Milton Van Dyke’s “An Album of Fluid Motion” is a rich historical contribution of flow visualisation.
A note about CFD stems from a photo that appears on page 23 for streaklines marked by dye released from upstream in a water tunnel for a NACA 64A015 airfoil at zero incidence, Re=7000 (laminar and steady flow). Van Dyke commented: “Appears to be unseparated”. Subsequent laminar RANS solution of S.R. Allmaras for streamlines clearly shows a separation bubble starting at nearly 60% chord. This is to stress the important insight one can get through CFD (although nowadays CFD methods are much more sophisticated).
Just as reminder, both stremlines and streeaklines are curves in space. While the former is defined as being parallel to the velocity vector at all times, the later is taken from the Lagrangian view as the it is a curve of fluid parcels for which all parcels posses the same point of origin somewhere upstream.
This is exactly why a streakline can be easily realized in a simple experiment marking the flow with a passive contaminant (such as dye in a liquid for the above case). Streamline on the other hand, is achieved by a mathematical endeavour, simply defined by constructing a curve which is parallel to the velocity field at all times.
In this case, the flow is steady, so both streamlines and streaklines from a fixed origin are simply particle paths (a Lagrangian standpoint again). For unsteady flows complication arises and they will usually look very different.
Kinematics is fun… 😉