May 2, 1892.] 



KNOWLEDGE 



93 



binding connection ? it may be direct, it may be indirect 

 and secondary only, but it must be real and effective. 



Consider that the period in question is practically some 

 0800 days. A magnetic storm does not last many hours ; a 

 Sunspot soon declines from its greatest development, or 

 soon passes away from the centre of the apparent disc. 

 Suppose we take an outside hmit, and give a period of two 

 days to a giant spot to exercise its influence, or a magnetic 

 storm to expend its violence ; what are the probabilities 

 against 3 out of 3400 of such periods of the one phenomenon 

 agreeing with 3 out of 3100 of the other, if they are not 

 related ? If 3400 numbers were placed in one box, and 

 3400 more in a second, and one from eacb box were 

 drawn at a time, what is the chance that the three highest 

 numbers would be drawn from the one box, simultaneously 

 with the three highest from the other, each to each, if the 

 matter had not been pre-arranged"? Indeed, we might 

 legitimately call the coincidence of April, 1882, a double 

 one, and ask the odds against the four highest numbers 

 from each box being so drawn. 



Between Sunspots and storms of the second magnitude 

 it is more difficult to make a satisfactory comparison, 

 because it is not so easy to frame a satisfactory definition 

 as to what constitutes a secondary disturbance. Never- 

 theless, the following brief table of large Sunspots seen 

 since the beginning of 1881 which were coincident with 

 considerable disturbances may prove of interest. The 

 spotted area is given in millions of square miles. 



and the general magnetic cycle is clearly established, 

 though even in minor irregularities the two curves closely 

 correspond, and though unusually large Sunspots are 

 answered by unusually violent magnetic storms, we cannot, 

 as yet, proceed farther and express the magnitudeorcharacter 

 of the magnetic disturbances in terms of the spotted area 

 of the sun, or of its principal groups at the time of 

 observation. The conclusion to my own mind seems to be 

 that though Sunspots are the particular solar phenomenon 

 most easily observed, we must not therefore infer that 

 their number and extent afford the truest indication of the 

 changes in the solar activity which produce the perturba- 

 tions we remark in our magnetic needles. 



Some of the above, those marked with an asterisk, may 

 fairly be taken as confirming, though with less definiteness, 

 the conclusion drawn from the correspondences between 

 the greatest spots and the greatest storms. But with the 

 others it is not so. Spots as important have been seen 

 upon the sim, and the magnets have scarcely fluttered, 

 and storms as distinct have occurred when there have 

 been only few spots, and those but small, upon the visible 

 disc of the sun. The table is important therefore, not as 

 adding to the weight of the evidence in favour of the con- 

 nection between Sunspots and magnetic disturbances, but 

 as emphasizing a point which must not be forgotten. 

 Though the diurnal and annual changes of terrestrial 

 magnetism conclusively prove the solar influence upon it, 

 though the connection between the general Sunspot cycle 



Utttcr. 



[The Editor Joes not hold himself responsible for the opinions or 

 statements of correspondents.] 



DARK NEBULOUS MASSES. 

 To the Editor of Knowledge. 



Dear Sir, — I see that you take for granted the existence of 

 dark nebulous masses. It appears the easiest hypothesis to 

 account for the curious features shown in the Milky Way ; 

 and the evidence for it seems strong, but is it yet sufficient, 

 or may there not be some other way of accounting for the 

 dark features '? The fact that the dark regions are so often 

 closely surrounded by stars shows that there are peculiar- 

 ities in the arrangement of the stars with respect to the dark 

 areas. Is it not, therefore, possible that the dark regions 

 . may be simply spaces for some reason devoid of stars ? 



With reference to the tree-like dark object — -(flower-like 

 I should be more inclined to call it) — referred to on page 

 230 in your December number, there is one feature which 

 is very striking to me that j-ou show only partially, and 

 that is a straight line extending in a slanting direction 

 from the bottom of the stalk to the left. You draw the 

 portion of the line to the right. The large photograph in 

 the December number shows this feature very distinctly as 

 three straight parallel lines of stars, of which the two 

 outer ones are by far the plainest. The passages between 

 them are very slender, the northern one so slender as to 

 give one almost the impression of a scratch on the 

 photograph. This shows the advantage of having copies 

 from more than one negative to compare, as this feature is 

 also visible in the photograph in your October number, 

 only not quite so distinct. Other less distinct lines 

 emanate from the structure, and give to my eye quite an 

 appearance of radiation, which your drawing, Fig. 1, 

 December number, only partly indicates. Nearly parallel 

 with the straight line from the lower part of the flower- 

 like structure but detached from it, there is another slender 

 dark straight line, also visible on both your photographs. 

 These dark lines remind one of the straight nebulous lines 

 in Mr. Roberts' photograph of the Pleiades. 



Yours faithfully, 



Sunderland. T. W. Backhoise. 



[A close exammatiou of the photographs shows that the 

 density of the photographic action has been less within 

 the dark areas referred to than over surrounding and 

 neighbouring regions. This shows that the appearance of 

 darkness is not an effect of contrast, and that there is 

 either less nebulosity within the regions surrounded by 

 stars, or that there is some opaque or semi-opaque matter 

 within the dark regions which cuts down the light of the 

 background. The dark areas are of various depths of black- 

 ness, as if some were due to masses of faint luminosity, while 

 others were due to an opaque fog in space. — A. C. Kanyard.] 



