102 BEPOET OF THE SECEETARY. 



left is a portion of that of the Yerkes Observatory. The 135-foot horizontal telescope 

 is under the canvas seen on the right. Its companion tube is hidden by it. On its 

 left extremity is seen, though on a very diminished scale, the coelostat mirror which 

 "fed" it (this is shown on an enlarged scale in the next illustration), and also the 

 box of the O-inch aperture camera equatorially mounted. Immediately in front of the 

 tube and cutting off a portion of the view are the sheds containing the apparatus for 

 the bolometric study of the corona and that for the large cameras for the 11-foot and 

 other lenses. The great Grubl) siderostat and other i:)ieces of apparatus are liidden 

 from view l)y the sheds. On the right of the sheds and immediately in the fore- 

 ground is the 5-inch ac-hromatic loaned by the United States Naval ObservatorJ^ 

 At the right hand extremity of the long tube is seen the photographic house, which 

 serves equally for the 135-foot lens and for that of the 38-foot focus lens, which latter 

 is in the tube inclined upward. 



Nature of the uhservation.s. — The chief aim of the ol^servations was the investigation 

 of the corona, and of this especially the inner portions. This investigation was three- 

 fold — photographic, bol(jmetric, and visual. In addition to these main objects there 

 was included the ])liotography of the sky near the sun for the discovery of jiossible 

 unknown bodies, an attempt to i)hotograph the "flasli spectrum" with an automatic 

 camera, and the observation of times of contact both by the ordinary visual methods 

 and by photography. 



Pieces of apparatus employed. — (a) Apparatus for photographic purposes: For the 

 direct jjhotography of the inner corona the 12-inch lens of 1,35 feet focus was used as 

 a horizontal telescope in connection with a cirfostat carrying an 18-inch plane mirror. 

 (Plate X.) Here^ire shown on a large scale the lens in (piestion, the extremity of 

 its c-anvas tuT)e, and the ccelostat and equat(jrially mounte<l camera descri])ed in the 

 preceding plate. It was also arranged to use with this lens just before second con- 

 tact an ol)jcctive prism forming a spectrum upon a plate moved each second by clock- 

 work, and tlaus suited to catch the "flash spectrum," so called. This necessitated a 

 second tube 135 feet long, inclined at about 8° to that used for the direct jihotographs 

 of the corona. Both tul)es were made of black canton flannel and were 42 inches 

 square with diaphragms of progressively increasing size 10 feet ajjart. Tlie two tubes 

 were fastened to trestlework and were covered by long canvas A tents. Nearly 1,000 

 yards of canvas and flannel were thus made up. The direct tube ended in a small 

 photographi(; house which had been prepared in sections and was transported from 

 Washington. Plates 30 inches S(iuare were here exposed. 



The direction of the long tubes was necessarily a matter of care, as they had to be 

 placed ])eforehanil wliere calculation showed the sun would Ije, and were incapable 

 of adjustment. 



Besides this gri-at liorizontal telescope, the 5-inch 3S-foot lens was als<j used for 

 obtaining inner coronal iihotographs. This lens was mounted u[)on a j^ole in such a 

 way as to be in line with the sun from the east window of the pliotographic house at 

 the moment of totality. A conical tul)e of white canvas, well blackened witliin, and 

 36 inches scjuare at its lower end, ran from the house to the lens, but was not attached 

 to tlu^ lens or its mounting. There being thus no provision for following tlu^ ai>par- 

 ent motion of the sun with the lens, a suitable^ motion was given to the j)hotograi)hic 

 plates by means of a water clock. With this instrument 11 by 14 inch plates were 

 emjdoyed. 



Upon the same instrument which carried the 1 8-inch mirror of the horizontal tele- 

 scope (shown in Plate X) was mounted equatorially a 0-inch i)hotographic lens of 

 1\ feet focus, provided with a conical tube, so that a considerable field was covered.^ 



' The focal curve of this lens was determined and it was intended to use a nest of 

 small plates so arranged as to be in focus over a large field. On the night ]:)efore the 

 eclipse;, however, it was so warm that the wax used to fasten the plates softened 

 repeatedly, and after several trials it was found necessary to use a flat plate, on which 

 the focus was good for perhaps 6° from tlic center. 



