ASTEROIDEA 77 



St. WS 791. 45° 38' S, 62° 57' W, 97-96 m., i specimen. 



St. WS 792. 45° 49' 30" S, 62° 20' 15" W, 102-106 m., dark green sand, i specimen. 



St. WS 793. 45° 53' S, 61° 35' W, 108-111 m., 2 specimens. 



St. WS 795. 46° 14' S, 60° 24' W, 157-161 m., I specimen. 



St. WS 796. 47° 49' 37" S, 63° 42?*' W, 106-113 m., I specimen. 



St. WS 797. 47° 45' 18" S, 64° 10' 30" W, 115-112 m., stones, 2 specimens. 



St. WS 801. 48° 26i' S, 61° 28' W, 165 m., 3 specimens. 



St. WS 805. 50° loj' S, 63° 29' W, 148 m., 2 specimens. 



St. WS 809. East of Porto Santa Cruz, Argentina, 49° 28J' S, 66° 29' W, 107-104 m., i specimen. 



St. WS 817. 52° 23' S, 64° 19' W, 191-202 m., 3 specimens. 



St. WS 818 A. 52° 31^' S, 63° 25' W, 272-278 m., I specimen. 



St. WS819A. 52° 41' 52" S, 62° 39' 30" W, 312-329 m., I specimen. 



St. WS 839. 53° 30' S, 63° 29' W, 403-434 m., I specimen. 



In the considerable series of specimens of all sizes there is much more uniformity of 

 structure than I found in north Pacific examples of C. crispaUis,^ of which I had 883 

 specimens. Owing to the great variability among these, and perhaps somewhat influenced 

 by Perrier's remarks,^ I questioned the validity of C. ausiralis, of which I had no ade- 

 quate material. It now seems to me possible that Perrier was not dealing with typical 

 australis but with variants arising from a mingling of aiistralis and proairotor, as his 

 specimens were taken, with the exception of one locality off Argentina, in the Magellan- 

 Cape Horn region (Beagle Canal, Franklin Canal, Murray Narrows). There is no great 

 difficulty in distinguishing typical australis from north Atlantic material. 



As Sladen (1889) has already pointed out, australis can be distinguished from ^ro- 

 ciirator by its constantly larger paxillae and fewer marginal plates. I have only two 

 specimens oi procurator, from 122 fathoms off Chile: R is 22 mm. and there are 17 and 

 18 marginal plates, while in comparable examples of australis there are 12 and 13. The 

 largest paxillae of the Chilean specimens have 6 or 7 spinelets (none in centre) and are 

 therefore similar to the type oi procurator. In australis the largest paxillae (in interradial 

 triangles) commonly have 9-1 1 peripheral and upward of 5-8 central spinelets, all 

 longer and stouter than in procurator. Naturally there is a big contrast in the size of the 

 paxillae. 



While australis looks more like crispatus than either, one resembles procurator. I 

 believe the first two are nevertheless distinct in the sense that any two nearly related 

 species of sea stars are distinct. Why two recognizable species of Ctenodisciis occupy the 

 two sides of southern South America is a puzzle. Ctenodiscus is very widely distributed 

 in the north and eastern Pacific, namely, from the Sea of Japan to the Bering Sea and the 

 Arctic Ocean, south along the American coast to central California, Gulf of California, 

 and Gulf of Panama. There are no records south of Panama until the home oi procurator, 

 off Chile, is reached. 



However, it may be recalled that Ceramaster patagonicus is found on the Alaskan coast 

 and in the southern Bering Sea. Odont aster crassus of California is closely similar to 

 O. grayi of the Falkland-Magellan region. The relatively gigantic Pedicellaster magister 



1 Fisher, 191 1, p. 32. ' Perrier, 1891, pp. 143, i44- 



