83 



graphically. It should be noted 

 that these photographs, which 

 are short-time exposures, indicate 

 not only the direction of motion 

 in any region, but also the velocity 

 of motion, which is obtained from 

 measurement of the length of 

 the lines, comparison being made 

 with the length of the lines in 

 the open channel where the 

 velocity is known. 



Some examples of the results 

 obtained are shown in the accom- 

 panying photographs, which are 

 taken from the report of the 



NATURE 



-De Havilland strut. 



plate in water (by methods 2 [a] 

 and [b]). Fig 7 shows the flow 

 past a strut in water (by method 

 2 [b] ). Fig. 8 shows the dead 

 region which exists, even at low 

 velocities, at the tail of an airship 

 model. 



The last figure is of some 

 interest, as it has been found 

 that where a dead region exists 

 at the tail of a fish-form model, 

 the resistance of the model is not 

 appreciably affected by the shape 

 of the tail within this region, 

 and so lone: as the tail is 



Advisory Committee for 

 Aeronautics, 1911-12. 



Figs. 1 and 2 show the 

 flow past an inclined 

 square plate in air at two 

 different velocities (by 

 method 1). Figs. 3 and 

 4 show the same types of 

 flow in water (by method 

 2 [a]). Figs. 5 and 6 

 show the flow past a 



Fig. 6.— Flat plate. 

 NO. 2265, VOL. 91] 



tail of an airship model. 



sufficiently blunt to cause the 

 formation of "dead air" 

 (as is usually the case in 

 airships, aeroplane hulls, struts, 

 etc.) it is convenient for con- 

 structional reasons to end the 

 tail rather abruptly once the 

 boundary of the dead region has 

 been passed. 



A dead region is always an 

 indication of high resistance, and 

 is therefore undesirable. 



It is hoped that observation of 

 the flow past models, together 

 with resistance measurements. 



