HYDR01DA. BALE. 35 



A pair of brandies originates from two successive iuternodes, 

 and they are therefore not opposite in the strict sense, though 

 to the naked eye, and when swathed, as they soon are, by the 

 enveloping supplementary tubes, they appear practically opposite. 

 They spring slightly towards the front, but soon bend back 

 sufficiently to bring both series virtually into one plane. The 

 occurrence of unpaired branches may be due in some cases to the 

 production of a single one only, but more often probably, to the 

 loss of one of a pair. As in all the allied forms seen by me, the 

 proximal part of a branch consists of sevei'al short internodes 

 (in this species sometimes as many as a dozen), each of which 

 bears a median sarcotheca only. Ritchie's statement that in 

 A. crucialis the branches bear on their proximal internodes a 

 series of TiydrotJiecce only, is probably a slip. 



The hydrothecre nearly resemble those of A. tasinait/r,/ 

 except in their size and proportions. Their length averages 

 about '45 mm. as against '37, but their median width in front 

 view is only about '15, while that of A. tasmanica is about '19. 

 A little variation occurs in the marginal outline, but in the most 

 typical specimens there is a minute denticle or undulation between 

 the anterior tooth and each of the lateral ones, while between 

 these and the lateral sarcotheca3 the border is nearly straight 

 except for two small convexities. The margin differs from that 

 of the next species in the presence of the angular lobes behind the 

 lateral sarcothecse. 



The anterior sarcotheca? are shorter, but more prominent than 

 those of A. tasmanica, but- in these characters, as well as in the 

 canaliculate aperture, they agree with those of A. mawocarpa. 



Some later specimens, received from the Australian Museum, 

 enable me to supplement the account of the gonosome, and to 

 recognize in the structure of the corbulze an instance of sexual 

 dimorphism not previously noticed. Torrey and Ann Martin, 

 have pointed out that in a number of species which are known 

 to possess two forms of corbulae those containing the female 

 gonophores are completely closed, except for a series of apertures 

 at the bases of the leaflets, while in the male type the leaflets are 

 separated for a portion of their length, or even throughout In 

 A. billardi the corbnlse of both sexes are, so far as I have seen, 

 closed throughout (always excepting the series of large openings 

 at the bases (if the leaflets), but there is a great difference in the 

 extent to which the protective armature is developed. In the 

 female corbula the distal sides of the leaves are continued past 

 the line of union with the next leaf and turned outward, being 

 continued into a free lateral \\ing which is widened upward in 

 the form of a large spreading secondary leaf, rising above the 



