500 Sir W. Kamsay and Dr. B. D. Steele on the 



In order to prevent the expansion of the scale the hot 

 volume-tube stood at some distance, and the level was 

 " carried ^^ from the external point on the volume-tube to the 

 scale by means of t\\ o mercury cups and a connecting tube. 

 The cups were placed fairly level by eye, and mercury was 

 poured in until the meniscus stood just a trace below the 

 level of the point. By means of a squeezer, compressing the 

 indiarubber tube which connected the two mercury cups, 

 the level of the mercury was so adjusted as accurately to 

 touch the point. The level of the mercury on the scale was 

 the same, and readings could be made of the distance between 

 the external point, and consequently of all the internal points, 

 and the height of the mercury in the manometer. 



The reading of the levels of mercury in the cup and in the 

 manometer was made with a telescope standing at a distance 

 of 10 feet. By a simple device due, we believe, to Prof. 

 Poynting, readings to an accuracy of O'Ol mm. were easily 

 made. It consists of a piece of truly plane parallel glass in 

 front of the telescope which can be rotated through an angle 

 by means of an attached pointer. The rotation of this plane 

 glass displaces the object viewed : it is easy to count the 

 divisions on an arc, ruled in sine divisions, which correspond 

 to 1 mm. of the scale- and in this way hundredths of a milli- 

 metre may be directly and simply estimated. The device 

 offers a simple and inexpensive substitute for a catheto meter, 

 and is not behind one of the best in accuracy. 



In order to prevent absorption of moisture during the 

 filling of the bulbs this operation was carried out in vacuo by 

 means of the apparatus shown in fig. 3. The small bulb .c is 

 about 2 cms. in diameter, there are sealed to it a number of 

 small weighing-bulbs y\ x is attached by the tube to a some- 

 what larger bulb z. 



After this apparatus is made a small quantity of phosphoric 

 anhydride is introduced into ~, and the whole is exhausted 

 and sealed at v. It is then allowed to stand for a length of 

 time so that all moisture introduced in making and sealing 

 the apparatus may be absorbed ; the point v is then broken 

 and the liquid with which the bulbs are to be filled is intro- 

 duced ; the apparatus is again exhausted and again sealed at 

 i\ Finally, after the liquid has stood in contact with the 

 anhydride for a number of hours, a portion of it is distilled 

 over into .^', and so into the bulbs y ; these are sealed ofi^" in 

 such a manner by a fine blow-pipe flame that a long capillary 

 stem remains attached to y, containing no liquid or foreign 

 o-as, but only the vapour of the contained liquid. The size 

 which y should possess for a given substance is calculated 



