May 2 [, 1903] 



NATURE 



69 



The kind of plate used is not very sensitive at about A. 5000, 

 and this may account for the absence of the band \ 5016, 

 which, however, is exceedingly faint in this Nova. H7 is 

 present, but scarcely strong enough to measure, and merges 

 into a brighter band which extends from A. 4347 to X 4371 

 <mean at A 4359). 



A reproduction of the spectrogram is given, and it is 

 seen that the spectrum corresponds to those of Nova 

 Aurigae and Nova Persei at the later stages of their develop- 

 ment. 



A very faint bright band in the spectrum of Nova 

 Geminorum in the region of the chief nebula lines is far 

 too weak to measure. 



Parallax of the Binary System 5 Equulei. — Mr. W. J. 

 Hussey publishes in Bulletin No. 32 of the Lick Observ- 

 atory the results of his calculation of the parallax of 

 •S Equulei, based on the micrometrical and spectroscopical 

 measurements made at the Lick Observatory during the 

 past three years. The method pursued is theoretically abso- 

 lute, for in no way is the result dependent upon the assump- 

 tion of values for comparison stars, as it is in the ordinary 

 method of calculating parallax. 



The formula used was published by Prof. A. A. Rambaut 

 {M.N. March, 1890), and gives the absolute parallax of a 

 system when the elements of the orbit, the relative velocity 

 of the components in the line of sight, and the orbital 

 velocity of the earth at the time are known. 



The determination of the elements of the orbit made at 

 Lick has led to the adoption of 57 years as the periodic 

 time of revolution ; using this value for the period, and 

 taking the mean distance as o"-28, the eccentricity as 0-46, 

 the apastron and periastron distances as o"-409 and o"i5i 

 respectively, the relative velocity in the line of sight, deter- 

 mined by the observers using the Mills spectrograph, as 

 ;2o 5 miles per second, and the orbital velocity of the earth 

 at the time as 18-2 miles per second, Mr. Hussey obtains 



ir = o'''o7i 

 as the parallax of this system, but states that this is 

 probably not the final value, for the elements may be 

 appreciably modified during the critical observations' it is 

 proposed to make during the next three years. 



Taking this value for the parallax and the mean distance 

 and period given above, the mass of the system becomes 

 1-89, the mass of the sun being taken as unity, and, as the 

 components are not quite equal in magnitude, the brighter 

 may have a mass equal to, but not greatly exceeding, that 

 of the sun. The mean distance of the components is about 

 four times that of the earth from the sun, but, owing to 

 the great eccentricity of the orbit, the actual distance at 

 periastron is just more than twice, and at apastron about 

 five times, that unit. As the spectra of the components 

 are both of the solar type, and as their masses are com- 

 parable with that of the sun, it might be reasonably assumed 

 that their densities do not differ to any great extent from 

 the density of that body. 



A 



A REGULATING OR RECORDING 

 THERMOMETER. 

 THERMOMETER which is capable of regulating the 

 temperature of a room with considerable accuracy, or 

 of keeping a continuous record of the temperature, is fre- 

 quently required in laboratory work. Such a thermometer 

 is described in the present article. Although there is little 

 essentially new in its construction, the details on which 

 success depends are the result of considerable practical ex- 

 perience, and as the manufacture of such an instrument 

 should be within the powers of most laboratories employing 

 a mechanic, it has been thought desirable to publish an 

 account of it. 



The estimation of temperature in this thermometer de- 

 pends on the alteration in shape of a piece of flat brass 

 tubing bent into spiral form and filled with a liquid possess- 

 ing a large coefficient of expansion. If one end of the tube 

 is fixed, the motion of the other end, magnified by a suit- 

 able arrangement of levers, serves as a measure of tempera- 

 ture. As the thermometer is intended for use within a 

 range of temperature of at most three or four degrees, we 



NO. 1751, VOL. 68] 



are not concerned with the equality of the graduations per 

 degree at different parts of the scale. 



The illustration (Fig. i) shows the general appearance of 

 the thermometer arranged as a recording instrument. The 

 brass tubing of which the spiral is formed has a section in 

 the shape of a very flat ellipse, the longer diameter being 

 J inch, the shorter 3/16 inch, while the thickness of the 

 wall is 002 inch. The tube is bent into the spiral form by 

 filling it with melted resin and bending it round a cylinder 

 8 inches in diameter, on which is cut a spiral groove. After 

 the resin has been removed by heating the tube, brass plugs' 

 are soldered into the ends, each plug having a central hole 

 for the purpose of filling the tube with liquid. In the 

 thermometer illustrated, these holes are shown closed by 

 steel screws. A simpler and more efficient plan is to solder 

 a short length of lead tubing into the brass plug. Then, 

 when the thermometer has been filled with liquid, the end 

 of the lead tube is pinched together and soldered. The 

 spiral can thus be hermetically sealed without loss of liquid. 



In order that the thermometer may acquire the tempera- 

 ture of the surrounding air as rapidly as possible, the surface 

 is increased by soldering to the spiral a strip of thin sheet 

 copper about four inches wide. The whole is painted dead 

 black. 



For filling the tube creosote has been found to answer 

 well. The process of filling the tube is the most trouble- 

 some part of the work, as it is difficult to get rid of the air 

 bubbles which cling to the interior. While it is being 

 carried out the tube should be placed in melting ice. 



Fig. I. — Recording Thermometer. 



Funnels may be attached to the open ends of the spiral to 

 facilitate the introduction of the liquid. When the tube is 

 nearly full, liquid should be poured into either end in turn 

 until the creosote rises in the other funnel free from air 

 bubbles. 



When the tube has been hermetically sealed, it is ready 

 for attachment to the stand. Instead of fixing one end 

 of the tube directly to the base board, it is fastened to one 

 flap of a common brass hinge, the other end of which is 

 screwed to the board. A hole is tapped in the upper flap and 

 fitted with a screw the point of which bears against the lower 

 flap, thus providing an adjustment for the distance between 

 the two. This is a very simple method of giving a small 

 alteration to the position of the fixed end of the spiral, and 

 so adjusting the pen to any desired height on the recording 

 cylinder. 



The free end of the spiral is attached by a connecting rod 

 of thin aluminium to a brass lever, half an inch in length, 

 fixed to the spindle that carries the tracing arm. The 

 length of the light arm which carries the pen is sixteen 

 inches. Thus the actual motion of the end of the spiral is 

 multiplied by the factor 32 at the recording drum. 



The bracket carrying the spindle is formed of two up- 

 rights of thin sheet brass, screwed and soldered to a thicker 

 base plate. The spindle itself is made of steel wire about 

 three-sixteenths of an inch in diameter ; the ends forming 

 the pivots are turned down to a somewhat smaller diameter 

 and ground into holes bored in the uprights. On the outer 

 side of each upright is screwed a short length of flat steel 



1 The plugs should be of drawn brass, as it is found that creosote gradu- 

 ally percolates through cast brass. 



