This is a hacked-together version of the nlme package which alters
the gls() and lme() functions to use great circle distance matrices
when assessing spatial autocorrelation. It was created following
instructions at:
https://stat.ethz.ch/pipermail/r-help/2012-October/325174.html

To use it, you need to remove.packages("nlme") and then:

install.packages(pkgs="path/to/nlmehaversine",
  repos=NULL, type="source")

The package should then compile and install. At that point, you can
run a gls() model with great circle spatial autocorrelation with a
call such as:

gls(ft, correlation=corSpher(form= ~ lon + lat,
  metric="rdist.earth"), data=hotsample)

It's advisable to uninstall this version of the package and
reinstall the base version when you don't specifically need it.

The original readme is included below.

-------------------

Copyright (C) 1999-2001 Jose C. Pinheiro and Douglas M. Bates
	
This program is free software; you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation; either version 2, or (at your option)
any later version.
 
This program is distributed in the hope that it will be useful, but
WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
General Public License for more details. 
  
You should have received a copy of the GNU General Public License
along with this file.  A copy of the GNU General Public License is
available at http://www.r-project.org/Licenses/

-------------------

This package contains the nlme (nonlinear mixed effects) package
version 3.1 for R.

Authors: Jose Pinheiro
         Douglas Bates <[email protected]>
         Saikat DebRoy