302 BULLETIN OF THE BUREAU OF FISHERIES. 



He divides the evidence necessary to prove the general theory of age determinations 

 on scales into (a) morphological, (b) experimental, and (c) statistical. To summarize 

 his conclusions: 



(a) Morphological : 



(i) The evidence necessary to prove that a broad band is formed in summer and 

 a narrow one in winter has not yet been produced. On this point he cites the insuffi- 

 ciency of Dahl 's and Johnston 's evidence. 



(2) " The scale can not be an accurate gauge of the lapse of time unless the zones, 

 besides being produced in their respective seasons, are always produced in response 

 to these seasons." 



(3) "The formation of these two different series of growth — rings or zones — takes 

 place in the winter half or summer half of the year, respectively" (quoting Dahl). 



(b) Experimental: The evidence of fishes of known age and kept under artificial 

 conditions is convincing as far as it goes (for the first two years), but can not be regarded 

 as convincing through the entire term of life until more work is done. 



(c) Statistical: "In studying the average sizes, average weights, and seasonal 

 occurrence of the different age groups and numerous other statistical relations, the 

 age data obtained from the scales give a rational and consistent result throughout." 



Under the caption, "The morphology of salmonoid scales" he classifies the different 

 circuli as complete circles, occurring in the earliest stages of the fish; crescentic or 

 incomplete circles, occurring in normal summer growth, and incomplete seasonal 

 crescentic ridges. These latter occur in the winter growth, and if his conclusions here 

 are reliable, consideration of these short circuli should be a valuable addition to the 

 methods already employed, but, of course, not necessarily applicable to any species 

 other than the salmon studied by him. 



In connection with his discussion of the "spawning mark," his conclusions may 

 be summarized as follows: 



1. It may be held as conclusive that the spawning mark is produced by changes 

 incidental to the act of spawning. 



2. The spawning mark is not caused b}- the mechanical vicissitudes of river life or 

 the act of spawning, as assumed by Johnston, Dahl, and others. In support of this 

 view he calls attention to the following points: (a) The spawning mark is produced 

 prior to entering the river and in some cases, long prior to spawning; (b) the fact that 

 the scale is imbedded deep in the dermal pocket would, alone, destroy the mechanical 

 attrition theory; (c) since it is known that gonads are developed partly by the absorption 

 of other tissues, it is not unreasonable to assume that the scales are among the tissues 

 so absorbed. 



3. It can not be taken as proved that the absence of the spawning mark is valid 

 evidence that the fish has not spawned. In this connection he cites the case of salmon 

 kept in aquaria at the Plymouth laboratory that were stripped for two successive seasons 

 without the formation of a spawning mark. This, he admits, may possibly be due to 

 the artificial conditions. 



4. It seems impossible accurately to define the spawning mark. Consequently the 

 personal element will enter into doubtful cases, and differences of opinion will result. 



