A CATALOG OF INTERESTING DIRICHLET SERIES

H. W. Gould and Temba Shonhiwa

Abstract. A Dirichlet series is a series of the form

∞
X f (n)
F (s) = ,
n=1
ns

where the variable s may be complex or real and f (n) is a number-theoretic

function. The sum of the series, F (s), is called the generating function of
f (n). The Riemann zeta-function
∞  −1
X 1 1
ζ(s) = = Πp 1 − s ,
n=1
ns p

where n runs through all integers and p runs through all primes is the special
case where f (n) = 1 identically. It is fundamental to the study of prime
numbers and many generating functions are combinations of this function.
In this paper, we give an overview of some of the commonly known number-
theoretic functions together with their corresponding Dirichlet series.

1. Introduction. There seems to be no convenient list of Dirichlet

series except for very large handbooks listing hundreds or thousands of
finite and infinite series of all kinds. Even there, many interesting series
may be missing and new ones turn up every so often. We believe that a
handy catalog of interesting Dirichlet series would be of use to researchers
and so we have gone through the literature [1, 2, 3, . . . , 43] and compiled
such a listing. In particular, the classic texts by Hardy and Wright [20],
McCarthy [25], and Titchmarsh [40] proved invaluable in this exercise. Our
list has been confined to what we believe to be useful and interesting series.
Additions and corrections to this initial catalog are welcomed.
By a Dirichlet series we mean any series of the form
∞
X f (n)
F (s) = .
n=1
ns

Most of the series we give converge for <(s) > 1, but we will just list the
real interval of convergence. The coefficients f (n) are commonly well-known
number-theoretic functions; and for those values of s for which the series
converges absolutely, the Dirichlet series serves as a generating function of
f (n). Further, if
∞ ∞
X f (n) X g(n)
F (s) = s
and G(s) = , for s > s0 ,
n=1
n n=1
ns

1
then
∞
X (f ∗ g)(n)
F (s)G(s) = , s > s0 ,
n=1
ns

where  
X n
(f ∗ g)(n) = f (d)g
d
d|n

is the Dirichlet convolution of f and g. Also, if the Dirichlet inverse f −1 of

f exists, then
∞
1 X f −1 (n)
= , s > s0 .
F (s) n=1 ns

Many of the formulas we list follow from the results above as well as from
the fact that when f (n) is a multiplicative function, then
∞
f (p) f (p2 ) f (p3 )
 
X f (n)
= Π p 1 + + + + · · · ,
n=1
ns ps p2s p3s

where p denotes primes.

The simplest such series is the classical Zeta function when f (n) = 1 iden-
tically:
∞
X 1
ζ(s) = ,
n=1
ns

convergent for s > 1.

For the sake of completeness we first give a list of some of the standard
number-theoretic functions we have found to occur in many Dirichlet series.
Let p always designate a prime number, (n, m) the greatest common divisor
of natural numbers n and m, and d a divisor of a natural number, as
indicated.

2. A Catalog Of Common Number-Theoretic Functions

N-1 pn = nth prime number, [20, page 5–6].
X
N-2 π(n) = 1 = the number of primes less than or equal to n, [20, page
p≤n
6].
X
N-3 τ (n) = 1 = the number of divisors of n, [20, page 239].
d|n

Here we have used the Greek letter tau for ‘Teiler’ (‘divisor’ in German).
Many British books commonly denote this function by d(n).

2
 N-4 τ (n, k) = the number of ways of expressing n as a product of k positive
factors (of which any number may be unity), expressions in which the
order of the factors is different being regarded as distinct.
In particular, τ (n, 2) = τ (n), [25, page 40].
X
N-5 σ(n) = d = the sum of the divisors of n, [20, page 239].
d|n
X
N-6 σk (n) = dk = the sum of the kth powers of the divisors of n. Thus,
d|n
σ1 (n) = σ(n), [20, page 239].
X
N-7 φ(n) = 1 = the number of integers less than or equal to n that
1≤j≤n
(j,n)=1

are relatively prime to n; commonly called Euler’s phi function (or

totient function), [25, page 1].

 1;
 if n = 1
N-8 µ(n) = 0; if p2 |n , commonly called the Möbius
 s
(−1) ; if n = p1 p2 · · · ps

function, [20, page 234].

 1;
 if n = 1
N-9 µk (n) = 0; if p2 |n , thus, µ1 = µ, [25, page 39].
 s k k k
(−1) ; if n = p1 p2 · · · ps


N-10 Jk (n) = the number of ordered k-tuples < a1 , a2 , · · · , ak > where

1 ≤ ai ≤ n for 1 ≤ i ≤ k and (a1 , a2 , · · · , ak , n) = 1, also known as the
X n

Jordan totient function. It is easily shown that Jk (n) = dk µ ,
d
d|n
[25, page 13].
log p; if n = pα for α ≥ 1

N-11 Λ(n) = , also known as the Von Man-
0; otherwise
goldt function, [25, page 247].
X
N-12 v(n) = 1 = the number of different prime factors of n. This is
p|n
sometimes denoted by ω(n), with ω(1) = 0. The letter ‘v’ comes from
the German word ‘verschieden’ meaning ‘different’, [17].
X
N-13 Ω(n) = 1 = the total number of prime factors of n counting repe-
pi |n
t
X
titions of a prime. Thus, if n = Πti=1 pα
i , then Ω(n) =
i
αi , [17].
i=1

3
N-14 λ(n) = (−1)r , where r = Ω(n), with λ(1) = 1, also called Liouville’s
function, [25, page 45].
X
N-15 Hk (n) = φ(e1 )φ(e2 ) · · · φ(ek ), called the Von Sterneck
1≤ei ≤n
[e1 ,e2 ,··· ,ek ]=n

function, [25, page 14]. Thus, H1 (n) = φ(n) and it may be shown
that Hk = Jk .
N-16 bxc = the greatest integer ≤ x, and its dual dxe = the least integer
≥ x, [3, page 72].
N-17 a(n) = the number of non-isomorphic Abelian groups with n elements.
a(n) is a multiplicative function studied by P. Erdős and G. Szekeres,
[13].
N-18 ζk (n) = nk , where k is a non-negative integer, [25, page 2].
N-19 β(n) = the number of integers j such that 1 ≤ j ≤ n and (j, n) is a
square, [25, page 25].
1; if n = 1

N-20 γ(n) = , also called the core function,
p1 p2 · · · pt ; if n = Πti=1 pα
i
i

[34].
N-21 φ(n, k) = the number of integers j such that 1 ≤ j ≤ n and (j, n) =
(n + k − j, n) = 1. It follows that, φ(n, 0) = φ(n), [25, page 35].
N-22 θ(n) = the number of ordered pairs < a, b > of positive integers such
that (a, b) = 1 and n = ab. It follows that θ(n) = 2ω(n) , [25, page 36].
N-23 θk (n) = the number of k-free divisors of n, where k is a positive integer,
greater or equal to 2, [25, page 37].
N-24 φ(x, n) = the number of integers j such that 1 ≤ j ≤ x and (j, n) = 1.
It follows that φ(n, n) = φ(n), [25, page 38].
N-25 Φk (n) = the number of integers j such that 1 ≤ j ≤ n and (j, n)k = 1,
where (a, b)k is the largest common kth power divisor of a and b, k ≥ 2,
[25, page 38]. Also known as Klee’s function. Clearly, (a, b)1 = (a, b)
and Φ1 = φ.
X n

N-26 ψk (n) = dk |µ |, where ψ1 = ψ is known as Dedekind’s function,
d
d|n
[25, page 41].
 1
|µ(n k )|; if n is a kth power
N-27 qk (n) = , where k is a positive integer,
0; otherwise
[25, page 41].
N-28 βk (n) = the number of integers j such that 1 ≤ j ≤ nk and (j, nk )k is
a 2kth power, [25, page 51].

4
 n
X
N-29 [k, n], where [k, n] = l.c.m of k and n, [17].
k=1
n
X n
X
N-30 (k, n) and its generalization f ((k, n)), where f is any number
k=1 k=1
theoretic function, [17].
1
(−1) 2 (n−1) if n is odd

N-31 χ(n) = , [25, page 27].
0; if n is even
N-32 R(n) = the number of ordered pairs of integers < a, b > such that
1 X
n = x2 + y 2 , [25, page 26]. It follows that, R1 (n) = R(n) = χ(d).
4
d|n

N-33 δk (n) = the greatest divisor d of n such that (d, k) = 1, where k is a

positive integer, [25, page 34].
X
N-34 ρk,s (n) = dk , also known as Gegenbauer’s function, [25,
d|n
n is an sth power
d
page 55].
2πia
N-35 For integers a and b, let e(a, b) = e b . Now let n be an integer and r
a positive integer and define
X
C(n, r) = e(nx, r).
1≤x≤r
(x,r)=1

Also known as Ramanujan’s sum, [20, page 247].

1; if n is an sth power

N-36 νs (n) = , [25, page 55].
0; otherwise
1; if n is k-free

N-37 ξk (n) = , [25, page 228].
0; otherwise
N-38 τk,h (n) = the number of ordered k-tuples < a1 , a2 , · · · , ak > of positive
integers such that each ai is an hth power and n = a1 a2 · · · ak , [25,
page 40].
X
N-39 ρ0k,t (n) = dk , [25, page 232].
d|n
d is a tth power

0
N-40 γ (n) = (−1)ω(n) γ(n), the inverse of the core function, [34].
0 X
N-41 σk (n) = λ(d)dk , where k is a non-negative integer, [25, page 235].
d|n

5
 X
N-42 φr (n) = j r , r = 0, 1, 2, 3, . . . , [17].
1≤j≤n
(j,n)=1

3. The Catalog of Dirichlet Series.

D-1
∞  −1
X 1 1
ζ(s) = = Π p 1 − , s > 1,
n=1
ns ps

[20, page 245]. More specifically,

∞
X 1 π2 22n−1 Bn π 2n
ζ(2) = 2
= and ζ(2n) = ,
n=1
n 6 (2n)!

where Bn denotes Bernoulli’s number.

∞
X µ(n) 1
D-2 s
= , s > 1, [20, page 250].
n=1
n ζ(s)
∞
X τ (n)
D-3 = ζ 2 (s), s > 1, [20, page 250].
n=1
ns
∞
X σ(n)
D-4 = ζ(s)ζ(s − 1), s > 2, [20, page 250].
n=1
ns
∞
X φ(n) ζ(s − 1)
D-5 s
= , s > 2, [20, page 250].
n=1
n ζ(s)
∞
X σk (n)
D-6 = ζ(s)ζ(s − k), s > 1, s > k, [20, page 250].
n=1
ns
∞
X log n 0
D-7 s
= −ζ (s), s > 1, [20, page 246].
n=1
n
∞ 0
X Λ(n) ζ (s)
D-8 =− , s > 1, [20, page 253–254].
n=1
ns ζ(s)
∞
X 2v(n) ζ 2 (s)
D-9 = , s > 1, [20, page 255].
n=1
ns ζ(2s)
∞
X |µ(n)| ζ(s)
D-10 s
= , s > 1, [20, page 255].
n=1
n ζ(2s)
∞
X τ 2 (n) ζ 4 (s)
D-11 = , s > 1, [20, page 29, 41, 255].
n=1
ns ζ(2s)

6
 ∞
X τ (n, k)
D-12 s
= ζ k (s), s > 1, [20, page 255].
n=1
n
∞
X ζ1 (n)
D-13 = ζ(s − 1), s > 2, [25, page 189].
n=1
ns
∞
X 1 1
D-14 s
= Πp>2 −s
= (1 − 2−s )ζ(s), s > 1 where we sum over odd
n=1
n 1 − p
n, [25, page 193].
∞
X ζk (n)
D-15 = ζ(s − k), s > k + 1, [25, page 226].
n=1
ns
∞  
X 1 1 ζ(s) 1
D-16 = − , s > 1, where we sum over n such that n
n=1
ns 2 ζ(2s) ζ(s)
is a product of an odd number of primes, [25, page 227].
∞ X 1
X ω(n)
D-17 = ζ(s) , s > 1, [25, page 227].
n=1
ns p
ps
∞
X θ(n) ζ 2 (s)
D-18 = , s > 1, [20, page 255].
n=1
ns ζ(2s)
∞ Pn
X ( k=1 [k, n]) 1 ζ(s − 1)
D-19 s
= {ζ(s − 2) + ζ(s − 3)}, s > 4, [18].
n=1
n 2 ζ(s − 2)
∞ Pn ∞
X ( k=1 f ((k, n))) ζ(s − 1) X f (n)
D-20 = , s > 2, [18].
n=1
ns ζ(s) n=1 ns
∞
X λ(n)θ(n) ζ(2s)
D-21 s
= 2 , s > 1, [25, page 227].
n=1
n ζ (s)
∞
X µk (n) 1
D-22 s
= , s > k1 , [25, page 228].
n=1
n ζ(ks)
∞
X Φk (n) ζ(s − 1)
D-23 s
= , s > 2, [33].
n=1
n ζ(ks)
∞
X ξk (n) ζ(s)
D-24 s
= , s > 1, [20, page 255].
n=1
n ζ(ks)
∞
X βk (n) ζ(s − k)ζ(2s)
D-25 s
= , s > k + 1, [25, page 229].
n=1
n ζ(s)

7
 ∞
X ψk (n) ζ(s − k)ζ(s)
D-26 s
= , s > k + 1, [19, 38].
n=1
n ζ(2s)
∞
X λ(n)Jk (n) ζ(s − k)ζ(s)
D-27 s
= , s > k + 1, [25, page 232].
n=1
n ζ(2s)
∞
X νk (n)
D-28 s
= ζ(ks), s > k1 , [25, page 228].
n=1
n
∞
X τk,h (n)
D-29 s
= ζ k (hs), s > h1 , [25, page 229].
n=1
n
∞
X ρk,t (n)
D-30 = ζ(s − k)ζ(ts), s > k + 1, [25, page 229].
n=1
ns
∞ ∞
X R(n) X (−1)m
D-31 s
= 4ζ(s)L(s), s > 1, where L(s) = , s > 1, [20,
n=1
n m=1
(2m + 1)s
page 256].
∞ 0
X λ(n)ρk,t (n) ζ(2s)ζ(2t(s − k))
D-32 = , s > k + 1, [25, page 232].
n=1
ns ζ(s)ζ(t(s − k))
∞ 0
X (−1)(k+1)ω(n) (γ )k (n)Jk (n) ζ(s − k)
D-33 s
= , s > 2k + 1, [12], chap-
n=1
n ζ(s − 2k)
ter X.
∞ 0
X γ (n)φ(n) ζ(s − 1)
D-34 s
= , s > 3, [12], chapter X.
n=1
n ζ(s − 2)
∞
X λ(n)τ (n, k − 1)τ (n2 γ k−2 (n)) ζ k (2s)
D-35 ω(n) s
= k+1 , s > 1, k ≥ 2, [12], chap-
n=1
(k − 1) n ζ (s)
ter X.
∞
X τ (n, k − 2)τ (n2 γ k−3 (n)) ζ k (s)
D-36 ω(n) s
= , s > 1, k ≥ 3, [12], chapter X.
n=1
(k − 2) n ζ(2s)
∞ 0
X σk (n) ζ(s)ζ(2(s − k))
D-37 s
= , s > k + 1, [25, page 235].
n=1
n ζ(s − k)
∞
X βh (n)σk (n) ζ(2s)ζ(s − h)ζ(s − h − k)ζ(2(s − k))ζ(2s − h − k)
D-38 s
= ,
n=1
n ζ(s)ζ(s − k)ζ(2(2s − h − k))
s > h + k + 1, [25, page 235].

8
 ∞ 0
X σh (n)σk (n) ζ(s)ζ(2(s − h))ζ(s − k)ζ(2(s − h − k))ζ(2s − h − k)
D-39 s
= ,
n=1
n ζ(s − h)ζ(s − h − k)ζ(2(2s − h − k))
s > h + k + 1, [6].
∞ 0
X λ(n)σh (n)σk (n) ζ(2s)ζ(2(s − k))ζ(s − h)ζ(2s − h − k)
D-40 s
= , s > h+
n=1
n ζ(s)ζ(2(2s − h − k))
k + 1, [6].
∞ 0 0
X σh (n)σk (n) ζ(s)ζ(2(s − k))ζ(2(s − h))ζ(s − h − k)
D-41 s
= , s > h+k +
n=1
n ζ(2s − h − k)ζ(s − h)ζ(s − k)
1, [6].
∞ 0 0
X λ(n)σh (n)σk (n) ζ(2s)ζ(2(s − h − k))ζ(s − h)ζ(s − k)
D-42 s
= , s > h+
n=1
n ζ(s)ζ(s − h − k)ζ(2s − h − k)
k + 1, [6].
∞ 0
X σk (n2 ) ζ(s)ζ(s − 2k)
D-43 s
= , s > 2k + 1, [25, page 237].
n=1
n ζ(s − k)
∞
X C(kn, r) X r

D-44 s
= ζ(s) d1−s (k, d)s µ , s > 1, [11].
n=1
n d
d|r
∞
X λ(n)C(n, r) ζ(2s) X 1−s r

D-45 s
= d λ(d)µ , s > 1, [11].
n=1
n ζ(s) d
d|r
∞
X λ(n)σh (n)σk (n) ζ(2s)ζ(2(s − h))ζ(2(s − k))ζ(2(s − h − k))
D-46 s
= ,
n=1
n ζ(s)ζ(s − h)ζ(s − k)ζ(s − h − k)ζ(2s − h − k)
s > h + k + 1, [25, page 232].
∞
X λ(n)σk (n) ζ(2s)ζ(2(s − k))
D-47 s
= , s > k + 1, [25, page 232].
n=1
n ζ(s)ζ(s − k)
∞  s 
X γ(n) p +p−1
D-48 = ζ(s)Π p , s > 1, [34, 42].
n=1
ns ps
∞
X λ(n)τ (n2 ) ζ 2 (2s)
D-49 s
= 3 , s > 1, [25, page 234].
n=1
n ζ (s)
∞
X λ(n)τ 2 (n) ζ 3 (2s)
D-50 s
= 4 , s > 1, [25, page 234].
n=1
n ζ (s)
∞
X σk (n2 ) ζ(s)ζ(s − k)ζ(s − 2k)
D-51 s
= , s > 2k + 1, [25, page 237].
n=1
n ζ(s(s − k))

9
 ∞
X Jk (n) ζ(s − k)
D-52 s
= , s > k + 1, [25, page 226].
n=1
n ζ(s)
∞
X λ(n) ζ(2s)
D-53 s
= , s > 1, [20, page 255].
n=1
n ζ(s)
∞
X (−1)n−1
D-54 s
= (1 − 21−s )ζ(s), s > 1, [40, page 21].
n=1
n
∞
X τ (n2 ) ζ 3 (s)
D-55 s
= , s > 1, [40, page 5].
n=1
n ζ(2s)
∞
τ 2 (n, k) 1 + p−s
X  
k
D-56 = ζ (s)Π P
p k−1 , s > 1, where Pk (x) is the kth
n=1
ns 1 − p−s
Legendre polynomial in x, [5, Vol. 3, page 170].
∞
X Λ(n) 1
D-57 = log(ζ(s)), s > 1, [40, page 57].
n=1
log n ns
∞
X σa (n)σb (n) ζ(s)ζ(s − a)ζ(s − b)ζ(s − a − b)
D-58 s
= , for s > 1, s − a >
n=1
n ζ(2s − a − b)
1, s − b > 1, and s − a − b > 1, [20 page 29, 43, 56].
∞
X (−1)n 1 − 22k−1
D-59 = |B2k |, k = 1, 2, 3, . . . , where Bk (x) is the
n=1
n2k (2k)!
Bernoulli polynomial of degree k in x, [3, pages 266-267].
∞
X 1 22k−1 π 2k (2π)2k
D-60 2k
= |B2k | = (−1)k−1 B2k , k = 1, 2, 3, . . . , with
n=1
n (2k)! 2(2k)!
Bn = Bn (0), [3, pages 266–267].
∞ 2k+1 Z 1
X 1 k−1 (2π)
D-61 = (−1) B2k (x) cot πxdx, [3, pages 266–
n=1
n2k+1 2(2k + 1)! 0
267].
∞
fk (n)
= (ζ(s) − 1)k , s > 1, where fk (n) is the number of represen-
X
D-62 s
n=1
n
tations of n as a product of k factors, each greater than 1 when n > 1,
the order of the factors being essential, [40, page 7].
∞
X f (n) 1
D-63 s
= , where f (n) is the number of representations of n
n=1
n 2 − ζ(s)
as a product of factors greater than unity, representations with factors
in a different order being regarded as distinct, and f (1) = 1. It must

10
 be noted that ζ(s) = 2 for some s = α a real number greater than 1,
so that the result is valid for s > α, [40, page 7].
∞
X Gn (a, b)
D-64 = ζ(s)ζ(s + a + b) − ζ(s + a)ζ(s + b), provided s > 1,
n=1
ns
s + a >X 1,  s + b >  1 and s + a + b > 1, where
n
a n
b
Gn (a, b) = da − db − , [16].
d|n
d d
√
d< n

∞
X a(n)
D-65 s
= Π∞
n=1 ζ(ns), s > 1, [13].
n=1
n
X 1 ∞
X µ(n)
D-66 s
= log ζ(ns), [40, page 12].
p
p n=1
n
∞ ∞
X v(n) X µ(n)
D-67 s
= ζ(s) log ζ(ns), [40, page 12].
n=1
n n=1
n
∞ ∞
X b(n) X φ(n)
D-68 s
= ζ(s) log ζ(s), where b(n) is the number of divisors
n=1
n n=1
n
of n which are primes or powers of primes, [40, page 12].
∞
X c(n) 1 − 21−s
D-69 = ζ(s − 1) , s > 2, where c(n) is the greatest odd
n=1
ns 1 − 2−s
divisor of n, [40, page 6].
∞ r+1  
X φr (n) 1 X r+1
D-70 =1+ Br+1−k ζ(s − k), s > k +
n=1
ns (r + 1)ζ(s − r) k
k=1
1, s > r + 1, [17].
D-71
∞ Pn k
j=1 [j, n]
X

n=1
ns
( k+1
)
1 X k + 1
= ζ(s − k) 1 + Bk+1−i ζ(s − k − i) ,
ζ(s − 2k)(k + 1) i=1 i

s > 2k + 1, k ≥ 1, [17].
We include the following curious formulas which are special cases of the
∞
X xn
function defined by P (x, s) = An s , which is both a power series and
n=1
n
a Dirichlet series.

11
 The following five infinite series may be found in the classic text by Whit-
taker and Watson, chapter seven, [41]. For further references, the reader
may also consult W. Spence [35] and L. J. Rogers [32].
∞
X xn
D-72 lim (1 − x)1−s = Γ(1 − s).
x→1
n=1
ns
∞
X xn 1
D-73 Let Φ(x) = 3
. Then, for |x| ≤ , this function satisfies the func-
n=1
n 2
tional equation,
 
x
Φ + Φ(x) + Φ(1 − x) − Φ(1)
x−1
π2 1 2
= log(1 − x) + {log(1 − x)} {log(1 − x) − 3 log x} .
6 6
D-74
 √ n
∞ 3− 5
X 2

n=1
n3
√ ! √ !)3
 ( 
2  −1 + 5 −1 + 5
= 6ζ(3) + π 2 log − 5 log 
15 2 2

= 0.4023504 . . . approximately.
∞
X sin nx x3 − π 2 x
D-75 (−1)n = .
n=1
n3 12
∞
X xn
D-76 Let H(x) = . This function satisfies the (Abel) functional equa-
n=1
n2
tion,
  
x(1 − y) y(1 − x)
H(x) + H(y) + H(xy) + H +H = 3H(1).
1 − xy 1 − xy
1
If we define L(x) = H(x) + log x × log(1 − x), then Leonard J.
2
Rogers [1906/07], [32] found the following remarkable relations:
L(x) + L(1 − x) = L(1),
   
x(1 − y) y(1 − x)
L(x) + L(y) = L(xy) + L +L ,
1 − xy 1 − xy
√ !
5−1 3 π2
L(x) + L(x2 ) = L(1), L = L(1) = ,
2 5 10
√ !
3− 5 2 π2
L = L(1) = .
2 5 15

12
 5. Some General Theorems.
∞
f (p) f (p2 ) f (p3 )
 
X f (n)
D-77 = Π p6| r 1 + + + + · · · , where r is any
n=1
ns ps p2s p3s
positive integer, (n, r) = 1 and f (n) is any multiplicative function, [25,
page 192].
∞
X f (n)
D-78 Let = F (s), s > s0 , say, where f is any arithmetic function.
n=1
ns
Then for a positive integer r,
∞ ∞
!
X f (n)C(n, r) X 1−s r
X f (md)
= d µ ,
n=1
ns d m=1 ms
d|r

s > s0 , [25, page 240].

∞ ∞ ∞ ∞
X an X An X xn X
D-79 ζ(s) s
= s
⇒ a n n
= An xn , [20, page 257].
n=1
n n=1
n n=1
1 − x n=1

D-80
∞ ∞
X an X Bn
ζ(s)(1 − 21−s ) s
=
n=1
n n=1
ns
∞ ∞
X xn X
⇒ an = B n xn ,
n=1
1 + xn n=1

[20, pages 247, 257].

∞ ∞
X 1 π2 X 2k + 5
D-81 3
= (−1)k B2k π 2k , posed by W. Jurkat [21]. It is
n=1
n 6 (2k + 3)!
k=0
a special case of:
∞
( )
1 (n) X (n) (n)
D-82 ζ(2n + 1) = ζ(2n) A − A2k ζ(2k) , where the A2k are ratio-
2 0
k=1
nal coefficients. This was published by Miklás Mikolás [27], 1953.
∞ ∞
X An p
X An (p)
D-83 Let F (s) = s
and [f (s)] = . Then
n=1
n n=1
ns

X p(a + c) + qa log n
(p + q log d)Ad (a)A nd (c) = An (a + c).
a+c
d|n

This was proved in Gould [15]. As shown there, this is a dual of a

corresponding identity for powers of power series.

13
 Acknowledgement. The authors wish to thank the anonymous referee
for suggestions which both improved the presentation as well as the utility
of this paper.

Ref erences
1. D. R. Anderson and T. M. Apostol, “The Evaluation of Ramanujan’s
Sum and Generalizations,” Duke Math. J., 20 (1953), 211–216.
2. T. M. Apostol, “Arithmetical Properties of Generalized Ramanujan
Sums,” Pacific J. Math, 41 (1972), 281–293.
3. T. M. Apostol, Introduction to Analytic Number Theory, Springer-
Verlag, New York, Heidelberg, Berlin, 1976.
4. T. M. Apostol, Modular Functions and Dirichlet Series in Number
Theory, Springer-Verlag, New York, Heidelberg, Berlin, 1976.
5. Bateman Project, Higher Transcendental Functions, Vol. 3.
6. S. Chowla, “On Some Identities Involving Zeta-functions,” J. Indian
Math. Soc., 17 (1928), 153–163.
7. E. Cohen, “An Extension of Ramanujan’s Sum,” Duke Math. J., 16
(1949), 85–90.
8. E. Cohen, “An Extension of Ramanujan’s Sum. III. Connections with
Totient Functions,” Duke Math. J., 23 (1956), 623–630.
9. E. Cohen, “Series Representations of Certain Types of Arithmetical
Functions,” Osaka Math. J., 13 (1961), 209–216.
10. E. Cohen, “Trigonometric Sums in Elementary Number Theory,”
Amer. Math Monthly, 66 (1959), 105–117.
11. M. M. Crum, “On Some Dirichlet Series,” J. London Math. Soc., 15
(1940), 10–15.
12. L. E. Dickson, History of the Theory of Numbers, Vol. I, Carnegie
Institution of Washington, 1919; reprinted, Chelsea Publishing Co.,
New York, 1952.
13. P. Erdős and G. Szekeres, “Über die Anzahl abelscher Gruppen
gegebener Ordnung und über ein verwandtes zahlentheoretisches Prob-
lem,” Acta Sci. Math. (Szeged), 7, 95–102, 1935.
14. D. L. Goldsmith, “A Generalization of Some Identities of Ramanujan,”
Rend. Mat., 6 (1969), 472–479.
15. H. W. Gould, “Coefficient Identities for Powers of Taylor and Dirichlet
Series,” Amer. Math. Monthly, 81 (1974), 3–14.

14
16. H. W. Gould, “A General Turan Expression for the Zeta Function,”
Canadian Math Bulletin, 6 (1963), 359–366.
17. H. W. Gould and T. Shonhiwa, “Functions of GCD’s and LCM’s,”
Indian Journal of Mathematics, 39 (1997), 11–35.
18. H. W. Gould and T. Shonhiwa, “A Generalization of Cesáro’s Function
and Other Results,” Indian Journal of Mathematics, 39 (1997), 183–
194.
19. J. Hanumanthachari, “A Generalization of Dedekind’s ψ− function,”
Math. Student, 40A (1972) 1–4.
20. G. H. Hardy and E. M. Wright, An Introduction to the Theory of
Numbers, Oxford University Press, 5th ed., 1979.
21. W. Jurkat, Problem No. 350, Jber DMV, 59 (1956), Abteilung 2 (Part
2), page 2.
22. D. M. Kotelyanskiǐ, “On N. P. Romanov’s Method of Obtaining Identi-
ties for Arithmetic Functions,” Ukrain Math. Zurnal, 5 (1953), 453–458
(Russian).
23. J. Lambek, “Arithmetic Functions and Distributivity,” Amer. Math.
Monthly, 73 (1966), 969–973.
24. E. Landau, Handbuch der Lehre von der Verteilung Primzahl, B.G.
Teubner, Leipzig, Berlin, 1909; reprinted, Chelsea Publishing Co., New
York, 1974.
25. P. J. McCarthy, Introduction to Arithmetical Functions, Springer-
Verlag, New York, 1986.
P f (n)
26. A. Mercier, “Identité pour 1≤n≤∞
ns ,” Canad. Math. Bull., 22
n≡1 (mod k)
(1979) 317–325.
27. M. Mikolás, “Über die Beziehung zwischen der Gammafunktion und
den trigonometrischen Funktionen,” Acta Math. Acad. Sci. Hungar.,
4 (1953), 143–157. CF. MR 15 (1954), p.525, and Zbl. für Math., 51
(1954), p.52.
28. G. Polyá and G. Szegö, Aufgaben und Lehrsátze aud der Analysis, Vol.
2, p. 129, Problems 65 and 66.
29. S. Ramanujan, “Some Formulae in the Analytic Theory of Numbers,”
Messenger of Math., 45 (1916), 81–84.
30. S. Ramanujan, “On Certain Trigonometrical Sums and Their Applica-
tions to the Theory of Numbers,” Trans. Camb. Philos. Soc., 22 (1918),
259–276.

15
31. D. Redmond and R. Sivaramakrishnan, “Some Properties of Specially
Multiplicative Functions,” J. Number Theory, 13 (1981), 210–227.
32. L. J. Rogers, “On Function Sum Theorems Connected with the Series
xn
P∞
n=1 n2 ,” Proc. London Math Soc., 4 (1907), 169–189.

33. U. V. Satyanarayana and K. Pattabhiramasastry, “A Note on the Gen-

eralized φ-function,” Math. Student, 33 (1965), 81–83.
34. T. Shonhiwa, “Core Function Characterisations of Number Theoretic
Functions,” Quaestiones Mathematicae Journal, 27 (2004), 185–194.
35. W. Spence, An Essay on Logarithmic Transcendents, London and Ed-
inburgh, 1809.
36. M. V. Subbaro, “Arithmetic Functions and Distributivity,” Amer.
Math. Monthly, 75 (1968), 984–988.
37. M. V. Subbaro and V. C. Harris, “A New Generalization of Ramanu-
jan’s Sum,” J. London Math. Soc., 4 (1966), 595–604.
38. D. Suryanarayana, “A Generalization of Dedekind’s ψ− function,”
Math. Student, 37 (1968) 81–86.
39. D. Suryanarayana, “The Greatest Divisor of n Which Is Prime to k,”
Math. Student, 37 (1969) 217–218.
40. E. C. Titchmarsh, The Theory of the Riemann Zeta Function, Oxford
University Press, 1951.
41. E. T. Whittaker, and G. N. Watson, A Course of Modern Analysis,
Cambridge University Press, Cambridge, England, 1927.
42. S. Wigert, “Sur Quelques Formules Asymptotiques de la Theorie des
Nombres,” Ark. Mat. Astron. Fys., 22B (1932) 1–6.
43. B. M. Wilson, “Proofs of Some Formulae Enunciated by Ramanujan,”
Proc. London Math. Soc., 21 (1923) 235–255.

16
Mathematics Subject Classification (2000): 11M41, 11A25

H. W. Gould
Department of Mathematics
West Virginia University
PO Box 6310
Morgantown, WV 26506-6310
email: Gould@math.wvu.edu
Temba Shonhiwa
School of Mathematics
University of the Witwatersrand
P. Bag 3
Wits 2050, South Africa
email: temba@maths.wits.ac.za

17