234 PHENOMENA OF FLIGHT IN THE ANIMAL KINGDOM. 



which we are conducted by the graphic as well as the other methods which, 

 we may employ. 



The apparatus on which the wings record their motions is the ordi- 

 nary registering apparatus, consisting of a metal cylinder covered with 

 smoked paper, to which a uniform rate of motion is imparted by clock- 

 work. Let ns suppose that, instead of the motions of the wings, we 

 would simply register the oscillations of a vibrating rod. For this pur- 

 pose the extremity of the rod is furnished with a little style which 

 touches the blackened paper with its point, and, as the different parts 

 of the movable cylinder pass successively before the point, the soot is 

 detached from the jilaces which it touches and a trace produced. If the 

 rod is not in vibration, it makes a long white rectilinear trace without 

 sinuosities, a straight line which, rolled upon the cylinder, constitutes a 

 circumference. If it is in vibrator}^ motion, its trajectory will be a 

 curved line, of which the sinuosities indicate all the circumstances of 

 the motion, its phases of elevation, it depressions — in a word, all its 

 movements — and consequently all the oscillations which the vibrating 

 rod executes in space will be faithfully reproduced on the paper. If 

 we would ascertain the frequency of the oscillations, it is sufficient to 

 know the rate at which the cylinder revolves. Ordinarily a tuning-fork 

 is employed, of which the number of vibrations is previously known, as, 

 for example, one hundred vibrations per second. This is made to write 

 its vibrations upon the registering cylinder below the line traced by the 

 vibrating rod, of which the number of A'ibrations are desired. The 

 comparison of the two tracings shows at once the number of the motions 

 of the tuning-fork back and forth ; that is to say, how many hundredths 

 of a second correspond to one oscillation of the rod; the number of mo- 

 tions of the vibrating body during a given time is thus known with 

 ^Tcat exactness. 



It is not, however, as easy to obtain the tracing from the wing of au 

 insect as from a vibrating rod, and this for several reasons. In the first 

 place it is very difficult to fix at the extremity of the wing a writing- 

 style ; however light it may be, the rapidity of the motion to which it 

 is submitted is sufficient in most cases co throw it off". If, however, 

 after many trials and much precaution we are able to retain it in its 

 place, a permanent cause of i^erturbation still exists from its very 

 presence. Under the influence of this incumberance the extent and fre- 

 quency of the strokes of the wing are evidently diminished. It is easy 

 to convince ourselves of this by taking a Macroglossa and fixing it in the 

 manner w^hich w^e have previously described ; that is, immovably between 

 two strips of cork, by means of a pin. Looking down upon it we per- 

 ceive the extreme limits traversed by the wing above and below which 

 we have called the dead-points. If some substance is applied to the 

 surface of the wing, we see by the effect of this burden, in diminishing 

 the play of the organ, the two limits of oscillation approach one another, 

 and the extreme upper position, which just now was almost vertical, in- 

 clines toward the horizontal. We may finally remark that it is only at 

 the cost of considerable chafing against the surface of the moving- 

 cylinder that we can obtain a complete tracing of the movement of the 

 wing. The wing cannot touch the cylinder except during a very short 

 instant of its stroke ; that is, the instant when the wing reaches pre- 

 cisely the distance from the body of the animal to the cylindrical sur- 

 face. The sj)herical figure which the margin of the wing describes in 

 space cannot have more than one i)oint in common with the blackened 

 cylinder. We can, therefore, only obtain, as the w^hole impression, a 

 series of points at more or less regular intervals ; and, if a more pro- 



