134 COLLIGATWE RELATIONS AND SCIENTIFIC LAWS 



that, under specified conditions, emits light of "wavelength" A: for 

 the Ka X-ray line we can then write A (Z — 1)- = 1.21 X 10~^. This 

 seems so purely theoretical, so abstract, that our first obligation is to 

 show that it can function as a coUigative relation. 



By 1869 the concept "relative atomic weight" had acquired a rea- 

 sonably defined denotation. Given certain chemical data, one could 

 compute "relative atomic weights"; given atomic weights, one could 

 predict certain chemical data. Beginning in 1869 Mendeleev, Meyer, 

 and others developed a classificatory chart in which, with a few 

 specific inversions, the chemical elements are arranged in order of 

 increasing atomic weight. The "atomic number" of an element is 

 simply the ordinal number expressing its position in the sequence of 

 the perfected classification. Giving exceedingly compact expression 

 to a great many relations among measurable properties of the ele- 

 ments, Mendeleev's "periodic table" is in part "theoretical," e.g., in its 

 implicit acceptance of the Daltonian atomic theory; and it has very 

 pronounced connotations, e.g., the nonultimacy of the chemical "ele- 

 ments." But through it the concept of "atomic number" acquires a 

 clear enough denotation. Given certain measurements made on an 

 element and its compounds, I can assign it an ordinal number and a 

 place in the chart; given the chart and the number of some element, I 

 can state the probable outcome of certain experiments not before 

 made on it. 



Turning now to the "wavelength of light," in the X-ray region a 

 denotation is fairly readily established with an instrument like the 

 crystal reflection spectrograph. The "wavelength" of light passed into 

 the instrument is calculated in a specified way from the location(s) 

 at which blackening is produced on a photographic plate that forms 

 part of the spectograph. An immense amount of theoretical work un- 

 derlies the instrument and the calculation. But, once we accept this 

 denotation, the concept has a perfectly clear linkage with experience. 

 \Mien I have seen the plate I can assign a definite wavelength (a) 

 to the light which produced a black mark on it; and when I predict a 

 certain \'alue of A I say where on the plate I expect to see a black 

 mark. 



Now we can come to grips with Moseley's law. We "excite" an 

 element under the specified conditions and, passing the light so 

 generated into our spectrograph, we find a determinate relation be- 

 tween the ordinal number of that element in the periodic classifica- 



