SPONGES, ZOOPHYTES, AND SEA-FIRS. 59 



represented in our own, division of labour is carried to a 

 much greater extent than in Hydractinia, and we have 

 floating colonies formed of many different kinds of persons. 

 These constitute the Siphonophora, and are exemplified by 

 such forms as the " Portuguese man-of-war," which is some- 

 times brought by the Gulf Stream to certain parts of our 

 coast. 



Another interesting point about our littoral Hydrozoa is 

 that, as we have already pointed out, they show a consider- 

 able range of variation in regard to the power of forming 

 a skeleton. While some, like Clava, form scarcely any 

 skeletal substance at all, in others, as for instance the 

 "bottle-brush," the tough coat is much more conspicuous 

 than the living zooids. But whether the coat be well 

 developed or not, it should be noted that it is always 

 horny, and never made of lime. There are a few Hydrozoa 

 which form limy coats (corals), but these do not occur 

 round our coasts. 



Lastly, we should note the relation of the zoophyte 

 colonies to the tiny swimming-bells so abundant in our 

 seas in late summer and early autumn. We have seen that 

 these medusoids arise from zoophyte colonies, and are the 

 reproductive persons of those colonies, and we have seen 

 also that while some zoophytes give rise to medusoids, 

 others bear sessile sporosacs. In some cases, as in the 

 Sertularians and Plumularians, this latter condition prevails 

 in a whole family ; while in other cases, as in the Campanu- 

 larians, closely related forms display the two conditions. 

 There seems no doubt that the production of swimming- 

 bells is the more primitive condition, and that this power 

 has been lost by such families as the Plumularians and 

 Sertularians. Probably its loss is associated with the fact 

 that the bells are very liable to be swept away by strong 

 currents to localities quite unsuitable for the hydroid stages, 

 and that distribution by means of minute larvae is as 

 effective and much less costly than the production of 

 swimming-bells. Nevertheless, we have forms like Obelia 

 geniculata and Campanularia flexuosa, which seem to live 

 under quite similar conditions, and are both extraordinarily 

 abundant; and of these one bears sporosacs and the other 

 true medusoids. Therefore, though we have much reason 



