18 



♦ KNOWLEDGE ♦ 



[November 1, 1888. 



ing combinations. Consequently the investigator, though 

 he may divide the cryptogram into groups of two letters, 

 will not succeed in getting at the actual combinations used, 

 but will get others quite different and of no use for his 

 purpose. If an investigator cannot individualise the cha- 

 racters of a crvptogiam, how is he to proceed to analyse it? 

 To further guard against certain vulnerable points of crypto- 

 grams in general I employ the following simple de\'ices : — 



In double letters the second letter is written by the 

 " Eepeater." Thus, tt=asv.v. E has two .signs, which are 

 to be used alternately. " End of Word " has three signs, 

 which are to be used alternately. Tlie words "And, The, 

 You " are each expres.sed by a single .sign. This also saves 

 time in writing. 



Now for a message. " We are prepaied to act " will be 

 written thus : — 



No. 3. sacoauvoscvssocuoucvcoouo 



scvssavsusasuuauoosoaasocx; 



This may at first appear longer than a single-letter system, 

 but it is really quicker both to write and to read, because 

 in the single-letter systems the sender must write down the 

 key underneath the whole of the message on a separate 

 piece of paper before commencing his work, and the receiver 

 must likewise write down the key underneath the whole of 

 the cryptogi-am before he can begin to decipher it. By my 

 method, however, the sender, with his code before him, can 

 jot down the message straight off without any preparation, 

 and the receiver can decipher it in a similar way with even 

 greater speed. 



There are some other devices and variations which may 

 be introduced into the system, but I have perhaps said 

 enough to explain the modus operandi. 



By-the-by, Bacon's five-letter cipher did not fulfil his 

 own condition. It was on the fixed-sign principle, and was 

 decipherable with but little difficulty. 



Edward Anderson. 



ON A GREAT BRANCHING STRUCTURE 

 IN THE CORONA OF AUGUST 29, 1886. 



IMONGST the photographs obtained by the 

 American and English observers who 

 visited the island of Grenada to observe the 

 eclipse of August 29, 188G, is one which, 

 according to the report of Professor W. H. 

 Pickering, shows some enormous structures 

 extending to a distance of nearly two solar 

 diameters from the .'iun's limb. 

 The most remarkable of these is a narrow jet in the south- 

 west quadrant which starts from the centre of a gi-oup of 

 rays which bend together forming what has been termed a 

 synclinal group, the outer rays of which are inclined inwards 

 from the i-adial, and then slightly curve together so as to 

 form a hollow cone. There are many such groups on photo- 

 (Traphs which have been taken during other eclipses, but 

 what is exceptional about this particular group is that from 

 its summit there rises a narrow nearly radial jet which 

 extends without curving to a height of forty-eight minutes 

 of arc (more than a solar diameter and a half), and then divides 

 into three branches, one of which falls over with a crook-like 

 curve towards the western etjuatorial region, another bends 

 over towards the South Pole, and the third reaches an 

 elevation of sixty minutes, and then falls back towards the 

 equatorial region. To the south of this gi-oup is another 

 oi^antic jet, which starts with an inclination of twenty 

 de<T«es or more from the solar i-adius, and bends over 



as if it belonged to the same synclinal group, or was under 

 the influence of similar forces to those which caused the 

 bending of the lower rays. This great ray bends over until 

 it appeai-s to cro.ss, or come up to, the hooked ray just below 

 the level at which it branches. The diagram is made from a 

 drawing which Professor W. H. Pickering kindly sent me 

 in a letter shortly after his return. 



The larger scale drawing (fig. 2) is made from the 

 frontispiece of the lieport of his Expedition recently 

 published. Unfortunately, there is only one photograph 

 showing this remarkable structure, but Mr. W. H. Wesley 

 has in the last number of the Obsfrvatori/ put together 

 the outline drawings of this corona made by three 

 observers, which show that as .seen with the naked eye 

 there wa-s a great projection at this part of the corona, and 

 the photographs of the English observers, Mr. Maunder and 

 Professor Schuster, corroborate Professor Pickering's photo- 

 graph as far as they go, that is, they show the synclinal 

 group of structures at the base of the great ray, but the 

 English photographs were taken with longer focussed 

 instruments than that with which Professor Pickering's 

 photogiaph was taken, and \he image thrown upon the 

 pliotugrapbic plate was Ies< brilliant though larger, conse- 



FlG. 1.— OniLINE FROM I'ROFESsOE PICKERING'S ORAWIM,. 



quently the upper and fainter parts of the corona which 

 registered itself with Professor Pickering's instrument can- 

 not be traced upon the English photographs. 



The structures in question are of a similar character to 

 coronal structures shown on other photographs ; for example, 

 the English photographs taken in India during the eclipse 

 of 1871 all show a hook-shaped ray, which, however, does 

 not extend to an altitude of more than eleven minutes above 

 the solar limb, where it divides into two branchas which 

 curl over in opposite directions. The main stem or jet is 

 not i-adial to the sun's limb, but is inclined to the normal at 

 an angle of more than twenty degrees. This curious jet, the 

 photographic evidence with respect to which is quite conclu- 

 sive, is shown in fig. 3. Its outline is somewhat hazy, as is 

 the case with the jet shown in Professor Pickering's photo- 

 graph, and it Ls seen as a somewhat brighter object on a 

 nebulous background. Though it appears small compared 

 with Professor Pickering's jet, it extends to an altitude 

 above the sun's limb gi eater than the moon's distance from 

 the earth, while the summit of Professor Pickering's jet 

 reaches an altitude of more than a million and a half miles 

 above the solar surface. My intention in this paper is to 



