184 PERCY SLADEN TRUST EXPEDITION 



14. Proboscidactyla tropica, Browne, 1904. 



Willsia sp. Huxley, 1877, p. 120, fig. 17. 

 Prohoscidactyla tropica, Browne, 1904, p. 727. 



Locality. Amirante Isles, Desroches Atoll. 100 — fms. 16 Oct. 1905, kk. 

 1 specimen. 



Huxley, when off the Louisiade Archipelago in 1849, found a medusa which he briefly 

 described and figured under the name of Willsia. In my revision of the Williadae (1904) 

 I recognised this medusa as a distinct species and called it Prohoscidactyla tropica. 



Bigelow (1909, p. 220) considers Prohoscidactyla tropica to be identical with 

 Prohoscidactyla ornata McCrady, 1857, var. stoloni/era, Maas, 1905. 



One of the characters, which I made use of to distinguish Prohoscidactyla tropica 

 from all the other species of the genus, was the form of a blastostyle bearing a cluster of 

 nematocysts at its free end, and upon the sides of this blastostyle the medusa-buds develop. 

 This blastostyle is similar in structure to the blastostyle of the hydroid Lar sahellarum, 

 upon which medusa-buds develop, which when set free belong to the medusoid genus Willia. 



Maas (1905) in his description of the variety stolonifera from the Malaysian region 

 makes no mention of this peculiar blastostyle, and if it had been present in his specimens 

 he could not have failed to have noticed it. 



Bigelow (1909, p. 219) states that Prohoscidactyla gemmifera (Fewkes) is only the 

 budding form of Prohoscidactyla ornata, McCrady, 1858. I cannot very well link 

 Prohoscidactyla tropica to Prohoscidactyla ornata by means of Prohoscidactyla gemmifera, 

 as Brooks (1880) states clearly that the stolons have branches and each branch terminates 

 with a medusa-bud. There is here also no mention made of any blastostyle-like stolon 

 with a terminal cluster of nematocysts. Mayer's figures (1910, PL 21) confirms Brooks's 

 statement. Under these circumstances I prefer to retain Prohoscidactyla tropica as a 

 distinct species until at least we know more about the life-histories of the other species 

 of the genus. 



The specimen in the "Sealark" collection is about 2 mm. in diameter, and has lost 

 its natural shape owing to contraction. The velum is very narrow. The stomach has 

 four perradial lobes, from which the four main radial canals leave to join the tentacles. 

 As in other species of the genus there is no circular canal. Each main radial canal has 

 the appearance of being twice dichotomously branched, each terminal branch going to a 

 tentacle. At the juncture of the first branch of each canal is situated a blastostyle-like 

 stolon. This blastostyle has rather the appearance of a tentacle, for at its free end there 

 is a large cluster of nematocysts. The medusa-buds are at different stages of development. 

 Two to four on each stolon, and the largest is nearly ready for liberation. They are 

 arranged round the sides of the stolon. Huxley, however, states that the buds are uni- 

 lateral. He saw them alive, whereas I have only a contracted specimen. In one of the 

 perradii at the junction of the second branch there is a second blastostyle-like stolon with 

 a small bud developing. So apparently the medusa has stolons on the junction of the 

 second branch as well as on the first. 



There are sixteen tentacles, rather thick, in a contracted condition, with a thick semi- 

 circular band of nematocysts on the inner side of the basal bulbs, which project into 



