524 Dr. A. L. Thomson : J^csulls of a Stwh/ of [Ibis, 



XII.— CONCLUSIONS REGARDING THE VALUE OF THE 

 METHOD OF BIRD-MARKING. 



It is thought that the results and conclusions set forth in 

 the preceding sections, together with the data collected by- 

 other workers, will be considered sufficient proof of the 

 value of tiie marking method as a means of obtaining a certain 

 ty{)e of fact regarding the problems of bird-migration. At 

 the same time it will be evident that the labour involved is 

 very great and that the percentage of marked birds recovered 

 is generally very small, while the results are in some cases 

 largely invalidated by inherent defects which the method 

 displays under certain circumstances. Various reservations 

 must accordingly now be made in appraising the method, 

 and the conclusion that must be drawn is that it would be 

 advisable to restrict its further practice to those lines which 

 have been shown to be comparatively fruitful in reliable 

 and interesting results. 



The continuance of promiscuous marking is likely, of 

 course, to yield a small proportion of isolated results that 

 will possess an undoubted interest. But it is questionable 

 whether records of this kind will ever adequately repay the 

 time, labour, and money expended on the actual task of 

 marking. Certainly these records will not equal in value 

 the results which can be obtained by more systematic study 

 on restricted and carefully selected lines. 



It has been stated in the introduction that three qualities 

 were considered essential, on a priori grounds, before a 

 species could be regarded as a suitable subject for this kind 

 of study : it must be available for marking in large numbers, 

 it must afford a good percentage of reappearance records, 

 and its migratory movements must })resent features worthy 

 of investigation. To these may now be added a fourth point, 

 namely that tlie circumstances must be such as will not tend 

 to invalidate the results by the introduction of too many 

 uncertain elements. 



Quantitative as well as qualitative analysis must be aimed at 

 in spite of the obstacles that seem to lie in the way. As has 



