1 8 ESDAILE, Results of tJie Saliuoii Scale Research. 



and "summer" fish represent two distinct races; this 

 would mean that spring fish bred spring fish, and summer 

 fish summer fish. However, the scales which have been 

 examined at the Manchester University do not appear to 

 lend any support to this theory. It is to other data that 

 we must now turn to settle the question satisfactorily; 

 such as the number of vertebrae, the number of fin rays 

 present in the various fish, or the form of the bones might 

 perhaps show some distinguishing features which would 

 definitely decide the matter. 



The fish that have spawned twice give even more 

 opposition to any such theory. When a fish is first a 

 spring fish, then a summer fish, and is finally caught in 

 the spring, the time of year would appear to be of no 

 importance as an indication of racial differences among 

 salmon. 



It should be noticed that there are no data such as 

 v/ere used by Professor Garstang (17) in his paper on 

 " Variations and Races of Mackerel," such as the number 

 of rays present in the various fins of fish from different 

 localities. Neither have we the accurate measurements 

 (of the sizes and distances of the various fins, etc.) similar 

 to those taken by Dr. Jenkinson in his recent paper (18) 

 on " Growth, Variability and Correlation in Young Trout." 

 The available data only record fluctuations, due in all 

 probability to the amount of food taken and to the 

 particular circumstances under which each fish has spent 

 its life. These data are of no use when we come to 

 consider whether the various runs of salmon are all of one 

 race or whether there are two distinct races which might 

 be characterised as " spring " and " summer " fish. 



In a work of this kind enough precaution can scarcely 

 be taken against generalising from {c\\ data. The records 

 uiicd in this paper were obtained from only one river and 

 from only a very small percentage of the total number of 



