306 THE VOYAGE OF H.M.S. CHALLENGER. 



subfamilies with four genera : (l) the Rhizophysidse (with Epibulla and Rhizophysa), 

 and (2) the Physalidse (with Alophota and Physalia, 25, pp. 33, 36). The same two 

 groups were accepted by Huxley, in his Oceanic Hydrozoa, 1859, as two separate 

 families of Physophoridse (9, p. 71). 



Physalia, as the largest and most splendid of all Siphonophorse, well known to all 

 travellers and sailors in the Tropics, celebrated by its peculiar form and swimming 

 locomotion, its brilliant colours and dangerous poison, has provoked a voluminous 

 literature (compare Eschscholtz, 1, p. 159 ; Offers, 79, p. 26 ; and Huxley, 9, p. 93). 

 But the greatest part of it is without scientific value, full of errors, and not supported 

 by accurate researches. Huxley rightly says, that " this department of zoological 

 literature makes one long for the advent of a Caliph Omar, and produces a sort of 

 unpleasant vertigo" (9, p. 99). Indeed, the knowledge of Physalia, although 

 examined and described by numerous observers, has remained very insufficient up to 

 our time. 



I myself had an excellent opportunity of observing living Physalidse. as well as 

 Rhizophysidse, during my residence in the Canary Island Lanzerote, in December 

 1866, and in January and February 1867. Among thousands of large Physalise, which 

 appeared at Christmas 1866 in the harbour of Arrecife, there were some interesting, 

 crestless, small, new forms, which are figured in PI. XXVI. of this Report as Alophota 

 giltschiana and Arethusa challengeri. Pis. XXIII. and XXIV. reproduce the figures 

 of two new genera of Rhizophysidse (Cannophysa with ordinate cormidia, and Necto- 

 physa with loose cormidia, both sexually mature), which I had drawn from life in 

 Arrecife in January 1867. But more interesting, as types of new families, may be the 

 two remarkable forms of Cystonects which I observed in December 1881 and January 

 1882 in Ceylon, and which are figured in PI. XXII. as Cystalia monogastrica and 

 Epibulia ritteriana. 



The collection of the Challenger contains a number of Physalise collected in different 

 parts of the Tropical Atlantic and Pacific, and besides some other Cystonects or frag- 

 ments of them. Among these is a very interesting deep-sea form, from the Tropical 

 Atlantic (Station 338), Salacia polygastrica (PI. XXV.) ; it is the type of a new family, 

 intermediate between the Rhizophysidse and Physalidse. 



Some other interesting new forms of Rhizophysidse, also inhabitants of the deep 

 sea, were described in 1878 by Studer as different species of Rhizojrfiysa (40, Taf. i.) ; 

 they represent, in my opinion, two different genera, Aurojihysa inerniis and Linophysa 

 conifera. The large form described by Studer as Bathyphysa abyssorum belongs 

 probably to the Forskalidse (compare p. 248). 



Some similar deep-sea forms, described recently by Fewkes (45), are too incomplete 

 and too insufficiently known to allow us to recognise their true position in the system. 

 Recently Chun has given some valuable contributions to our knowledge of the 



