Arxiv:2010.15657V1 [Math.NT] 29 Oct 2020

LOW-DEGREE PERMUTATION RATIONAL FUNCTIONS OVER FINITE FIELDS

ZHIGUO DING AND MICHAEL E. ZIEVE

Abstract. We determine all degree-4 rational functions f(X) ∈ Fq(X) 1 which permute P (Fq), and answer two questions of Ferraguti and Micheli about the number of such functions and the number of equivalence classes of such functions up to composing with degree-one rational functions. We also determine all degree-8 rational functions f(X) ∈ Fq(X) 1 which permute P (Fq) in case q is suﬃciently large, and do the same for degree 32 in case either q is odd or f(X) is a nonsquare. Further, for several other positive integers n < 4096, for each suﬃciently large q we determine all degree-n rational functions f(X) ∈ Fq(X) which permute 1 P (Fq) but which are not compositions of lower-degree rational functions in Fq(X). Some of these results are proved by using a new Galois- theoretic characterization of additive (linearized) polynomials among all rational functions, which is of independent interest.

1. Introduction Let q be a power of a prime p.A permutation polynomial is a polynomial f(X) ∈ Fq[X] for which the map α 7→ f(α) is a permutation of Fq. Such polynomials have been studied both for their own sake and for use in various applications. Much less work has been done on permutation rational functions, namely rational functions f(X) ∈ Fq(X) which permute 1 P (Fq) := Fq ∪ {∞}. However, the topic of permutation rational functions seems worthy of study, both because permutation rational functions have the same applications as permutation polynomials, and because of the con- struction in [24] which shows how to use permutation rational functions over Fq to produce permutation polynomials over Fq2 . Surprisingly little is known about permutation rational functions: for instance, the recent paper arXiv:2010.15657v1 [math.NT] 29 Oct 2020 [5] contains the first classification result in this subject, namely a classification of degree-3 permutation rational functions. Very recently [15] classified degree-4 permutation rational functions. In this paper we give much sim- pler proofs of these results, and classify permutation rational functions in many other degrees when certain additional conditions hold. In particular, we answer [5, Problems 9.1 and 9.2].

Date: October 30, 2020. The authors thank the referees for helpful comments, and in particular for informing them of [15]. The second author thanks the National Science Foundation for support under grant DMS-1601844. 1 2 ZHIGUO DING AND MICHAEL E. ZIEVE

Recall that the degree of a nonconstant rational function f(X) ∈ Fq(X) is max(deg a, deg b) for any coprime a, b ∈ Fq[X] such that f = a/b. The statements of our results use the following terminology:

Definition 1.1. We say that nonconstant f, g ∈ Fq(X) are equivalent if f = µ ◦ g ◦ ν for some degree-one µ, ν ∈ Fq(X). 1 Plainly if f, g ∈ Fq(X) are equivalent then f permutes P (Fq) if and 1 only if g permutes P (Fq). For completeness, we begin with the following essentially immediate result. 1 Lemma 1.2. Every degree-one f(X) ∈ Fq(X) permutes P (Fq). A degree- 1 two f(X) ∈ Fq(X) permutes P (Fq) if and only if q is even and f(X) is equivalent to X2. The following result is a more conceptual version of [5, Thms. 5.1 and 6.2]. 1 Theorem 1.3. A degree-three f(X) ∈ Fq(X) permutes P (Fq) if and only if it is equivalent to one of the following: (1) X3 where q ≡ 2 (mod 3) −1 3 (2) ν ◦ X ◦ ν where q ≡ 1 (mod 3) and for some δ ∈ Fq2 \ Fq we have ν(X) = (X − δq)/(X − δ) and ν−1(X) = (δX − δq)/(X − 1) 3 (3) X − αX where 3 | q and either α = 0 or α is a nonsquare in Fq. Remark. The functions ν, ν−1 in case (2) satisfy ν−1 ◦ν = X = ν ◦ν−1, and 1 also ν(P (Fq)) is the set Λ of (q +1)-th roots of unity in Fq2 [24, Lemma 3.1]. −1 n 1 n Thus ν ◦ X ◦ ν permutes P (Fq) if and only if X permutes Λ, i.e., −1 n (n, q + 1) = 1. Moreover, ν ◦ X ◦ ν is in Fq(X)[4]. These functions ν−1 ◦ Xn ◦ ν are instances of the general class of Rédeifunctions; see [4]. Remark. The functions in (1) and (3) of Theorem 1.3 are members of well- n known classes of permutation polynomials. Specifically, X permutes Fq when (n, q − 1) = 1. An additive polynomial is a polynomial of the form Pr pi i=0 αiX with αi ∈ Fq and p := char(Fq); any such polynomial permutes Fq if and only if it has no nonzero roots in Fq. (Additive polynomials are sometimes called “linearized polynomials” or “p-polynomials”.) Theorem 1.3 was proved in [5] via a long and complicated argument in- volving computer calculations of prime components of certain ideals, among other things. We give two very short non-computational proofs of Theo- rem 1.3 using different methods than [5]; we hope that the new understand- ing provided by these proofs will help readers adapt our methods to address further questions. We then treat the much more difficult case of degree-4 permutation rational functions, which cannot be classified using the methods of [5]. 1 Theorem 1.4. A degree-four f(X) ∈ Fq(X) permutes P (Fq) if and only if one of the following holds. LOW-DEGREE PERMUTATION RATIONAL FUNCTIONS 3

(1) q is odd and f(X) is equivalent to X4 − 2αX2 − 8βX + α2 X3 + αX + β 3 for some α, β ∈ Fq such that X + αX + β is irreducible in Fq[X] 4 2 (2) q is even and f(X) is equivalent to X +αX +βX for some α, β ∈ Fq 3 ∗ such that X + αX + β has no roots in Fq (3) q ≤ 8 and f(X) is equivalent to a rational function in Table 1 on page 19. Moreover, f(X) is exceptional (cf. Definition 1.5) if and only if (1) or (2) holds. Remark. The functions in (2) of Theorem 1.4 are additive polynomials. The functions in (1) are a new class of permutation rational functions (although the characteristic zero analogue of these functions appears in [11, Thm. 7.1]). At the end of Section5 we will give a single form which com- bines (1) and (2). Remark. The referee kindly informed us that Hou has independently and simultaneously classified degree-4 permutation rational functions [15]. We note that Hou’s result lists more functions than Theorem 1.4: in addition to the functions in Theorem 1.4, Hou lists a family of examples over each finite field of characteristic 3, and many additional low-degree examples. Our result implies that Hou’s characteristic 3 families are equivalent to special cases of (1) of Theorem 1.4, and that all of Hou’s low-degree examples are equivalent to the low-degree examples in Table 1. Hou’s proof is quite long and computational, while ours is much shorter and more conceptual, using a completely different approach based on Galois theory. Moreover, it does not seem to be substantially quicker to deduce Theorem 1.4 from Hou’s result than to prove Theorem 1.4 directly. We apply our Theorem 1.4 to answer [5, Problems 9.1 and 9.2] about the number of degree-4 permutation rational functions in Fq(X) and the number of equivalence classes of such functions; these questions cannot be resolved easily from Hou’s result.

Definition 1.5. A rational function f(X) ∈ Fq(X) is exceptional if f(X) 1 permutes P (Fq` ) for infinitely many integers `. 1 1 Since bijectivity of f(X) on P (Fq` ) implies bijectivity of f(X) on P (Fq), 1 we see that every exceptional rational function in Fq(X) permutes P (Fq). The following quantitative converse was proved in [13, Thm. 2.5]: 1 Lemma 1.6. If f(X) ∈ Fq(X) has degree n ≥ 2 and permutes P (Fq), where √ q > 2(n − 2)2 + 1, then f(X) is exceptional. This inequality holds in particular when q ≥ 4n4. In light of Lemma 1.6, for any fixed n the study of degree-n permutation rational functions over Fq reduces to the study of exceptional rational 4 ZHIGUO DING AND MICHAEL E. ZIEVE functions for all but finitely many values of q. Our next result classifies exceptional rational functions of degree 8. Note that any rational function equivalent to an exceptional rational function is exceptional.

Theorem 1.7. If f(X) ∈ Fq(X) is exceptional and deg(f) = 8 then q is even and f(X) is equivalent to an additive polynomial. The same conclusion holds if f(X) ∈ Fq(X) is a degree-8 permutation rational function with q > 732. The following result classiﬁes exceptional rational functions of degree 32, modulo the classiﬁcation of degree-16 exceptional rational functions in characteristic 2.

Theorem 1.8. A degree-32 rational function f(X) ∈ Fq(X) is exceptional if and only if q is even and f(X) is equivalent to either 2 (1) g(X ) for some exceptional g ∈ Fq(X) of degree 16, or (2) L1 ◦ µ ◦ L2 for some degree-one µ ∈ Fq(X) and some exceptional additive L1,L2 ∈ Fq[X]. Our final results address indecomposable exceptional rational functions, which are defined as follows. Definition 1.9. For any field K, a rational function f(X) ∈ K(X) of degree at least 2 is indecomposable if it cannot be written as g(h(X)) with g, h ∈ K(X) of degree at least 2. The importance of indecomposability comes from the following classical result [6, Thm. 1]:

Lemma 1.10. A rational function f(X) ∈ Fq(X) of degree at least 2 is exceptional if and only if f = g1 ◦ g2 ◦ · · · ◦ gr for some indecomposable exceptional gi ∈ Fq(X). In light of Lemma 1.10, in order to classify exceptional rational functions it suﬃces to classify indecomposable exceptional rational functions.

Theorem 1.11. If f(X) ∈ Fq(X) is an exceptional rational function of degree 128 then q is even. If in addition f(X) is indecomposable then f(X) is equivalent to an additive polynomial. In Theorem 1.12 we will prove similar results for degrees that are not prime powers, which turns out to be an easier situation to address via our methods. In the case of prime power degree, the degrees 2, 3, 4, 8, 32, and 128 in the above results are the only degrees for which our method of proof yields a conclusion of the desired form. It seems conceivable that one might be able to classify prime-degree exceptional rational functions by further developing the approach used for polynomials in [9, Thm. 8.1] and [17, App.], but we do not pursue that problem here. However, with current techniques it seems quite diﬃcult to classify degree-9 exceptional rational functions, for instance. We note that there exist indecomposable additive LOW-DEGREE PERMUTATION RATIONAL FUNCTIONS 5 exceptional polynomials of any prescribed prime power degree, and there exist other types of indecomposable exceptional rational functions in many classes of prime power degrees. Theorem 8.3 of [5] asserts that there are no exceptional rational functions of degree 6 with nonzero derivative. By Lemmas 1.10 and 1.2, a degree-6 0 exceptional f(X) ∈ Fq(X) is indecomposable if and only if f (X) 6= 0, so the nontrivial portion of [5, Thm. 8.3] is the assertion that there are no indecomposable exceptional rational functions of degree 6. We prove the following vast generalization of this result:

Theorem 1.12. Suppose f(X) ∈ Fq(X) is an indecomposable exceptional rational function whose degree n satisfies n < 4096 and n is not a prime power. Then n ∈ {28, 45, 325, 351, 496, 784, 819, 1225, 1456, 2025, 3321}. Remark. For the benefit of the reader, we mention several previous results which were overlooked by the authors of [5]. The result [5, Thm. 3.2] is a weaker version of [2, Thms. 4 and 5]; the latter result is generalized in [7, Thm. 1 and Prop. 1], [9, General Exceptionality Lemma, p. 185], and [13, Thm. 1.1]. The result [5, Lemma 4.1] is a weaker version of a very special case of [13, Thm. 2.5]. Groups satisfying conditions (1)–(3) on [5, p. 872] were studied in [8,9, 11]. Many papers have explicitly produced all rational functions with prescribed monodromy groups, for instance [9, 12, 14, 16, 17, 23]. In this paper we have chosen to use elementary arguments even when shorter arguments were possible if one used more advanced tools. We did this in order to make this paper accessible to the largest possible audience, since we hope to entice members of the permutation polynomial community to study permutation rational functions. This paper is organized as follows. In the next section we quickly review background material on exceptionality and monodromy groups. In Section3 we prove Lemma 1.2 and Theorem 1.3. In Section4 we prove a characterization of additive polynomials among all rational functions, and use it to describe the indecomposable exceptional rational functions of degree 8, 32, and 128. In Section5 we prove Theorem 1.4 – which is the most difficult result in this paper, given what was known previously – and then we use this result in Section6 to answer [5, Problems 9.1 and 9.2] about the number of degree-4 permutation rational functions, and the number of equivalence classes of such functions. In Section7 we prove Theorems 1.7, 1.8, and 1.11, and finally in Section8 we prove Theorem 1.12.

2. Background material In this section we recall some basic facts and tools, giving proofs when we cannot ﬁnd suitable references.

2.1. Separable rational functions. 6 ZHIGUO DING AND MICHAEL E. ZIEVE

Definition 2.1. For any field K, a rational function f(X) ∈ K(X) is separable if f(X) ∈/ K(Xp) where p := char(K). Note that constant rational functions are not separable. We will often use without comment the following equivalent characterizations of separable rational functions: Lemma 2.2. Let K be a field, let x be transcendental over K, and pick f(X) ∈ K(X) of degree n > 0. Then [K(x): K(f(x))] = n, and the following are equivalent: (1) f(X) is separable (2) f 0(X) 6= 0 (3) the field extension K(x)/K(f(x)) is separable (4) the numerator of f(X) − t has no multiple roots in the algebraic closure of K(t), where t is transcendental over K. Proof. Write f(X) = a(X)/b(X) with coprime a, b ∈ K[X], and put t := f(x) and G(X) := a(X) − tb(X). Then G(X) ∈ K[X, t] is a degree-n polynomial in X which is not divisible by any nonconstant polynomial in K[X]. Since G(X) is also a degree-1 polynomial in t, we see that G(X) is irreducible in K[X, t], so by Gauss’s lemma it is irreducible in K(t)[X]. Since G(x) = 0, this shows that G(X) is a nonzero constant times the minimal polynomial of x over K(t), so since K(t, x) = K(x) it follows that [K(x): K(f(x))] = n. Thus (3) does not hold if and only if G(X) has multiple roots in the algebraic closure K(t) of K(t), or equivalently G(X) and G0(X) have common roots in K(t), i.e., gcd(G(X),G0(X)) 6= 1. Since gcd(G(X),G0(X)) divides G(X) in K(t)[X], and G(X) is irreducible in K(t)[X], it follows that (3) does not hold if and only if gcd(G(X),G0(X)) = G(X). Since deg(G0) < deg(G), the latter condition says that G0(X) = 0, i.e., a0(X) = tb0(X), which in turn says that a0(X) = 0 = b0(X). Writing p := char(K), it follows that (3) does not hold if and only if a, b ∈ K[Xp], i.e., if and only if f(X) is not separable. Plainly if f(X) is not separable then f 0(X) = 0. Conversely, if f 0(X) = 0 then, since f 0(X) = (b(X)a0(X) − a(X)b0(X))/b(X)2, it follows that b(X)a0(X) = a(X)b0(X); but then coprimality of a(X) and b(X) implies a(X) | a0(X), so that a0(X) = 0, and likewise b0(X) = 0, whence f(X) is not separable. The definitions immediately imply

Lemma 2.3. If f(X) ∈ Fq(X) is indecomposable but not separable then p f = µ ◦ X where p := char(Fq) and µ(X) ∈ Fq(X) has degree 1.

Every nonconstant f(X) ∈ Fq(X) has a unique expression as f(X) = pr g(X ) where p := char(Fq), r ≥ 0, and g(X) ∈ Fq(X) is separable. Since p X permutes Fq` for all `, we see that f(X) is exceptional if and only if g(X) is exceptional. Thus the study of exceptional rational functions reduces to the separable case. LOW-DEGREE PERMUTATION RATIONAL FUNCTIONS 7

2.2. Monodromy groups. For any field K and any separable f(X) ∈ K(X) of degree n > 1, let x be transcendental over K, let Ω be the Galois closure of K(x)/K(t) where t := f(x), and let L be the algebraic closure of K in Ω. Then A := Gal(Ω/K(t)) and G := Gal(Ω/L(t)) are called the arithmetic monodromy group and the geometric monodromy group of f(X), respectively. Write S for the set of conjugates of x over K(t), and put A1 := Gal(Ω/K(x)) and G1 := Gal(Ω/L(x)). We will use the following classical result, in which K denotes an algebraic closure of K: Lemma 2.4. With notation as above, G and A are transitive subgroups of Sym(S) and G is a normal subgroup of A with A/G =∼ Gal(L/K), where in addition |S| = n. In particular, if K is finite then A/G is cyclic. Moreover, (1) f(X) is indecomposable if and only if A is primitive (cf. Defini- tion 2.5) (2) If K is finite then the following are equivalent: (a) f(X) is exceptional (b) if H(X,Y ) ∈ K[X,Y ] divides the numerator of f(X) − f(Y ), and H(X,Y ) is irreducible in K[X,Y ], then H(X,Y ) equals α(X − Y ) for some α ∈ K∗ (c) A1 and G1 have exactly one common orbit on S.

Definition 2.5. A subgroup A of Sn is primitive if A is transitive and there are no groups strictly between A and the stabilizer of 1 in A. Proof. By Lemma 2.2 we have [L(x): L(t)] = n = [K(x): K(t)], so since f(x) = t it follows that the numerator of f(X) − t is irreducible in L(t)[X]. Thus the first two sentences follows from basic Galois theory. Item (2) is a weaker version of [2, Thms. 4 and 5]. Although (1) is known, we do not know a reference proving it in this setting, so we provide a proof. By the Galois correspondence, A is primitive if and only if there are no fields strictly between ΩA = K(t) and ΩA1 = K(x). By Lüroth’stheorem [21, Thm. 2], every field between K(t) and K(x) has the form K(h(x)) for some h(X) ∈ K(X), where since t ∈ K(h(x)) we have t = g(h(x)) with g ∈ K(X), so that f = g ◦ h. Conversely, if f = g ◦ h with g, h ∈ K(X) then [K(x): K(h(x))] = deg(h) and [K(h(x)) : K(t)] = deg(g). Thus f(X) is decomposable if and only if there is a field strictly between K(t) and K(x). Remark. The above proof of (1) is an algebraicization of the topological proof in [20, §II] of the analogous result over the complex numbers. 2.3. Degree-one rational functions. We will often use the following results without explicit comment. Lemma 2.6. For any field K, any degree-one µ(X) ∈ K(X) induces a 1 1 bijective function µ: P (K) → P (K). Proof. For any β ∈ K, the numerator of µ(X) − β is a nonzero polynomial whose degree is 0 if β = µ(∞) and 1 otherwise. Thus in any case β has 8 ZHIGUO DING AND MICHAEL E. ZIEVE

1 a unique µ-preimage in P (K). Also the unique µ-preimage of ∞ is ∞ if µ(X) is a polynomial, and otherwise is the unique root of the denominator of µ(X). 1 Lemma 2.7. For any ﬁeld K, any pairwise distinct α1, α2, α3 ∈ P (K), 1 and any pairwise distinct β1, β2, β3 ∈ P (K), there is a unique degree-one µ(X) ∈ K(X) such that µ(αi) = βi for each i.

Proof. In case α1 = β1 = ∞, this follows from the fact that there is a unique line through two points in K × K. To deduce the general case, pick degree- one ρ, τ ∈ K(X) with ρ(∞) = α1 and τ(β1) = ∞, and note that µ(αi) = βi if and only if ν := τ ◦ µ ◦ ρ maps the ρ-preimage of αi to τ(βi). Corollary 2.8. For any ﬁeld K, and any degree-one µ(X) ∈ K(X), there is a unique degree-one µ−1(X) ∈ K(X) such that µ−1 ◦ µ = X = µ ◦ µ−1. Proof. If µ−1 ◦ µ = X then µ−1 maps µ(α) 7→ α for each α ∈ {0, 1, ∞}. 1 1 Conversely, since deg(µ) = 1, the function µ: P (K) → P (K) is injective, so that µ(0), µ(1), and µ(∞) are pairwise distinct. By Lemma 2.7, there is a unique degree-one µ−1(X) ∈ K(X) such that µ−1(µ(α)) = α for each α ∈ {0, 1, ∞}. Then µ−1 ◦µ and µ◦µ−1 are degree-one rational functions ﬁx- ing 0, 1, ∞ and µ(0), µ(1), µ(∞), respectively, so each of these compositions agrees with X at three points and hence equals X. Remark. Explicitly, if µ(X) = (αX + β)/(γX + δ) then µ−1(X) = (δX − β)/(−γX + α).

2.4. Automorphisms of K(x). For any ﬁeld K, write ΓK for the set of degree-one rational functions in K(X), and note that ΓK is a group under the operation of functional composition, by Corollary 2.8. The following two results are immediate. Lemma 2.9. Let K be a ﬁeld, and let x be transcendental over K. For −1 each µ(X) ∈ ΓK , let σµ : K(x) → K(x) map f(x) 7→ f(µ (x)). Then σµ is in AutK (K(x)), and the map µ(X) 7→ σµ is an isomorphism ΓK → AutK (K(x)).

In light of the above result, we will identify AutK (K(x)) with ΓK when- ever convenient. We sometimes identify ΓK with PGL2(K) via the following result. Lemma 2.10. For any ﬁeld K, the map α β αX + β → γ δ γX + δ induces a surjective homomorphism φ: GL2(K) ΓK whose kernel consists of the constant multiples of the identity matrix, so that φ induces an isomorphism PGL2(K) → ΓK . LOW-DEGREE PERMUTATION RATIONAL FUNCTIONS 9

2.5. Fixed fields and Galois closures. Lemma 2.11. Let K be a field, let x be transcendental over K, and let G be a finite subgroup of AutK (K(x)). If f(X) ∈ K(X) has degree |G| and G fixes f(x) then the fixed field K(x)G equals K(f(x)). Proof. By basic Galois theory and Lemma 2.2 we have [K(x): K(x)G] = |G| = deg(f) = [K(x): K(f(x))]. G G Since K(x) ⊇ K(f(x)), it follows that K(x) = K(f(x)). Lemma 2.12. Let K be a field, let x be transcendental over K, and let f(X) ∈ K(X) be a separable rational function of degree n. Let Ω be the Galois closure of K(x)/K(f(x)), and let L be the algebraic closure of K in Ω. If the geometric monodromy group G of f(X) has order n then Ω = L(x). Proof. Writing t := f(x), we have [Ω : L(t)] = |G| = n = [L(x): L(t)], which implies the result since Ω ⊇ L(x). 2.6. Branch points. Definition 2.13. For any algebraically closed field K and any nonconstant 1 f(X) ∈ K(X) of degree n, a branch point of f(X) is an element β ∈ P (K) 1 which has fewer than n distinct f-preimages in P (K). Lemma 2.14. If β ∈ K is a branch point of f(X), then either β = f(∞) or β = f(γ) for some γ ∈ K such that f 0(γ) = 0. Any separable rational function has only finitely many branch points. Proof. Write f(X) = a(X)/b(X) where a, b ∈ K[X] are coprime. If β ∈ K\{f(∞)} is a branch point of f(X) then c(X) := a(X)−βb(X) is a degree- n polynomial having fewer than n distinct roots, so c(X) has a multiple root γ. Thus c0(γ) = 0, so that a0(γ) = βb0(γ), whence b(γ)a0(γ) − a(γ)b0(γ) b0(γ) · (b(γ)β − a(γ)) f 0(γ) = = b(γ)2 b(γ)2 b0(γ) · (β − f(γ)) = − = 0, b(γ) where we note that b(γ) 6= 0 since a(γ)/b(γ) = β 6= ∞. If f(X) is separable then f 0(X) 6= 0, so the numerator of f 0(X) is a nonzero polynomial and hence has only finitely many roots.

Lemma 2.15. If f(X) ∈ Fq(X) \ Fq then the q-th power map permutes the branch points of f(X), and also permutes the f-preimages of any element 1 of P (Fq). Proof. This holds because the multiplicity of α as an f-preimage of f(α) q q q equals the multiplicity of α as an f-preimage of f(α ) = f(α) . 10 ZHIGUO DING AND MICHAEL E. ZIEVE

3. Degree at most 3 In this section we give quick elementary proofs of Lemma 1.2 and The- orem 1.3. We give two proofs of the latter result, one being a very short proof using basic Galois theory and the other a slightly longer proof using essentially nothing.

Proof of Lemma 1.2. Lemma 2.6 shows that degree-one rational functions 2 1 are bijective. If q is even then clearly X permutes P (Fq), so that also any function equivalent to X2 is a permutation rational function. Conversely, we now suppose that f(X) ∈ Fq(X) is a degree-2 permutation rational 1 function. Write α := f(0) and β := f(∞), so that α, β ∈ P (Fq), and α 6= β by the permutation hypothesis. Letting µ(X) ∈ Fq(X) be a degree- one rational function which maps α and β to 0 and ∞, respectively, it 1 follows that fb(X) := µ(f(X)) permutes P (Fq) and ﬁxes 0 and ∞. Thus 2 fb(X) = (X + γX)/b(X) for some γ ∈ Fq and some nonzero b(X) ∈ Fq[X] of degree at most 1. By the permutation condition, the only zero of fb(X) in 1 Fq is 0, so we must have γ = 0. Likewise the only pole of fb(X) in P (Fq) is ∞, so b(X) has no roots in Fq and thus b(X) is constant (since deg(b) ≤ 1). 2 ∗ 2 Hence fb(X) = δX for some δ ∈ Fq, so f(X) is equivalent to X . Since fb(1) = fb(−1), the permutation condition implies 1 = −1 so q is even. Theorem 1.3 asserts that all degree-3 permutation rational functions are equivalent to X3 or an additive polynomial or a member of the following class of rational functions which is not as widely known:

Definition 3.1. A Rédeifunction is a rational function f(X) ∈ Fq(X) for which there exist two points in Fq2 \Fq which each have a unique f-preimage in Fq. These were introduced in a non-conceptual way in [19]. The following result from [4] describes basic properties of Rédeifunctions. Here Λ is the group of (q + 1)-th roots of unity in Fq2 .

Lemma 3.2. A rational function f(X) ∈ Fq(X) of degree n > 0 is a Rédei n function if and only if f = µ ◦ X ◦ ν for some degree-one µ, ν ∈ Fq2 (X) 1 1 such that µ(Λ) = P (Fq) and ν(P (Fq)) = Λ. If f(X) ∈ Fq(X) is a Rédei function of degree n > 0, then the following are equivalent: 1 (1) f(X) permutes P (Fq) (2) f(X) is exceptional (3)( n, q + 1) = 1. A degree-n Rédeifunction is indecomposable if and only if n is prime. For any n and q, there is a unique equivalence class of degree-n Rédeifunctions −1 n q in Fq(X), and it includes ν ◦ X ◦ ν where ν(X) = (X − δ )/(X − δ) for any δ ∈ Fq2 \ Fq. LOW-DEGREE PERMUTATION RATIONAL FUNCTIONS 11

We now give two proofs of Theorem 1.3, one assuming familiarity with basic facts about Galois theory and degree-one rational functions, and the other assuming nothing.

Proof of Theorem 1.3. The remarks after Theorem 1.3 show that the func- 1 tions in (1)–(3) permute P (Fq), so that also any function equivalent to 1 one of these permutes P (Fq). Henceforth assume that f(X) ∈ Fq(X) has 1 degree 3 and permutes P (Fq). We may assume q > 9, since the result is easy to verify via computer when q ≤ 9. Thus f(X) is exceptional by p Lemma 1.6. If f(X) is inseparable then f(X) = h(X ) with p := char(Fq) and h(X) ∈ Fq(X), so that p = 3 and deg(h) = 1, whence f(X) is equivalent to X3. Henceforth assume f(X) is separable. By Lemma 2.4, the arithmetic and geometric monodromy groups A and G of f(X) are transitive subgroups of S3 such that G is a proper subgroup of A, so that G = A3 and A = S3. Thus Fq2 (x)/Fq2 (f(x)) is Galois of degree 3, with Galois group generated by an order-3 automorphism of Fq2 (x) which fixes each element of Fq2 and maps x 7→ µ(x) for some degree-one µ(X) ∈ Fq2 (X). It follows that µ(X) has order 3 under composition. Thus there is some degree-one ν ∈ Fq(X) such that θ(X) := ν−1 ◦µ◦ν is either X +1 (if 3 | q) or ωX with ω of order 3 (if 3 - q). Since fb(x) := f ◦ ν(x) is fixed by the Fq-automorphism σ of Fq(x) which maps x 7→ θ(x), and the order of σ is 3 which equals [Fq(x): Fq(fb(x))], it follows that the fixed field of σ is Fq(fb(x)). Since plainly σ fixes g(x), 3 3 where g(X) := X − X if 3 | q and g(X) = X if 3 - q, it follows that there 3 is a degree-one ρ(X) ∈ Fq(X) such that ρ◦fb(X) = g(X). If g(X) = X −X then f(X) has a unique branch point, and this branch point has a unique f-preimage, so that both the branch point and its preimage must be in 1 P (Fq). Up to equivalence over Fq, we may assume that both of these points are ∞, so that f is equivalent to a(X)◦(X3 −X)◦b(X) for some degree-one a, b ∈ Fq[X], whence f(X) is equivalent to an additive polynomial over Fq. 3 Henceforth assume 3 - q, so that f(X) is equivalent over Fq to X , and thus f(X) has exactly two branch points, each of which has a unique preimage. If 1 the preimages of the branch points are in P (Fq) then, up to equivalence, we may assume that the unique root of f(X) is 0 and the unique pole of f(X) 3 ∗ is ∞, so that f(X) = γX with γ ∈ Fq. Finally, suppose that δ ∈ Fq \Fq is a preimage of a branch point of f(X). Since the q-th power map permutes the set of preimages of branch points, it follows that δ ∈ Fq2 and the preimages of branch points are δ and δq. Thus the branch points of f(X) are f(δ) and q q f(δ ) = f(δ) , which are in Fq2 \ Fq. Hence f(X) is a Rédeifunction, so the conclusion follows from Lemma 3.2. Our second proof of Theorem 1.3 uses the following lemma.

Lemma 3.3. Let f(X) ∈ Fq(X) be a separable exceptional rational function of degree 3. Then f(µ(X)) = f(X) for some degree-one µ(X) ∈ Fq2 (X) \ Fq(X) having order 3 under composition. 12 ZHIGUO DING AND MICHAEL E. ZIEVE

Proof. Let G(X,Y ) be the numerator of f(X) − f(Y ), so that G(X,Y ) ∈ Fq[X,Y ] has X-degree 3 and Y -degree 3, and G(X,Y ) is not divisible by any nonconstant element of Fq[X] or Fq[Y ]. By Lemma 2.4, every irreducible factor of G(X,Y ) in Fq[X,Y ] which remains irreducible in Fq[X,Y ] is a constant times X − Y . Note that X − Y divides G(X,Y ), and write G(X,Y ) = (X − Y )H(X,Y ) where H(X,Y ) ∈ Fq[X,Y ] has X-degree 2 and Y -degree 2. Thus H(X,Y ) must factor in Fq[X,Y ] as the product of two irreducible polynomials H1,H2 ∈ Fq[X,Y ] which each have X-degree 1 and Y -degree 1. Hence Hi(X,Y ) is the numerator of X −µi(Y ) for some degree- one µi(X) ∈ Fq(X). Let Γf be the set of all degree-one ν(X) ∈ Fq(X) for which f ◦ν = f, so that Γf is closed under composition, and also Γf is closed under the map ψ : ν(X) 7→ ν(q)(X) which raises all coeﬃcients of ν(X) to the q-th power. Since Γf is also the set of degree-one ν(X) ∈ Fq(X) for which the numerator of X − ν(Y ) divides G(X,Y ), we have Γf = {X, µ1(X), µ2(X)}, and Lemma 2.4 implies Γf ∩ Fq(X) = {X}. Thus if µ1(X) = X then since (q) (q) µ2 (X) ∈ S we must have µ2 (X) = µ2(X), so that µ2(X) ∈ Fq(X) and 3 hence µ2(X) = X. In this case the numerator of f(X)−f(Y ) is (X −Y ) , so 1 that every element of Fq \f(∞) has a unique f-preimage in P (Fq), and hence is a branch point of f(X), which contradicts separability by Lemma 2.14. Hence µ1(X) 6= X, so µ1(X) ∈/ Fq(X). Since S = {X, µ1(X), µ2(X)} is closed under ψ, it follows that ψ interchanges µ1(X) and µ2(X), so that µ1(X) ∈ Fq2 (X). Since S is closed under composition, µ1(X) has order 3 under composition.

Alternate proof of Theorem 1.3. Just as in the start of the first proof, it suffices to show that every separable exceptional f(X) ∈ Fq(X) of degree 3 is equivalent to a function in (1)–(3). By Lemma 3.3, there is a degree-one µ(X) ∈ Fq2 (X) \ Fq(X) which has order 3 under composition and satisfies f(µ(X)) = f(X). The numerator of µ(X) − X has degree 2 if µ(∞) 6= ∞ and degree at most 1 otherwise, so the set ∆ of fixed points of µ(X) in 1 P (Fq) has size 1 or 2. Since µ(X) has order 3 and f(µ(X)) = f(X), if 1 β ∈ P (Fq) is not fixed by µ(X) then β, µ(β), and µ(µ(β)) are three distinct f-preimages of f(β), so they comprise all f-preimages of f(β), whence f(β) is not a branch point of f(X), and also f(β) ∈/ f(∆). Hence the elements of ∆ are the only f-preimages of f(∆), so (since |∆| ≤ 2) each element of f(∆) is a branch point of f(X). Thus f(∆) is the set of branch points of f(X), and hence is preserved by the q-th power map. Moreover, if some δ ∈ ∆ is 1 q q q not in P (Fq) then, since f(δ ) = f(δ) is in f(∆), we have ∆ = {δ, δ }; since δq has the same multiplicity as an f-preimage of f(δq) as does δ as an f- preimage of f(δ), while the elements of ∆ are the only f-preimages of f(∆), it follows that δ is the unique f-preimage of f(δ), so that f(δ) ∈ Fq2 \ Fq. 1 1 First assume µ(X) has a unique fixed point δ in P (Fq). Then δ ∈ P (Fq), 1 so also f(δ) ∈ P (Fq). Pick degree-one ν, θ ∈ Fq(X) such that ν(δ) = ∞ = θ(f(δ), and put g := θ ◦ f ◦ ν−1 and ρ := ν ◦ µ ◦ ν−1. Then g ◦ ρ = g where LOW-DEGREE PERMUTATION RATIONAL FUNCTIONS 13 g(X) is in Fq(X) and has degree 3, with ∞ being the unique g-preimage of ∞, so that g(X) is a polynomial. Also ρ(X) is a degree-one rational function in Fq2 (X) \ Fq(X) with order 3 under composition, and ∞ is the unique fixed point of ρ(X). Thus ρ(X) = X + γ for some γ ∈ Fq2 \ Fq, and we must have char(Fq) = 3 since ρ(X) has order 3. Since g(X) is a degree-3 polynomial satisfying g(X + γ) = g(X) we have g(X) = a(X) ◦ (X3 − γ2X) 2 for some degree-one a ∈ Fq[X]. Since g(X) ∈ Fq[X], this implies γ ∈ Fq, so 3 2 that f(X) is equivalent (over Fq) to X − δX where α := γ is a nonsquare in Fq. 1 Henceforth assume that |∆| = 2. Now suppose ∆ ⊆ P (Fq), and pick −1 ν ∈ Fq(X) of degree one such that ν(∆) = {0, ∞}. Then g := f ◦ ν is −1 a degree-3 rational function in Fq(X), and ρ := ν ◦ µ ◦ ν is a degree-one rational function in Fq2 (X) \ Fq(X) with order 3 under composition, where ρ fixes 0 and ∞. Thus ρ(X) = ωX where ω ∈ Fq2 \ Fq has order 3, so since 3 g ◦ρ = g we have g = θ ◦X for some degree-one θ ∈ Fq(X), whence f(X) is 3 equivalent to X . Plainly ω ∈ Fq2 \ Fq has order 3 just when q ≡ 2 (mod 3). 1 q Finally, suppose ∆ contains an element outside P (Fq), so that ∆ = {δ, δ } for some δ ∈ Fq2 \ Fq such that f(δ) ∈ Fq2 \ Fq and δ is the unique f- preimage of f(δ), while also δq is the unique f-preimage of f(δ)q. Put ν(X) := (X − δq)/(X − δ) and θ(X) := (X − f(δ)q)/(X − f(δ)), so that g := θ ◦ f ◦ ν−1 has 0 as its unique root and ∞ as its unique pole, whence 3 ∗ −1 3 g(X) = γX for some γ ∈ Fq2 . Thus f(X) = θ ◦γX ◦ν = τ ◦fb(X) where τ := θ−1 ◦ γX ◦ ν and fb(X) := ν−1 ◦ X3 ◦ ν. It is easy to check that ν(X) 1 and θ(X) map P (Fq) bijectively onto the set Λ of (q + 1)-th roots of unity 1 3 in Fq2 . Since f(X) permutes P (Fq), it follows that g(X) = γX permutes Λ, so that γ ∈ Λ and (3, q + 1) = 1. Since ρ := ν ◦ µ ◦ ν−1 is a degree-one rational function in Fq2 (X) with order 3 under composition, and ρ fixes 0 ∗ and ∞, we know ρ(X) = ωX where ω ∈ Fq2 has order 3, so char(Fq) 6= 3 which implies q ≡ 1 (mod 3). Thus τ(X) is a degree-one rational function 1 1 in Fq2 (X) which permutes P (Fq), so that τ(X) takes values in P (Fq) at 0, −1 1, and ∞, whence τ(X) ∈ Fq(X). This implies fb(X) = τ ◦ f is in Fq(X), −1 3 so that f(X) is equivalent to fb(X) = ν ◦ X ◦ ν.

Remark. Our proofs of Theorem 1.3 (and also the proof in [5]) rely on computer calculations to show that there are no non-exceptional degree-3 permutation rational functions in Fq(X) when q ≤ 9. This can also be shown without a computer, as follows. For any f(X) ∈ Fq(X) of degree 3, it is easy to show that the normal closure of Fq(x)/Fq(f(x)) has genus at most 1. We will show in a subsequent paper that there are no non- exceptional indecomposable permutation rational functions f(X) ∈ Fq(X) (of any degree) for which the normal closure of Fq(x)/Fq(f(x)) has genus at most 1. 14 ZHIGUO DING AND MICHAEL E. ZIEVE

4. Additive polynomials In this section we prove a Galois-theoretic characterization of additive polynomials among all rational functions, and use it to describe the in- r decomposable exceptional rational functions in Fq(X) of degree 2 when r ∈ {3, 5, 7}. The results of this section are also used in our treatment of degree-4 exceptional rational functions in the next section. Pr pi Deﬁnition 4.1. A polynomial in Fq[X] is additive if it has the form i=0 αiX with αi ∈ Fq and p := char(Fq).

For each positive integer s, an additive polynomial in Fq[X] induces a homomorphism from the additive group of Fqs to itself. Since a homomorphism from a ﬁnite group to itself is bijective if and only if it has trivial kernel, this yields the following characterization of exceptional additive polynomials.

Lemma 4.2. If f(X) ∈ Fq[X] is additive then the following are equivalent: (1) f(X) is exceptional (2) f(X) permutes Fq ∗ (3) f(X) has no roots in Fq. The main result of this section is the following Galois-theoretic characterization of additive polynomials.

Proposition 4.3. If f(X) ∈ Fq(X) is separable with geometric monodromy group G then the following are equivalent: (1) |G| = deg(f), and deg(f) is a power of char(Fq) (2) f(X) is equivalent to an additive polynomial in Fq[X]. The proof uses the following easy classical result [18, Thm. 8 of Ch. 1].

Lemma 4.4. If f(X) ∈ Fq[X] is squarefree then the roots of f(X) form a group under addition if and only if f(X) is additive. Proof of Proposition 4.3. It is well-known that (2) implies (1), so we assume (1). Writing n := deg(f), we may assume n > 1 since otherwise the result is immediate. Write p := char(Fq), let x be transcendental over Fq, and put t := f(x). Let Ω be the Galois closure of Fq(x)/Fq(t), and let FQ be the k algebraic closure of Fq in Ω, where Q = q . Since n = |G|, Lemma 2.12 implies Ω = FQ(x). Thus G is a subgroup of Γ := AutFQ (FQ(x)), and Lemma 2.9 shows that Γ consists of the maps h(x) 7→ h(µ(x)) with µ(X) ∈ 3 FQ(X) of degree one. In particular, Γ has order Q − Q, and one Sylow p-subgroup of Γ is the group U consisting of the maps h(x) 7→ h(x + α) with α ∈ FQ. Since G is a p-group contained in the finite group Γ, basic group theory shows that G is contained in a Sylow p-subgroup of Λ, and that any two Sylow p-subgroups are conjugate. Thus there is some σ ∈ Λ such that σ−1Gσ is contained in U. The elements of Gb := σ−1Gσ are thus h(x) 7→ h(x + α) with α varying over an order-n subgroup H of FQ. The fixed field (x)Gb contains r(x) := Q τ(x) = Q (x + λ), where FQ τ∈Gb λ∈H LOW-DEGREE PERMUTATION RATIONAL FUNCTIONS 15 r(X) is additive by Lemma 4.4. Since deg(r) = |Gb|, by Lemma 2.11 we Gb G have FQ(x) = FQ(r(x)). It follows that FQ(x) = FQ(σ(r(x))), so if σ G maps h(x) to h(µ(x)) for some degree-one µ(X) ∈ FQ(X) then FQ(x) = G FQ(r(µ(x))). Since FQ(x) = FQ(t) = FQ(f(x)), this implies f = ν ◦ r ◦ µ 1 for some degree-one ν ∈ FQ(X). Since ∞ is the unique element of P (Fq) 1 which has a unique r-preimage, ν(∞) is the unique element of P (Fq) which q has a unique f-preimage. But since f(X) ∈ Fq(X), also ν(∞) has a unique q 1 f-preimage, so that ν(∞) = ν(∞) and thus ν(∞) ∈ P (Fq). Likewise the unique f-preimage of ν(∞) is µ−1(∞), which must equal its q-th power 1 and hence lies in P (Fq). Thus there are degree-one ρ, θ ∈ Fq(x) such that ρ(ν(∞)) = ∞ and θ(∞) = µ−1(∞). Since ∞ is fixed by the degree-one rational functions ρ ◦ ν and µ ◦ θ in FQ(X), it follows that ρ ◦ ν and µ ◦ θ are degree-one polynomials. Since r(X) is additive, this implies that g := (ρ◦ν)◦r ◦(µ◦θ) is an additive polynomial plus a constant. But g(X) equals ρ ◦ f ◦ θ, so since ρ, f, θ ∈ Fq(X) it follows that g(X) = α + L(X) for some −1 −1 α ∈ Fq and some additive L(X) ∈ Fq[X]. Thus f = ρ ◦ (X + α) ◦ L ◦ θ is equivalent to an additive polynomial. Remark. The polynomial case of Proposition 4.3 implies the main portions of [1, Thm. 3], [3, Thm. 1.1], and [23, Thm. 2.10(a)].

Lemma 4.5. If K is an algebraically closed ﬁeld with char(K) 6= 2, and x is transcendental over K, then AutK (K(x)) has no subgroup isomorphic to 3 2 (C2) , and every subgroup of AutK (K(x)) isomorphic to (C2) is conjugate to hσ, τi where σ(x) = −x and τ(x) = 1/x.

∼ r Proof. Let G be a subgroup of Γ := AutK (K(x)) such that G = (C2) with r ∈ {2, 3}. As in Lemma 2.9, we identify Γ with the group ΓK of degree-one rational functions in K(X) under the operation of functional composition. Pick some σ ∈ G of order 2. Then σ ∈ K(X) is a degree-one rational function having order 2 under composition, so σ 6= X. Since the numerator of σ − X has degree 2 if σ(∞) 6= ∞ and degree at most 1 otherwise, σ 1 has either one or two fixed points in P (K). If σ has just one fixed point then some conjugate of σ in ΓK has ∞ as its unique fixed point, and hence equals X + α with α ∈ K∗, so the order of σ is char(K) and hence is not 2, contradiction. Thus σ has two fixed points, so there exists µ ∈ ΓK which −1 maps these fixed points to 0 and ∞. Then σb := µσµ fixes 0 and ∞, −1 and has order 2, so σb = −X. Note that σb is in Gb := µGµ , which is r isomorphic to (C2) . Since Gb is abelian, each τ ∈ Gb \ hσbi must commute with σb, and hence must permute the fixed points 0, ∞ of σb. If τ fixes 0 and ∞ then τ = αX with α ∈ K∗, and since τ has order at most 2 it follows that α ∈ {1, −1} so τ ∈ hσbi, contradiction. Thus τ interchanges 0 and ∞, so τ = β/X with β ∈ K∗. Pick one such τ, and note that any ∗ ρ ∈ Gb \ hσbi must have the form ρ(X) = γ/X with γ ∈ K , so that Gb −1 −1 contains τρ = βγ X, whence βγ ∈ {1, −1} so that ρ ∈ hσ,b τi. Thus 16 ZHIGUO DING AND MICHAEL E. ZIEVE

2 Gb = hσ,b τi, so that r = 2. Finally, for ν := δX where δ = β, we have −1 −1 −1 ν Gνb = hν σν,b ν τνi = h−X, 1/Xi. Proposition 4.6. If f(X) ∈ Fq(X) is an indecomposable exceptional rational function of degree 2r with r ∈ {3, 5, 7} then q is even and f(X) is equivalent to an additive polynomial. Proof. Since f(X) is indecomposable, it must be separable by Lemma 2.3. Let A and G be the arithmetic and geometric monodromy groups of f(X), r so Lemma 2.4 implies that A is a primitive subgroup of Sn with n := 2 , G is a transitive normal subgroup of A with cyclic quotient, and the one- point stabilizers A1 and G1 have a unique common orbit. By testing all normal subgroups G of all primitive subgroups A of Sn, we ﬁnd that the ∼ r only possibility is that G = (C2) . If char(Fq) = 2 then Proposition 4.3 implies that f(X) is equivalent to an additive polynomial. Henceforth assume char(Fq) > 2. Let x be transcendental over Fq, write t := f(x), let Ω be the Galois closure of Fq(x)/Fq(t), and let FQ be the algebraic closure of Fq in k Ω, where Q = q . Lemma 2.12 implies Ω = FQ(x), so G is a subgroup of ∼ r Aut ( Q(x)), which in turn embeds into Aut ( Q(x)). But G = (C2) FQ F FQ F with r > 2, contradicting Lemma 4.5. 5. Permutation rational functions of degree 4 In this section we prove Theorem 1.4.

Lemma 5.1. Suppose q is odd and f(X) ∈ Fq(X) is a separable degree- 4 rational function whose geometric monodromy group G is isomorphic to 2 2 −2 (C2) . Then f = µ ◦ (X + X ) ◦ ν for some degree-one µ, ν ∈ Fq(X). The branch points of f(X) are µ(∞), µ(2), and µ(−2), each of which has exactly two f-preimages.

Proof. Since |G| = deg(f), if x is transcendental over Fq then the extension Fq(x)/Fq(f(x)) is Galois with Galois group G. By Lemma 4.5, there is −1 some ρ ∈ Aut ( q(x)) such that G := ρ Gρ equals hσ, τi, where σ(x) = Fq F b −x and τ(x) = 1/x. Here ρ(x) = ν(x) for some degree-one ν ∈ Fq(X). Gb −1 The fixed field Fq(x) contains both fb(x) := f(ν (x)) and g(x), where 2 −2 Gb g(X) := X + X . Since deg(fb) = deg(g) = |Gb| = [Fq(x): Fq(x) ], Gb Lemma 2.11 implies Fq(x) = Fq(g(x)). Thus fb(X) = µ◦g for some degree- 1 one µ ∈ Fq(X), so that f = µ ◦ g ◦ ν. For any α ∈ P (Fq), the images of α under the four elements of Gb all have the same image under g(X), so that if g(α) is a branch point of g(X) then some nonidentity element of Gb fixes α, whence either α = −α (so α ∈ {0, ∞}) or α = 1/α (so α ∈ {1, −1}) or α = −1/α (so α ∈ {i, −i} where i2 = −1). Since g(X) induces a surjective 1 1 map P (Fq) → P (Fq), it follows that every branch point of g(X) is in g({0, ∞, 1, −1, i, −i}) = {∞, 2, −2}. Conversely it is clear that each of ∞, 2, and −2 has two g-preimages. Since deg(µ) = deg(ν) = 1, it follows that LOW-DEGREE PERMUTATION RATIONAL FUNCTIONS 17 the branch points of f(X) are µ(∞), µ(2), and µ(−2), which are distinct and which each have two f-preimages.

Lemma 5.2. Suppose q is odd and f1, f2 ∈ Fq(X) are degree-4 exceptional rational functions. Then there exist exactly three degree-one µ ∈ Fq(X) such that µ ◦ f1(X) and f2(X) have the same branch points, and for each such µ(X) there is exactly one ν ∈ Fq(X) for which µ ◦ f1 ◦ ν = f2.

Proof. Note that f1(X) is separable since char(Fq) - deg(f1). Let A and G be the arithmetic and geometric monodromy groups of f1(X), so Lemma 2.4 implies A is a subgroup of S4 and G is a transitive normal subgroup of A with cyclic quotient such that the one-point stabilizers A1 and G1 have a unique common orbit. It is easy to check that the only possibility is A = A4 2 and G = (C2) , so that [A : G] = 3. By Lemma 5.1, f1(X) has exactly three 1 branch points, each of which has exactly two f1-preimages in P (Fq), and 2 −2 f1 = ρ1 ◦g◦θ1 for some degree-one ρ1, θ1 ∈ Fq(X), where g(X) := X +X . 1 Since f1(X) is exceptional, in particular f1(X) permutes P (Fq). If a branch 1 1 point of f1(X) is in P (Fq) then it has two preimages in P (Fq) and exactly 1 one preimage in P (Fq); but this is impossible since the set of f1-preimages 1 of any element of P (Fq) is preserved by the q-th power map. Hence no 1 branch point of f1(X) is in P (Fq). Since the q-th power map preserves the set of branch points of f1(X), it follows that the three branch points q q2 of f1(X) are α, α , α for some α ∈ Fq3 \ Fq. Likewise f2 = ρ2 ◦ g ◦ θ2 q q2 and the branch points of f2(X) are β, β , β for some β ∈ Fq3 \ Fq. Thus for any degree-one µ(X) ∈ Fq(X) such that µ ◦ f1(X) and f2(X) have the same branch points, we must have µ(α) = βqj for some j ∈ {0, 1, 2}, and then µ(αq) = µ(α)q = βqj+1 and likewise µ(αq2 ) = βqj+2 . Conversely, for each j ∈ {0, 1, 2} there is a unique degree-one µj(X) ∈ Fq(X) such that qi qi+j q (q) q µj(α ) = β for each i ∈ {0, 1, 2}. Write µj(X) = µj (X ) where (q) µj (X) is obtained from µj(X) by raising all coeﬃcients to the q-th power. (q) qi qi−1 q qi+j (q) Then µj (α ) = µj(α ) = β , so that µj (X) and µj(X) take the same values as one another at each of the three elements αqi . It follows that (q) µj (X) = µj(X), so that µj(X) ∈ Fq(X). Here µj ◦ f1(X) and f2(X) have the same branch points, so the three functions µj(X) comprise all degree- one µ(X) ∈ Fq(X) for which µ ◦ f1(X) and f2(X) have the same branch points. Fix j and put µ := µj. Let ∆ := {∞, 2, −2} be the set of branch points of g(X). The q-th power map transitively permutes the set of branch points of f1(X), which is ρ1(∆) since f1(X) = ρ1 ◦ g ◦ θ1(X). Upon replacing ρ1 (q`) (q`) and θ1 by ρ1 and θ1 for a suitable `, we may assume that ρ1(∞) = α; note that this replacement preserves the identity f1 = ρ1 ◦ g ◦ θ1, since ` (q ) (q`) 2 f1 = f1 and g = g. Further, since −g(iX) = g(X) when i = −1, we may replace ρ1 and θ1 by ρ1(−X) and iθ1(X) if necessary in order to 18 ZHIGUO DING AND MICHAEL E. ZIEVE

q assume that ρ1(2) = α . Likewise, we may assume that ρ2(∞) = µ(α) and q ρ2(2) = µ(α ), so that ρ2(δ) = µ(ρ1(δ)) for each δ ∈ ∆. Since ρ2(X) and µ(ρ1(X)) are degree-one rational functions which agree at three points, we have ρ2(X) = µ(ρ1(X)), whence −1 −1 f2 ◦ θ2 = ρ2 ◦ g = µ ◦ f1 ◦ θ1 .

Let S be the set of all degree-one η ∈ Fq(X) for which µ ◦ f1 ◦ η = f2. Then −1 S contains θ := θ1 ◦ θ2, and Gal(Fq(X)/Fq(X)) permutes S. Moreover, for −1 any degree-one η ∈ Fq(X) we have η ∈ S if and only if f1 ◦ θ ◦ η = f1, or equivalently θ ◦ η−1 ∈ G. In particular, we have |S| = 4. If η(X) ∈ G 1 lies in Fq(X) then, since f1 ◦ η = f1, injectivity of f1(X) on P (Fq) implies 1 that η(X) fixes each element of P (Fq), so that η(X) has at least four fixed points which implies η(X) = X. Since Gal(Fq(X)/Fq(X)) permutes G, and G is a four-element set with a unique fixed point under this map, it follows that the three nonidentity elements of G comprise a single orbit under this −1 map, and are in Fq3 (X) \ Fq(X). For any η ∈ S \{θ} we know that θ ◦ η is a nonidentity element of G, and hence is in Fq3 (X) \ Fq(X). Thus at least one of θ or η is not in Fq4 (X). Since Gal(Fq(X)/Fq(X)) permutes the four- element set S, it follows that the orbits of this action have sizes 1 and 3, and the size-1 orbit consists of the unique ν ∈ Fq(X) for which µ◦f1 ◦ν = f2.

Proof of Theorem 1.4. Pick f(X) ∈ Fq(X) of degree four. If f(X) is insep- 1 arable then char(Fq) = 2, and Lemma 1.2 implies f(X) permutes P (Fq) if 2 2 and only if f(X) is equivalent to X ◦µ◦X for some degree-one µ ∈ Fq(X). 2 2 Since X ◦ µ = ν ◦ X for some degree-one ν ∈ Fq(X), this says f(X) is equivalent to X4. Henceforth assume f(X) is separable. 1 If f(X) permutes P (Fq) but f(X) is not exceptional, then Proposition 1.6 implies q ≤ 81. For q ≤ 81, a computer search shows that Table 1 contains representatives for all equivalence classes of non-exceptional degree-4 bijective rational functions. It remains to determine the exceptional functions f(X). Let A and G be the arithmetic and geometric monodromy groups of f(X), so Lemma 2.4 implies A is a subgroup of S4, and G is a transitive normal subgroup of A with cyclic quotient, where the one-point stabilizers A1 and G1 have a unique common orbit. By inspection, the only possibility is A = A4 and 2 G = (C2) . If q is even then Proposition 4.3 implies f(X) is equivalent to an additive polynomial, and by Lemma 4.2 this polynomial is exceptional ∗ just when it has no roots in Fq. Henceforth assume q is odd. By Lemma 5.2, there is at most one equivalence class of degree-4 exceptional rational functions over Fq, so it remains only to show that the functions in (1) of Theorem 1.4 are exceptional. Pick 3 α, β ∈ Fq for which X + αX + β is irreducible, and put X4 − 2αX2 − 8βX + α2 f(X) := . X3 + αX + β LOW-DEGREE PERMUTATION RATIONAL FUNCTIONS 19

Table 1. Non-exceptional degree-4 permutation rational functions over Fq (the stabilizer size is deﬁned in Proposi- tion 6.2)

q f(X) Conditions Stabilizer size

X4 + αX3 + X 8 α3 + α = 1 6 X2 + X + 1

7 X4 + 3X 3

X4 + X + 1 2 1 5 X + 2 X4 + X3 + 1 3 X2 + 2 X4 + ωX 6 X3 + ω2 X4 + X2 + X 2 4 ω + ω = 1 2 X3 + ω X4 + ωX2 + X 2 X3 + X + 1

X4 − X2 + X 1

X4 + X + 1 3 1 X2 + 1 X4 + X3 + 1 3 X2 + 1

X4 + X3 + X 1 2 X4 + X3 + X 2 X2 + X + 1

3 Let γ1, γ2, γ3 be the roots of X + αX + β, so that γi ∈ Fq3 \ Fq. Then the numerator of f(X) − f(Y ) is 3 Y 2 (X − Y ) XY − γi(X + Y ) − α − 2γi . i=1 Since each of the above factors has Y -degree 1, and plainly the numerator of f(X) − f(Y ) has no factor of the form Y − δ with δ ∈ Fq, we see that 20 ZHIGUO DING AND MICHAEL E. ZIEVE each factor is irreducible in Fq[X,Y ]. But none of the factors other than X − Y is a constant multiple of a polynomial in Fq[X,Y ], so that f(X) is exceptional by Lemma 2.4. Remark. The polynomials f(X) in case (1) of Theorem 1.4 have the un- usual property that the branch points of g(X) := f(X)/4 are precisely the three elements of g−1(∞) \ {∞}, namely the three roots of X3 + αX + β. Remark. Cases (1) and (2) of Theorem 1.4 may be combined into a single case which covers both even and odd characteristic. Namely, it is easy to deduce from Theorem 1.4 that a separable f(X) ∈ Fq(X) of degree 4 is exceptional if and only if f(X) is equivalent to X4 − 2αX2 − 8βX + α2 X3 + αX + β 3 for some α, β ∈ Fq such that X + αX + β is irreducible in Fq[X].

6. The number of degree-4 permutation rational functions In this section we answer [5, Problems 9.1 and 9.2], which for each q ask for the number of equivalence classes of degree-4 permutation rational functions over Fq, an explicit representative for each class, and the total number of degree-4 permutation rational functions over Fq. Proposition 6.1. The number of equivalence classes of exceptional f(X) ∈ Fq(X) of degree 4 is (1)1 if q is odd (2)( q + 4)/3 if q ≡ 2 (mod 6) (3)( q + 8)/3 if q ≡ 4 (mod 6). If q is odd then any function in (1) of Theorem 1.4 represents the unique equivalence class. If q is even then a system of distinct representatives for 4 4 2 3 the classes consists of X , all polynomials X +X +αX with α ∈ Fq \{β + 4 4 2 β : β ∈ Fq}, and if q ≡ 4 (mod 6) then in addition X + γX and X + γ X ∗ for a single prescribed non-cube γ ∈ Fq. The number of equivalence classes of non-exceptional degree-4 permutation rational functions f(X) ∈ Fq(X) is (1)0 if q > 8 (2)1 if q = 7 (3)2 if q ∈ {2, 5} (4)3 if q ∈ {3, 8} (5)5 if q = 4. Representatives for the distinct classes are all the entries in Table 1, using all possible values for α and ω, except that for each of the two choices of ω the third entry for q = 4 yields the same equivalence class. Proof. The non-exceptional case follows from Theorem 1.4 and routine com- putations. The exceptional case for q odd follows from Theorem 1.4 and Lemma 5.2. Now suppose q is even. By Theorem 1.4, each exceptional LOW-DEGREE PERMUTATION RATIONAL FUNCTIONS 21 f(X) ∈ Fq(X) of degree 4 is equivalent to an additive polynomial L(X) ∗ 2 with no roots in Fq. If L(X) is inseparable then L = h(X) ◦ X for some h(X) ∈ Fq[X] of degree 2, and plainly h(0) = 0; since L(X) has no roots ∗ ∗ in Fq, also h(X) has no roots in Fq, so that h(X) is a monomial and thus L(X) is equivalent to X4. If L(X) is separable then, by composing on both ∗ sides with polynomials of the form δX with δ ∈ Fq, we see that L(X) is 4 2 ∗ 4 equivalent to either X + X + αX with α ∈ Fq or X + γX with γ varying ∗ ∗ 3 over a set of coset representatives for Fq/(Fq) . The condition that these ∗ 3 polynomials have no roots in Fq says that α ∈ ∆ := Fq \{β + β : β ∈ Fq} ∗ 3 and γ∈ / (Fq) . Here |∆| = b(q + 1)/3c by [22, Prop. 4.6]. It remains only to show that equivalent separable degree-4 monic additive polynomials f, g with degree-2 coefficient in {0, 1} are equal. So suppose that g = µ ◦ f ◦ ν −1 −1 for some degree-one µ, ν ∈ Fq(X). Since g (∞) = {∞} = f (∞), and any element of Fq has four distinct preimages under each of f(X) and g(X), we see that both µ(X) and ν(X) must fix ∞, and hence must be degree-one polynomials. If neither f(X) nor g(X) has a degree-2 term then the ratio of their degree-1 coefficients is a cube so f(X) = g(X). If at least one of f(X) or g(X) has a degree-2 term then equating coefficients of X4 and X2 shows that ν(X) and µ(X) are monic, and then since f(X) and g(X) are additive it follows that f(X) = g(X). Problem 9.2 in [5] asks for an explicit formula for the number of degree-4 permutation rational functions over Fq. To this end, let Γ be the group of degree-one rational functions over Fq under the operation of functional composition. Then the equivalence classes of nonconstant rational functions over Fq are precisely the orbits of Γ×Γ on the set Fq(X)\Fq under the group action (µ(X), ν(X)): g(X) → µ ◦ g ◦ ν−1. By the orbit-stabilizer theorem, in order to compute the size of an equivalence class it suffices to compute the size of the stabilizer of any prescribed element of the class.

Proposition 6.2. If f(X) ∈ Fq(X) is a degree-4 permutation rational function, then the number of pairs (µ(X), ν(X)) of degree-one rational functions in Fq(X) for which µ ◦ f ◦ ν = f is as follows: (1)3 if q is odd and f(X) is exceptional (2) q if q is even and f(X) is equivalent to an additive polynomial having a degree-2 term 4 ∗ (3)3 q if q ≡ 4 (mod 6) and f(X) is equivalent to X + γX with γ ∈ Fq (4) q3 − q if q is even and f(X) is equivalent to X4 (5) the stabilizer size listed in the corresponding entry of Table 1, if f(X) is equivalent to a rational function in Table 1 Proof. If f(X) is non-exceptional, and hence is equivalent to a function in Table 1, then this is a simple computation. Now assume f(X) is exceptional. If q is odd then the result follows from Lemma 5.2 by putting f1 = f2 = f. Henceforth assume q is even, so that f(X) is equivalent to an additive 22 ZHIGUO DING AND MICHAEL E. ZIEVE

∗ polynomial having no roots in Fq. Since equivalent functions yield the same number of pairs (µ(X), ν(X)), we may assume that f(X) is a monic additive 4 polynomial. If f(X) = X then for any degree-one ν(X) ∈ Fq(X) there is a unique degree-one µ(X) ∈ Fq(X) such that µ ◦ f ◦ ν = f, so there are q3 − q pairs (µ, ν). If f(X) 6= X4 then f −1(∞) = {∞}, while any element of Fq has at least two distinct f-preimages in Fq. Thus if degree- one µ, ν ∈ Fq(X) satisfy µ ◦ f ◦ ν = f then both µ(X) and ν(X) must fix ∞, ∗ so that µ(X) = γX + δ and ν(X) = αX + β with α, γ ∈ Fq and β, δ ∈ Fq. By equating coefficients, we see that µ ◦ f ◦ ν equals f(X) if and only if δ = γf(β), γ = 1/α4, and α4−j = 1 for each j occurring as the degree of a term of f(X). This yields the stated formulas. Remark. In case q is odd and f(X) is given by (1) of Theorem 1.4, one can explicitly describe the pairs (µ(X), ν(X)) with µ ◦ f ◦ ν = f. Let γ be 3 q a root of X + αX + β, so that γ1 := γ ∈ Fq3 \ Fq, and put γ2 := γ and q2 γ3 := γ . It suffices to describe the pairs with µ(4γ1) = 4γ2, since the other pairs are then (µ ◦ µ, ν ◦ ν) and (X,X). Put −(3β + δ)X + 4α2 µ(X) := 4 3αX + 24β − 4δ where 2 2 2 2 2 2 δ := γ1 γ2 + γ2 γ3 + γ3 γ1 and := γ1γ2 + γ2γ3 + γ3γ1 .

Writing Tr for the trace map from Fq3 to Fq, put 2 2 3q +2q+1 2 q +2q+1 −( + 3β) + Tr((γ1 − γ2) 2 ) X + α − α Tr((γ1 − γ2) 2 ) ν(X) := . q2+2q+3 3αX + δ + 3β + Tr((γ1 − γ2) 2 ) Then µ(X), ν(X) ∈ Fq(X) satisfy µ(4γ1) = 4γ2 and µ ◦ f ◦ ν = f. Formulas for the number of degree-4 permutation polynomials over Fq follow immediately from the previous two results, via the orbit-stabilizer theorem. Corollary 6.3. The number of degree-4 permutation rational functions over Fq is as follows. (1)( q3 − q)2/3 if q odd and q > 7 (2) q(q − 1)(q + 2)(q3 + 1)/3 if q even and q > 8 (3) 222768 if q = 8 (4) 75264 if q = 7 (5) 24000 if q = 5 (6) 8160 if q = 4 (7) 1536 if q = 3 (8) 78 if q = 2. Remark. Of all the counting results in this section, we ﬁnd that the most illuminating one is Proposition 6.2, about the size of the stabilizer of a permutation rational function under the stated group action. LOW-DEGREE PERMUTATION RATIONAL FUNCTIONS 23

7. Exceptional rational functions of degrees 8, 32, and 128 In this section we prove Theorems 1.7, 1.8, and 1.11. Proof of Theorem 1.7. It suﬃces to prove the ﬁrst sentence of Theorem 1.7, since this implies the second sentence by Lemma 1.6. Let f(X) ∈ Fq(X) be exceptional of degree 8. If f(X) is indecomposable then the conclusion follows from Proposition 4.6. Thus we may assume f(X) = g(h(X)) with g, h ∈ Fq(X) of degree at least 2. Plainly g(X) and h(X) are exceptional. The degrees of g(X) and h(X) are 2 and 4 in some order, which by Lemma 1.2 implies that q is even and one of g(X) and h(X) is equivalent to X2. It follows that f 0(X) = 0, so that f(X) is inseparable and thus 2 f(X) = u(X ) with u(X) ∈ Fq(X) of degree 4, where again u(X) is exceptional. By Theorem 1.4, u(X) = µ ◦ L ◦ ν where µ, ν ∈ Fq(X) have degree 4 2 one and L(X) = X + αX + βX with α, β ∈ Fq where L(X) has no roots ∗ 2 2 in Fq. Then ν ◦ X = X ◦ ρ for some degree-one ρ(X) ∈ Fq(X), so that 2 2 f(X) = µ ◦ L ◦ X ◦ ρ, whence f(X) is equivalent to L(X ) as desired. Proof of Theorem 1.8. Let f(X) ∈ Fq(X) be exceptional of degree 32. Propo- sition 4.6 implies the conclusion if f(X) is indecomposable, so we may assume that f = g ◦ h with g, h ∈ Fq(X) both of degree at least 2. Then g(X) and h(X) are exceptional. If either g(X) or h(X) has degree 2 then Lemma 1.2 implies q is even and either g0(X) = 0 or h0(X) = 0, so that 0 2 f (X) = 0, whence f = u(X ) where u(X) ∈ Fq(X) is exceptional. Finally, if {deg(g), deg(h)} = {4, 8} then Theorem 1.7 implies q is even, so Theo- rems 1.4 and 1.7 imply that both g(X) and h(X) are equivalent to additive polynomials. Proof of Theorem 1.11. Let f(X) ∈ Fq(X) be exceptional of degree 128. By Proposition 4.6, if f(X) is indecomposable then q is even and f(X) is equivalent to an additive polynomial. Now assume that f = g ◦ h for some g, h ∈ Fq(X) of degree at least 2. Then g, h are exceptional. Since one of g(X) and h(X) has degree 2r with r ∈ {1, 3, 5}, the combination of Lemma 1.2 and Theorems 1.7 and 1.8 implies that q is even. 8. Indecomposable exceptional rational functions of non-prime power degree In this section we prove Theorem 1.12.

Lemma 8.1. If f(X) ∈ Fq(X) is a separable exceptional rational function of degree n ≥ 5, then the arithmetic monodromy group of f(X) is not in {An,Sn}. Proof. Let A and G be the arithmetic and geometric monodromy groups of f(X), and suppose that A ∈ {An,Sn}. Since n ≥ 5, we know that An is simple. Since G is a transitive, we have |G| ≥ n. Next, G0 := G ∩ An satisﬁes [G : G0] ≤ [A : An] ≤ 2, so that |G| ≤ 2|G0| and thus |G0| ≥ |G|/2 ≥ n/2 > 1. Since G is normal in A, also G0 is normal in An, so that 24 ZHIGUO DING AND MICHAEL E. ZIEVE

G0 = An. But the stabilizer of 1 in An has orbits {1} and {2, 3, . . . , n}, so these are also the orbits of the stabilizer of 1 in each of A and G. This contradicts exceptionality by Lemma 2.4.

Proof of Theorem 1.12. Suppose f(X) ∈ Fq(X) is an indecomposable exceptional rational function of degree n, where n < 4096 and n is not a prime power. Then f(X) is separable by Lemma 2.3. Let A and G be the arithmetic and geometric monodromy groups of f(X), so that A is a subgroup of Sn and G is a transitive normal subgroup of A with cyclic quotient. By Lemma 2.4, indecomposability of f(X) implies that A is primitive, and exceptionality implies that if A1 and G1 are the subgroups of elements of A and G which ﬁx the element 1 of {1, 2, . . . , n} then A1 and G1 have a unique common orbit on {1, 2, . . . , n}. Since n > 1 (by indecomposability) and n is not a prime power, we have n ≥ 6, so Lemma 8.1 implies A/∈ {An,Sn}. Fi- nally, for each candidate n, we use the following Magma code to test whether there are groups satisfying the conditions required of A and G. for n in [2..4095] do if not IsPrimePower(n) then for A in PrimitiveGroups(n) do if A notin {Alt(n),Sym(n)} then A1:=Stabilizer(A,1); for GG in NormalSubgroups(A) do G:=GG‘subgroup; if #G gt 1 and G ne A and IsCyclic(A/G) and 1 eq #{i:i in Orbits(Stabilizer(G,1))|i in Orbits(A1)} and IsTransitive(G) then n; continue n; end if; end for; end if; end for; end if; end for;

Remark. The reason Theorem 1.12 restricts to n < 4096 is that Magma’s database of primitive groups consists of the primitive groups of degree less than 4096. If this database were extended to larger degrees then one could extend Theorem 1.12 to larger degrees via the same proof.

References [1] A. Bremner and P. Morton, Polynomial relations in characteristic p, Quart. J. Math. Oxford (2) 29 (1978), 335–347. [2] S. D. Cohen, The distribution of polynomials over finite fields, Acta Arith. 17 (1970), 255–271. [3] S. D. Cohen, The factorable core of polynomials over finite fields, J. Austral. Math. Soc. (Ser. A) 49 (1990), 309–318. [4] Z. Ding and M. E. Zieve, Rédeifunctions, in preparation. LOW-DEGREE PERMUTATION RATIONAL FUNCTIONS 25

[5] A. Ferraguti and G. Micheli, Full classification of permutation rational functions and complete rational functions of degree three over finite fields, Des. Codes Cryptogr. 88 (2020), 867–886. [6] M. D. Fried, Arithmetical properties of function fields (II) The generalized Schur problem, Acta Arith. 25 (1974), 225–258. [7] M. Fried, On a theorem of MacCluer, Acta Arith. 25 (1974), 121–126. [8] M. Fried, Galois groups and complex multiplication, Trans. Amer. Math. Soc. 235 (1978), 141–163. [9] M. D. Fried, R. Guralnick and J. Saxl, Schur covers and Carlitz’s conjecture, Israel J. Math. 82 (1993), 157–225. [10] R. M. Guralnick and P. Müller, Exceptional polynomials of affine type, J. Algebra 194 (1997), 429–454. [11] R. M. Guralnick, P. Müllerand J. Saxl, The rational function analogue of a question of Schur and exceptionality of permutation representations, Mem. Amer. Math. Soc. 162 (2003). [12] R. M. Guralnick, J. E. Rosenberg and M. E. Zieve, A new family of exceptional polynomials in characteristic two, Annals of Math. 172 (2010), 1367–1396. [13] R. M. Guralnick, T. J. Tucker and M. E. Zieve, Exceptional covers and bijections on rational points, Int. Math. Res. Notices 2007 (2007), 19 pp. [14] R. M. Guralnick and M. E. Zieve, Polynomials with PSL(2) monodromy, Annals of Math. 172 (2010), 1321–1365. [15] X.-D. Hou, Rational functions of degree four that permute the projective line over a finite field, arXiv:2005.07213v2, 16 Aug 2020. [16] A. A. Klyachko, Monodromy groups of polynomial mappings, in: Studies in Number Theory, 6, Izdat. Saratov. Univ., Saratov (1975), 82–91. [17] P. Müller, A Weil-bound free proof of Schur’s conjecture, Finite Fields Appl. 3 (1997), 25–32. [18] O. Ore, On a special class of polynomials, Trans. Amer. Math. Soc. 35 (1933), 559– 584. [19] L. Rédei, Uber¨ eindeutig umkehrbare Polynome in endlichen Körpern, Acta Univ. Szeged. Sect. Sci. Math. 11 (1946), 85–92. [20] J. F. Ritt, Prime and composite polynomials, Trans. Amer. Math. Soc. 23 (1922), 51–66. [21] A. Schinzel, Polynomials with Special Regard to Reducibility, Cambridge Univ. Press, Cambridge, 2000. [22] G. Turnwald, A new criterion for permutation polynomials, Finite Fields Appl. 1 (1995), 64–82. [23] G. Turnwald, Some notes on monodromy groups of polynomials, in: Number Theory in Progress, vol. 1, de Gruyter, Berlin (1999), 539–552. [24] M. E. Zieve, Permutation polynomials on Fq induced from Rédeifunction bijections ∗ on subgroups of Fq , Monatsh. Math., to appear. Available at arXiv:1310.0776v2, 7 Oct 2013.

School of Advanced Study, Hunan Institute of Traffic Engineering, Hengyang, Hunan 421001 China Email address: [email protected]

Department of Mathematics, University of Michigan, 530 Church Street, Ann Arbor, MI 48109-1043 USA Email address: [email protected] URL: http://www.math.lsa.umich.edu/∼zieve/