12 " ENDEAVOUR " SCIENTIFIC RESULTS, 



it even happens sometimes that they may be closer on one side 

 than the other, twelve on one side for example occupying as 

 much space as thirteen on the other, Allman first referred to 

 Thuiaria such species as his T. sertularioides, a true Sertularia, 

 solely on the ground that several pairs of hydrotheca? were 

 cai-ried on a single internode. He afterwards retracted this 

 determination, but some observers have adhered io it, often 

 however somewhat arbitrarily, applying the rule in some cases 

 and not in others. The fact is that in some of the most typical 

 species of Sertularia the occurrence of internodes of the second 

 and third order is common, and such is the case with S. pumila 

 itself, which nevertheless is always regarded as a typical 

 representative of the genus. 



In the few Australian species of Thuiaria known to me T. 

 lata, T. fenestrata, T. yiiadridens, T. sinuosa, T. sitlarticulata 

 I find a uniform type of stem-division. Each internode supports 

 a pinna and three hydrotheca?, but the arrangement is quite 

 different from that which exists in S. elongata and its allies. 

 the pinna is given off from the middle of the internode, there is 

 a hydrotheca below and another above it, and a single one on the 

 opposite side. There is no paired arrangement, so that the 

 unpaired condition of the hydrotheca? is common to the stem and 

 pinnae. In one or two of these species the nodes may be 

 indistinguishable on the older parts of the stem, but the arrange- 

 ment of the pinna? and the hydrotheca? is as described, and on 

 the more recently-formed portions the internodes are distinct. 

 Some of the hydroids here associated are now ranked by most 

 observers under the genus Sertularella ; to me, however, they 

 appear a thoroughly homogeneous group, and I am quite unable 

 to find any distinction of sufficient importance to justify their 

 separation. T. lata and T. quadridois doubtless have a very 

 Sertularella-like aspect, owing to the regular alternation of the 

 hydrotheca 3 , which is a necessary result of the narrowness of the 

 pinna? forcing them into that position. In the broader forms of 

 T. quadridens, where the two series have room to develop 

 independently of each other, the arrangement is sometimes 

 subalternate rather than strictly alternate. Those varieties of 

 S. divaricata in which several hydrotheca? are carried on a single 

 internode approximate very closely to T. lata, yet there is a 

 radical difference readily perceived by studying the development 

 in the terminal portions of the pinna?. Even in those foi'ins of 

 S. divaricata in which the T It tti aria-like structure is most 

 apparent var. suldicliotottta for example the newly-formed 

 portions are, so far as I have seen, divided into the single-celled 

 internodes characteristic of the genus. But the nodes, though 

 unmistakeable, are very slightly marked, and as the perisarc 

 thickens with growth many of them become obliterated. This 



