Taxicabs and Sums of Two Cubes

From MaRDI portal





This is a magnificent introduction to the beautiful world of elliptic curves. The author motivates the subject through Hardy's taxicab number 1729, which Ramanujan delighted in as being the smallest positive integer that is the sum of two cubes of positive integers in two different ways \((1729= 1^ 3+ 12^ 3= 9^ 3+ 10^ 3)\). Silverman asks whether we can find other such integers, and integers which can be written in more ways as the sum of two such cubes. He notes that from any \(k\) distinct rational points on a curve \(x^ 3+ y^ 3= a\) we can find \(k\) ways of writing some integer \(A\) as the sum of two cubes of integers, for we take \(A= az^ 3\) where \(z\) is the least common multiple of the denominators of all the \(x\)'s and \(y\)'s. He notes that we can take \(k\) arbitrarily large if we choose \(a\) so that the curve has positive rank. However this is really cheating; we shouldn't be allowed to multiply through in this way, so the question is then asked assuming \(A\) to be cubefree. Currently the best that is known is that we can write some such \(A\) in three distinct ways (e.g. \(436^ 3+ 167^ 3= 423^ 3+ 228^ 3= 414^ 3+ 255^ 3\)); no cubefree \(A\) is known that can be written in four distinct ways as the sum of cubes of positive integers. The author uses these considerations to discuss the Mordell-Weil theorem, the question of effective algorithms to compute the rank, the question as to whether the rank is unbounded as one goes through a family of twists, and how the number of integral points on a curve is related to the rank. This is the perfect \textit{Monthly} paper, because it should succeed in motivating the beginner to learn a lot more about a very important subject.











This page was built for publication: Taxicabs and Sums of Two Cubes

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4695133)