NOVKMBKR 2, IS*!.] 



SCIENCE. 



589 



greatest convexity, farthest away from the 

 vestibule. This decussation of e and /, like 

 tiie twining inosculation of / and d, is well 

 known. It may not be so generally under- 

 stood that there is (in the eagle ; I do not 

 know whether or not in birds generally) a 

 thii-d extra-vestibular communication of the 

 bony canals. My sections show this perfectly. 

 The groat loop of d, sweeping past the decus- 

 sation of e and/, is thrown into a cavity com- 

 mon to all three. Bristles threaded through 

 t ' 'f the three canals can all be seen 



in t , JJt, crossing one another through this 

 curious extra-vestibular chamber. I call it 

 the trivia, or 'three-way place.' It is just 

 where, in fig. 0, the three niomliranous canals 



oussale, — midway between the letters e,/, 

 and c. It does not, of course, follow, tliat 

 the contained membranous canals intercom- 

 municate liere, and it apjiears from Ibsen's 

 figures that the}- do not. The ampullar dilata- 

 tions of the ends of the canals are well marked. 

 The anatomy of associate soft parts is explained 

 to some extent under fig. 7. 



The endolymph maj- contain otoliths simi- 

 lar to those great concretions called ' ear- 

 stones ' in fishes. The equilibrating function 

 of the labyrinth and its fluid appears to have 

 been determined mainly from experiments 

 upon birds. The apparatus may be likened to 

 the glass tubes filled with water and a bubble 

 of air, bj- a combination of which a survej-or, 

 for example, is enabled to adjust his theodolite 

 true ; for a bird somehow knows how the liquid 

 stands in these self-registering lexelling tulies, 

 and adjusts itself accordingly. Observations 

 made upon pigeons show, that, " when the 

 membranous canals are divided, very remark- 

 able disturbances of equilibrium ensue, which 

 vary in character according to the seat of the 

 lesion. When the horizontal canals are di- 

 vided, rapid movements of the head from side 

 to side in a horizontal plane take place, along 

 with oscillation of the eyeballs, and the animal 

 tends to spin round on a vertical axis. "When 

 the posterior or inferior vertical canals are di- 

 \-ided, the head is moved rapidly backwards 

 and forwards, and the animal tends to exe- 

 cute a backward .somersault, head over heels. 

 When the superior vertical canals arc divided, 

 the head is moved rapidly forwards and back- 

 wards, and the animal tends to execute a for- 

 ward somersault, heels over head. Combined 

 section of the various canals causes the most 

 bizarre contortions of the head and body." — 

 (Ferrier, Fund, of the brain, 187(;, [). .")7.) 

 Injury of the canals does not cause loss of 

 hearing, nor does loss of equilibrium follow de- 



struction of the cochlea. Two diverse though 

 intimately connected functions are thus pre- 

 sided over b}- the acoustic nerve, — audition 

 and equilibration. 



The wonderful and endlessly varied songs 

 of birds may acquire for us a new significance, 

 now that we understand tiie mechanism b}- 

 which these engaging creatures derive pleas- 

 ure from their own music. Though no two 

 things can well be conceived more different 

 than an anatomical disquisition and a bird- 

 song, either maj- be made to subserve the pur- 

 pose of a truer appreciation of the other ; and 

 there may be physiological aspects of even a 

 'Christmas carol.' Elliott Coues. 



Washington, Christmag, 1882. 



WHIRLWINDS, CYCLONES, AND TORNA- 

 DOES.^— I. 



The general circulation Xti the winds is at 

 times interrupted b}- local and teniporar}- dis- 

 turbances of very varied size and strength, to 

 which the general name of ' storms ' is given. 

 Their most constant features arc, a more or less 

 pronounced inward spiral whirling of the air 

 near the ground, feeding an up-draught at the 

 centre, an<l an outflow above ; and a progressive 

 motion from place to place, along a tolerably 

 well-defined track. Clouds, and generally rain 

 as well, accompany the larger storms. 



It is our object to explain how these disturb- 

 ances arise, to examine the causes and methods 

 of their peculiar action, and to study their dis- 

 tribution in time and place. With this end in 

 view, the small dust-whirlwinds that commonly 

 arise in the hot dry air of deserts will be first 

 considered. Next will come the great hurri- 

 canes and typhoons of the tropical seas, and 

 the less violent rotary storms of our own lati- 

 tudes, all of which maj- be grouped together as 

 cyclones. The tornadoes and water-spouts, 

 sliowing a ])eculiar concentration of power over 

 a very limited area, will be discussed last. 



The dry whirlwinds in flat desert regions 

 suddenly interrupt the calmness of the air, 

 .-ind begin turning, catching up dust and sand, 

 and carrying them upwards through the spiral 

 vortex to a height of many hundred feet. Thej- 

 are therefore not at all like tliose win'rls formed 

 about our street-corners at the meeting of 

 opi)osing currents of lilustering wind, or the 

 eddies of greater strength seen in windy moun- 

 tain regions ; for they arise in a time of quiet, 

 and begin their motion without apparent cause. 

 Hence- we must, at the outset, inquire into the 



• Based on a series of lectures delivered at the Lowell Insti- 

 tute, Boston, In January, 1883. 



