70 DISCOVERY REPORTS 



loop. The sperm duct passes almost straight forward beside the rectum. No ovary was found in any 

 zooid examined, but the zooids are known to be hermaphrodite. 

 Distribution. Eastern Australia; North Island, New Zealand. 



Subfamily Polycitorinae Michaelsen, 1904 



Berrill (1950) proposed the subfamily Holozoinae for the three genera, Distaplia Delia Valle, 1881, 

 Holozoa Lesson, 1830, and Sycozoa Lesson, 1830, and pointed out that these genera formed a very 

 natural group of closely related forms. Many authors have had difficulty in deciding whether these 

 three genera are in fact separate. Thus Hartmeyer (1909, in Bronn) regarded Distaplia as a synonym 

 of Holozoa, but maintained Sycozoa. Van Name (1945) allowed the separation of Holozoa mainly on 

 the grounds of the shape of the colony and the alleged pelagic phase in its adult life, but was evidently 

 not entirely satisfied that it is distinct from Distaplia. Brewin (1953) united the two genera under the 

 nomen conservandum Distaplia, at the same time maintaining the separation of Sycozoa, and estab- 

 lishing a new genus Hypsistozoa. I agree with Brewin's decision to unite Holozoa and Distaplia, but 

 the position of Sycozoa is more difficult to determine. 



Before Brewin's (1953) paper, the colony of Sycozoa had been described as typically having one 

 terminal common cloacal opening, but she found in living material from New Zealand waters a ' very 

 regular arrangement of the common cloacal apertures around the periphery of the distal part of the 

 head, a region entirely free of zooids . . . '. As I shall show below, this is certainly not the case in the 

 ' Discovery ' specimens, in which examination of many well-preserved colonies, both of S. sigillinoid.es 

 and S. georgiana, showed one terminal common cloacal opening. Brewin emended the generic 

 description to agree with her New Zealand material, and acceptance of the emended diagnosis would 

 exclude both the type species S. sigillinoides and also S. georgiana. This seems to be unjustified, and 

 I suggest that the arrangement of the common cloacal openings be disregarded. 



The question is further complicated by the structure of a new species, Sycozoa anomala, in the 

 ' Discovery ' collections, from north of New Zealand. The main reason for separating Sycozoa from 

 Distaplia has hitherto been the unisexual nature of the colonies of Sycozoa and their hermaphrodite 

 nature in Distaplia. But the new species, S. anomala, from the ' Discovery ' collections, which in all 

 other respects agrees well with other species of Sycozoa, has hermaphrodite colonies as well as uni- 

 sexual ones. The distinction between the two genera thus seems to break down. In 5. anomala the 

 position of the ovary also is somewhat intermediate between the genera, and the logical course might 

 be to unite Sycozoa with Distaplia, the name Sycozoa having priority. But as there are many well- 

 known species bearing the name Distaplia, this course would undoubtedly lead to great confusion, 

 and I have retained both Distaplia and Sycozoa. 



Genus Sycozoa Lesson, 1830 



There are several reasons why separation of species of Sycozoa is difficult, one of the more im- 

 portant being ignorance of the changing forms of the colony during its normal annual cycle, and under 

 different conditions of environment. Thus S. ramulosa (Herdman) and S. umbellata (Michaelsen) may 

 be two forms of S. sigillinoides. The characters of S. ramulosa which Arnback (1950) used to distinguish 

 it from S. sigillinoides are: (1) its longer body and especially the branchial sac; (2) the extension of the 

 testis beyond the intestinal loop; (3) the small number (one to three) of eggs in each brood pouch. 



The shape of the body depends greatly on its state of contraction and is rather a difficult character to 

 determine. The size of the testis varies according to the degree of sexual development of the zooid and 

 is therefore also unreliable. It may be noted that, among the ' Discovery ' material, there are colonies 

 combining character (2) of S. ramulosa with character (3) of S. sigillinoides (e.g. St. 1230). Moreover, 



