This is really off topic, but:
It just occurred to me that there are cases where the representation of a
CONSERVATIVE E field by continuous lines tangent to the field does not
allow the interpretation of the line density (per unit perpendicular area)
as a proportional measure of the magnitude of E. This occurs inside
continuous charge distributions, where the divergence of E is non-zero.
E.g.: inside an isolated sphere of constant charge density (rho):
E(r) = (rho*r)/(3*epsilon) so that the total "E-flux" piercing an interior
sphere is (4*PI*r^2)*E(r) = (4*PI*rho)/(3*epsilon) * r^3 (just Gauss'
law).
So the E-flux is a function of r and cannot be represented by continuous
lines. To properly represent this flux, the NUMBER of lines must
continuously increase with r, because of the continuous source rho.