

MKI.ON Will NTS. 381 



characteristic of the raid-zone in the lowest species of tin- gmu.-. a- M. springeri (text-fin. 237, 

 p. 231). Again, proceeding dorsally, in multiporus we next find two (-..11111111-- of isolated plate- 

 in each half-area (Plate 57, fig. 1, area J; text-fig. 245); these with the demi- and occluded plates 

 form a stage with eight columns of plates, which is chaniet eristic of the species M. indianensis 

 (text-fig. 237, p. 231). At the mid-zone in multiporus, and for some distance below and above 

 the same, -the full species character of ten columns of plates exists (Plate ."><>, fig. 4; Plate :>7. 

 fig. 1). Thus do typical stages in development repeat the characters seen as adult features 

 in lower genera, or lower species in the genus. As a case of extreme regressive radial variation, 

 in area B of text-fig. 245 and Plate 57, figs. 1, 3, it is seen that a few plates vent rally near the 

 peristomal border are primaries, completely crossing the half-areas. This condition was not 

 seen in any other specimen in the genus; as a reversionary stage, it is comparable to the 

 typical adult character of Palaeechinus and to the youthful ventral character of Maocoya, 

 Lovenechinus, and Oligoporus (text-fig. 237, p. 231), (p. 19). 



Turning to the dorsal portion of the ambulacrum in Melonechinus multiporus, we find 

 that close to the ocular in the placogenous zone in area J (Plate 56, fig. 6; Plate 57, fig. 2; text- 

 fig. 245) there are two ambulacral plates only. In this zone there may be three plates as shown 

 in other areas, but my figures show that one primary plate in each half-area may be found, 

 when as a feature this may be considered a localized stage comparable to the adult typical 

 character of Palaeechinus, and to a similar dorsal localized stage in lower genera or other 

 species in the genus (text-fig. 237, p. 231). Proceeding ventrally, in multiporus we find next 

 a stage with four plates in a row (Plate 56, fig. 6), which is comparable to the typical char- 

 acter of Lovenechinus (text-fig. 237, p. 231). Again, passing ventrally, we next find a sta^e 

 with six plates in a row (Plate 56, fig. 6) ; this is comparable to the typical character of 

 Melonechinus springeri (text-fig. 237, p. 231). Again, passing ventrally, we next find a stage 

 with eight plates in a row (text-fig. 245; Plate 57, fig. 1), which is comparable to the typical 

 character in M. indianensis. Thus do dorsal localized stages, passing ventrally in the progres- 

 sive development of plates built late in life, present stages which are quite comparable to tho-e 

 seen in youthful development, also to the typical adult characters of lower genera or species 

 in the genus (p. 366). 



Jackson and Jaggar (1896) worked out the development of the interambulacra in this 

 species, illustrating it by a number of figures and also by tabulations. I have nothing to add 

 to that except to give the arrangement in an exceptionally perfect specimen, lately in Pro; 

 Hambach's and now in Mr. Frank Springer's collection. This specimen (Plate 55, figs. 1, 2) 

 is not very large, but is free from all matrix and distortion so that I am able to represent it spread 

 out by the Loven method with the restoration of only a few plates in the basicoronal row (Plate 

 57, fig. 1). Before describing the interambulacra of this specimen, attention is called to the 

 diagrammatic text-fig. 246. This is selected to show a typical interambulacral area in this 



