1. Introduction
The symmetry group of a link 
L is defined to be the mapping class group 
 (or 
) of the pair 
. The study of this symmetry group is a classical topic in knot theory, and these groups have now been computed for prime knots and links in several ways. Kodama and Sakuma [
1] used a method in Bonahon and Siebenmann [
2] to compute these groups for all but three of the knots of 10 and fewer crossings in 1992. In the same year, Weeks and Henry used the program SnapPea to compute the symmetry groups for hyperbolic knots and links of 9 and fewer crossings [
3]. These efforts followed earlier tabulations of symmetry groups by Boileau and Zimmermann [
4], who found symmetry groups for non-elliptic Montesinos links with 11 or fewer crossings.
We consider a different group of symmetries of a link 
L given by the image of the natural homomorphism
      
      Since these symmetries record an action on 
L itself (and only record the orientation of the ambient 
), we will call them “intrinsic” symmetries of 
L to distinguish them from the standard symmetry group and denote them as 
, cf. Definition 4.5. Following Whitten [
5], we denote all possible intrinsic symmetries as 
; see Definition 4.1.
Unlike the elements in the Sym group, which may be somewhat difficult to describe explicitly, each of the elements in  corresponds to an isotopy of L which may exchange the position of some components, which may reverse the orientation of the ambient space—this mirrors any diagram of L, and which may reverse orientations of some components. Neither the Boileau–Zimmerman or the Henry–Weeks-SnapPea method gives much insight into what those isotopies might look like. In addition, it is worth noting that SnapPea is a large and complicated computer program, and while its results are accurate for the links in our table, it is always worthwhile to have alternate proofs for results that depend essentially on nontrivial computer calculations.
In this spirit, the present paper presents an elementary and explicit derivation of the 
 groups for all links of 8 and fewer crossings. We rule out certain isotopies using elementary and polynomial invariants to provide an upper bound on the size of 
 for each link in our table and then present explicit isotopies generating 
 starting with the configurations of the link in Cerf’s table of alternating oriented links [
6] or (for nonalternating links) Doll and Hoste’s table [
7]. For two links in our table, 
 and 
, an additional construction is needed to rule out certain “component exchange” symmetries using satellites and the Jones polynomial. This shows that the polynomial invariants are powerful enough to compute 
 for these links. For one link in our table, 
, even this construction does not rule out all the isotopies outside the symmetry group and we must fall back on the hyperbolic structure of the link to compute the symmetry group. We give the first comprehensive list of 
 groups that we know of, though Hillman [
8] provides examples of various two-component links (including some split links) with symmetry groups equal to 12 different subgroups of 
.
Why are the 
 groups interesting? First, it is often more natural to consider the restricted group 
 than the generally larger 
. Sakuma [
9] has shown that a knot 
K has a finite symmetry group if and only if 
K is a hyperbolic knot, a torus knot, or a cable of a torus knot. Thus for many [
10] knots, the group 
 contains infinitely many elements which act nontrivially on the complement of 
L but fix the link itself. We ignore such elements, which lie in the kernel of the natural homomorphism 
. In fact, even when 
 is finite, we give various examples below where 
 has nontrivial kernel. It is often difficult to describe an element of 
 in 
 explicitly, but it is always simple to understand the exact meaning of the statement 
.
As an application, if one is interested in classifying knots and links up to oriented, labeled ambient isotopy, it is important to know the symmetry group 
 for each prime link type 
L, since links related by an element in 
 outside 
 are not (oriented, labeled) ambient isotopic. The number of different links related by an element of 
 to a given link of prime link type 
L is given by the number of cosets of 
 in 
. If we count these cosets instead of prime link types, the number of actual knots and link types of a given crossing number is actually quite a bit larger than the usual table of prime knot and link types suggests. (See 
Table 1.)
Second,  seems likely to be eventually relevant in applications. For instance, when studying DNA links, each loop of the link generally has a unique sequence of base pairs which provide an orientation and an unambiguous labeling of each component of the link. In such a case, the question of whether two components in a link can be interchanged may prove to be of real significance.
Last, we are interested in the topic of tabulating composite knots and links. Since the connect sum of different symmetry versions of the same knot type can produce different knots (such as the square knot, which is the connect sum of a trefoil and its mirror image, and the granny knot, which is the connect sum of two trefoils with the same handedness), keeping track of the action of 
 is a crucial element in this calculation. We treat this topic in a forthcoming manuscript [
11].
  2. The Symmetry and Intrinsic Symmetry Groups
As we will describe below, the group 
 was first studied by Whitten in 1969 [
5], following ideas of Fox. They denoted this group 
 or 
, where 
 is the number of components of 
L. We can write this group as a semidirect product of 
 groups encoding the orientation of each component of the link 
L with the permutation group 
 exchanging components of 
L (cf. Definition 4.1), finally crossed with another 
 recording the orientation of 
:
	  It is clear that an element 
 acts on 
L to produce a new link 
. If 
, then 
 and 
L are the same as sets (but the components of 
L have been renumbered and reoriented), while if 
 the new link 
 is the mirror image of 
L (again with renumbering and reorientation). We can then define the symmetry subgroup 
 by
For knots, 
. Here the five subgroups of 
 correspond to the standard descriptions of the possible symmetries for a knot, as shown in 
Table 2.
For links, the situation is more interesting, as the group 
 is more complicated. In the case of two-component links, the group 
 is a non-Abelian 16 element group isomorphic to 
. The various subgroups of 
 do not all have standard names, but we will call a link 
purely invertible if 
, and say that components 
 have a 
pure exchange symmetry if 
. For two-component links, we will say that 
L has pure exchange symmetry if its two components have that symmetry. The question of which links have this symmetry goes back at least to Fox’s 1962 problem list in knot theory, cf. Problem 11 in [
13] and [
14]. For example, 
 (the Hopf link) has pure exchange symmetry while we will show that 
 does not.
We similarly focus on the pure invertibility symmetry: we show that 45 out of the 47 prime links with 8 crossings or less are purely invertible. The two exceptions are 
 (the Borromean rings) and 
; we show that both of these are invertible using some nontrivial permutation, 
i.e., 
 for some 
. (Whitten found examples of more complicated links which are not invertible even when a nontrivial permutation is allowed [
15].)
Increasing the number of components in a link greatly increases the number of possible types of symmetry. 
Table 3 lists the number of subgroups of 
; each different subgroup represents a different intrinsic symmetry group that a 
-component link might have. We note that if 
 is the symmetry subgroup of link 
L, then the symmetry subgroup of 
 is the conjugate subgroup 
. Therefore, it suffices to only examine the number of mutually nonconjugate subgroups of 
 in order to specify all of the different intrinsic symmetry groups. 
Table 3 also lists the number of conjugacy classes of subgroups of 
, and the number of these which appear for prime links of 8 or fewer crossings.
  4. The Whitten Group
We begin by giving the details of our construction of the Whitten group 
 and the symmetry group 
. Consider operations on an oriented, labeled link 
L with 
 components. We may reverse the orientation of any of the components of 
L or permute the components of 
L by any element of the permutation group 
. However, these operations must interact with each as well: if we reverse component 3 and exchange components 3 and 5, we must decide whether the orientation is reversed before or after the permutation. Further, we can reverse the orientation on the ambient 
 as well, a process which is clearly unaffected by the permutation. To formalize our choices, we follow [
5] to introduce the Whitten group of a 
-component link.
Definition 4.1 Consider the homomorphism given bywhere  is defined asFor  and , we define the Whitten group 
 as the semidirect product  with the group operationWe will also use the notation  to refer to the Whitten group .    4.1.  Link Operations
Given a link L consisting of  oriented knots in , we may order the knots and write
Consider the following operations on L:
		Let 
 be a combination of any of the moves (1), (2), or (3). We think of 
 as an element of the set 
 in the following way. Let
        
        and
        
        Lastly, let 
 be the permutation of the 
 associated to 
. To be explicitly clear, permutation 
p permutes the labels of the components; the component originally labeled 
i will be labeled 
 after the action of 
.
For each element, 
 in 
, we define
        
        where 
 is 
 with orientation reversed, 
 is the mirror image of 
 and the 
 appears above if and only if 
. Note that the 
ith component of 
 is 
 the possibly reversed or mirrored 
th component of 
L. Since we are applying 
 instead of 
 to 
 we are taking the convention of first permuting and then reversing the appropriate components.
Example 4.2 Let  and . Then, .
 Example 4.3 Let  and . Then, . Since we have reversed the orientation on , note that  will be the mirror image of L as well.
 We now confirm that this operation defines a group action of the Whitten group  on the set of links obtained from L by such transformations.
Proposition 4.4 The Whitten group  is isomorphic to the group .
 We know that L is a disjoint union of  copies of  denoted . Further, the mapping class groups of  and  are both , where the elements  correspond to orientation preserving and reversing diffeomorphisms of  and . In general, the mapping class group of  disjoint copies of a space is the semidirect product of the individual mapping class groups with the permutation group . This means that  and the Whitten group  has a bijective map to .
It remains to show that the group operation * in the Whitten group maps to the group operation (composition of maps) in 
. To do so, we introduce some notation. Let 
 and 
. We must show
        
        where 
 is the operation of the Whitten Group, 
.
Then,
        
        where 
 denotes the 
i-th component 
 of 
.
        
		Note that 
, which implies
        
		Now, 
 and acts on 
L as
        
We have dropped the notation for mirroring throughout the proof, because the two links clearly agree in this regard. The element  preserves the orientation of  if and only if , i.e., if either both or neither of  and  mirror L.
We can now define the subgroup of  which corresponds to the symmetries of the link L.
Definition 4.5 Given a link, L and , we say thatL admits 
 when there exists an isotopy taking each component of L to the corresponding component of  which respects the orientations of the components. We define as the Whitten symmetry group 
of L,  The Whitten symmetry group  is a subgroup of , and its left cosets represent the different isotopy classes of links  among all symmetries . By counting the number of cosets, we determine the number of (labeled, oriented) isotopy classes of a particular prime link.
Next, we provide a few examples of symmetry subgroups. Recall that the first Whitten group  has order four and that  is a non-Abelian 16 element group.
Example 4.6 Let , the figure eight knot. Since , we have , so the figure eight knot has full symmetry. There is only one coset of  and hence only one isotopy class of  knots.
 Example 4.7 Let , a trefoil knot. It is well known that  and , but , so we have . This means that the two cosets of  are  and , and there are two isotopy classes of  knots. A trefoil knot is thus invertible.
 Example 4.8 Let , whose components are an unknot  and a trefoil . In Section 7.2, we determine all symmetry groups for two-component links, but we provide details for  here. Since the components  and  are of different knot types, we conclude that no symmetry in  can contain the permutation . Since , we cannot mirror L, i.e., the first entry of  cannot equal . The linking number of L is nonzero, so we can rule out the symmetries  and  by Lemma 7.6. Last, L is purely invertible, meaning isotopic to . Thus,  is the two element group . There are 8 cosets of this two element group in the 16 element group , so there are 8 isotopy classes of  links.  We now prove
Proposition 4.9 The Whitten symmetry group  is the image of  under the map .
 Given a map , we see that if f is orientation-preserving on , then it is homotopic to the identity on  since . This homotopy yields an ambient isotopy between L and , proving that . If f is orientation-reversing on , it is homotopic to a standard reflection r. Composing the homotopy with r provides an ambient isotopy between L and , proving that . This shows .
Now suppose . The isotopy from L to  generates an orientation-preserving (since it is homotopic to the identity) diffeomorphism  which either fixes L or takes L to . In the first case, . In the second, the map .
  5. The Linking Matrix
For each link L, our overall strategy will be to explicitly give isotopies for certain elements of the symmetry subgroup , generate the subgroup containing those elements, and then rule out the remainder of  using invariants. For three- and four-component links, a great deal of information about  can usually be obtained by considering the collection of pairwise linking numbers of the components of the link.
We recall a few definitions:
Definition 5.1 Given a n-component link, with components , we let the  linking matrix of L be the matrix  so that  and  where  is the linking number of  and . We let  denote the set of  symmetric, integer-valued matrices with zeros on the diagonal.
 The linking number can also be computed by counting the signed crossings of one knot over another. Among minimal crossing number diagrams of alternating links, the following three numbers are also useful link invariants:
Definition 5.2 The (overall) linking number  of L is half the sum of the entries of the linking matrix. (This is half of the intercomponent signed crossings of L.) The writhe  is the sum of all signed crossings of L. The self-writhe  is the sum of the intracomponent signed crossings of L. Clearly, .
 Murasugi and Thistlethwaite separately showed that writhe was an invariant of reduced alternating link diagrams [
19,
20]. Since linking number is an invariant for all links, self-writhe is also an invariant of reduced alternating diagrams. We will utilize this invariance to rule out certain symmetries of links, cf. Lemma 7.6.
The Whitten group 
 acts on the set of linking matrices 
. Further, for a given link 
L, the symmetry subgroup 
 must be a subgroup of the stabilizer of 
 under this action. This means that it is worthwhile for us to understand this action and make a classification of linking matrices according to their orbit types. We start by writing down the action:
      
Proposition 5.3 The action of the Whitten group  on n-component links gives rise to the following  action on the set of  linking matrices :  Equation (
1) reminds us that
      
	  This means that the 
ith component of 
 is component 
 of 
L. Recall that linking number is reversed by changing the orientation of either curve or the ambient 
, which proves that we should multiply by 
 as claimed.
Corollary 5.4 For , let . The action of  on  can be writtenwhere  as sets.  In principle, this description of the action provides all the information one needs to compute orbits and stabilizers for any given matrix (for instance, by computer). However, that brute force approach does not yield much insight into the structure of the problem. We now develop enough theory to understand the situation without computer assistance in the case of three-component links.
  5.1.  Linking Matrix for Three-Component Links
We first observe that there is a bijection between 
 and 
 given by
        
		We would like to understand the action of 
 on 
 by reducing it to the natural action of the simpler group 
 on 
.
Proposition 5.5 The action of  on  descends to the natural action of  on  via the surjective homomorphism  defined bywhere  We first check that 
. Now 
. This means that
        
       since 
 for any permutation 
p. But
        
		This proves that 
f is a homomorphism. In order to show that 
f is surjective we will compute the kernel. Consider the image 
. Then the product 
 since 
.
Note that direct computation shows that the pre-image of a general element  is given by  and . So suppose that . It is clear that  and . Since , if , then  and . Likewise, if , then  as well, and .
Since  is a group of order 96 and the kernel of f has order 2, the image of f has order 48. Since the target group  also has order 48, we conclude that f is surjective, as claimed.
By Corollary 5.4, the 
 action on 
 maps each entry 
 in the linking matrix to 
, 
i.e., 
        
By the definition of 
f, the natural action of 
 on 
 is
        
		This triple corresponds precisely to the new linking matrix (
3) obtained from the 
 action, so we have shown the two actions correspond.
We are now in a position to classify  linking matrices according to their orbit types, and compute their stabilizers in . The stabilizer of a linking matrix  as a subgroup of  under the group action of Proposition 5.3 is the pre-image under the homomorphism f of Proposition 5.5 of the stabilizer of the corresponding triple  under the natural action of  on . Since the kernel of f has order 2, stabilizers in  are twice the size of the corresponding stabilizers in .
There are 10 orbit types of triples  under this action. To list the orbit types, we write a representative triple in terms of variables a, b, and c which are assumed to be integers with distinct nonzero magnitudes. To list the stabilizers, we either give the group explicitly as a subgroup of  or provide a list of generators in the form . One of these groups, , is more complicated and is described below.
The group 
 is a 6 element group isomorphic to 
 (or 
) given by
        
Using the pre-image formula in the proof of Proposition 5.5, it is easy to compute the stabilizer of a given linking matrix in  directly from the table above; we simply conjugate by a permutation to bring the corresponding triple into one of the forms above and then apply the pre-image formula.
We can now draw some amusing conclusions which might not be obvious otherwise, such as:
Lemma 5.6 If L is a three-component link and any element of  reverses orientation on  then at least one pair of components of L has linking number zero.
 The stabilizer of  includes an element of the form  if and only if some element  in the stabilizer of the corresponding triple  has  since we showed in the proof of Proposition 5.5 that  equaled .
A negative  will switch the sign of the linking number ; to stabilize the triple  there must be an even number of sign changes unless some . Hence, if L has some mirror symmetry, i.e., one with , then  which produces an odd number of sign changes, so some linking number .
Example 5.7 We will see that the linking matrix for  has corresponding triple in the form . This means that the stabilizer of this linking matrix is a group of order 12 isomorphic to  in  conjugate to the pre-image of the stabilizer . Using the pre-image formula of Proposition 5.5, we can explicitly compute
      
        
      
      
      
      
      Conjugating this subgroup by , we obtain the stabilizer of :
      
        
      
      
      
      
     We know that  is a subgroup of this stabilizer; actually, it equals the stabilizer, which we show in Section 8.4.    5.2. Linking Matrix for Four-Component Links
For four-component links, we will need to develop a different observation. There are only three prime four-component links with 8 or fewer crossings, and fortunately they all possess a particular type of linking matrix. While the situation seems too complicated to make a full analysis of the 
 action on the 6 nonzero elements of a general 
 linking matrix, it is relatively simple to come up with a theory which covers our cases. We first give a correspondence between certain 
 linking matrices and elements of 
:
		Equivalently, we let
        
		If we let 
 act on the matrix 
 as usual, matrices in this form are fixed by the subgroup with permutations in an 8 element subgroup of 
 isomorphic to 
 which we will call 
.
Proposition 5.8 Let  denote the subgroup  (isomorphic to ) of . The action of the subgroup  on  descends to the natural action of  on  via the homomorphismwhere  is defined as  A series of easy but lengthy direct computations show that this homomorphism has a kernel of order 4 given by
        
        and the image of 
f in 
 is 
. The pre-image of such an element is the 4 element set 
, where 
 and 
 are arbitrary.
To show that the action of 
G on 
 descends to the natural action of 
 on 
, 
i.e., 
        
		we need only check that
        
       for all 
. This is another straightforward, if lengthy, computation.
We now need to find stabilizers for a few carefully chosen linking matrix types.
Lemma 5.9 The stabilizers in  of , , and  are all 8 element groups isomorphic to . These produce 32-element stabilizers for the corresponding linking matrices in  isomorphic to . The individual stabilizers in  are in the form  where the  are determined uniquely by p. For , the . In the other two cases, the pattern of signs is more intricate. We give the subgroups explicitly below in Table 7.  As before, we can draw some conclusions about links from this theory which we might not have noticed otherwise. For example,
Corollary 5.10 If L is a four-component link with linking matrix in the form of Equation (4), then no element of  exchanges components 1 and 3 without also exchanging components 2 and 4.    7. Two-Component Links
This section records the symmetry group 
 for all prime two-component links with eight or fewer crossings; there are 30 such links to consider. Our results are summarized in 
Section 7.1, which names and lists the symmetry groups which appear (see 
Table 8). We count how frequently each group appears by crossing number in 
Table 9. The symmetry group for each link is listed in 
Table 4 and 
Table 10, by group and by link, respectively. Proofs of these assertions appear in 
Section 7.2  7.1.  Symmetry Names and Results
The Whitten group 
 of all possible symmetries for two-component links is a non-Abelian 16 element group isomorphic to 
. The symmetry group 
 of a given link must form a subgroup of 
. There are 27 mutually nonconjugate subgroups of 
; of these possibilities, only seven are realized as the symmetry subgroup of a prime link with 9 or fewer crossings (see 
Table 8). An eighth appears as the symmetry subgroup of a 10-crossing link.
Question 7.1 Do all 27 nonconjugate subgroups of  appear as the symmetry group of some (possibly composite, split) link? Of some prime, non-split link?
 Hillmann [
8] provided examples for a few of these symmetry subgroups, but some of his examples were split links. Here are the groups we found among links with 8 or fewer crossings.
The first seven nontrivial groups in 
Table 8 are realized as the symmetry group of a link with nine or fewer crossings, while 
 appears to be the symmetry group of a 10-crossing link. We know of no nontrivial links with full symmetry but speculate that they exist. We now give three tables of results. 
Table 9 records the frequency of each group. 
Table 10 lists the prime two-component links of eight or fewer crossings by symmetry group. And 
Table 4 lists each two-component link and its corresponding Whitten symmetry group.
Theorem 7.2 The symmetry groups for all prime two-component links up to 8 crossings are as listed in Table 4.  The proof of this theorem is divided into five cases, based on the five symmetry groups that appear in 
Table 4; these proofs are found in 
Section 7.2. Many of our arguments generalize to various families of links. As this paper focuses on these first examples, we ask for the reader’s understanding when we eschew the most general argument in favor of a simpler, more expedient one.
  7.2.  Proofs for Two-Component Links
Below, we attempt to provide a general framework for determining symmetry groups for two-component links. Since  is a subgroup of , the order of the symmetry group must divide . Our strategy begins by exhibiting certain symmetries via explicit isotopies. With these in hand, we next use various techniques to rule out some symmetries until we can finally determine the symmetry group . These techniques generally involve using some link invariant to show . Among link invariants, the linking number and self-writhe (for alternating links) are easily applied since they count signed crossings; we also use polynomials and other methods.
We focus on the 30 prime links with eight or fewer crossings. Our first results indicate which of these 30 links have either a pure invertibility or a pure exchange symmetry, which we prove explicitly by exhibiting isotopies. Recall that a link is purely invertible if reversing all components’ orientations produces an isotopic link; a link has pure exchange symmetry if swapping its two components is an isotopy.
Lemma 7.3 Via the isotopies exhibited in Figure B1, Figure B2, and Figure C10, Figure C11, Figure C12, Figure C13, Figure C14, Figure C15 and Figure C16 in Appendix B.1 and Appendix C.1, the following 17 links have pure exchange symmetry:i.e.,  belongs to the symmetry group of each of these links.  As we determine symmetry groups, we will establish that the remaining 13 links in consideration do not have pure exchange symmetry.
Cerf [
6] states that all prime, alternating two-component links of 8 or fewer crossings are invertible, though this may involve exchanging components. Via the isotopies exhibited in 
Figure B3 and 
Figure C17, 
Figure C18, 
Figure C19, 
Figure C20, 
Figure C21, 
Figure C22, 
Figure C23 and 
Figure C24 of 
Appendix B.2 and 
Appendix C.2, we extend Cerf’s result to non-alternating links, and we show that the invertibility is pure (
i.e., without exchanging components). To obtain invertibility for 
, combine the results of 
Figure B4 and 
Figure B5, which show that each of its components can be individually inverted.
Lemma 7.4 All 30 prime two-component links with eight or fewer crossings are purely invertible.
 We note that the pure exchange and pure invertibility symmetries, corresponding to Whitten elements  and , respectively, generate the subgroup  of . This implies our first result about link symmetry groups.
Lemma 7.5 Any two-component link, such as those listed in Equation (6), that has both pure exchange symmetry and (pure) invertibility, must have  as a subgroup of its symmetry group .  By examining signed crossings of a link, we calculate its linking number and self-writhe; if one of these is nonzero, we may rule out some symmetries.
Lemma 7.6. - 1. 
- If the linking number , then . 
- 2. 
- For L alternating, if the self-writhe , then . 
- 3. 
- For L alternating, if  and , then . 
 Consider the effect of each symmetry operation upon linking number (see Proposition 5.3): mirroring a link or inverting one of its components will swap the sign of the linking number, while exchanging its components fixes the linking number. As for the self-writhe of a link, it is fixed by inverting any component or exchanging the two components; however, mirroring a link swaps the sign of .
Thus, the elements of  that will swap the sign of a linking number are precisely those of the form  with . If the linking number  is nonzero, these cannot possibly produce a link  isotopic to the original link L, so these eight elements are not part of . The remaining eight symmetry elements form , which proves the first assertion.
Self-writhe is an invariant of reduced diagrams of alternating links, and any symmetry operation which mirrors the link will swap the sign of . If the self-writhe  is nonzero, then no element which mirrors, i.e., , can lie in . The remaining elements form .
The last assertion follows as an immediate corollary of the first two. If both hypotheses are satisfied, then .
Lemma 7.7 Let L be a two-component link.
- 1. 
- If L is purely invertible, then . 
- 2. 
- If the components of L are different knot types, then . 
- 3. 
- If both hypotheses above are true, and - (a) 
- if , then  is either  or . 
- (b) 
- if L is alternating and , then  is either  or . 
 
 If the components of L have different knot types, then no exchange symmetries are permissible; the permutation  never appears in . Hence the symmetry group  is contained in the “No exchanges” group .
Combining these two results with the previous lemma proves the third assertion. If the linking number is nonzero and the components of L have different knot types, then . If L is also purely invertible, then . This implies that the order of  equals 2 or 4, so it is either  or .
If instead self-writhe is nonzero and the first two hypotheses hold, then . This implies that the order of  equals 2 or 4, so it is either  or .
With these five lemmas in hand, we are now prepared to begin proving Theorem 7.2, which we treat by each symmetry group.
  7.2.1. Links with Symmetry Group 
Claim 7.8 Links  and  have symmetry group .
 All of these links are purely invertible, so . Also, all of them have components of different knot types and nonzero linking numbers; thus by Lemma 7.7, their symmetry groups are either  or .
Three of the alternating links  have nonzero self-writhe, so we apply Lemma 7.7 again. We conclude that they have only the purely invertible symmetry, and  is their symmetry group.
For the remaining three links in this case 
, consider the action of the Whitten element 
. We consider the Jones polynomials of 
L and 
. They are unequal, as demonstrated below, which implies 
 is not isotopic to 
L. Thus 
, so it must be 
.
          
  7.2.2. Links with Symmetry Group 
Claim 7.9 Links  and  have symmetry group .
 All five of these links appear in our list Equation (
6) of pure exchange symmetry links; also, they all are purely invertible and have nonzero linking numbers. Lemmas 7.3 and 7.6 imply that 
.
For each link, the Conway polynomials differ for 
L and 
, where 
. Thus each link cannot have 
 as its symmetry group and must therefore have 
. We display the Conway polynomials in 
Table 11 below.
Claim 7.10 Links  and  have symmetry group .
 These links have both pure exchange and pure invertibility symmetries; they also have nonzero self-writhes and linking numbers. By Lemmas 7.3 and 7.6, their symmetry group must be .
  7.2.3. Links with Symmetry Group 
Claim 7.11 Links  and  have symmetry group .
 These links are purely invertible, comprised of different knot types, and have self-writhe ; thus, Lemma 7.6 implies .
Figure C3 and 
Figure C7 exhibit an isotopies which shows that 
 for each link, which means 
 is their symmetry group.
 Claim 7.12 Links  and  have symmetry group .
 These links are purely invertible and comprised of different knot types; thus, Lemma 7.6 implies .
First, we use the Jones polynomial to rule out mirror symmetry, 
i.e., the element 
 does not lie in 
. That means order of the subgroup 
 is between 2 and 7; hence it is either a 2 or 4 element subgroup. Here are the Jones polynomials: 
Next, for each link we depict an isotopy which reverses the orientation of just one component, i.e., we show either  or  lies in . This means  is the symmetry group for these three links.
Claim 7.13 Links  and  have symmetry group .
 These links are purely invertible and have nonzero self-writhe; thus Lemma 7.6 implies .
We take the satellites, 
 of the first and second component of 
L, for 
, as shown in 
Figure 1 and 
Figure 2, and then compute the Jones polynomial for each.
          
Since the Jones polynomials of the two different satellites are not equal for either link L, we have that L is not isotopic to  for  by Lemma 6.3. Thus, .
Figure C4 and 
Figure C8 exhibit isotopies that show 
 for both of these links. Therefore, we conclude 
 for these two links.
   7.2.4. Links with Symmetry Group 
Claim 7.14 Links  and  have symmetry group .
 These two links have the pure exchange and pure invertibility symmetries, and their self-writhes are nonzero; thus Lemma 7.6 implies 
. We show, in 
Figure B4 and 
Figure B5, that each symmetry group includes either 
 or 
, neither of which is an element of 
. Therefore, we conclude 
 for these two links.
  7.2.5. Links with Symmetry Group 
Claim 7.15 Links  and  have symmetry group .
 These three links have the pure exchange and pure invertibility symmetries, and their linking numbers are nonzero; thus Lemma 7.6 implies .
Figure C1, 
Figure C2, and 
Figure C5 display the isotopies which show 
 lies in the symmetry group for each of these three links. Since this element is not in 
, we may conclude all three links have symmetry group 
.
   8. Three-Component Links
There are 14 three-component links with 8 or fewer crossings. In this section, we determine the symmetry group for each one; 
Table 5 summarizes the results. We obtain 11 different symmetry subgroups inside 
, which represent 7 different conjugacy classes of subgroups (out of the 131 possible).
For each link, our first task is to calculate the linking matrix. Then, we utilize 
Table 12 to determine the stabilizer of this matrix within 
; we know that the symmetry group 
 must be a subgroup of this stabilizer. From there, we proceed by ruling out other elements using polynomial invariants and by exhibiting isotopies to show that certain symmetries do lie in 
 until we can discern the symmetry group.
Here are the results, listed in terms of generators for each group. We use the following notation for common group elements:
- PI, for pure invertibility, i.e., the element  
- PE, for having all pure exchanges, i.e., all elements  where  
We note that all but two of these links are purely invertible, even though PI might not be part of a minimal set of generators. Neither the Borromean rings  or the link  are purely invertible. Both of these links are, however, invertible. The Borromean rings can be inverted using any odd permutation p, i.e., they admit the symmetry ; the link  is invertible using .
Claim 8.1 The subgroup  is the 12 element group isomorphic to  generated by pure exchanges and pure invertibility.
 The linking matrix for 
 is
	  
      which is in the standard form 
. We know that 
 is a subgroup of the stabilizer of this matrix under the action of 
 on linking matrices. Consulting 
Table 12, we see that this stabilizer is the group in the claim. We must now show that all these elements are in the group. 
Figure B2 and 
Figure D2 show that
      
	  Since any 3-cycle and 2-cycle generate 
, we have the rest of the pure exchanges as well. 
Figure D1 shows that this link is purely invertible, completing the proof.
Claim 8.2 The subgroup  is the 48 element group where  is in the group if either
- (a) 
-  and p is an even permutation, or 
- (b) 
-  and p is odd. 
 Figure B5, 
Figure D3, and 
Figure D4 tell us that 
 contains the elements
      
      which clearly obey the rules in the claim. In fact, they generate a group of 48 such elements. Since the order of 
 must divide 
, it is either these 48 elements or it is all of 
. But in [
21], Montesinos proves that 
 is not purely invertible. Thus, 
 cannot be in 
, which completes the proof.
 Claim 8.3 The subgroup  is the 12 element group isomorphic to  given by .
 Figure D5 and 
Figure D6 imply that 
 contains
      
	  These three elements generate the 12 element group of the claim. Now the linking matrix for 
 is
	  
      which is in the standard form 
. Consulting 
Table 12, we see that this means 
 divides 12, the order of the stabilizer. Since we already have 12 elements in the symmetry group, it must equal the stabilizer, which completes the proof.
 Claim 8.4 The subgroup  is the 12 element subgroup isomorphic to  that is conjugate to  by .
 The linking matrix for 
 is
	  
      which corresponds to the triple 
 and has the same orbit type as 
. Thus, the stabilizer of this linking matrix is a 12 element group conjugate to the stabilizer 
 by 
. 
Figure D7 and 
Figure D8 show that
      
	  These elements generate a 12 element group, so this stabilizer is the entire symmetry group of 
, as claimed. We note that this stabilizer was worked out explicitly as Example 5.7.
Claim 8.5 The subgroup  is the 4 element group isomorphic to  generated by pure invertibility and .
 The linking matrix for 
 is
	  
      which is in the standard form 
. Consulting 
Table 12, we see that the stabilizer of this linking matrix in 
 has order 4. But 
Figure B3 and 
Figure D9 show stabilizer elements
      
	  which means that we also have 
. Therefore these three elements, plus the identity, must form the symmetry group 
 .
Claim 8.6 The subgroup  is the 4 element group isomorphic to  generated by the pure exchange  and pure invertibility.
 The linking matrix for 
 is
	  
      which is in the standard form 
. Consulting 
Table 12, we see that the stabilizer of 
 has order 4. 
Figure B4 and 
Figure D10 show
      
      and these generate the 4 element subgroup of the claim, which equals the stabilizer.
Claim 8.7 The subgroup  is the 12 element group isomorphic to  generated by pure exchanges and pure invertibility.
 The linking matrix for 
 is
	  
      which is in the form 
. Consulting 
Table 12, we see that the stabilizer of 
 has order 12, and hence 
 divides 12. Now 
Figure D12 and 
Figure D13 show that
      
	  Since the cycles 
 and 
 generate all of 
, we know that all 6 of the pure exchanges are in 
. 
Figure D11 shows that 
 is purely invertible as well, completing the proof.
Finally, we note that we have encountered this symmetry group before, as .
Claim 8.8 The subgroup  is the 4 element group isomorphic to  generated by pure invertibility and .
 The linking matrix for 
 is
	  
      which is in the standard form 
. Consulting 
Table 12, we see that the stabilizer of 
 is the 8 element group 
, which is generated by 
, and pure invertibility. Hence, 
 divides 8. 
Figure B3 and 
Figure D14 show that the latter two of these generators, namely pure invertibility and 
, are in 
.
To finish the proof, we now show that the third stabilizer generator 
 does not lie in the symmetry group. Applying it to 
, we get a link with HOMFLYPT polynomial
      
	  However, the base 
 has HOMFLYPT polynomial
      
	  This means that 
 and hence that 
 is generated by 
 and pure invertibility, as claimed.
Claim 8.9 The symmetry group  is the four element group isomorphic to  generated by  and .
 Figure D15 and 
Figure D16 imply that the four element group given above is a subgroup of 
. Now the linking matrix for 
 is
	  
      which is in the standard form 
. This means that the symmetry group must be a subgroup of the 16 element pre-image of 
 in 
. Computing this pre-image, we observe next that if we apply any of the group elements
      
      in this pre-image to our link, we get a link with Jones polynomial
      
      while the base link has Jones polynomial
      
	  This rules out those 8 elements, leaving us with a subgroup of 8 possible elements remaining. Interestingly, even framing the various components of the knot and using the satellite lemma does not allow us to rule out any of the remaining symmetries. However, this link is hyperbolic, and Henry and Weeks[
3] use SnapPea to show that the symmetry group has four elements. Thus the four element subgroup of 
 generated by the isotopies in 
Figure D15 and 
Figure D16 must be the entire group.
 Claim 8.10 The subgroup  is the 8 element group isomorphic to  which is generated by  Figure B3, 
Figure D17, and 
Figure D18, respectively, imply that the three generators above lie in the symmetry group 
. Now the linking matrix for 
 is
	  
      which corresponds to the triple 
 and has the same orbit type as 
. Consulting 
Table 12, we see that the stabilizer of the linking matrix is precisely the group above, and hence equals 
, which completes the proof.
 Claim 8.11 The subgroup  is the 4 element group isomorphic to  generated by the pure exchange  and pure invertibility.
 Figure B4 and 
Figure D19 show that 
, which implies that the subgroup above is contained in 
. Now the linking matrix for 
 is
	  
      which is in the standard form 
. Consulting 
Table 12, we see that 
 divides 4. Since we already have 4 elements in the group, this completes the proof.
 Finally, we note that we have encountered this symmetry group before, as .
Claim 8.12 The subgroup  is the four element subgroup isomorphic to  generated by pure invertibility and .
 Figure D20 and 
Figure D21 show that 
, which implies that 
 contains the claimed group. Now the linking matrix for 
 is
	  
      which is in the standard form 
. Consulting 
Table 12, we see that 
 divides 4. This completes the proof, since we already have 4 elements in the subgroup.
 Finally, we note that we have encountered this symmetry group before, as .
Claim 8.13 The subgroup  is the 8 element group isomorphic to  generated by pure invertibility, the pure exchange  and .
 Figure B4, 
Figure D22, and 
Figure D23 show that 
, which implies that the 8 element subgroup generated by these elements is a subgroup of 
. Now the linking matrix for this link is
	  
      which is in the standard form 
. Consulting 
Table 12, we see that 
 divides 16. Working out these 16 elements as in the case of 
, we see that if we apply any of the elements
      
      to 
, we get a link with Jones polynomial
      
      while the Jones polynomial of the base 
 link is
      
	  This leaves only the subgroup claimed. We note that while 
 and 
 have the same stabilizer, the symmetry group 
 is a proper subgroup of 
.
 Claim 8.14 The subgroup  is the 4 element group isomorphic to  generated by pure invertibility and the pure exchange .
 Figure B3 and 
Figure D24 show that 
 and 
 are in 
. Now the linking matrix for 
 is
	  
      which has the standard form 
. Consulting 
Table 12, we see that this matrix has an 8 element stabilizer in 
 given by the inverse image of 
 under the map 
f. Now we observe that if we apply any of the four elements
      
      in this stabilizer to 
, we get a link with Jones polynomial
      
      while the base 
 link has Jones polynomial
      
	  This rules out all but the four element subgroup above, completing the proof.
   9. Isotopies for Four-Component Links
There are three prime four-component links with 8 crossings. They are quite similar in appearance, with only some crossing changes distinguishing them. Their symmetry computations are made somewhat more difficult by the fact that we are working in the 768 element group . All three of these links are composed of four unknots linked together so that component 1 is linked to 2 and 3, and 2 and 3 are linked to 4.
Here are the symmetry groups for these links, listed in terms of generators for each group. Again, we denote the purely invertible symmetry, i.e., element , by PI; all three links admit this symmetry.
Our approach mimics the one used for three-component links. After calculating the linking matrix, we utilize 
Table 7 to determine the stabilizer of this matrix within 
; we know that the symmetry group 
 must be a subgroup of this stabilizer. Next, we show some elements are in the symmetry group by exhibiting isotopies and rule others out using polynomial invariants until we can discern the symmetry group.
Claim 9.1 The symmetry subgroup for  is the 16 element group isomorphic to  given by the -orientation-preserving elements of the inverse image .
 The linking matrix for 
 is
      
     which corresponds to the standard form 
. By Lemma 5.9, 
 must be a subgroup of the 32 element stabilizer 
 of the linking matrix.
Figure E1, 
Figure E2, and 
Figure B2 show that pure invertibility and 
 and 
 are all in 
. Together, these generate the 16 element group of the claim.
 We must show that the 16 
-orientation-reversing elements of the stabilizer (
i.e., elements of the form 
) are not in the symmetry group. If we apply any of these 16 elements to the base link, we obtain a link with Jones polynomial
      
But the Jones polynomial of the base 
 is
      
This rules out all 16 of these remaining elements, which proves the claim.
Claim 9.2 The symmetry subgroup for  is the 16 element group isomorphic to  given by the -orientation-preserving elements of .
 The linking matrix for 
 is
      
     which corresponds to the standard form 
 (remember that the ordering of elements given by Equation (
4) is not obvious). By Lemma 5.9 the stabilizer of this linking matrix in 
 is a 32 element subgroup isomorphic to 
.
Figure E3 and 
Figure E4 show that 
 and 
 are in 
. Together, these generate the 16 element group of the claim.
 We must show that the 16 
-orientation-reversing elements of the stabilizer (
i.e., elements of the form 
) are not in the symmetry group. If we apply any of these 16 elements to the base link, we obtain a link with Jones polynomial
      
      while the base link has Jones polynomial
      
	  This rules out these 16 remaining elements, so the claim is proven.
Claim 9.3 The symmetry subgroup for  is the 32 element group isomorphic to  given by .
 The linking matrix for 
 is
      
      which corresponds to the standard form 
. By Lemma 5.9 the stabilizer of this linking matrix in 
 is a 32 element subgroup isomorphic to 
. This group is generated by the isotopies in 
Figure E5, 
Figure E6, 
Figure E7, and 
Figure E8, which show that pure invertibility, along with elements
      
  10. Comparison of Intrinsic Symmetry Groups with Ordinary Symmetry Groups for Links
We now compare our results on intrinsic symmetry groups to the existing literature on symmetry groups for links. Henry and Weeks [
3,
22] report 
 groups for hyperbolic links up to 9 crossings, while Boileau and Zimmerman [
4] computed 
 groups for non-elliptic Montesinos links with up to 11 crossings, and Bonahon and Siebenmann computed 
 for the Borromean rings link (
) as an example of their methods in [
2] (Theorem 16.18).
Comparing all this data with ours, we see that
Lemma 10.1 Among all links of 8 and fewer crossings with known  groups, the Whitten symmetry group  is not isomorphic to  only for the links in Table 13.  Our results provide some data on symmetry groups of torus links as well.
Lemma 10.2 For the ,  and  torus links (, , and ), we have . For the Hopf link, the  torus link, we know that .
 Goldsmith computed [
23] the “motion groups” 
 for torus knots and links in 
. Combining her Corollary 1.13 and Theorem 3.7, we see that for the 
 torus link 
, the subgroup 
 of orientation preserving symmetries is homeomorphic to 
 if 
 is odd, and an index two quotient group of 
 if 
 is even. The motion group itself is either 
 or an 8-element quaternion group. But in either case 
.
Now the motion group by itself does not provide any information about the orientation reversing elements in Sym. However, any such element in  would map to an element in  which reversed orientation on . Since we have already shown that there are no such elements in  for , , , we see that for these links . For the Hopf link such an orientation reversing element does exist in . So,  is a  extension of  and thus is isomorphic to 
By Proposition 4.9, the group 
 is the image of 
 under a homomorphism, and hence a quotient group of 
. Further, if 
 has only orientation-preserving elements (on 
) then 
 does as well. Thus we know something about the 
 groups of all our links; 
Table 14 summarizes the new information provided by our approach.