EEPOET OF THE SECEETARY. 77 



occasional assistance in matters of a physical or astronomical nature connected with 

 the Institution. Among experimental undertakings of this kind may be mentioned 

 the rough measurement of the absorption of certain substances for infra-red radia- 

 tions, made at the request of some correspondents of the Institution. 



Personnel. — No changes have been made in the permanent staff of the Observatory. 

 Doctor Gilbert completed his temporary services on August 15, 1902, and Dr. J. R. 

 Benton tilled a temporary ajipointment from September 16, 1902, to November 30, 

 1902. 



2. The new horizontal telescope and other apimratns. 



Referring to my report of last year, it will be recalled that preliminary attempts 

 had been made to measure the absorption of the gases of the solar envelope by holo- 

 graphic study of an enlarged solar image, and that it -was your intention to continue 

 the work so as to include the holographic study of sunspot spectra, but that these 

 researches were temporarily laid aside till a more suitable arrangement for forming 

 and guiding the solar image could be obtained. Much study and experiment 

 has been devoted to this matter in the past year, and as a result a horizontal reflect- 

 ing telescope of 20 inches aperture and 140 feet focus has been obtained,- and pro- 

 vided with a tube in which the air can be thoroughly stirred to overcome "boiling," 

 in accordance with the experiments reported last year. To "feed" this horizontal 

 telescope a inodification of the coelostat has been devised which is believed to be 

 before untried, and which renders this simple instrument so well adapted for 

 the purpose of solar research that it is hoped that the device will approve itself 

 elsewhere. 



A large instrument of this type has been constructed by the J. A. Brashear Com- 

 pany, of Allegheny, Pa., and will form a part of the Astrophysical 01:)servatory 

 exhibit at the Louisiana Purchase Exposition at St. Louis in 1904. All the above 

 apparatus, including the coelostat, long-focus mirror, tube and air-stirring devices, 

 and three great piers for the coelostat, "concave mirror, and bolometric apparatus for 

 the study of the image, are now in use. The accompanying illustration, PI. IV, 

 shows the great coelostat and a portion of the tube which incloses the beam from 

 the coelostat to the concave mirror, 55 feet north, and thence south and under the 

 coelostat to the plane of the focal image. It will be noted by the reader that the 

 beam is reflected in the plane of the meridian from the first plane mirror mounted 

 on a polar axis which turns half as fast as the earth, and that a second reflection 

 occurs at the surface of a second plane mirror, adjustable about two horizontal axes, 

 and also cai)able of moving bodily, by means of tracks, east and west and north and 

 south. Thus the second mirror can receive the beam at any hour of any day of the 

 year, and reflect it in any desired direction. In practice a nearly horizontal and 

 northerly direction is chosen. 



PI. Ill is from the working drawing of the tube, which is an acute V in general 

 shape, with a longer branch of circular cross section extending from the concave 

 mirror on the northern pier to the focus, 140 feet distant on the southern, and with 

 a shorter branch uniting with the longer at its northern end, but proceeding south- 

 ward and inclined upward at an angle of 6° and ending at the coelostat, 55 feet dis- 

 tant. This shorter branch is circular at its northern end, but broadens out to an 

 elliptical cross section, as shown, in order to inclose the beam for the east and west 

 positions of the second coelostat mirror. Both branches of the tube are of galvanized 

 iron, with two walls separated by an air sjiace \\ inches thick all around. The inner 

 tube is blackened and is provided with diaphragms. 



In Plate III is also shown the air ducts which are employed for stirring in the great 

 tube. Starting from the blowing engine, which is a 29-inch circular fan blower, with 

 direct-connected 22-horsepower electric motor, making 700 revolutions per minute, 

 the blast is carried by a 20-inch main to a point near the middle of the tube, where 



