Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

Let $f$ be a homomorphism defined on a finite group $G$, and let $H$ is the subgroup of $G$. Then show that $$ \left [f(G) : f(H)\right] \text{ divides } \left [G : H\right].$$

I know $$\left [G : H\right] = o(G)/o(H);$$

if $o(H) = n$ then $o(G) = kn$ so $o(G)/o(H) = k$.

Likewise $$ \left [f(G) : f(H)\right] = o(f(G)) / o(f(H)).$$

I am stuck here.

Is this the right way of doing this problem?

share|cite|improve this question
What is $g$? An element of $G$ would be a fishy thing to say. – user21436 Apr 6 '12 at 9:38
Is $f(g)$ a typo for $f(G)$? – Brian M. Scott Apr 6 '12 at 9:40
@Brian Wouldn't that render this problem trivial? Not that I know of a way to make it non-trivial, : ) – user21436 Apr 6 '12 at 9:41
@BrianM.Scott NO, f(g) is not a typo for f(G) – faisal Apr 6 '12 at 9:50
@faisal Then what is $g$? – user21436 Apr 6 '12 at 9:54

You are near, you only need to apply the first isomorphism theorem, and deduce from it that $$ |f(G)|\cdot|\ker f|=|G|$$ And $$ |f(H)|\cdot |\ker f\cap H|=|H|$$ Then, you have only to notice that $\ker f\cap H$ is a subgroup of $\ker f$, and therefore $$\frac{|\ker f|}{|\ker f\cap H|}$$ is an integer by Lagrange's theorem. Thus divisibility is obtained.

share|cite|improve this answer
How did you get the second step? – faisal Apr 6 '12 at 11:06
Note that $f:G\rightarrow G$ when restricted to $H$ gives an homomorphism $f_{|H}:H\rightarrow G$ whose kernel is $ker f_{|H}=ker f\cap H$. – Josué Tonelli-Cueto Apr 6 '12 at 11:12
How does your last step shows that the [f(G):f(H)] divides [G:H]? – faisal Apr 6 '12 at 11:31
When dividing the two wqualitis, you have $$|f(G):f(H)|\frac{|ker f|}{|ker f\cap H|}=\frac{|f(G)|}{|f(H)|}\frac{|ker f|}{|ker f\cap H|}=\frac{|G|}{|H|}=|G:H|$$ from where divisibility is obtained since $\frac{|ker f|}{|ker f\cap H|}$ is an integer. – Josué Tonelli-Cueto Apr 6 '12 at 11:32
Sorry for the typo. – Josué Tonelli-Cueto Apr 6 '12 at 11:37

We can show something a little more general. Let $G$ and $K$ be two groups, and let $\phi:G\rightarrow K$ be a surjective homomorphism. If $H$ is a subgroup of $G$ with finite index $n$, then $\phi(H)$ has finite index in $K$ and its index divides $n$.

Consider the left cosets of $\phi(H)$. Since $\phi$ is surjective, each coset of $\phi(H)$ is of the form $\phi(x)\phi(H)$ for some $x\in G$. But $\phi(x)\phi(H)=\phi(xH)$ as subsets of $K$. Thus the cosets of $\phi(H)$ are images of cosets of $H$ under $\phi$. This can be interpreted as $\phi$ inducing a surjective map from the cosets of $H$ onto the cosets of $\phi(H)$. Let's give this function a name $\phi^*$. To be explicit, $\phi^*$ is the function from the set of cosets of $H$ onto the cosets of $\phi(H)$ which sends $xH$ to $\phi(xH)=\phi(x)\phi(H)$.

In general, if $X$ is a finite set, and there is a surjective set function from $X$ onto a set $Y$, then $Y$ is also finite.

Thus there only finitely many cosets of $\phi(H)$ and is thus of finite index in $K$.

Now we need to show that $[K:\phi(H)]$ divides $[G:H]$. This essentially follows from the fact that $\phi^*$ dishes out the cosets of $H$ evenly onto the cosets of $\phi(H)$.

In general, if $X$ and $Y$ are two finite sets, and $f:X\rightarrow Y$ is a surjective set function such that $f^{-1}(y_1)$ is equinumerous with $f^{-1}(y_2)$ for any $y_1$ and $y_2\in Y$, then the cardinality of $Y$ divides the cardinality of $X$.

Now we need to show that this lemma can be applied to our case here. That is, we need to show that $(\phi^*)^{-1}(C_1)$ is equinumerous to $(\phi^*)^{-1}(C_2)$ for any two cosets $C_1$ and $C_2$ of $\phi(H)$. It suffices to show that $(\phi^*)^{-1}(C)$ is equinumerous to $(\phi^*)^{-1}(\phi(H))$ for each coset $C$ of $\phi(H)$.

But what is $(\phi^*)^{-1}(\phi(H))$? With all the $\phi$s it can be difficult to untangle what it is. $(\phi^*)^{-1}(\phi(H))$ is the set of all cosets of $H$ of the form $xH$ where $x\in\ker \phi$. This is because $\phi(xH)=\phi(x)\phi(H)=e_K\phi(H)=\phi(H)$. Thus $\phi^*(xH)=\phi(H)$ and $xH\in (\phi^*)^{-1}(\phi(H))$. This is one inclusion, you can show the other.

Now let $C$ be a coset of $\phi(H)$. Then we have $C=\phi(x)\phi(H)$ for some $x\in G$. We have that $(\phi^*)^{-1}(C)$ is the set of all cosets of $H$ of the form $xyH$ where $y\in\ker\phi$. This set is in bijection with the set $(\phi^*)^{-1}(\phi(H))$ which we determined was the set of all cosets of $H$ of the form $yH$ with $y\in\ker H$. And the bijection is given by sending the coset $yH$ to the coset $xyH$, i.e., multiply each coset on the left by $x$. It's inverse is given by multiplying on the left by $x^{-1}$.

We have shown what we needed to use the lemma, thus $[K:\phi(H)]$ divides $[G:H]$.

share|cite|improve this answer

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.