All the successive numbers which serve to form the triangular,
quadrangular or polygonal numbers, are called gnomons.
– Theon of Smyrna (XXIII, Lawlor & Lawlor, 1979, p. 25)
This article highlights the author’s teaching experience in a university history of mathematics classroom focusing on the exploration of figurate numbers. Three mathematics history textbooks were adopted throughout the semester (Heath, 1921; NCTM, 1989; Katz, 2009); the present report focuses on students’ exploration of the figurate numbers using the historical resources as outlined in Thomas Heath's A History of Greek Mathematics: From Thales to Euclid (1921), together with the popular mathematical manipulative, Cuisenaire rods. The History of Mathematics course in which these activities were used consisted of students majoring in Mathematics and Mathematics Education (prospective secondary mathematics teachers) who had completed the prerequisite courses of Calculus & Analytic Geometry and College Geometry. The activities were conducted during two 75-minute class periods.
Figure 1. Cuisenaire rods
When modeling counting numbers from 1 to 10 with Cuisenaire rods, each colored rod represents a different number: the white rod represents 1, the red rod represents 2, the light green rod represents 3, the purple represents 4, and so on. This way, rods from white to orange are allotted whole number values 1 to 10, respectively (see Figures 1 and 2).
Figure 2. Cuisenaire rods
In all explorations, students worked in groups of two or three. The exploration began with the reading of a passage on triangular numbers from Thomas Heath's A History of Greek Mathematics: From Thales to Euclid (1921):
The particular triangle which has 4 for its side is mentioned in a story of Pythagoras by Lucian. Pythagoras told someone to count. He said 1, 2, 3, 4, whereon Pythagoras interrupted, “Do you see? What you take for 4 is 10, a perfect triangle and our oath." (p. 77)
After reading this passage and viewing the accompanying diagram (see Figure 3a), students modeled the first four triangular numbers using the Cuisenaire rods in accordance with the passage in slightly different ways (see Figures 3b-d). They then proved the triangular number formula \[1+2+3+\cdots+n={\frac{1}{2}}n(n+1)\] via mathematical induction.
a (1921, p. 76) |
|
||
b | |||
c | |||
d |
Figure 3: Triangular numbers \(1, 3, 6, 10\) using Cuisenaire rods
Prior to the construction of square numbers, class time was dedicated to the modeling of the consecutive odd numbers \(1,3,5,\dots\) as symmetric L-shaped gnomons (see Figure 4a):
All this was known to Pythagoras. The odd numbers successively added were called gnomons; this is clear from Aristotle's allusion to gnomons placed round 1 which now produce different figures every time (oblong figures, each dissimilar to the preceding one), now preserve one and the same figure (squares). (1921, p. 77)
Students noted that the only Cuisenaire rod that is already in the desired gnomon format is the white rod representing 1. Students in all groups indicated that they must combine various Cuisenaire rods to form gnomons representing odd numbers larger than 1. This activity led to a diversity of representations (see Figures 4b-e). Through a series of debates and considerations, students in all groups eventually agreed on the model in Figure 4e as the one in agreement with the passage they discussed.
a (1921, p. 77) |
|||
b | |||
c | |||
d | |||
e |
Figure 4: Odd integers \(1, 3, 5, 7, 9, 11\) with Cuisenaire rods
Having already modeled the gnomons of square numbers (see Figure 4e), it was time to combine them to form a square number at each step while "preserv[ing] one and the same figure (squares)" (1921, p. 77).
Consider the square with \(n\) dots in its side in relation to the next smaller square \({(n-1)}^2\) and the next larger \({(n+1)}^2.\) Then \(n^2\) exceeds \({(n-1)}^2\) by the gnomon \(2n-1,\) but falls short of \({(n+1)}^2\) by the gnomon \(2n+1.\) Therefore the square \({(n+1)}^2\) exceeds the square \({(n-1)}^2\) by the sum of the two gnomons \(2n-1\) and \(2n+1,\) which is \(4n.\) (1921, p. 81)
Using the template given in the text (see Figure 5a), students developed a gnomonic representation of the first six square numbers (see Figure 5b). For \(n\ge2,\) we can see the desired representation showing that the square \({(n+1)}^2\) minus the square \({(n-1)}^2\) equals the sum of the successive gnomons \(2n-1\) and \(2n+1\) at each step (see Figure 5c).
a (1921, p. 81) |
|||
b | |||
c |
Figure 5: Square numbers with Cuisenaire rods
The next task was to model oblong numbers; that is, numbers of the sequence
\[2,\,\,2+4, \,\,2+4+6,\dots, \,\,2+4+6+\cdots+2n.\]
For this purpose, students first focused on representation of the even integers with the Cuisenaire rods. As was the case for the odd numbers, students offered different visualizations of the even numbers \(2, 4, 6,\dots\) (see Figure 6).
a | ||
b |
Figure 6: Even integers \(2, 4, 6, 8, 10, 12\) with Cuisenaire rods
Students then studied the text along with the accompanying template (see Figure 7a) for the purpose of generating the oblong number sequence:
While the adding of the successive odd numbers as gnomons round 1 gives only one form, the square, the addition of the successive even numbers to 2 gives a succession of 'oblong' numbers all dissimilar in form, that is to say, an infinity of forms (1921, p. 83).
When it came to the modeling of the oblong numbers, the majority of the students selected the model proposed in Figure 6b as the most suitable representation in agreement with the description in the historical resource:
It is to be noted that the word έτερομήκης (‘oblong’) is in Theon of Smyrna and Nicomachus limited to numbers which are the product of two factors differing by unity, while they apply the term προμήκης ('prolate', as it were) to numbers which are the product of factors differing by two or more (1921, p. 83).
Figure 7b depicts the oblong numbers generated based on the gnomonic structure given in Figure 6b. Other groups preferred the simplified \(n\times(n+1)\) rectangular representations in which they used the minimum number of Cuisenaire rods corresponding to each oblong number (see Figure 7c).
a (1921, p. 82)
|
|||
b | |||
c |
Figure 7: Oblong numbers \(2, 6, 12, 20, 30, 42\) with Cuisenaire rods.
Students in all groups recognized the relation between the \(n\times n\) square number pattern (see Figure 5b) and the \(n\times(n+1)\) rectangular oblong number pattern (see Figure 7), which they easily verified by the distributive property as \(n\times(n+1)=n^2+n\). That is, the \(n\)th oblong number is \(n\) more than the \(n\)th square number (see Figure 8).
Figure 8: Square-oblong relation: the \(n\)th oblong number = the \(n\)th square number \(+\,n.\)
After students noted the relationship between square and oblong numbers depicted in Figure 8, the next task was to model any oblong number \(n(n+1)\) as the sum of two equal triangular numbers; see Figure 9a from Thomas Heath's A History of Greek Mathematics: From Thales to Euclid (1921). All students immediately noted the one-half relation between an oblong number and the corresponding triangular number (both analytically and visually) and suggested an oblong number modeling based on two congruent triangular numbers (see Figures 9b and 9c).
a (1921, |
|||
b | |||
c |
Figure 9: Oblong-triangular relationship with Cuisenaire rods
Upon establishing the oblong-triangular relationship, students focused on the theorem of Theon, which states that any square number is expressible as the sum of two consecutive triangular numbers. As Thomas Heath explained, referring to Figure 10a:
In this case, as is seen from the figure, the sides of the triangles differ by unity, and of course \[{\frac{1}{2}}n(n-1)+\frac{1}{2}n(n+1)=n^2\quad{\sf{\small{(1921, pp. 83{\mbox{-}}84).}}}\]
Figures 10b and 10c depict Cuisenaire rod representations of Theon’s theorem.
a (1921, |
|||
b | |||
c |
Figure 10: Theon’s theorem, the sum of two consecutive triangular numbers is a square, or \(\frac{1}{2}n(n-1)+\frac{1}{2}n(n+1)=n^2,\) with Cuisenaire rods
As "quoted by Plutarch and used by Diophantus" (1921, p. 84), the theorem stating
that 8 times any triangular number plus 1 makes a square … is equivalent to the formula
\[{8\cdot\frac{1}{2}}n(n+1)+1=4n(n+1)+1={(2n+1)}^2\quad{\sf\small{{(1921, p. 84).}}}\]
Students first verified the formula given in the text; they then focused on generating Cuisenaire rod patterns modeling the given template (see Figure 11a):
It may easily have been proved by means of a figure made up of dots in the usual way. Two equal triangles make up an oblong figure of the form \(n(n+1),\) as above. Therefore we have to prove that four equal figures of this form with one more dot make up \({(2n+1)}^2.\) The annexed figure representing \(7^2\) shows how it can be divided into four 'oblong' figures \(3\cdot4\) leaving \(1\) over. (1921, p. 84)
Figures 11b and 11c depict Cuisenaire rod representations of this ancient theorem. Whereas one group of students focused on the “4 oblong plus 1” approach (see Figure 11b), other students were more explicit in their representations by embracing the “8 triangular plus 1” approach (see Figure 11c).
a (1921, |
|
b | |
c |
Figure 11: "Eight times any triangular number plus one makes a square" with Cuisenaire rods
Figurate numbers whose "figures" are regular polygons, such as triangular numbers and square numbers, are also known as polygonal numbers. Prior to the modeling of the pentagonal and hexagonal numbers using the Cuisenaire rods, students in groups first studied the excerpt on the theory of polygonal numbers, credited to Nicomachus, from Thomas Heath's A History of Greek Mathematics: From Thales to Euclid (1921):
The gnomons of triangles are therefore the successive natural numbers. Squares are obtained by adding any number of successive terms of the series of odd numbers, beginning with \(1,\) or \[1,\,\,3,\,\,5,\,\,\dots,\,2n-1,\dots.\] The gnomons of squares are the successive odd numbers. Similarly the gnomons of pentagonal numbers are the numbers forming an arithmetical progression with \(3\) as common difference, or; \[1,\,\,4,\,\,7,\,\,\dots,\,1+(n-1)\cdot3,\dots;\] and generally the gnomons of polygonal numbers of \(a\) sides are \[1,\,\,1+(a-2),\,\,1+2(a-2),\,\,\dots,\,1+(n-1)(a-2),\dots\] and the \(a\)-gonal number with side \(n\) is \[1+1+(a-2)+1+2(a-2)+\cdots +1+(n-1)(a-2)=n+{\frac{1}{2}}n(n-1)(a-2).\] The general formula is not given by Nicomachus, who contents himself with writing down a certain number of polygonal numbers of each species up to heptagons. (p. 106)
This summative activity proved beneficial in that students appeared to excel in their understanding of the word “gnomon.” Some groups went back to the exploration of the triangular numbers and explained that every Cuisenaire rod in the formation of a triangular number could be thought of as a gnomon whose combination produces the triangular number itself: Gnomons of triangular numbers are consecutive positive integers. For instance, the gnomons of the triangular number 10 are 1, 2, 3, and 4.
Nicomachus wrote that any square number is the sum of two consecutive triangular numbers and (without modern notation) that
an \(a\)-gonal number of side n is the sum of an \((a-1)\)-gonal number of side \(n\) plus a triangular number of side \((n-1),\) i.e. \[n+{\frac{1}{2}}n(n-1)(a-2)=n+\frac{1}{2}n(n-1)(a-3)+\frac{1}{2}n(n-1)\quad{\sf{\small{(1921, p. 106).}}}\]
Letting \(a=5\) in the first half of Nicomachus' theorem, students obtained that a pentagonal number of side \(n\) is the sum of a square number of side \(n\) plus a triangular number of side \(n-1.\) From this, together with the template given in the text (see Figure 12a), students then came up with a diversity of visualizations of the pentagonal number sequence using the Cuisenaire rods (see Figures 12b-f).
a (1921, |
|||
b | |||
c | |||
d | |||
e | |||
f |
Figure 12: Pentagonal numbers 1, 5, 12, 22, 35, 51, 70 with Cuisenaire rods
Letting \(a=6\) in the first half of Nicomachus' theorem, students obtained that a hexagonal number of side \(n\) is the sum of a pentagonal number of side \(n\) plus a triangular number of side \(n-1.\) From this, together with the template given in the text (see Fig. 13a), students then came up with a diversity of visualizations of the pentagonal number sequence using the Cuisenaire rods (see Figures 13b-f).
a (1921, |
|||
b |
|||
c |
|||
d |
|||
e |
|||
f |
|||
g |
|||
h |
|||
i |
|||
j |
Figure 13: Hexagonal numbers 1, 6, 15, 28, 45, 66, 91 with Cuisenaire rods
We summarize the figurate numbers explored in the module along with their interrelationships modeled throughout the exploration (see Tables 1-2).
Notation | Figurate Number | Gnomonic Formula |
\(T_n\) | \(n\)th triangular number | \(T_n=1+2+3+\cdots+n\) |
\(O_n\) | \(n\)th oblong number | \(O_n=2+4+6+\cdots+2n\) |
\(S_n\) | \(n\)th square number | \(S_n=1+3+5+\cdots+(2n-1)\) |
\(P_n\) | \(n\)th pentagonal number | \(P_n=1+4+7+\cdots+(3n-2)\) |
\(H_n\) | \(n\)th hexagonal number | \(H_n=1+5+9+\cdots+(4n-3)\) |
Table 1: Figurate number notations and gnomonic formulas
Notation | Explicit Formula in \(n\) | Relations to Other Figurate Numbers |
\(T_n\) | \(T_n=\frac{n(n+1)}{2}\) | \(T_n=T_{n-1} + n\) |
\(O_n\) | \(O_n= {n(n+1)}\) |
\(O_n=2T_n\) \(O_n=S_n+n\) |
\(S_n\) | \(S_n=n^2\) |
\(S_n= T_{n-1}+ T_n\) \(S_n=O_{n-1}+n\) |
\(P_n\) | \(P_n=\frac{n(3n-1)}{2}\) |
\(P_n=S_n + T_{n-1}\) \(P_n= n+ O_{n-1}+ T_{n-1}\) \(P_n=n+ 3T_{n-1}\) \(P_n= T_n+ O_{n-1}\) |
\(H_n\) | \(H_n={n(2n-1)}\) |
\(H_n=P_n + T_{n-1}\) \(H_n=S_n+ 2T_{n-1}\) \(H_n= S_n+ O_{n-1}\) \(H_n=T_n+ 3T_{n-1}\) \(H_n= n+ 4T_{n-1}\) \(H_n= n+2O_{n-1}\) \(H_n= T_{2n-1}\) |
Table 2: Figurate number formulas and relationships
Heath, Thomas L. (1921). A History of Greek Mathematics: From Thales to Euclid (Volume I). Oxford: Clarendon Press. (Also available as a paperback from Dover Publications since 1981 and on Google Books.)
Katz, Victor J. (2009). A History of Mathematics: An Introduction (3rd edition). Addison-Wesley.
Lawlor, R. & Lawlor, D. (1979). Mathematics Useful for Understanding Plato, by Theon of Smyrna, Platonic Philosopher. San Diego: Wizards Bookshelf.
National Council of Teachers of Mathematics (1989). Historical Topics for the Mathematics Classroom (revision of 1969 edition edited by J.K. Baumgart, D.E. Deal, B.R. Vogeli, A.E. Hallerberg). Reston, VA: NCTM.