348 ANNUAL KEPOET SMITHSONIAN INSTITUTION, 1910. 



second uses a grating, two slits and a camera of the same length 

 as the first; the third has an arrangement differing from the two 

 preceding; finally, the fourth, the most powerful, has three slits 

 and prisms and grating. It consists first of a spectrograph having 

 a camera of 7 meters, and, as in the classical apparatus of Rowland, 

 allows the isolation of very fine lines. But its ordinary solar image 

 would require too long an exposure. It is therefore received by a 

 second spectrogTaph which reduces it to the desired size and eliminates 

 the diffused interior light. The final image of the sun is of any di- 

 ameter desired; and by means of a special contrivance it shows the 

 entire solar disk, a condition not fulfilled in other spectrohelio- 

 graphs of great dispersion. The customary diameters of the sun's 

 images are 6 and 4 centimeters. 



This apparatus, with the two spectrographs, has a total length of 

 14 meters, and under these conditions rests stationary. It is, indeed, 

 the first spectroheliograph in which all the parts except the photo- 

 graphic plate remain at rest. The movable parts, the plate and the 

 astronomical objective, are put in motion at the desired rate by 

 synchronous electric motors and transformers for special speeds. 



The agreement of the movements is assured by electrical means, 

 which do not depend upon distance, and this arrangement is pre- 

 sented as a general solution of the spectroheliographic problem. 

 Each of the four spectroheliographs has its special advantages, and 

 the passage from one to another may be made in a few moments. 

 The observer has thus at his disposal varied means for his investi- 

 gations. In a general manner the spectroheliograph of two slits and 

 a length of 3 meters has a large image, rich in detail. The three- 

 slit spectroheliograph of 14 meters gives, with a longer exposure, a 

 smaller image, but one much more pure (that is, inore monochro- 

 matic) ; it allows the isolation of the finest lines. 



The researches with this apparatus have been made by a young 

 astronomer of this observatory, M. d'Azambuja, whose name is 

 associated with mine. 



5. THE DISCLOSURE OF THE UPPER K3 LAYER OF CALCIUM. 



In 1908 we were able to isolate the narrow dark central line, K3, of 

 calcium, and therefore the upper stratum of that vapor. Figure 1, 

 which shows the K line and its components, will indicate the progress 

 accomplished. Until now the spectroheliograph used had isolated 

 the ensemble of the two bright lines (Ko), which include K3; the 

 slit width was then ninety one-hundredths Angstrom. The resulting 

 image, called by us the K03 image, was a composite of the layers K, 

 and K3, the much brighter Ko layer predominating. Now with the 

 great spectroheliograph we are able to isolate easily with slits of 



