6 BARNARD— PHOTOGRAPHIC OBSERVATIONS [April 25. 



much brighter and more definite. Their junction has bodily moved 

 outward for quite a distance. The north ray of the four has closed 

 in on the one close south of it. A broad light region on the south 

 edge of the northern of the two middle rays has drifted outwards 

 and is less marked. 



August 14. — There is one broad widening stream in the Juvisy 

 photograph, with two lesser ones symmetrically placed on each side. 

 In the Yerkes plate there is a general resemblance to the other; 

 though the tail is made up of three broad streamers, they are much 

 further out. It looks as if the three had drifted out and fused 

 together more or less. The whole system of tails has bodily receded 

 from the comet. 



August ip. — In the Yerkes plate the head has become relatively 

 smaller. The tail has spread out very greatly, especially on the 

 south side. There is less structure than in the Juvisy plate. 



August 20. — In the Juvisy plate a principal ray divides to the 

 north and joins a dark space behind the head. In the Yerkes plate 

 this ray and dark space have both moved outwards. The head and 

 neck are also narrower. 



On July II (which date we will treat specially) a bright con- 

 densation 1:^° back from the head is strongly shown on both the 

 Yerkes and Lick plates and can be seen on the Juvisy plate, but it 

 is faint and cannot be located with very great accuracy on this last 

 picture. A plate made at the Lick on July 10 seems to show this 

 same object somewhat nearer to the comet or about f as far out 

 as on the eleventh. It is noticeable on all three photographs of July 

 II that this condensation was receding from the comet, at the same 

 time that it was following slowly towards the sun. From the 

 appearance I am inclined to think that it is the same object which 

 is visible on the plate of July 10. If so, then it must have left the 

 comet on or about July 7. Between this condensation and the head 

 of the comet on July 11 the tail is very faint but continuous. In 

 reality this mass is the near end of a bright strip of the tail about 3° 

 long. The object on the Lick plate of July 10 is joined to the head 

 by a bright, strongly defined connection, of which the condensation 

 is only an inconspicuous part. In the interval between July 10 and 



