Foresys --- a very nice static analysis tool from Simulog in France. Does many of the same things that FORCHECK does. A nice plus is that code metrics are included; their value may be debated, but they are included if you want them. Also has a GUI; this is the only linux static analysis tool I have seen (for fortran that is) that has a GUI. Very nice, can easily take you with one click to the definition, or source code, for a variable or function on which you click in a source window. A trial version is available online.
NAGWare Tools --- Software from NAG for transformation and analysis of Fortran-77 and Fortran-95 code. More information here.
The Fortran Programmer's Workshop This tool can be thought of as an interactive "make" facility. It has methods for organizing source code into libraries, building various programs from this code, running programs, and other code-maintenance facilities. There is a command-line version as well as a new graphical version which uses Tcl/Tk. You can visit the Fortran Programmer's Workshop Web Page for more information. I have no experience with this tool, as I am a diehard make fan.
SNiFF+ --- A "source code engineering tool" which I know
is in use at CERN, has now been released for Linux. Note that the
version that you can download only works on glibc2.1 systems (like
RedHat 6.0). I have not yet tried it, but here is what they sent
Description: SNiFF+ is the premier Source Code Engineering tool. SNiFF+ provides an integrated collection of source code analysis, browsing, navigation, comprehension, editing, build, and configuration management tools for C, C++, FORTRAN, Java, and CORBA IDL developers. With support for Windows and all major UNIX platforms, SNiFF+ enables developers in cross-platform and multi-language environments to utilize the same tool set and GUI.Visit their web site to download an evaluation copy and find more information.
asa --- This is an output filter for your Fortran output files. It looks at the Hollerith carriage-control strings in the first character position in each line (those pesky "0", "1", and "+" characters) and converts them into control codes understood by programs like "more" and by Unix printer software. This code should also work on other Unix systems.