DESCRIPTION OF SPECIES 83 



Zoom. The zooid is much as described by Arnback, but there is only a relatively short waist joining 

 the thorax and abdomen. This condition may have resulted from contraction. 



Larva. No fully developed larvae were found in any of the colonies, but the large colony from 

 St. 190 had some advanced embryos with the tail differentiated. These embryos had a length of i-i to 

 i-2 mm. 



Remarks. Sluiter (191 2) described a species C. antarcticus, which Arnback (1950) retained, but 

 thought would very probably prove to be a form of C. dellechiajei, if more material could be studied. 



The only way of distinguishing C. dellechiajei and C. antarcticus, other than by their distribution, 

 is by the shape of the spicules. All antarctic specimens so far examined, including the ' Discovery ' 

 ones, have spicules with a distinctly crenelated margin (Text-fig. 23 A). Specimens from other regions 

 have the margin smooth or minutely notched (Text-fig. 23 B, C), but there is sufficient variation in 

 the size and distinctness of the notches to render doubtful the value of this character for separating 

 species. I therefore propose that the present specimens, together with those previously called 

 C. antarcticus, should be recognized as a form antarctica nom.nov. of C. dellechiajei. 



Distribution. The form antarctica is recorded only from the vicinity of Graham Land. 



Subfamily Atapozoinae Brewin, 1956 



Genus Atapozoa Brewin, 1956 

 Atapozoa marshi Brewin (Text-fig. 24) 

 Atapozoa marshi Brewin, 1956c, p. 31, fig. 1. 



Occurrence. St. 935 : North Island, New Zealand, 84 m. 



Colony (Text-fig. 24 A, B). There are two specimens. The larger has a cylindrical stalk 1-3 cm. 

 long and an ovoid head i-o cm. long. In the smaller specimen the stalk is only 0-3 cm. in length and 

 the head is globular, with a diameter of about 1 -o cm. In each the stalk is encrusted with white shelly 

 debris but the head is bare. The zooids, which are confined to the head, are seen through the test as 

 orange-brown bodies. The test is semi-transparent and firm, but contains no 'kotballen' (masses of 

 foreign matter) which Brewin found in her specimens. 



Zoom (Text-fig. 24 C). Large zooids in which the abdomen is well developed reach a total length 

 of 3-4 mm.; zooids with a short abdomen are little more than half that length. The siphons are both 

 provided with six distinct lobes. Eighteen to twenty longitudinal muscles are present on each side of 

 the thorax, and converge to form a more compact band of muscle which extends down the whole 

 length of each side of the abdomen. Some zooids have a short vascular process (v.pr.) at the lower end 

 of the abdomen. A small thin-walled projection from the base of the dorsal side of the thorax (b.p.) 

 is present in some zooids and may represent the brood pouch described by Brewin. The whole body 

 is of an orange-brown colour when preserved in alcohol. 



Branchial sac. Details are difficult to distinguish owing to the contraction of the thorax. The oral 

 tentacles are closely spaced and number at least twenty. There are three rows of stigmata but the 

 number in each row could not be determined; they probably exceed twenty. The hooked dorsal 

 languets on the two transverse bars are quite prominent. 



Gut. The oesophagus (oe.) is nearly twice as long as the stomach, and in this respect differs from 

 Brewin's type specimens most zooids of which had a short oesophagus. The stomach (st.) is smooth- 

 walled and the intestinal loop short, without any apparent divisions. In the type specimens the anus 

 was smooth-edged; in the 'Discovery' material the anus (a), which lies opposite the second row of 

 stigmata, has two lips. 



Gonads. Gonads were not developed in the present specimens. 



