Manchester Meiuous, Vol. hi. (1912), No. l\. 7 



are taken they all have the same number of peroniclia. 

 The total number of annuli varies from y6 to 121, this 

 range of variation being due not to variation in any one 

 particular peronidium, but to variation in each peronidium. 

 This great variation is not due to a casual variation of 

 the scale, but is caused, to some extent, as the figures show, 

 by definite local differences. That is to say, speaking 

 generally, the nuuibcr of annuli in each peronidium increases 

 front the head to the adipose fin, on both dorsal and ventral 

 sides of the lateral line, and then decreases again tozvards 

 the tail. It would be of very great interest to study 

 more scales from each of these seven positions in order 

 to ascertain whether these differences were merely 

 abnormalities of the particular scales which had happened 

 to be taken for examination or whether other scales from 

 the same positions would also show the same variations. 



While looking out and preparing the material for 

 this test it was noticed that the scales from different 

 parts of the body differed very much in size and 

 shape. Thus it was found that scales from position (i) 

 approximated generally to one type, while those from 

 (2), as a whole, differed from those from (i), and could be 

 called type 2. This was found to be the case with each 

 set of scales from the various positions on the fish ; there 

 were as many types as positions from which scales 

 were examined. These t}'pes are in some cases so 

 distinct that after a little practice one can pick out the 

 scales from the different positions without referring to 

 the numbers of the slides. 



There is always great difficulty in making up these 

 sets of scales. About thirty scales were taken from each 

 position on the fish, but on examination it was frequently 

 found that twenty or more had imperfect centres, so that 

 there might not be even three scales from any particular 



