82 ROBERT TRACY JACKSON ON ECHINI. 



resorbed. In all of these cases the ambulacral plates of the peristome are strongly imbricated 

 adorally, and, as in Bothriocidaris, the pores are nearly or quite superposed dorso-ventrally. 



A third type of peristome is where there are both ambulacral and non-ambulacral plates 

 on the buccal membrane. The ten primordial ambulacral plates surround the mouth, at least 

 as far as known, and are succeeded by rows of two or more plates in each ambulacral area 

 aborally, separated by more or fewer interradial non-ambulacral plates. Such a structure 

 is seen in Archaeocidaris (text-fig. 47) and Melonechinus (text-fig. 48) in the Palaeozoic. In 

 Archaeocidaris there are two vertical columns of ambulacral plates and many columns of 

 scale-like non-ambulacral plates increasing in number aborally to the margin of the actinostome, 

 where they have a greater number of plates than exist in the basicoronal row of the inter- 

 ambulacrum. Both ambulacral and non-ambulacral plates on the peristome imbricate strongly 

 adorally. This structure has been made out more or less fully in four species of Archaeocidaris, 

 namely, A. wortheni (Plate 9, fig. 6), A. rossica (Plate 11, fig. 1), A. agassizi (Plate 13, fig. 3), 

 and A. urii (Plate 15, fig. 1). In Melonechinus (text-fig. 48; Plate 56, figs. 7, 8) there are 

 two plates in each ambulacral area around the mouth, but passing outward, they increase to 

 m,any plates in each row before reaching the corona. There are but three non-ambulacral 

 plates in each interradial area on the buccal membrane. This is based on a single specimen, but 

 it is probable that a similar structure may exist in other members of the family of the Palaeechi- 

 nidae. Ambulacral and non-ambulacral plates of this type on the peristome are a character of 

 the Cidaridae alone among Recent Echini. The plates imbricate strongly adorally in both areas 

 (text-fig. 39, p. 75), as in Archaeocidaris, emphasizing the primitive character of this struc- 

 ture. In the peristome of Eucidaris (text-fig. 46; Plate 2, fig. 6) there are two plates in each 

 row in each ambulacrum. In the interradial areas there may be one plate in each row, or 

 passing outward, this may increase to two plates in a row; or in Phyllacanthus (Plate 2, fig. 18), 

 as shown by A. Agassiz and Clark (1907a), there may be as many as five plates in a row in 

 each area interradially on the border of the actinostome. While in adult Eucidaris tribuloides 

 there are many rows of very low scale-like ambulacral plates on the peristome (text-fig. 46), in 

 a young specimen of 5 mm. diameter (Plate 2, fig. 6) there are only four rows, and the plates 

 are high, hexagonal in outline, resembling the similar plates in Bothriocidaris. 



Mr. Agassiz, in the Revision of the Echini (1874, p. 646), compared the actinostome of 

 Cidaris to the whole ambulacral and interambulacral system of Palaeozoic Echini. As he 

 said, "Obliterate the two rows of coronal plates [of Cidaris] and we have a spherical urchin 

 composed of a series of plates corresponding in every respect to the hypothetical Palaechinus. 

 It consists, namely, only of an abactinal system, .... of an ambulacral and interambulacral 

 system, composed of hexagonal and pentagonal plates, perfectly flexible, and of an oral opening 

 through which we find teeth projecting." Mr. Agassiz (1892, p. 72) again said in his mono- 

 graph on Calamocrinus diomedae, "The structure of the test of Bothriocidaris is represented 



