My boss'll probably laugh till he falls over if he ever reads this review. He's an atmospheric physicist (translation: "a meteorologist who doesn't want to be in the same category as the empty suits on the evening news,") who programmed a lot in FORTRAN before he changed over to management. He likes to goad C programmers showing signs of prima-donnaism by saying he could write XYZ in FORTRAN, no matter what XYZ happens to be.What earthly use is FORTRAN these days, you may ask (apart from any unlucky soul who has to maintain some hunk of ancient code), and thus, what use is this book? Well, that depends.
In the last numerical analysis class I took, I opted to do the programming assignments in FORTRAN, after one or two aggravating forays at writing them in C with a limited amount of time. Translating the algorithms into FORTRAN required a lot less work (duh, you say; the language exists for FORmula TRANslation). Still, with a time limit for homework assignments, it can be useful.
(In real life, a mathematician attached to a project I'm familiar with designs algorithms, but he's not a programmer, so his assistant has to implement them. Well, the mathematician actually writes them in FORTRAN, and the assistant gets to translate them into C...)
This particular book is, of course, somewhat elderly, as the title will tell you, but that can actually be in your favour. It's *extremely* easy to follow, and quite short, so it's easy to find what you need. And no matter how old, unmaintained, and just plain obsolete the FORTRAN compiler you're dealing with happens to be, this book will support you in writing code that can work with it.
Even though I don't work with FORTRAN these days, I keep it as a reference just in case, because it's worth its weight in platinum if you need to write something in the language.