84 



PROF. H. G. SEELEY 0^" THE 



verse position vertical above the femur, and is in a way imperforate. 

 It is noticeable that, while the sacral ribs form 5 lateral but- 

 tresses to support the acetabular part of the ilium, several pre- 

 sacral ribs (in no way modified) extend between the externa] 

 armour and the ilium, for the more anterior are seen through the 

 thin pre-acetabular plate of the ilium. And this indicates that 

 armour, ribs, and ilium blended late in life. 



There is a very close resemblance in the ilium between Pola- 

 cantJms and Omosaurus ; for Omosaurus has a similar large hori- 

 zontal acetabulum, while the pre-acetabular plate is of similar form, 

 and extends horizontally forward and outward, and the post- 

 acetabular part is similar. There is also a manifest resemblance 

 in the forms of the sacral ribs, and in their mode of attachment to 

 the sides of the vertebrae to which they belong, although in Pola- 

 cantlius they tend to a slightly more anterior position. In Omo- 

 saurus, however, external armour on the trunk is unknown. 



Other resemblances to Polacantlms are found in the Gosau fossil 

 Cratceomus. When that ty])e was described in 1881 I had not seen 

 any of the remains of this armoured AVealden fossil. But I drew 

 attention to a figure by the late Mr. J. E. Lee ^ from the Wealden 

 of the Isle of Wight, which shows a dermal plate with tubercles 

 upon an osseous base, as similar to armour ' of Cratasomiis.^ Mr. 

 Lee's fossil may have belonged to Polacantliiis or a nearly allied 

 genus. The dorsal vertebrEe of these genera are very similar, as is 

 evident on comparing Phil. Trans. Hoy. Soc. 1881, pi. Ixx., and 

 Quart. Journ. Geol. Soc. vol. xxxvii. pi. xxx. fig. 3. The femur 

 in both corresponds in type and in many details. If there had 

 been any grounds for referring the remarkable plates to Cra- 

 tceomus ^ which Bunzel had figured as the ilium of Danuhiosaurus 

 the resemblance might have been carried further. It has been shown 

 that those remains consist of a rib-like lower part, with which is 

 blended a superimposed thick smooth armour, like that of a Chelo- 

 nian. Without re-examination it may not be possible to form a 

 definite judgment on these remains, but there seems some likelihood 

 that they may prove to be portions of the ilia of Cratceomus with 

 conflueiit smooth armour extending forward from the acetabulum, 

 Polacantlms, however, has no such smooth scutes, and all the 

 armour of Crata^omus which can be compared with it is either 

 tuberculate or rugose with vascular impressions. 



Oruithischians with smooth armour, however, existed in the 

 Wealden deposits. And the British Museum acquired in the 

 Beckles Collection a portion of the acetabular region of such a 

 fossil which differs from Polacantlms in the far greater thickness 

 of the ilium above the acetabulum, in the forms of the confluent 

 sacral ribs constricted from front to back, and in the absence of 

 all trace of ornament from the smooth and relatively thin dermal 

 shield above the ilium as preserved. The species is larger than 



1 Ann. Mag. Nat. Hist. vol. xi. (1848) pi. i., and ' Note-book of a Naturahst.' 

 ^ Quart. Journ. Geol. Soc. vol. xxxvii. (1881) pi. xxviii. 

 3 Ibid. p. 694. 



