i8 
The specimen from Stat. 274 has 20 arms iio mm. long; there are four IIIBr series 
present, three internal and one external, paired with one of the internal ; the cirri are XVII, 
23, — 25, 20 mm. long; the longest cirrus segments are about a third again as long as broad. 
The example from Stat. 282 has 20 arms. 
The individual from Stat. 285 is very small with 10 arms. 
Of the four from Stat. 294 the largest has 14 arms 35 mm. long; all of the IIBr 
series are 4 (3-4-4); another has 11 arms 35 mm. long; the single IIBr series is 4 (3 + 4); 
the other two have 10 arms. 
The specimen from Stat. 299 has 19 arms 80 mm. long; there are no IIIBr series; 
the cirri are XVII, 18 — 20, 15 mm. long. 
Of the two from Banda the larger has about 26 arms about 105 mm. long; three of 
the IIBr series are 2; the IIIBr series are 3 (2 -f- 3); the cirri are XV, 24 — 26, 22 mm. to 
25 mm. long. 
The very fine example from the reef at Haingsisi has 20 arms 140 mm. long; all ot 
the IIBr series are present, and all are 4 (3 + 4); the cirri are XIII, 20 — 21, 15 mm. long; 
the longest cirrus segments are not more than one third again as long as broad ; the animal 
is very robust: the brachials are very short with strongly produced distal edges. 
From the coral reef at Noimini (Stat. 296) there is a small specimen with 15 arms; 
one of the IIBr series is 2. 
Dr. Reichensperger has recently described, under the name of Capillaster clat'ki, a 
new species of this genus from the Aru Islands based upon two specimens dredged by Dr. 
Merton near Pulu Bambu in 10 metres; the cirri of the supposed new form are XXVIII — 
XXXIV, 20 — 21, 15 mm. long, stout, the longest segments (fifth-seventh) slightly longer than 
broad; the arms are 27 — 35, 85 mm. to 100 mm. long. 
The essential character of Capillaster clarki is the shortness of the proximal cirrus 
segments. The numerous specimens from the Aru Islands in the "Siboga" collection are for 
the most part characterized by rather unusually short cirrus segments, but as a similar shor- 
tening of the longer proximal cirrus segments is not infrequent in examples of the species 
from other localities I do not believe that C. clarki can be maintained as a separate form. 
The smaller of Dr. Reichensperger 's two specimens has highly irregular division series, 
as in the type of C. viariae. But this same extraordinary irregularity is occasionally found in 
undoubted specimens of C. nmltiradiata^ and, rather more commonly, in the large species of 
the allied genus Nemaster which represents Capillaster in the Atlantic. 
The ten armed young" of this species may be distinguished from those of the species of 
Comaster or of Comanthus by the long and slender cirri with a slight distal taper which, as 
commonly found, are nearly or quite straight. The cirri of the young of the species of Comaster 
and Comanthus are shorter with fewer segments, and the distal half is always strongly 
recurved. 
