the notches with the thumb nail while the hand is going from the body, and in the second movement you scratch the notches with the nail of the second finger when the hand is coming toward the body, thus producing two different vibrations.  In order to make it work perfectly (?) you must of course say “skidoo” when you begin the first movement, and then, no matter how fast the little arm is revolving when changed to the second movement, you must say “skidee” and the arm will immediately stop and begin revolving in the opposite direction.  By using the magic words the little arm will obey your commands instantly and your audience will be mystified.  If any of your audience presume to dispute, or think they can do the same, let them try it.  You will no doubt be accused of blowing or drawing in your breath, and many other things in order to make the arm operate.  At least it is amusing.  Try it and see.  —­Contributed by Charles Clement Bradley, Toledo, Ohio.

* * * * * * *

The foregoing article describing the “Skidoo-Skidee Trick” appeared in a recent issue of Popular Mechanics.  I have been told that a similar arrangement is used by a tribe of Indians in the state of Washington, by the Hindoos in India, and one friend tells me that they were sold on the streets of our large cities many years ago.

This toy interested me so much that I have made an investigation into the causes of its action, and I think the results may be of interest.

To operate, one end of the notched stick is held firmly in the left hand, while with the right hand a nail or match stick is rubbed along the notched edge, at the same time pressing with the thumb or finger of the moving hand against the oblique face of the stick.  The direction of rotation depends upon which face is pressed.  A square stick with notches on edge is best, but the section may be circular or even irregular in shape.  The experiments were as follows: 

1.  A rectangular stick had notches cut on one face.  When the pressure was applied upon a face normal to the first, no rotation resulted.  If the pressure was upon an edge, rotation was obtained.

2.  Irregular spacing of the notches did not interfere with the action.  The depth of the notches was also unimportant, although it should be suited to the size of the nail for best results.

3.  The hole in the revolving piece must be larger than the pin; if there is a close fit no rotation is obtained.

4.  The center of gravity of the revolving piece must lie within the hole.  If the hole is not well centered the trick cannot be performed.

5.  If the stick be clamped in a vise no results are obtained; with this exception:  if the stick has enough spring, and the end clamped is far enough away from the notched portion, the rotation may be obtained.

The above experiments led me to the conclusion that the operation of the device is dependent upon a circular motion of the pin, and this was confirmed by the following experiments.  The action is somewhat similar to swinging the toy known as a locust around with a slight circular motion of the hand, It is necessary to show here that a slight circular motion is sufficient to produce the result and, secondly, that such motion can be produced by the given movements of the hands.