260 



The Growth-Rings on Herring Scales. 



polarised light (in conjunction with certain stains) it is this lower lamellar 

 layer (not, as in the previous case, the' upper striated layer) which focusses 

 our attention. Under these conditions the scale appears to be made up of a 

 number of concentric bands alternately dark and light. 



These bands represent the outcrops of successive lamellae and are not of 

 equal width but show distinct zoning. Further, if suitable comparisons be 

 made, the limit of the narrow zone will be found to correspond with the 

 transparent ring, just as did the narrow zone of the concentric stria? on the 

 dorsal scales ; whilst, in section, the lamella? which come to the surface at 

 this point not only show very narrow outcrops (as we should expect from 

 our surface view), but show diminished bulk as well. It is clear, therefore, 

 that at the time of the formation of the ridge of the upper layer, the lamellar 

 development of the lower layer was at a minimum. 



The physical explanation of the light and dark bands seen under polarised 

 light is to be found in the minute structure of the lamellar layer. The fibres 

 which compose this layer are arranged in two distinct series, one radial, the 

 other concentric ; each series is proper to a single lamella, and the lamellae 

 themselves are arranged in such a manner that the fibres of contiguous 

 lamella? are at right angles. 



As a result of numerous experiments with stained and unstained scales 

 under varying conditions of light, it seemed most probable that the light and 

 dark bands owed their appearance to reflection from the corrugated surfaces 

 of successive outcrops, the corrugations being produced by the individual 

 fibres. That this is the true explanation of the phenomenon is supported by 

 the fact that an effect precisely similar to that of a scale under polarised light 

 may be obtained by engraving upon a piece of copper plate a series of radial 

 and concentric lines in such a manner as to simulate in their arrangement 

 the fibres on the upper surface of the lower layer of a herring scale as 

 ascertained by dissection. 



Apart, however, from the physical explanation of the polarisation effect it 

 is obvious that the method affords a most valuable check on the reading of 

 scales by ordinary light and affords important corroborative evidence, when 

 taken in conjunction with that adduced already, of the truth of the view that 

 the transparent rings do indeed mark recurring periods of minimum growth. 



