UMBELLULIDAE 263 



U. Jordani Nutting 1908, p. 564, PI. 42, fig. 3. 



U. lovia+U. magniflora Nutting 1909, p. 710, 712, PI. 87, fig. 9. 



U. antarctica+ U. rigida+ U. kollikeri Kiikenthal & Broch 191 1, pp. 286, 292, PI. 15, fig. 15; PI. 26, fig. 16. 



U. magniflora + U. carpenter i Nutting 1912, pp. 42, 43. 



U. carpenteri Kiikenthal 1912, p. 345, PI. 23, fig. 22. 



U. lindahli+U. encrinus ambigua+U. carpenteri + U. rigida+U. antarctica + U . kollikeri +U. jordani +U. loma 



Kiikenthal 19 15, pp. 48-50. 54. ^gs. 55-7. 

 U. antarctica+ U. Weberi+ U. Jordani Hickson 1916, pp. 128, 131, 133, PI. 3, fig. 20; PI. 4, fig. 24; text-fig. 29B, 



D, G. 



U. encrinus var. lindahli Jungersen 19166, p. 500. 



U. encrinus var. lindahli Jungersen 1919, p. 1 158. 



U. encrinus var. lindahli Broch 1929, p. 149. 



U. lindahli Deichmann 1936, p. 268. 



U. encrinus Madsen 1948, p. 17. 



U. carpenteri Broch 1957, p. 357, PI. i, figs. 2-4, text-figs. 4-6. 



Under the name of Umbellida carpenteri, Kolliker (1880) gave a description of 5 specimens (A-E) 

 from the 'Challenger' stations 156 and 157. Unfortunately, he did not say precisely from which 

 station each specimen originated, and all of them have been merged in the same container. However, 

 he mentioned that the specimens A-E exhibit 'a very interesting gradation from a bilateral to an 

 apparently irregular arrangement of the polyps'. Kolliker probably used the largest specimen E as 

 the main base for his description. Specimen C is lost, but the specimens A, B, D and E are in the 

 British Museum (Nat. Hist.). All are figured in Kolliker's paper. 



According to Kolliker, the type specimen E has eight polyps, arranged in the form of a rosette, and 

 surrounding a ninth middle polyp Uke a cup; the axis terminated not in the ninth, but in one of the 

 eight. It is, however, not possible to tell with certainty where the axis terminates if specimen E is 

 examined. It does not end in the (dorsal) wall of the upper autozooid (see text-fig. 3, i), but is bent 

 inwards below its base, disappearing into the tissues of the rachis below the autozooids. This is the 

 condition met with in U. magniflora (see below). Dissection would be necessary to determine whether 

 the central, or the dorsal autozooid of the cluster is the primary polyp. In appearance the specimens 

 resemble those from the Discovery collections which are intermediate between typical carpenteri and 



magniflora specimens. 



In specimen D (Kolliker 1880, PI. 10, fig. 39) all autozooids have contracted tentacles. In the same 

 way as in Cormdaria (Broch 1935) and Solenopodium (Broch and Horridge 1957) the tentacles seem 

 to be laid together, stretched upwards and contracted like those of a slug by inverting the distal (oral) 

 part of the body wall. It is interesting to note in this connexion that polyps with contracted tentacles 

 have very seldom been found among over a hundred specimens of Umbellula which have passed 

 through my hands. The specimen D (text-fig. 3, 2) has eight autozooids in one slightly irregular 

 whorl. The conspicuously smaller, primary polyp is distinctly characterized by the termmal part of 

 the axis running up to about the middle of its (dorsal) wall. The primary polyp has, on the other 

 hand, been displaced a little to the inner side of the whorl.'widently by the outgrowth of the two 

 adjacent secondary autozooids. 



Specimen B (length 105 mm.) figured by Kolliker (1880, PI. 10, fig. 39«) has 3 autozooids, the median 

 of which is the primary polyp, the axis continuing in its (dorsal) wall till about one-third below the 

 tentacle bases; there is a small knob where it terminates, but there seems to have been no free end of 

 the rachis. The specimen differs a little from the youngest developmental stages from the Swedish 

 Deep-Sea Expedition (Broch 1957), but corresponds with two small specimens from Discovery 

 St. 181. 



