1. Let r(n) be the number of representations of n as a sum of two squares, d(n) the number of divisors of n, and

where γ is Euler's constant. Thus P(x) is the error term in the problem of the lattice points of a circle, and Δ(x) the error term in Dirichlet's divisor problem, or the problem of the lattice points of a rectangular hyperbola.

This paper is an attempt to apply the methods of the theory of algebras to the more general problem of the structure of semi-groups, i.e. systems with one composition, satisfying the associative law.

In a previous paper, under the same title, I considered the problem of how far apart two consecutive primes can be. The present paper is concerned with the opposite question. How near together can large primes lie? The published literature on this subject is scanty and, though interesting, is mainly negative in character. It appears to be very difficult to give any answer that is not trivial, or that is at all illuminating.

Generalities. Suppose that

is real and L2(0, ∞), that

is its Mellin transform, and that

Then K(1)(x) generates a “Watson transformation”, with a “Fourier theorem”

valid for every F(x) of L2(0, ∞).

Some theorems in the operational calculus, taking definite integration as the fundamental operator, are proved for discrete systems. It is suggested that it is physically more satisfactory to regard the solution for a continuous system as the limit of the solutions for a set of discrete systems, rather than as the solution of a partial differential equation. The necessary and sufficient condition for the validity in this sense of the usual operational method for continuous systems appears to be that the discrete systems considered shall not tend to infinite instability; that is, that all poles of the Bromwich integrand shall always have real parts less than some fixed positive quantity. The correction for finiteness of the number of degrees of freedom is examined for a case analogous to heat conduction and found to be unimportant.

This paper is a continuation of four others under the same title†. The paragraphs are numbered following on to those of the fourth paper of the series. In § XXIV we show that, if λ(A) is a linear functional, then there exists a resolution Eμ such that λ(A) = ∫μdr(AEμ), and if B = ∫μdEμ is bounded, then λ(A) = τ(AB)‡ for all A, where τ is the trace. This implies that τ(A) is a linear functional, and that the conjugate space ℒ, i.e. the space of the linear functionals, has a subset ℒ′ which is in (1, 1) correspondence with the original set of operators, and that in this correspondence the linear functional τ(A) is associated with the unit operator.

Two theorems by the late Prof. F. Morley have aroused special interest, one being the chain of circle properties discovered by De Longchamps and independently by Pesci, Grace and by Morley himself, and the other the remarkable property that if the angles of any triangle are trisected, then the trisecting lines meet in pairs to form an equilateral triangle. In a previous paper‡ the present writer has shown that the circles making up the chain may be derived from a rational normal Cn in [n] by the following process.

1. The description “approximate”, as applied to properties of a measurable function at a point, has come to mean, roughly speaking, “with the neglect of sets of measure zero”. If a function has, at a point x0, a certain property, such as differentiability or continuity, then it has the same property in the approximate sense, but not conversely.

It is known † that, if G is any p-group (group whose order is a power of the prime p) the class of which is less than p, then the order of a product of elements of G cannot exceed the orders of all the factors. This result in general ceases to hold when the class of G exceeds p − 1. In the present paper, we prove the following general results.

Alexander (1, 2) has introduced certain topological invariants of a manifold which arise from the intersections of cycles of non-complementary dimensions, and he points out that they are not derivable from the Betti and torsion numbers, nor from the fundamental group. In the present paper we consider some topological invariants of this type on an algebraic surface, and, although we cannot define them completely, we show that they are intimately connected with the multiplications of the period matrix of the simple integrals of the first kind. We are then led to a concept which we call the “intersection group” of the surface, which is, by its definition, topologically invariant, and we show that it is also invariant under birational transformations. The proofs are based on Lefschetz's theory of cycles for an algebraic surface (4) and some simple properties of the period matrix of an algebraic curve. The results obtained here have a number of applications to the theory of ∞3 correspondences between algebraic surfaces, as we propose to show in a later paper.

Suppose that is an integral modular form of dimensions − κ, where κ > 0, and Stufe N, which vanishes at all the rational cusps of the fundamental region, and which is absolutely convergent for The purpose of this note is to prove that

The notation employed is that of my second paper under the same general title* I refer to this paper as II.

This paper gives a criterion for determining whether real, non-zero solutions of a linear differential equation of the second order have an infinite or a finite number of zeros, or, in short, are oscillatory or non-oscillatory, as the independent variable tends to infinity.

1. The determination of an explicit formula for the postulation of a multiple variety of given characters for primals of sufficiently large order is one which seems to have received attention only in special cases. The postulation of a multiple curve for surfaces in ordinary space is known*, and Roth† has obtained the postulation of a multiple surface in [4], free from singularities, for primals in that space. Apart from these results, and the trivial case of isolated multiple points, the only general results seem to be those of Torelli‡ and Giambelli§ for the case in which the multiple variety is a Vh which is the complete intersection of r − h primals in [r].

The presence of H2 molecules reduces the temperature of the cosmical cloud to a value that is small compared with the estimate given by Eddington. The necessary conditions for the molecules to persist at the capture radius of hot stars are investigated in the present paper, and it is shown that provided that the density of the cosmical cloud is sufficiently high the molecules will not suffer appreciable dissociation, and that radiation pressure will have only a negligible effect on the hydrogen. The critical density for a typical B star appears to be about 5 × 10−21 g. per c.c.

1. Any number x can be expressed uniquely in the form

where the xr's are positive integers and where xr < r. In the present paper we consider the set of numbers Eb for which the xr's are bounded, so that 0 ≤ xr < b say, where b also is an integer. We prove that this set has dimension function

h(t) = b−u,

where t = euu−u−½(b − 1)(2π)−½, in the sense of Hausdorff.

In a paper published in the Proceedings of the London Mathematical Society, (2), 40 (1936), 143, Welchman introduced the idea of fundamental scrolls from which all special scrolls may be obtained by projection, and he developed the theory of fundamental line scrolls. These fundamental line scrolls are generated by joins of pairs of corresponding points in a cyclic (1, 1) correspondence of period 2 on a canonical curve. In this paper I generalize some of Welchman's results, and then go on to consider certain fundamental scrolls which have a particular interest.

The elastic scattering of Yukawa's particles, or mesons, by a static electric field is recalculated by considering the scattering as a process in which the incident meson is absorbed and another meson, i.e. the scattered meson, is created, and Laporte's result is confirmed. This method promises an easier handling of the second approximation.

A similar dependence of the angular distribution of the emitted quanta upon the spin of the incident mesons in a process like Y− + P → N + hν is found. Thus, the angular distribution contains a dependence upon the azimuthal angle in general. For the reverse process, the angular distribution of the emitted mesons, both longitudinal and transverse, contains also a dependence upon the azimuthal angle, which disappears only when the incident light is circularly polarized.

The writer wishes to thank Prof. W. Pauli for calling his attention to the azimuthal dependence in the problem of elastic scattering and Dr C. Møller for help given during the preparation of the paper.

The stability of lattices is discussed from the standpoint of the method of small vibrations. It is shown that it is not necessary to determine the whole vibrational spectrum, but only its long wave part. The stability conditions are nothing but the positive definiteness of the macroscopic deformation energy, and can be expressed in the form of inequalities for the elastic constants. A new method is explained for calculating these as lattice sums, and this method is applied to the three monatomic lattice types assuming central forces. In this way one obtains a simple explanation of the fact that the face-centred lattice is stable, whereas the simple lattice is always unstable and the body-centred also except for small exponents of the attractive forces. It is indicated that this method might be used for an improvement of the, at present, rather unsatisfactory theory of strength.

In this paper we consider the slow two-dimensional motion of viscous liquid past a sharp edge projecting into the stream, the motion being one of uniform shear apart from the disturbance caused by the projection. A special form is assumed for the boundary so that a method lately developed by N. Muschelišvili can be used in solving the biharmonic equation; a simple expression in finite terms is found for the stream function ψ1. Fig. 1 shows the section ABC of the fixed boundary of the liquid, the equation of the curve ABC is given in § 2, and ψ1 in § 3.

The interaction between an external electromagnetic field and a nucleus, including its exchange meson field, has recently been investigated by several authors (1). The interaction between a vector potential A and a nucleus has been found to be expressible in terms of the electric and magnetic multipole moments of the latter. It is the object of this note to discuss the corresponding interaction with a scalar potential V, and its connexion with previous results.