: Added a new function to calculate mantel.correlog() rams,
      updated lisa and lisa.nc, added a new spatial.plot() function.
      Compliant with new CRAN checks

1.1-5: Some minor changes to three plotting functions

1.1-4: The auxcilliary function cor2() was changed

1.1-3: error in partial.mantel.test corrected

1.1-2: Sncf tweaked so that the cross-covariance function will use the
       predicted y-intercept if z and w are sampled at different locations.
       Otherwise the mean of the diagonal of the cross-correlation matrix
       is used (as before).
       
       quiet argument added to all functions to surpress bootstrap counter
       (if desired)

1.1-1: R-2.6 compliant, cran compliant

1.1-0: Added lisa, added fixed bug in correlog
       (when resamp=0). nc.correlog and nc.lisa renamed correolg.nc and lisa.nc,
       and appropriate permutation tests implemented.
       Various minor fixes in documentation

1.0-9: Calculation of distance classes corrected in
       correlog and nc.correlog to make it consistent with spdep
       (thanks to Ingolf Kuhn)

       Correct some grammatical errors in the documentation.

1.0-8: A function for partial Mantel test added

1.0-7: Jitter argument included in Sncf.srf

1.0-6: Documentation w.r.t. latlong argument corrected
       (thanks to Amit Vora)

1.0-5: A 'circ' argument was added to Sncf and mSynch functions
       to allow calculations with angular data (in radians);
       cor2 function updated for 1.9

       Some inconsequential garbage in the code was cleaned up.

       Compliant with R 1.9

       Deleted some old obsolete functions (oldSncf2D, inner.na)

1.0-4: Two functions (spline.correlog.2d, plot.cc.offset)
       and one dataset (lbm) added

1.0-3: cor2 auxilliary function changed in response to R-1.8
       changes in cor-syntax

1.0-2: FF-filtering fixed (thanks to Ben Bolker)

1.0-0: First propper R packaging

Previous Version History:
June 16, 2003: spline.correlog enhanced to calculate
    e-intercepts

June 16, 2003: rmvn.spa updated to take an optional nugget.

April 26, 2003: spline.correlog.2d the unvariate analogue to m2d
    currently in beta testing.

October 11, 2002: m2d an alternative to Sncf2d incorporated in
    core source.

March 21, 2002: A couple of R/S+ inconsistencies corrected.
    (atan of S+ subsumes atan2 of R). library(modreg) is
    automatically called under R.

March 17, 2002: A number of problems corrected to ensure
    S4 compliance. (functions now run under S-plus 6,
    and R 1.4)

March 17, 2002: x-intercept calculation updated in all
    functions to ensure more meaningful calculations
    under cross-correlation (and under other cases of
    locally negative correlation) (as suggested by Eric
    Seabloom)

March 17, 2002: a permutation test for significance added
    to correlog. A new function mantel.test for a
    permutation based mantel test added [currently in beta]

March 17, 2002: gcdist and inner.na functions written as separate
    functions.

October 10, 2001: Install page rewritten

September 20, 2001: Error in calculations of cross-correlation
    corrected  across all functions. (Thanks to Patrick Tobin).

September 2, 2001: Incorporated code of Tim Keitt for numerically
    improved  algorithm for calculating spatial distances for
    latitude/longitude data

September 1, 2001: error fixed in calculation of cross-correlation
    function for Sncf.srf.

August 12, 2001: mSynch have the added ability to calculate
    cross-correlation

August 5-13, 2001: Sncf2D made R-compliant.

July 10-13, 2001: Recoded to be R compliant. All functions (except
    Sncf2D) are fully R compliant!

May-June, 2001: All functions have the added ability to calculate
    cross-correlation, cross-correlograms, cross-covariances, etc.

April 24, 2001: corrected bug in sncf.q file

March 22, 2001: nc.lisa added. rmvn.spa added.

March 21, 2001: Reformating and recompilation of the help-files.

October 08, 2000: An option 'latlon' was added to Sncf allowing
    geographic coordinates to be given in latitude and longitude.
    The correct distance matrix (in km) is calculated.

August 16, 2000: Sncf.na consolidated with Sncf; Sncf2D.na consolidated
    with Sncf2D. Code branches to pairwise deletion of missing
    values (if na.omit = T)

August 16, 2000: coded mSync to calculate the mean regional correlation
    (with a  bootstrap CI) from a spatiotemporal data-panel.

July 17, 2000: Sncf2d (and Sncf2D.na) for the anisotropic correlation
    functions have been included in the core distribution. Sncf.cc
    to calculate the cross-correlation functions of sympatric
    data-panels. ncf (and related) are, now, deleted from the
    distribution.

June 13, 2000: rewrite and recode with perticular reference to html
    help files. Sncf, Sncf.srf, and Sncf.na is now distributed as
    part of the core library.

April 19, 2000: coded Sncf.srf which is like Sncf, but for spatiotemporal
    data-panels  from stationary random field (marginal expectation
    and variance identical for n*p random variates). [Currently
    distributed in in beta]

April 18, 2000: coded Sncf.na which is like Sncf, but handles (through
    pair-wise deletion) spatiotemporal data-panels with missing
    data. [Currently distributed in beta]

April 18, 2000: coded Sncf (eventually to replace ncf) which have the
    added funcionality of calculating regional synchrony ('cbar'
    with CI), and cbar-intercept (the distance at which synchrony
    decends to the region wide average). [Currently distributed in
    beta: sncf.q]

February 14, 2000: a error in the ncf-function description was corrected

January 21, 2000: An error in the newton search for the x-intercept and
    e-intercept was corrected. (Thanks to Ole Chr. Lingjaere)

January 20, 2000: An error in the fourier filter was corrected.
    (Thanks to Ole Chr. Lingjaere)

January 20, 2000: An error in the bootstrap algorithm was corrected.
    (Thanks to Peter Thrall)

December 07, 1999: An error in the bootstrap algorithm was corrected

November 12, 1999: Initial code uploaded
