CEPHALASPID^. 



89 



Powriei, and otlicrs less satisfactory of the same subgenus. What is said here, there- 

 fore, must be accepted as based merely on that limited amount of evidence. 



In the first place, it is probably a very constant character that the scabs of the body 

 present the same tubercular ornamentation as do the head-shields, since it is observed in 

 all the specimens discovered, and is parallelled in the Heterostraci by the identity of orna- 

 mentation of the scales and shield of Picraspis. The body in those forms known is very 

 thin, small, and tapering, probably embracing but a small portion of the viscera, which were 

 rather covered in by the so-called //cY/cZ-shield. The scales (in Eucophalaspis) are in four 

 chief series on either side the me.dian line — a dorsal, of oblong rectangular form ; a lateral, 

 of much longer proportions ; a marginal, projecting from these at an angle ; and a ventral, 

 meeting in the median line on the ventral aspect, as do the dorsal on the dorsal aspect. In 

 Eiicephalaspis the scales are in a single roio in each series, with the exception, probably, of 

 the ventral series, where they appear to be broken up into four or more (PI. XI, fig. 2). 



Fig. 14. 



Loricaria plafystuma, Giintlier. 



The Specimen referred to in the Arbroath Museum is the only one which clearly shows 

 this remarkable series of scales. Posterior to the dorsal fin the three well-defined rows 

 of scales, visible when the fish is in a lateral position, become broken up and confused in 

 a continuous armour of rhomboid scales. The manner of the breaking up of the series I 

 do not know, since no specimen is well preserved in this part ; but it may be conjectured 

 from the numerous analogous cases among Siluroids. Prof. Agassiz was led to suppose 

 that the dorsal series of scales was double on cither side, by the observation of the rhom- 

 boid scales belonging to the region behind the dorsal fin, and probably also by Mr. Dinkel's 

 very imaginative representation of the profile specimen (fig. 1, pi. I, in the ' Poissons 

 fossiles;' and in ' Siliu'ia,' 18G7, pi. 30, fig. 3), now in the possession of Sir Philip 



