Fauna Associated with Crinoids of Tropical Coral Reef, etc. 83 



other than the spines on the dactyli of the thoracic legs, but the animals 

 are thigmotropic and swim back immediately to the host when they 

 have been detached. 



Galathea elegans Adam and White. (Plate 1, Fig. 5.) 



G. elegans, ADAM AND WHITE, Voyage of the Samarang. 



G. longirostris, DANA. 



G. deflexipous, HASWELL, Proc. Linn. Soc. New South Wales, vol. vi; MIERS, Zool. Coll. H. M. S. 



Alert. 

 G. longirostris, SOUTHWELL, Anomura, Ceylon Pearl Oyster Rep. Roy. Soc., part V, p. 220, 1906. 



In earlier literature there are only three records of the occurrence of 

 galatheids on crinoids. The first is that of Dana, who originally 

 described G. longirostris from a crinoid dredged at Fiji; the second is 

 that of Haswell, who founded a new species G. deflexipous, associated 

 with Synalpheus comatularum on an unidentified crinoid. The third 

 likewise concerns G. longirostris, which is mentioned by Southwell as 

 obtained near Ceylon, clinging to Antedon bella. 



G. elegans Adam and White, G. longirostris Dana, and G. deflexipous 

 are all very closely related. They all possess a long rostrum with from 

 5 to 10 small lateral denticles, a very typical dorsal ornamentation, and 

 have a similar colour scheme, with longitudinal stripes of pigment on 

 the dorsum. The first-named species has not been noted as occurring 

 on crinoids, but this may be the fault of the collectors. Balss, in 

 describing a specimen from Japan, which he assigns with some doubt to 

 G. elegans, says with regard to its occurrence: " Wahescheinlich an 

 Comatuliden, wie es Haswell von der nahe verwandten G. deflexipous 

 und Southwell von G. longirostris Dana angeben. Darauf weist die 

 bunte Farbung hin, die wohl als eine mimetische zu deutenist." Prob- 

 ably the whole group of related species are crinoid dwellers. If not, it 

 will be strange that so marked a pattern should occur on a free-living 

 galatheid, when that is a character so definitely associated with crinoid 

 commensalism, both here and in other groups. 



The three species are principally defined with regard to the characters 

 of the rostrum; thus: 



G. elegans: G. longirostris: G. deflexipous: 



Rostrum more than half the length of the Rostrum fully as Differs from elegans only in 



rest of the carapace, " with 7 small den- long as half the the fact that the rostrum 



ticulations on each side" (Haswell). carapace ; 'mi- is entirely deflexed. 



But Balss gives the latter number as 9. nutely 5-6 serru- 



But most important of all, the original late." 

 figure only shows 5 or 6. 



With regard to the third species Miers says : 



"In more than one of the specimens in the Museum collection the rostrum 

 is slightly deflexed, and I think that G. deflexipous Haswell, from Albany 

 Passage, should be regarded merely as a marked variety of G. elegans." 



In my own specimens the number of denticulations on each side is 

 from 6 to 7 ; in one individual there were 6 on one side, 7 on the other. 



