Metrisable Lie algebras have been defined by Tsou and Walker (1957). Their definition is adopted below in § I.

The object of the present paper is to display something of the geometrical background of such algebras, particularly for those taken over the field of real numbers.

§ I is introductory. In § 2 appears a statement, made as brief as possible because it is wholly classical, of the relationship between the vector-space of the Lie algebra L and the associated affine and projective spaces A and P. Some properties of metrisable Lie algebras are then examined in terms of the geometry of P, which provides an (n –I)-dimensional map of the n-dimensional algebra L. It is assumed throughout the paper that the Lie algebras under discussion are non-abelian, since the projective map of an abelian algebra presents nothing of interest.

As the present work is intended as no more than a preliminary, it is confined, so far as its applications are concerned, to a discussion of metrisable algebras of dimensions 3 and 4 and to one example of an algebra of dimension 6.

I have been privileged in preparing the paper to have access to the typescript of the paper by Tsou and Walker referred to above, and also to the doctoral thesis (1955) of the former. I am greatly indebted to them both, and also to Dr Paul Cohn and to a referee for suggestions regarding certain details of presentation.

In this paper are considered certain numbers called moduli associated with a rectangular matrix with complex elements. These moduli have to satisfy a set of conditions analogous to those satisfied by the modulus of a complex number. For a complex rectangular matrix A = (aij) it is shown that R(A), C(A) and | A |° are moduli of A where:

where cmin(H) and cmax(H) denote, respectively, the minimum and maximum characteristic values of the hermitian matrix H, A* being the transpose conjugate of A. Using the various properties of a modulus of a matrix and taking R(A), C(A) and | A |° as the moduli, a number of known results about the characteristic values of a matrix are obtained and extended. Relations between |A|°, |A |° R(A), p(A) and C(A), ɣ(A) are also studied. These relations provide a number of results about estimates of bounds of characteristic values of sums and products of matrices.

The bivariate distribution corresponding to the univariate negative binomial, and the corresponding distribution when sampling is without replacement, are investigated. Formulas are derived for the factorial moment generating functions and for the regression equations, which are linear.

Every algebraic equation can be uniformized by automorphic functions belonging to a certain group of bilinear transformations. In certain cases, such as for hyperelliptic equations, this group is a subgroup of the monodromic group of a differential equation of the form

where R(z) is a rational function which, in general, contains unknown parameters as coefficients. A conjecture of E. T. Whittaker regarding the values of these parameters for the hyperelliptic case is proved for a wide variety of algebraic equations whose branch points possess certain symmetric properties, and is extended to equations of higher type. In several cases, the uniformizing functions belong to subgroups of the groups of the Riemann-Schwarz triangle functions.

A result proved elsewhere concerning the structure of linear algebras of genus one is used (i) to classify Jordan algebras of genus one, and (ii) to obtain necessary and sufficient conditions for an algebra of genus one to be simple. The class of simple Jordan algebras of genus one and dimension not less than 4 over an algebraically closed field is shown to coincide with the class of simple Jordan algebras of degree two.

Canonical forms of the four-dimensional complex Lie algebras are obtained by considering the roots of certain well-defined vectors of the algebras. A complete set of characters of the algebras is also given, enabling any given four-dimensional complex Lie algebra to be identified with one of the canonical forms.

The function ɸx = log x satisfies the functional equation ɸxy = λɸx + μɸy + ϰ, where in this case x, y are complex variables, λ = μ = i, and ϰ = 2πi, o or − 2πi according as σ < − π, − π < σ < π, π < σ, where σ = arg x + arg y. Generalizing this situation, let A be a linear algebra with basis e1, …, en over the real or complex field and let ɸx be a complexvalued function of the hypercomplex variable x = Σξiei, i.e. of the n real or complex variables ξi. Assume that the gradient ∂ɸx, i.e. the column vector of partial derivatives {∂ɸx/∂ξi}, exists at a general point of A. Then ɸx is called an entropic function if it satisfies a functional equation of the above-mentioned form and obeys certain other postulates, ϰ being a step function of the two hypercomplex variables. Values of the constants λ, μ (complex numbers, not both zero) for which a solution exists are entropic roots of A. They are usually discrete.