TRANSACTIONS OF SECTION A. 553 



ptobletn ; but when compared with the Cambridge observations, if the rotation sug- 

 gested by Sir David Gill exists, the plates taken early should give different results 

 for the Cambridge ' magnitude-equation ' from those taken later. The effective 

 interval available is only about five years (very small), and the etfective range of 

 magnitudes about three. But since this implies a difference of 0' 0012 x 5 x 3 

 = 0''018, the inquiry is not hopeless, even now. 



4. The comparison of a large number of plates gives a quantity about as large 

 as that found by Sir David Gill, but of the opposite siffn. If this sign is correct 

 the phenomenon cannot be a rotation. 



5. Assuming it correct, we can explain the facts by assuming that the stars 

 of the Milky Way are different in size from others, so different that where the 

 Milky Way appears in (say) the N. hemisphere faint stars, which we take to be 

 more remote than brighter ones, may be really nearer. The effect of the sun's 

 motion in space would then no longer be symmetrical about its apex, and the 

 disturbance would have opposite signs in the two hemispheres, since the Milky 

 Way appears in belts of R. A. on opposite sides of the sphere. 



4. A Standard Scale/or Telescopic Observations. By Percival Lowell. 



1. Present State. — At present there exists no criterion among astronomers for 

 the weight to be attached to any given observation due to the atmospheric condi- 

 tions under which it is made. Yet these atmospheric conditions are among the 

 most important factors entering into an astronomic observation. They are far more 

 to the point than the size of the instrument. For our telescopes have long since 

 outstripped the conditions under which they are put to work ; the great bar to 

 advance to-day, whether visually, photographically, or spectroscopically, being not 

 instrument but atmosphere. Each man realises this, but marks his own work on 

 his own scale, as if he should take his own foot as the unit of length. 



2. Difficulties of this Condition. — In consequence no absolute value is assign- 

 able to any man's work, and no comparison between different men's work is possible 

 whether in accuracy or credibility. The practical outcome is that the only test is 

 the test of time, and while the world is waiting for confirmation of any new result 

 just so many years are lost. 



As important is the incapacity to leave permanent records of observations 

 capable of being compared with newer ones as time rolls on. 



3. A Change necessary. — A change in this state of things is imperatively 

 needed. It is time a standard scale for observations were introduced similar to 

 what the metric system is, that it may do what that does for physics generally. 



4. Possibtlitij of a Criterion. — Until lately such a scale has not been feasible 

 owing to ignorance of the conditions upon which it must be based. Studies how- 

 ever directed to that end, first at Arequipa and then at Flagstaff", during the past 

 few years have resulted in the knowledge of the conditions which constitute good 

 or bad seeing, and have thus enabled an absolute scale to be constructed. 



6. The Criterio7i. — The basis of the matter lies in the discovery that systems of 

 ■waves traverse the air, several of these systems being present at once at various 

 levels above the earth's surface. The waves composing any given system are 

 constant in size and difl'er for the different currents all the way from a fraction of 

 an inch to several feet in length. If the wave be less than the diameter of the 

 object-glass from crest to crest, the image is confused by the unequal refraction 

 from the different phases of the wave. If the wave be longer than this a bodily 

 oscillation of the whole image results. The first is fatal to good definition, the 

 second to accurate micrometric measurement. 



It is possible to see these waves by taking out the eye-piece and putting one's 

 eye in the focus of the instrument when the tube is pointed at some sufficiently 

 bright light. It is further possible to measure their effect by careful noting of the 

 character of the spurious disc and rings made by a star and the extent of the swing 

 of the image in the field of view. By combining the amount of confusion with the 



