June 15, 1882] 



NATURE 



163 



A NEW THERMOGRAPH 1 



T'HE instrument under consideration is a thermograph for 

 ■*■ recording the atmospheric temperature, the fluctuations of 

 which are much less regular and more frequent than one who 

 has not made a study of it would suppose. It records the tem- 

 perature directly from the column of mercury in the tube of a 

 thermometer by dots or perforations upon a sheet of paper 

 previously ruled with degrees and hours. 



Its principal parts are, as shown in Fig. I of plate : 



1. A thermometer in the form of an ordinary mercury ther- 

 mometer, but open at the top of the tube, and having a wire 

 entering the bulb and connected to one pole of a battery, the 

 other pole of which is connected to the mechanism of the 

 iastrument. 



2. An upright cylinder revolving by clockwork, covered with 

 a paper which is divided vertically into twenty-four parts by 



■ lines representing the hours, and horizontally by lines represent- 

 ing the degrees. 



3. A bar raised and lowered by mechanism driven by clock- 

 work, furnished below with a needle entering the tube of the 

 thermometer, and carrying a pencil — or preferably a point — 

 driven forward by a small electro-magnet when the circuit is 

 closed by the needle entering the mercury, and then making a 

 mark at the proper place upon the paper and indicating the 

 temperature. 



The bar carrying the needle rises about half an inch from the 

 point at which the needle leaves the mercury, and then descends 

 until the needle again touches the mercury, whether that in the 

 meantime shall have risen or fallen, when the point makes its 

 mark upon the paper and the bar again commences to rise. 



This movement is accompli-hed by the mechanism shown in 

 the draw ing, of which only the wheel E, gearing into the rack 

 upon the needle-bar, is shown in Fig. 1, but which i- show n in 

 full and U| on an enlarged scale in Fig. 2, which i- a top view. 

 The two wheels a and l! are mov^d by clockwork (not shown), 

 and are constantly revolving- in opposite direction?, as indicated 

 by the arrow-. The-e wheels are not attached to the -haft u on 

 which the wheel E is fixed, but are attached to sleeves which 

 move without affecting that wheel except when they are joined 

 to it by the clutches c or D. They are so geared that when the 

 wheel E is joined to them, its rim moves at the rate of half an 

 inch per minute. Upon the shaft with the wheel g i- also a 

 loose sleeve F, which is free when the clutch c is not in action, 

 but which moves with that wheel when that dutch is on. 



The levers actuating the two clutches unite and move upon a 

 common pivot, from which point they extend as an arm, which 

 is capable of a lateral movement between two stops, bringing 

 one or the other of the clutches into action. 



Opposite to the wheel E, the needle-bar passe- through a 

 guide, which is furnished on the back with a small wheel taking 

 the thrust of the gear and reducing friction. For a lower guide, 

 the needle-bar is furnished on each side with a rod parallel to 

 the needle, and of nearly the same length. These rods are at 

 such distance apart that they pass clear of the thermometer tube. 

 They are not shown in the drawing, as they would lie directly 

 in front of and behind the needle and tube. 



The teeth of the clutches are partly V-shaped and partly 

 square, or nearly so, as shown in Fig. 3 ; that is, they have 

 slightly tapered sides but V-shaped points and bases, so that 

 they enter freely, as entirely V-shaped teeth would do, and w hen 

 in action they have no outward thrust. The V. shaped base 

 strengthens the tooth and admits the point of the opposite tooth. 



A very small spring on each side of the -leeve B holds it out 

 of gear while the clutch c is off. 



Beneath the clutch arm is a pressure spring, one end of which 

 presses against the end of the arm, and the other against a plate 

 moving upon the same pivot with the arm, which plate also is 

 capable of a lateral movement between its stops. 



If this spring-plate is moved in either direction to its -top, 

 carrying with it the base of the spring, the clutch-arm will be 

 moved in the other direction, and the clutch on that side will be 

 brought into action ; and if the position of the spring-plate with 

 the base of the spring be rever-ed, the position of the clutch- 

 arm will be reversed — that clutch will be disengaged, and the 

 other one will be engaged — the wheel E being moved, and the 

 needle-bar raised or lowered accordingly. 



To the sleeve F is attached an arm which is connected by a 

 draft-rod to the spring-plate. 



1 A paper by G. Morgan Fldrid^e, read at the stated meeting of the 

 Franklin Institu e ' Con-ntuted. ry the Ant' 



When the clutch C is in action — as shown in the drawing- 

 connecting the wheel a with the wheel E and the sleeve F, raising 

 the needle-bar, the arm of the sleeve F draws upon the spring- 

 plate — moving to that side the base of the reversing-spring, 

 which, when its base has pas-ed the line between the pivot and 

 the end of the clutch-arm, presses that arm to the other side, 

 disengaging that clutch, loosing the sleeve F, engaging the other 

 clutch, and reversing the motion of the needle-bar, which now 

 descends. 



The length of the arm on the sleeve F is such that when the 

 needle-bar has risen half an inch the spring-plate is moved over, 

 and the clutch-action is reversed. 



When, by descending, the needle is brought in contact with 

 the mercury and a circuit is made, the large electro-magnet, thus 

 vitalised, attracts its armature, which is attached to a lever con 



nected with and drawing upon the spring- plate, and moves the 

 base of the reversing spring to that side, changing the position 

 of the clutch-arm, and reversing the action of the clutches and 

 the movement of the needle-bar, while at the same time the 

 recording-point upon the needle bar is, by its electro-magnet, 

 driven into the paper, and the temperature is recorded upon the 

 scale. 



The sleeve F, being loose, yields to the movement of the 

 spring-plate, and is afterwards held by its clutch, and acts as 

 before. 



The action of the large electro-magnet is supplemented by 

 that of a spring drawing upon the same side of the spring- 

 plate, whose strength is such that it is not quite sufficient of 

 itself to overcome the thrust of the reversing spring, but whose 

 force is greatest when that of the electro-magnet, by reason of 



