BARNACLES FROM DEEP-SEA CABLES. 283 



It is strange to find a barnacle such as Heteralepas gigas 



on the sea-bottom in comparatively deep water, for it has all the 

 appearance of a pelagic form. The specimens in the Raffles Mu- 

 seum, however, were undoubtedly attached to a telegraph cable, as 

 is proved by an examination of the material adhering to their bases. 

 They have been preserved in alcohol for many years and are now 

 dull and formless objects, shrivelled out of all resemblance to their 

 natural form, but much less uniformly contracted than the type- 

 specimen. In life they were in all probability even larger than 

 they are now and their integument must have been smooth, trans- 

 parent and swollen ; they must have borne a close external resem- 

 blance to the true Ale pa*, which is found on the surface, as a rule 

 depending from the bells of medusae (see Bee. Ind. 31 us., X, p. 2?6, 

 pi. xxxiii, fig. 2 ) . 



Both Scalp ellum stearnsi and 8. persona, which are not closely 

 allied species, display a tendency to get rid of the calcareous valves 

 or plates on the capitulum and to substitute for them a homogeneous 

 cartilaginous or thick membraneous investment. In all families of 

 the true Cirripedia Pedunculata a similar tendency occurs and re- 

 occurs in certain genera and species. It may be correlated either 

 with a deep-sea or a pelagic existence or with semi-parasitic habits 

 and therefore affords a rare instance of parallel evolution in which 

 convergence is connected not with similar but with diverse modes of 

 life. In the genus Scalpellum itself we find two if not three dif- 

 ferent manifestations of this curious tendency, which may perhaps 

 be regarded as an ultimate reversion to a primitive condition. In 

 the first place we may note a number of species of comparatively 

 small size (e.g. S. laccadivicum , Annandale = 8. polymorphum, 

 Hoek, and 8. larvale, Pilsbry) with compressed capitula and very 

 delicate valves in which there is great variation in the development 

 of these plates. Even when they are most degenerate the mem- 

 brane that covers them is not thick. In some cases individuals are 

 known in which the greater part or the whole of the capitular sur- 

 face is covered by the valves, while others occur in which the valves 

 are reduced by an excavation of their lower margins until (as in 8. 

 lambda, Annandale) they may all have a form approaching to that 

 of the Greek letter /. In such forms the valves of the young are, 

 at any rate in some species, more complete than those of the adults. 

 At the other extreme we find a little group of very large species 

 such as 8. giqanteum, 8. persona and 8. alcockianum, with 

 stout, more or less inflated capitula and with relatively thick valves 

 almost completely concealed beneath a thick cartilaginous invest- 

 ment. In such species the condition of the valves seems to be 

 much more stable than in the other group; in 8. giganteum they 

 are relatively large, in the two Oriental species very small. But 

 in these latter their area though small is not, except in the 

 terga and to a less extent in the carinal latera, reduced by ex- 

 cavation of the margins of the plates, but by a general reduction in 

 size. Such forms as S. stearnsi and 5. gnivelii are to some extent 



E. A. Soc, No. 74. 1916. _ 



