The Network Of Knowledge

                Science   Technology   Engineering   Mathematics   History   Thought   Belief   The Modern World
                

                        Omnia Exeunt In Mysterium

Durfee Square

The Durfee square of a partition is the largest square of nodes that can fit inside the Ferrers Graph of a partition. Durfee squares were named after the American mathematician William Pitt Durfee (5 February 1855 – 17 December 1941), who was a student of James Joseph Sylvester (3 September 1814 – 15 March 1897).

For example, the partition shown below has a Durfee square of size \(3\times 3\).

Figure 1: Ferrers graph of the partition \(6+5+4+2+2+1\), with \(3\times 3\) Durfee square.
Figure 1: Ferrers graph of the partition \(6+5+4+2+2+1\), with \(3\times 3\) Durfee square.

In a letter to the mathematician Arthur Cayley (16 August 1821 – 26 January 1895), Sylvester remarked that
"Durfee's square is a great invention of the importance of which its author has no conception."

Durfee squares play a part in the combinatorial proof of many partition identities (see the page on Combinatorial Proofs of Partition Identities, for example). Durfee squares also play a part in the analytic proof of some \(q\)-series identities, for example \begin{equation*} 1+ \sum_{k=1}^{\infty} \frac{q^{k^2}}{\prod_{j=1}^{k}(1-q^{j})^2} = \frac{1}{\prod_{j=1}^{\infty}(1-q^{j})}. \end{equation*}

k

For a fixed non-negative integer \(r\), one can ask for the largest \(k \times (k+r)\) rectangle of nodes that can fit inside the Ferrers Graph of a partition. This the Durfee rectangle of size \(k \times (k+r)\) of the partition.

Durfee rectangles also play a part in proving various combinatorial and analytic identities.

Back to the main Partitions page

Page created by Oideachas, LLC