PlanetMath (more info)
 Math for the people, by the people.
Encyclopedia | Requests | Forums | Docs | Wiki | Random | RSS  
Login
create new user
name:
pass:
forget your password?
Main Menu
Owner confidence rating: High Entry average rating: No information on entry rating
[parent] proof of snake lemma (Proof)

Suppose we are given a commutative diagram

$\displaystyle \begin{xy} *!C\xybox{ \xymatrix{ 0\ar[r] & A_1\ar[d]_{\alpha}\ar[... ...\ar[r] & 0 \ 0\ar[r] & A_2\ar[r] & B_2\ar[r] & C_2\ar[r] & 0 \ } } \end{xy}$
with exact rows. We wish to prove that the sequence
$\displaystyle 0\to\ker\alpha\to\ker\beta\to\ker\gamma\to {\mathrm{coker}}\alpha\to{\mathrm{coker}}\beta\to{\mathrm{coker}}\gamma\to 0$
is exact.1

First we claim that if any square

$\displaystyle \begin{xy} *!C\xybox{ \xymatrix{ X_1\ar[d]_{\varphi}\ar[r] & Y_1\ar[d]_{\psi} \ X_2\ar[r] & Y_2 } } \end{xy}$
is commutative, then there are well-defined morphisms $ \ker\varphi\to\ker\psi$ and $ {\mathrm{coker}}\varphi\to{\mathrm{coker}}\psi$. For example, if $ x\in\ker\varphi$, then the square
$\displaystyle \begin{xy} *!C\xybox{ \xymatrix{ x\ar@{\vert->}[d]_{\varphi}\ar@{\vert->}[r] & y\ar@{\vert->}[d]_{\psi} \ 0\ar@{\vert->}[r] & 0 } } \end{xy}$
must commute, and so the image of $ x$ in the top row must be in $ \ker\psi$. The proof of the claim for cokernels is similar. Thus we have two sequences,
$\displaystyle 0\to\ker\alpha\to\ker\beta\to\ker\gamma$ and $\displaystyle {\mathrm{coker}}\alpha\to{\mathrm{coker}}\beta\to{\mathrm{coker}}\gamma\to 0,$
each of which inherits being a complex from the original diagram.

Suppose $ x\in\ker\beta$ is sent to $ 0\in\ker\gamma$. By exactness, $ x$ has a preimage $ x'\in A_1$. Because the diagram

$\displaystyle \begin{xy} *!C\xybox{ \xymatrix{ x'\ar@{\vert->}[d]_{\alpha}\ar@{\vert->}[r] & x\ar@{\vert->}[d]_{\beta} \ y'\ar@{\vert->}[r] & 0 } } \end{xy}$
is commutative and the bottom morphism is injective, $ y'=0$ and so $ x'\in\ker\alpha$. So the sequence
$\displaystyle 0\to\ker\alpha\to\ker\beta\to\ker\gamma$
is exact. The proof of the claim for the cokernel sequence is similar.

So now all we need to do is find a connecting morphism $ \ker\gamma\to{\mathrm{coker}}\alpha$ such that the resulting sequence is exact at both of those points.

Suppose $ x''\in\ker\gamma$. Then $ x''$ has at least one preimage in $ B_1$. So let $ x$ and $ \widehat{x}$ be preimages of $ x''$. Thus $ \widehat{x}-x\mapsto 0$ and so by exactness has a preimage $ x'\in A_1$. By commutativity of the diagram, $ \beta(x)$ has a preimage $ y'$, which is unique by injectivity of the morphism $ A_2\to B_2$. But we know that the square

$\displaystyle \begin{xy} *!C\xybox{ \xymatrix{ x'\ar@{\vert->}[d]_{\alpha}\ar@{... ...}[d]_{\beta} \ \alpha(x')\ar@{\vert->}[r] & \beta(\widehat{x}-x) } } \end{xy}$
is commutative. We wish to define $ \ker\gamma\to{\mathrm{coker}}\alpha$ by $ x''\mapsto y'+{\mathrm{im}}\alpha$. Observe that
$\displaystyle y'+\alpha(x')\mapsto\beta(x)+\beta(\widehat{x}-x)=\beta(\widehat{x}),$
and so the choice of preimage of $ x''$ does not affect which cokernel element we ultimately select. So now we have our connecting morphism. By applying this definition we see that
$\displaystyle \ker\beta\to\ker\gamma\to{\mathrm{coker}}\alpha\to{\mathrm{coker}}\beta$
is a complex.

Suppose $ x''\in\ker\gamma$ is sent to 0 by the connecting morphism. Thus we have a diagram

$\displaystyle \begin{xy} *!C\xybox{ \xymatrix{ x'\ar@{\vert->}[d]_{\alpha} & x\... ...[d]_{\gamma} \ y'\ar@{\vert->}[r] & \beta(x)\ar@{\vert->}[r] & 0 } } \end{xy}$
which is commutative. Let $ \widehat{x}$ be the image of $ x'$ under the morphism $ A_1\to B_1$. Exactness of the diagram implies that $ x-\widehat{x}$ is a preimage of $ x''$. But $ \beta(x-\widehat{x})=0$. So the kernel-cokernel sequence is exact at $ \ker\gamma$. The proof that it is exact at $ {\mathrm{coker}}\alpha$ is similar.



Footnotes

... exact.1
This proof was reconstructed without any notes, but the style of the proof is influenced by a presentation by Edgar Enochs of the zig-zag lemma.


Anyone with an account can edit this entry. Please help improve it!

"proof of snake lemma" is owned by mps.
(view preamble)

View style:

See Also: long exact sequence (of homology groups)

Other names:  proof of zig-zag lemma, proof of serpent lemma
Keywords:  exact sequence, kernel-cokernel sequence

This object's parent.
Log in to rate this entry.
(view current ratings)

Cross-references: implies, points, injective, preimage, diagram, complex, similar, cokernels, image, morphisms, well-defined, zig-zag lemma, proof, commutative diagram
There is 1 reference to this entry.

This is version 1 of proof of snake lemma, born on 2004-02-14.
Object id is 5578, canonical name is ProofOfSnakeLemma.
Accessed 6184 times total.

Classification:
AMS MSC18G35 (Category theory; homological algebra :: Homological algebra :: Chain complexes)

Pending Errata and Addenda
None.
Discussion
Style: Expand: Order:
forum policy

No messages.

Interact
post | correct | update request | add example | add (any)