WorksMathematics
Abelian covers and the index of the centre in a group
Abstract
Let be the largest number of abelian subgroups needed to cover a group in which every set of more than elements contains two distinct elements that commute, and let be the largest index of the centre under the same constraint. Erdős asked for the growth of , Neumann proved that both quantities are finite, and Pyber bounded both and exponentially. We determine for every positive integer and prove that whenever the constraint admits a nonabelian group. The covers are built directly from the commutator geometry of a scalar quotient, and the comparison with is made afterwards. Beyond finitely many small cases, both maxima are attained by the same groups, those isoclinic to an extraspecial -group of order , confirming Pyber's prediction for the central index in this range.
Introduction
How many abelian subgroups are needed to cover a group in which every sufficiently large set contains two commuting elements? The aim is to turn this restriction on noncommutation into a sharp bound for the size of an abelian cover. Write for the largest size of a set of pairwise noncommuting elements, called a clique, and for the least size of an abelian cover. A third invariant, the central index , measures the size of the quotient on which commutation is determined. These satisfy
a clique meets each abelian subgroup in at most one point, and each central coset lies in .
Erdős asked for the optimal bound on under the constraint [7 Problem 26]. Neumann [15] proved that this constraint bounds the central index. Thus both extremal functions
are finite. Pyber [16] proved an exponential upper bound for , hence also for . Extraspecial 2-groups give the exponential lower bound attributed to Isaacs [3 p. 40].
The covering question also has a graph formulation. In the noncommuting graph , the vertices are the noncentral elements and edges join noncommuting pairs. Then for nonabelian , and : a colour class generates an abelian subgroup, and a cover gives a colouring. Faber, Laver and McKenzie [8] developed this connection, while Mason [14] and Bertram [3] studied quantitative relations between covers and noncommuting sets. See also Abdollahi, Akbari and Maimani [1] for the graph-theoretic perspective. Thus the covering problem asks for the optimal bound on the chromatic number of noncommuting graphs in terms of their clique number.
Here we determine for every , classify its maximizers for , and identify its exact relation to . The covering profile and its equality case are proved directly. The comparison with uses only the central-index theorem [13 Theorems 1.1 and 1.2(v)]. For , the maximizers all have binary symplectic commutation geometry, in the dimension specified below.
Theorem 1.1 (The covering profile). For every .
Explicitly, for all , where
For , the groups attaining under are exactly those isoclinic to an extraspecial -group of order . These are also the groups attaining .
Why a square root appears. The identity in Theorem 1.1 reflects the geometry of the extremizers. Let . We call a group a finite-field determinant model if its central quotient is the additive group of and its commutator map is
where is a nonzero -linear map onto the derived subgroup, written additively. The determinant itself and its scalar trace versions are both included. The -lines give a spread, and every such group satisfies
(Corollary 3.3). The theorem shows that, for each , a group of this type simultaneously maximizes the covering number and the central index under . In the small cases the relevant fields have orders ; for , the models are the binary scalar trace models. Thus a common geometric construction explains the relation between the two extremal functions.
For , put , so . The function is nondecreasing, with and . In particular : this holds at and persists in steps of two.
The idea of the proof. The proof compares cliques and covers in the fibres of a quotient map. Call the largest clique size in a fibre its width. If the kernel is abelian, a fibre can be covered by exactly that many abelian subgroups. More generally, an abelian subgroup of index in the kernel gives a cover of size at most times the width. A clique in the quotient allows us to combine cliques in its fibres, so the sum of their widths is at most . Averaging over quotient cliques then bounds the total number of subgroups in the cover (§ 2).
For a -group, a scalar commutator form provides a symplectic quotient. In rank two, an abelian radical already gives by Lemma 4.2. In the remaining cases, the symplectic group allows us to average fibre cliques (§ 4). When covering the fibres separately gives too many subgroups, we instead use subspaces isotropic for all commutator forms (§ 5). The model is the symplectic spread, which also constructs the sharp examples in § 3. For nonnilpotent groups, § 6 uses affine geometry in a minimal nonnilpotent quotient for the solvable case. A centralizer descent applies to every nonnilpotent group when ; nonsolvable groups with are treated by the small-clique classification. For , each construction gives a strict inequality except in the binary symplectic model with one commutator form. Tracking where equality is possible therefore identifies the maximizers.
The main distinction is between a scalar commutator form, which certifies noncommutation, and the full commutator map, which determines whether a subgroup is abelian. The former provides large cliques; the latter provides common isotropic subspaces and hence covers. Keeping both roles explicit allows the same quotient estimates to handle groups of different nilpotency classes. The case maps in § 4–§ 6 indicate where each estimate is used.
Classification input. The nonsolvable small-clique case uses the structural classification in [2 Theorem 1.1]; we do not claim a proof independent of the classification of finite simple groups (CFSG). The centralizer descent in § 6.1 uses only the classical Baer–Suzuki theorem.
Passing to finite groups. Two groups are isoclinic if isomorphisms of their central quotients and derived subgroups preserve the commutators of lifts of central cosets. A finite -group is extraspecial when its centre and derived subgroup coincide and have order . For an extraspecial 2-group the commutator is a nondegenerate alternating form with values in its derived subgroup of order two. Conversely, if has order two and is elementary abelian of order , its commutator defines such a form. An isometry to the form of an extraspecial group of order gives an isoclinism. We call this the binary symplectic model of dimension .
By Zarrin’s theorem [22 Theorem 1.1(2)], a group of finite clique number is isoclinic to a finite group. Isoclinism preserves and , since the central quotient and the commutation of its lifts determine them. To see that it preserves , enlarge each member of an abelian cover by the centre. The resulting subgroups are exactly the lifts of subgroups of the central quotient on which the commutator is trivial. We may therefore work with finite groups throughout.
Write . A noncommuting pair extends to a clique . Thus clique number two never occurs, and clique number one characterizes abelian groups. We use and write for a cyclic group of order . The derived subgroup is ; is generated by the -th powers. Centralizers of subsets are taken pointwise. Vector-space dimensions are over the indicated field; a line and a plane have vector dimensions one and two. Projective dimensions are specified explicitly. Orthogonality refers to the chosen scalar form unless a subscript specifies another form. Logarithms without a subscript are natural.
Cliques and covers in cosets
For a subset of a group, write for the largest size of a clique contained in . We call the width of the coset .
Lemma 2.1 (The abelian coset formula). If is abelian, then
Moreover, can be covered by exactly that many abelian subgroups.
Proof. For , the elements commute if and only if , by cancelling the two products and using commutativity in . Thus they commute exactly when lie in the same coset of in . A clique takes at most one element from each resulting class, and any such choice is a clique. Each class consists of commuting elements and generates an abelian subgroup. Conversely, the clique gives a lower bound on the number of abelian subgroups in a cover.
If has an abelian subgroup of index , partition into cosets of that subgroup. Each has width at most . By the lemma, each can be covered by at most that many abelian subgroups. Thus costs at most . Throughout, the cost of a cover means its number of abelian subgroups.
Lifting a subgroup cover
For nonabelian , fix an abelian cover of a subgroup . For each coset , its translates , with a member of this cover, cover . Each translate costs at most subgroups. If fails to centralize , an element of extends a maximum clique in . If centralizes , the translate costs one. Thus
Adding widths across fibres
One fibre per nontrivial cyclic subgroup of the quotient suffices for the cover. To control their total width, we average over families whose cliques can be joined.
Lemma 2.2 (Clique and cover bounds in a quotient). Let be a surjection with kernel . Write , and let be the index of an abelian subgroup of .
(i) Two generators of the same cyclic subgroup of have fibres of the same width, covered by the same families of abelian subgroups.
(ii) If every point of fails to commute with every point of for , then
(iii) If contains a generator of every nontrivial cyclic subgroup of , then
Suppose a random subset of contains each with probability at least . If lifts of its distinct points are always noncommuting, then .
Proof. For (i), choose an exponent inducing the required power on the cyclic subgroup and coprime to , using the Chinese remainder theorem. The power map is a bijection between the fibres and preserves the cyclic subgroup generated by each element. This preserves both commutation and membership in abelian subgroups. For (ii), take a maximum clique in each inverse image. Their union is a clique in . For (iii), cover and each selected fibre, then use (i). For every realization of the random subset, the cliques in its fibres can again be joined. Their total size is at most , and its expectation is at least .
Lifting the cyclic subgroups of gives . For a -group of central index this cost is at most
since each nontrivial cyclic subgroup has at least generators.
Centralizers in a narrow coset
The factor makes the quotient estimate useful only if the kernel has a large abelian subgroup. To find one, we look inside a narrow coset. The centralizers of a maximum clique cover that coset, so one of them meets a large part of the kernel. The same cover also controls the intersection of the centralizers.
Lemma 2.3 (Centralizer partitions). Let be a maximum clique in a finite subset of . The sets , , cover . If the sum of their sizes is at most , they partition into commuting sets.
Proof. An uncovered point would extend . If the sizes sum to at most , the cover is a partition. A noncommuting pair in the part at would replace and enlarge , since neither point commutes with any other vertex.
We call a cover irredundant if deleting any member leaves an uncovered point.
Lemma 2.4 (Centralizers in a coset). Let , , and let be a maximum clique in . Then
The sets , , form an irredundant coset cover. Each satisfies , and some satisfies . If is a nonabelian -group, then .
Proof. Suppose centralizes but fails to commute with . Replace each vertex that commutes with by . The replacements remain in the coset and preserve mutual commutators because centralizes . Now no vertex commutes with , so we can adjoin , a contradiction. Thus centralizes all of , which is equivalent to belonging to .
The parts of the cover in Lemma 2.3 are , and each vertex belongs only to its own part. At least one part has size or more, giving the index bound. Multiplication by carries a clique in into with its commutation relations unchanged. Translating the cover by on the left and then inverting gives an irredundant cover of by right cosets of the .
If is a nonabelian -group, a one-part cover would contradict the intersection formula. Each subgroup in the cover is therefore proper. At most parts could cover only by partitioning into cosets of index . Two distinct index- subgroups are normal and multiply to , so any two of their cosets intersect. Hence all the subgroups in the partition would coincide. Their intersection lies in , which would make cyclic.
Tomkinson’s bound for irredundant coset covers [20] now gives
In particular a coset of a nonabelian subgroup has width at least three.
Suppose centralizes an abelian subgroup of index in . We can use the fibre width to cover itself. For each , the elements of commuting with are exactly those commuting with . The abelian coset formula then gives
Covering the cosets of in therefore costs at most .
Symplectic geometry and the sharp examples
When an alternating form determines commutation in a central quotient, cliques bound the admissible profile and isotropic subspaces lift to abelian subgroups. We construct both.
An alternating form is symplectic if the only vector pairing to zero with every vector is zero. A clique for is a set whose distinct vectors pair nontrivially. In dimension two, the basic example is the determinant pairing. Two nonzero vectors pair to zero exactly when they lie on the same line. Thus a clique is obtained by choosing one nonzero vector on each line. The following lemma extends this count to larger dimensions. For odd primes, the lower bound is due to Chin [5], in the equivalent formulation for extraspecial -groups. The binary maximum is due to Isaacs [3 p. 40]. We include proofs for completeness.
Lemma 3.1 (Symplectic cliques). Let be symplectic of dimension over . There is a clique of size . For , the maximum is , and every maximum clique spans and has sum zero.
Proof. To combine cliques in orthogonal summands, choose and use . The translate retains its original pairings, and its pairings with the remaining vertices of are those of . The new clique has size . Starting with vertices per symplectic plane gives vertices.
Over , a clique of size has a Gram matrix with diagonal zero and all other entries one. Its kernel consists of the vectors whose entries all equal their sum. The rank is therefore for even and for odd . It cannot exceed , so . At equality the rank is , so the clique spans . Its sum is orthogonal to the clique and is therefore zero.
We call a maximum clique for a binary symplectic form a symplectic simplex. A subspace on which an alternating form vanishes on every pair is called totally isotropic. When commutation in a central quotient is determined by such a form, the abelian subgroups containing the centre are exactly the lifts of these subspaces.
Lemma 3.2 (A cover from a symplectic spread). Suppose the central quotient of is a vector space of dimension over , and commutation of lifts is determined by a symplectic form on that space. Then .
Proof. A totally isotropic subspace satisfies , so . It contains at most nonzero vectors. Dividing the total by gives the lower bound .
For the cover, write and view as a space over . Use the form
The trace pairing is nondegenerate, so this form is symplectic. Every -line is totally isotropic because the determinant vanishes on it. The lines partition the nonzero vectors of . Such a partition into totally isotropic subspaces of dimension is called a symplectic spread. All nondegenerate alternating forms over of dimension have symplectic bases, so the partition gives a spread in the original quotient. Lifting its members gives the required cover.
Corollary 3.3 (Finite-field determinant models). Every finite-field determinant model with central quotient satisfies .
Proof. Write . The lines over are isotropic for the determinant and hence for its image under . Their preimages form an abelian cover. For the converse, choose a linear functional on the derived subgroup such that . The form is nondegenerate over : for , the map is onto . Every abelian subgroup, enlarged by the centre, therefore has isotropic image of dimension at most . Counting nonzero vectors as in Lemma 3.2 requires at least members.
Groups attaining the bounds
The constructions above give all lower bounds in Theorem 1.1. Let be the allowed clique size.
For the abelian model attains the bound. For , take , the group of upper unitriangular matrices, with . Its central quotient is , and the commutator is the determinant . The lines lift to an abelian cover; one nonzero point on each line gives a clique of size .
For , use an extraspecial 3-group with a four-dimensional symplectic central quotient. By the duality between the symplectic generalized quadrangle and the parabolic quadric , a set of pairwise nonorthogonal projective points corresponds to a partial spread of the quadric. Tallini’s bound [19]2, together with the nonexistence of ovoids in for odd , bounds its size by ; see [12], Theorem 3.3 and § 6. At this is seven, attained by combining two plane cliques. The clique number is therefore seven, and the spread in Lemma 3.2 gives an optimal cover of size ten.
For , take an extraspecial 2-group of order with . Its central index is , its clique number is , and its covering number is , by Lemmas 3.1 and 3.2.
The scalar quotient of a -group
A finite nilpotent group is a product of its Sylow subgroups. We first cover each prime-power factor using a scalar commutator form. We will combine their covers at the end of § 5.
Case map. For a scalar radical , write . The binary cases are treated in § 5; the odd-prime cases below are organized as follows. In the last row, .
| Radical and prime | Range or constraint | Covering argument |
| Any radical, | All | Binary case map in § 5. |
| Abelian noncentral, odd | ; or | Lemma 4.2; otherwise the fibre average in § 4.2. |
| Central, one component, odd | ; or , | The spread of Lemma 3.2, with . |
| Central, one component, | ; or | The ten-member cover; otherwise thirteen-point cliques and the spread. |
| Central, several components, odd | Lemma 4.6; the common centralizer planes in dimension four. | |
| Nonabelian, odd | Lemma 4.5 and the fibre average. |
Table 2.
Proposition 4.1 (Nilpotent groups). If is nilpotent and , then . For , equality requires the binary symplectic model of dimension .
The quotient given by one commutator form
Let be a nonabelian -group and set . A single scalar commutator value will suffice to certify noncommutation of lifts.
The commutators are central modulo . The abelian -group is nontrivial: otherwise the lower central series would be constant from , contrary to nilpotence. Choose a nonzero homomorphism , and also write for its composite with the quotient map on .
Applying to commutators gives the pairing
Centrality of the commutators in makes the pairing additive in each argument. Its radical is the subgroup
It is normal and contains . Hence is a vector space over , and the induced form is symplectic by the definition of . We call a scalar commutator form and its scalar radical. A nonzero pairing in certifies noncommutation of every pair of lifts, so a scalar clique lets us add its fibre widths.
Scalar radicals need not be central, even in class two: tests only , whereas requires for every . For example, in , projecting the commutator onto the first component gives the nonabelian radical ; here is the dihedral group of order eight.
Lemma 4.2 (The terminal rank-two case). If a scalar commutator form of has rank two and an abelian radical , then .
Proof. For , the perpendicular of in the symplectic plane is its own line. Thus is abelian. Two commuting elements outside have the same centralizer: each centralizer is abelian and hence contained in the other.
Choose one representative outside for each distinct centralizer. They form a clique, and their centralizers cover . If they cover as well, the clique and cover have the same size. Otherwise an uncovered point of extends the clique, and adding extends the cover. Hence .
Adding the widths of the fibres
Write for the dimension of . Select one nonzero vector on each line of . Its fibre has width , and a cover of that fibre serves the whole line by Lemma 2.2(i). To bound the total width, take the lines of a symplectic clique of size and move them by a uniformly chosen symplectic transformation. The selected vectors on the resulting lines still form a clique. Transitivity on lines gives inclusion probability . Lemma 2.2(iii) now gives
If has an abelian subgroup of index , the same lemma gives the common bound
Lemma 4.3 (A clique meeting a hyperplane once). In a symplectic space of dimension over , every hyperplane meets some clique of size in exactly one point.
Proof. For a prescribed hyperplane , choose with , and let be the orthogonal complement of . Take a clique of size in , with if . Then
is a clique. The first part inherits the pairings of . In the second part they are differences of distinct coefficients of , and the cross pairings are all . Only lies in .
Lemma 4.4 (Coset covers of -groups). If is a finite -group and the right cosets , , form an irredundant cover of , then .
Proof. We argue by induction on . At each step, we restrict the cover to a coset disjoint from the members already selected. The restricted cover uses fewer members, so induction controls the index of the next intersection. The case is immediate.
For , all are proper. Choose a maximal subgroup containing . It is normal of index . Right-translate the cover so that misses , and choose an irredundant subcover of , with index set . Put . Since , induction gives .
Write . If the intersection of the subgroups selected so far is not yet , the selected cosets do not cover , by irredundancy. Choose a point that they miss. They miss the whole coset , because lies in every selected subgroup. Restrict the remaining cover to and translate by . An irredundant subcover, indexed by , consists of cosets of in . Its indices are all new, so induction bounds the next intersection by
Each step adds a nonempty set of previously unused indices, so the process terminates at . Since , multiplying the index bounds gives
If the index remains unused, then and , so the exponent is at most . If it is used, then , since , and gives the same bound.
Lemma 4.5 (Fibres over a noncentral radical). Let be the radical of a scalar commutator form of rank on , and put .
If is abelian and noncentral, then .
If is nonabelian and , then the width of each fibre is at least and at least . Consequently , and if .
Proof. For (1), choose and a hyperplane of containing . By Lemma 4.3, some symplectic clique of size meets the hyperplane once. Each of its other vertices has a fibre of width at least by (1), since every lift moves . The remaining fibre contributes at least one.
For (2), let be the width of . Lemma 2.4 gives and an irredundant cover of by cosets. Their subgroups intersect in , of index at most by Lemma 4.4. This index is also , giving the second width bound. Summing the widths over a clique of size gives . If , choose moving a point of . Then and has width at least . Some symplectic clique of size contains , by transitivity on nonzero vectors, giving the last bound.
Odd primes
Choose a scalar form of least nonzero rank. The fibre average will suffice except in the central ternary cases, where we use larger cliques or a common isotropic spread.
Abelian noncentral radical. Lemma 4.2 gives when . For , Lemma 4.5 gives . To compare the fibre-cover bound with the profile, note that decreases for , since each step multiplies it by . Using , we may compare at . There , so
The factor in parentheses is at most . For this follows from . For use . The first inequality holds at and persists in steps of two, since .
Central radical and one commutator component. If , then has class two and is elementary abelian. Its commutator subgroup is elementary abelian as well, because . Write for the alternating commutator map on with values in . Composing with a linear functional on gives a scalarization of . When has dimension one, a symplectic spread gives exactly subgroups. In a scalar plane the spread costs exactly . For and , we compare the spread size with the profile at the scalar clique bound. If is even, gives . If is odd, and
Here decreases as increases in steps of two.
At and , the spread costs ten and the clique number is seven by § 3.1. For , we need larger cliques than the plane count provides. Ceria, De Beule, Pavese and Smaldore [4 Theorem 3.7] give thirteen projective points forming a clique in dimension six over : more generally, for , their points
lie in the -subgeometry , where , of and are pairwise nonorthogonal for the alternating form , since
Taking gives the required thirteen points. Write , with , and combine such cliques with plane cliques by Lemma 3.1. This gives . Thus , since and the bound grows faster than the spread cost.
Central radical and several commutator components. A subspace isotropic for one form may contain points with noncommuting lifts. The cover must therefore use subspaces isotropic for every commutator form. To compensate for any increase in their number, we use a second scalarization to construct a larger clique. The first form separates the fibres, and the second lets us take a whole line of points in most of them.
Lemma 4.6 (Cliques from two commutator forms). Let be alternating bilinear over , with values spanning , and . If some scalarization is nondegenerate, then and there are vectors whose -values on distinct pairs are all nonzero.
Proof. Take an independent functional on and put . Write . The operator is nonscalar because the values of span . Thus some are independent, since an operator preserving every line is scalar.
On the symplectic quotient , the functional is well defined and nonzero, since . This also shows that the quotient has positive dimension, so . Choose a clique of size meeting its kernel once, by Lemma 4.3. Lift its vertices to and replace every lift outside the kernel by the line . The -pairings between different fibres are still nonzero. Within a line the -pairing is a nonzero multiple of , so all its points can be used. This gives vertices.
Applying the lemma to gives . Lifting the lines gives a cover of size . For and , the inequality gives . Since and , , so the line count is below . For , the bound gives , while the line count is less than .
Let . The least-rank choice implies that every nonzero scalarization of on the four-space is nondegenerate. The Pfaffian bound [10], Lemma 3 gives , and hence . For , the map is onto: a functional annihilating its image would give a nonzero scalarization with in its radical. Thus is a plane. It is isotropic for , since it is spanned by and by a vector commuting with . For each nonzero we then have , with equality by dimension. The planes partition the nonzero vectors of ; their lifts give an abelian cover. Since by Lemma 4.6, this cover is strictly below the profile.
Nonabelian radical. Write with , and put and . Lemma 4.5 gives . At most cyclic lifts cover , and one such lift has index at most in . Since , (4) gives
The ratio of the last bound to is largest at , because decreases. Since , this gives
Here : , and decreases for odd . For , use and , which holds at and persists in steps of two. Then
since the two factors decrease for and , respectively. If , we have . If , then and , since decreases for . This suffices when , as .
It remains to take . Here , and the first cost bound gives . If , Lemma 4.5 gives . At , the inequality reduces to . This holds at and persists since . It also holds at , where has the same value, and persists in steps of two, since doubles. If , then and cyclic lifts give
The middle inequality starts at and persists as increases; the last uses . Thus for odd . When , the bounds above are all strict, including the costs and ten.
Binary groups and the equality model
Case map. Write . First use cyclic lifts when . In the remaining range , test the radicals in the order shown below. Here .
| Radical or commutator | Range | Covering argument |
| Any | Cyclic lifts; the small values are included below. | |
| Some nonabelian radical | The centralizer-cover estimate in § 5.1. | |
| All radicals abelian; rank two occurs | All remaining | Lemma 4.2: . |
| Abelian noncentral radical; all ranks | The uniform fibre average. | |
| Same | Lemma 5.1; separate fibres, except for , , handled by Lemma 5.2. | |
| All radicals central; one component | All remaining | Lemma 3.2; the equality model. |
| All radicals central; several components | ; or | The amplified clique forces the cyclic-lift range; otherwise Lemma 5.4. |
Table 3.
For the first row, cyclic lifts give when . If , then : the cost is at most three for , and seven for , which requires .
In the remaining rows, a nonabelian radical raises every fibre width; an abelian noncentral radical raises widths in directions that act nontrivially. With central radicals, we use the commutator forms themselves.
A nonabelian radical
Suppose some scalar radical is nonabelian. Put and write and . Thus , with and .
The radical can be covered by cyclic lifts, and one such lift gives an abelian subgroup of index at most in . Thus the fibre bound gives
By Lemma 4.5, the width of is at least . If , the widths over a simplex give , since . This contradicts .
Now . Conjugation on factors through . The simplex spans , so at least one of its lifts acts nontrivially on . Number the lifts so that is such a lift. The simplex vertices sum to zero, so the first still span . Thus . A coset of this intersection is determined by its cosets modulo the , giving
The first fibres therefore contribute at least vertices beyond their baseline widths, and the last fibre contributes at least one more. Hence
Together with , this gives , using . Put and . The clique bound gives . The cover estimate and give
It remains to compare with . At , the required inequality is , which holds since . Thereafter doubles with each unit of , whereas grows by one. Thus .
An abelian noncentral radical
Assume now that all scalar radicals are abelian. If some scalar form has rank two, we have by Lemma 4.2. For , this is strictly below . If all scalar ranks are at least four, choose a noncentral radical of rank , if one exists, and use (4) with . Write , so the fibre cost divided by is at most
For the ratio decreases with , starting at . When , we have . The averaging bound alone is insufficient here, since it does not exclude individual fibres of large width. We first show that every fibre has width at most two. Covering the fibres separately then suffices except at with , where we build a cover of the central quotient by common isotropic subspaces.
Lemma 5.1 (Width reduction in small scalar rank). Let be an abelian noncentral scalar radical in a finite -group , with . Assume all nonzero scalar forms have rank at least four. If , then , , and every fibre of has width at most two.
Proof. Step 1: cliques from two fibres. Write additively and set , where is right conjugation on . Since is abelian, this action factors through the elementary abelian group . In particular the actions commute, and the notation , for , is unambiguous. The width of the fibre at is , by Lemma 2.1. The commuting classes in are the cosets of . For distinct fibres , , two elements , commute exactly when
Indeed, and ; all the factors after belong to the abelian subgroup . Suppose both fibres have width two and their kernels differ. Start with a noncommuting pair. Vary by an element of and by an element of . Neither change affects the equation, and each gives a second class in its own fibre. This yields a four-point clique.
More generally, fibres with different actions give at least three clique points. A fibre of width at least four suffices on its own. If one fibre has width one and the other width two, choose both classes in the latter. Equation (5) lets us choose a point in the inactive fibre that commutes with neither. If both widths are two and the kernels agree, the two operators have distinct images of order two. The equation has at most one solution in the pair of class labels, so we can use both classes in one fibre and one class in the other.
Step 2: restricting the scalar dimension. Choose a direction acting nontrivially. The points of a symplectic simplex in each lift to two directions , . Their actions differ, since and . Each pair supplies at least three points by Step 1, and the scalar form separates different pairs. Thus , which leaves and .
Step 3: linearizing the action. We next show that every fibre has width at most two. The first step is to make the action a vector space of maps. If the directions of width at most two lay in a hyperplane, the simplex in Lemma 4.3 would give at least vertices. This contradicts , so those directions span . Their action operators have images of order at most two. Every other action preserves each such image because the actions commute. An image of order two has no nonidentity automorphism, so it is fixed pointwise. Thus for the spanning directions, and makes all operators sums of the spanning ones. Their images generate the central elementary abelian group , and . The common kernel in of all the operators is , since the centre lies in every scalar radical. Thus and are binary vector spaces. The assignment is now linear, and its image is spanned by maps of rank one. A fibre has width 2 iff .
Step 4: excluding wider fibres except at one endpoint. If some direction has rank , repeat the clique construction in . Each pair of lifts now supplies four vertices. Indeed, if both widths are at most two, the maps must have rank one and distinct kernels, since their sum has rank at least two. The two-fibre argument applies. Hence . In the range , this leaves only , .
Step 5: excluding the endpoint , . At this endpoint, we need two rank-one maps with different images and different kernels. If no pair differed in both respects, the rank-one maps would share an image line or a kernel. Indeed, two with distinct images would have the same kernel, and every other rank-one map would have to share it. Their sums would then all have rank at most one, contrary to the existence of a map of rank at least two.
In the four-space , take the plane spanned by two directions inducing . Each translate of the plane contains a direction of width at least four. To see this, choose inputs with , and . The outputs are independent. In each translate, the two possible values at differ by , and the two at differ by . The choices are independent. If every pair were dependent, any nonzero value at would span a line containing both values at , and hence containing . Both values at would then lie on , contrary to their difference .
The four translates give at least four wide directions. A totally isotropic subspace in the symplectic four-space has at most three nonzero vectors. Two wide directions therefore pair nontrivially. Put . We complete to a simplex with the three directions for . They pair nontrivially with and with each other.
At least two of these last three directions act nontrivially. If two acted trivially, the kernel of the action map would have dimension at least two. Rank-nullity then gives ; equality holds since contains . Its only rank-two map would be , so . The actions at would then be . Two trivial actions would force all of to act trivially. But then would act through just the one map , contradicting . The simplex therefore gives at least clique vertices.
This excludes rank at least two throughout the range under consideration. Every fibre has width at most two.
Covering each fibre separately now costs at most . For , this is below . Only and remain. Here we replace the separate covers by subspaces isotropic for every commutator form.
Lemma 5.2 (Nine common isotropic subspaces). Let be an abelian noncentral scalar radical in a finite -group , with . Suppose every nonzero scalar form has rank at least four, every fibre has width at most two, and . Then has a cover by nine common isotropic subspaces. Consequently .
Proof. Step 1: the five-dimensional quotient. We first prove that and that has class two with central squares. This reduces the problem to an alternating map on a five-dimensional binary space. We then construct the cover from a Kronecker block when there are two scalar forms, and from the incidence of common isotropic planes when there are three.
Put and retain the action notation from the preceding proof. These operators form a space of maps from to , each of rank at most one. First we show ; a larger quotient would give an eleven-point clique. If , the kernels of the rank-one maps cannot all coincide, since they intersect in zero. The maps must therefore share an image line. Here we use the following rank-one observation: if and have different image lines and different kernels, their sum has rank two. Thus any two rank-one maps in share an image or a kernel. Once two have different kernels, they share an image, and comparison with both shows that every other nonzero map has that same image. Their action functionals span the dual of . Project these functionals onto two coordinates, giving a surjection with kernel .
Choose . Then and has dimension one, so maps onto . Write for the vectors of , and choose coordinates with . The second coordinate is then a nonzero functional on the symplectic plane . It is one at two of the three nonzero points and zero at the third. In each pair of lifts, adding toggles the first coordinate. The two pairs over the points with second coordinate one thus carry the labels , and the remaining pair carries .
In the first two pairs, the action functionals are nonzero and distinct because their projected labels are. Over they therefore have distinct kernels. Each pair gives four clique points by the two-fibre argument in Lemma 5.1. In the last pair the actions differ, so the same lemma gives at least three points. The three nonzero points of a symplectic plane pair nontrivially, so the points form a clique, contrary to .
Thus . We show that the central quotient is an elementary abelian space of dimension five, so we can look for a cover by common isotropic subspaces. The action of on the group of order two is trivial. If , projection of the commutator onto gives a nonzero scalar form on . It is well defined because . Moreover , so the projection kills : this is one of the scalar forms to which the rank hypothesis applies. On the four-space , it is therefore nondegenerate. Choose acting nontrivially on . Its perpendicular for this form contains two vectors with nonzero pairing, by rank at least four. For lifts of these vectors, conjugation by changes by central factors, so it fixes . Yet represents the nonzero class in , whose representatives are moved by . This proves that has class two.
Its squares lie in . Since , the commutator is trivial if either square is central. If both are noncentral, the squares agree modulo , so again commutes with . Thus every square is central, and shows that is elementary abelian.
Step 2: the alternating map on the quotient. Write , and let be the commutator map. Write for the common centralizer of in . The original scalar form vanishes on the nonzero space , so . All nonzero scalar forms have rank four.
Step 3: two scalar forms. If , Scharlau’s decomposition [17] gives a single singular Kronecker block of dimension five. A singular block of dimension one would give a nonzero common radical. If a singular block had dimension three, the remaining dimension two could contain neither another singular block (there is no common radical) nor a regular block nondegenerate for all three nonzero combinations: the space of alternating forms on a two-space has dimension one. Some nonzero combination would consequently have rank at most two, contrary to the rank-four hypothesis. The only possible decomposition is therefore the single five-dimensional singular block. A singular Kronecker block of a prescribed odd dimension is unique up to simultaneous equivalence of the two alternating forms [17]. It is therefore enough to construct the cover in any model of this block. To exhibit such a model, take , a two-dimensional subspace , and a surjection . On the pair
has zero common radical. Multiplication by is onto , and cannot lie in for . Each nonzero scalarization has rank four because the pairing between and has rank two. It therefore models the same Kronecker block. The space and the eight graphs , , partition the nonzero vectors into common isotropic subspaces.
Step 4: at most three scalar forms. Suppose . We look for a three-space among the radical vectors of the scalar forms. There is one nonzero radical vector per nonzero scalar form, since every form has rank four.
If there were four independent forms, the fifteen nonzero combinations would have at most fifteen radical vectors. At a nonzero vector outside that set, the four functionals obtained by evaluation at are independent: a linear relation would put in the radical of a nonzero combination. Their common kernel has dimension and contains , so it is exactly . Two distinct vectors outside the set therefore cannot commute even for these four forms. There are at least such vectors, contrary to . Thus .
Step 5: counting the common isotropic planes. Let be the set of radical vectors of the seven nonzero forms. Outside , evaluation has rank three, so the common centralizer is a plane. Each such vector lies on just that one common isotropic plane: an isotropic plane containing must lie in , which already has dimension two. Take one vector outside from each plane that meets the complement of . If two representatives commuted, they would span the unique isotropic plane through either of them. They would therefore come from the same plane. The selected representatives thus form a clique.
To see how many planes this leaves inside , count the planes through a nonzero vector . If nonzero forms have radical , the annihilator of has size . Rank-nullity gives . The planes through correspond to the nonzero vectors of , so there are
Each of the seven nonzero scalar forms has exactly one nonzero radical vector, so , even if several forms have the same radical. Summing over gives incidences. Each plane has three nonzero vectors, hence there are exactly fifteen common isotropic planes. The clique bound leaves at least five planes with all their nonzero vectors in .
Step 6: forcing an isotropic three-space. These five planes force the three-space. Two must meet, since has at most seven points. If they meet at , their span has dimension three and already contains five vectors of . A plane not contained in uses at least two vectors outside , and there are at most two available in . Those two vectors determine the plane, so at most one of the five planes lies outside . Among the at least four planes in , one avoids , since only three can pass through . It is isotropic, and commutes with all of by the construction of . Together this plane and span a common isotropic three-space.
Step 7: partitioning the remaining vectors. For , we have , so . These seven vectors exhaust . Since each accounts for one of the seven scalar forms, for . Every vector outside lies on a unique common isotropic plane, and this plane meets only at zero. These planes partition the 24 vectors outside into eight triples, giving eight planes in addition to .
In either case the nine subspaces lift to nine abelian subgroups. Lemma 5.1 applies to the present hypotheses and gives . Together with , this yields .
Central radicals
If every scalar radical is central, then has class two, is elementary abelian of even dimension , and every nonzero scalarization of the commutator is nondegenerate. If the commutator has one component, the spread gives , with . For these cover sizes are and fit the stated profile. For , this group has the binary symplectic model and .
With several forms, we again use one to separate fibres and a second to enlarge the clique inside them. Here we can triple each contribution by choosing a plane on which the forms behave differently. Take independent forms . There is a plane isotropic for and nondegenerate for . To find it, write , choose with independent, and choose with . Then has the desired properties. In , choose a binary symplectic clique of size , using the singleton if . Lift each vertex to a representative orthogonal to for , which is possible because is nondegenerate. Adding the three nonzero vectors of to this representative gives three pairwise -nonorthogonal lifts. Lifts of different vertices are already separated by . Thus
When , this implies , so cyclic lifts suffice. In the smaller dimensions, we use subspaces isotropic for the full commutator. In dimensions four and six, spreads or common centralizers supply them. In dimension eight, we first cover disjoint common isotropic planes, then pair commuting points among the remaining vectors. The planes reduce the number of subgroups enough to obtain a strict bound.
Lemma 5.3 (Regular pencils in dimensions four and six). Let be independent alternating forms on a -dimensional space over , with or . If every nonzero form in their span is nondegenerate, then can be identified with so that its -lines are isotropic for both forms.
Proof. In Scharlau’s regular decomposition [17], a block with elementary divisor has dimension . Here there are no singular blocks or linear elementary divisors, since every nonzero combination is nondegenerate. In particular each irreducible has degree at least two. The sum of the quantities over the regular blocks is . Since or , two blocks would already contribute at least four, and in one block would also contribute at least four. Thus there is exactly one block, with and . On this block the two forms are scalarizations of the determinant on . The -lines are therefore isotropic for both.
Lemma 5.4 (Small binary central quotients). Let have class two and elementary abelian central quotient of dimension . Suppose that there are at least two independent scalar commutator forms, and that every nonzero form is nondegenerate. Then , where . For the inequality is strict.
Proof. Step 1: the common-centralizer dimension. Write , let be the commutator map on , and put and . The Pfaffian bound [10 Lemma 3] gives .
For every , the evaluation map is onto: otherwise a nonzero functional annihilating its image would give a scalar form with in its radical. Hence every common centralizer has dimension .
Step 2: dimension four. If , then , and the regular pencil gives five blocks by Lemma 5.3. The scalar clique has size five, so .
Step 3: dimension six, a common spread or a third form. If , then or and (6) gives . With two forms, the regular pencil gives nine blocks. With three forms and , first apply the regular pencil lemma to two forms, giving nine common isotropic three-spaces. On each three-space the restriction of the third alternating form has rank zero or two. In rank zero keep the whole space. In rank two its radical is a line, and the inverse images of the three lines of the two-dimensional quotient are isotropic planes covering the three-space. Thus each of the nine blocks costs at most three. This costs at most .
Step 4: dimension six, the range . With three forms, put . Each has dimension three by the evaluation calculation above. We show that all these spaces are common isotropic, unless there is an eleven-point clique. If all are common isotropic, and , then . Both have dimension three, so . Thus two such spaces meeting off zero coincide. They partition the 63 nonzero vectors into blocks of seven, giving nine blocks.
Otherwise choose with nonisotropic and a scalar form nonzero on . In the symplectic space , the plane is nondegenerate. Translate the three nonzero vectors of by a nonzero and add the two nonzero vectors of other than . The five vectors lie outside . Within each set the pairings are unchanged, and across the sets they are the nonzero pairings with . Lift each vector to its two translates by . The form separates lifts of different vectors. If is one of the chosen lifts, its image lies outside , so and . Hence and fail to commute with one another, and both fail to commute with . The five doubled directions together with therefore give eleven vertices, a contradiction.
Step 5: dimension eight. Here by (6). If , cyclic lifts suffice since . For , pairing commuting points could leave up to singletons. We first choose disjoint common isotropic planes. Each covers three points in one subgroup, offsetting the cost of one singleton.
Every nonzero vector lies on the same positive number of common isotropic planes. Indeed, is onto for , so those planes are the lines in , of dimension . There are planes in total, and each chosen plane meets at most planes off zero, including itself. After choices, at most planes are forbidden. A greedy choice therefore gives planes with pairwise disjoint sets of nonzero vectors. Cover their points in subgroups, then pair commuting points among the rest until no pair remains. The unpaired points form a clique and hence number at most . The total cost is at most
Products of Sylow subgroups
An abelian group needs one subgroup. If only one Sylow subgroup is nonabelian, the abelian factors preserve , and the isoclinism class. Otherwise split into two nonabelian direct factors. Their clique numbers satisfy , and multiplying the two covers gives an upper bound for the product. We use for , direct from the formula for (for , use ). Unless , we have , so
For the cost is nine. Induction on the number of nonabelian factors gives , since is nondecreasing.
Abelian covers of nonnilpotent groups
Proposition 6.1 (Nonnilpotent groups). If is nonnilpotent and , then , with strict inequality for .
Case map. We argue by strong induction on . The nilpotent case is already available, as is the profile for smaller clique numbers.
| Group or quotient | Clique range | Covering argument |
| Any nonnilpotent group | Centralizer descent and induction on the smaller clique number. | |
| Nonsolvable group | A central extension of of index at most two. | |
| Affine, cyclic complement; abelian | ; or | Cover the nilpotent translation preimage; otherwise the quadratic bound. |
| Affine, cyclic complement; nonabelian | , | Lemma 6.3 and the weighted fibre estimate. |
| Affine, nonabelian complement | Affine degree nine and a cover from the widths of translation conjugates. |
Table 1.
Here is the affine degree, a complement-fibre width, and a selected fibre width, as defined below.
Centralizer descent for large clique numbers
The selection argument is the one in [13 Proposition 4.3]; we give the proof because it uses no central-index bound. We use the classical Baer–Suzuki theorem [9 Theorem 3.8.2]: if is a -subgroup and is a -group for every , then . Its nilpotent form says that if every two conjugates of a cyclic subgroup generate a nilpotent group, its normal closure is nilpotent. Indeed, apply the theorem to the -part of a generator for each prime ; each such part lies in , so the generator lies in . See also [11 pp. 281–282] for this formulation; the stronger extensions proved there using the classification of finite simple groups are not needed.
We first record two elementary facts used in the descent. A nonnilpotent group has clique number at least four. Indeed, if , (3) with the whole group as its coset gives . A central quotient cannot be nontrivial cyclic, so its only nontrivial possibilities are and . The latter has a four-point clique (the three transpositions and a 3-cycle), which lifts to . The former makes nilpotent of class at most two.
Also, if and centralize one another, then
To see this, multiply every vertex of a clique in by every chosen lift of a clique in . For distinct quotient vertices, a commutator from is noncentral in , whereas any cancelling commutator from would lie in . For equal quotient vertices, noncommutation is inherited from . The products are distinct for the same reason.
Lemma 6.2 (Centralizer descent). Let be a finite nonnilpotent group and put . There is a nonnilpotent subgroup whose centralizer satisfies
Proof. Put , , and . Take a maximum clique in , of size . Its centralizers cover . For with , we have , so . Since , summing gives
Some therefore has . By Baer–Suzuki, has a conjugate such that is nonnilpotent. The -orbit of avoids , so it has size at most . Its stabilizer is , giving . The quotient is nonnilpotent, since a central extension of a nilpotent group is nilpotent. It has a clique of size four. Now (7) gives .
The numerical cost of descent. Let and put . For the centralizer in Lemma 6.2, induction gives , and (2) gives
For this bound is at most . Increasing by four quadruples , while grows by at most a factor two and grows by a factor at most . Thus for .
Nonsolvable groups with at most 44
Let be nonsolvable with , and let be its solvable radical. Write . In alternatives (1) and (2) of [2 Theorem 1.1], with or , where . Here , since is perfect, and , since is nonabelian simple. Taking in these two cases, all four alternatives give with and isomorphic to , , or . The last two are excluded by and [1 Lemma 4.4]. Thus .
The nonidentity elements of have as centralizers its five Klein four subgroups, ten subgroups of order three and six subgroups of order five. Two commuting nonidentity elements have the same centralizer, since these centralizers are abelian. Choosing one element from each gives .
The intersection is solvable and normal in , so its image in is trivial. Conversely is abelian and normal in , hence . In particular . For each , the map from to its centre is a homomorphism. It is trivial by perfectness, so . By (7), , forcing to be abelian.
Put . Then and . The nontrivial cyclic subgroups of have abelian preimages in and cover it. Since , we have . By (2),
Indeed , and in steps of two the left side grows by less than a factor two while doubles.
The affine quotient
Let be solvable and nonnilpotent with . Choose a nonnilpotent quotient of smallest order. Every proper quotient of is nilpotent. Take a nonnormal maximal subgroup $L of . It has trivial core, since a nontrivial core would give a smaller quotient with a nonnormal maximal subgroup, hence a smaller nonnilpotent quotient. The primitive solvable structure theorem [6, Theorem A.15.6} gives , with and a faithful irreducible action of on . We view as the affine map . Thus consists of translations and of maps fixing the origin. The inverse images in of affine maps are the fibres whose widths we will add.
Since acts faithfully and is nonnilpotent, we have . The complement is nilpotent by minimality, and its largest normal -subgroup is trivial. Thus . A clique of lifts to a clique of . Any nonidentity central element acts without nonzero fixed points, since its fixed space is -invariant and the action is faithful. The elements with linear part form a clique, to which any nonzero translation can be adjoined. Hence .
When the complement is abelian. The algebra generated by is a field by irreducibility, so is cyclic of order dividing . If the linear part is nonidentity, then is invertible, so the affine map has a unique fixed point. It therefore belongs to exactly one of the point stabilizers, which are the conjugates of . An affine map commuting with it must fix that point. Thus nonidentity maps from different stabilizers do not commute, and a nonzero translation commutes with none of them. This lets us add clique sizes across the corresponding fibres.
Let be the preimage of the translations. Write for the largest width of a nonzero translation fibre and for the largest width of a nonidentity complement fibre. By conjugacy a fibre of width occurs over each point stabilizer. Choose one there and a translation fibre of width . The cliques in these fibres join by Lemma 2.2(ii). Writing for the contribution from the complements, we get .
By Lemma 2.2(i), select one translation fibre per line and one generator per nontrivial cyclic subgroup of each complement. There are translation fibres and fibres per complement. In particular , the number of inversion orbits on nonidentity elements. For an abelian subgroup of index in , the quotient estimate gives
If is odd, then and . If is even, then and .
Abelian kernel and cyclic complement
Write , so . For we have , since the square starts at and grows by less than a factor two in each step.
Suppose first that . For every nonidentity complement element, every lift centralizes by the abelian coset formula (1). The conjugates of the complement generate . Indeed, for a nonidentity complement element , the map is invertible on the translations, and its values arise as commutators with . Because is itself abelian, it follows that . Thus is nilpotent, as it is a central extension of an abelian group, and the preimage of each cyclic complement is abelian.
A maximum clique in can be chosen in . If is nonabelian, its maximum cliques avoid the centre. If is abelian, choose one lift of a nonzero translation. It follows that . The cover of together with the abelian complement preimages costs , where . If , then and the cost is at most . For and , use to bound the cost by . If , the complements leave , so . The cost is at most , since . Thus the bound is strict for .
Now suppose . The complement fibres contribute at least clique vertices and the translation fibres contribute at least one, so . In the fibre cover (8), with , those contributions give a cost . Each further vertex costs at most subgroups. Since , there are at most further vertices. Hence
The bounds on give : for odd use , and for even use and . Consequently
This proves when . The only smaller value is , where and the same bound gives .
Nonabelian kernel and cyclic complement
A narrow fibre supplies a large abelian subgroup of the kernel and controls the cost of its cosets.
Lemma 6.3 (Covering the kernel from one fibre). Let be nonabelian and let have width . Assume for each with . There is an abelian subgroup of index , centralized by an element of , with
Proof. As in § 2, if centralizes then the cosets of in cost at most in all. We find inside a centralizer in the cover of a maximum clique in . The centralizers in have intersection and satisfy by (3).
The uniform construction. For , if all centralizers have index at least , their cosets partition the fibre into commuting sets by Lemma 2.3. Each centralizer is then abelian of index . Otherwise choose of index . It satisfies by Lemma 2.4. If is nonabelian, a cover by at most abelian subgroups contains one of index at most : if the largest has order , their union has size at most . If is abelian, take itself. For , this gives .
Small widths. Only widths three and four need sharper information about the central quotient. For , the noncyclic quotient has order four or six. In the latter case it is , whose automorphisms are inner. As is normal, one can choose a representative of that centralizes in . Translating the three transpositions and a 3-cycle by this representative gives a four-point clique in the image of , a contradiction. Thus and . Some centralizer has index at most three by Lemma 2.4. Its index divides four and is at least two, hence equals two. Its quotient by is cyclic, so this centralizer is abelian.
For , the construction above gives if , and in the partition case. If , we need an abelian subgroup of index at most two in . The same centralizer intersection gives . If has an element of order at least three, its cyclic subgroup has index at most two, and its preimage is abelian. Otherwise the quotient is elementary abelian of order at most eight. A line suffices when its order is two or four, and itself suffices if it is abelian. At order eight, two distinct nonzero classes must have commuting lifts, since . They span a plane with abelian preimage of index two. In all cases .
The resulting bounds. For , the index bounds give . For , the profile gives , so
The last expression decreases from since . The same index bounds give : for , both and are at most . ◻
Take . Both widths are at least three, and ensures . The induction hypothesis gives the cover bound required in Lemma 6.3. With its subgroup , the fibre estimate (8) becomes
The translation fibres contribute at least clique vertices, and the complements at least . At these widths, the cover bound is , including the subgroups for the kernel. The bounds on give for either parity of . Put .
For fixed , each vertex beyond increases the cover bound by at most , since . As , the linear bound
grows by a factor at most per extra vertex. The lower bound for grows faster, by a factor . Comparing at the starting size therefore gives
Finally, increasing by one adds vertices to the starting clique size. The factor increases by a factor of at most two, whereas the exponential denominator increases by . Thus the last bound is largest at , where , and is at most by Lemma 6.3.
Nonabelian complement
Choose . Its fixed-point-free action on gives . We enlarge a clique in this centralizer through the translation fibres.
Take a maximum clique in . For each , use the two complement elements . Distinct pairs are completely noncommuting modulo . Coprimeness gives
Fixing the component and varying the component gives a clique of size over . When , choose its component nonzero. Commutation with an element , where commutes with , requires
Projection onto shows that this clique is completely noncommuting with both fibres over and over : both act without nonzero fixed vectors on .
The fixed spaces of and intersect trivially. If both are nonzero, the two cliques have sizes with product at least , so together give at least vertices. If one is zero, its full fibre gives vertices. Use such a pair for vertices if it exists. Otherwise adjoin the full fibre over . Each of at least further pairs contributes . Thus
A nonabelian requires . It has a nonabelian Sylow subgroup, so its order is divisible by the cube of a prime. Since , its order divides , the part of coprime to . For this product is , and , respectively. None has a prime cube divisor. The remaining possibilities are and . The latter gives , already treated in § 6.1. Thus , , and is a -group of order at most , since and .
The kernel is abelian. Otherwise every fibre has width at least three by (3), and a clique of size at least in would give . Write for the width of the fibre over . For , its conjugates by translations are the elements with . Distinct such elements do not commute, since . Their fibres have the same width, so . Summing over the elements with linear part , and then over , gives
If and , its eigenvalues are , since the action is coprime and the field is . Thus and . There are at most such elements by the determinant homomorphism. For the other we have , so the sum is at most . Add one fibre for each of the four lines in and include itself. By Lemma 2.2,
since and . This completes the proof of Proposition 6.1.
Attaining the profile
Proof of Theorem 1.1. Propositions 4.1 and 6.1 give . Since is nondecreasing, . The groups in § 3.1 attain these covering bounds, so .
Fix and suppose with . Then , so and . By Propositions 4.1 and 6.1, has the binary symplectic model of dimension . Thus is odd and this dimension is . Conversely, the extraspecial model in this dimension attains the bound by the symplectic spread.
The central-index theorem [13 Theorems 1.1 and 1.2(v)] gives for , , and for . Thus for . For , its equality case is the same binary symplectic model just obtained for the covering problem.
Extremal profiles and equality for individual groups. The identity compares two maxima under the same clique bound. The equality cases show what that bound selects. For , the determinant on and its binary trace both give central index and covering number . The same field lines give an optimal cover in each model. Their clique numbers, however, are and , respectively. Thus the clique constraint can distinguish commutator maps that give the same central index and the same covering number.
For an individual group satisfying , let be the largest order of the image of an abelian subgroup in . In the tight case, , and every optimal cover induces a partition of the nonidentity elements of the central quotient. In the loose case, , and every optimal cover has two members whose images share a nonidentity element. The loose case occurs even for groups of class two with elementary abelian central quotient. The open question is which incidence properties of the abelian images determine whether exactly members are needed in this regime.
Statements and declarations
Funding. The author received no specific funding for this work.
Competing interests. The author declares no competing interests.
Data availability. No datasets were generated or analysed during this study.
Author contributions. The sole author carried out the research and wrote the manuscript.
Declaration of generative AI and AI-assisted technologies
OpenAI ChatGPT assisted with mathematical discussion, reference checks and LaTeX editing.
The author takes responsibility for the mathematical arguments, references and final text.
References
- [1]A. Abdollahi, S. Akbari and H. R. Maimani, Non-commuting graph of a group, J. Algebra 298 (2006), 468–492. doi:10.1016/j.jalgebra.2006.02.015.DOI
- [2]A. Abdollahi, A. Azad, A. Mohammadi Hassanabadi and M. Zarrin, On the clique numbers of non-commuting graphs of certain groups, Algebra Colloq. 17 (2010), 611–620. doi:10.1142/S1005386710000581. arXiv:0903.0692.DOI
- [3]E. A. Bertram, Some applications of graph theory to finite groups, Discrete Math. 44 (1983), 31–43. doi:10.1016/0012-365X(83)90004-3.DOI
- [4]M. Ceria, J. De Beule, F. Pavese and V. Smaldore, On large partial ovoids of symplectic and Hermitian polar spaces, J. Combin. Des. 31 (2023), no. 1, 5–22. doi:10.1002/jcd.21864. arXiv:2203.04553.DOI
- [5]A. Y. M. Chin, On non-commuting sets in an extraspecial p-group, J. Group Theory 8 (2005), 189–194. doi:10.1515/jgth.2005.8.2.189.DOI
- [7]P. Erdős, Some unsolved problems, in B. Bollobás and A. Thomason (eds.), Combinatorics, Geometry and Probability: A Tribute to Paul Erdős, Cambridge University Press, Cambridge, 1997, pp. 1–10. doi:10.1017/CBO9780511662034.004.DOI
- [8]V. Faber, R. Laver and R. McKenzie, Coverings of groups by abelian subgroups, Canad. J. Math. 30 (1978), 933–945. doi:10.4153/CJM-1978-081-1.DOI
- [9]D. Gorenstein, Finite Groups, Harper & Row, New York, 1968.
- [10]R. Gow and R. Quinlan, On the vanishing of subspaces of alternating bilinear forms, Linear Multilinear Algebra 54 (2006), 415–428. doi:10.1080/03081080500361264.DOI
- [11]R. M. Guralnick and G. R. Robinson, On extensions of the Baer–Suzuki theorem, Israel J. Math. 82 (1993), 281–297. doi:10.1007/BF02808114.
- [12]A. Klein and K. Metsch, New results on covers and partial spreads of polar spaces, Innov. Incidence Geom. 1 (2005), 19–34. doi:10.2140/iig.2005.1.19. Corrections, Innov. Incidence Geom. 11 (2010), 237–240.
- [13]G. Lecomte, The exact central-index profile of groups with bounded noncommutativity, preprint, 2026. arXiv:2611.00943.
- [14]D. R. Mason, On coverings of a finite group by abelian subgroups, Math. Proc. Cambridge Philos. Soc. 83 (1978), 205–209. doi:10.1017/S0305004100054463.
- [15]B. H. Neumann, A problem of Paul Erdős on groups, J. Austral. Math. Soc. Ser. A 21 (1976), 467–472. doi:10.1017/S1446788700019303.
- [16]L. Pyber, The number of pairwise non-commuting elements and the index of the centre in a finite group, J. London Math. Soc. (2) 35 (1987), 287–295. doi:10.1112/jlms/s2-35.2.287.
- [17]R. Scharlau, Paare alternierender Formen, Math. Z. 147 (1976), 13–19. doi:10.1007/BF01214270.DOI
- [18]G. Tallini, Fibrations through lines in a nonsingular quadric Q₄,q of PG(4, q), Atti Accad. Peloritana Pericolanti Cl. Sci. Fis. Mat. Natur. 66 (1988), 127–146 (1989).
- [19]G. Tallini, Blocking sets with respect to planes in PG(3, q) and maximal spreads of a nonsingular quadric in PG(4, q), Mitt. Math. Sem. Giessen 201 (1991), 141–147.
- [20]M. J. Tomkinson, Groups covered by finitely many cosets or subgroups, Comm. Algebra 15 (1987), 845–859. doi:10.1080/00927878708823445.
- [22]M. Zarrin, On non-commuting sets and centralisers in infinite groups, Bull. Aust. Math. Soc. 93 (2016), 42–46. doi:10.1017/S0004972715000696.