Twenty plus years ago when I started working in this profession there were maybe only a couple of books on Software Testing, and even 12 years ago there were only a few more in the mix, and up until about 2000 there were only about 20+ books on the market. 

Nowadays though it seems everyone and their brother has written a book of some type on Software Testing.  This is good in the fact that we have different subjects within testing now being covered.  Topics range from basic knowledge and techniques to test automation (functional, unit and performance) to test management to specific tools or applications (like SAP), and other points in between. 

Performing an Amazon search I came up with a list of about 250+ books, of which (after trudging through each listing page) about 200 or a little more (I didn't do an exact count, give me a break) were books specifically on software testing or associated subjects (Unit, Debugging, etc.).  Looking through the list I noticed the majority of books were published in the last five to six years.  Wow!

Now the bad news (I'm a Tester for crap sake, not doing my job if I don't bring this up), the majority of the books cover the area of general processes and techniques for Software Testing.  Now how many times can you rehash the same subject trying to build a better mouse trap?!  This leads me into my topic, how many books do we need?  This is worrysome because now we have swung to the other end of the pendulum and run the risk of confusion.  I mean which books are really going to be useful. 

Also, as part of the plethora of books you have new definitions of old terms and methods.  This unto itself leads to confusion and much heated debate, and typically for no good reason.  Look at the definition and use of the term Bug / Defect.  I'm not going to go into it here, but you can see my response, and a smart aleck one at that thank you very much, on other forums.  Basically it was my attempt at humor via George Carlin (genius of making fun of terms).

This whole thing is just plain stupid when we debate vocabulary and definitions of terms.  No wonder other groups look at us with disdain.  But back on topic...

I can only think of and recommend a few books that I would say are original and relevant to our profession.  The first being "The Art of Software Testing" by Glenford Myers, still a classic and contains all that is necessary to learn the theory of Software Testing.  Second, "Testing Computer Software" by Kaner et al.  This book takes the theory and puts it all to practical use.  Third, "Managing the Testing Process" by Rex Black.  Again, straight forward and has practical application.  Fourth, "Software Metrics: Establishing a Company-Wide Program" by Robert Grady & Deborah Caswell.  An older book, but has a lot of great information on testing metrics and their implementation.  Still relevant to this day.  Fifth, "Automated Software Testing" by Elfriede Dustin et al.  A great book on automation of software testing and a process to implement it.  The follow up book "Implementing Automated Software Testing" is a great read also, and has a lot of practical information for any project.  Sixth, "Software Testing In The Real World" by Edward Kit.  This book is a great look at implementation and alignment of testing within a company.  Shows different models of how testing can be done and what role(s) the Test staff may play. 

Now I'm sure some of you will argue my list, but these are books that were first to have new ideas and make points that are still relevant today.  I have read all of these books and they remain in my library.  There are some other books that some may argue are "classics", but to me they are rehashes of the ones listed. 

If anything thing these are good books to start off with, and the others need to be read & utilized as best determined.  So again, how many books do you need, or can read, on this subject of Software Testing.  And which ones are really going to help you do your job.  Our work is about practical application and not theory.