REPORT ON THE SIPHONOPHORjE. 185 



of corms were described by Quo)' and Gaimard (2, 19, 20) and by Lesson (3, 22) ; but 

 their figures are so unnatural and incomplete, and their descriptions so superficial and 

 devoid of scientific understanding of the subject, that they have only produced extra- 

 ordinary confusion and numerous mistakes. 



Eschscholtz, the founder of the System der Acalephen (182 ( J), first established the 

 family Physophoiidae, and distinguished it from the other two families of his Siphono- 

 phorte (Diphyidae and Velellidiu) by this definition : — "The soft body bears at its upper 

 end a swimming-bladder filled with air." He distinguished (1, p. 141) ten different 

 genera ; three of these, however, belong to the Cystonectse, and one to the Calyconectse, 

 so that six remain; of these, Apohmia and Aihorybia are types of two separate famdies ; 

 Agahna and Stepkomomia belong to the Agalmidse ; Physophora and Discolahe to the 

 Discolabidie. The system founded by Eschscholtz was much extended, but not advanced, 

 by Lesson, who in his Acalephes (1843) gave a most confused compilation of all 

 descriptions published up to his time. Brandt (in 1835) founded the two famdies 

 Agalmidaj and Anthophysidse (25). The first good anatomical description of a Physonect 

 was published in 1841 by Milne-Edwards, who illustrated the Mediterranean Stephanomia 

 ( = Forshdia) contorta (71). Another excellent paper on Agalmopsis elegans was 

 written in 1846 by Sars (27, L). 



A more accurate knowledge of the peculiar organisation of the Physonectaa, and a 

 more natural explanation of their complicated structure, was not acquired before the 

 sixth decade of this century. At this time Kolliker (4), Vogt (6), Leuckart (5 and 8), 

 Gegenbaur (7 and 10), and Huxley (9) so greatly advanced our knowledge by a series 

 of excellent illustrations and accurate descriptions, that most succeeding observers have 

 only been able to add single particulars. Claus, in his monographs of Physophora 

 hydrostatica (34), Halistemma tergestinum (74), and Agalmopsis utricularia (75), 

 advanced mainly our histological knowledge of the Physonectge ; as did afterwards, 

 more especially, Korotneff (1884), but, unfortunately, without sufficient knowledge of 

 their morphological and systematic relations (50). 



My own observations on the Physonectse were commenced in 1859 in Messina, and 

 advanced much in 1866 during my residence at Lanzerote in the Canary Islands. I 

 found here, and stdl more in 1881 in Ceylon, and during my voyage in the Indian 

 Ocean, the opportunity of examining a number of interesting new forms and even new 

 types of Physonects (Circalia, Atlwria, Dicymba, Cnjstallodes, Anthemodes, Lychna- 

 galma, NectaUa, Discolahe, &c, Pis. XI.«-XXL). The Challenger collection, however, 

 contained only very few specimens of Physonectse which were preserved well enough 

 for description ; only scattered fragments and detached parts (nectophores, bracts, 

 siphons, gonophores) were found in many of the bottles. 



Kelying on these extended observations, and comparing the numerous scattered 

 descriptions and figures of former observers, I was enabled to establish the new system 



(ZOOL. CHALL. EXP. — PART LXXVII. — 1888.) Hhlih 24 



