2 74 



NATURE 



[Au§- 5, 1875 



the instruments used, having especially in view the ex- 

 planation of the advantages of the horizontal telescope 

 and its origin. 



I should not have thought it worth while to make any 

 communication on this subject, but since it has been a 

 matter of discussion in the French Academy, and several 

 pamphlets have been written on the subject, it may not 

 be uninteresting to explain here the connection of the 

 Observatory of Harvard College with it. 



In the spring of 1869, when it became necessary to 

 begin preparations for observations of the Solar Eclipse 

 of August 7th of that year, my attention was called espe- 

 cially to the subject of Solar Photography for the first 

 time. Mr. Warren De La Rue in England, and Mr. 

 Rutherford in America, had devoted themselves almost 

 exclusively to astronomical photography for many years, 

 and they were the authorities on this subject. The 

 methods employed by them were substantially the same ; 

 each used an equatorial telescope with clock move- 

 ment, and enlarged the image formed by the object-glass 

 by means of another system of lenses, and photographed 

 this magnified image. Mr. De La Rue corrected his 

 object-glass to secure as nearly as possible the foci 

 of chemical and visual rays, and Mr. Rutherford cor- 

 rected his so as to obtain the best echromatic combi- 

 nation of the chemical rays without regard to the visual 

 focus. 



Mr. De La Rue first undertook a series of daily photo- 

 graphs of the sun at Kew some time previous to i860. 

 In making my own preparations for the Solar Eclipse of 

 1869, I collected what information I could about the 

 extent and brightness of the corona, and the nature 

 of Mr. De La Rue's photoheliograph which he em- 

 ployed in observing the total eclipse of i860 in Spain. 

 In the first part of the volume of 'the " Philosophical 

 Transactions" for 1869, I found a paper which was read 

 May 31, 1868, which contains the results of the first at- 

 tempt to measure the heliographical positions and areas 

 of sun-spots observed with the Kew photoheliograph. 

 From the examinations of the measurements in this 

 paper I became convinced that no trustworthy measures 

 of photographs taken in this way could be made. The 

 magnified image is so much distorted by the eyepiece, or 

 the equivalent system of lenses used to form the image in 

 the camera, that no satisfactory scale could be obtained ; 

 in fact the scale was found to vary irregularly from the 

 centre to the circumference of the image ; and even if this 

 irregular scale could be investigated, a slight displace- 

 ment of the centre of the picture from the axis of the 

 telescope would introduce confusion. Mr. De La Rue's 

 method of investigation consisted in photographing the 

 pinnacle of a pagoda which was composed of rings and 

 chains of known dimensions, and then attempting to find 

 the scales of the different parts of the pictures from the 

 images of this pinnacle. 



The result, as I have said, satisfied me that this was a 

 method to be avoided. The difficulty arising from the 

 distortion of the image, and the apprehension that the 

 light of the. corona might be so enfeebled by enlargement 

 that it would not make an impression on the plate, deter- 

 mined me to photograph the image in the principal focus 

 of the object-glass. 



All of the many other parties fitted out for photograph- 

 ing the total phase of this eclipse followed the method of 

 De la Rue and Rutherford ; the expedition from the Ob- 

 servatory of Harvard College was the only one that suc- 

 ceeded in getting a picture of the corona. The method 

 of De la Rue was employed in the preceding eclipse of 

 1 868, and no photograph of the corona was secured. I 

 mention these facts simply to show how little the disad- 

 vantages of enlarging the image by an eye-piece were 

 appreciated. In the next eclipse no party went into the 

 field with De la Rue's plan ; every one of them photo- 

 graphed in the principal locus, but this time, on account 



of the weather, the American party in Spain alone suc- 

 ceeded in getting the corona. 



In 187 1, in India, this method was again followed by 

 all the parties, and was successful. My preparatory ex- 

 periments in 1869 were made with an equatorial of 7 feet 

 focal length, which gave an image of about three-fourths 

 of an inch, and with the great equatorial of 24 feet, which 

 gave an image of 2} inches diameter. 



Measurements o! photographs of the smaller image 

 seemed to indicate that under a microscope an accuracy 

 comparable with that of the best meridian circles was 

 attainable ; but believing that a larger image would be 

 better, I thought that four inches would be a convenient 

 size. In order to get such an image free from the distor- 

 tion of the eye-piece, I must have a telescope 40 feet in 

 length. Immediately on my return from the eclipse of 



1869, I ordered a lens of Messrs. Clark and Sons, of 40 

 feet focus, and a micrometer capable of measuring con- 

 veniently an image of four inches diameter. Thus the 

 long telescope was adopted to escape the distortion. 



Then of course the difficulty of mounting and handling 

 a telescope of this length, especially when extreme pre- 

 cision in measuring was the main object, naturally pre- 

 sented itself. To obviate this I resorted to the very 

 simple expedient of placing the telescope horizontally, so 

 that it need not be moved at all, and reflecting the light 

 of the sun through it by means of a plane mirror. 

 This seemed likely to meet all of the difficulties of the 

 case ; the well-known methods of mounting and directing 

 collimators rendered the utmost degree of accuracy at- 

 tainable in directing such a telescope, and by putting the 

 object-glass on one pier and the camera on another, using 

 a tube which should touch neither, only for excluding the 

 light, all disturbance of the focus by the expansion of the 

 long tube was avoided. 



Other information obtained by my preparatory experi- 

 ments had an important bearing upon my plan at this 

 time. I had found how difficult it was to get an exposure 

 of the plate short enough. It became necessary to reduce 

 the apertures of the equatorials to one or two inches, 

 and then throw the slide across with a strong sprmg. 

 From this I derived two important suggestions : one, 

 that a heliostat driven by clockwork was not indispen- 

 sable, as the picture would be instantaneous, so that the 

 motion of the sun during the exposure would be of no 

 consequence. The other was that I might reduce the 

 light by using a transparent glass reflector, and not be 

 compelled to reduce the aperture of the telescope so much. 

 By these means the cost of the experiment was greatly 

 reduced, saving the expense both of a heliostat and of 

 a silvered mirror. Messrs. Clark and Sons did not get 

 the apparatus ready for use at the Observatory until July 



1870, although it was tried at their shop previously. 

 A series of daily photographs was begun with it, July 4, 

 1870, and has been kept up with little interruption to the 

 present time. 



At this time and for a year or two after, I had not 

 heard of this method being thought of by any one. No 

 one of my acquaintances seemed to have any knowledge 

 of any other claimant of the method. Mr. Rutherford, 

 with whom I had frequent communication, and who had 

 been occupied with the subject for twenty years, regarded 

 it as new and original. It was described in Mr. 

 Lockyer's paper in 1870, and attributed to me; Mr. 

 Newcomb, in the latter part of 1872, speaks of it 

 as a method devised by me, and in successful opera- 

 tion for several years, and also independently proposed 

 by Faye. Lord Lindsay adopted it for his expedi- 

 tion to the Mauritius. Mr. De La Rue in several com- 

 munications down to 1873 spoke of it as the method of 

 the American Astronomers. It was afterwards, about 

 this time, called the method invented by Foucault and 

 Prof. Winlock independently. Then, in 1873, I received 

 a book by M. Edmond Dubois, claiming it as a French 



