Abstract
It is shown that the stress tensor can be symmetrised by a proper redefinition of the (linear) momentum density, even if an intrinsic angular momentum is present. The difference between 'intrinsic' and 'external' angular momentum turns out to be a matter of definition. There is no independent hydrodynamic variable due to the angular momentum conservation law. Some expressions for the (linear) momentum density are discussed and the use of the symbol Cij for different quantities by various authors is clarified. The tensor Cij contains two independent parameters by Galilean invariance. Angular monentum conservation including the dipole-dipole interaction is discussed. The orbit wave spectrum shows a quadratic k dependence in the hydrodynamic region and different results are obtained only outside that region or by the use of non-hydrodynamic variables. In two appendices the mode spectrum (with unclamped normal fluid) is discussed and Graham's free energy is explicitly written down.