5*18 TRANSACTIONS OF THE AMERICAN INSTITUTE. 



tjie effect of the rotatory motion upon the projectile may be the 

 same, and thus cause the circularity of their orbits. Now have 

 we not a little clearer perception, how a rotatory force can de- 

 flect a planet into a curvilinear orbit than an attractive force can ? 

 But both forces may act together and assist each other. 



But may not this same rotatory force throw out the waters of 

 the ocean, on the conjunction and opposition sides of the earth, 

 ftnd be a cause also of the tides, which, being impeded by the 

 continents, may be the cause again of their happening at every 

 position of the sun and moon. 



Mr. Dibben said that there was no such leverage in rotary mo- 

 tion. The rotary motion of globular balls is caused simply by 

 the eccentricity of the center of gravity. 



The President inquired whether a cannon ball was ever made 

 without this eccentricity. 



Mr. Dibben said that Mr. Hotchkiss had a method of centering 

 his ball to accomplish that result. It is easy to determine upon 

 which side of a round ball is its center of gravity, by immersing 

 it in mercury. 



Mr. Parkhurst said that, according to the law illustrated last 

 week, there is a tendency in the lengthened ball for rifled cannon 

 to take the shortest axis of rotation, and showed by experiment 

 that a model of these balls would thus change its axis of rota- 

 tion, even in opposition to the force of gravitation. In practice, 

 however, this tendency may be overcome if the rotary motion is 

 not too rapid, by placing the center of gravity forward of the 

 center of the mass, or by the feathering of the rear of the ball 

 from the action of the expanding gases upon the leaden belt, so 

 that the atmospheric resistance shall counterbance the lateral 

 centrifugal force, the progressive momentum will tend very much 

 to retard the change of axis. 



THE WEIGHT OF THE ATMOSPHERE. 



Dr. Vander Weyde exhibited a new process for weighing the 

 atmosphere. In the common method, the air is not directly 

 weighed, for when we weigh the flask containing the air, we do 

 Iiot weigh the air in it. We first weigh the flask by itself, and 

 then the flask minus the amount of air abstracted from it ; and 

 it is very difficult to determine precisely how much air is ab- 

 stracted, because we cannot easily make the vacuum perfect. 

 The new method reverses the process. A flask being made suffi- 



