I have already taken up to Multivariable Calculus, Linear Algebra and Diff Eq. I want to learn Calculus on Manifolds by myself, could you recommend a good introductory book on this subject?
Should I look at other things first, like topology, to get a better background?
What you're really asking for is a textbook giving a modern presentation of vector calculus/calculus of functions of several variables. Of necessity,there's going to be a lot of overlap between such textbooks and differential topology books. Indeed, I think eventually separate books on both subjects will be obsolete and there'll be unified presentations of both.
The standard books for learning this material are Calculus On Manifolds by the legendary Micheal Spivak and Analysis on Manifolds by James Munkres. Spivak's book is basically a problem course with quite a few pictures. It's quite rough going,but it's worth the effort if you've got the patience. Munkres is more of a standard textbook and covers the same material with much more detail.
Notorious for it's level of difficulty is Advanced Calculus by Lynn Loomis and Shlomo Sternberg, now available for free at Sternberg's website,which is a huge gift to all mathematics students of all levels. This book was written for an honors course in advanced calculus at Harvard in the late 1960's and it's unimaginable that they actually taught UNDERGRADUATES this material at this level. Then again,these were honor students at Harvard University in the late 1960's-argueably the best undergraduates the world has ever seen. In any event,for mere mortals,this is a wonderful first year graduate text and probably the most complete treatment of the material that's ever been written. It even ends with an abstract treatment of classical mechanics. But you better make sure you got a firm grasp of undergraduate analysis of one variable and linear algebra first.
Similar in content,but easier and much more modern is J.H. Hubbard and B.B. Hubbard. Vector Calculus, Linear Algebra, and Differential Forms. Beautifully written,wonderfully illustrated with many,many applications,philosophical digressions and unusual sidebars, like Kantorovich's theorem and historical notes on Bourbaki, this is the book we all wish our teachers had handed us when we first got serious about mathematics.Even if you're using a "purer"treatment like Spivak, it's a book you simply must have. It's a book anyone can learn something new from.
Lastly,I want to mention something. The very first course trying to present to undergraduates the essentials of calculus on manifolds was given at Princeton University in the early 1960's and was taught by Norman Steenrod,along with Nickerson and Spence. Spivak was an undergraduate in that course and famously,it inspired his textbook. Although those notes circulated for many years, they've never been published in book form.
Until now. The venerable Dover Books has acquired the rights to the notes and they will finally be officially published by them this summer. I've never seen the notes in detail-but given their heritage,they're well worth checking out.Also,since Dover is putting them out in a very cheap edition,there's no good reason not to own a copy. So be on the lookout for that.
That should be more then enough to get you started-good luck!
A good place to start learning about manifolds is to read a book by Spivak called "Calculus on Manifolds." This book is very small, and the first three chapters are a short review on multivariate calculus, but I should say that the focus is much more on mathematical rigor than a normal calculus course so it would be worth your time to look through that I imagine. The last chapter introduces manifolds, how to integrate on them, and eventually culminates in the modern version of Stoke's Theorem.
If you want however to get a much more in depth view on manifolds, you will have to learn some topology. A good free online book to learn from, that I myself originally used, is called "Topology Without Tears." It can be downloaded for free at: http://uob-community.ballarat.edu.au/~smorris/topology.htm
It covers about everything a first course in topology covers, while being fairly straightforward and easy to understand. It is especially good for people who have not dealt with mathematical proof very much yet, because the author is very clear and detailed in his explanations and often explains about common proof techniques.
If you prefer a different book, or a more commonly used book than "Topology Without Tears," Munkre's Topology book is fine.
I can recommend two books on introductory manifolds which will be a lot more in depth than Spivak, but are still at the introductory level:
John Lee's "Introduction to Smooth Manifolds" and Boothby's "Introduction to Differentiable manifolds"
These books are about Smooth Manifolds, which are a type of manifold in which calculus can be done.
Lastly, if you learn topology and think you want to learn about manifolds from a topological point of view
William Fulton's "Algebraic Topology"
is a good place to look.
I hope this helps! I've read a lot of books, so if you are looking for anything in particular that is slightly different from what I've said please feel free to email me.
Munkres (the author of the very clear text on Topology) wrote a book called Analysis on Manifolds. This is basically like Spivak, but twice as long and with more pictures. Everyone has fond memories of Spivak after they've used it, because of how small and "cool" it is. But at the end of the day (especially for self-learning) I think Munkres is probably more straightforward and less frustrating.
http://www.amazon.com/Analysis-Manifolds-Advanced-Books-Classics/dp/0201315963
May be this : Spivak,Calculus-Manifolds-Approach
It is an reprint from year 1995, although maybe an more new could also you serve.
Spivak and Munkres are both not bad, but I think Advanced Calculus by Loomis and Sternberg is even better. It rigorously covers linear algebra, calculus in severable variables, metric spaces, and multilinear algebra. Then it moves on to manifolds and integration, and ends with applications to classical mechanics.
Moreover, it's free! See http://www.math.harvard.edu/~shlomo/
If you are looking for a book with tons of proofs, I would strongly recommend J.H. Hubbard and B.B. Hubbard. Vector Calculus, Linear Algebra, and Differential Forms.
Super good as an introduction book if you can find someone to go to when you are stuck. http://www.amazon.com/Vector-Calculus-Linear-Algebra-Differential/dp/0130414085
I'd just like to chime in to express my agreement with some of the main points above. Spivak is an elegant little book with good exercises but years ago I pooped out I think in the midst of his characteristically terse treatment of differential forms, wedge products and all that. Munkres' "Analysis on Manifolds," on the other hand, covers much of the same material but in a much more leisurely, chatty and well-motivated way, with lots of examples. And with a much more friendly appearance than Loomis and Sternberg, at least the latest edition of the latter. L and S is indeed a bit intimidating, and goes into much more detail and theory than either Spivak or Munkres. E.g. where Munkres hops right into the idea of the derivative of a function between R^m and R^n, but with some friendly motivation using the ordinary derivative, and then the directional derivative, L&S spend pages and pages on theorems about norms. Perhaps L&S might make a good reference or reading after Munkres. So I tentatively plan to study Munkres. Then if all goes well, and with more topology and other review, on to John Lee's "Introduction to Smooth Manifolds," which seems like a relatively friendly treatment on a much higher level.
Differential forms and connections by Darling is not too bad. I can't remember how basic it is though. He does spend a lot of time going through what differential forms are. My approach to learning is to grab a bunch of books and read each one in turn until i get confused or stuck and then to go to the other to see if it makes more sense. There is also topology from the differentiable viewpoint by milnor. but this is more on smooth manifolds than calculus.
