A MONOGRAPH OF THE EXISTING CRINOIDS 299 



he has seen from Mauritius, the locality of Lamarck's original specimen, that ho 

 would most assuredly never have given them the specific name carinata. He noted 

 that, as is also the case with the Red Sea form (audouini), the terminal portions of 

 the arms have stiff er pimiules and a less feathery appearance than is the case with 

 the Brazilian variety. 



In the relative development of the carination along the dorsal surface of the arms 

 Brazilian specimens exhibit an extraordinary amount of diversity. The carination 

 is rarely so slight as in extreme specimens from Africa, and is often greatly exaggerated, 

 the median tubercles standing up from the dorsal side of the brachials as tall spatulate 

 processes. This condition is correlated with the development of an opposing spine 

 and a serrate distal border on the brachials and pinnule segments which imparts to 

 the animal a peculiar rough feeling. I have examined specimens of this variety taken 

 with others of the usual type, and all of those from deep water off St. Lucia are of this 

 form. It also occurs occasionally among specimens from Mauritius (see p. 298). 



In Brazilian specimens not of full size Pi is often much longer than P-j. This 

 condition seems merely to indicate immaturity. 



A small specimen collected by the Challenger at Bahia was described by Dr. 

 P. H. Carpenter under the name of Antedon dubeni as follows: 



The centrodorsal is a slightly convex pentagonal disk. The cirri arc XXX-XL, 

 12-15; the segments are smooth, the outer stouter than those at the base, laterally 

 compressed, and rather longer than broad. The radials are almost entirely concealed. 

 The IBri are oblong and are not imited laterally. The IBr2 (axiUaries) are acutely 

 triangular. The 10 arms are 40-45 mm. long. The first two brachials are tolerably 

 similar in shape, oblong or subtriangular, the second being rather the longer. A 

 few brachials after the second syzygy may be triangular, but they soon become quad- 

 rate with the sutures but little inclined so as to be somewhat squarish in outline, 

 becoming elongated toward the arm tips. The lower and middle brachials may 

 overlap more or less, but the distal parts of the arms are almost smooth. Syzygies 

 occur between brachials 3 + 4, 9 + 10, and 14 + 15, and distally at intervals of 2-7 

 muscular articulations. Pj is over 10 mm. long and is composed of 25-30 elongated 

 segments. P^ is about half its size with 12-15 segments. Pj and Pt are of about 

 the same length, and the pinnules following gradually increase, becoming very long 

 and slender distally. There is a variable amoimt of calcareous plating between the 

 rays and a few scattered granules on the ventral surface of the disk, especially in the 

 anal interradius. The disk is about 7 mm. in diameter. Sacculi are very abundant 

 along the pinnule ambulacra. Dr. Gisl^n gave additional notes on this specimen 

 which he examined at the British Museum in August, 1925. He said that the centro- 

 dorsal is flattened with the dorsal pole convex and one-third the diameter of the base. 

 The cirri are arranged in a partially double row. The cirri are about XX, 15, 9-10 

 mm. long, and have a small opposing spine. The arms are about 30 mm. long. Pi 

 is 5.5 mm. long with 15 segments. Pj is 4.3 mm. long with 12 segments. P3 is 4 

 mm. long with 12 segments. The discrepancies between the descriptions of Car- 

 penter and Gisl6n arc explained by the fact that Carpenter introduced into his descrip- 

 tion a few characters taken from the type specimen of Antedon dubenii which he had 

 before him at the time ho WTOte. 



