Maximal matroids and counterexamples
Extracted source text ยท arXiv 2606.14663
\documentclass[11pt, a4paper]{article}
\usepackage[utf8]{inputenc}
\usepackage{amsfonts, amssymb, amsmath, amsthm}
\usepackage{mathtools}
\usepackage{enumerate}
\usepackage{graphicx}
\usepackage{fullpage}
\usepackage{verbatim}
\usepackage{hyperref}
\usepackage[normalem]{ulem} % Crossing out an older text
\usepackage{authblk}
\hypersetup{colorlinks=true, linkcolor=red, urlcolor=blue, citecolor=blue, pdfpagemode=UseNone, pdfstartview=FitH, bookmarksopen=true}
\hypersetup{pdftitle=Title}
\renewcommand{\baselinestretch}{1.1}
%%%%%
%%%%% THEOREMS DEFINITIONS
%%%%%
\newtheorem*{theorem*}{Theorem}
\newtheorem{theorem}{Theorem}
\newtheorem{dtheorem}[theorem]{Disproved Theorem}
\newtheorem{qtheorem}[theorem]{Questionable Theorem}
\newtheorem*{lemma*}{Lemma}
\newtheorem{lemma}[theorem]{Lemma}
\newtheorem{problem}[theorem]{Problem}
\newtheorem*{proposition*}{Proposition}
\newtheorem{proposition}[theorem]{Proposition}
\newtheorem*{fact*}{Fact}
\newtheorem{fact}[theorem]{Fact}
\newtheorem*{question*}{Question}
\newtheorem{question}[theorem]{Question}
\newtheorem{conjecture}[theorem]{Conjecture}
\newtheorem*{corollary*}{Corollary}
\newtheorem{corollary}[theorem]{Corollary}
\newtheorem*{claim*}{Claim}
\newtheorem{claim}[theorem]{Claim}
\theoremstyle{remark}
\newtheorem*{remark*}{Remark}
\newtheorem{remark}[theorem]{Remark}
\newtheorem{example}[theorem]{Example}
\theoremstyle{definition}
\newtheorem*{definition*}{Definition}
\newtheorem{definition}[theorem]{Definition}
\newtheorem*{observation*}{Observation}
\newtheorem{observation}[theorem]{Observation}
%%%
%%% MATH DEFINITIONS
%%%
\newcommand{\abs}[1]{\left|#1\right|} % absolute value
\newcommand{\lnorm}[2]{{\left\|#2\right\|_#1}}% an l_p norm
\newcommand{\norm}[1]{\left\|#1\right\|} % a norm
\newcommand{\R}{\mathbb{R}}
\newcommand{\Z}{\mathbb{Z}}
\newcommand{\N}{\mathbb{N}}
\newcommand{\Q}{\mathbb{Q}}
\newcommand{\F}{\mathbb{F}}
\newcommand{\CC}{\mathcal{C}}
\newcommand{\HH}{\mathcal{H}}
\newcommand{\II}{\mathcal{I}}
\newcommand{\MM}{\mathcal{M}}
\newcommand{\NN}{\mathcal{N}}
\newcommand{\RR}{\mathcal{R}}
\renewcommand{\SS}{\mathcal{S}}
\newcommand{\UU}{\mathcal{U}}
\newcommand{\XX}{\mathcal{X}}
\newcommand{\YY}{\mathcal{Y}}
\newcommand{\ZZ}{\mathcal{Z}}
\newcommand{\rest}[1]{|_{#1}} % restriction to a set
\newcommand{\mb}[1]{\mathbf{#1}}
\newcommand{\set}[2]{\left\{#1, \ldots, #2\right\}}
\newcommand{\intr}{\mathop{\rm int}\nolimits} %%interior of a set
\newcommand{\p}{\mathbf{p}}
\newcommand{\bv}{\mathbf{v}}
\newcommand{\fV}{\mathfrak{V}}
\newcommand{\cL}{\mathcal{L}}
\newcommand{\hyper}[1]{H_{#1}}
\newcommand{\rigid}[1]{R_{#1}}
\newcommand{\cfactor}[1]{C_{#1}}
\DeclareMathOperator{\susp}{Susp}
\DeclareMathOperator{\diam}{diam}
\DeclareMathOperator{\sgn}{sgn}
\DeclareMathOperator{\interior}{int}
\DeclareMathOperator{\supp}{supp}
\DeclareMathOperator{\crno}{cr}
\DeclareMathOperator{\sd}{sd}
\DeclareMathOperator{\lk}{lk}
\DeclareMathOperator{\conv}{conv}
\DeclareMathOperator{\SPAN}{span}
\DeclareMathOperator{\Part}{part}
\DeclareMathOperator{\wsat}{wsat}
\DeclareMathOperator{\mwsat}{w}
%\DeclareMathOperator{\sign}{sign}
\DeclareMathOperator{\Ex}{Ex}
\DeclareMathOperator{\cone}{cone}
\DeclareMathOperator{\rank}{rank}
\DeclareMathOperator{\rk}{rk}
\DeclareMathOperator{\shift}{shift}
\DeclareMathOperator{\val}{val}
\DeclareMathOperator{\cyc}{cyc}
\DeclareMathOperator{\evec}{\bf e}
\DeclareMathOperator{\vvec}{\bf v}
\newcommand{\ellb}{\neg{\ell}}
% Geometric entities for the gluing
\DeclareMathOperator{\andd}{and}
\newcommand{\vand}{v_{\andd}}
\newcommand{\fand}{f_{\andd}}
\newcommand{\fl}[1]{f_{#1}}
\newcommand{\flb}[1]{f_{\neg{#1}}}
\newcommand{\llb}{{\ell,\neg{\ell}}}
\newcommand{\fllb}[1]{f_{#1,\neg{#1}}}
\newcommand{\pllb}[1]{p_{#1,\neg{#1}}}
\newcommand{\blocker}[1]{B_{#1,\neg{#1}}}
\newcommand{\bledge}[1]{b_{#1,\neg{#1}}}
\newcommand{\vllb}[1]{v_{#1,\neg{#1}}}
\newcommand{\Sllb}[1]{S_{#1,\neg{#1}}}
\newcommand{\sllb}[1]{s_{#1,\neg{#1}}}
\newcommand{\flc}[2]{f_{#1,#2}} % literal first, clause second
\newcommand{\plc}[2]{p_{#1,#2}} % literal first, clause second
\newcommand{\cl}[1]{\overline{#1}}
% Gadgets
\newcommand{\gand}{{\bf A}}
\newcommand{\gcla}{{\bf C}}
\newcommand{\gvar}{{\bf V}}
\newcommand{\gadg}{{\bf G}}
\newcommand{\gcho}{{\bf H}}
\newcommand{\by}{{\bf y}}
\newcommand{\bx}{{\bf x}}
\newcommand{\cla}{{\bf C}}
\newcommand{\var}{{\bf V}}
\newcommand{\cho}{{\bf CH}}
\newcommand{\bloc}{{\bf B}}
%\newcommand{\bb}[1]{{\mathbf{#1}}}
\newcommand{\avec}{{\mathbf a}}
\newcommand{\bvec}{{\mathbf b}}
\newcommand{\rvec}{{\mathbf r}}
\newcommand{\kvec}{{\mathbf k}}
\newcommand{\nvec}{{\mathbf n}}
\newcommand{\unitvec}{{\mathbf 1}}
\newcommand{\svec}{{\mathbf s}}
\newcommand{\lvec}{\boldsymbol{\ell}}
\newcommand{\tvec}{{\mathbf t}}
\newcommand{\Dvec}{{\mathbf D}}
%\newcommand{\lip}{{\vdash}}
\newcommand{\lip}{{\llcorner}}
\newcommand{\bbeta}{\boldsymbol{\beta}}
\newcommand{\eepsilon}{\boldsymbol{\varepsilon}}
\newcommand{\symdif}{{\bigtriangleup}}
\newcommand{\crossout}[1]{{\color{red}\sout{#1}}}
\newcommand{\red}[1]{\begingroup\color{red}#1\endgroup}
\newcommand{\blue}[1]{\begingroup\color{blue}#1\endgroup}
\newcommand{\magenta}[1]{\begingroup\color{magenta}#1\endgroup}
\newcommand{\emptyfrac}[2]{\genfrac{}{}{0pt}{}{#1}{#2}}
\title{Maximal matroids and counterexamples\thanks{D. B. is supported by AARMS postdoctoral fellowship. M. T. is supported by the
GA\v{C}R grant 25-16847S.}}
\author[1]{Denys Bulavka}
\author[2]{Martin Tancer}
\affil[1]{\small Department of Mathematics and Statistics, Dalhousie University, Halifax, Nova Scotia, Canada.}
\affil[2]{\small Department of Applied Mathematics, Faculty of Mathematics and Physics,
Charles~University, Prague, Czech Republic.}
\date{}
\begin{document}
\maketitle{}
\begin{abstract}
Jackson and Tanigawa conjectured that the rigidity matroid $\RR^d_n$ and the hyperconnectivity matroid $\HH^d_n$ are the unique maximal
matroids in the posets of $\{K_{d+2}, K_{d+2,d+2}\}$-matroids and $\{K_{d+2},K_{d+1,d+1}\}$-matroids, respectively. We disprove these conjectures
by showing the existence of maximal matroids that are distinct from the proposed candidates.
\end{abstract}
\section{Introduction}
In a recent paper~\cite{jackson24-maximal}, Jackson and Tanigawa initiated a
systematic study of maximal matroids with prescribed sets of circuits.
They conjectured that certain well known matroids may be described as unique
maximal matroids among those with a particularly simple
prescribed set of circuits.
When trying to prove some of these conjectures, we actually arrived
at counterexamples, which is the content of this article.
\paragraph{An upper bound on the rank function of a matroid.}
Following~\cite{jackson24-maximal}, let $\XX$ be a family of subsets from a
finite set $E$. An \emph{$\XX$-matroid} on $E$ is a matroid whose ground set is $E$ and each
set in $\XX$ is a circuit. We consider matroids on $E$ ordered with respect to
inclusion. (Here and throughout the article we identify a matroid $M$ with the collection of its
independent sets.) Given $\XX$ as above, a \emph{proper $\XX$-sequence} is a sequence
$\SS = (X_1, \dots, X_k)$ of sets in $\XX$ such that, for every $i \in \{2,\dots, k\}$, $X_i$ is not a subset of
$X_1 \cup \cdots \cup X_{i-1}$. For $F
\subseteq E$, set $\val(F, \SS) := |F \cup X_1 \cup \cdots \cup X_k| - k$. Then
$\val_\XX \colon 2^E \to \Z$ is defined as
\[
\val_\XX(F) = \min \{ \val(F, \SS)\colon \SS \hbox{ is a proper
$\XX$-sequence}\}.
\]
For any proper $\XX$-sequence $\SS$, the value $\val(F,\SS)$ is an upper bound on the rank of $F$ in every
$\XX$-matroid:
\begin{lemma}[{\cite[Part~of~Lemma 3.3]{clinch22-rigidity2}}]
\label{l:lower-bound}
Let $M$ be an $\XX$-matroid on $E$ and $F\subseteq E$. Then $\rk_M(F)\leq
\val(F,\SS)$ for any proper $\XX$-sequence $\SS$. In particular, this implies that
$\rk_M(F) \leq \val_\XX(F)$.
\end{lemma}
Jackson and Tanigawa~\cite[Lemma~1.2]{jackson24-maximal} further showed that if
$\val_\XX$ is submodular and there exists some $\XX$-matroid, then there is a
unique maximal $\XX$-matroid with its rank function coinciding with $\val_\XX$.
The function $\val_\XX$ is interesting because it could describe the rank
function of various well known matroids as explained below.
\paragraph{Circuits in specific abstract rigidity matroids.}
From now on let us assume that $E$ is the collection of edges of the complete graph
$K_n$, for some $n$. Given a collection of graphs $G_1, \dots, G_k$, with a
slight abuse of the notation, by setting $\XX = \{G_1, \dots, G_k\}$ we actually
mean that $\XX$ consists of all subgraphs of $K_n$ isomorphic to one of the
graphs $G_1, \dots, G_k$.
Recently Clinch, Jackson and Tanigawa~\cite{clinch22-rigidity1,clinch22-rigidity2}
showed that the (generic) cofactor matroid\footnote{This matroid is denoted as
$\CC^1_2(K_n)$ in~\cite{clinch22-rigidity1}.} $\CC^3_n$ is the unique maximal $K_5$-matroid and
$\val_{K_5}$ is its rank function. Jackson and Tanigawa~\cite{jackson24-maximal} studied the existence of
maximal $\XX$-matroids for several classes $\XX = \{G_1, \dots,
G_k\}$ as above. They conjectured that various specific abstract rigidity
matroids~\cite{graver91-abstractrig} can be described as maximal $\XX$-matroids for a particularly simple family $\XX$
independent of $n$. In order to state these conjectures we recall that
$\HH^d_n$ stands for the \emph{(generic) $d$-hyper\-connectivity matroid}\footnote{In Kalai's
paper this matroid is denoted $\HH_d^n$ but we prefer to be consistent
with~\cite{jackson24-maximal}.} on $n$ vertices introduced by
Kalai~\cite{kalai85-hyperconnectivity}, $\RR^d_n$ stands for the \emph{(generic) $d$-rigidity
matroid}\footnote{As far as we understand the notation
in~\cite{jackson24-maximal}, this matroid is denoted either $\RR_d(K_n)$ or
$\RR_n^d(K_n)$.} on $n$ vertices~\cite{maxwell64-rigidity,gluck74-rigidity,asimow78-rigidity,asimow79-rigidity},
and $\CC^d_n$ stands for the \emph{(generic) $d$-cofactor matroid}\footnote{This matroid is usually denoted
as $\CC^{d-2}_{d-1}(K_n)$~\cite{whiteley96-geometry,jackson24-maximal}.} on $n$ vertices ~\cite{whiteley88-cofactor,
whiteley91-cofactor, alfeld93-cofactor, whiteley96-geometry}.
\begin{conjecture}{\cite[Conjecture 6.6.a]{jackson24-maximal}}
\label{c:hyperconnectivity}
Let $d\geq 1$ and $\XX=\{K_{d+2},K_{d+1,d+1}\}$. Then, $\HH_n^d$ is the
unique maximal $\XX$-matroid and $\val_\XX$ is its rank function.
\end{conjecture}
We remark that $n$ is not really quantified in~\cite[Conjecture
6.6.a]{jackson24-maximal}; however, it is meaningful to assume $n \geq 2d + 2$,
because then $K_n$ contains a copy of $K_{d+1,d+1}$.
The conjecture is true for $d=1$ since in this case the hyperconnectivity matroid
$\HH_n^1$ coincides with the cycle matroid, which is also a count matroid.
In this case, $\HH_n^1$ is the unique maximal $\{K_3,K_{2,2}\}$-matroid, in particular, because it is the unique maximal
$K_3$-matroid~\cite[Theorem 4.3]{jackson24-maximal}, and $\val_{K_3}$ is its rank function.
Although the conjecture is open for $\HH_n^2$, Bernstein provided a combinatorial characterization of its independent sets~\cite{bernstein17-2hyper}.
Here we disprove Conjecture~\ref{c:hyperconnectivity} for $d\geq 3$:
\begin{theorem}
\label{t:rank-hd}
Let $d\geq 3$, $n\geq 2d+2$ and $\XX = \{K_{d+2}, K_{d+1,d+1}\}$. Then, $\HH_n^d$ is not the unique
maximal $\XX$-matroid. In particular, $\val_\XX$ is not the rank function of $\HH_n^d$.
\end{theorem}
We remark that it is still possible that $\HH_n^d$ could be described as the
unique maximal $\XX$-matroid for some finite set $\XX$ independent of $n$. It
is also possible that there exists a unique maximal $\{K_{d+2},
K_{d+1,d+1}\}$-matroid. Theorem~\ref{t:rank-hd} only implies that if such a
matroid exists, it is not $\HH_n^d$.
\begin{conjecture}{\cite[Conjecture 6.3]{jackson24-maximal}}
\label{c:rigidity}
Let $d\geq 3$ and $\XX=\{K_{d+2},K_{d+2,d+2}\}$. Then, $\RR_n^d$ is
the unique maximal $\XX$-matroid and $\val_\XX$ is the rank function of $\RR_n^d$.
\end{conjecture}
We disprove Conjecture~\ref{c:rigidity} for $d\geq 5$ and $n \geq 2d + 4$ (when $K_n$
contains a copy of $K_{d+2,d+2}$):
\begin{theorem}
\label{t:rank-rd}
Let $d\geq 5$, $n\geq 2d+4$ and $\XX=\{K_{d+2},K_{d+2,d+2}\}$. Then, $\RR^d_n$ is not the
unique maximal $\XX$-matroid. In particular, $\val_\XX$ is not the rank function of $\RR^d_n$.
\end{theorem}
Again, it is still possible that $\RR_n^d$ could be described as the
unique maximal $\XX$-matroid for some finite set $\XX$ independent of $n$ as
well as it is possible that there exists a unique maximal $\{K_{d+2},
K_{d+2,d+2}\}$-matroid.
\paragraph{Outline.}
The remainder of the article is organized as follows. In Section~\ref{s:preliminaries} we review
some known matroidal constructions and the
necessary properties about the hyperconnectivity, rigidity and cofactor matroids.
In Section~\ref{s:xxmatroid} we review the properties of $\XX$-uniform matroids introduced in~\cite{jackson24-maximal}, and study their behavior when intersected with another matroid.
In Section~\ref{s:counter-umax} we present two counterexamples proving Theorem~\ref{t:rank-hd} and Theorem~\ref{t:rank-rd}.
\section{Properties of several abstract rigidity matroids}
\label{s:preliminaries}
\begin{table}
\begin{center}
\begin{tabular}{cccc}
& $\HH_n^d$ & $\RR_n^d$ & $\CC^d_n$\\
$K_{d+2}$ & circuit & circuit & circuit\\
$K_{d+1,d+1}$ & circuit & independent ($d\geq 2$) & independent ($d\geq 2$)\\
$K_{d+2,d+2}$ & dependent & circuit ($d\geq 3$) & independent ($d\geq 4$) \\
\end{tabular}
\end{center}
\caption{
The statements in the table hold for $d\geq 1$, whenever $d$ is not specified.
For $d=1$ all these matroids coincide with the cycle matroid~\cite{whiteley96-geometry}.
The claims on $\HH_n^d$ are proved in~\cite{kalai85-hyperconnectivity}.
The claims regarding $\RR_n^d$ for $d=2,3$ are proved in~\cite[Sections 2,3, and 9]{whiteley96-geometry},
while for $d\geq 4$ are provided in~\cite[Examples 11.1.6, 11.2.4]{whiteley96-geometry}.
The claims regarding $\CC_n^d$ for $d=2,3,4$ are given in~\cite[Sections 5, 10 and 11.3]{whiteley96-geometry},
while for $d\geq 5$ are provided in Lemma~\ref{l:bipartite_plus_edge}.
}
\label{t:circuits}
\end{table}
\paragraph{Hyperconnectivity matroid.}
We skip the technical definition of the hyperconnectivity matroid
from~\cite{kalai85-hyperconnectivity}; however, we point out several important
properties of this matroid, all of them proved
in~\cite{kalai85-hyperconnectivity}. Given two isomorphic graphs $G$ and $H$ on
at most $n$ vertices, $G$ and $H$ have the same rank in $\HH_n^d$,
i.e., $\HH_n^d$ is a \emph{symmetric matroid}~\cite{kalai90-symmetric}.
In particular, $G$ is a circuit in $\HH_n^d$ if and only if $H$ is a
circuit in $\HH_n^d$. In addition, the property of being a circuit does not depend
on $n$: (any isomorphic copy of) $G$ is a circuit in $\HH_n^d$ if any only if (any isomorphic copy of)
$G$ is a circuit in $\HH_m^d$ assuming that $m,n \geq |V(G)|$. It is known
that $K_{d+2}$ and $K_{d+1,d+1}$ are circuits in $\HH_n^d$ (for $n \geq 2d+2$);
see Table~\ref{t:circuits}. In particular, $\HH_n^d$ is a $\{K_{d+2},
K_{d+1,d+1}\}$-matroid.
Given a graph $G$, by $\cone(G)$ we denote the \emph{cone over $G$}
which is obtained from $G$ by adding a (new) vertex and connecting it with all
other vertices of $G$. For $k\geq 2$, by $\cone^k(G)$ we denote the
\emph{$k$-th cone} over $G$ defined iteratively as $\cone^{k-1}(\cone(G))$.
For ease of notation, we set $\cone^0(G)=G$.
The following corollary is a consequence of the fact that coning transforms circuits
of $\HH_n^d$ into circuits of $\HH_n^{d+1}$~\cite[Corollary 5.2]{kalai85-hyperconnectivity}.
\begin{corollary}
\label{c:cone-hbip}
The cone $\cone^k(K_{d+1-k,d+1-k})$ is a circuit in $\HH_n^d$
for every $d\geq 1$ and $k=0,\dots,d$ (where $n$ is at least the size of the circuit).
\end{corollary}
In particular, $\cone(K_{d,d})$ is a circuit in $\HH_n^d$. The core of
the proof of Theorem~\ref{t:rank-hd} is to build another $\{K_{d+2},
K_{d+1,d+1}\}$-matroid where $\cone(K_{d,d})$ is independent.
\paragraph{Rigidity matroid.}
We also skip the technical definition of the rigidity matroid $\RR_n^d$. It
shares many properties with the hyperconnectivity matroid. Two isomorphic
graphs have again the same rank in $\RR_n^d$. Likewise, $G$ is a circuit in
$\RR_n^d$ if any only if (any isomorphic copy of) $G$ is a circuit in $\RR_m^d$
assuming that $m,n \geq |V(G)|$, see~\cite{asimow78-rigidity}.
For $d\geq 3$, the well known circuits in $\RR_n^d$
are $K_{d+2}$ and $K_{d+2,d+2}$ (for $n \geq 2d+4$); see Table~\ref{t:circuits}.
From this we get that $\RR_n^d$ is a $\{K_{d+2}, K_{d+2,d+2}\}$-matroid.
Similarly as for the hyperconnectivity matroid, the following corollary is a consequence of the fact
that coning transforms circuits of $\RR_n^d$ into circuit in $\RR_n^{d+1}$~\cite[Theorem 5]{whiteley83-conerigidity}.
\begin{corollary}
\label{c:cone-rbip}
The cone $\cone^k(K_{d+2-k,d+2-k})$ is a circuit in $\RR_n^d$
for every $d\geq 3$ and $k=0,\dots,d-3$ (where $n$ is at least the size of the circuit).
\end{corollary}
In particular, $\cone(K_{d+1,d+1})$ is a circuit in $\RR_n^d$. Similarly as
in the case of the hyperconnectivity matroid, the core of
the proof of Theorem~\ref{t:rank-rd} is to build another $\{K_{d+2},
K_{d+2,d+2}\}$-matroid where $\cone(K_{d+1,d+1})$ is independent.
\paragraph{Cofactor matroid.}
We also skip the technical definition of the cofactor matroid $\CC^d_n$.
It shares many of the properties with the hyperconnectivity and rigidity matroids since
all of these are abstract rigidity matroids, e.g., the ranks of $\HH_n^d$, $\RR_n^d$
and $\CC_n^d$ are all equal to $dn - \binom{d+1}{2}$ and all have $K_{d+2}$ as a circuit.
It is also well behaved with respect to the cone operation:
\begin{lemma}[{\cite[Theorem 11.3.3]{whiteley96-geometry}}]
\label{l:cofactor-cone}
Let $G$ be a graph, then $G$ is independent in $\CC^d_n$ if and only if
$cone(G)$ is independent in $\CC^{d+1}_n$ (provided that $n \geq |V(G)|+1$).
\end{lemma}
The matroids $\HH^d_n, \RR^d_n$ and $\CC^d_n$ in general differ if we look at their
circuits.
For example, it is known that the complete bipartite graph $K_{d+1,\binom{d+1}{2}}$ is dependent in
$\HH_n^d$ when $d\geq 2$, see Table~\ref{t:circuits}. However, this graph is independent in
$\CC_n^d$~\cite[Example 11.3.12]{whiteley96-geometry} for $d\geq 1$.
\begin{lemma}
\label{l:bipartite_plus_edge}
\begin{enumerate}
\item Let $d \geq 3, n \geq 2d+2$ and let $H$ be a subgraph of $K_n$
obtained from $K_{d+1,d+1}$ by adding an edge (of $K_n$) to it. Then $H$ is independent in $\CC^d_n$.
\item Let $d \geq 4, n \geq 2d+4$ and let $H$ be a subgraph of $K_n$
obtained from $K_{d+2,d+2}$ by adding an edge (of $K_n$) to it. Then $H$ is independent in $\CC^d_n$.
\end{enumerate}
\end{lemma}
We will need the first item of the lemma in the proof of
Theorem~\ref{t:rank-hd} in order to build a suitable $\{K_{d+2},K_{d+1,d+1}\}$-matroid.
Similarly, the second item will be used in the proof of Theorem~\ref{t:rank-rd}
in order to build a suitable $\{K_{d+2},K_{d+2,d+2}\}$-matroid.
For the proof of Lemma~\ref{l:bipartite_plus_edge} we need the definition of a $(k,d)$-extension of a graph.
For it, let us consider a pair of integers $0\leq k \leq d$, a graph $G$, a $k$-matching $\evec = \{e_1,\dots, e_k\}$ in $G$, i.e., a matching of size $k$,
and a set of vertices $V =\{v_1,\dots, v_{d-k}\}$ in $G$ disjoint from all
the edges in $\evec$.
The \emph{$(k,d)$-extension of $G$ (with respect to $\evec$ and $V$)} is a new graph $G'$ obtained from $G$
by adding to it a new vertex $u$, connecting $u$ to the endpoints of the edges appearing in the matching $\evec$ as well as
to the vertices in $V$, and finally removing from it the edges appearing in $\evec$.
It is known that the independent sets in $\CC_n^d$ are preserved after performing
$(0,d),(1,d)$ and $(2,d)$-extensions~\cite[Lemma 11.3.6, Theorem 11.3.10,
Theorem 11.4.1]{whiteley96-geometry}.
\begin{proof}
In both cases, let us assume that $H$ is obtained from the complete graph with
parts $\{a_1, \dots, a_k\}$ and $\{b_1, \dots b_k\}$ where $k \in
\{d+1,d+2\}$ by adding an edge $e$ to it. If $e$ meets the complete graph,
without loss of generality, we can assume that it meets it in $a_1$ (and
possibly $a_2$).
We start with the complete graph on $\{a_1,\dots,a_d, b_1,b_2\}$ minus the
edge $\{a_{d-1},a_{d}\}$. This graph, call it $G_1$, is independent as the complete graph
$K_{d+2}$ is a circuit in $\CC^d_n$
We perform $(0,d)$-extensions in $G_1$ at vertices $b_3, \dots, b_d$
connecting them to $a_1, \dots, a_d$, obtaining a graph $G_2$, independent in
$\CC^d_n$. Next we perform one more $(0,d)$-extension at $a_{d+1}$,
connecting it to $b_1,\dots, b_d$, obtaining a graph $G_3$, still independent
in $\CC^d_n$.
Now we perform a $(1,d)$-extension in $G_3$ at the vertex $b_{d+1}$, connecting it
to $a_1, \dots, a_{d+2}$ at the cost of removing the edge $a_1a_3$ from
$G_3$. This way we obtain a graph $G_4$ independent in $\CC^d_n$.
If we are in the case of the first item of the lemma, we are essentially done.
If $e$ is the edge $a_1a_2$, then $H$ is a subgraph of $G_4$. If $e$ is not
the edge $a_1a_2$, then we can perform further one or two $(0,d)$-extensions
in vertices of $H$ and we obtain a supergraph of $H$ independent in
$\CC^d_n$.
It remains to consider the second item of the lemma, assuming $d \geq 4$. We
perform one more $(1,d)$-extension at the vertex $a_{d+2}$, connecting it to
$b_1, \dots, b_{d+1}$ while removing the edge $b_1b_2$, obtaining a graph
$G_5$. Next, we perform a $(2,d)$-extension at the vertex $b_{d+2}$
connecting it to $a_1, \dots, a_{d+2}$ while removing the edges $a_1a_4$ and
$a_2a_3$, obtaining $G_6$ independent in $\CC^d_n$.
Now we finish as in the previous case: If $e$ is the edge $a_1a_2$, then $H$
is a subgraph of $G_6$. If $e$ is not
the edge $a_1a_2$, then we can perform further one or two $(0,d)$-extensions
in vertices of $H$ and we obtain a supergraph of $H$ independent in
$\CC^d_n$.
\end{proof}
\begin{remark}
As the reader might have noticed, if $d$ is large, the subgraph of $G_5$ (or
$G_6$) induced by vertices $a_1, \dots, a_d$ still contains many edges (and
pairs of disjoint edges). Let $G$ be $K_{d+2, d+ \binom d2/2}$ if $\binom d2$ is
even and $G$ be $K_{d+2, d+ \lceil\binom d2/2\rceil}$ if $\binom d2$ is odd.
By a suitable strategy of applying
$(2,d)$-extensions, starting with $G_5$, possibly followed by a single
$(1,d)$-extension, it is possible to achieve that the graph $G$ is
independent in $\CC^d_n$ for $d \geq 4$. This generalizes the well known fact that $K_{6,7}$ is
independent in $\CC_n^4$~\cite[Theorem 11.4.2]{whiteley96-geometry}.
If, in addition, $n = 2d+2+
\lceil\binom d2/2\rceil$, then $G$ is a basis (by comparing the number of
edges of $G$ with the rank of $\CC^d_n$).
\end{remark}
\section{Intersecting a matroid with an $\XX$-uniform matroid}
\label{s:xxmatroid}
\paragraph{Matroid intersection.}
Let us denote by $\CC(M)$ the circuits of a matroid $M$.
We observe that if an intersection $N=M_1\cap M_2$ is a matroid,
then we can express its circuits and the rank function in terms
of $M_1$ and $M_2$ as follows.
\begin{lemma}
\label{l:intersection-crank}
Let $M_1,M_2$ be matroids on the same ground set such that $N=M_1\cap M_2$ is a matroid as well. Then, $$\CC(N)= (\CC(M_1)\cap M_2) \cup (M_1\cap \CC(M_2))\cup (\CC(M_1)\cap \CC(M_2))$$ and $$r_N(X) = \min \{r_1(Y)+r_2(X\setminus Y)\colon Y\subseteq X\}.$$
\end{lemma}
We remark that we do not really need a formula for the rank function in
the lemma above. We add it for completeness.
\begin{proof}
Let $C\in \CC(N)$. Then, since $C\setminus x\in N$ for every $x\in C$, we
have that $C$ is either a circuit or independent in $M_1$ and $M_2$.
Vice versa, if $C$ is either a circuit or independent in $M_1$ and
$M_2$, and, in addition, $C \not\in N$, then $C \in \CC(N)$.
The given decomposition covers all the possible combinations.
Now, let $X\subseteq E$ then its rank in $N$ is given by $$r_N(X) = \max \{|I| \colon I\in M_1|_X \cap M_2|_X\} = \min\{r_1(Y)+r_2(X\setminus Y) \colon Y\subseteq X\}$$ where the second equality follow from~\cite[Corollary 11.3.16]{oxley11-matroid}.
\end{proof}
\begin{corollary}
Fix a ground set $E$ and $\XX_1,\XX_2$ finite families of subsets of $E$.
Let $M_1$ an $\XX_1$-matroid, and $M_2$ an $\XX_2$-matroid such that $N=M_1\cap M_2$ is a matroid.
Then, $\XX_i\subseteq M_j\cup \CC(M_j)$ for $i\neq j$ if and only if $N$ is an $(\XX_1\cup \XX_2)$-matroid.
Moreover, if for $i=1,2$ the matroid $M_i$ is the unique maximal $\XX_i$-matroid for $i=1,2$,
then $N$ is the unique maximal $(\XX_1\cup \XX_2)$-matroid.
\end{corollary}
\begin{proof}
The matroid $N$ is an $(\XX_1\cup \XX_2)$-matroid if and only if $\XX_1\cup \XX_2 \subseteq \CC(N)$.
By Lemma~\ref{l:intersection-crank} this last condition holds if and only if
$$\XX_i\subseteq (\CC(M_1)\cap M_2) \cup (M_1\cap \CC(M_2))\cup (\CC(M_1)\cap \CC(M_2)).$$
Since $\XX_i\cap M_i =\emptyset$ and $\XX_i \subseteq \CC(M_i)$, we have that $\XX_1\cup \XX_2 \subseteq \CC(N)$ holds if and only if
$\XX_i \subseteq M_j \cup \CC(M_j)$, as wanted.
Now, assume that $M_i$ is the unique maximal $\XX_i$ matroid for $i =
1,2$. Let $N'$ be an $(\XX_1\cup \XX_2)$-matroid. In particular, $N'$ is a $\XX_i$-matroid
and consequently, by uniqueness and maximality of $M_i$, we have that $N'\subseteq M_i$ for $i=1,2$.
From this it follows that $N'\subseteq M_1\cap M_2 = N$.
\end{proof}
The intersection of two matroid as defined in this section is not always matroid.
Nash-Williams introduced the following closely related construction.
For two matroids $M_1,M_2$ on $E$ the family $M_1\vee M_2 = \{I_1\cup I_2 \colon I_1\in M_1, I_2\in M_2\}$
is again a matroid, see~\cite[Theorem 11.3.1]{oxley11-matroid}.
By matroid duality we obtain that $M_1\wedge M_2 = (M_1^*\vee M_2^*)^*$ is a matroid as well.
We observe that the inclusion $M_1\wedge M_2 \subseteq M_1\cap M_2$ holds in general.
Indeed, let $B$ be a basis of $M_1\wedge M_2$, then $B=E\setminus B^*$ with $B^*$ a basis of $M_1^*\vee M_2^*$.
Since $B^*$ is independent in $M_1^*\vee M_2^*$ then $B^*=I_1\cup I_2$ for $I_j\in M_j^*$.
For $j=1,2$, let $I_j\subseteq B_j^*$ be a basis in $M_j^*$. Then, $B^*\subseteq B^*_1\cup B^*_2 \subseteq B^*$ where the last inclusion
follows from the fact that $B^*$ is a maximal independent set in $M_1^*\vee M_2^*$.
Since $B_j^* = E\setminus B_j$ with $B_j$ a basis in $M_j$ we have that $B = B_1\cap B_2$.
However, the reverse inclusion $M_1\wedge M_2 \supseteq M_1\cap M_2$ does not need to be satisfied in general.
For it, consider $\UU_2$ the uniform matroid of rank $2$. We have that $\UU_2\cap \UU_2 = \UU_2$, while $\UU_2\wedge \UU_2 = \{\emptyset\}$ is the matroid whose only independent set is the empty set.
\paragraph{Intersection of $\XX$-uniform matroids.}
Jackson and Tanigawa~\cite{jackson24-maximal} introduced the notion of a $\XX$-uniform matroid.
For it, let $\XX$ be a finite uniform family of (finite) sets, i.e., all of its members have the same size $k\geq 1$,
and set $E = \bigcup_{X \in \XX} X$.
We define the $\XX$-uniform system as
\[
\UU_\XX = \{ F \subseteq E \colon |F| \leq k \hbox{ and there is no } X \in
\XX \hbox{ with } X \subseteq F\}.
\]
We say that the family $\XX$ as above is \emph{union-stable} if for every $X_1,
X_2 \in \XX$ with $X_1 \neq X_2$ and $e \in X_1 \cap X_2$ either $|(X_1 \cup X_2) - e| > k$ or
there exist $X\in \XX$ such that $X \subseteq (X_1\cup X_2) - e$.\footnote{Thinking of $\XX$ as a set of circuits, these
conditions are essentially the circuit exchange axiom.
}
\begin{lemma}[{\cite[Lemma 2.5]{jackson24-maximal}}]
\label{l:union-stable}
Let $\XX$ be a uniform family of sets. Then, $\XX$ is union-stable if and only if $\UU_\XX$ is a matroid.
\end{lemma}
\begin{example}
\label{ex:bipartite}
Let $d\geq 2$, $n\geq 2d$, and $\XX=\{K_{d,d}\}$ be given by all the isomorphic copies of $K_{d,d}$ in $K_n$.
We claim that $\UU_{K_{d,d}} \coloneqq \UU_\XX$ is a matroid. For it, we verify that the family $\XX$ is union-stable.
Indeed, any two distinct element $X_1,X_2\in \XX$ either share all the vertices or not. In the former
case there are at least two new edges while in the latter the number of edges increases at least by $d$.
In both cases $|(X_1\cup X_2) \setminus e| > d^2$ for every $e\in X_1\cap X_2$.
\end{example}
Let $M$ be a matroid with the ground set $E$, a family $\XX$ of subsets of $E$
is \emph{compatible with} $M$ if for every $X\in \XX$, $e\in X$ and $f\in E$,
the set $(X\setminus e)\cup f \in M$.
\begin{lemma}
\label{l:intersection}
Let $M$ be a matroid with ground set $E$, and $\XX$
a uniform family of subsets of $E$, all of whose elements have the same size $k\geq 1$, such that $\UU_\XX$ is a matroid.
If $\XX$ is
compatible with $M$, then $N=M\cap \UU_\XX$ is a matroid.
\end{lemma}
\begin{proof}
We proceed by verifying the independent set axioms for $N$.
The family of sets $N$ is not empty since both $M$ and $\UU_\XX$ contain the emptyset.
Let $I_1,I_2\in N$ with $|I_1|< |I_2|$.
Since $M$ is a matroid there exists $e\in I_2\setminus I_1$ such that
$I_1\cup e\in M$. If $I_1\cup e\in \UU_\XX$ we are done.
Assume that this is not the case, then by definition of $\UU_\XX$ either $|I_1\cup e| > k$
or there exists $X\in \XX$ such that $X\subseteq I_1\cup e$.
The former option is not possible since $|I_1\cup e| \leq |I_2| \leq k$.
In the latter case we obtain that $|X|\leq |I_1\cup e| \leq |I_2|\leq |X|$ and consequently $X=I_1\cup e$,
i.e., $I_1 = X\setminus e$.
Because $\UU_\XX$ is a matroid, there exists $f\in I_2\setminus I_1$
such that $I_1\cup f \in \UU_\XX$. We notice that $f\neq e$ since we assumed
that $I_1 \cup e \notin \UU_\XX$.
Since $\XX$ is compatible with $M$ we can conclude that $(X\setminus e) \cup f = I_1\cup f$ is independent in $M$,
and so it is independent in $N$ as well.
\end{proof}
\begin{example}
\label{ex:bip-comp-small}
For $d\geq 3$ and $n\geq 2d+2$, let $\XX = \{K_{d+1,d+1}\}$ denote the family of all the
isomorphic copies of $K_{d+1,d+1}$ in $K_n$. We claim that $\CC_n^d \cap \UU_{K_{d+1,d+1}}$ is
a matroid. From Example~\ref{ex:bipartite} we already know that $\UU_{K_{d+1,d+1}}$ is a matroid.
By Lemma~\ref{l:bipartite_plus_edge}, $\XX$ is compatible with
$\CC_n^d$. Consequently, Lemma~\ref{l:intersection} implies that $\CC_n^d \cap \UU_{K_{d+1,d+1}}$ is
a matroid.
\end{example}
\begin{example}
\label{ex:bip-comp}
For $d\geq 4$ and $n\geq 2d+4$, let $\XX = \{K_{d+2,d+2}\}$ denote the family of all the
isomorphic copies of $K_{d+2,d+2}$ in $K_n$. We claim that $\CC_n^d\cap \UU_{K_{d+2,d+2}}$ is a matroid.
Similarly as above, it follow from Example~\ref{ex:bipartite} that $\UU_{K_{d+2,d+2}}$ is a matroid.
By Lemma~\ref{l:bipartite_plus_edge}, $\XX$ is compatible with
$\CC_n^d$. Consequently, Lemma~\ref{l:intersection} implies that $\CC_n^d \cap \UU_{K_{d+2,d+2}}$ is
a matroid.
\end{example}
\section{Counterexamples to the unique maximal candidates}
\label{s:counter-umax}
\begin{proof}[Proof of Theorem~\ref{t:rank-hd}]
Let $d\geq 3$ and $\XX=\{K_{d+2},K_{d+1,d+1}\}$.
We show that $\HH_n^d$ is not the unique $\XX$-maximal matroid.
For it, let $N = \CC_n^d\cap \UU_{K_{d+1,d+1}}$. It follows from Example~\ref{ex:bip-comp-small} that $N$ is a matroid.\\
{\bf Claim: $N$ is an $\XX$-matroid.}
By Lemma~\ref{l:intersection-crank} it follows that $N$ is an $\XX$-matroid since $K_{d+2}\in \CC(\CC^d_n)\cap \UU_{K_{d+1,d+1}}$ and
$K_{d+1,d+1}\in \CC_n^d\cap \CC(\UU_{K_{d+1,d+1}})$.\\
{\bf Claim: there is a maximal $\XX$-matroid containing $N$.}
The poset of all $\XX$-matroids containing $N$ is not empty since
$N$ is in it. Set $N'$ to be a maximal element in this poset.\\
{\bf Claim: $N'$ is different from $\HH_n^d$.}
Now, for $d\geq 3$ the graph $K_{d,d}$ is independent in $\CC_n^{d-1}$~\cite[Example 11.3.12]{whiteley96-geometry}.
By Lemma~\ref{l:cofactor-cone} we can conclude that $G = \cone(K_{d,d})$ is
independent in $\CC_n^d$.
By cardinality, $G$ is independent in $\UU_{K_{d+1,d+1}}$, and so it is independent in $N \subseteq N'$.
However, by Corollary~\ref{c:cone-hbip}, the graph $G$ is dependent in $\HH_n^d$ and
consequently $N'\neq \HH_n^d$. It follows that $\HH_n^d$ is not
the unique maximal $\XX$-matroid.
Finally, by~\cite[Lemma~1.2]{jackson24-maximal}, the function $\val_\XX$ is not the rank function $\HH_n^d$ since
otherwise $\HH_n^d$ would be the unique maximal $\XX$-matroid.
\end{proof}
\begin{proof}[Proof of Theorem~\ref{t:rank-rd}]
Let $d\geq 5$ and $\XX = \{K_{d+2},K_{d+2,d+2}\}$.
We show that $\RR_n^d$ is not the unique $\XX$-maximal matroid.
For it, let $N = \CC_n^d\cap \UU_{K_{d+2,d+2}}$. It follows from Example~\ref{ex:bip-comp} that $N$ is a matroid.\\
{\bf Claim: $N$ is an $\XX$-matroid.}
By Lemma~\ref{l:intersection-crank} it follows that $N$ is an $\XX$-matroid
since $K_{d+2}\in \CC(\CC_n^d)\cap \UU_{K_{d+2,d+2}}$ and, by Lemma~\ref{l:bipartite_plus_edge}, $K_{d+2,d+2}\in \CC_n^d\cap \CC(\UU_{K_{d+2,d+2}})$.\\
{\bf Claim: there is a maximal $\XX$-matroid containing $N$.}
The poset of all $\XX$-matroids containing $N$ is not empty since
$N$ is in it. Set $N'$ to be a maximal element in this poset.\\
{\bf Claim: $N'$ is different from $\RR_n^d$.}
Now, for $d\geq 5$ the graph $K_{d+1,d+1}$ is independent in $\CC_n^{d-1}$, see Table~\ref{t:circuits}.
By Lemma~\ref{l:cofactor-cone} it follows that $G=\cone(K_{d+1,d+1})$ is independent in $\CC_n^d$.
By cardinality, $G$ is independent in $\UU_{K_{d+2,d+2}}$, and so it is independent in $N \subseteq N'$.
However, by Corollary~\ref{c:cone-rbip} the graph $G$ is a circuit in $\RR_n^d$ and consequently $N'\neq \RR_n^d$.
It follows that $\RR_n^d$ is not the unique maximal $\XX$-matroid.
Finally, by~\cite[Lemma~1.2]{jackson24-maximal}, the function $\val_\XX$ is not the rank function $\RR_n^d$ since
otherwise $\RR_n^d$ would be the unique maximal $\XX$-matroid.
\end{proof}
\section{Questions}
\paragraph{Circuits in the hyperconnectivity matroid.}
We disproved Conjecture~\ref{c:hyperconnectivity} due to the fact
that $\cone(K_{d,d})$ is a circuit in $\HH_n^d$ while we found a $\{K_{d+2},
K_{d+1,d+1}\}$-matroid where $\cone(K_{d,d})$ is independent. This suggests that
the list of prescribed circuits in $\XX$ order to get $\HH_n^d$ as a unique
maximal $\XX$-matroid is incomplete. Maybe, it is possible to fix
Conjecture~\ref{c:hyperconnectivity} by adding circuits from
Corollary~\ref{c:cone-hbip}:
\begin{question}
Let $d\geq 1$ and $\XX=\{\cone^k(K_{d+1-k,d+1-k})\colon k
\in\{0,\dots,d\}\}$. Is it true that $\HH_n^d$ is the
unique maximal $\XX$-matroid and $\val_\XX$ is its rank function?
\end{question}
We remark that $\XX$ above contains, in particular, $K_{d+2}$ and $K_{d+1,d+1}$
by setting $k$ to $d$ and $0$ respectively. We also remark that, in principle,
it might be even possible to remove some elements from $\XX$ above in order to
answer the question affirmatively. For example, for $d=1$, we get $\XX = \{K_3,
K_{2,2}\}$. However, $\HH_1^n$ is already a unique maximal $K_3$-matroid (as we
mention in the introduction). In general, it may, for example, turn out that it
is not necessary to include $\cone^{d-1}(K_{2,2})$. However, we include all cones in $\XX$ for a convenient formulation.
\paragraph{Rank of $\{K_{d+2},K_{d+1,d+1}\}$-matroids.}
In the proof of Theorem~\ref{t:rank-hd}, we show that $N = \CC^d_n \cap
\UU_{K_{d+1,d+1}}$ is a $\{K_{d+2},K_{d+1,d+1}\}$-matroid different from
$\HH_n^d$.
\begin{question}
\label{q:extend_hyper}
Is $\CC^d_n \cap \UU_{K_{d+1,d+1}}$ contained in a $\{K_{d+2},K_{d+1,d+1}\}$-matroid of rank $dn - \binom{d+1}2$?
\end{question}
In order to motivate the question above, let us remark that both, the
affirmative or the negative answer could have interesting consequences.
If the answer is affirmative, then this could yield a construction of a new
abstract $d$-rigidity family, as discussed, for example, in~\cite{tyomkyn26}.
If the answer is negative, then we get an even stronger counterexample to
Conjecture~\ref{c:hyperconnectivity}: There is no unique maximal
$\{K_{d+2},K_{d+1,d+1}\}$-matroid. Indeed, let us consider a maximal
$\{K_{d+2},K_{d+1,d+1}\}$-matroid $N'$ containing $N$ and a maximal
$\{K_{d+2},K_{d+1,d+1}\}$-matroid $H'$ containing $\HH^d_n$. Then $N'$ and $H'$
differ as the rank of $N'$ is less than $dn - \binom{d+1}2$ while the rank of
$H'$ is at least the rank of $H$ which is $dn -
\binom{d+1}2$; see~\cite{kalai85-hyperconnectivity}.
In addition, if the answer is negative, there is a hope to save the maximality
part of Conjecture~\ref{c:hyperconnectivity} by adding a restriction on the
rank:
\begin{question}[\cite{tyomkyn26pc}]
Let $d\geq 1$ and $\XX=\{K_{d+2},K_{d+1,d+1}\}$. Is it true that $\HH_n^d$ is
the unique maximal $\XX$-matroid among $\XX$-matroids of rank $dn -
\binom{d+1}2$?
\end{question}
\paragraph{Analogous questions for the rigidity matroid.} For all the questions
above, there are analogous questions for the rigidity matroid. Given that the
motivation is essentially the same, we only pose the questions. (In the case of
the first question, we have to be a bit careful regarding the list of circuits
we have.)
\begin{question}
Let $d\geq 1$ and $\XX=\{\cone^k(K_{d+2-k,d+2-k})\colon k
\in\{0,\dots,d-3\}\} \cup \{K_{d+2}\}$. Is it true that $\RR_n^d$ is the
unique maximal $\XX$-matroid and $\val_\XX$ is its rank function?
\end{question}
\begin{question}
Let $d \geq 4$. Is $\CC^d_n \cap \UU_{K_{d+2,d+2}}$ contained in a $\{K_{d+2},K_{d+2,d+2}\}$-matroid of rank $dn - \binom{d+1}2$?
\end{question}
\begin{question}[\cite{tyomkyn26pc}]
Let $d\geq 3$ and $\XX=\{K_{d+2},K_{d+2,d+2}\}$. Is it true that $\RR_n^d$ is
the unique maximal $\XX$-matroid among $\XX$-matroids of rank $dn -
\binom{d+1}2$?
\end{question}
\section*{Acknowledgment}
We would like to thank Mykhaylo Tyomkyn for a fruitful discussion regarding
open questions.
\bibliographystyle{alpha}
\bibliography{main}
\end{document}
\begin{thebibliography}{CJT22b}
\bibitem[AR78]{asimow78-rigidity}
L.~Asimow and B.~Roth.
\newblock The rigidity of graphs.
\newblock {\em Trans. Amer. Math. Soc.}, 245:279--289, 1978.
\bibitem[AR79]{asimow79-rigidity}
L.~Asimow and B.~Roth.
\newblock The rigidity of graphs. {II}.
\newblock {\em J. Math. Anal. Appl.}, 68(1):171--190, 1979.
\bibitem[ASW93]{alfeld93-cofactor}
P.~Alfeld, L.~L. Schumaker, and W.~Whiteley.
\newblock The generic dimension of the space of {$C^1$} splines of degree
{$d\geq 8$} on tetrahedral decompositions.
\newblock {\em SIAM J. Numer. Anal.}, 30(3):889--920, 1993.
\bibitem[Ber17]{bernstein17-2hyper}
D.~I. Bernstein.
\newblock Completion of tree metrics and rank 2 matrices.
\newblock {\em Linear Algebra Appl.}, 533:1--13, 2017.
\bibitem[CJT22a]{clinch22-rigidity1}
K.~Clinch, B.~Jackson, and S.~Tanigawa.
\newblock Abstract 3-rigidity and bivariate {$C^1_2$}-splines {I}: {W}hiteley's
maximality conjecture.
\newblock {\em Discrete Anal.}, pages Paper No. 2, 50, 2022.
\bibitem[CJT22b]{clinch22-rigidity2}
K.~Clinch, B.~Jackson, and S.~Tanigawa.
\newblock Abstract 3-rigidity and bivariate {$C^1_2$}-splines {II}:
{C}ombinatorial characterization.
\newblock {\em Discrete Anal.}, pages Paper No. 3, 32, 2022.
\bibitem[Glu75]{gluck74-rigidity}
H.~Gluck.
\newblock Almost all simply connected closed surfaces are rigid.
\newblock In {\em Geometric topology ({P}roc. {C}onf., {P}ark {C}ity, {U}tah,
1974)}, Lecture Notes in Math., Vol. 438, pages 225--239. Springer,
Berlin-New York, 1975.
\bibitem[Gra91]{graver91-abstractrig}
J.~E. Graver.
\newblock Rigidity matroids.
\newblock {\em SIAM J. Discrete Math.}, 4(3):355--368, 1991.
\bibitem[JT24]{jackson24-maximal}
B.~Jackson and S.~Tanigawa.
\newblock Maximal matroids in weak order posets.
\newblock {\em J. Combin. Theory Ser. B}, 165:20--46, 2024.
\bibitem[Kal85]{kalai85-hyperconnectivity}
G.~Kalai.
\newblock Hyperconnectivity of graphs.
\newblock {\em Graphs Combin.}, 1(1):65--79, 1985.
\bibitem[Kal90]{kalai90-symmetric}
G.~Kalai.
\newblock Symmetric matroids.
\newblock {\em J. Combin. Theory Ser. B}, 50(1):54--64, 1990.
\bibitem[Max64]{maxwell64-rigidity}
J.~C. Maxwell.
\newblock L{.} on the calculation of the equilibrium and stiffness of frames.
\newblock {\em The London, Edinburgh, and Dublin Philosophical Magazine and
Journal of Science}, 27(182):294--299, 1864.
\bibitem[Oxl11]{oxley11-matroid}
J.~Oxley.
\newblock {\em Matroid theory}, volume~21 of {\em Oxford Graduate Texts in
Mathematics}.
\newblock Oxford University Press, Oxford, second edition, 2011.
\bibitem[Tyo26a]{tyomkyn26pc}
M.~Tyomkyn, 2026.
\newblock Private communication.
\bibitem[Tyo26b]{tyomkyn26}
M.~Tyomkyn.
\newblock On plane rigidity matroids.
\newblock {\em arXiv preprint arXiv:2602.11892}, 2026.
\bibitem[Whi83]{whiteley83-conerigidity}
W.~Whiteley.
\newblock Cones, infinity and {$1$}-story buildings.
\newblock {\em Structural Topology}, (8):53--70, 1983.
\newblock With a French translation.
\bibitem[Whi88]{whiteley88-cofactor}
W.~Whiteley.
\newblock The union of matroids and the rigidity of frameworks.
\newblock {\em SIAM J. Discrete Math.}, 1(2):237--255, 1988.
\bibitem[Whi91]{whiteley91-cofactor}
W.~Whiteley.
\newblock The combinatorics of bivariate splines.
\newblock In {\em Applied geometry and discrete mathematics}, volume~4 of {\em
DIMACS Ser. Discrete Math. Theoret. Comput. Sci.}, pages 587--608. Amer.
Math. Soc., Providence, RI, 1991.
\bibitem[Whi96]{whiteley96-geometry}
W.~Whiteley.
\newblock Some matroids from discrete applied geometry.
\newblock In {\em Matroid theory ({S}eattle, {WA}, 1995)}, volume 197 of {\em
Contemp. Math.}, pages 171--311. Amer. Math. Soc., Providence, RI, 1996.
\end{thebibliography}