204 W. K. BROOKS. 



I for one cauuot believe that hydroid cormi acquired locomotor 

 medusae lo distrUmte the species, and that then these medusae lost 

 their locomotor habit, and the medusoid structure which they had 

 slowly acquired, and that they sank into sessil, degenerated, bud- 

 like gonophores. 



If this degeneration were rare and exceptional, or if it had taken 

 place only once or twice, or if it were an adaptation to any modern 

 change in the ocean it would not be so remarkable, but in reality it 

 it is one of the most noteworthy peculiarities of hydroids, and it has 

 taken place over and over again. In every group of hydroids, in 

 the tubularians, in the campanularians, in the hydrocorals and in the 

 siphonophores, there are species or genera or larger groups with de- 

 generated gonophores, and there is ample evidence that this has not 

 been inherited from a common source, but that it has been acquired 

 again and again. Hydractinia, with its sessil degenerated gono- 

 phores is clearly related to podocoryue with its free medusa (dys- 

 morphosa). Tubularia with medusa buds is closely related to 

 corymorpha with its free medusa (steenstrupia). Laomedsea stands 

 in the same relation to obelia ; bougainvillia to heterocordyla and 

 so on, and by far the best ^example in the whole animal kingdom 

 of independent modification along parallel lines, is to be found in 

 the degeneration of the sexual locomotor raedusEe of compound 

 hydroids. Among the most widely distributed genera and species 

 are some of those in which this degeneracy is most complete, such 

 as the species of eudendrium, hydractinia, laomedsea, cordylophora, 

 and hydra, and I think we may dismiss the idea that the medusa 

 was originally acquired for the purpose of distributing the species, 

 and with it the idea that it has been produced by division of labor, 

 which is disproved by the following facts. 



The four great groups of craspedota, the narcomedusse, the tracho- 

 medusse, the anthomedusse and the leptomedusre agree with each 

 other in the possession of a gelatinous bell, a muscular sub-umbrella, 

 and velum, and in the presence of tentacles and sense-organs on the 

 bell margin, and wc are therefore justified in assuming that these 

 common structural characteristics were acquired before these groups 

 diverged from each other; that they are older than the modern 

 genera and families, and older than the differentiation of the hydroids 

 into tubularians and campanularians. 



