(a) Bernoulli's principle: as we move along a streamline in the steady flow of an incompressible, non-viscous fluid, the sum of the pressure P, the kinetic energy per unit volume 21ρv2 and the potential energy per unit volume ρgh remains constant:
P+21ρv2+ρgh=constant
(b) Consider a fluid of density ρ flowing steadily through a pipe of varying cross-section and height, as shown. In region 1 (BC) the area is A1, the height h1, the speed v1 and the pressure P1; in region 2 (DE) they are A2, h2, v2 and P2.

Consider the fluid that initially lies between B and D. In a short time Δt the fluid at B moves to C, a distance v1Δt, and the fluid at D moves to E, a distance v2Δt.
Work done by pressure. The fluid behind pushes on our fluid with a force P1A1 through a distance v1Δt, doing work W1=P1A1v1Δt=P1ΔV. By the equation of continuity, A1v1=A2v2, the same volume ΔV passes through DE. There the fluid ahead pushes back with a force P2A2 through v2Δt, so the work done on our fluid is −P2A2v2Δt=−P2ΔV. The net work done on the fluid is
W=(P1−P2)ΔV
Change in energy. The flow is steady, so the fluid between C and D is in the same state at the start and at the end. The net effect is that a mass Δm=ρΔV has moved from region 1 to region 2:
ΔK=21ρΔV(v22−v12),ΔU=ρgΔV(h2−h1)
Work–energy theorem. With no energy lost to friction,
(P1−P2)ΔV=21ρΔV(v22−v12)+ρgΔV(h2−h1)
Dividing by ΔV and rearranging,
P1+21ρv12+ρgh1=P2+21ρv22+ρgh2
Since 1 and 2 are any two points along the flow, P+21ρv2+ρgh=constant. This is Bernoulli's equation. The derivation assumes that the fluid is incompressible (its density is the same everywhere and no elastic energy is stored), that it is non-viscous (no energy is lost to internal friction) and that the flow is steady.
(c) For a fluid at rest, v1=v2=0, and the equation becomes
P1+ρgh1=P2+ρgh2⇒P1−P2=ρg(h2−h1)
Taking point 2 at the free surface of a liquid open to the air (P2=Pa) and point 1 at a depth h=h2−h1 below it gives P=Pa+ρgh, the pressure at a depth h in a liquid at rest.
Honey is highly viscous. As it flows through a long, narrow tube, its layers move with different speeds and exert large frictional forces on one another, so much of the work done by the pressure difference turns into heat instead of kinetic or potential energy. The derivation assumed that no energy is lost in this way, so for honey P+21ρv2+ρgh does not stay constant along the tube: in a horizontal tube of uniform bore the pressure falls along the tube even though the speed does not change.