Nature's Operations which Man is competent to Study. 465 



very much simpler for what is really going on. However, 

 Group D is represented in our diagram as including another 

 sub-section, w, going 10,000 times farther still ; in order by 

 this extension to provide for the possibility of future dis- 

 coveries which we hope may some day be realised. 



Very little is known about the events going on within 

 chemical atoms, of which we have found that the range is to 

 be measured in tenthet-decimetres, tenthet- centimetres, or 

 tenthet-miilimetres, and even the fact that there are such 

 events lies near the limit of our knowledge ; and yet these 

 excessively minute quantities can be dealt w T ith accurately 

 when they present themselves as differences of wave-length. 

 This is truly astonishing, when we remember that we are here 

 measuring lengths that are from 100,000 to 1,000,000 times 

 smaller than the most minute interval that can be detected by 

 the microscope — as much smaller than a micron as a tenth or 

 hundredth of an inch is less than three-quarters of a mile. 

 Nevertheless these lengths can be determined with precision 

 because the position of a line in the spectrum depends on its 

 wave-length, and the difference of the wave-lengths of the 

 closest lines which can be photographed as double is exces- 

 sively small ; and again, because two rays with a still smaller 

 difference of wave-length may give rise to interference effects 

 which can be detected by the interferometer. By the spectro- 

 meter measures can be carried at all events as far as the 50th 

 of a tenthet-metre, i. e. as far as to one or two tenthet-centi- 

 metres, while with the interferometer determinations can 

 probably be carried one step of our scale farther, i. e. to one 

 or two tenthet-miilimetres. Here, for the present, our powers 

 end : and we cannot fail to be impressed b}' the extraordinary 

 accuracy which has been attained in measuring wave-lengths 

 by the methods spoken of above. It is a degree of accuracy 

 which ascertains the length of a wave of light within a 

 millionth of its entire length, thus equalling and even sur- 

 passing the best results obtained when comparing with 

 excessive care international standard yards or metres ; in 

 which a determination within one fif'thet (the 100,000th) of 

 the whole length is probably the most that can be fully 

 relied on. 



Group B {Planetary Intervals). 



We have next to direct our attention to Nature's operations 

 on a great scale, and first to Group B which deals with events 

 within the solar system. This group, like the others of our 

 survey, may conveniently be divided into sub-sections — u, r, 

 and iv. 



Phil. May. S. 5. Vol. 48. No. 291. Nov. 1899. 2 K 



