PERISCHOECHINOIDA. 253 



ambulacra wide with in each area about eight columns of very irregular thin polygonal plates 

 which bear small primary and secondary tubercles and spines. 



The apical plates are doubtful; but a madreporite is described by Wyville Thomson and 

 Gregory in an interambulacrum. As this is a unique condition in Echini and morphologically 

 difficult to understand, it may not be too much to suggest that in the squeezing of this rather 

 badly preserved material, it may have got displaced. According to Wyville Thomson, the 

 anus with apparently periproctal plates is in an interambulacrum, making it an irregular type. 

 The lantern is well developed, and Gregory shows that it has typical echinoid pyramids. The 

 material I have seen is so crushed and badly preserved that it is very difficult to make out much 

 structure. There seems no reason for considering this type as presenting any possible connec- 

 tion with starfishes, or as a primitive type. Its four columns of ambulacral plates, and especi- 

 ally its eccentric anal area, would lead one to consider it a specialized and irregular rather than a 

 primitive echinoid. 



Lower Ludlow, Silurian, Leintwardine, England; a very fine specimen is in the Museum of 

 Practical Geology, London 7,385, one of Sir Wyville Thomson's originals. This is shown in 

 Plate 16, fig. 4, from a photograph kindly sent me by Dr. Kitchin. There are several specimens 

 in the British Museum, part of which Dr. Bather kindly loaned me for study. Oxford Museum 

 (Sollas). 



Order PERISCHOECHINOIDA M'Coy. 



Perischoechinida M'Coy, 1849, p. 251; 1854, p. 114; Jackson, 1896, p. 239. 



Tessdati Desor, 1858, p. 152. 



Perischoechinidae Loven, 1874, p. 39; A. Agassiz, 1874, p. 644; 1881, p. 3; Zittel, 1879, p. 481. 



This large order consists of regular Palaeozoic Echini in which the periproct is within the 

 oculo-genital ring. The test is of various shapes and horizontal outline throughout the ambitus, 

 but is usually spheroidal. Ambulacral areas are narrow, or wide, with in each area from two 

 to twenty columns of plates, all simple and all bearing one pore-pair each. There are from 

 three to fourteen columns of plates in each interambulacral area. Usually the plates of the 

 adambulacral columns are pentagonal, and those of the median columns hexagonal, except 

 where new columns are introduced, and in young plates dorsally. Plates may be thick, or thin, 

 and bear primary perforate tubercles and spines with secondaries, or the latter only. Plates 

 of the corona may be imbricate or not; when imbricate, ambulacral plates imbricate adorally 

 and bevel under the adambulacral plates laterally. Interambulacral plates imbricate aborally 

 and from the center laterally and over the adjacent ambulacral plates. When plates are not 

 imbricate, then the sides of the plates are as nearly parallel as is mechanically possible in a 

 curved test, but they may be very far from parallel, as in the thick-plated Melonechinus. When 

 there is no imbrication, the ambulacral plates may bevel over the adambulacrals, as is char- 



