July 12, 1883] 



NA TURE 



255 



STELLAR PHOTOGRAPHY AT HARVARD 



AT the meeting of the Astronomical Society which was 

 held on June 8 last, Prof. Pickering of Harvard 

 College Observatory, so well known for his stellar obser- 

 vations, and who is a Foreign Associate of the Society, 

 attended and gave an interesting account of the work 

 which has been done during the past few years at his 

 observatory. 



Some few years ago Prof. Pickering took up the work 

 of determining the intensity of the light of the principal 

 stars by eye observation, without taking the question of 

 colour into consideration, work which has been already 

 dwelt upon in this journal. For this purpose he used a 

 photometer, completing his observations, which number 

 some 90,000, about a year ago, and a large part of his results 

 are already in print. The published results of the more 

 important investigators of star magnitudes, from the time 

 of Almagest and Lufi, have al-o been reduced. Sir W. 

 JHerschel's observations, which appeared almost a century 

 ago in the Philosophical Transactions, have likewise been 

 taken in hand at Harvard Observatory and completely 

 discussed. Sir John Herschel's works, the " Uranometria 

 Nova," the " Durchmusterung," as well as many other 

 works in the same field, have also been made use of in 

 preparing the Harvard Catalogue, which therefore shows 

 those ca'es in which the photometric observations carried 

 out by Prof. Pickering differ from the results obtained by 

 other observers, when their observations are reduced to 

 the same system. These eye observations of stars having 

 been completed, Prof. Pickering, in conjunction with his 

 brother, Mr. W. H. Pickering, has taken up stellar photo- 

 graphy from the same point ot view. By this means a com- 

 parison is obtained between the brightness of the star as 

 seen by the e\e, and its brightness as determined by its 

 greater or less action upon the photographic plate ; and 

 by a comparison of photographs taken on different nights 

 any variation in brightness may be detected ; whilst the 

 exact positions of stars may of course be more accu- 

 rately and permanently recorded than by eye observa- 

 tions. Mr. A. A. Common recently, by taking photo- 

 graphs of the nebula in Orion on different nights and 

 comparing them, has thus teen able to detect a probable 

 variation in one of the stars in the nebula, and in 1858 

 Professor George P. Bond, by measuring the diameters of 

 stars in photographs was able to determine the relative 

 brightness of the two stars which form the double f Ursae 

 Majoris. 



But the work at Harvard University was to do more 

 than this. The stars which Prof. Bond examined were 

 close together. Prof. Pickering wished to compare stars 

 far removed from each other. For this purpose the ordi- 

 nary method of stellar photography, by which photographs 

 are taken at the foci of large telescopes, would not suffice. 

 These photc graphs only comprise a small region of but 

 one or two degrees in diameter. A different method was 

 therefore employed in the Harvard observations. A 

 wholly different form to the ordinary equatorial telescope 

 was used. It is not unusual to construct photographic 

 cameras to take pictures of buildings which subtend an 

 angle of 6o° or even 90 . But when applied to the stars, 

 however, the images at the edges are very poor, and 

 only very small apertures can be used. It has, however, 

 been found that some of the best lenses for pictutes can 

 be obtained covering a circle of 20 3 diameter without 

 serious distortion, and at the same time large apertures 

 can be used, thus reducing the time of exposure. In 

 order to still further this work, Mr. W. H. Pickering in- 

 vestigated the sensitiveness of various phutographic 

 plates, and obtained some so sensitive that stars of the 

 fifth and sixth magnitude have been photographed without 

 using clockwork, they forming dots or making lines, as their 

 images pass across the photographic plate, the length of 

 these lines depending of course upon the time during which 



the plate is exposed. If the plate be exposed during ten 

 seconds a distinct dot is obtained, whilst an exposure of 

 thirty seconds causes a short line to be formed. The 

 plates used at Harvard Observatory are six by eight 

 inches. They are divided into six equal parts, each part 

 being in turn exposed. By this means six regions of the 

 heavens, each about 15 square, may be photographed on 

 one plate ; and by a variation in the dot and line system 

 employed, sometimes taking the dot and sometimes the 

 line first, three pictures maybe taken on a single division 

 of one of the plates without giving rise to any confusion. 

 Instead of simply six, therefore, eighteen photographs are 

 taken on one of these plates, so that on a single plate 

 a portion of ihe heavens of more than three hours' 

 right ascension, and extending from 30 S. to 6o° N., may 

 be included. Since each portion of the plate covers a 

 region of about 15°, the camera mounting has a series of 

 notches or stops, by which it may be instantly moved 

 through that amount either of right ascension or 

 declination. 



When photographing the following is the exact method 

 employed. The first exposure takes the region between 

 30°and 15 south declination, and between one hour and a 

 half and half an hour west of the meridian. First the plate 

 is exposed for ten seconds, and each star records itself by a 

 dot. The plate is then covered for ten seconds ; next it is 

 exposed for a period of thirty seconds, and each star makes 

 a line on the plate. By means of the clamping arrange- 

 ment to which we have referred the plate is then moved 

 through one hour in right ascension. This takes up the re- 

 maining few seconds of the minute, so that the taking of 

 the next photograph begins with the first second of 

 another minute. The camera is then on the meridian. 

 The same part of the plate is again exposed, and in order 

 to distinguish this series of stars from those first photo- 

 graphed, this time the plate is exposed first during thirty 

 seconds, and then during ten, so that the result is a line 

 followed by a dot. This gives the second series. But 

 the same portion of the plate may be again used. The 

 remaining ten seconds of the second minute, like those 

 of the first, are spent in moving the camera through 

 another hour of right ascension. Then a fresh ex- 

 posure is made for thirty seconds, a line simply being 

 obtained without a dot, and this completes the series. 

 The first class of images is in dots and lines, the second 

 in lines and dots, the third is recognised by the presence 

 of lines alone. The thirty seconds which remain of the 

 third minute are employed in exposing a second portion of 

 the plate, and changing the position of the camera, which 

 now takes in the region from 15° S. to the equator. The 

 same process is then gone through again, three exposures 

 as before being made in three different positions of right 

 ascension. By continuing this process, taking three 

 photographs on each of the six portions into which the 

 plate is divided, the whole region included between the 

 declinations of — 30° and + 6o°, and between three hours 

 of right ascension, ii hours on each side of the meridian, 

 bemg one eighth of the whole heavens, excluding the cir- 

 cumpolar stars, will be photographed on one plate, the whole 

 operation occupying but eighteen minutes. With regard 

 to those stars in the vicinity of the Pole, some othei 

 method will have to be adopted. Thus much for one 

 branch of the work — and an important branch — carried 

 on at Harvard Observatory. 



Another portion of their work consists in the prepara- 

 tion of a photographic map of the entire heavens. The 

 method just desciibed, in which clockwork is dispensed 

 with, only enables those stars whose magnitude is not less 

 than five or six to be photographed, and stars of a less mag- 

 nitude than this must of course be included in a map of the 

 heavens. The camera in this work, therefore, is driven 

 by clockwork. By this means stars of the eighth magni- 

 tude record their images on the photographic plate, and 

 as many as 200 are visible in the paper print within a 



