ELECTRIC AND MAGNETIC SENSES 525 



did not significantly change during the individual sessions or from day to 

 day. This, we felt, allowed us to treat the trials as independent choices, 

 despite the facts that the field was reversed only once a day and in 

 systematic order, that is, each night before the next morning's series of 10 to 

 20 trials. In a follow-up series, we changed the direction of the field again on 

 a daily basis, but this time in random order. Under this regime, our most 

 active stingray made, without further training, 120 correct vs 64 incorrect 

 choices in a period of 15 days (P < 0.001). 



In earlier tests, frequent field reversals appeared to confuse the animals. 

 Yet, after refining both the setup and experimental procedure and 

 progressively gaining more experience, we recently began a series in which 

 the direction of the magnetic field was altered randomly from trial to trial to 

 provide the strongest evidence possible. In these most crucial experiments, 

 the field was set by a third person, while the two observers did not know 

 whether to feed or punish until after the stingrays had made their choice. 

 The change of the field was made during the commotion of feeding or 

 punishment by first slowly turning the coil current down to zero and then 

 up again in either the normal or reversed direction. After being released at 

 the north or south of the pool, the animals usually swam about briefly, often 

 stopping at the entrance of both the east and west enclosures before making 

 their final decision. Under these double-blind conditions, our fastest 

 performing ray at times attained scores as high as 9 out of 10, and upon 

 completion of the series totaled 101 correct vs 53 incorrect choices, which 

 again is significant at the extremely conservative level of P < 0.001. 



Since we know that marine elasmobranchs, when cruising through the 

 earth's magnetic field, inevitably induce electric fields that (1) are well 

 within the sensitivity and frequency range of their acute electric sense and (2) 

 strictly correlate with the magnetic compass direction in which the fishes are 

 heading, it appears reasonable to assume that the stingrays' geomagnetic 

 orientation reflects the animals' proposed electromagnetic sensory abilities. 

 Granted this inference to be correct, the ampullae of Lorenzini undoubtedly 

 are the pertinent receptors. They not only are the sense organs that render 

 sharks, skates, and rays sensitive to electric fields, but they also have the 

 proper spatial arrangement and physical disposition for the reception of the 

 induced electromagnetic voltage gradients (Kalmijn 1973, 1974). As yet, 

 attempts to electrophysiologically demonstrate the receptors' sensitivity to 

 motion through the earth's magnetic field have suffered from a miscon- 

 ception of the physics involved (Andrianov et al. 1974; Akoev et al. 1976). 

 A change of magnetic flux imposed on a stationary animal in a stationary 

 environment is principally different from a change of flux resulting from an 

 animal's motion with respect to its environment in the presence of a 

 constant magnetic field. Nevertheless, a preview of unpublished data 

 indicates that conclusive evidence is to be expected soon (Brown and 

 Ilyinsky 1977). 



In the ocean, electric fields of geophysical and geochemical origin may 

 also play an important role in the elasmobranchs' life. Thus, in addition to 

 the magnetic experiments, we are training the stingray Urolophus halleri to 



