What are the units, irreducibles, and primes of the ring ${\mathbb{Z}}\sqrt n $, the set of all numbers $a + b\sqrt n $ where a and b are integers and n is a fixed positive square-free integer? In the ring ${\mathbb{Z}}$, primes and irreducibles are synonymous and its units are ±1. ${\mathbb{Z}}\sqrt n $ is wilder, and our modest goal here is to catalogue all such numbers for ${\mathbb{Z}}\sqrt {{\rm{[10]}}} $, where a and b range from 0 to 10; the result appears in Figure 1. Here are a few teasers that may induce a reader to read on: $3 + \sqrt {10} $ is a unit; 2, 3, 5, and 7 are irreducibles, but not 31; and 7 is the least positive integer that is prime in both ${\mathbb{Z}}$ and ${\mathbb{Z}}\sqrt {{\rm{[10]}}} $.

Spirals are among the most intriguing of geometrical patterns, frequently encountered in the world around us at all scales, from the cosmic spiral arms of galaxies to the microscopic structure of the DNA molecule. For centuries, humans have studied these patterns, classifying them, giving them different names, and describing them mathematically. The most common definition of a two-dimensional spiral is a curve on the plane traced by a point that winds around a certain fixed point (called the spiral's pole), while monotonically approaching or receding from it, depending on the direction of motion.

When defining a group, do we need to include closure? This is a detail that is often touched upon when the notion of a group is introduced to undergraduates. Should closure be listed as an axiom in its own right, or should it be regarded as an inherent property of the binary operation? There is no clear answer to this question, although there are firm opinions on both sides. Indeed, a very brief survey of group theory textbooks found in [1, pp. 458-459] suggests that there is a rough 50 : 50 split between authors who include closure explicitly and those who do not. In this Article, we go back to the beginning of the twentieth century to provide some historical perspective on this problem.

While the notion of roots of a quadratic polynomial is rudimentary in high school mathematics, that of its fixed points is uncommon. A real or complex number is a fixed point of a polynomial p (x) p (θ) = θ. The fact that the notion of fixed point of polynomials is not commonly covered in high school or undergraduate mathematics is surprising because the relevance of the fixed points of a quadratic can be demonstrated easily via iterative methods for the approximation of such numbers as , when the quadratic formula offers no remedy.

Consider the problem of proving that, for any positive numbers x, y and z,

$${\rm{9 (}}{{\rm{x}}^{\rm{3}}}{\rm{ + }}{{\rm{y}}^{\rm{3}}}{\rm{ + }}{{\rm{z}}^{\rm{3}}}{\rm{) }} \ge {\rm{ (x + y + z}}{{\rm{)}}^{\rm{3}}}{\rm{ }}{\rm{.}}$$

This is an example of a type of inequality that frequently occurs in Olympiad-style problems, [1]. These problems may involve symmetric functions of more or fewer variables than the three used here. However, three variables are commonly used and appear to give appropriately difficult problems without making excessive computational demands.

We consider the second order homogeneous linear differential equation (H)

$${ ay'' + by' + cy = 0 }$$

We shall present some new proofs concerning the distribution of the digits in decimals. We finish with a proof of Borel's normal number theorem. The proof given here is more direct and straightforward than the one given in [1]. This theorem is concerned with the distribution of the digits within infinite decimals. The actual statement of the theorem will be given later after some intermediate results have been established. The theorem is not only unexpected but seemingly counter-intuitive. In order to follow the arguments given here it is useful to have a numerical example as we will count the decimals and then their digits. We shall deduce the results for a general base b.

In this Article, we derive a method for classifying all positive integer solutions of the equation x2 + ry2 = z2, where r is a given positive rational. In order to simplify notation, such a solution with x = a, y = b and z = c will be denoted by the ordered triple (a, b, c) and, in all that follows, the term solution will be taken to mean a positive integer solution of the above equation. The method employed will be similar to that used in [1] to find all integer triangles containing an angle whose cosine is known.

Which is your favourite formula in triangle geometry? Mine is definitely the formula for the area of triangle OIH . It is well known that the perpendicular bisectors to the sides of any triangle are concurrent at a point O (centre of the circumcircle), that the angle bisectors to the vertex angles are concurrent at a point I (centre of the incircle), and that the altitudes are concurrent at a point H. If the triangle is not isosceles, then these three points are all different and uniquely determine a new triangle OIH (see Figure 1), whose area can be expressed in terms of the sides a, b, c of the original triangle. I derived such a formula 20 years ago, and later found out that it had been studied a century earlier.

In a previous paper [1], I described a method for computing sums of powers via a matrix composed from Pascal's triangle. (The motivation for this came from a mathematical puzzle posed by Paul Maslanka in the Guardian.) Since then I have read or viewed a number of papers on sums of powers. Some were nearly four hundred years old and the most recent one was published this year. One rather obscure work I stumbled across was ‘On the sums of powers of natural numbers’ by the Japanese mathematician Mr T. Kariya, published in Tokyo in 1907. If any readers know more of Mr Kariya I would be glad to hear about him.

Magic squares have long been popular in recreational mathematics. Their potential for introducing students to ideas in linear algebra was recognised over forty years ago in [1] and later in [2]. More recently they have proved to be a fascinating topic for undergraduate exploration, especially when students have access to a computer algebra package [3]. Some results on powers of magic square matrices can be found in [4], [5] and [6]. (Readers who google the title ‘Odd magic powers’ of Thompson’s paper [5] will be treated to a wide variety of non-mathematical exotica!)

When considering ‘convergence’ many people think of number sequences or contexts arising from calculus. But there are also interesting phenomena of convergence – easy to visualize – arising in elementary geometry. Some of them are so elementary that they can be dealt with at school, for instance an example of iteration that is described in [1, p. 171f] (see also [2, p. 59], [3, p. 222f] and [4, p. 42ff]).

There are recursive expressions (see [1]) for sequentially generating the integer solutions to Pell's equation: p2 − Dq2 = 1, where D is any positive non-square integer. With known positive integer solution p1 and q1 we can compute, using these recursive expressions, pn and qn for all n > 1. See Table in [2] for a list of smallest integer, or fundamental, solutions p1 and q1 for D≤ 128. These (pn, qn) pairs also form rational approximations to that, as noted in [3, Chapter 3], match with convergents (Cn = pn / qn) of the Regular Continued Fractions (RCF, continued fractions with the numerator of all fractions equal to 1) for .

The launch and subsequent motion of a projectile provide a context for several quantities that yearn to be optimised. Most notable is the horizontal range of the projectile, a problem dating back to Galileo and still studied in modern times; see, for example [1], [2], [3], [4]. In a different direction, the articles [5] and [6] provide a solution to the problem of finding the angle of launch that results in the trajectory of longest arc length.

Two common methods of approximating difference problems are by rounding or truncating the minuend and the subtrahend before calculating the difference. Since rounding generally gives a better approximation of a number than truncation, one might expect that rounding also generally gives a better approximation in difference problems. We show that intuition to be false.

The number πis perhaps the most famous irrational number. This constant is equal to the ratio of the circumference of a circle to its diameter. One of the most well-known mathematical problems of antiquity, which is related to π, is how to construct by using a ruler and compasses a square which has the same area as a circle. This particular problem cannot be solved, due to the fact that π is a transcendental number, which means that it cannot be obtained as the root of a polynomial equation with rational coefficients. It was Euler in the 18th century who established the notation π.