Skip to content
Publicly Available Published by De Gruyter November 17, 2017

Metabelian thin Beauville p-groups

  • Norberto Gavioli EMAIL logo and Şükran Gül
From the journal Journal of Group Theory

Abstract

A non-cyclic finite p-group G is said to be thin if every normal subgroup of G lies between two consecutive terms of the lower central series and |γi(G):γi+1(G)|p2 for all i1. In this paper, we determine Beauville structures in metabelian thin p-groups.

1 Introduction

Beauville surfaces [4, p. 159] are complex surfaces of general type constructed from two orientable regular hypermaps of genus at least 2, with the same automorphism groups. A finite group which arises as an automorphism group linked to a Beauville surface is called a Beauville group.

A group-theoretical formulation of Beauville groups can be given as follows. Given a finite group G and a pair of elements x,yG, we define

Σ(x,y)=gG(xgygxyg),

that is, the union of all subgroups of G which are conjugate to x, to y or to xy. Then G is a Beauville group if and only if the following conditions hold:

  1. G is a 2-generator group.

  2. There exists a pair of generating sets {x1,y1} and {x2,y2} of G such that Σ(x1,y1)Σ(x2,y2)=1.

Then {x1,y1} and {x2,y2} are said to form a Beauville structure for G.

By using this characterization, one can study whether a given finite group is a Beauville group. Since the beginning of this century, there has been a considerable interest in Beauville groups; see, for example, the recent survey papers [5, 8, 17]. In 2000, Catanese [7] showed that a finite abelian group is a Beauville group if and only if it is isomorphic to Cn×Cn, with n>1 and gcd(n,6)=1. On the other hand, a remarkable result, proved independently by Guralnick and Malle [15] and by Fairbairn, Magaard and Parker [9] in 2012, is that every non-abelian finite simple group other than A5 is a Beauville group.

If p is a prime, not much was known about Beauville p-groups until very recently (see [1] and [5]). For p=2 and 3, early examples of Beauville p-groups were given in [12] where the orders of the groups are 212 and 312. Even earlier, in [3] Bauer, Catanese and Grunewald constructed a Beauville 2-group of order 27. In [1] Barker, Boston and Fairbairn classified Beauville p-groups of order at most p4 and they found estimates for the number of Beauville groups of orders p5 and p6. They also determined the smallest Beauville p-group for every prime p.

Also in [1], it was shown that there are non-abelian Beauville p-groups of order pn for every p5 and every n3. Recently in [20], Stix and Vdovina constructed infinite series of Beauville p-groups, for every prime p, by considering quotients of ordinary triangle groups (von Dyck groups). In particular, this gives examples of non-abelian Beauville p-groups of arbitrarily large order. On the other hand, the first explicit infinite family of Beauville 2-groups was constructed in [2].

In [11], Fernández-Alcober and Gül extended Catanese’s criterion in the case of p-groups from abelian groups to a much wider family of groups, including all p-groups having a “good behavior” with respect to taking powers, and in particular groups of class at most p-1. As an application of this result, they determined all Beauville p-groups of order p5, for p5, and of order p6, for p7. Also, they extended the characterization of split metacyclic p-groups given in [20] to all metacyclic p-groups.

On the other hand, by [1] it is known that the proportion of 2-generator groups of order p5 which are Beauville tends to 1 when p. Later on, in [11] it has been shown that the proportion tends to 25 when p for groups of order p6. However, the question of what proportion of 2-generator p-groups are Beauville groups is still open.

A non-cyclic finite p-group G is said to be thin if every normal subgroup of G lies between two consecutive terms of the lower central series and for all i1, |γi(G):γi+1(G)|p2. Indeed, these groups are 2-generator. Thus it is natural to ask whether they are Beauville or not. Furthermore, the study of thin p-groups is also motivated by the fact that they usually give examples of groups whose power structures are not so well-behaved. Well-known examples of thin p-groups are p-groups of maximal class and quotients of the Nottingham group. In [11], all Beauville quotients of the Nottingham group were determined. Thanks to the ill-behaved power structure, an infinite family of Beauville 3-groups was given by considering quotients of the Nottingham group over 𝔽3. In [10], the existence of Beauville structures in the most important families of groups of maximal class, in particular in metabelian p-groups of maximal class, was studied.

The goal of the present paper is to complete the study of Beauville structures in metabelian thin p-groups. In this paper, we will exclude p-groups of maximal class from our consideration of thin groups. This means, in particular, that the prime 2 is excluded [16, Theorem III.11.9]. Then according to [6, Theorem A], if G is a metabelian thin p-group, then cl(G)p+1. If the class is <p, then the existence of Beauville structures can be determined by using [11, Corollary 2.6]. Thus we restrict to groups of class p or p+1. The main result of this paper is as follows.

Theorem A.

Let G be a metabelian thin p-group of class p or p+1 for p5. Then there are four cases in which there is a Beauville structure:

  1. cl(G)=p and |γp(G)|=p2.

  2. cl(G)=p+1.

  3. cl(G)=p, |γp(G)|=p and Gp=γp-1(G).

  4. cl(G)=p, |γp(G)|=p, Gp=γp(G) and G has at least three maximal subgroups of exponent p.

Notation.

We use standard notation in group theory. If G is a group, H,KG and NG, then HK(modN) means HN/N=KN/N. If p is a prime, then we write Gpi for the subgroup generated by all powers gpi as g runs over G, and Ωi(G) for the subgroup generated by the elements of G of order at most pi. The exponent of G, denoted by expG, is the maximum of the orders of all elements of G.

2 Proof of the main result

In this section, we give the proof of Theorem A. We begin by giving some properties of metabelian thin p-groups. Firstly, we recall the following more general result of Meier-Wunderli (see [18, Theorem 3]). If G is a metabelian 2-generator p-group, then

(2.1)Gpγp(G).

Observe that the only thin abelian p-group is the elementary abelian group of order p2 and we refer to its lattice of normal subgroups as a diamond. It then follows that if G is thin, then G/G is elementary abelian of order p2, and hence G is 2-generator. Also, the lower and upper central series of a thin p-group coincide [6, Corollary 2.2].

Note that if G is a thin p-group and gγi(G)γi+1(G), then (see [6, Lemma 2.1])

(2.2)γi+1(G)=[g,G]γi+2(G).

Let G be a metabelian thin p-group. By [6, Theorem A], we have the following:

  1. γp+1(G) is cyclic and γp+2(G)=1.

  2. The lattice of normal subgroups of G consists of a diamond on top, followed by a chain of length 1, at most p-2 diamonds, plus possibly another chain of length 1.

As a consequence, cl(G)p+1 and |G|p2p. We next recall the power structure of a metabelian thin p-group.

Lemma 2.1.

Let G be a metabelian thin p-group, and let l be the largest integer such that Gpγl(G). Then 3lp, γl+1(G) is cyclic, γl+2(G)=1 and γ2(G)pγl+1(G).

Proof.

See [6, proofs of Lemmas 1.2, 1.3 and 3.3]. ∎

The next corollary follows directly from Lemma 2.1.

Corollary 2.2.

Let G be a metabelian thin p-group. Then |Gp|p3.

Lemma 2.3.

Let G be a metabelian thin p-group such that its lattice of normal subgroups ends with a chain, i.e. monolithic. Then the order of Gp cannot be p2.

Proof.

If G is of class p+1, then Gp=γp(G), and hence |Gp|=p3. Thus we assume that cl(G)=cp. Next observe that if M is a maximal subgroup of G, then cl(M)<cp, and so M is regular.

Now suppose, on the contrary, that |Gp|=p2. Consider the quotient group G¯=G/γc(G), which is regular. Then |G¯p|=p, and hence |G¯:Ω1(G¯)|=p. Write Ω1(G¯)=M/γc(G) for some maximal subgroup M of G. Since G¯ is regular, it follows that expΩ1(G¯)=p, and hence Mpγc(G). This implies that |Mp|=|M:Ω1(M)|p as M is regular. It then follows that GΩ1(M), and thus expG=p.

On the other hand, if M is an arbitrary maximal subgroup of G, we have GM and since expG=p, we obtain GΩ1(M)M. It follows that |Mp|=|M:Ω1(M)|p. Since G is thin, this implies that Mpγc(G) for any maximal subgroup M. But Gp=MpM maximal in Gγc(G). Thus |Gp|p, which is a contradiction. ∎

We finally recall a commutator relation between the generators of G. More specifically, if G is a metabelian thin p-group, then to every xGG there corresponds a y such that G=x,y and

(2.3)[y,x,x,x][y,x,y,y]h(modγ5(G)),

where h is a quadratic non-residue modulo p (see [6, Theorem B]).

Before we proceed to prove Theorem A, we will determine which metabelian thin 3-groups are Beauville groups.

Theorem 2.4.

A metabelian thin 3-group is a Beauville group if and only if it is one of SmallGroup(35,3), SmallGroup(36,34), or SmallGroup(36,37).

Proof.

Let G be a metabelian thin 3-group. Then |G|36. Note that the smallest Beauville 3-group S=𝚂𝚖𝚊𝚕𝚕𝙶𝚛𝚘𝚞𝚙(35,3) is of order 35, and it is the only Beauville 3-group of that order [1]. Furthermore, by using the computer algebra system GAP [21], it can be seen that S is metabelian thin. Thus, if |G|=35, then G is a Beauville group if and only if GS. We next assume that |G|=36. It has been shown in [1] that there are only three Beauville 3-groups of order 36, namely S=𝚂𝚖𝚊𝚕𝚕𝙶𝚛𝚘𝚞𝚙(36,n) for n=34,37,40. However, if n=40, then S is not thin since |Z(S)|=9 and thus Z(S)γ4(S). On the other hand, if n=34 or 37, then again by using the computer algebra system GAP, one can show that S is a metabelian thin 3-group. Consequently, G is a Beauville group if and only if GS for n=34 or 37. ∎

Thus we assume that p5. Let G be a metabelian thin p-group with cl(G)=p or p+1. Then we have three cases: G is of class p+1, or is of class p and |γp(G)|=p2, or is of class p and |γp(G)|=p. In the first two cases, we have Gpγp(G). It then follows from (2.1) that Gp=γp(G). Note that we also have γ2(G)pγp+1(G), by Lemma 2.1. On the other hand, in the last case if l is the largest integer satisfying Gpγl(G), then l=p-1 or p and hence γp(G)Gpγp-1(G).

Our first step is to calculate the pth powers of xty modulo γp+1(G) for all 0tp-1 if G=x,y and γ2(G)pγp+1(G). To this purpose, we need the following lemma.

Lemma 2.5 ([19, Lemma 6]).

Let G be a metabelian p-group and x,yG. Set σ1=y and σi=[σi-1,x] for i2. Then

(xy)p=xpypσ2(p2)σp(pp)z,

where

z=i=1p-1j=1p-1[σi+1,jσ1]C(i,j)

and

C(i,j)=k=1p-1(ki)(kj).

Lemma 2.6.

Let G be a metabelian thin p-group such that γ2(G)pγp+1(G). If x and y are the generators of G satisfying (2.3), then for all 0tp-1,

(2.4)(xty)p(xp)typ[y,x,p-2y]-2t1-ht2[y,x,p-3y,x]2t21-ht2(modγp+1(G)).

Proof.

By Lemma 2.5, we have

(xty)p=(xp)typ[y,xt](p2)[y,p-1xt](pp)i=1p-1j=1p-1[y,ixt,jy]C(i,j).

Since γ2(G)pγp+1(G), it then follows that

(xty)p(xp)typ[y,p-1xt]1i,ji+jp-1[y,ixt,jy]C(i,j)(modγp+1(G)).

Note that for i+j>0, C(i,j) is the coefficient of uivj in k=0p-1(1+u)k(1+v)k, where

k=0p-1(1+u)k(1+v)k((u+v)+uv)p-1(modp).

In the previous expression the monomials of total degree less than p appear only in (u+v)p-1r=0p-1(-1)rurvp-r-1(modp), and hence

C(i,j){0(modp)if i+j<p-1,(-1)i(modp)if i+j=p-1

Thus the condition γ2(G)pγp+1(G) implies that

(xty)p(xp)typi=1p-1[y,ixt,p-i-1y](-1)i(modγp+1(G)).

On the other hand, notice that for 1tp-1,

(2.5)[y,xt,xt,xt][y,xt,y,y]ht2(modγ5(G)).

Since G is metabelian, for every a,bG and cG we have [c,a,b]=[c,b,a], and this, together with (2.5), yields that

[y,ixt,p-i-1y](-1)i{[y,xt,p-2y]-(ht2)s-1(modγp+1(G)),i=2s-1,(ht2)s-1(modγp+1(G)),i=2s,

and hence (modulo γp+1(G))

(xty)p(xp)typ([y,x,p-2y]-t[y,x,p-3y,x]t2)s=1(p-1)/2(ht2)s-1.

Note that since h is a quadratic non-residue, we have

s=1p-12(ht2)s-1=21-ht2.

Consequently, we get

(xty)p(xp)typ[y,x,p-2y]-2t1-ht2[y,x,p-3y,x]2t21-ht2(modγp+1(G))

for 0tp-1, as desired. ∎

Lemma 2.7.

Let G be a metabelian thin p-group such that |γp(G)|p2 and let x and y be the generators of G satisfying (2.3). Then for every t0{0,1,p-1} there exist at most three distinct t{0,1,p-1} such that

(2.6)(xt0y)p(xty)p(modγp+1(G)).

Proof.

Since |γp(G)|p2, it follows that Gp=γp(G) and γ2(G)pγp+1(G), by Lemma 2.1. Also note that |γp(G):γp+1(G)|=p2. Notice that, as a consequence of (2.2), we have that l=[y,x,p-2y] and m=[y,x,p-3y,x] are linearly independent modulo γp+1(G). Thus (l,m) is a basis of γp(G) modulo γp+1(G). If we set xplαmβ(modγp+1(G)) and yplγmδ(modγp+1(G)) for some α,β,γ,δ𝔽p, then by (2.4), we have

(xty)plγ+αt-2t1-ht2mδ+βt+2t21-ht2(modγp+1(G)).

Observe that as rational functions in t, neither

f(t)=γ+αt-2t1-ht2

nor

g(t)=δ+βt+2t21-ht2

are zero.

We now fix t0{0,1,p-1}. Then (2.6) holds if and only if there exists λ𝔽p* such that

f(t)=λf(t0)andg(t)=λg(t0).

If f(t0)=0 or g(t0)=0, then we have f(t)=0 or g(t)=0, that is,

(1-ht2)(γ+αt)-2t=0

or

(1-ht2)(δ+βt)+2t2=0.

Otherwise, we have f(t)f(t0)=g(t)g(t0). Then g(t0)f(t)-f(t0)g(t)=0, that is,

g(t0)((1-ht2)(γ+αt)-2t)-f(t0)((1-ht2)(δ+βt)+2t2)=0,

which is a polynomial in t of degree 3. Thus in every case, there are at most three distinct t{0,1,p-1} such that (xt0y)p(xty)p(modγp+1(G)). ∎

Lemma 2.8.

Let G be a metabelian thin p-group such that γ2(G)pγp+1(G). If M is a maximal subgroup of G and a,bMG, then

apbp(modγp+1(G)).

Proof.

If we write b=aic for some cG and for some integer i not divisible by p, then by the Hall–Petrescu collection formula (see [16, III.9.4]), we have

(aic)p=apicpc2(p2)c3(p3)cp,

where cjγj(a,c)γj+1(G). Thus we have (aic)papi(modγp+1(G)), and hence apbp(modγp+1(G)). ∎

Remark 2.9.

If we replace x with x*, where x*GG is not a power of x, there exists a corresponding y* satisfying (2.3). Then x(x*)t0y*,GG for some 0t0p-1, and according to Lemma 2.8, we have

xp((x*)t0y*)p(modγp+1(G)).

It follows from Lemma 2.7 that there exist at most three distinct t{0,1,p-1} such that xp(xty)p(modγp+1(G)).

The following corollary is an immediate consequence of Lemmas 2.7 and 2.8.

Corollary 2.10.

Let G be a metabelian thin p-group such that |γp(G)|p2. If M is a maximal subgroup of G, then there exist at most two maximal subgroups M1, M2 different from M such that MpM1pM2p(modγp+1(G)).

Before we present the main result, we also need the following remark.

Remark 2.11.

Let G be a finite 2-generator p-group. Then we can always find elements x,yGΦ(G) such that x,y and xy fall into the given three maximal subgroups of G. Let M1, M2 and M3 be three maximal subgroups of G. Choose the elements xM1Φ(G) and yM2Φ(G). Since each element in the set {xyj1jp-1} falls into different maximal subgroups, there exists 1jp-1 such that xyjM3Φ(G). Thus if we put x*=x and y*=yj, then elements in the triple {x*,y*,x*y*} fall into the given three maximal subgroups.

We are now ready to prove Theorem A. We deal separately with the cases in the theorem.

Theorem 2.12.

Let G be a metabelian thin p-group with cl(G)=p such that |γp(G)|=p2, where p5. Then G has a Beauville structure in which one of the two triples has all elements of order p2.

Proof.

We divide our proof into three cases depending on the number of maximal subgroups whose pth powers coincide, and in every case, we take into account Corollary 2.10 and Remark 2.11. First of all, note that since G has at most three maximal subgroups of exponent p and p5, there are at least three maximal subgroups of exponent p2.

Case 1. Assume that there is a one-to-one correspondence between maximal subgroups Mi of exponent p2 and Mip. Choose a set of generators {x1,y1} such that o(x1)=o(y1)=o(x1y1)=p2.

Case 2. Assume that there exist three maximal subgroups of exponent p2 such that their pth power subgroups coincide. Then choose a set of generators {x1,y1} such that x1,y1 and x1y1 fall into those maximal subgroups.

In both Case 1 and Case 2, since p5, we can choose another set of generators {x2,y2} so that each pair of elements in {xi,yi,xiyii=1,2} is linearly independent modulo G by Remark 2.11.

Case 3. Assume that we are not in the first two cases. Then there exist two maximal subgroups M1, M2 of exponent p2 such that M1p=M2p and MpM1p for all other maximal subgroups M.

Let us first deal with p7. We start by choosing a set of generators {x1,y1}, where x1M1 and y1M2 are such that o(x1y1)=p2, say x1y1M3. Then there might be a maximal subgroup M4 such that M3p=M4p (note that there is no other i3,4 satisfying Mip=M3p, otherwise we are in Case 2). Since p7, we can choose another set of generators {x2,y2} so that x2,y2,x2y2M4 and each pair of elements in {xi,yi,xiyii=1,2} is linearly independent modulo G, by Remark 2.11.

If p=5, then by using the construction of metabelian thin p-groups in [6] and the computer algebra system GAP, one can show that there is no metabelian thin 5-group of class 5 such that |γ5(G)|=52 and fifth powers of maximal subgroups coincide in pairs. Thus in Case 3, there exists a maximal subgroup, say M3, of exponent 52, where M5M35 for all maximal subgroups M with MM3. Then choose sets of generators {x1,y1} and {x2,y2} so that x1M1, y1M2 and x1y1M3 and each pair of elements in {xi,yi,xiyii=1,2} is linearly independent modulo G.

We claim that, in every case, {x1,y1} and {x2,y2} form a Beauville structure for G. If A={x1,y1,x1y1} and B={x2,y2,x2y2}, then we need to show that

(2.7)agbh=1

for all aA, bB and g,hG. Note that o(a)=p2 for every aA. Assume first that o(b)=p. If agbh1 for some g,hG, then bhag, and hence aG=bG, which is a contradiction, since a and b are linearly independent modulo G. Thus we assume that o(b)=p2. If (2.7) does not hold, then (ag)p=(bh)p, which contradicts the choice of b. ∎

In order to deal with the case cl(G)=p+1, we need the following lemma.

Lemma 2.13 ([13, Lemma 4.2]).

Let G be a finite group and let {x1,y1} and {x2,y2} be two sets of generators of G. Assume that, for a given NG, the following hold:

  1. {x1N,y1N} and {x2N,y2N} is a Beauville structure for G/N,

  2. o(u)=o(uN) for every u{x1,y1,x1y1}.

Then {x1,y1} and {x2,y2} is a Beauville structure for G.

Theorem 2.14.

Let G be a metabelian thin p-group with cl(G)=p+1, where p5. Then G has a Beauville structure.

Proof.

By Theorem 2.12, G¯=G/γp+1(G) has a Beauville structure in which one of the two triples has all elements of order p2, i.e. they have the same order in both G and G¯. Then we can apply Lemma 2.13 and thus G is a Beauville group. ∎

We next analyze the case cl(G)=p and |γp(G)|=p. Recall that we have γp(G)Gpγp-1(G), and thus there are two possibilities:

  1. Gp=γp-1(G),

  2. Gp=γp(G).

Observe that by Lemma 2.3, Gp cannot be a proper subgroup of γp-1(G) of order p2.

Theorem 2.15.

Let G be a group in case (i). Then G has a Beauville structure.

Proof.

First of all, notice that there exists a pair of generators a and b of G such that ap and bp are linearly independent modulo γp(G). By the Hall–Petrescu formula, we have

(atb)p=atpbpc2(p2)cp(pp),

where cjγj(at,b). Since γ2(G)pγp(G), by Lemma 2.1, we get

(atb)patpbp(modγp(G))

for 1tp-1. Observe that, similarly to Lemma 2.8, for every maximal subgroup M, mM and cG , we have (mc)pmp(modγp(G)). It then follows that the power subgroups Mp are all distinct modulo γp(G).

On the other hand, since G¯=G/γp(G) is of class p-1, it is a regular p-group such that |G¯p|=p2. According to [11, Corollary 2.6], G¯ is a Beauville group since p5. From the observation above, all elements outside G are of order p2 in both G and G¯. Then we can apply Lemma 2.13 to conclude that G is a Beauville group. ∎

Theorem 2.16.

Let G be a group in case (ii). Then G has a Beauville structure if and only if it has at least three maximal subgroups of exponent p.

Proof.

If the number of maximal subgroups of exponent p is less than three, then Ω1(G) is contained in the union of at most two maximal subgroups. Since |Gp|=p, it then follows from [11, Proposition 2.4] that G has no Beauville structure.

On the other hand, if at least three maximal subgroups have exponent p, then we choose a triple in which all elements have order p. Since p5, we can choose another triple such that each pair of elements in the union of the two triples is linearly independent modulo G. Then G has a Beauville structure. ∎

An example of a metabelian thin p-group having at least three maximal subgroups of exponent p can be found in [14, pp. 63–67].


Communicated by Nigel Boston


Award Identifier / Grant number: MTM2014-53810-C2-2-P

Funding source: Basque Government

Award Identifier / Grant number: IT974-16

Funding statement: The second author is supported by the Spanish Government, grant MTM2014-53810-C2-2-P, and the Basque Government, grant IT974-16. The first author is a member of INdAM-GNSAGA.

Acknowledgements

We would like to thank Carlo M. Scoppola for helpful discussions and suggestions. Also we wish to thank G. Fernández-Alcober for valuable comments. The second author would like to thank the Department of Mathematics at the University of L’Aquila for its hospitality while this research was conducted.

References

[1] N. Barker, N. Boston and B. Fairbairn, A note on Beauville p-groups, Exp. Math. 21 (2012), no. 3, 298–306. 10.1080/10586458.2012.669267Search in Google Scholar

[2] N. Barker, N. Boston, N. Peyerimhoff and A. Vdovina, An infinite family of 2-groups with mixed Beauville structures, Int. Math. Res. Not. IMRN 2015 (2015), no. 11, 3598–3618. 10.1093/imrn/rnu045Search in Google Scholar

[3] I. C. Bauer, F. Catanese and F. Grunewald, The classification of surfaces with pg=q=0 isogenous to a product of curves, Pure Appl. Math. Q. 4 (2008), no. 2, 547–586. 10.4310/PAMQ.2008.v4.n2.a10Search in Google Scholar

[4] A. Beauville, Surfaces algébriques complexes, Astérisque 54, Société Mathématique de France, Paris, 1978. Search in Google Scholar

[5] N. Boston, A survey of Beauville p-groups, Beauville Surfaces and Groups, Springer Proc. Math. Stat. 123, Springer, Cham (2015), 35–40. 10.1007/978-3-319-13862-6_3Search in Google Scholar

[6] R. Brandl, A. Caranti and C. M. Scoppola, Metabelian thin p-groups, Quart. J. Math. Oxford Ser. (2) 43 (1992), no. 170, 157–173. 10.1093/qmath/43.2.157Search in Google Scholar

[7] F. Catanese, Fibred surfaces, varieties isogenous to a product and related moduli spaces, Amer. J. Math. 122 (2000), no. 1, 1–44. 10.1353/ajm.2000.0002Search in Google Scholar

[8] B. Fairbairn, Recent work on Beauville surfaces, structures and groups, Groups St Andrews 2013, London Math. Soc. Lecture Note Ser. 422, Cambridge University Press, Cambridge (2015), 225–241. 10.1017/CBO9781316227343.014Search in Google Scholar

[9] B. Fairbairn, K. Magaard and C. Parker, Generation of finite quasisimple groups with an application to groups acting on Beauville surfaces, Proc. Lond. Math. Soc. (3) 107 (2013), no. 4, 744–798. 10.1112/plms/pds097Search in Google Scholar

[10] G. A. Fernández-Alcober, N. Gavioli, Ş. Gül and C. M. Scoppola, Beauville p-groups of wild type and groups of maximal class, preprint (2017), https://arxiv.org/abs/1701.07361. Search in Google Scholar

[11] G. A. Fernández-Alcober and Ş. Gül, Beauville structures in finite p-groups, J. Algebra 474 (2017), 1–23. 10.1016/j.jalgebra.2016.11.007Search in Google Scholar

[12] Y. Fuertes, G. González-Diez and A. Jaikin-Zapirain, On Beauville surfaces, Groups Geom. Dyn. 5 (2011), no. 1, 107–119. 10.4171/GGD/117Search in Google Scholar

[13] Y. Fuertes and G. A. Jones, Beauville surfaces and finite groups, J. Algebra 340 (2011), 13–27. 10.1016/j.jalgebra.2011.04.027Search in Google Scholar

[14] Ş. Gül, Beauville structure in p-groups, Ph.D. thesis, Middle East Technical University, 2016. Search in Google Scholar

[15] R. Guralnick and G. Malle, Simple groups admit Beauville structures, J. Lond. Math. Soc. (2) 85 (2012), no. 3, 694–721. 10.1112/jlms/jdr062Search in Google Scholar

[16] B. Huppert, Endliche Gruppen. I, Grundlehren Math. Wiss. 134, Springer, Berlin, 1967. 10.1007/978-3-642-64981-3Search in Google Scholar

[17] G. A. Jones, Beauville surfaces and groups: A survey, Rigidity and Symmetry, Fields Inst. Commun. 70, Springer, New York (2014), 205–225. 10.1007/978-1-4939-0781-6_11Search in Google Scholar

[18] H. Meier-Wunderli, Metabelsche Gruppen, Comment. Math. Helv. 25 (1951), 1–10. 10.1007/BF02566442Search in Google Scholar

[19] R. J. Miech, Metabelian p-groups of maximal class, Trans. Amer. Math. Soc. 152 (1970), 331–373. 10.2307/1995575Search in Google Scholar

[20] J. Stix and A. Vdovina, Series of p-groups with Beauville structure, Monatsh. Math. 181 (2016), no. 1, 177–186. 10.1007/s00605-015-0805-9Search in Google Scholar

[21] The GAP Group, GAP- Groups, Algorithms, and Programming, Version 4.7.5, 2014, http://www.gap-system.org. Search in Google Scholar

Received: 2016-3-16
Revised: 2017-10-16
Published Online: 2017-11-17
Published in Print: 2018-3-1

© 2018 Walter de Gruyter GmbH, Berlin/Boston

Downloaded on 1.6.2024 from https://www.degruyter.com/document/doi/10.1515/jgth-2017-0040/html
Scroll to top button