SACCAMMININAE 327 



bulbous or club-shaped, or irregularly branching, with terminal apertures. In this 

 final stage the wall is reduced to almost papery thinness and is very fragile. No perfect 

 specimens in this stage have been found, nor does it seem probable that they would 

 survive the treatment of cleaning the material. A sufficient number of fragments repre- 

 senting the final stage, ranging up to 2-00 mm. in length, associated with specimens of the 

 second stage with fractured terminal tube was, however, found at WS 225 to justify the 

 foregoing reconstruction of the third or final stage of growth. The height of the complete 

 organism is probably 4-5 mm. Perfect specimens may yet perhaps be found in sheltered 

 crevices of sponges, etc. 



One or two specimens have been seen with two diverging tubes on the same basal 

 " pad ", but these appear to be abnormal. The shape of the basal pad varies according to 

 its locus adhaerendi. On stones and shells it is invariably circular and very depressed, but 

 when the organism has started growth on the stem of a zoophyte, the pad increases in 

 size conformably with the surface of attachment and becomes elongated and more 

 rapidly conical. Such specimens were found at WS 225, but were very uncommon. 

 Intermediate forms of the basal pad also occur. 



In one very interesting specimen which we figure, the basal pad had probably been 

 attached to some soft organism which had decayed. The branching passages of the 

 earliest chamber are clearly seen through a thin chitinous film, which covers the base of 

 the pad, and leaves no doubt as to the astrorhizid character of the organism. It measures 

 I -20 mm. in diameter. 



Dendronina papillata is evidently very closely related to Dendronina limosa var. hutnilis 

 (H.-A. & E. 1922, TN, p. 81, pi. ii, figs. 7-9), but diff'ers from it in the spreading 

 character of the basal pad, and by an invariably sessile habit. 



The species is not uncommon at WS 225 and 242, less frequent at the other stations. 



The removal of this species from the genus Dijfusilina does not aff^ect our original 

 diagnosis of that genus and its genotype D. hiimilis. 



Sub-family SACCAMMININAE 

 Genus Psammosphaera, F. E. Schulze, 1875 



60. Psammosphaera fusca, Schulze (Plate VIII, figs. 1-4 and Plate XVII, figs. 4-6). 



Psammosphaera fiisca, Schulze, 1874-5, 1^' P- ^'S' pl- "> ^S- ^■ 



Psammosphaera fusca, Brady, 1884, FC, p. 249, pi. xviii, figs. 1-8. 



Psammosphaera fusca, Heron-Allen and Earland, 1912, etc., NSG, 1913, p. i, pis. i-iii. 



Twenty-nine stations: 230, 388; WS 71, 73, 76, 77, 78, 79, 80, 83, 84, 87, 88, 90, 91, 92, 93, 94, 

 108, 109, 210, 215, 217, 221, 225, 246, 248, 431, 433. 



The species is generally distributed and often common, and it presents an unusual 

 amount of variation. What may be described as the normally spherical and regularly 

 constructed type occurs at 230 and WS 71, 80, 92, 109. A similar form, but neatly con- 

 structed almost entirely of coarse sand grains occurs at WS 225. Fine material only is 

 employed for a nearly spherical form at WS 84. At WS 210, the species was represented 



