70 ^^6 Philippine Journal of Science i9i6 



tive absorption of light by organic compounds, a ready explana- 

 tion of the occurrence of selective absorption of light in dihy- 

 droterephthalate is at hand. The double bonds of the ring are 

 fixed by the addition of a molecule of hydrogen to the terephtha- 

 late; consequently no selective absorption of light waves can 

 arise from the vibration of the bonds in the benzene ring, but 

 must arise from oscillation of the double bonds of the carbonyl 

 groups. 



Baly and Collie ^^ state in regard to nitrobenzene — 



It seems probable that this very great absorption is due to some form 

 of strained condition having been set up in the benzene molecule until the 

 making and breaking of linking between the carbon atoms has ceased in 

 nitrobenzene, and it seems not unlikely therefore that the presence of the 

 active residual affinity as possessed by ketonic oxygen in the position 

 tends to lock up or restrain the internal motions of the benzene ring. 



The theory of partial valency as formulated by Pratt has the 

 advantage of being simple enough to be understood readily, of 

 portraying an easily grasped picture of the condition in the mole- 

 cule, and of being broad enough to explain all the recorded 

 instances of compounds possessing selective absorption, while 

 the explanation of Baly and Collie, quoted above, presents a 

 much more complex picture, which would be grasped only with 

 exceeding difficulty and would not readily explain the occurrence 

 of selective absorption in such compounds as dihydroterephthalic 

 ester and diketohexamethylene. However, even they are agreed 

 that the carbonyl oxygen has an important influence on the selec- 

 tive absorption of the compound. 



To determine if the carbonyl groups alone possess the quality 

 of selective absorption of light, diketohexamethylene, in which 

 the bonds of the ring are all fixed, thus precluding their in- 

 fluence on the selective absorption, was photographed in the 

 light of the iron-nickel arc. As predicted by Gibbs and Brill *• 

 in accordance with the theory of partial valency, this compound 

 shows selective absorption giving a refrangible band heading 

 at 1/a=3,500 and does not show only general absorption as noted 

 by Hartley." 



Baeyer and Noyes " report that diketohexamethylene reacts 

 with acetylchloride in ether solution, forming a white preci- 

 pitate in the presence of sodium alcoholate and that, therefore, 

 it may exist in two forms, the enol and keto. This acetyl com- 



"Journ. Chem. Soc. (1905), 87, 1340. 



"Loc. cit. 



"Journ. Chem. Soc. (1898), 73, 598. 



"Ber. d. deutsch. chem. Ges. (1889), 22, 2168. 



