Validity, and Residual Affinity. 311 



forms " recognizable by the notation 



£££.££ 



or even -, __. ■ i i 



3 H d 3 J 



corresponding to the five trivalent groupings 



^"a> X ft , X ft| , X A(( , X y 



of my previous paper. Thus we have an organic compound 

 radical, long viewed as one of the " elements " of organic 

 chemistry, possessing several clearly distinguished allotropic 

 forms, as it were. On the other hand, C 2 H 5 is, so far as our 

 strong evidence goes, a genuine organic " element " without 

 such forms. May there not be some analogous mystery in 

 the case of our present elementary radicals which has yet to 

 be solved, and of which, perhaps, the spectroscope — as fore- 

 shadowed by the almost uncanny researches of Dr. Crookes 

 and Prof. Norman Lockyer — already gives us indications ? * 

 Thus Prof. NordenskiokTs experiments indicate that the crude 

 mixture ,of yttria, erbia, ytterbia, &c, which Dr. Crookes 

 briefly calls gadolinia, has a constant molecular weight, just 

 as X a , Xp, Xy has. Yet the oxide of gadolinium is known to 

 be the oxide of a non-simple body. Meanwhile there is surely 

 enough to cause us to hope for the further great future of what 

 seems to be to some the untenable, if not exploded, theory of 

 valency. 



Oxford, 28th January, 1888. 



POSTSCEIPT. 



The importance and difficulties of this subject of validity is 

 seen by Professor Lothar Meyer devoting over fifty pages to 

 it in his ' Modern Theories/ which I have, since writing my 

 two papers, re-read with deep interest after several years' 

 interval. It may be convenient to note briefly our chief 

 differences of opinion. 



(1) Oxygen is, as usual, taken as strictly divalent. The 

 result appears to be that chlorine is not only mono- and triva- 

 lent, but "probably heptavalent with regard to negative 



* Kemembering that oxygen is characterized above all other elements 

 by its extreme chemical sociability for others, it is noteworthy that if we 

 add its atomic weight to the atom-weights of the seven elements of the 

 first period, we obtain those of the second period ! 

 Y 2 



