194 Dr. a. Hodgkinson on 



or to interference. Both these phenomena are known to be 

 dependent on structural arrangements of a magnitude com- 

 mensurate with the wave-length of light, and light tends to 

 break down, so far as its image-forming capabilities go, 

 when acted on by structures of such small dimensions. 

 Now the microscope is essentially dependent for its effect 

 on image formation, and hence the possibility of its inade- 

 quacy under the above conditions. Failing other methods, the 

 microscope has been employed, and that by most careful and 

 reliable observers, for the purpose of determining the cause 

 of colour in many of the most striking colour-producing 

 structures, e.g., iridescent feathers, innumerable species of 

 gaudy insects, opal, mother of pearl, and the like. In 

 the varying results of these observations we have one of 

 the most convincing proofs of the inadequacy of the 

 microscope alone to reveal this cause. A few instances of 

 these varying results may be cited out of innumerable 

 examples. The changing colours of the opal are by one 

 observer attributed to a structure of fine lines, by another to 

 thin plates, whilst a third holds them due to both these 

 causes. Again the iridescent hues of the feathers of hum- 

 ming-birds, sun-birds, and various other tropical birds, as 

 also the brilliant and changing tints of innumerable insects 

 are, by almost all observers, considered due to a structure 

 of fine lines. In some of these instances, as for example in 

 the case of scales from the wings of Lepidoptera, and from 

 the elytra of some beetles, and various parts of other insects, 

 the fact that lines andmarkingsdo exist, as we shall presently 

 see, would seem at first sight to confirm this assumption. 

 Whilst fully recognising the existence of lines and markings 

 in these and numerous other instances, and whilst admitting 

 that the most brilliant diffraction colours are produced by 

 them, facts will be adduced to show that the colours pro- 

 duced by these lines and markings are either imperceptible 

 in the natural condition of these objects, or, if apparent, so 



