38 JOHAN KI^R. [sec. arct. exp. fram 



Thelodns like forms. Palaeaspis, Pteraspis and kindred forms con- 

 stitute a related group, which give us a co-ordinate family Pteraspidae 

 with a similar origin. 



Within the family Psammosteidae we can elucidate the development 

 by the series Thelodus — Drepanaspis — Psammosteus, which as 

 concerns the outer form of the body appears to constitute a parallel to 

 the series Thelodns — Ateleaspis — Akeraspis — Cephalaspis. 



Coccosteidae. 



Great Coccossteomorph form. 



(PI. II, fig. 3-4). 



The fish horizon at Skrap Valley provides us with a somewhat 

 large, beautifully ornamented skeleton plate. (S. I. 7), which is probably 

 a portion of a large, new Coccossteomorph fish. 



Description. Figs :-> — 4 on plate II give a better representatioa 

 then any long description can do of this specimen. The upper surface 

 is beautifully ornamented with large, round, slightlv crenelated tubercles, 

 which often have a diameter of 2 mm. 



The borders of the plate appear to be the original one on three 

 sides; on the fourth there is an irregular fracture line. On the smooth 

 inner side we see faithest down to the left a strongly developed facet, 

 by which the plale must be assumed to have been superimposed over 

 an adjoining plate. 



The maximum length of the plate is 67 mm. At the edge which 

 is seen turned up to be in the figure, the thicknes is 3 — 4 lum. 



Farthest to the right, whei-e the plate however possibly has suffered 

 erosion, this diminishes to 2 mm. and then rises to 7 mm. just inside 

 the broadest part of the facet. 



The plate is thus extremely massive and strong. 



Observations. The determination of this plate has caused 

 me a great deal of labour, altbougb I have been unable to arrive at 

 anv entirely satisfactory result. The scul[iture may be said to be 

 coarsely coccosteid. The inner struclure agrees entirely with this con- 

 clusion. Therefor il would appear to be certain that we have before ns 

 a plate of a large coccosteomorph fish. But which? This question 

 is not easy to solve, especially when we possess no good material for 

 comparison wilh known forms to aid us in this task. 



