78 MADRAS FISHERIES BULLETIN VOL. XV, 



small scales covering the base — not reckoned in the lateral line 

 count — are not always sharply defined from the scales of the trunk. 

 The count ends with the last scale having a sensory canal. The 

 first scale counted is that immediately above the base of the pec- 

 toral fin, and behind the operculum. Care must be taken to account 

 for those scales which have been lost, and if these are many, one 

 must rely on the scale pockets, which are often not easy to deal 

 with. The number in the lateral line series of all specimens in the 

 same species is not quite constant; but within a certain range (at 

 most 3 or 4) there is constancy. 



2. The lateral transverse series.— Speaking generally, for purposes 

 of counting the number of rows of scales in the lateral transverse 

 series, the mid-ventral and mid-dorsal line is omitted; thus only 

 truly lateral scale series are taken into account. 



Between the two dorsal fins, however, we frequently find no 

 mid-dorsal scales present ; not a single scale is median for this 

 length, the median series being represented by two short lateral 

 series, which do not generally extend beyond the limits of the fins. 

 Since their limits are so restricted, these short series have not been 

 included in the count of the lateral transverse series. 



The mid-dorsal series between the two dorsal fins may, 

 however, be represented by any number up to six scales — which 

 figure practically represents a complete series between the fins in 

 question ; one, two, three, etc., only may be present, before the series 

 divides into two lateral ones. The number of scales present in this 

 mid-dorsal line does not appear to be distinctive of species, but 

 rather to be an individual variation. 



Occasionally, the two intercalated lateral series mentioned 

 do extend a little beyond the origin of the second dorsal fin, even 

 to its posterior limit ; and anteriorly, they may include the 

 elongated scales found at the base of the first dorsal fin. But this 

 series seldom proceeds further forward than the first dorsal fin. 



Mid-ventrally, too, it is not always a simple matter to decide 

 which line of scales can be regarded as mid-ventral ; short inter- 

 calated series are sometimes present, and sometimes a lateral series 

 "gives out " before the vent is reached. But here also, complete 

 series only have been taken into account. 



The number of rows in the lateral transverse series very 

 frequently differs on the two sides of the same specimen ; but 



