262 



related to pseudo-digitata. Judging from Dr. Mortensen's prepa- 

 rations, I have no doubt in identifying the Suez-specimens as pseudo- 

 digitata. The fact that the specimen with tentacles has only ten, may 

 be due to an anomaly, as all the complete specimens at hand 

 have 12 tentacles. 



Protankyra pseudo-digitata was previously reported only from 

 Bohol and Makasser. The collection at hand shows that it is more 

 widely distributed, occurring also in the Gulf of Siam and in the 

 Red Sea. On the other hand I do not suppose that it occurs at 

 Japan, where it is represented by bidentata. 



Protankyra petersi (Semper). 



Synapta petersi Semper. 1868. Die Holothurien. pag. 230, PI. 39 fig. 12. 

 Protankyra petersi. 0stergren. 1898. Das System der Synaptiden. p. 1 17. 



Clark. 1908. The Apodous Holothurians. pag. 108. PI. 



IV fig. 15—16. 



In the collections of the Copenhagen Museum there are some 

 preparations made by C. Semper, containing spicules of some of 

 his more interesting Holothurians. Among these preparations there 

 is one labelled: Synapta Petersi S. Amboina, including some few 

 anchors and plates of this species, which must belong to the type- 

 specimen, the only one known till now. 



Semper has not given any figure of the anchor-plates in petersi, 



and he only writes: "Die Anker sie sind sehr eigenthümlich 



gebildet, asymmetrisch; der eine kurze und sich bis auf den Stiel 

 niederbeugende Ast trägt 3 — 4 kleine Zähne, der andre lange vom 

 Stiel abtredende ist ganz glatt. Ausserdem kleine kreuzförmige Kör- 

 perchen. Die Ankerplatten sind denen von Synapta pseudo-digitata 

 ähnlich". Semper is not quite right in saying that the plates in 

 petersi are like the plates in pseudo-digitata. The outline of the 

 plates in petersi is rather like that of the small plates in pseudo- 

 digitata, but the plates differ distinctly, those in petersi being thick, 

 and consisting of several layers of spiny network, which is densest 

 close to the articular end of the plate (Fig. 51. 1). Thus it is quite 

 erroneous whei Clark writes that "The curious grotesque anchors 

 are accompanied by normal plates perforated with numerous smooth 

 holes". The anchor-plates in the preparation at hand measure ca. 

 300 fi in length. 



