I haven't followed this closely enough to know if it's been completely worked over, yet. I'm including an Email from my son, who is fairly good at problems like this. My apologies if this is now old material. --BC ---------------------------------------------------- Hi, Dad, Actually, this is very similar to a question at STS. For n > 2, it's possible to do it in two trips. The basic idea is that you can connect the cables into groups of different sizes with the patch cord on one end and, when you get to the other end, you use the battery and wire to check each pair of cables and see if it lights up. For example, with 11 cables, you might start on side A and patch 4 together, 3 together, pair 2 up, and leave 2 unconnected. When you get to side B, you use your battery & bulb, choose any cable, and try to form a circut with all of the other cables in turn and look at how many cause the bulb to light up. If 2 do, then you can say "oh, it looks like this cable must be connected to 2 others on the other end, therefore it's part of the group of 3" Then you patch the cables together on side B in the same manner and run back to side A and use you battery & bulb. Here, when you patch together your new groups, you want the cables that were in the same group on side A to be in distinguishable groups on side B. i.e. for the 4 cables which were connected on side A, and are therefore indistinguishable on side B, you want to put one in a new group of 4, one in a new group of 3, one in a new group of 2, and leave one unconnected. You want the cables to be in groups of different sizes, so that they can be easily distinguished from each other. The 'unconnected' cables also form a group--but, because they are unconnected, may be any size and still be distinguishable. For all values except for T-1 and T-2, where T is a triangular number, you can form your groups such that they are all easily distinguishable. For those two cases, however, you have to fudge things slightly--but it can still be done. For example, suppose that you are solving the nasty case of 14 cables. You may want to connect 4 cables into group A, 3 cables into group B, 3 cables into group C, 2 into group D, and leave two unconnected (E). Now, on the other side, using the battery and lightbulb, you can determine which cables are in A, which are in B or C, which are in D, and which are in E. Connect the cables into groups as follows: 1: A, B(C), C(B), D, E 2: A, B(C), D, E 3: A, B(C), C(B) 4: A, C(B) (unconnected) Now, when you get back, one of the cables from either B or C is going to be unconnected. This by itself is enough to tell you which group is B and which group is C. Hope that this makes sense.