96 THE CANADIAN NATURALIST. [Vol. X. 



tubercles, the surface is marked by certain superficial grooves, 

 which are represented in the wood-cut by unbroken lines. The 

 general direction of most of these grooves is longitudinal, and 

 the most strongly murked are those which run from the antero- 

 lateral (e. €.) to the postero-lateral angles (f^ a.) and which are 

 nearly parallel to the sides of the shield. Sutures scarcely per- 

 ceptible : their apparent outlines being indicated in the figure by 

 dotted lines. 



Fost-dorsomedian j^i'fte. Convex along the median line but 

 highest in the centre, from which point there is a downward slope 

 in every direction, the lateral slopes being most abrupt. Outline 

 oblong but narrowing posteriorly so as to form a short beak. 

 Anterior end somewhat rounded, sides parallel for more than two- 

 thirds of their length, then attenuating rapidly into a point with 

 obliquely concave sides. Maximum breadth equal to about one- 

 third the entire length ; apex of the beaked extremity curved 

 slightly upwards. Tubercles arranged concentrically but not in 

 distinct rows, those in the centre being the smallest, and those 

 near the circumference being both distant and of comparatively 

 larae size. 



Veiitromcdian platQ. Flat; subrhomboidal, but with all the 

 sides unequal and the margins of two of them (the right ante- 

 riorly and the left posteriorly) shallowly concave. Posterior ex- 

 tremity rather more produced than the anterior ; length about 

 one-third greater than the breadth. Tubercles arrano;ed in dis- 

 tinct rows on three sides, but not on the left side of the posterior 

 half, where they are nearly all isolated, those towards the centre 

 being comparatively large and those near the centre very minute 

 and densely crowded. 



Pre-vejitrolateral plates. Flat ; longitudinally subreniform, a 

 little longer than bro:id ; outer margin concavely emarginate and 

 inflected. Tubercles isolated, crowded and arranged obscurely 

 in concentric, subparallel lines. 



More than twenty well preserved and tolerably perfect speci- 

 mens of the central shield have been collected, besides numerous 

 fragments, but the suborbital plate is invariably absent, and the 

 rostral plate is only preserved in place in one or two instances. 

 The whole of these shields, too, appear to have been flattened by 

 pressure, and if so, they may once have been longer in proportion 

 to their breadth than they now are, and the anterior sinus into 

 which the rostral plate fits, may have been narrower and deeper. 



