Mf'/iterruneun '!'< rvia (iu<l Miiionon. 5 



of four and :i half scries. In all cases the iriercasc of the 

 zoojcial tube is very gradual. 



Seeiiif^ at what an early stage preparations are made for 

 the form of the colony, and what radical differences there 

 are almost from the heginning, it would seem strange if the 

 colonial fornj did not give any assistance in classification; an<l 

 when the ovieells are known in most genera, then studying 

 zo(eoial, zoarial, and ovicell characters together will show us 

 their importance, hut it does nor fidlow that the value will 

 be the same in all genera. We are finding an increasing 

 uuuihcr of species which may occur either adnate or erect. 



Tervia inx'gularis (}>lci\ei;h'\ii\). (PI. I. ligs. 1-9.) 



For syuonyius, see Miss Jelly's Catalogue uiulcr Fi/isparsa, a.nd add : — 



Terrid irret/ulan's, Jull. & Calv., Brv. de Camp, dt! riliruiidell'', ])p. 1 14, 

 I'u, pi. xiv. Hf,'. 7 (l!H)3) : Waiera, " Mar. Fauua Jirit. E. Africa," 

 Proc. Zool. Soc. 1914, p. Si'-i, pi. iv. tig. 8. 



Idm'nifa imijularis, Ilaswell, " Pol. Qiieen.slaud," Proc. Linn. Soc. N. 

 S. Walt's, vol. V. p. '.'>') (1H80) ; Waters, " liry. X. S. Wales," pt. iii., 

 Ami. it MaJ,^ Nat. Hi.st. ser. 5, vol. xx. p. 2.jo (1887) ; " Ovieells dl" 

 Cvclos. Hrv," Journ. Linn. Soc, Zool. vol. xx. p. 279, pi. xiv. 

 fi.^rs. "), 6 (1'888). 



Filiitpaisa irrei/ularis, Calvet, Trav. & Talisman, p. 472; Norman, 

 'Madeira,* p. 279, typical; var. siiperbo, p. 27!>, pi. xxxiv. ligs. 1-;') 

 (UH"»i»); IJarrosa, " Bri. de la Estacioii de Biologia Mar. de Santander, " 

 Trab. del Mas. de ciencias. nat. No. o, p. o7 (l!M2); Friedl, II., 

 " liry. des Adrift," Zool. Anzeifror, vol. xlix. pp. 22"j, 2(58 (1917). 



Teruia foUni, Calvet, " Camp, du Caudan," Ann. de I'L'uiv. de Lyons, 

 p. 2Go, pi. vii. Ii^^ 3 (IS'JtJ). 



/'«V/5/)«r.<(/ i'flri«w<, Neviani, Cont. Bri. fo.«s. Ital. p. 43, pi. iv. fi;.'. 21 

 (1891). 



Tervia jelhjfe, Harmer, Poly, of the 'Siboga'Exp. pt. 1, p. 143, pi. xi. 



tifr8."l-3 (191")^ 



This is abundant near Capri, and the zooecia, as well as the 

 internodes, are very long. The short internodes oi Fifisparso, 

 var. jtennat a, Normau, hardly seem to be T. irregularis. On 

 the anterior surface there is often a good dcuil of inegularity 

 in the position of the zotecia, aiul the outer zooecium shows 

 on the dorsal surface (PI. I. fi-s- 1, 2, -4, 5, 14). It was 

 placed under Filisparsa until the dorsal ovicell was found, 

 and it has much in common with Hornera. I cannot feel 

 tliat we yet know the limits of Tervia, for T. (jluhidifera. 

 ('anu & l^issler *, T. /lyrifera, C & B., 7'. tuiiti<ia,C. & M., 

 as described, have com[)lete series as in h/i/toma, and Mould 

 not fall under Tervia as described liy .1 ullieii, for he sa\ s 

 this genus difl'ers from Idmonea by the presence of a certain 

 * Loc. cil. pp. 790, 791, 792. 



