An oriented triple system of order \(v\), denoted OTS\((v)\), is said to be \(d\)-cyclic if it admits an automorphism consisting of a single cycle of length \(d\) and \(v-d\) fixed points, \(d\geq 2\). In this paper, we give necessary and sufficient conditions for the existence of \(d\)-cyclic OTS\((v)\). We solve the analogous problem for directed triple systems.

Let \(A_m(n, k)\) denote the number of permutations of \(\{1, \ldots, n\}\) with exactly \(k\) rises of size at least \(m\). We show that, for each positive integer \(m\), the \(A_m(n, k)\) are asymptotically normal.

Let \(G\) be a graph of order \(n\) and \( X\) a given vertex subset of \(G\). Define the parameters:
\(\alpha(V) = \max\{|S| \mid S\}\) is an independent set of vertices of the subgraph \(G(X)\) in \(G\) induced by \(X\)
and
\(\sigma_k(X) = \min\{|\Sigma_{i=1}^{k}d(x_i)| \mid \{x_1,x_2,\ldots,x_k\} \}\) is an independent vertex set in \( G[X]\)
A cycle \(C\) of \(G\) is called \(X\)-longest if no cycle of \(G\) contains more vertices of \(X\) than \(C\). A cycle \(C’\) of \(G\) is called \(X\)-dominating if all neighbors of each vertex of \(X\setminus V(C)\) are on \(C\). In particular, \(G\) is \(X\)-eyclable if \(G\) has an \(X\)-cycle, i.e., a cycle containing all vertices of \(X\). Our main result is as follows:
If \(G\) is \(1\)-tough and \(\sigma_3(X) \geq n\), then \(G\) has an \(X\)-longest cycle \(C\) such that \(C\) is an \(X\)-dominating cycle and \(|V(C) \cap X| \geq \min\{|X|. |X| + \frac{1}{3}\sigma_3(X) – \sigma(X)\}\), which extends the well-known results of D. Bauer et al. [2] in terms of \(X\)-cyclability. Finally, if \(G\) is \(2\)-tough and \(\sigma_3(X) \geq n\), then \(G\) is \(X\)-cyelable.

In 1992, Mahmoodian and Soltankhah conjectured that, for all \(0 \leq i \leq t\), a \((v, k, t)\) trade of volume \(2^{t+1} – 2^{t-i}\) exists. In this paper we prove this conjecture and, as a corollary, show that if \(s \geq (2t – 1)2^t\) then there exists a \((v, k, t)\) trade of volume \(s\).

We prove two new characterization theorems for finite Moufang polygons, one purely combinatorial, another group-theoretical. Both follow from a result of Andries Brouwer on the connectedness of the geometry opposite a flag in a finite generalized polygon.

Cyclonomial coefficients are defined as a generalization of binomial coefficients. It is proved that each natural number can be expressed, in a unique way, as the sum of cyclonomial coefficients, satisfying certain conditions. This cyclonomial number system generalizes the well-known binomial number system. It appears that this system is the appropriate number system to index the words of the lexicographically ordered code \(L^q(n, k)\). This code consists of all words of length \(n\) over an alphabet of \(q\) symbols, such that the sum of the digits is constant. It provides efficient algorithms for the conversion of such a codeword to its index, and vice versa.

We investigate the connections between families of graphs closed under (induced) subgraphs and their forbidden (induced) subgraph characterizations. In particular, we discuss going from a forbidden subgraph characterization of a family \(\mathbb{P}\) to a forbidden induced subgraph characterization of the family of line graphs of members of \(\mathbb{P}\) in the most general case. The inverse problem is considered too.

A family of double circulant quasi-cyclic codes is constructed from the incidence matrices of difference sets associated with hyperplanes in projective space. A subset of these codes leads to a class of doubly-even self-orthogonal codes, and two classes of self-dual codes.

All nonisomorphic \(2\)-\((21, 6, 3)\) designs with automorphisms of order \(7\) or \(5\) were found, and the orders of their groups of automorphisms were determined. There are \(33\) nonisomorphic \(2\)-\((21, 6, 3)\) designs with automorphisms of order \(7\) and \(203\) with automorphisms of order \(5\).

Let \(G\) be a graph with even order \(p\) and let \(k\) be a positive integer with \(p \geq 2k + 2\). It is proved that if the toughness of \(G\) is at least \(k\), then the subgraph of \(G\) obtained by deleting any \(2k – 1\) edges or \(k\) vertices has a perfect matching. Furthermore, we show that the results in this paper are best possible.