and its Reproduction. 315 



productive buds are borne in clusters of three or four on the 

 upper portion of the body, and when matm-e detach themselves 

 as free medusiform zooids [jylanohlasts'^) ; they are destitute 

 of an ectothecal covering (a character which they share with the 

 gonozooids of Clavatella, Corymorpha^ and Gladonemo) ^ and 

 are therefore freely exposed to the surrounding water. In an 

 early stage of development the buds are much elongated 

 (PL XIX. fig. 1), and take on their hemispherical form as 

 they approach maturity. 



The planoblast (PI. XIX. figs. 3, 4), at the time of its libera- 

 tion, is almost hemispherical in form ; the umbrella is perfectly 

 colourless and destitute of thread-cells. The digestive sac or 

 manubrium is very mutable in shape; normally it is sub- 

 cylindrical, and somewhat swollen at the base, with a slightly 

 lobate mouth. Six radiating canals ti'averse the umbrella, 

 terminating on the margin in as many oval bulbs of a brownish 

 colour, from which six smooth tentacles originate. Both ocelli 

 and lithocysts are wanting 5 but halfway between every two 

 tentacles a minute sac occurs on the margin of the umbrella, 

 containing two or three glittering bodies, which appear to be 

 thread-cells (PL XIX. fig. 6). The planoblast, when detached, 

 bears with it a portion of the peduncle which had formed the 

 bond of connexion between it and the parent stock ; this sur- 

 vives as a somewhat conical process above the base of the 

 manubrium (PL XIX. fig. 3, .r), but is no doubt absorbed 

 after a time. 



Six is an unusual number for the radiating canals ; amongst 

 the British Hydroida it is met witli only in ClavateUa (which 

 has also occasionally four) and in the genus WUlsia of Forbes. 

 The smooth tentacles (which closely resemble those of the 

 polyjiite), the absence of the customary organs of sense, and 

 the minute marginal sacs with thread-cells may also be noted 

 as significant characters f. 



There can be no doubt that the genus Lar must stand as 

 the type of a distinct family amongst the Hydroida Athecata, 

 which will present features as strongly marked as those of any 

 group in the suborder. Indeed the important modification in 

 the structure of the polypite has scarcely a parallel within the 

 limits of the whole order. A question, however, arises as to 



* Planoblast (wanderiug bud) is a happy and expressive term intro- 

 duced by AUman to desigiiate the free gouozooid. 



t Allnian considers it probable that the marginal sac is the origin of 

 " what in the adult Medusa would become an inteiTadial marginal tentacle " 

 (' Monograph of the ({ymnoblastic or Tubularian Ilydroids,' part ii. p. -)27). 



The substance of this paper was communicated to Prof. /Vllman by 

 letter, and is incorporated in the second part of his ' Monograpli ' just 

 issued by the Ray Society. 



23* 



