Calculus On Manifolds Solutions

Working Through Spivak's Problem Sets

The book most people mean when they say Calculus on Manifolds Solutions is Spivak's short but brutal text. The problems are where the actual learning happens. The proofs in the main text are elegant but terse, and they assume you already understand a lot of what's going on. You won't get there without grinding through the exercises. Most people hit a wall around Chapter 3 when differential forms and the generalized Stokes theorem show up together, and that's usually where they start looking for solutions online. I spent weeks trying to work through every problem in Spivak on my own during a grad qualifying exam prep. What I learned was that the solution manuals floating around are a mixed bag. Some are correct but skip critical steps. Others have genuine errors, especially in the later chapters dealing with de Rham cohomology and applications of Stokes theorem. The ones worth your time typically show every boundary computation and explain why a form is closed but not exact, or vice versa. The best resources I found were either lecture notes from courses that assign Spivak as a primary text or carefully worked solutions hosted on university webpages. MIT OpenCourseWare has materials related to this, and a few professors post their own solution keys. The student-made ones on various math forums tend to be more complete but less carefully checked. I learned to verify every answer by plugging it back into the relevant theorem. If your solution to a Stokes theorem problem gives a boundary integral that doesn't match the surface integral, something is wrong, and usually it's an orientation issue.

One specific problem that tripped me up was showing that a certain k-form on R^n \ {0} is closed but not exact, and computing its cohomology class. The standard approach uses a deformation retraction to the sphere S^{n-1} and pullback via the inclusion map. I kept getting the sign wrong on the orientation of the boundary because I wasn't tracking which direction the normal vector pointed relative to the standard orientation of R^n. The workaround was to explicitly write out the parameterization of the sphere, compute the pullback form component by component, and compare against the volume form. This took about twenty minutes of careful coordinate computation instead of the ten seconds it should have taken if I'd just drawn the orientation diagram correctly at the start.

The Core Structure of the Problems

Spivak organizes the material in a way that builds quickly. You start with linear algebra review, then move to integration on chains, then the fundamental theorem as a special case of Stokes, and finally the full generalized version with differential forms. The problems in each section reinforce the previous material while introducing new technical machinery. The change of variables theorem comes early and is proved using the partition of unity argument. This is one area where solution manuals tend to gloss over the partition of unity step, which is actually the hardest conceptual leap for most students. When you see a problem asking you to prove something about differential forms, first identify which theorem it's asking you to apply. Most problems in Chapters 2 and 3 are direct applications of either the change of variables theorem, Stokes theorem, or both. The trick is setting up the chain and the form correctly. I've seen people lose points not because they didn't understand the theorem but because they parametrized the manifold incorrectly or got the orientation backward. These are mechanical errors but they're expensive on exams.

Common Pitfalls

Orientation is the biggest source of mistakes. When you integrate a form over a chain, the orientation of the chain matters, and it's easy to flip it without noticing. The other common error is forgetting that the exterior derivative increases degree by one. Students sometimes apply d twice and expect zero without checking that the form is smooth enough, or they confuse d with the Lie derivative or the codifferential. These are distinct operators and mixing them up leads to completely wrong answers. Another issue is the difference between closed and exact. A form can be closed everywhere on a domain but still not exact if the domain has nontrivial topology. Spivak makes you prove this several times using explicit counterexamples, and the solutions need to show the cohomology computation carefully. The de Rham cohomology of S^n is zero in all dimensions except 0 and n, and problems asking you to verify this require constructing explicit primitives or showing none exist. This is where the weaker solution manuals fail because they just state the result without the construction.

What Works

The most efficient approach is to work the problems in order and not skip the early ones even if they feel like review. The linear algebra chapter problems establish notation and conventions that Spivak uses throughout. If you skip them, you'll be confused later about whether a k-form is a section of the k-th exterior power or something else. The integration on chains chapter is where you learn to think about manifolds as domains of integration, which is essential for everything after. For the differential forms section, writing out the coordinate expression of every form you encounter helps. Spivak works in a very abstract style, but the computations are fundamentally about alternating tensors in coordinates. Expanding omega = sum f_I dx^I and applying d term by term removes most of the ambiguity. This is especially important when dealing with wedge products of forms with multiple terms, where the sign rules become easy to mess up under pressure. When you're stuck on a problem, check whether it's asking you to construct something explicitly or to prove existence. Construction problems usually have a direct computational path. Existence problems often require an argument by contradiction or an appeal to a previously proved theorem. I've found that misclassifying the problem type is another common source of wasted time. You'll spend an hour trying to compute something when a one-paragraph topological argument solves it.

Limitations of Available Solutions

Most freely available solutions online cover only the easier problems. The harder ones, particularly in Chapter 4 where Spivak discusses applications to vector calculus and some basic Riemannian geometry, are sparsely covered. If you need complete coverage, you're better off sourcing solutions from a specific professor's course page or working through a supplementary text like Tu's "An Introduction to Manifolds," which covers similar material with more detailed examples and exercises with solutions. Even the good solutions have a tendency to assume familiarity with point-set topology that Spivak doesn't always build up. Compact support, partitions of unity, and submanifold definitions appear without full justification. If you're reading Spivak without a topology background, you'll hit gaps that no solution manual fills. In that case, pairing it with Lee's "Introduction to Smooth Manifolds" for the foundational material makes a significant difference in how quickly you can work through the problem sets. The tradeoff is time. Doing both texts properly takes considerably longer than Spivak alone, but the completeness is worth it if you're preparing for comprehensive exams or plan to use this material in research.