I have been looking at a problem in magnetostatics about sheets of infinite surface current, positioned along the xy plane. Now clearly there is an Amperian loop parallel to the yz-plane. I understand why there would be no x component of the magnetic field, but I don't understand why there would be no z (vertical) component. According to Griffiths, they 'cancel out' or through a clever argument he uses: the direction of the current cannot affect the vertical component. Either way, I don't understand the logic at work here. Would appreciate some feedback. Thanks.