26 



JOHAN 



[SEC. ARCT. EXP. FRAM 



sculpture and interior structure from the Psammosteus remains previously 

 described. In certain respects they are nearer to the well-known Drepanaspis 

 Gemundenensis of TRAQUAIR'S description, but they cannot be combined 

 with that form. I will return to this matter subsequently, and simply 

 state here in what manner the specimens described may be orientated. 

 I make the very probable assumption that Drepanaspis is a closely 

 related form, that provides us with a representation of the habits of the 

 body in the case of Psammosteus. 



The specimen described in 1 must therefore be assumed to be a 

 fragment of one of the large central body plates, either dorsal or 

 ventral. 



The fragment described in 2 probably belonged to one of the lateral 

 plates, as the denticles on these in the case of Drepanaspis are usually 

 extended. 



The little plate described in 5 shows signs of correspondance to the 

 small, intermediary plates in Drepanaspis. 



The fragment described in 4 is more difficult to determine. It is 

 evidently portion of an element with an oblique, spine-shaped construc- 

 tion, corresponding to those known from the Livland Psammosteus form. 

 It might be regarded as either a dorsal spine or a lateral spine, for the 

 posterior corners of the lateral flaps in these forms possibly emerged in 

 sharp spines or horns, instead of being bluntly pointed as in the case 

 of Drepanaspis. 



Horizon. All the remains in our possession of this form come 

 from the fish horizon in Skrap Valley. 



Psammosteus complicates new species. 

 (PI. II, fig. 8; PI. V, figs 1-3; PI. VI, figs 1-3). 



A number of Psammosteus fragments in my material prove to be 

 so different from the above and other forms described, both as regards 

 the interior and the exterior character, that they must be regarded as 

 a new species to which I have given the name complicatus. 



The following peculiarities appear to be most characteristic; 



1. The denticles vary greatly both in size and in form. On certain 

 plates presumably the large body plates, - - they are large, round, and 

 regular, with central top and distinct ribs; in others, -- presumably 

 portions of the corner plates (cornua) -- they are small, flat, oval, or 

 rhombic, sometimes greatly extended and faintly ribbed. 



