All groups G considered in this paper are finite and all representations of G are defined over the complex numbers. The reader unfamiliar with projective representations is referred to [6] for basic definitions and elementary results.

The work of Coxeter, Longuet-Higgins and Miller (1953) and of Skilling (1975) is extended to give a complete list of uniform compounds of uniform polyhedra. Symmetry relationships between polyhedra and their compounds are discussed.

In this paper we prove that the Hausdorff dimension of the set of (nondegenerate) singular two-dimensional vectors with uniform exponent μ in (1/2, 1) is equal to 2(1 − μ) for μ ⩾ $\sqrt2/2$, whereas for μ < $\sqrt2/2$ it is greater than 2(1 − μ) and at most equal to (3 − 2 μ)(1 − μ)/(1 − μ + μ2). We also establish that this dimension tends to 4/3 (which is the dimension of the set of singular two-dimensional vectors) when μ tends to 1/2. These results improve upon previous estimates of R. Baker, joint work of the first author with M. Laurent, and unpublished work of M. Laurent. Moreover, we prove a lower bound for the packing dimension, which appears to be strictly greater than the Hausdorff dimension for μ ⩾ 0.565. . . .

Let G be a connected real Lie group and let [gfr ] be its Lie algebra. We shall denote by [qfr ] ⊂ [gfr ] the radical of [gfr ]. Let [gfr ] = [qfr ] [ltimes ] [sfr ] (where [sfr ] is semisimple or 0) be a Levi decomposition of [gfr ] (cf. [11]). When [sfr ] ≠ 0 we can apply the Iwasawa decomposition on [sfr ] (cf. [8]) [sfr ] = [nfr ] [oplus ] [afr ] [oplus ] [kfr ], where [nfr ] is nilpotent and [afr ] is abelian and normalizes [nfr ] so that [nfr ] [oplus ] [afr ] is a soluble algebra. Since [nfr ] [oplus ] [afr ] normalizes [qfr ] it is clear that [rfr ] = [qfr ] [oplus ] [nfr ] [oplus ] [afr ] is a soluble Lie algebra of [gfr ]. By Lie's theorem (cf. [11]) we can find a basis on [rfr ]c = [rfr ] [otimes ] C for which the adjoint action of [rfr ] on [rfr ]c takes a triangular form. Let us denote by λ1(x); λ2(x), …, λn(x), x ∈ [rfr ] the corresponding eigenvalues. The λj's can be identified with elements of Homℝ([rfr ], C) and are called the roots of the adjoint action of [rfr ]. Let us denote by [Lscr ] = {L1, …, Lk} the set of the non zero real parts of the λj's. We shall say that the group G is a B-group if [Lscr ] ≠ &0slash; and if there exist α1, …, αk [ges ] 0, [sum ]kj=1 αj = 1, such that [sum ]kj=1 αjLj = 0. Otherwise we say that G is an NB-group. It can be shown that the above definition is independent of the particular choice of the Levi and Iwasawa decompositions that are used (cf. [13]).

We shall denote by dlg = dg (resp. drg) the left (resp. right) Haar measure on G and by m(g) = drg/dlg the modular function.

Let [Xscr ] = {X1, X2, …, Xn} be left invariant fields on G that verify the Hörmander condition (cf. [15]) and let Δ = −[sum ]X2j be the corresponding sub-Laplacian. Δ is formally self adjoint on the Hilbert space L2(G, drg) and the spectral gap of Δ is defined by

formula here

The Busemann–Petty problem asks whether origin-symmetric convex bodies in ℝn with smaller central hyperplane sections necessarily have smaller volume. The answer is affirmative if n ≤ 4 and negative if n ≥ 5. We study this problem in the complex hyperbolic n-space ℍnℂ and prove that the answer is affirmative for n ≤ 2 and negative for n ≥ 3.

1. A set of functions {øn (x)} is said to be closed L over an interval (a, b) if for an f (x) belonging to L

implies that f(x) = 0 almost everywhere. Here f(x) is a complex valued function of the real variable x.

Let X, Y be real Banach spaces, and let = {f(λ):λ ∈ m} be an m-parameter family of bounded linear operators from X to Y, with f(λ) depending continuously on λ. The cases m = 1 and m = 2 are considered, and conditions on are found which determine the null space of f(λ) for all λ near a given λ0 such that f(λ0): X → Y is a Fredholm operator. The results obtained are shown to be of particular interest in perturbed bifurcation theory.

It has been known for some time ((6), p. 270; (4), theorem 9·19) that if is a Boolean topos, then the full subcategory Kf of Kuratowski-finite objects in is again a topos. For a non-Boolean topos , however, Kf need not be a topos, as can be seen when is the Sierpinski topos ((1), example 7·1); on the other hand, two other full subcategories of , coinciding with Kf when is Boolean, suggest themselves as candidates for a subtopos of finite objects. Of one of these, the category dKf of decidable K-finite objects in , the Main Theorem of (1) asserts that it is always a (Boolean) topos. The other is the category sKf of -subobjects of K-finite objects. The inclusions

dKf ⊆ Kf ⊆ sKf are clear.

are clear.

The notion of cycle-free partial order (CFPO) was defined in [9] and the class of sufficiently transitive CFPOs was investigated, and in many cases, classified. In [10] a complete classification was given for the countable 3- or 4-CS-transitive CFPOs having a finite chain and embedding an infinite ‘alternating chain’ ALT. Here, the major case is that of the so-called ‘skeletal’ CFPOs, in which the CFPO itself forms a mere skeleton of the structure of the whole picture, which is more accurately provided by its Dedekind–MacNeille completion. It was asserted that a similar classification should also be possible in the infinite chain case (still with the countability restriction), and it is the object of the present paper to finish this task.

The overall plan of the work is similar in style to [10]. The CFPOs we study can all be construed as arising from a chain, suitable ‘adorned’ with instructions as to how to branch (and repeatedly). An obvious difference from the finite chain case is that, this time, many of the points in the chain will actually survive as elements of the structure; whereas in [10], since the chains of the structure had to have length 2, only the endpoints would. On the whole, the treatment is parallel, with just a small number of additional cases.

We recall the main ideas from [10], quote results that are needed, and concentrate on aspects of the structures which are new or specific to the infinite chain case. We conclude by remarking on some direct connections between the finite and infinite chain cases.

The conjecture of Brown, Erdős and Sós from 1973 states that, for any k ≥ 3, if a 3-uniform hypergraph H with n vertices does not contain a set of k +3 vertices spanning at least k edges then it has o(n2) edges. The case k = 3 of this conjecture is the celebrated (6, 3)-theorem of Ruzsa and Szemerédi which implies Roth’s theorem on 3-term arithmetic progressions in dense sets of integers. Solymosi observed that, in order to prove the conjecture, one can assume that H consists of triples (a, b, ab) of some finite quasigroup Γ. Since this problem remains open for all k ≥ 4, he further proposed to study triple systems coming from finite groups. In this case he proved that the conjecture holds also for k = 4. Here we completely resolve the Brown–Erdős–Sós conjecture for all finite groups and values of k. Moreover, we prove that the hypergraphs coming from groups contain sets of size $\Theta (\sqrt k )$ which span k edges. This is best possible and goes far beyond the conjecture.

A quandle is the algebraic distillation of the second and third Reidemeister moves, on unframed links. Quandles have been studied by many people including Kauffman[K], Fenn and Rourke[F-R] and Joyce [J]. Joyce defines the fundamental quandle, which is a classifying invariant of irreducible, unframed links. Fenn and Rourke, who study generalised quandles or racks in [F—R], show that the natural map from the fundamental quandle of a knot K to the knot group is never injective if K is a connected sum. We now prove that the natural map from the fundamental quandle to the fundamental group of the complement of a link in S3 is injective if and only if the link is prime.

This paper continues the study initiated in [10] of the Markov branching process (MBP) subject to an independent Poisson process of catastrophe times at which the population size is reduced by random decrements whose distribution is independent of the current population size. More generally, let (Xt:t ≥ 0) be the Feller process on the non-negative integers ℕ+ having the generator

Let Si: ℝd → ℝd for i = 1, …, n be contracting similarities, and let (p1, …, pn) be a probability vector. Let K and μ be the self-similar set and the self-similar measure associated with (Si,pi)i. For q ∈ ℝ and r > 0, define the qth covering moment and the qth packing moment of μ by

[formula here]

where the infimum is taken over all r-spanning subsets E of K, and the supremum is taken over all r-separated subsets F of K. If the Open Set Condition (OSC) is satisfied then it is well known that

[formula here]

where β(q) is defined by [sum ]ipqirβi(q) = 1 (here ri denotes the Lipschitz constant of Si). Assuming the OSC, we determine the exact rate of convergence in (*): there exist multiplicatively periodic functions πq, Πq: (0,∞) → ℝ such that

[formula here]

where ε(r) → 0 as r[searr ]0. As an application of (**) we show that the empirical multi-fractal moment measures converges weakly:

[formula here]

where, for each positive r, Er is a (suitable) minimal r-spanning subset of K and Fr is a (suitable) maximal r-separated subset of K, and [Hscr ]q,β(q)μ and [Pscr ]q,β(q)μ are the multifractal Hausdorff measure and the multifractal packing measure, respectively.

If M is a group, Z(M) its integral group ring and AM the augmentation ideal, then following Passi we can form the Abelian groups

It is well known that every orientation preserving homeomorphism of the plane $\mathbb{R}^2$ with a $k$-periodic orbit ${\cal O}$ ($k \geq 2$) necessarily has a fixed point (see [4] or [5], [7], [11]). It is then interesting to know if ${\cal O}$ is linked with one of these fixed points. The difficulty in answering this question depends a lot on the precise sense we give to the word “linked”. Before discussing briefly this point, let us observe this general problem has a very natural counterpart in the framework of orientation reversing homeomorphisms of the sphere $\mathbb{S}^2$. Indeed, every such homeomorphism possessing a $k$-periodic orbit ${\cal O}$ ($k\geq 3$) also has a 2-periodic orbit ([2]) and one can ask if ${\cal O}$ is linked with one of these 2-periodic orbits. The aim of this paper is to try to answer this question.

1. If the canonical series of an algebraic curve of genus p is compounded of an involution of sets of points on the curve then the involution must be rational and of order two, and the canonical model is a repeated rational normal curve of order p − 1 with 2p + 2 branch points. An analogous question suggests itself for algebraic surfaces. Under what conditions is the canonical model of a surface, whose geometric genus is not less than two, a multiple surface, and what in this case are the properties of the simple surface on which the multiple surface is based? In this paper we give particular examples of multiple canonical surfaces and attempt to go some way towards the solution of the general problem.

Introduction. Let sn denote the infimum of the set of real numbers X, where x belongs to X if, and only if, x is the Besicovitch dimension of the residual set of a packing of n-spheres into the unit n-cube In, of Rn. In recent work ((1)) I have shown that s2 is greater than one, and, quite naturally, I have since been asked whether or not the proof can be generalized to prove the analogous result, sn greater than n − 1, in Rn. Whilst it is clear, in theory, that this could be done, in practice the details might become rather complicated. However, such a generalization is unnecessary, for the result sn > n − 1 is a trivial consequence of combining the result s2 > 1 with the following theorem.

The concept of a local system of a set W is defined in ((8), p. 166) and ((12), p. 126). Recall that a set ℒ of subsets of W is a local system if Uℒ = W and ℒ is directed in the following sense: for every finite set H1, …, Hn of elements of ℒ, there is an M∈ℒ such that Hi⊂M for 1≤i≤n. If Σ is a class of groups, L(Σ) is the class of all groups G that possess a local system of Σ subgroups. Σ satisfies the local theorem, or is L-closed, if L(Σ)⊂Σ. Many classes of groups which satisfy the local theorem are discussed in ((12), pp. 126–144).