152 Haakon Schetelig. [No. 8 



5. General ehronologieal conelusions compared with the 



typology. 



From the recorded finds I think it possible to make out, 

 roughly, some main periods of the typological development expressed 

 by absolute dates in centuries. Of course the dates are always 

 given with the reservation before mentioned, and the discussion of 

 the date in each of the above instances may have given results of 

 no great exactness, as the graves in most cases contain objects of 

 different age. But when a general progress of the typological 

 development has been confirmed by the study of the flnds, it is 

 allowed also to combine typological features with evidences drawn 

 from the graves in order to obtain a doser ehronologieal evidence 

 than it is possible to obtain from the graves alone. When a form 

 has passed through a so rich typological development as is the 

 case with the cruciform brooches, we may be sure that none of the 

 distinct stages of the development has lasted for any long period; 

 the development is to be considered only an unbroken series of 

 small changes which gradually produced varieties very different from 

 the first form. Thus it must be evident that the date of a certain 

 stage of development is most exactly stated by the earliest appear- 

 ance of it, and that specimens of the same form from graves of a 

 considerably later date are to be regarded as specimens which were 

 old already at the time when they were buried in the grave. With 

 a sufficient knowledge of all the forms and of the other contents 

 of the graves where they are found, it is thus possible to decide 

 that, for instance, a brooch was made about the middle of the 5th 

 cent. but was not buried till after the beginning of the 6th cent. 

 In the following statements it must be remembered, however, that 

 I have regard only to the Norwegian material, as my knowledge 

 of the material from other countries does not permit me to draw 

 reliable conelusions. 



I have arranged the periods of the cruciform brooches from 

 Norway in the following way: 



I. 350—400. A. D. Early cruciform brooches, before the be- 

 ginning of a separate development in the different districts. 

 The foot is long and narrow; the bow is long and approaches 

 to the semicircular form; the head-plate is small and the side- 

 knobs are fixed upon the axis of the spring-coil. Perhaps 



