150 



KNOWLEDGE ♦ 



[May 1, 1888. 



I must express the fullest agreement : only, it has always 

 seemed to me that when an observer of established reputa- 

 tion has made a series of striking observations, respectful 

 attention involves careful examination of his work. If I 

 were myself devoted to observation, and \mable (as is the 

 case with many of our best observers) to give much time to 

 analysis, I should not be very grateful to those who claimed 

 the acceptance of my observations precisely as they stood, 

 without inquiry into their significance, or such due com- 

 parison intei- se and with observations made by others as 

 would be essential to their satisfactory interpretation. To 

 this I may add that the basis of such analysis should always, 

 in my opinion, be admitted analogies. I know of no 

 theoi-y now accepted as sound which ever had any other 

 foundation. 



I regard Schiaparelli's observation as one of the most 

 interesting ever made by the telescopist. At the same time, 

 I consider his " double canals " as having no existence in 

 nature. " But," says the purely inductive philosopher, 

 " he has seen them, and therefore they must be objective 

 realities." Hevelius said the same of the star-discs which 

 his telescopes showed him, and became quite angry (for an 

 astronomer) when his measurements of those discs were 



Fic. 2. — Construotiou for determining tlic axial pose of Mars on 

 tlie meridian, Jlay ], 18S1. 



lookeA on with doubt. If his telescope had been a little 

 better he would have seen not only seemingly well-defined 

 and measiu'able star-discs, but a series of very obvious and 

 unmistakable rings round each star. And probably the 

 more simply inductive philosophers of his time would have 

 been indignant (philosophically, of course) at those who 

 would not admit the enfeebling, to say the least of it, of 

 long-admitted solar analogies, and give such respectful 

 attention to the observations of the eminent astronomer as 

 would have been involved in the supposition that other 

 suns than ours have immense rings round them, so situate 

 through some strange influence that they are none of them 

 foreshortened, but all appear exactly circular. 



The interest of Schiaparelli's observations resides (for me) 

 in the circumstance that they indicate the existence of an 

 analogy between Mars and the earth, which, though long 

 suspected, had never before been demonstrated. Unques- 

 tionably not duplicate canals on Mars, nor phenomena due 

 to our own atmosphere, the double marks were as unques- 

 tionably seen ; and we need not bo in a hurry to say that 

 " explanation is set at defiance," since there is nothing 

 suggestive of inherent inexplicability in these appearances. 



On the contrar}-, we seem guided easily towards an inter- 

 pretation which promises to remove every ditticulty. 



Parallel lines apparently seen where no parallel lines can 

 be reasonablj' supposed to be, suggest certain optical phe- 

 nomena in which we see parallel lines as the optical images 

 of lines really single, circles as the optical images of points, 

 and other optical products — as they may be called since they 

 are not optical illusions — which by no means correspond 

 with the real nature of the object under obsei'vation. The 

 optical image of an exceedinglj' fine bright line on a rela- 

 tively dark ground observed through a telescope is a broader 

 bright line, on either side of which run two parallel lines 

 much less bright, and outside these again other still fainter 

 lines. Under particular conditions of relative lustre in the 

 source of light the eye would recognise only the two rela- 

 tively dark parallel bands on cither side of the median 

 bright streak of measurable width (depending on the aper- 

 ture of the telescope employed). 



Is it not, on the whole, more likely (to say the least of it) 

 that what Sohiaparelli has taken for sets of double canals 

 are simply the two relatively dark streaks on either side of 

 the bright diffraction images of exceedingly fine luminous 

 streaks on Mars, than that those objects arc really "double 

 canals" on toe enormous scale imagined, or that the 



Fio. 3.. 



-Meridians atid lalitude-pamllels of Mars rn the meridian, 

 May 1, 1888, for an inverting telescope. 



phenomena are otherwise wholly inconsistent with " admitted 

 terrestrial analogies 1 " 



The varying aspect of these objects would be readily 

 explained on this supposition, nay, wculd be a necessary 

 consequence of the hypothesis that the luminous streaks ai'e 

 Martian rivers. Moreover, the change of ajipearance would 

 be likely to occur at precisely the season when Schiaparelli 

 has found that it does occur. 



Rivers on Mars would be fivr too delicate objects to be 

 seen with their own proper outlines even with the most 

 powerful telescopes, whether they were bright streaks on a 

 darker ground or dark streaks on a lighter ground. It is 

 not likely that there is any river so large as the Nile, probably 

 there is not one so large as the Danube or the Volga on the 

 ])lanet Mars, and a river would have to be at least twenty 

 times as wide as the Nile, taking the average of the lowest 

 five hundred miles of its flow, to be seen with the telescope 

 at Mars's distance. Usually a river if visible would be seen 

 as a dark streak on the lighter background of the continent, 

 just as the Martian seas are usually darker than the lands. 

 Such a river would be perceptible, though not actually 

 visible with its true outlines, through a good telescojie, as a 

 dusky streak many times broader than the river itself. But 



