Global Analysis

Differential forms in Analysis, Geometry und Physics

I. Agricola, Th. Friedrich

We thank all colleagues who pointed out mistakes and misprints. Any more hints are welcome, do not hesitate to send me an email!

Known mistakes so far:

p. 6 ff: The volume form should be defined with a different normalisation, thus leading to several corrections:

l.-9: Put a factor $(-1)^q$ in front of the determinant.

l. -4: The volume form should be defined as $\sigma_1\wedge\ldots\wedge\sigma_n$ independent of the signature, i.e. the factor $(-1)^q$ should be discarded.

p. 7, l. -9 and -11: Again, remove both factors $(-1)^q$.

p. 8, l. 2,3: remove the signs $(-1)^q$ and $(-1)^{2q}$ up front, respectively. Notice that the final result, l. 4 remains the same!

p. 15, l. 8: the $dx^i$ on the right hand side should be $dx^j$.

p. 30, l. 7 and 8: it should be twice $dx1\wedge\ldots\wedge dx^k$ (i.e., the upper index $n$ should be $k$).

p. 24, l. -10: This formula requires a proof, of course (not difficult). Since it's a useful one, I guess we should have stated it as a lemma.

p. 31, proof of Thm 7: same thing, every $dx^n$ should be replaced by $dx^k$.

p. 32, l. 11: Integrate over $[0,1]^{k+1}$ instead of $[0,1]^k$.

p. 33, l. 3 and 4: before the last integral sign in l.3 and the first integral sign in l.4, a factor $(-1)^{j+\alpha}$ is missing.

P. 35, l.3-4: The argument is right, albeit a bit obscure. The implicit function theorem is not needed, one just observes that $A$ cannot carry $(k+1)$-forms, and that's it.

p. 51, condition (2) of Thm 2: replace "bijective" by "injective".

p. 58, l. -1: the function "g" in this formula should be "h" (there is no function called g here).

p. 127, l. 13-15: in a), there is a typo, the equation should read $(e^{2t} - x^2) + x dx/dt =0$. b) ist ok, c) should be deleted (not integrable by any elementary means).

p. 127, l. -3: Add the initial condition that $A$ should coincide with a given matrix $A_0$ for $t=0$, $A(0)=A_0$.

p. 150, Theorem 13 (2): the second $V_1$ in the subscript should be $V_2$.

p. 184, Theorem 42 (2): the second $V_1$ in the subscript should be $V_2$.

p. 178, l. 7/9: replace "meridian" by "parallel circle" (translation mistake).

p.223, l. 16: "and equality holds if $G$ is connected" (translation mistake).

p. 226, exercise 5: Add assumption that G is connected.

p.227, exercise 8 b): replace fraction by $\frac{az+b}{\bar{b}z+\bar{a}}$ (the coefficients in the denominator should be interchanged).


I. Agricola / Th. Friedrich, last update: 30.10.2018 agricola@mathematik.uni-marburg.de