89 



times a little toriilose, about 30// in average diameter, only just 

 below the dichotomy reaching 40 ,« or a little more, rather strongly 

 constricted above the dichotomy and above the constriction 

 often a single moniliform swelling (Fig. 72 a, b), nearer the sur- 

 face the filaments first taper (about 14 — 11 f^) becoming irre- 

 gularly torulose and often somewhat constricted ; towards the 

 apices they again grow thicker, the apices being subclavate and 

 obtuse and reaching a thickness of about 19 — 25,«, seldom even 

 27 // (Fig. 72 d, e). 



Of forms characterized by such clavate apices Mr. & Mrs. 

 Gepp mention three species, namely: A. obscura, A, cla<^atiramea 

 and A. Ridleyi. 



Of these the first mentioned species has much thicker fila- 

 ments, these being nearly cylindric ; the habit of the plant is also 

 very different from my species. 



In A. clavatirajyiea the filaments are more torulose remin- 

 ding one in a way more of A. Geppii, but the filaments are much 

 thicker and more straight ; also the habit of this Australian plant 

 is quite different. 



And this is far more the case with the .4. Ridleyi with its 

 tufted, very irregularly branched growth. 



St. Jan: Off Maho Bay in a depth of about 16 meters. 



To this form most probably belongs a quite small plant 

 found in the sea to the west of Water Island in a depth of about 

 30 meters. It only reaches a height of about 534 cm. and has a 

 short, 1 cm. high, vertical rhizome covered with sand ; the stipe 

 is l%cm. long and ^ cm. broad, simple, expanding suddenly 

 into the broad oblong frond. This is 2i/4cm. high and 314 cm. 

 broad, of a tawny green colour and obscurely zonate. 



The filaments in the flabellum nearly cylindric, sometimes a 

 little torulose about 20 — -30// thick, near the surface somewhat 

 thinner 15 — 20/-^ and more torulose and irregularly bent some- 

 times also in the apices a little thicker again. 



Mrs. Gepp to whom I have shown the specimen during a 

 visit to London in 1911 was of the opinion, that it might perhaps 

 be referred to the Avr. Elliottii known from Grenada and it cannot 

 be denied that it shows much likeness with this species but it 

 differs by the more torulose, irregularly bent and tapering surface 

 filaments. 



4. Avrainvillea asarifolia Borgs. 



BoRGESEN, F., The species of Avrainvilleas hitherto found on the shores 

 of the Danish West Indies (Videnslc. Meddel. fra den naturh. Foren. i Kbhvn., 



