APPENDIX IV. 127 



benzene vibration still exists, and that solids in solution can be studied 

 in the infra-red, just as in the ultra-violet. 



Naphthalene. CioHj. 



Naphthalene was also examined at 3.25 fi. Its solubility in CCI4 is 

 unusually great, e. g., the present solution contained 0.25 gram per cubic 

 centimeter of CCI4. This is sufficient to make a homogeneous solid 

 film 0.28 mm. thick (sp. gr. = 1.15 ; cell = 0.6 mm.). 



This compound is of considerable importance, for its molecule is 

 formed by the condensation of two benzene nuclei. As a result, its 

 chemical properties are quite different from that of benzene ; but its 

 absorption spectrum shows the 3.25 fi band of benzene, with which it 

 is also comparable in general transmission. As a whole, for the region 

 investigated, this compound shows that the vibration of the benzene 

 nucleus has not been disturbed. 



AZOBENZENE. CeHs — N=N — CeHs. 



This compound is also quite soluble in CCI4, so that a saturated solu- 

 tion contained about 0.2 gram per cubic centimeter of the liquid. As 

 with the preceding compounds, the rock-salt cell was 0.6 mm. thick. 



The curve shows the 3.25 fi benzene band, indicating that the presence 

 of the N atoms does not disturb the vibration of the benzene nucleus. 

 This solution is far more opaque than the preceding, showing the effect 

 of the introduction of the N atoms, just as in other compounds contain- 

 ing nitrogen. 



Moreover, this solution is of a reddish-brown color, showing an 

 absorption band in the visible spectrum, as well as in the infra-red. 

 This shows that both are due to an intramolecular disturbance, and dis- 

 approves the idea of a resonance of small particles (of molecular dimen- 

 sions, as for, example, Kosonogoff's butterfly scales and Wood's metal 

 films) in the optical region, as distinguished from the intramolecular 

 resonance in the infra-red. Of course, one might say that the infra- 

 red band, e. g., the 3.25 ju, band, is also a resonance effect. Possibly it 

 is, but the evidence of a resonance of small, electrically charged parti- 

 cles whose capacity, and hence whose periods (like a condenser) , depend 

 upon their proximity to similarly charged particles (the closer the par- 

 ticles the greater the capacity, hence the slower the period, and hence 

 the farther is the absorption band shifted toward the longer wave- 

 lengths), is somewhat contradictory in the infra-red. 



The best evidence in favor of this idea of a resonance effect is in the 

 xylenes, in which for the ortho, which has the two CHg-groups nearest 

 together, the large absorption bands lie farthest toward the long wave- 

 lengths, while in the para, in which the two CHg-groups are farthest 



