UMBELLULIDAE 253 



This very characteristic species has dimorphic spicules: (i) Spicules up to 3 mm. long form a 

 narrow band along the aboral side of the pinnulae adjoining the strong band of similar large spicules 

 along the aboral side of the tentacle stem, a short way below its base on the anthocodia. (2) Small 

 fusiform spicules crowded in the surface tissues of autozooids and rachis. All the spicules are round 

 in section, they are never three-flanged. The axis is round in section and remarkably feebly calcified 

 in contrast to the heavy armoured autozooids with their calcareous spicules. 

 (Holotype: British Museum (Nat. Hist.). Reg. No. 1881. 2. 11. 21.) 



Umbellula thomsonii Kolliker 1874 

 U. thomsonii Kolliker 1874a, p. 13. 

 U. thomsonii Kolliker 1874ft, P- i°> P'- i> figs. 1-5. 



U. thomsonii+ U. Giintheri+ U. leptocaiilis + U. simplex Kolliker 1880, pp. 18-20, PI. 9, figs. 34-36. 

 Non-U. giintheri A. Agassiz 1888, p. 314. 

 Non-U. giintheri Studer 1894, p. 57. 

 U. crassiflora Roule 1905, p. 456. 



U. intermedial U. rosea+ U. Kollikeri Thomson & Henderson 1906, pp. 94, 95, 97, PI. 5, fig. 5. 

 U. Hendersoni Balss 1910, p. 68, footnote. 

 U. Giintheri Broch 1913, p. 3, Pi. i, fig. i. 



U. thomsonii+ U. giintheri+ U. rosea+ U. crassiflora+ U. intermedia+ U. simplex Kukenthal 1915, pp. 52, 54, 56-58. 

 U. aciailifera J. Stuart Thomson 1915, p. 20, PI. 2, fig. 2, text-fig. (on p. 21). 

 U. Giintheri Hickson 1916, pp. 120 et seq. 

 U. aciculifera J. Stuart Thomson 1923, pp. 49 et seq. 

 U. giintheri+ U. thomsonii Deichmann 1936, pp. 268, 269. 

 U. giintheri Broch 1957, p. 351, PI. i, fig. i, text-figs. 1-3. 



This very characteristic species is easily identified by its three-flanged spicules (Broch 191 3) and 

 its strongly calcified, quadrangular rigid axis. 



Discussing the synonymy of U. giintheri Kolliker 1880, I (Broch 1957) suggested the possibility 

 that both U. leptocaiilis Kolliker and U. simplex Kolliker (1880) had been founded on young specimens 

 of U. giintheri. An examination of the original specimens in the British Museum has fully corrobo- 

 rated the suggestion. 



In the same place (1957, p. 357) the position of U. thomsonii Kolliker 1874 was also touched upon, 

 but owing to discrepancies in the different papers of Kolliker (1874^, 18746, 1880) and the inter- 

 pretations of Kiikenthal (191 5), I found the problem so confused that it was impossible to solve it 

 without a study of Kolliker's types, although most statements indicated U. giintheri to be a synonym 

 of U. thomsonii. 



Kolliker based his descriptions of U. thomsonii on two specimens from the ' Challenger ' Expedition 

 (1880, p. 19, specimens A and B). The smaller specimen B could not be found in the collections of the 

 British Museum;* according to Kolliker its height was 270 mm., but the specimen was defective. 

 Specimen A has accordingly been selected as the lectotype of U. thomsonii. Kolliker wrote that 

 the specimen was 895 mm. high and complete. However, the upper part with the autozooid cluster 

 has now been broken oflF just below the swollen rachis. The upper limit of the holdfast cannot be 

 ascertained with certainty, and the basal ' bulb ' is contracted and not discernible. 



Text-fig. 2 shows the position of the autozooids ; it is a little at variance with the descriptions both 

 of Kolliker (1874^, 18746, 1880) and Kiikenthal (1915). Originally there was one primary polyp 

 (hatched in text-fig. 2) and twelve secondary autozooids, but two of the latter (stippled in text-fig. 2) 

 had been cut away. The primary polyp and nine secondary autozooids are arranged in a whorl, and 

 can be regarded as a direct continuation of the developmental sequence demonstrated by me (1957, 

 * Presumably the specimen was kept by Kolliker as a reward for working out the collection. 



