My Math Forum groups and homo/iso/automorphisms

 Abstract Algebra Abstract Algebra Math Forum

 August 13th, 2007, 09:08 PM #1 Member   Joined: Mar 2007 Posts: 57 Thanks: 0 groups and homo/iso/automorphisms How many automorphisms are there of Z, and Z_n? (i.e. How many isomorphisms are there from Z to Z, and how many are there from Z_n to Z_n?) How many homomorphisms are there from Z to G, where G is any group of order n? What's the easiest way to describe/list/write out all of them? How many homomorphisms are there from ZxZ to S_n? (Perhaps not in general but juse show a few cases.) Let V = Z/2 x Z/2 be Klein four group. How many homomorphisms are there from V to S_2, S_3, and S_4, in each case? (Z is the set of integers, Z_n = Z mod n, S = "permutations")
 August 14th, 2007, 05:47 AM #2 Senior Member   Joined: Nov 2006 From: I'm a figment of my own imagination :? Posts: 848 Thanks: 0 I'm assuming that an automorphism creates a bijection of a set to itself. If I'm interpreting this correctly, the number of automorphisms of Z should be Z!, which is equivalent to 2^Z=2^aleph_0=c (n! rises faster than 2^n but slower than 2^(2^n)). The number of automorphisms of Z_n should be n!. I'm not going to try to guess at the remaining questions .
 August 14th, 2007, 06:49 AM #3 Global Moderator   Joined: Dec 2006 Posts: 20,931 Thanks: 2207 I suspect aptx4869 meant the group (Z, +) rather than the set Z.
 August 14th, 2007, 05:37 PM #4 Member   Joined: Mar 2007 Posts: 57 Thanks: 0 Oops. Yes, sorry, all of them are meant to be groups so Z was meant to be (Z, +), Z_n was meant to be (Z_n, +_n), G was meant to be (G, *), S_n was meant to be (S_n, composition), etc. Automorphisms are bijective and needs to satisfy the homomorphism property, so we need the identity element to be mapped to itself in all cases. So far I have found for: Z: Due to the homomorphism property, we need f(1)=1 or f(1)=-1 as it is necessary for the map to be surjective since if we have |x|>|y| we must also have |f(x)| > |f(y)|. Therefore there are only two automorphisms: f(x) = x which is the identity mapping, and f(x) = -x Z/2: only one automorphism which is the identity mapping Z/3: there are two automorphisms Z/4: there are two automorphisms Z/5: there are four automorphisms ... Z/n for n>2 I think there should be 2*int{ (n-1)/2 } automorphisms, where int{.} is the integer function / floor function / removing the non-integer part...... though I haven't proved it yet, and what's more - I could be wrong about that. Correct me if I'm wrong.
 September 19th, 2007, 03:41 AM #5 Newbie   Joined: Sep 2007 Posts: 3 Thanks: 0 You are correct about automorphisms of Z, but incorrect about Z/nZ. Since 1 generates Z/nZ a homomorphism is completely determined by the image of 1. Thus there are |Z/nZ| = n homomorphisms (by the way this answers your question about the number of homomorphisms from Z to G, there is exactly one for each element a in G, given by k \mapsto a^k). Since Z/nZ is finite, a homomorphism is injective iff it is surjective, so it is enough to show that it is surjective. If f(1) = a, the map is surjective iff a is relatively prime to n. Hence the number of automorphisms is phi(n) (where phi is the Euler phi-function). In case you didn't know: phi(n) = \prod_i (p_i - 1)p^(e_i - 1), where n = \prod_i p_i^(e_i) is the prime number decomposition of n. As you see this is consistent with your claims about Z/nZ, n = 2...5. Hope it helps /d

 Tags groups

,

,

,

### homo iso and automorphism

Click on a term to search for related topics.
 Thread Tools Display Modes Linear Mode

 Similar Threads Thread Thread Starter Forum Replies Last Post Math Amateur Abstract Algebra 0 October 29th, 2013 03:46 PM Sheila496 Abstract Algebra 0 October 20th, 2011 09:45 AM polysot Abstract Algebra 1 February 12th, 2011 05:31 PM lime Abstract Algebra 3 October 25th, 2010 07:30 AM HairOnABiscuit Abstract Algebra 1 December 12th, 2009 11:55 AM

 Contact - Home - Forums - Cryptocurrency Forum - Top