1898- 1 '.102. iNo.33.] UPPER DEVONIAN FISH REMAINS. 23 



speaking of new species or not. The variations of the sculpture in 

 Drepanaspis strongly confirm Pander's opinion that the four species 

 given by Agassiz simply represent various dermal plates of one and the 

 same species. 



Should this be the case there appears to be little justification for 

 placmg Psammosteiis remains from other regions as new species, simply 

 on account of differences in the external ornamentation. 



My own microscopic investigations of both the material from Elles- 

 mere Land and some Psammosteiis remains from other regions from 

 Liv Land, however, have shown that there are certain differences in 

 the microscopic structure which are very much more constant than 

 the exterior sculpture that varies greatly even in one and the same 

 species. 



1 believe that for the present this provides us with the best method 

 of differentiating species within this family, — which is still very im- 

 perfectly known, — and also of controlling the circumstance of certain 

 apparently constant difference in sculpture and their value or otherwise 

 for purposes of classification. 



I will therefore first describe the existing material from Ellesmere 

 Land, and later on proceed to compare the same with forms previously 

 known. 



Psammosteus a,rcticas new species. 

 (PI n, figs 5—7; PI. IIL figs 1-4; PL V, figs. 4-5). 



Under this name I include a series of fragments the sculpture of 

 which varies considerably with respect to the dimensions and form of 

 the denticles, which, however are remarkable for the following outer and 

 inner characteristics. — 



L The surface denticles are most often high and markedly ribbed, 

 and have very prominent peripheral points. 



2, The pulp cavities of the denticles are very wide, and are 

 situated with wide openings in connection with the underlying vascular 

 canals. 



These fragments must therefore be assumed to belong to one and 

 the same species. 



Description. We may now proceed to a more detailed descrip- 

 tion of the most important of the specimens. 



\. The best of the fragments is a little skeleton plate with 

 extremely well preserved sculpture (S. I. 20). 



