I was looking through a book on magnetism and it had the attached diagrams of field lines and stated that the classical diagram of field lines exiting the N pole and entering the S pole is incorrect. They claim their diagram is the correct representation of the magnetic field lines. Their rationale is that if you take a long bar magnet, there is a strong magnetic attraction at the ends but very little in the middle. But isn't this observation due to field gradients? That is, the field lines curve at the ends of a bar magnet and the field is strongly inhomogeneous, while at the middle of the bar magnet the field lines are almost parallel and the field is almost homogeneous. Thank you.