342 ANNUAL REPOET SMITHSONIAN INSTITUTION, 193 3 



black or blackish it is possible to see through the binoculars that 

 these fins do not vibrate while they are outstretched during the 

 actual air flight. 



The fact that the expanded paired fins do not vibrate after the fish 

 leaves the water surface proves that the caudal fin does not beat in 

 the air (another fantas}'^ of some authors). Furthermore, the edge 

 of this fin can be seen to be clear-cut while a fish volplanes away in 

 line with one's vision. 



THE CONTROL OF FLIGHT DIRECTION 



I conclude from my observations, therefore, that a flying fish when 

 in the air makes no effort by any flapping or vibration of the fins 

 to add to the velocity it gained by the sculling of the tail during 

 the swimming under water and during the taxi at the surface. That 

 it does, however, use the fins in the air to control the direction of 

 the flight is certain. This control is most often seen when a fish 

 avoids the obstruction of the ship's bow by flying around it in a 

 distinct curve, and even more spectacularly when one, headed di- 

 rectly for the ship's side, turns off at right angles in a curve of short 

 radius. Wlien a flying fish approaches even closer to the ship, it 

 dives into the water, to take advantage of the greater resistance of 

 that medium in making the shorter turn necessary to avoid a col- 

 lision with the ship. The control is so nearly perfect that I have 

 never seen one strike a vessel by day, though they do so when blinded 

 by a light at night, and I have found one published record of a fly- 

 ing fish striking the side of a ship. Occasionally I have seen them col- 

 lide, not with a ship but with one another (as though to heighten 

 the analogy of their flight to that of airplanes). Flying fishes, 

 therefore, seem to be lacking either in perfect flight control or in 

 perfect air vision. 



The flying fishes of the Gypselur^us type, like airplanes again, 

 show a vertical as well as a horizontal control of flight direction. 

 This is beautifully evident as they maintain a course nearly parallel 

 with a choppy water surface, just skimming over sharp-peaked 

 crests and dropping to a lower level over the intervening troughs. 

 As already noted, the diving to avoid collision with a ship likewise 

 indicates a vertical control. 



The claim has been made that flying fishes invariably take off into 

 the wind, but I have seen them scoot away toward all points of the 

 compass, w^hile a strong wind remained in one quarter. The power 

 and speed of the taxi seem sufficient to hurl them into the air, when 

 the pelvic planes are expanded, no matter what may be the angle of 

 the taxi to that of the wind. But after the fishes start their flight, 

 especially as they begin to lose momentum, their direction is much 



