Supplementary Notes on Direct Products of Groups and

the
Fundamental Theorem of Finite Abelian Groups

Definition. Let G1 , G2 , . . . Gn be groups. We can make the Cartesian product

G1 × G2 × · · · × Gn into a group via coordinate-wise multiplication:
(g1 , g2 , . . . , gn ) · (g10 , g20 , . . . , gn0 ) = (g1 g10 , g2 g20 , . . . , gn gn0 ).
It is absolutely routine to check that with the above operation, G1 × G2 × · · · × Gn
is a group, called the (external) direct product of G1 , G2 , . . . Gn .
In case G1 , G2 , . . . , Gn are abelian groups written additively, then it is more
customary to write the direct product as G1 ⊕ G2 ⊕ · · · ⊕ Gn , and refer to this as
an (external) direct sum.
To “internalize” this notion, recall first that if G is a group with normal sub-
groups H, K / G, then the product HK is a subgroup of G. More generally, if we
have normal subgroups H1 , H2 , . . . , Hn / G, then the product H1 H2 · · · Hn is also a
subgroup of G. Sometimes this represents G as a direct product, as follows:

Theorem 1. [Internal Direct Product Theorem] Let H1 , H2 , . . . , Hn be normal

subgroups of G, and assume that
(i) G = H1 H2 · · · Hn ,
(ii) For each i = 1, 2, . . . , n, we have Hi ∩ (H1 · · · Hi−1 Hi+1 · · · Hn ) = {e}.
Then G ∼ = H1 × H2 × · · · × Hn ; in fact the mapping
φ : H1 × H2 × · · · × Hn −→ G
given by φ(h1 , h2 , . . . , hn ) = h1 h2 · · · hn is an isomorphism. When this happens,
we shall slur the distinction between “internal” and “external” and simply write
G = H1 × H2 × · · · × Hn , keeping in mind that H1 , H2 , . . . , Hn are actually
subgroups of G.

Proof. We claim first that if hi ∈ Hi and hj ∈ Hj with i 6= j, then hi hj = hj hi . But

this is easy: since both Hi and Hj are normal in G, we see that the “commutator”
hi hj h−1 −1
i hj ∈ Hi ∩ Hj . But condition (ii) above clearly implies that Hi ∩ Hj = {e},
and so it follows that hi hj = hj hi , as required. Now define
φ : H1 × H2 × · · · × Hn −→ G
by setting φ(h1 , h2 , . . . , hn ) = h1 h2 · · · hn . Because of what we just proved above, φ
is a homomorphism. By condition (i), φ is onto. Finally, if (h1 , h2 , . . . , hn ) ∈ ker φ,
then h1 h2 · · · hn = e implies that h−1 i = h1 · · · hi−1 hi+1 · · · hn ∈ Hi ∩(H1 · · · Hi−1 Hi+1 · · · Hn ) =
{e}, i.e., hi = e. Since i was arbitrary, we conclude that (h1 , h2 , . . . , hn ) =
(e, e, . . . , e), and so φ is also injective. This proves the theorem.

Please note that for abelian groups written additively, there is an obvious analogue
of the above result.

We shall now turn our attention to finite abelian groups. Basically we shall
try to prove that every finite abelian group can be decomposed into cyclic groups.
Without further stipulations, we can’t really hope for a uniqueness result, basically
because of the

Theorem 2. [Chinese Remainder Theorem] Let m and n be relatively prime

integers and let Zm , Zn , Zmn be cyclic groups of the given orders. Then Zm × Zn ∼
=
Zmn .

Proof. Let Zmn be generated by the element z. Then the elements z1 = z n , z2 = z m

have orders m and n, respectively. Set H1 = hz1 i, H2 = hz2 i (cyclic groups of orders
m and n, respectively), and note that the hypotheses of the above theorem are met,
i.e., Zmn ∼
= H1 × H2 .
Notice that the Chinese Remainder Theorem allows us to consolidate many direct
product decompositions of cyclic groups into fewer such products, or to reconfigure
direct products, as the following examples indicate:
Z6 × Z7 ∼
= Z42 ,
Z2 × Z5 × Z7 ∼
= Z70 ,
Z6 × Z35 ∼
= Z2 × Z3 × Z35 ∼
= Z2 × Z105 .

Now let A be a finite abelian group, which we continue to write multiplicatively.

Let |A| = pa11 pa22 · · · pakk be the decomposition of the order of A into distinct prime
factors. We set n = |A|, and set
n n n
n1 = a1 , n2 = a2 , . . . , nk = ak .
p1 p2 pk
Note that while the integers n1 , n2 , . . . , nk are not pairwise relatively prime, they
are collectively relatively prime, meaning that there is no integer greater than 1
than divides all of them. This implies, since Z is a principal ideal domain, that
the ideal (n1 , n2 , . . . , nk ) = (1) = Z, which further implies that there exist integers
s1 , s2 , . . . , sk ∈ Z, with s1 n1 + s2 ns + · · · sk nk = 1.
 Using the above, we can prove the following fundamental result.

Theorem 3. [Primary Decomposition Theorem] Let A be a finite abelian group

of order |A| = n = pa11 pa22 · · · pakk , and let Pi be the pi -Sylow subgroup of A, i =
1, 2, . . . , k. Then A = P1 × P2 × · · · × Pk .

Proof. We shall show that the hypotheses of Theorem 1 are satisfied. Note first
that since A is abelian, then every subgroup of A is normal. In particular, the Sylow
subgroups P1 , P2 , . . . , Pk are all normal. Furthermore, if a ∈ A is an element
whose order is a power of pi , then it follows from Sylow’s Theorem (the “covering”
part) that a is contained in a pi -Sylow subgroup. But there is only one pi -Sylow
subgroup since all are conjugate and any one is normal. Therefore, it follows that
a ∈ Pi .
Next, let a ∈ A be an arbitrary element. Using the result above, we have
s1 n1 + s2 ns + · · · sk nk = 1, for suitable s1 , s2 , . . . , sk ∈ Z, and so
a = a1 = as1 n1 +s2 ns +···sk nk = as1 n1 as2 n2 · · · ask nk .
ai
However, (asi ni )pi = (an )si = esi = e and so asi ni ∈ Pi , for i = 1, 2, . . . , k. This
already proves that A = P1 P2 · · · Pk . That Pi ∩ (P1 · · · Pi−1 Pi+1 · · · Pk ) = {e} should
be obvious. This proves the theorem.

At this stage, we see that the decomposition of a finite abelian group into a
direct product of cyclic groups can be accomplished once we show that any abelian
p-group can be factored into a direct product of cyclic p-groups.

Theorem 4. Let P be a finite abelian p-group of order pm . Then there exist

powers e1 , e2 , . . . , er with e1 ≥ e2 ≥ · · · ≥ er such that P ∼
= Zpe1 × Zpe2 × · · · × Zper .
Proof. We shall argue by induction on the order of P . First, let x1 ∈ P be an
element of maximal order pe1 in P . Clearly, e1 ≤ m. Set Z1 = hx1 i. By induction,
the quotient group A/Z1 is a direct product of cyclic groups:
A/Z1 = hy2 Z1 i × hy3 Z1 i × · · · × hyr Z1 i,
where o(y2 Z1 ) = pe2 , . . . , o(yr Z1 ) = per , e2 ≥ e3 ≥ · · · ≥ er . Note that since x1 has
maximal order in A, it follows that e1 ≥ e2 .
Now we shall make some adjustments. Fix i, where 2 ≤ i ≤ r, and consider
ei
the cyclic subgroup hyi Z1 i ≤ A/Z1 . Since o(yi Z1 ) = pei we have yip ∈ Z1 , say
ei e −e ei e −e e1
yip = xk1i . Thus (xk1i )p 1 i = (yip )p 1 i = yip = e and so pe1 |ki pe1 −ei . Thus pei |ki ,
ei ei
say ki = li pei , so yip = xk1i = x1li p . Now the adjustment: we set xi = yi x−l 1 ,
i
ei ei ei
so xpi = yip x−l 1
ip
= e, and no smaller exponent kills xi . Doing this for each
i = 2, 3, . . . , r produces elements x2 , x3 , . . . , xr ∈ P such that o(xi ) = pei , i =
2, 3, . . . , r. Set Z2 = hx2 i, Z3 = hx3 i, . . . , Zr = hxr i. To finish the proof, we need
to show:
(i) P = Z1 Z2 · · · Zr ,
(ii) Zi ∩ (Z1 · · · Zi−1 Zi+1 · · · Zr ) = {e}.
Let x ∈ P . From the product decomposition above for A/Z1 , we have powers
f2 , f3 , . . . , fr with
(y2 Z1 )f2 (y3 Z1 )f3 · · · (yr Z1 )fr = xZ1 ,
so it follows that y2f2 · · · yrfr = xx0 , for some x0 ∈ Z1 . Thus, since yi = xi xl1i , we get
xf22 xf33 · · · xfrr = xx00 ,
for some x00 ∈ Z1 . Now let f1 be such that xf11 = (x00 )−1 , and conclude that
xf11 xf22 · · · xfrr = x, proving part (i).
To prove part (ii), note that it suffices to prove that if
xf11 xf22 · · · xfrr = e,
then pei |fi , for i = 1, 2, . . . , r. Again, since xi = yi x−l 1 , i = 2, 3, . . . , r we get an
i

equation of the form

x0 y2f2 y3f3 · · · yrfr = e,
for some x0 ∈ Z1 , from which we conclude that
(y2 Z1 )f2 (y3 Z1 )f3 · · · (yr Z1 )fr = eZ1 ;
this implies that pei |fi , i = 2, 3, . . . , r. But then it follows immediately that xf11 = e,
and so pe1 |f1 , as well. This completes the proof.

If we use the above theorem, together with the Chinese Remainder Theorem, we
get the following ”existence” theorem.

Theorem 5. [Existence of Invariant Factors] Let A be a finite abelian group, and

let n = |A| be the order of A. Then there exists a decreasing sequence n1 ≥ n2 ≥
· · · ≥ ns , with ns |ns−1 , ns−1 |ns−2 , . . . , n2 |n1 , and cyclic subgroups Z1 , Z2 , . . . , Zs
of orders n1 , n2 , . . . , ns , respectively with A = Z1 × Z2 × · · · × Zs . The numbers
n1 , n2 , . . . , ns are called invariant factors of A.

It is pretty easy to see how the direct product decompositions of the Sylow
subgroups in the Primary Decomposition Theorem can be synthesized into a direct
product decomposition of the type indicated above. For example, if we had a
group A = P1 × P2 × P3 × P4 , (direct decomposition into Sylow subgroups), with
P1 ∼
= Z8 × Z4 , P2 ∼
= Z9 × Z3 , P3 ∼
= Z125 × Z5 × Z5 × Z5 , and P4 ∼
= Z11 , (note here
that I’ve written Zn for a cyclic group of order n; sorry about being somewhat
inconsistent in my usage), then we would have

A∼
= Z8 × Z4 × Z9 × Z3 × Z125 × Z5 × Z5 × Z5 × Z11 ,
and so a reorganized direct product decomposition would look like:

A∼
= Z8·9·125·11 × Z4·3·5 × Z5 × Z5 ,
and so the “invariant factors” are n1 = 8 · 9 · 125 · 11 = 99, 000, n2 = 4 · 3 · 5 =
60, n3 = 5, n4 = 5.

To obtain a uniqueness result for the direct product decomposition, we shall first
discuss the question of “cancellation” in a direct product decomposition. Basically,
the question is this: If we have an isomorphism of direct product groups,
G1 × H1 ∼
= G2 × H2 ,
with H1 ∼= H2 , is it true that G1 ∼
= G2 ? The naive student will quickly respond with
an affirmative answer; yet the answer is no, making the cancellation question rather
subtle. As a counterexample to the cancellation problem, consider the infinite direct
product of the infinite cyclic group Z:
Z × (Z × · · · ) ∼
= Z × Z × (Z × · · · );
note that if we could cancel off the right hand factors, when we would end up with
Z∼= Z × Z, which is false (Z is cyclic, but Z × Z is not). Therefore we must treat
this question with some care.
Before turning to the main cancellation result, let’s state and prove a very simple
lemma, and recall another important result.

Lemma. Suppose we have groups and normal subgroups: M / A, N / B. Then

M × N / A × B, and
(A × B)/(M × N ) ∼
= (A/M ) × (B/N ).

Proof. Clearly the first statement above is true. As for the isomorphism, define
φ : A × B → (A/M ) × (B/N ), by setting φ(a, b) = (aM, bN ). Note that φ is a
surjective homomorphism with kernel M × N .
Theorem 6. [Noether Isomorphism Theorem] Let G be a group and let H, K ≤ G,
with K / G. Then
HK/K ∼ = H/(H ∩ K).
In particular, if we have H, K / G and H ∩ K = {e}, then (H × K)/K ∼
= H.
We turn now to the main result. I’ve pretty much followed the treatment given
by R. Hirshon, in his paper Decomposition of groups, in the American Mathe-
matical Monthly, vol. 76 (1969), pp. 1037-1039.

Theorem 7. [Cancellation of Finite Groups] Assume that we have an isomorphism

H ×L∼= K × G, where L ∼ = G and are finite groups. Then H ∼= K.
Proof. We shall regard the above direct product as internal. Assume that φ :
H × L → K × G is an isomorphism. Thus, if K 0 = φ(H), G0 = φ(L), then we
have K 0 × G0 = K × G; the task, then, is to show that K ∼ = K 0 . We now set
X = K × G = K 0 × G0 ; since we have regarded the direct products as internal,
then K, K 0 , G, G0 are all normal subgroups of X, and G ∼
= G0 .
We divide the proof into two steps.
Step 1. If K 0 ∩ G = {e}, or if K ∩ G0 = {e}, we shall show that K 0 ∼ = K. In
the first case, we have K 0 G = K 0 × G ≤ X = K 0 × G0 . However, |G| = |G0 |
and so the index of K 0 in both K 0 × G and K 0 × G0 is the same. Therefore,
K 0 × G = K 0 × G0 = K × G, and so
K0 ∼
= (K 0 × G)/G = (K × G)/G ∼ = K.
The argument is similar in case K ∩ G0 = {e}.
Step 2. Induction on |G|.) Set M = K ∩ G0 , N = K 0 ∩ G, and note that M, N / X,
and that M ∩ N = {e}. We have
X/(M × N ) = (K × G)/(M × N ) ∼ = (K/M ) × (G/N ),
and
X/(M × N ) = X/(N × M ) = (K 0 × G0 )/(N × M )
∼
= (K 0 /N ) × (G0 /M ).
Therefore (K/M ) × (G/N ) ∼
= (K 0 /N ) × (G0 /M ). But G ∼
= G0 and so
G × (K/M ) × (G/N ) ∼ = G0 × (K 0 /N ) × (G0 /M ). (*)
We now work on the right hand side of (*):
G0 × (K 0 /N ) × (G0 /M ) ∼
= ((G0 × K 0 )/N ) × (G0 /M )
= ((G × K)/N ) × (G0 /M )
∼
= (G/N ) × K × (G0 /M )
∼
= K × (G/N ) × (G0 /M ),
In other words,
G0 × (K 0 /N ) × (G0 /M ) ∼
= K × (G/N ) × (G0 /M ). (**a)
In an entirely similar way,
G × (K/M ) × (G/N ) ∼
= K 0 × (G0 /M ) × (G/N ). (**b)
Now apply (*) to the left hand sides of (**a) and (**b) and infer that
K × (G/N ) × G0 /M ) ∼
= K 0 × (G/N ) × (G0 /M ).
By induction we may successively cancel (G0 /M ) and then (G/N ), resulting in
the desired isomorphism: K ∼
= K 0.

Corollary. [Uniqueness of Invariant Factors] Let A be a finite abelian group

with invariant factors n1 , n2 , . . . , ns , and m1 , m2 , . . . , mt . Then s = t and mi =
ni , i = 1, 2, . . . , s.

Proof. By assumption, we have

Zn1 × Zn2 × · · · × Zns ∼

=A∼
= Zm1 × Zm2 × · · · × Zmt ,
where ni+1 |ni , i = 1, 2, . . . , s − 1, and mj+1 |mj , j = 1, 2, . . . , t − 1. Note that
if n1 6= m1 , say m1 < n1 , then since A ∼ = Zm1 × Zm2 × · · · × Zmt , we conclude
m1
that a = e for all a ∈ A. However, as A contains a subgroup isomorphic with
Zn1 , then A has an element of order n1 , a contradiction. Therefore, n1 = m1 .
Now use induction on |A| and apply Theorem 7, to cancel off the Zn1 factors
and infer that the remaining invariant factors all agree.

Example 1. The possible isomorphism types of abelian groups of order 16 are:

Z16 , Z8 × Z2 , Z4 × Z4 , Z4 × Z2 × Z2 , Z2 × Z2 × Z2 × Z2 .

Example 2. The possible isomorphism types of abelian groups of order 100 are
Z100 , Z50 × Z2 , Z20 × Z4 , Z10 × Z10 .