Abstract: Multicrossings, which have previously been defined for classical knots and links, are extended to virtual knots and links. In particular, petal diagrams are shown to exist for all virtual knots, providing a new combinatorial framework for studying these topological objects.
Introduction
In classical knot theory, knots are typically represented by projections where only two strands cross at any given point, known as double crossings. However, recent advancements have introduced multicrossings, where $n$ strands intersect at a single point. This paper, "Virtual multicrossings and petal diagrams for virtual knots and links," extends this concept to virtual knots—a generalization of classical knots that represent embeddings in thickened surfaces. The authors prove that every virtual knot can be represented by a single, complex "petal" diagram.
Defining Valid Virtual Multicrossings
A primary challenge in virtual knot theory is avoiding "forbidden moves"—sequences of moves that, if permitted, would render every virtual knot equivalent to a simple unknotted circle. To generalize multicrossings safely, the authors define a "valid" virtual $n$-crossing by imposing a specific constraint: no subset of three strands can contain exactly one virtual and two classical crossings.
This constraint ensures that when a multicrossing is "resolved" into standard double crossings, the resulting configurations remain equivalent under the allowed moves of virtual knot theory. As shown in Figure 1, a multicrossing can be viewed as a compact representation of a sequence of double crossings along a vertical axis.
Classification and Combinatorial Counting
The researchers identify three fundamental types of virtual triple crossings based on the nature of their constituent double crossings:
- Type I: Three classical crossings.
- Type II: One classical and two virtual crossings.
- Type III: Three virtual crossings.
To determine the number of distinct virtual $n$-crossings, the authors employ Burnside’s Lemma and Bell numbers. They treat the strands as elements in a set partition where equivalence denotes a classical crossing. Their enumeration formula shows that the number of types grows rapidly with $n$: while there are only 5 types of triple crossings, a 10-crossing has over 5.8 million distinct types.
Petal Universality
The paper's most significant result is Theorem 5.1, which establishes that every virtual knot possesses a virtual petal diagram. A petal diagram is a projection with exactly one central multicrossing and a series of non-nested loops (petals) radiating from it.
Using an algorithmic proof based on signed Gauss codes, the authors demonstrate that any virtual knot, regardless of its complexity in a standard projection, can be "squashed" into this single-point representation. Furthermore, they establish that the "virtual petal number" (the minimum number of petals needed) is bounded linearly by the number of classical crossings in the knot's standard projection.
Future Directions
While this work proves the existence of petal diagrams for all virtual knots, several questions remain open:
- Efficiency: The current algorithm provides a proof of existence but does not necessarily find the minimal number of petals.
- Classical vs. Virtual: It is unknown if allowing virtual multicrossings could reduce the petal number of a classical knot.
- Lower Bounds: Finding a linear lower bound for the petal number across infinite families of knots remains a significant challenge for future research.
Summary updated