But if we start dividing sexuality into classes based on preferences (whether or not they're genetic or learned), the number of classes quickly becomes unmanageably large, with possibly one class per person.  Can any two people have exactly the same preferences with exactly the same emotional power behind those preferences?  (Hmm.  It strikes me that we need a scale for measuring "emotional power", but this probably shows my ignorance of psychology.)

For many years, I've thought instead of sexuality being a spectrum that has at least two dimensions, if not more.  Class divisions imply that some one is either completely this or that, but a spectrum allows for very small variations in preferences.  In fact, I don't even like using the words "homosexual" or "heterosexual", because those are located at the extreme ends of the spectrum.  And I don't like "bisexual" because it implies that someone has no preference at all.

If someone wants to classify himself or herself as, for instance, a submissive who prefers his own sex but is not averse to having sex with the women, we might be able to locate that set of preferences on a multi-dimensional spectrum.  But calling that person a bisexual who is into BDSM doesn't seem accurate enough.

I realize that artificial divisions can make it easy when we need to generalize about people with similar preferences.  I just think its useful to remember that sexuality is not a set of discrete preferences.  We are more like a rainbow.

Now, as for the case of twins: isn't that something like masturbation?  Not being a twin, I can't say, but...

Again, please let me say how sorry I am for that syrupy "rainbow" remark.  It just came out.  I couldn't help it.