HYDKOIHA. BALE. 11 



In S. recta I have seen no internodes with more than one pair, 

 but I have met with two specimens only, both very .small. 



X. unguiculata stands by itself in its excessive variability. By 

 far the most abundant form is abont the same size as the small 

 variety of 8. elmnjahi and is a typical 8<'iii.(lni-i<i, with very few 

 exceptions to the rule that each internode supports a single pair. 

 Somewhat larger forms differ from these exactly as described 

 under 8. elon</at/i, to which this species is, so far, strictly parallel, 

 but beyond these we meet with a series of varieties leading up to 

 forms in which there are no short internodes, and a pinna may 

 support a great number of hydrothecas divided between one or 

 two internodes only. Such forms are in every respect (except in 

 the stem-internodes) typical Tlmim-in', yet it is impossible to find 

 any distinct line of demarcation between them and the ordinary 

 varieties. 



Thai ur! i.i heteromorpha, according to Airman's account, scarcely 

 differs from 8. innjiiiculnta except in the absence of the bidentate 

 margin of thehydrotheca. It is not known however if it exhibits 

 such an extreme tendency to variability as is found in the 

 allied species. Allmau, recognizing that the hydroid combined 

 the characters of more than one genus, placed it under Thninrin, 

 on the ground that the characters of that genus predominated. 

 Among these characters he included the arrangement of the 

 stem-internodes, but in this he was mistaken, as the arrangement 

 is that of a typical Sertulnria, only that the internodes are 

 double, each being formed by the complete coalescence of two of 

 the stem-internodes described in the foregoing remarks. In 

 S. unguiculata single and double internodes occur on the same 

 shoot, and often mixed indiscriminately, and I have little doubt 

 that further examination will prove that the same condition 

 occurs in Allman's species. As I shall presently show there is 

 also a characteristic Thuiarian stem-internode, different in 

 arrangement from that found in the group we are coiisideiiug. 



All the preceding forms ai'e undoubtedly, in my opinion, true 

 Sertidariie, being distinguished from Thuiaria by the paired 

 condition of the hydrotheese, both on the stem and the pinna 1 , 

 and by the presence, and frequently the preponderance, of 

 internodes with a single pair. The occurrence of inter- 

 nodes of higher orders on all but the smallest specimens is 

 exactly similar to what takes place in Sertularella divaricata* 

 Here the rule is that every intei'node bears a single hydrotheca, 

 but in specimens of elongated habit we find that not only on the 

 stem, but on parts of the pinna), several hydrothecte may 

 be borne on a single internode. But no one proposes there- 

 fore to remove the species from the genus Sertularella, In a 

 typical Tlimariii the hydrotheca? are not paired but biserial, and 

 an internode very often bears unequal numbers on the two sides ; 



