MASTERS EXAMINATION IN MATHEMATICS

PURE MATH OPTION, Fall 2018

Full points can be obtained for correct answers to 8 questions. Each numbered question
(which may have several parts) is worth 20 points. All answers will be graded, but the score
for the examination will be the sum of the scores of your best 8 solutions.

Use separate answer sheets for each question. DO NOT PUT YOUR NAME ON
YOUR ANSWER SHEETS. When you have finished, insert all your answer sheets into
the envelope provided, then seal it.

Any student whose answers need clarification may be required to submit to an oral exam-
ination.

Algebra

A1. (a) List all non-isomorphic Abelian groups of order 108.

(b) List all non-isomorphic Abelian groups of order 108 that have an element of order 9.
(c) How many subgroups of order 3 does Z36 × Z3 have? Justify your answer.

Solution.

(a) By the classification theorem for Abelian groups, since 108 = 22 ×33 , we list all possible
products of a group of order 4 and a group of order 27, which in turn must be products of
cyclic groups. Thus the possibilities are

• Z2 × Z2 × Z27
• Z2 × Z2 × Z3 × Z9
• Z2 × Z2 × Z3 × Z3 × Z3
• Z4 × Z27
• Z4 × Z3 × Z9
• Z4 × Z3 × Z3 × Z3

(b) Letting G = H × K where |H| = 4 and |K| = 27, the order of an element (a, b) is the
least common multiple of the order of a and the order of b. So If |(a, b)| = 9, one has a = 0
and b is an element of K of order 9. This can happen if K = Z3 × Z9 or K = Z27 , and there
are no restrictions on H. Thus the list is

• Z2 × Z2 × Z27
• Z2 × Z2 × Z3 × Z9
• Z4 × Z27
• Z4 × Z3 × Z9
1
 (c) We count elements of order 3. Suppose (a, b) ∈ Z36 × Z3 is of order 3. Then (3a, 3b) =
(0, 0). Thus a = 0, 12, or 24, and b = 0, 1, or 2. Since (a, b) 6= (0, 0), we have 8 possibilities for
(a, b). Each subgroup of order 3 contains two of them, with no overlaps between subgroups,
so there are a total of 4 subgroups of order 3.
√
3
√
3
√
3
√
3
A2. Find the minimal polynomial of 2+ 4 over Q and determine [Q( 2 + 4) : Q].

Solution.
√3
√3
√
Let α = 2 + 4. Since α ∈ Q( 3 2) and is expected to √ be irrational,
√ the expectation
√ √ is
3 3 3 3 3 3
that [Q(α) : Q] = 3. Expanding directly we see that α = ( 2 + 4) = 2 + 6( 2 + 4) +
4 = 6 + 6α, so that we have
α3 − 6α − 6 = 0
By the Eisenstein Irreducibility Criterion with p = 2 or p = 3, we see that the polynomial
α3 − 6α − 6 is irreducible. Hence [Q(α) : Q] = 3 and α has minimal polynomial α3 − 6α − 6
over Q.

A3. (a) Let R be a commutative ring with prime characteristic p. Show that ϕ : R → R,
defined for any x ∈ R by ϕ(x) = xp , is a ring homomorphism.
(b) Give an example of a commutative ring R with characteristic 4 such that
ϕ : R → R, defined for any x ∈ R by ϕ(x) = x4 , is not a ring homomorphism. Justify your
answer.

Solution.

(a) First, φ(xy) = (xy)p = xp y p = φ(x)φ(y) by commutativity of φ. Secondly, φ(1) = 1p =

1. Lastly, we have
p
p
X p!
(x + y) = xi y p−i
i=0
i!(p − i)!
p!
Since p divides the numerator of i!(p−i)! but not the denominator whenever i 6= 0 or p, the
i p−i
coefficient of x y in characteristic p for such terms is zero and we have

(x + y)p = xp + y p

Thus we have φ(x + y) = φ(x) + φ(y), and we have verified that φ is a ring homomorphism.

(b) Let R = Z4 [x]. Then φ(x+1) = (x+1)4 = x4 +4x3 +6x2 +4x+1 = x4 +2x2 +1, which
is different from x4 + 1 = φ(x) + φ(1). Thus in this situation φ is not a ring homomorphism.

Complex Analysis
2
C1. Suppose f (z) is analytic on a domain D. Show that if f (z) is also analytic on D, then
f is constant.

Solution. Write f (z) = u(x, y) + iv(x, y) in terms of its real and imaginary parts. Then
we have that f (z) = u(x, y) + iv(x, y) and f (z) = u(x, y) − iv(x, y) are both analytic on D.
So by the Cauchy-Riemann equations applied to f (z), we have ∂x u = ∂y v and ∂y u = −∂x v
on D, while by the Cauchy-Riemann equations applied to f (z), we have ∂x u = −∂y v and
∂y u = ∂x v on D. The equations ∂x u = ∂y v and ∂x u = −∂y v give ∂y v = ∂x u = −∂y v on D,
so that ∂y v = 0 on D. Since ∂x u = ∂y v, we also have ∂x u = 0 on D.
The other two equations similarly give ∂y u = ∂x v = 0 on D. Hence ∇u and ∇v are both
identically zero on D. Hence u and v are constant functions, implying f (z) is constant on
D.

C2. Find all z ∈ C such that the principal value of z i is equal to i.

Solution. By definition, z i = eiLog(z) . Writing z = reiθ for θ ∈ (−π, π], we have Log(z) =
πi πi
ln r + iθ. Thus z i = eiLog(z) = ei ln r−θ = e−θ ei ln r . Also, i = e 2 . For e−θ ei ln r to equal e 2 ,
π
we must have e−θ = 1 and ln r = π2 + 2πk for some k ∈ Z. Hence θ = 0 and r = e 2 +2πk for
π
some k ∈ Z. We conclude that the z for which z i = i are exactly the z of the form e 2 +2πk
for some k ∈ Z.

C3. Using residues, determine the value of the following integral.

Z ∞
sin x
dx
−∞ x2 + 2x + 2

Solution. The integral is the imaginary part of

∞
eix
Z
dx
−∞ x2 + 2x + 2
The residue theorem applies to integrals of this form. The zeroes of the denominator occur at
x = −1 ± i, and only the zero −1 + i occurs above the x axis. Hence by the residue theorem,
eiz eiz
the integral is given by 2πi times the residue of z2 +2z+2 = (z−(−1+i))(z−(−1−i)) at z = −1 + i.
eiz
Since the pole is of order 1, the residue is given by inserting z = −1 + i into (z−(−1−i))
,
i(−1+i)
obtaining e 2i . So the integral is 2πi times this, or πei(−1+i) = πe e−i = πe (cos 1 − i sin 1).
Hence the original integral is the imaginary part of this, or − πe sin 1.

Number Theory

d = n2 .
Q
N1. Determine all natural numbers n such that d|n

Solution.
3
 Q
First note that the desired formula holds for n = 1. If n > 1, one can evaluate d|n d by
pairing up each divisor d with nd , obtaining a factor of n for each such pair. In the case where
Q τ (n)
n is not a perfect square, one obtains d|n d = n 2 in this fashion, where τ (n) denotes the
√
total number of divisors of n. If n is a perfect square, one can exclude n from this product
τ (n)−1 √
and pair like before, obtaining n 2 . Incorporating the n factor one once again obtains
Q τ (n)
the formula d|n d = n 2 . So the question reduces to determining the n for which τ (n) = 4.

To find a formula for τ (n), let n = pa11 ...pakk denote the prime factorization of n. then each
factor of n has a unique expression pb11 ...pbkk for 0 ≤ bi ≤ ai . Hence the total number of such
factors is ki=1 (ai + 1). Thus we need to find all n for which ki=1 (ai + 1) = 4. Since each
Q Q
ai ≥ 1, we must have k ≤ 1, and there are Q two possibilities k = 1 and a1 = 3, and k = 2 and
a1 = a2 = 1. Thus the n > 1 for which d|n d = n2 are the n for which n = p3 for a prime p
and the n of the form n = pq for distinct primes p and q.

N2. Find the rational solutions to x2 − y 2 = 1 by writing x − y = m/n and x + y = n/m.

Solution.

Following the hint, we write x − y = m/n and x + y = n/m where n and m are relatively
2 +m2
prime. Adding the two equations and dividing by 2 gives x = 21 ( m n
+m n
) = n2mn , and
1 n m n2 −m2
subtracting the two equations and dividing by 2 gives y = 2 ( m − n ) = 2mn . The fact that
2 +m2 2 −m2
n and m are relatively prime implies that the fractions n2mn and n2mn are also in reduced
form.
2 2
Conversely one readily computes that for any nonzero integers m and n, one has x = n2mn +m
2 −m2
and y = n2mn provide a solution to x2 − y 2 = 1. Hence the rational solutions to x2 − y 2 = 1
2 +m2 n2 −m2
are exactly the pairs (x, y) = ( n2mn , 2mn ) for m, n nonzero integers.

N3. State the Law of Quadratic Reciprocity. Determine all primes p such that x2 +3x+1 ≡ 0
mod p has a solution.

Solution.

The law of quadratic reciprocity states that if p and q are distinct odd primes one has
  
p q p−1 q−1
= (−1) 2 2
q p

Here ( pq ) denotes the Legendre symbol, which is equal to 1 if p is a quadratic residue mod q,
and is equal to −1 otherwise.

We would like to complete the square to analyze solutions to x2 + 3x + 1 ≡ 0 mod p. We

treat the case p = 2 separately due to the denominator of 2 arising in this process. Direct
verification for x = 0, 1 reveals that x2 + 3x + 1 ≡ 0 mod p has no solutions when p = 2.

4
 So we suppose p > 2. Completing the square, we have that if x2 +3x+1 ≡ x2 +(p+3)x+1 ≡
0 (mod p) we must have
 2
p+3 (p + 1)(p + 5)
x+ ≡ mod p
2 4
Since p is odd, this formula having a solution mod p is equivalent to the statement that
(p + 1)(p + 5) ≡ 5 is a quadratic residue mod p. One way for this to happen is for p = 5 in
which case x = 1 mod p solves. Otherwise, we can use quadratic reciprocity.

Letting q = 5, quadratic reciprocity tells us that

  
p 5
= (−1)p−1
5 p
Since p is odd, we have (−1)p−1 = 1 here. Thus 5 is a quadratic residue mod p exactly
when ( p5 ) = 1. We go through the list of possibilities and see that We have ( p5 ) = 1 if and
only if p ≡ 1, 4 mod 5. Thus our final answer is that x2 + 3x + 1 ≡ 0 mod p has a solution
if and only if p = 5, or p = 1 or 4 mod 5.

Real Analysis

R1. Let f be a differentiable function function on interval [0, 3] such that f 0 is continuous
on the same interval. Suppose we know that f (0) = 3, f (2) = −2, f (3) = 1.
• Show that there exists a point a ∈ [0, 3] such that f (a) = 0;
• Show that there exists a point b ∈ [0, 3] such that f 0 (b) = 3;
• Show that there exists a point c ∈ [0, 3] such that f 0 (c) = 1;

Solution. By the IVT, since f (0) = 3 and f (2) = −2 the function must attain any value in
between, in particular, there exists a point a ∈ [0, 2] such that f (a) = 0. By the mean value
theorem there exists b ∈ [2, 3] such that f 0 (b) = f (3)−f
3−2
(2)
= 3. Also, by the MVT there exists
f (2)−f (0)
a point d ∈ [0, 2] such that f (d) = 2−0 = − 2 . Applying the IVT to the derivative f 0 ,
0 5

we conclude that there should exist a point at which f 0 take intermediate value 1 between 3
and − 52 .

R2. Prove using the definition of a limit that

cos(n)
lim √ =0
n→∞ n+1
√
lim h + 12 = 4
h→4

Solution. Let us fix an  > 0. We need to find N such that for all n > N we have
 
 cos(n) 
√
 n + 1  < .

5
Using that | cos(n)| ≤ 1 we estimate
 
 cos(n)  1
√
 n + 1  ≤ √n + 1 .


We thus will aim at a stronger inequality

1
√ < .
n+1
Solving for n we obtain
1
n > 2 − 1.

1
So, if we choose N = [ 2 ] then for all n > N we guarantee the desired inequality.

Let us fix an  > 0. We need to find δ > 0 such that for all |h − 4| < δ we have
√
| h + 12 − 4| < ε.
√
Let us multiply and divide by the conjugate h + 12 + 4:
√ |h − 4|
| h + 12 − 4| = √
h + 12 + 4
If δ1 = 1, then h > 3, and hence we continue
|h − 4| |h − 4|
√ <√ < |h − 4|.
h + 12 + 4 15 + 4
We need to make the last expression < ε. So, if δ2 = ε, then the condition |h − 4| <
δ2 guarantees this. Picking δ = min{ε, 1} we guarantee with the final and intermediate
inequalities. This implies that if |h − 4| < δ, then
√
| h + 12 − 4| < ε.

R3.
R1
• Let f : [0, 1] → R be a continuous function such that 0 f (x)2 dx = 0. Prove that
f (x) = 0 for all x ∈ [0, 1].
• Give an example (proof
R 1 is2 not required) of a bounded Riemann integrable function
g : [0, 1] → R with 0 g(x) dx = 0 and g(m/n) > 0 for all integers 1 ≤ m ≤ n.

Solution. Let f : [0, 1] → R be continuous. Assume, towards contradiction, that for some
a ∈ [0, 1] one has v = f (c) 6= 0. Let  < |v|/2. By continuity, there is some δ > 0 so that for
x ∈ (c − δ, c + δ) ∩ [0, 1] we have |f (x) − v| < ; in particular |f (x)| > |v| −  > |v|/2. Choose
an interval [a, b] ⊂ (c − δ, c + δ) ∩ [0, 1] with a < b. We have
Z 1 Z b
2
f (x) dx ≥ f (x)2 dx ≥ (b − a) · v 2 /4 > 0.
0 a
This contradiction proves that f (x) = 0 for all x ∈ [0, 1].
LetRg(x) = 0 for irrational x ∈ [0, 1], and g(m/n) = 1/n is known to be Riemann integrable
1
with 0 g(x) dx = 0. Since 0 ≤ g(x) ≤ 1 on [0, 1], it follows that 0 ≤ g(x)2 ≤ g(x) and so
R1
0
g(x)2 dx = 0.
6
 Topology

T1. Let X be a first countable topological space, and A ⊂ X. Show that x ∈ Ā if and only
if there exists a sequence xn ∈ A such that xn → x. Then give an example to show that the
first countability hypothesis is necessary.

Solution. First, if there is such a sequence then for any neighborhood U of x we know
that ai is in U for all i sufficiently large. In particular, there is some ai in U so A ∩ U is
non-empty. Thus x ∈ Ā.
In the other direction, suppose x ∈ Ā. By the first countability assumption there is a
countable basis {Ui } at x so that Ui ⊂ Uj if i ≥ j. (To see this, start with any countable
basis {Vi } at x and define Ui = ∩k≤i Vk .)
Since x ∈ Ā we know that A ∩ Ui is non-empty. Choose xi ∈ A ∩ Ui . Thus xi ∈ A for all
i. We claim xi → x.
If V is any neighborhood of x, by definition of a basis at a point there exists j so that
Uj ⊂ V . Thus for all i ≥ j we have xi ∈ Ui ⊂ Uj ⊂ V . This shows xi → x.
To see that the first countability hypothesis is needed, consider X = R with the countable
complement topology. Let A be the interval [0, 1]. Since A is uncountable, we have A = X
in this topology. However, no sequence in A converges to a point outside of A: Let xn ∈ A
and let x 6∈ A. Then R \ {xn | n ∈ N} is an open neighborhood of x (since its complement is
countable) that contains no terms from the sequence. Hence xn 6→ x.

T2. Let X be a connected topological space. Show that X N , with the product topology, is
connected.

Solution. Suppose X N = A ∪ B for A and B disjoint nonempty open sets. Let a =

(a1 , a2 , . . .) be a point of A and b = (b1 , b2 , . . .) be a point of B. By the definition of the
product topology, there is a finite sequence U1 , U2 , . . . , Un of open subsets of X with ai ∈ Ui
such that V = U1 × U2 × · · · × Un × X × X × X · · · ⊂ A.
Consider a sequence of points, {si }, in X N where
si = (a1 , a2 , . . . , ai−1 , bi , bi+1 , bi+2 , · · · )
By construction si ∈ V for i > n, in particular si ∈ A for i large. On the other hand,
s1 = b ∈ B. Thus there exists k so that sk ∈ B and sk+1 ∈ A. Note that sk and sk+1 agree
in all coordinates except the k th . Consider the map f : X → X N defined by
x 7→ (a1 , a2 , · · · , ak−1 , x, bk+1 , bk+2 , · · · )
Thus f (ak ) = sk+1 ∈ A and f (bk ) = sk ∈ B. However, since X is connected its image can
intersect only one of A or B, a contradiction.

T3. If Suppose that f : X → Y is a continuous, surjective, open map, and that X is locally
compact. Show that Y is locally compact.

Solution. Recall that a space S is locally compact if, for every s ∈ S there is an open set
U and a compact set K with s ∈ U ⊂ K.
7
 Let y ∈ Y . By surjectivity of f there exists x ∈ X with f (x) = y. Since X is locally
compact, there is an open neighborhood V of x and a compact set L so that x ∈ V ⊂ L. Let
U = f (V ) and K = f (L). By construction we have y ∈ U and U ⊂ K. Since f is an open
map, U is open. Since f is continuous and L is compact, K is compact. As y was arbitrary,
this shows Y is locally compact.

Note: There are other definitions of “locally compact” in common use that are equivalent
to the one given above if X is Hausdorff, but which are different in general. The conclusion
of this exercise holds for these as well, with a substantially similar proof.

Logic

L1. Let T be a theory and σ a sentence. Suppose that σ holds in every infinite model of T .
Show that there is some natural number k such that σ holds in every model of T of size at
least k.

Solution. Suppose otherwise. Then for every natural number k, there is a model of T of
size at least k, in which σ does not hold.
For every k > 1, define the sentence
^
λk := (∃x1 )(∃x2 )...(∃xk ) xi 6= xj .
i<j

Let Γ = T ∪ {¬σ} ∪ {λk : k > 1}. We claim that Γ is consistent.

Suppose that ∆ is a finite subset of Γ; let k be the largest number such that λk ∈ ∆, if
such exists, and any natural number otherwise. By our assumption there is a model A of
T ∪ {¬σ} of size at least k. Then A |= ∆. I.e. ∆ is consistent.
By compactness it follows that Γ is consistent. Let A |= Γ. But then A is an infinite model
of T ∪ {¬σ}. Contradiction.

L2. Suppose that L = {P } is a language with a two place predicate P (and equality). Let
A = (Z, P A ), where ha, bi ∈ P A iff |a − b| = 1. I.e. A looks like an infinite graph:
... ←→ · ←→ · ←→ · ←→ ...
Show that there is an elementarily equivalent structure B that is not connected. (Here
connected means that for every two elements a, b there is a path between them, i.e. a
sequence hp0 , ..., pn−1 i such that p0 = a, pn−1 = b and for each i < n − 1, hpi , pi+1 i ∈ P A .)

Solution. Add two constant symbols to L to obtain L∗ = {P, c, d}. In L∗ , for every n > 0,
let λn be the formula
^
∀x0 , ...∀xn−1 ( xi P xi+1 → (x0 6= c ∨ xn−1 6= d)).
i<n−1

I.e. λn says that there is no hp0 , ..., pn−1 i, such that for each i < n − 1, hpi , pi+1 i ∈ P A with
p0 = c and pn−1 = d. Let Γ be the theory of A together with {λn | n > 0} and the sentence
c 6= d. We claim that Γ is consistent.
8
 Suppose that ∆ is a finite subset of Γ. Let n be the largest such that λn ∈ ∆. Let A∗ be
∗ ∗
obtained from A, by assigning cA = 0 and dA = n + 1. Then A∗ |= ∆.
By Compactness, it follows that Γ is consistent. So let B∗ be an L∗ -structure with B∗ |= Γ.
Then its reduct to L is an elementarily equivalent structure to A that is not connected.

L3. (a) Prove the Los-Vaught Test: Suppose that L is a countable language and T is an
L-theory that has no finite models and is κ-categorical for some infinite cardinal κ. Prove
that T is complete. (Hint: Use the Löwenheim-Skolem Theorem.)

(b) Suppose that L = {E}, where E is a binary relation symbol. Let T be the L-theory that
asserts that the interpretation of E is an equivalence relation with precisely two equivalence
classes, both of which are infinite. Prove that T is complete.

Solution. (a) Let M, N |= T ; we must show that M ≡ N . By assumption, M and

N are both infinite. Thus, by the Löwenheim-Skolem Theorem, there are M 0 ≡ M and
N 0 ≡ N such that |M 0 | = |N 0 | = κ. (M 0 is either an elementary extension or an elementary
substructure of M , depending on whether or not |M | ≤ κ or |M | ≥ κ; likewise for N 0 ). By
κ-categoricity, M 0 ∼
= N 0 . Thus, we have
M ≡ M0 ∼ = N 0 ≡ N,
and so M ≡ N .

(b) Note that T has no finite models and is ℵ0 -categorical: if (M, E M ) and (N, E N ) are two
countable models of T , then E M = E1 t E2 and E N = E10 t E20 , where both E1 and E2 are
countably infinite. Thus, by mapping E1 bijectively onto E10 and E2 bijectively onto E20 , we
have an isomorphism (M, E M ) → (N, E N ). So, by the Los-Vaught test, we have that T is
complete.