540 



KNOWLEDGE & SCIENTIFIC NEWS. 



[September, 1906. 



The purified dry high pressure air then enters the 

 outer coils (seven in number). It passes through the 

 three outer ones lirst. At the bottom of the apparatus 

 there is a receptacle connecting the three coils together 

 with the four otiiers; this receptacle serves the purpose 

 of separating the carbonic acid gas. After the air has 

 left tne four coils, it enters at the tojj ol the central part 

 of the liqueher. i-lere the air compressed passes through 

 a numlx^r of very fine copper tuljes to the bottom, and 

 to a number of high pressure expansion valves. By 

 these the air is reduced from 2,500 lbs. per square inch 

 to 125 lbs. to the square inch, with a consequent great 

 reduction of temperature. Thence returning over- the 

 small tubes it cools the high pressure air in them to 

 such an extent that the air liquefies partly in the tubes 

 and partly in the expansion valves. 



'J'he air which does not liquefy on expansion then 

 returns over the small copper tubes, which are enclosed 

 in a large copper vessel and connected with a pipe at 

 the top leading to the turbine, where the air under 125 

 lbs. pressure is again expanded as nearly as possible 

 to atmospheric pressure. It is thus doing work in the 

 turbine, and thereby cooling itself to a temperature 

 many degrees greater than before entering the turbine 

 under pressure. 



This cooling reagent in the form of exhaust air from 

 the turbine re-enters the liquefiers at the top of the ap- 

 paratus, circulating in the opposite direction to the high 

 pressure air within the coils. This highl3'-compressed 

 air passes over the four copper coils, in a downward 

 direction, and then returns in the upward direction over 

 the three first large coils of the liquelier. Thence it 

 circulates as previously explained over one of the two 

 moisture separators, and then to the conduct pipe to the 

 low pressure cylinder of the air compressor, together 

 with the atmospheric air drawn from the dust separator. 

 The power which is saved by- expanding air in the tur- 

 bine is utilised for operating a small air compressor, 

 and the air is utilised for cooling the high pressure air. 



ASTRONOMICAL. 



By Charles P. Butlek, A.R.C.Sc. (Lond.i, F.R.P.S. 



Prominences Observed during 1905 — 

 January I to June 30. 



The Director of the Kodail<anal Solar Physics Observatory in 

 South India has recently issued a second list of prominences, 

 from observations made during the first half of 1905. The 

 present record is rendered more complete by the fact that the 

 prominences photographed with the spectroheliograph have 

 been indexed, in addition to those recorded visually. The 

 two conditions, of course, represent conditions of matter of 

 two distinctly separate elements, the prominences recorded 

 visually in the C(Ha) line being due to hydrogen, while those 

 recorded photographically in the H line are due to calcium. 

 Very interesting and important discussions arise from the 

 minute comparative examination of these two sets of observa- 

 tions, and it is stated that from fifteen months' observations it 

 appears that near the time of sun-spot maximum the promi- 

 nences, as seen in hydrogen, agree, as a rule, very closely in 

 form with those photographed in calcium light. There are a 



few prominences seen in the hydrogen radiation which are 

 not tound on the photographs taken with calcium light, and 

 there are a larger number photographed in calcium which 

 have not been seen in hydrogen. It is expressly stated, how- 

 ever, that it cannot be assumed without much further exami- 

 nation that all those photographed but not seen were actually 

 absent from the hydrogen radiation. In many cases it is pro- 

 tiable that the bright background of sky covered with thin 

 clouds prevented their detection visually. Again, while, in 

 general, the calcium and hydrogen prominences agree closely, 

 there are frequently very marked differences. The chief of 

 these is that whereas the visually-observed hydrogen promi- 

 nences consist very often of a network of fine jets or filaments, 

 the calcium prominences show a more continuous structure. 



Of considerable importance from a physical standpoint 

 is the statement that spectroheliograms ot the disc of the sun 

 occasionally show prominences extending to a considerable 

 distance iiisidt: the limb, either (i) as an area of dark flocculi or 

 (2) as an area less dark than the surrounding area, thus indi- 

 cating greater absorption. 



The tables contain details of more than 2,000 prominences 

 determined during the half-year, giving the time of observation 

 (Madras .Mean Time), latitude, position on limb, and height in 

 seconds of arc, an appendix of remarks being added for notes 

 of special phenomena. — Kodaikanal (Observatory Bulletin. 

 No. 5. 



Sun-Spots of 1906. 



In the August number of the Observatory Mr. E. W. 

 Maunder starts an interesting experiment in the publication of 

 advance astronomical data. It is well known that the Green- 

 wich serial spot numbers are standard for many solar workers, 

 but hitherto these have not been available until the complete 

 measures and reductions had been made, generally some six 

 months after the end of the current year. By permission of 

 the .■\stronomer Royal Mr. .Maunder has made a preliminary 

 examination of the solar photographs as soon as the necessary 

 supplemental negatives have arrived from India, and num- 

 bered the spot groups at once. In an accompanying table he 

 gives such serial numbers for the spots from December 24, 

 1905, to March 27, igo6. Additional information includes the 

 duration of the spot, date of passing central meridian, longi- 

 tude and latitude, the two latter being estimations only. In 

 making use of this information it is to be understood that, 

 while the numbering of the groups is definitive, the other 

 particulars relating to the groups have only been derived from 

 a preliminary examination of the photographs, and may sub- 

 sequently be modified when measured correctly. This infor- 

 mation, however, will be widely welcomed by workers de- 

 siring to use thefc statistics in connection with other work. 



New Iron Arc Wave Length Standards. 



In consequence of a decision by the Solar Union at Oxford 

 in September, 1905, recommending the establishment of a new 

 system of reference standards of wave length, MM. Fabry 

 and Buisson have undertaken the task, and their results for 

 the region X36oo-6500 have recently been published. The 

 source of light was the iron arc, produced between poles 8 mm. 

 diameter by a current of 36 amperes at a pressure of 120 

 volts. Each line selected for standardisation has been com- 

 pared directly with the green ray of mercury given by a 

 Cooper-Hewitt mercury lamp, this green line having itself 

 been rigorously compared with the standard lines of 

 cadmium. All these new determinations have been made by 

 photography. To eliminate any errors due to expansion of the 

 interferential apparatus the mercury comparison was photo- 

 graphed both before and after the exposure to the iron arc. 

 The number of rays measured and tabulated to three 

 decimals (thousandths of a tenth metre) is 84, each having 

 been determined several times on different plates. In the 

 region near \5S00, where there is rather a large gap without 

 iron lines, four lines of nickel have been included. The wave 

 lengths given are based on those of Michelsoii and Benoit for 

 the green and red lines of cadmium, and like them, are the 

 values in air at 15° C. and 760 mm. pressure. — Comptes 

 Reiidus 143, p. 165. 



New Fornn of Spectroheliograph. 



Experiments with various forms of spectroheliograph have 

 led Messrs. G. Millochau and M. Stefanik to design a new 



