MUSEUM OF COMPARATIVE ZOOLOGY. 295 



but as the specimens increase in size, the anal branch separates from it. 

 The odd ambulacral pores are at first two single rows of pores, which 

 by closer crowding eventually alternate, but are not arranged in pairs. 



The young Agassizia, a quarter of an inch in length, is a flat elliptical 

 Spatangoid resembling Gualteria. The peripjetalous and lateral fascioles 

 have the same general limits as in the adult, but the arrangement of the 

 pores in all the ambulacra is identical ; there is but a single pore for 

 each ambulacral plate, as it exists in the anterior pair and odd ambulacra 

 of the adult ; the ambulacral grooves are not yet formed, the anterior 

 groove alone being slightly indicated ; the mouth is not labiate. 



The great number of Spatangoid genera established upon differences 

 in the subanal fasciole, the existence or absence of the anal branch, the 

 depth of the ambulacral grooves, the confluence or distinctness of the 

 lateral ambulacra, all based upon characters subject to great variation 

 during growth, show the necessity of a careful revision of the whole 

 group of Spatangoids with the data here furnished ; and such closely 

 allied genera as Maretia, Spatangus, Hemipatagus, and Macropneus- 

 tes ; Eupatagus, Plagionotus, and Metalia; Meoma and Linthia ; Agas- 

 sizia, Prenaster, and Periaster ; Gualteria and Brissopsis ; Tripylus, 

 Desoria, Abatus, and many others, must be re-examined and critically 

 revised before we can attempt an arrangement of Spatangoids into 

 natural families. 



The subordinal divisions usually adopted since their introduction by 

 Albin Gras do not seem satisfactory, if tested by our present infor- 

 mation. In the first place, the whole classification is based upon the 

 separation of the anus from the abactinal system. From what the 

 Embryology of Echini has taught us, the position of the anus has not 

 the physiological importance attributed to it by authors who have so 

 generally received this classification. The unstable position it occupies 

 in the same animal at different stages of growth — at one stage opening 

 next to the mouth, then on the margin, and finally opening in the 

 central part of the apical system in the adult — should make us hesitate 

 to adopt a single anatomical feature as our sole guide. In the first 

 place the order of Perischocchinida?, a most natural one, is founded 

 upon characters derived from the structure of the interambulacral and 

 ambulacral systems. The other two suborders, regular and irregular, 

 usually recognized, can scarcely be called natural. The suborder of 

 regular Echini is more satisfactory than the other, though, from what I 



