60 CONTRIBUTIONS TO CANADIAN PALAEONTOLOGY. 



Some ten specimens have been found, some of them doubtfully referred 

 here, of which two of the best, occurring on a single small slab, have been 

 selected for illustration. The species is the first referred to in my note on 

 sigillarian articulates appended to Sir William Dawson's account of his 

 recent explorations* as having "perfectly flat segments showing only a very 

 'slight and jiarrow transverse ridge at the anterior margin, occupying not 

 more than one-fourth of the segment." In the form of the segment it is 

 comparable to Xylohius mazonus from Illinois, the anterior ridge having 

 the effect of a strap around the body (see fig. 6). The flatness of the seg- 

 ments is no doubt due in part to crushing, but the effect is to impress one 

 with the belief that the body was broader than high. Some specimens 

 seem to indicate that the transverse ridge was generally half as broad as 

 the remaining portion of the segment, and was separated from it by a 

 suture, when viewed from the under side of the dorsal scutes, so that the 

 body may be said to be made up of shorter and more elevated, and 

 longer and more depressed flattened segments. The surface itself of the 

 segments appears to be perfectly smooth and shows no pigns whateyer 

 of frustra ; the dorsal scute of the largest specimen when laterally ex- 

 panded and crushed is O-eS"" long and S-YS""" broad. Several specimens 

 show more or less connected fragments, making together a length of from 

 25 to 40°"°, and there can hardly be reason to doubt from all the appear- 

 ances taken together that the creature reached at least a length of from 

 60 to TO""- Remgjins of serially connected ventral, scutes show that 

 these were nearly as broad as the dorsal and twice as numerous. The 

 absolute smoothness of the dorsal scutes, however, shows that the genus 

 cannot be referred to Euphoberia, although in every thing but the arma- 

 ture (so far as the fragments go) the relation is close. 



Archiulns lyelli sp. nov. 



PI. IV., figs. 3, 7. 



This additional species from the same locality as the preceding affords 

 no better material for study than it, but indicates as clearly the presence 

 of a hitherto unknown form. It is. the second of the species referred to 

 in the note above alluded to as smaller than the preceding, and having 

 " shorter and more simple segments, made slightly concave by the gentle 

 elevation of both front and hind margins, but with no anterior ridge." 



Four specimens are referred to this species, of which two are figured. 

 They all indicate a small species perhaps S-S"" long (the longest connected 

 fragment is perhaps 20°"° long) and 2°"" broad. The segments appear to 

 average about four times as broad as long, to be smooth and entirely des- 



'*Phil. Trans. Roy. Soc, Lond., 1882, 649. 



