If I can offer my 2 cents...
Think of it this way. The electron doesn't follow a nice, planetary orbit like the cartoon pictures we learned in earlier classes. The Heisenberg was mentioned earlier. We can, with some accuracy, determine the position of an electron. If we measure the position of an electron, we can plot it in 3 dimensions, and it will be a point at a given point in time.
If we measure the position at a different time, it will be in a different position. Perhaps a little closer to the nucleus, perhaps not. We can plot that point, too, and now we have 2 points where we determined the electron to be.
If we keep plotting the position of the electron, eventually it fills out a 3-dimensional shape - at least 95% of the time. This shape that is created by plotting the position of the electron is the atomic orbital.
Orbitals aren't physical shapes... think of them as areas of high probability. It is statistically probably that we will find an electron within the volume defined by the atomic orbital.
Try this, see if it helps...
http://www.upei.ca/~chem342/Resources/Reviews/Molecular_Orbital_Tutorial.pdfThe p,d,f g,h, etc, orbitals are spherical.
That is, the sum of the px,py and pz are spherical. The sub-orbitals, are however not spherical
Very interesting... I hadn't thought of it that way. Makes sense, though.