 Next Article in Journal
Scattering of Electromagnetic Waves by Many Nano-Wires
Previous Article in Journal
On the Class of Dominant and Subordinate Products
Article

# On the Distribution of the spt-Crank

1
Department of Mathematics, The Pennsylvania State University, University Park, PA 16802, USA
2
Institute for Advanced Study, School of Natural Sciences, Einstein Drive, Princeton, NJ 08540, USA
3
Stanford University, Department of Mathematics, Bldg 380, Stanford, CA 94305, USA
*
Author to whom correspondence should be addressed.
Mathematics 2013, 1(3), 76-88; https://doi.org/10.3390/math1030076
Received: 16 February 2013 / Revised: 10 April 2013 / Accepted: 10 April 2013 / Published: 28 June 2013

## Abstract

Andrews, Garvan and Liang introduced the spt-crank for vector partitions. We conjecture that for any n the sequence ${\left\{{N}_{S}\left(m\text{},\text{}n\right)\right\}}_{m}$ is unimodal, where ${N}_{S}\left(m\text{},\text{}n\right)$ is the number of S-partitions of size n with crank m weight by the spt-crank. We relate this conjecture to a distributional result concerning the usual rank and crank of unrestricted partitions. This leads to a heuristic that suggests the conjecture is true and allows us to asymptotically establish the conjecture. Additionally, we give an asymptotic study for the distribution of the spt-crank statistic. Finally, we give some speculations about a definition for the spt-crank in terms of “marked” partitions. A “marked” partition is an unrestricted integer partition where each part is marked with a multiplicity number. It remains an interesting and apparently challenging problem to interpret the spt-crank in terms of ordinary integer partitions.
Keywords:

## 1. Introduction and Statement of Results

The $spt$-function, introduced by the first author , counts the total number of appearances of the smallest parts in the partitions of n. For example, $spt ( 4 ) = 10$ because the partitions of 4 are $4 , 3 + 1 , 2 + 2 , 2 + 1 + 1 , 1 + 1 + 1 + 1$. The first author established the following remarkable congruences
$spt ( 5 n + 4 ) ≡ 0 ( mod 5 ) spt ( 7 n + 5 ) ≡ 0 ( mod 7 ) spt ( 13 n + 6 ) ≡ 0 ( mod 13 )$
These congruences bear a striking resemblance to Ramanujan’s congruences for the usual partition counting function $p ( n )$, namely
$p ( 5 n + 4 ) ≡ 0 ( mod 5 ) p ( 7 n + 5 ) ≡ 0 ( mod 7 ) p ( 11 n + 6 ) ≡ 0 ( mod 11 )$
The rank statistic of a partition was defined by the second author  as the largest part of the partition minus the number of parts. Let $N ( m , n )$ denote the number of partitions of n with rank m and $N ( m , t , n )$ denote the number of partitions of n with rank congruent to m modulo t. The second author conjectured, and Atkin and Swinnerton-Dyer  proved, that
$N ( k , 5 , 5 n + 4 ) = p ( 5 n + 4 ) 5 for 0 ≤ k ≤ 4 N ( k , 7 , 7 n + 5 ) = p ( 7 n + 5 ) 7 for 0 ≤ k ≤ 6$
Therefore, the rank provides a combinatorial interpretation of Ramanujan’s congruences modulo 5 and 7. Moreover, the second author observed that the rank is not sufficient to decompose Ramanujan’s congruence modulo 11, and he conjectured the existence of a statistic called the “crank” that would explain all three congruences.
Garvan  found the crank statistic for vector partitions and together with the first author  presented a definition for the crank of a ordinary partition, namely
$crank ( λ ) : = largest part of λ if o ( λ ) = 0 μ ( λ ) - o ( λ ) else$
where $o ( λ )$ is the number of 1s in the partition λ and $μ ( λ )$ is the number of parts of λ strictly larger than $o ( λ )$. Let $M ( m , n )$ be the number of partitions of n with crank m and $M ( m , t , n )$ be the number of partitions of n with crank congruent to m modulo t. Garvan proved 
$M ( k , 5 , 5 n + 4 ) = p ( 5 n + 4 ) 5 for 0 ≤ k ≤ 4 M ( k , 7 , 7 n + 5 ) = p ( 7 n + 5 ) 7 for 0 ≤ k ≤ 6 M ( k , 11 , 11 n + 6 ) = p ( 11 n + 6 ) 11 for 0 ≤ k ≤ 10$
Hence, the crank provides a combinatorial interpretation of all three of Ramanujan’s congruences for the partition counting function.
Recently Garvan, Liang and the first author  defined the $spt$-crank and used it to provide a combinatorial interpretation of the $spt$-congruences modulo 5 and 7. To describe the $spt$-crank we introduce the set of vector partitions, denoted by V. Then V is the Cartesian product
$V = D × P × P$
where $D$ is the set of partitions into distinct parts and $P$ is the set of all integer partitions. For $π = ( π 1 , π 2 , π 3 ) ∈ V$, let $π = π 1 + π 2 + π 3$, where $·$ is the sum of the parts of a partition. If $π = n$ we say that π is a vector partition of n. The crank of a vector partition is defined as $# ( π 2 ) - # ( π 3 )$, where $# ( · )$ is the number of parts in an integer partition.
To define the $spt$-crank we introduce the set of S-partitions. Let
$S : = { π = ( π 1 , π 2 , π 3 ) ∈ V : 1 ≤ s ( π 1 ) < ∞ and s ( π 1 ) ≤ min s ( π 2 ) , s ( π 3 ) }$
where $s ( π )$ is the smallest part in the partition. For $π ∈ S$ define a weight $ω 1$, by $ω 1 ( π ) = ( - 1 ) # ( π 1 ) - 1$. If $π ∈ S$ has crank m, then we refer to the $spt$-crank as $ω 1 ( π )$. Define
$N S ( m , n ) : = ∑ π ∈ S , π = n , crank ( π ) = m ω 1 ( π )$
and $N S ( m , t , n ) = ∑ k ≡ m ( mod t ) N S ( k , n )$. Garvan, Liang and the first author  establish the following combinatorial interpretation of the $spt$-congruences modulo 5 and 7
$N S ( k , 5 , 5 n + 4 ) = spt ( 5 n + 4 ) 5 for 0 ≤ k ≤ 4 N S ( k , 7 , 7 n + 5 ) = spt ( 7 n + 5 ) 7 for 0 ≤ k ≤ 6$
In a second paper , they prove a number of basic results about these values. For instance,
$N S ( m , n ) = N S ( - m , n )$
and, surprisingly,
$N S ( m , n ) ≥ 0$
Later, a simpler proof of this result was given by the second author .
Table 1 suggests that the sequence ${ N S ( m , n ) } m$ is (weakly) unimodal. Precisely, we give the following conjecture.
Conjecture 1.1. For each $m ≥ 0$ and $n ≥ 0$ we have
$N S ( m , n ) ≥ N S ( m + 1 , n )$
Remark. Chen, Ji, and Zang have announced a proof of this conjecture [].
Table 1. A table of values of $N S ( m , n )$.
Table 1. A table of values of $N S ( m , n )$.
$n \ m$0123456789
01
11
211
3111
42211
522211
6443211
75443211
877653211
91098653211
1013131110753211
11171615121075321
122424211814117532
1331292723191411753
14404036322621151175
155351484135272115117
1669686256463829221511
This property is not true for the ordinary rank or crank statistic. For example,
$N ( n - 1 , n ) = N ( n - 3 , n ) = 1 and N ( n - 2 , n ) = 0$
for all $n > 2$ and a similar statement holds for the crank. Our first statement reinterprets this conjecture in terms of the rank and crank. Define the cumulative density functions of the rank and crank as follows:
$N ≤ m ( n ) : = ∑ r ≤ m N ( r , m ) and M ≤ m ( n ) : = ∑ r ≤ m M ( r , m )$
Theorem 1.2. For all $n > 1$ and any $m ≥ 0$ we have
$N S ( m , n ) ≥ N S ( m + 1 , n )$
if and only if
$N ≤ m ( n ) ≥ M ≤ m ( n )$
Remark. The statement that
$N ≤ m ( n ) ≥ M ≤ m ( n )$
is true for each n was conjectured by Bringmann and Mahlburg .
Remark. Kaavya  conjectured that $N ≤ 0 ( n ) = N ( 0 , n ) ≥ M ( 0 , n ) = M ≤ 0 ( n )$ for all n.
This theorem leads to a good heuristic reason to believe that the $spt$-crank is unimodal. Define the moments of the rank and crank statistic by
$N 2 ℓ ( n ) : = ∑ m m 2 ℓ N ( m , n ) and M 2 ℓ ( n ) : = ∑ m m 2 ℓ M ( m , n )$
Remark. Since $N ( - m , n ) = N ( m , n )$ and $M ( m , n ) = M ( - m , n )$ the odd moments of these statistics are zero.
Garvan  conjectured that
$M 2 ℓ ( n ) > N 2 ℓ ( n )$
for each $ℓ > 0$ and n, despite the fact that
$N 2 ℓ ( n ) ∼ M 2 ℓ ( n ) as n → ∞$
This conjecture says that while the rank and crank are distributed asymptotically the same, the crank distribution is slightly “wider” for any fixed n. The first author  proved that $spt ( n ) = 1 2 ( M 2 ( n ) - N 2 ( n ) )$, which yields the $ℓ = 1$ case of Garvan’s conjecture. Garvan  later proved his own conjecture by introducing higher order $spt$-functions. As a result, we expect that
$N ≤ m ( n ) ≥ M ≤ m ( n )$
which is by Theorem 1.2 is equivalent to Conjecture 1.1.
The next theorem provides an asymptotic result supporting Conjecture 1.1.
Theorem 1.3. For each $m ≥ 0$ we have
$N ≤ m ( n ) ∼ M ≤ m ( n ) ∼ ( 2 m + 1 ) π 48 2 n 3 2 exp π 2 n 3 as n → ∞$
Moreover, we have
$N ≤ m ( n ) - M ≤ m ( n ) ∼ ( 2 m + 1 ) π 2 192 3 n 2 exp π 2 n 3 as n → ∞$
Remark. This result implies that for and fixed m and sufficiently large n we have
$N S ( m , n ) > N S ( m + 1 , n )$
Remark. For fixed m one may obtain an expansion $N ≤ m ( n ) ∼ ( 2 m + 1 ) π 48 2 n 3 2 exp π 2 n 3 1 + ∑ r ≥ 1 β r n r 2$ as $n → ∞$ with computable $β r$.
We close this section by giving an asymptotic for the distribution of the numbers $N S ( m , n )$. Let
$N S , k ( n ) : = ∑ m m k N S ( m , n )$
be the moments of the $spt$-crank.
Remark. By (1.1) the odd moments will be identically zero.
To define the asymptotic result we give the following definitions: Define
$γ ( a , b , c ) : = ( 2 ( a + b + c ) ) ! ( a + 1 ) ! ( 2 b ) ! ( 2 c ) ! B 2 b 1 2 B 2 c 1 2 ( - 1 ) a + c 4 - a - c π - a ( 3 a + 1 - 1 )$
where $B n ( x )$ is the nth Bernoulli polynomial. Define the Kloosterman sum
$K k ( n ) : = ∑ 0 ≤ h < k ( h , k ) = 1 ω h , k e - 2 π i h n k$
where $ω h , k : = exp π i s ( h , k )$ with
$s ( h , k ) : = ∑ μ ( mod k ) μ k h μ k$
the Dedekind sum, and
$x : = x - ⌊ x ⌋ - 1 2 if x ∈ R \ Z 0 if x ∈ Z$
is the sawtooth function.
Theorem 1.4. As $n → ∞$ we have
$N S , 2 ℓ ( n ) = 1 2 ∑ k < n K k ( n ) k ∑ a + b + c = ℓ k a + 1 γ ( a , b , c ) ( 24 n - 1 ) c + a 2 - 1 4 I 1 2 - a - 2 c π 24 n - 1 6 k + O n 2 ℓ + ϵ$
where $I ν$ denotes the modified Bessel function of order ν.
Remark. Using the asymptotic $I ν ( x ) ∼ 1 2 π x e x$ as $x → ∞$ we have
$N S , 2 ℓ ( n ) = 3 π ( - 1 ) ℓ B 2 ℓ 1 2 ( 24 n ) ℓ - 1 2 exp π 2 n 3 1 + O 1 n$
Since $spt ( n ) = N S , 0 ( n ) ∼ 1 2 π 2 n exp π 2 n 3$ and $B 2 ℓ 1 2 = - B 2 ℓ 2 2 ℓ - 1 - 1 2 2 ℓ - 1$, we have
$N S , 2 ℓ ( n ) spt ( n ) ( 6 n ) ℓ ∼ ( 2 2 ℓ - 2 ) B 2 ℓ$
The results of Bringmann, Mahlburg, and the third author  show that
$N 2 ℓ ( n ) p ( n ) ( 6 n ) ℓ ∼ M 2 ℓ ( n ) p ( n ) ( 6 n ) ℓ ∼ ( 2 2 ℓ - 2 ) B 2 ℓ$
Therefore, the $spt$-crank (after normalization) has the same distribution as the rank and crank of a partition. This distribution is known to be the same as the distribution of difference of two independent extreme value distributions. See the results of Diaconis, Janson, and the third author  for details.
In Section 2 we prove Theorem 1.2. In Section 3 we use the results of Bringmann, Mahlburg, and the third author  on the moments of the rank and crank statistics to establish Theorem 1.4. In Section 4 we use the circle method to calculate the asymptotics of Theorem 1.3. Finally, in Section 5 we discuss the $spt$-crank in terms of ordinary integer partitions. It seems a challenging and interesting problem to find an interpretation of the $spt$-crank in terms of ordinary integer partitions.

## 2. Generating Functions for $N S ( m , n )$

In this section we prove Theorem 1.2. Garvan, Liang and the first author (Corollary 2.5 of ) give
$∑ n ≥ 1 , m ∈ Z N S ( m , n ) z m q n = z - 1 ( 1 - z - 1 ) 2 ∑ n = 0 ∞ ∑ m ∈ Z N V ( m , n ) - N ( m , n ) z m q n$
where $N V ( m , n )$ is the number of vector partitions with crank m. Note that $N V ( m , n ) = M ( m , n )$ for $n > 1$. Formal q-series manipulations lead to the following: for any $m ≥ 0$ we have
$∑ n = 1 ∞ N S ( m , n ) q n = ∑ n > 1 ∞ ∑ ℓ ≥ m ( ℓ - m ) N V ( ℓ , n ) - N ( ℓ , n ) q n$
For example, when $m = 0$ we obtain
$∑ n = 1 ∞ N S ( 0 , n ) q n = ∑ n = 1 ∞ ospt ( n ) q n$
where
$ospt ( n ) = ∑ ℓ ≥ 0 ℓ M ( ℓ , n ) - N ( ℓ , n )$
The $ospt$ function is the difference of “first” moments of the crank and rank distributions, see . From (2.1) we have
$∑ n = 1 ∞ ( N S ( m , n ) - N S ( m + 1 , n ) ) q n = ∑ n > 1 ∑ ℓ > m ( N V ( ℓ , n ) - N ( ℓ , n ) ) q n$
Remark. This also follows from (37) of  and Equations (4.1) and (4.2) below.
Using the symmetry of the rank and crank statistics, and the fact that $∑ m N ( m , n ) = ∑ m M ( m , n ) = p ( n )$ we have
$∑ n > 0 ( N S ( m , n ) - N S ( m + 1 , n ) ) q n = ∑ n > 0 ∑ ℓ > m ( N V ( ℓ , n ) - N ( ℓ , n ) ) q n = ∑ n = 1 ∞ 1 2 ∑ ℓ N V ( ℓ , n ) - N ( ℓ , n ) ) + ∑ - m ≤ ℓ ≤ m ( N ( m , n ) - N V ( m , n ) ) q n = 1 2 ∑ n = 1 ∞ ∑ - m ≤ ℓ ≤ m ( N ( m , n ) - N V ( m , n ) ) q n (2.3)$
This establishes Theorem 1.2

## 3. Asymptotics for the Moments of the $spt$-Crank Statistic

In this section we will calculate the asymptotic for the moments of the $spt$-crank statistic. This calculation uses the results of  and establishes Theorem 1.4. For details see .
Let
$S ( x ; q ) = ∑ n = 1 ∞ ∑ m N S ( m , n ) x m q n = - 1 ( 1 - x ) ( 1 - x - 1 ) C ( x ; q ) - R ( x ; q )$
where $C ( x ; q )$ is the crank generating function and $R ( x ; q )$ is the rank generating function. Notice that
$S ( e 2 π i u ; q ) = ∑ k ≥ 0 ( 2 π i u ) k k ! ∑ n = 1 ∞ ∑ m m k N S ( m , n ) q n$
By the symmetry of the statistic we have $∑ m m k N S ( m , n ) = 0$ for all n when k is odd. We define $S k ( q ) = ∑ n = 1 ∞ ∑ m m k N S ( m , n ) q n$ to be the S-crank moment generating functions and $N S , k ( n ) = ∑ m m k N S ( m , n )$ to be S-crank moments weighted by $ω 1$.
The proof of Theorem 1.4 follows in a straightforward way from the results of  and a simple modification of some of the lemmas therein.
Throughout the remainder of this section let $z ∈ w i t h$ Re $( z ) > 0$ and $0 ≤ h < k$ with $( h , k ) = 1$ We define $[ a ] b$ the inverse of $a$ modulo $b$. Moreover, for fixed $h$ and $k$ we let $q = e 2 π i k h + i z$. Define $χ ( h , [ - h ] k , k )$ to be the multiplier of the Dedekind eta-function. In particular,
$χ ( h , [ - h ] k , k ) = i - 1 2 ω h , k - 1 e - π i 12 k [ - h ] k - h$
Finally, we define
$f ν ( u ; z ) : = e ν π u 2 z sin ( π u ) sinh π u z$
Proposition 3.1 (Section 3.2 of ). In the notation above
$C ( e 2 π i u ; q ) = - i 3 2 e π i 12 k ( h - [ - h ] k ) χ - 1 ( h , [ - h ] k , k ) e π 12 k 1 z - z z - 1 2 f k ( u ; z ) + ∑ r = 0 ∞ a r ( z ) u r r !$
where $a r ( z ) ≪ z 1 2 - r e - α k Re 1 z$ for some $α > 0$ independent of k.
Proposition 3.2 (Proof of Proposition 3.5 of ). In the notation above
$R ( e 2 π i u ; q ) = - i 3 2 e π i 12 k ( h - [ - h ] k ) χ - 1 ( h , [ - h ] k , k ) e π 12 k 1 z - z z - 1 2 f 3 k ( u ; z ) + ∑ r = 0 ∞ a r ( z ) u r r !$
where $a r ( z ) ≪ k 1 2 z 1 2 - r$.
Combining Propositions 3.1 and 3.2 and (3.1) we have the following lemma.
Lemma 3.3. In the notation above,
$S ( e 2 π i u ; q ) = - 1 4 i 3 2 e π i 12 k ( h - [ - h ] k ) χ - 1 ( h , [ - h ] k , k ) e π 12 k 1 z - z z - 1 2 e 3 π k u 2 z - e π k u 2 z sin ( π u ) sinh π u z + ∑ r = 0 ∞ a r ( z ) u r r !$
where $a r ( z ) ≪ k 1 2 z 1 2 - r$.
Taylor expanding the expression in Lemma 3.3 with respect to u and using (3.2) give asymptotics for $S k ( q )$. The circle method can now be used to turn those asymptotics for the generating functions into asymptotics for the coefficients. Applying the following theorem gives Theorem 1.4. The theorem is a general circle method result, which is a slight modification of Theorem 4.1 of .
Theorem 3.4. Assume that
$F r , ℓ e 2 π i k ( h + i z ) = ∑ n c r , ℓ ( n ) e 2 π i k ( h + i z )$
is holomorphic function of z satisfying
$F r , ℓ e 2 π i k ( h + i z ) = - i 3 2 e π i 12 k ( h - [ - h ] k ) χ - 1 ( h , [ - h ] k , k ) e π 12 k 1 z - z ∑ a + b + c = ℓ C ( a , b , c ) z - p ( a , b , c ) + E r , ℓ , k ( z )$
with $E r , ℓ , k ( z ) ≪ r , ℓ k 1 2 z - 1 2 - 2 ℓ$, $C ( a , b , c )$ are some constants and $p ( a , b , c )$ is a polynomial in a, b, and c. Then
$c r , ℓ ( n ) = 2 π ∑ k ≤ n K k ( n ) k ∑ a + b + c = ℓ C ( a , b , c ) ( 24 n - 1 ) p ( a , b , c ) 2 - 1 2 I 1 4 - p ( a , b , c ) π 24 n - 1 6 k + O n 2 ℓ + ϵ$

## 4. The Circle Method and False Theta Functions

In this section we consider the cumulative density functions of the rank and crank. We show that these generating functions are partial theta functions times the partition generating function. Obtaining an asymptotic expansion for the coefficients of such a generating function via the circle method is classical (see , for example). We have the following well known generating functions for $N ( m , n )$ and $M ( m , n )$
$∑ n ≥ 0 N ( m , n ) q n = 1 ( q ) ∞ ∑ n > 0 ( - 1 ) n + 1 q n ( 3 n - 1 ) 2 + m n ( 1 - q n )$
and
$∑ n ≥ 0 M ( m , n ) q n = 1 ( q ) ∞ ∑ n > 0 ( - 1 ) n + 1 q n ( n - 1 ) 2 + m n ( 1 - q n )$
Fine  showed that
$R ≤ m ( q ) = ∑ n N ≤ m ( n ) q n = 1 ( q ) ∞ 2 ∑ n = 0 ∞ ( - 1 ) n q 3 n 2 + n 2 + m n - 1$
Similarly, we have from (4.2)
$C ≤ m ( q ) = ∑ n N ≤ m ( n ) q n = 1 ( q ) ∞ 2 ∑ n = 0 ∞ ( - 1 ) n q n 2 + n 2 + m n - 1$
Remark. This shows that the generating function for each cumulative density function is a partial theta functions times the partition generating function.
Note that
$q n 2 + n 2 + m n = q 1 2 ( n 2 + 2 m n + n ) = q 1 2 ( n 2 + 2 m n + n + m + 1 4 ) - 1 2 m 2 + m + 1 4 = q 1 2 m + n + 1 2 2 - 1 2 m + 1 2 2$
So we have
$C ≤ m ( q ) = 1 ( q ) ∞ 2 ( - 1 ) m q - 1 8 2 m + 1 2 ∑ n > 2 m ∞ - 4 n q n 2 8 - 1$
where $- 4 ·$ is the Kronecker symbol. Similarly, we have
$R ≤ m ( q ) = 1 ( q ) ∞ 2 q - 1 24 2 m + 1 2 ∑ n > 2 m ∞ χ m ( n ) q n 2 24 - 1$
where
$χ m ( n ) = + 1 n ≡ 2 m + 1 ( mod 12 ) - 1 n ≡ 2 m + 7 ( mod 12 ) 0 else$
We set $q = e - s$ and consider the asymptotic as $s → 0 +$.
The following proposition is a slight variation of a proposition of Lawrence and Zagier . Since the proof is analogous and standard, we do not include it here.
Proposition 4.1 (p. 98 of ). Let $C : Z →$ be a periodic function with mean value 0. Then for each $m ≥ 0$ the L-series $L m ( s , C ) e = ∑ n > m ∞ C ( n ) e - s$ ($R e ( s ) > 1$) extends holomorphically to all of and the function $∑ n > m ∞ C ( n ) e - n 2 t$ ($t > 0$) has the asymptotic expansion
$∑ n > m ∞ C ( n ) e - n 2 t ∼ ∑ r = 0 ∞ L m ( - 2 r , C ) · ( - t ) r r !$
as $t → 0 +$. The numbers $L m ( - r , C )$ are given explicitly by
$L m ( - r , C ) = - M r r + 1 ∑ n = ( m + 1 ) m + M C ( n ) B r + 1 n M ( r = 0 , 1 , … )$
where $B k ( x )$ denotes the kth Bernoulli polynomial and M is any period of the function $C ( n )$. Moreover, these expansions are valid in the region $t < 2 π M$.
This proposition readily yields an asymptotic for the infinite series in (4.3) and (4.4).
Proposition 4.2. With $q = e - s$ we have the following asymptotic expansions valid in the region $s < π 6$.
$∑ n > 2 m ∞ - 4 n q n 2 8 ∼ ( - 1 ) m 1 2 + 2 m 2 - 1 2 ( - s ) 8 + 8 m 4 - 12 m 2 + 5 2 ( - s ) 2 8 2 · 2 + ⋯ ∑ n > 2 m ∞ χ m ( n ) q n 2 24 ∼ 1 2 + 2 m 2 - 4 m - 5 2 ( - s ) + 1 2 ( 2 m - 5 ) ( 2 m + 1 ) ( 4 m 2 - 8 m - 41 ) s 2 2 + ⋯$
Using $( q ) ∞ - 1 = s 2 π e π 2 6 s - s 24 ( 1 + O ( s N ) )$ for any $N > 0$ (this follows from Euler–Maclaurin summation formula or the modularity of the Dedekind eta-function, see  page 53), we see that
$C ≤ m ( e - s ) ∼ s 2 π e π 2 6 s - s 24 ( 2 m + 1 ) 4 s + ( 2 m + 1 ) 16 s 2 + ⋯$
and
$R ≤ m ( e - s ) ∼ s 2 π e π 2 6 s - s 24 ( 2 m + 1 ) 4 s + 3 ( 2 m + 1 ) 16 s 2 + ⋯$
A standard application of the circle method (see, for instance, Wright  for a similar situation) gives the theorem.

## 5. Some Guesses for the $spt$-Crank

This section collects some observations concerning the values of $N S ( m , n )$. In particular, we are concerned with defining the $spt$-crank in terms of partitions (perhaps with their parts marked by the multiplicity).
A marked partition means a pair $( λ , k )$ where λ is a partition and k is an integer identifying one of its smallest parts. If there are s smallest parts then the $k = 1 , 2 , ⋯ , s$. Evidently, a good first approximation for the $spt$-crank is
$F ( λ , k ) : = p - k if p > 0 1 - k if p = 0$
where p is the number of parts in λ greater than or equal to k. If $T ( n , m )$ is the number of marked partitions of n with $F ( λ , k ) = m$, then the difference
$D ( n , m ) : = T ( n , m ) - N S ( n , m )$
is zero for most of the possible values of n and m. Table 2 and Table 3 give the values of $T ( n , m )$ and $D ( n , m )$ for small $n ≤ 12$.
Table 2. A table of values of $T ( m , n )$.
Table 2. A table of values of $T ( m , n )$.
$n \ m$−9−8−7−6−5−4−3−2−10123456789
1 1
2 111
3 11111
4 1122211
5 112312211
6 11244343211
7 1123553443211
8 112368666653211
9 112357118888653211
1011235812151011111110753211
Table 3. A table of values of $D ( m , n )$.
Table 3. A table of values of $D ( m , n )$.
$n \ m$−11−10−9−8−7−6−5−4−3−2−101234567891011
1 0
2 000
3 00000
4 0000000
5 0001−10000
6 00010−100000
7 000011−2000000
8 000012−1−1−1000000
9 0000013−1−2−10000000
10 00000124−3$- 2$−200000000
11 000000144−2−4−2−100000000
120000001264−5−4−3−1000000000
The differences $D ( n , m )$ are non-zero and have a simple regular behavior in the central angle between the two lines
$n = 3 m + 5 and n = - 2 m + 2$
and zero everywhere else. Additionally, the numbers become periodic on the boundaries. (This is hard to tell from the Table 2, but easy to see from a larger table.) The left side boundary has period 2 and the right hand boundary has period 3. After removing those periodic parts, there are two more boundary lines
$n = - 3 m + 3 and n = 4 m + 11$
which separate the regions where the numbers are periodic from the regions where they are not. So it is easy to conjecture that there is a series of boundaries $N = k m + c$, $N = - k m + c$, for each integer k, separating regions with period $k - 1$ from regions with period k.
Finally, we speculate that a definition of the $spt$-crank may be different depending on the size of the smallest part. It remains a challenge to find a definition of the $spt$-crank for ordinary partitions.

## References

1. Andrews, G.E. The number of smallest parts in the partitions of n. J. Reine Angew. Math. 2008, 624, 133–142. [Google Scholar] [CrossRef]
2. Dyson, F.J. Some Guesses in the Theory of Partitions; Eureka: Cambridge, UK, 1944; Volume 8, pp. 10–15. [Google Scholar]
3. Atkin, A.O.L.; Swinnerton-Dyer, P. Some properties of partitions. Proc. London Math. Soc. 1954, 3, 84–106. [Google Scholar] [CrossRef]
4. Garvan, F.G. New combinatorial interpretations of Ramanujan’s partition congruences mod 5, 7, and 11. Trans. Am. Math. Soc. 1988, 305, 47–77. [Google Scholar] [CrossRef]
5. Andrews, G.E.; Garvan, F.G. Dyson’s crank of a partition. Bull. Am. Math. Soc. (N.S.) 1988, 18, 391–407. [Google Scholar] [CrossRef]
6. Andrews, G.E.; Garvan, F.; Liang, L. Combinatorial interpretations of congruences for the $spt$-function. Ramanujan J. 2012, 29, 321–338. [Google Scholar] [CrossRef]
7. Andrews, G.E.; Garvan, F.; Liang, L. Self-conjugate vector partitions and the parity of the $spt$-function. Acta Arithmetica 2013, in press. [Google Scholar] [CrossRef]
8. Dyson, F.J. Partitions and the grand canonical ensemble. Ramanujan J. 2012, 29, 423–429. [Google Scholar] [CrossRef]
9. Chen, W.Y.C.; Ji, K.Q.; Zang, W.J.T. Proof of the Andrews-Dyson-Rhoades conjecture on spt-crank. Preprint at http://arxiv.org/abs/1305.2116.
10. Bringmann, K.; Mahlburg, K. Inequalities between Crank and Rank Moments. Proc. Am. Math. Soc. 2009, 137, 2567–2574. [Google Scholar] [CrossRef]
11. Kaavya, S.J. Crank 0 partitions and the parity of the partition function. Int. J. Number Theory 2011, 7, 793–801. [Google Scholar] [CrossRef]
12. Garvan, F.G. Congruences for Andrews’ smallest parts partition function and new congruences for Dyson’s rank. Int. J. Number Theory 2010, 6, 1–29. [Google Scholar] [CrossRef]
13. Garvan, F.G. Higher order $spt$-functions. Adv. Math. 2011, 228, 241–265. [Google Scholar] [CrossRef]
14. Bringmann, K.; Mahlburg, K.; Rhoades, R.C. Asymptotics for rank and crank moments. Bull. Lond. Math. Soc. 2011, 43, 661–672. [Google Scholar] [CrossRef]
15. Diaconis, P.; Janson, S.; Rhoades, R.C. Note on a partition limit theorem for rank and crank. Bull. London Math. Soc. 2013. submitted for publicaiton. [Google Scholar] [CrossRef]
16. Bringmann, K.; Mahlburg, K.; Rhoades, R.C. Taylor coefficients of mock-jacobi forms and moments of partition statistics. Proc. Camb. Phil. Soc. 2013. submitted for publicaiton. [Google Scholar] [CrossRef]
17. Andrews, G.E.; Chan, S.H.; Kim, B. The odd moments of ranks and cranks. JCT(A) 2013, 120, 77–91. [Google Scholar] [CrossRef]
18. Wright, E.M.; Stacks, I.I. Quart. J. Math. Oxford Ser. (2) 1971, 22, 107–116. [CrossRef]
19. Fine, N.J. Basic Hypergeometric Series and Applications with a Foreword; Andrews, G.E., Ed.; Mathematical Surveys and Monographs, 27; American Mathematical Society: Providence, RI, USA, 1988. [Google Scholar]
20. Lawrence, R.; Zagier, D. Modular forms and quantum invariants of 3-manifolds. Sir Michael Atiyah: A great mathematician of the twentieth century. Asian J. Math. 1999, 3, 93–107. [Google Scholar]
21. Zagier, D. The Dilogarithm Function. In Frontiers in Number Theory, Physics, and Geometry; Springer: Berlin, Germany, 2007. [Google Scholar]