Bull. not. Hist. Mus. Land. (Zool.) 65(1): 51-72 



Issued 24 June 1999 



A review of the genus Bargmannia Totton, 1954 

 (Siphonophorae, Physonecta, Pyrostephidae) 



p.r. pugh ky (^\ 8t? n . , V) 



Southampton Oceanography Centre, Empress Dock, Southampton, Hants, S014 3ZH, UK 



CONTENTS 



Introduction 00 



Family Pyrostephidae Moser, 1925 00 



Genus Bargmannia Totton, 1954 00 



Bargmannia elongata totton. 1954 00 



Bargmannia amoena sp. nov 00 



Bargmannia lata Mapstone, 1998 00 



Bargmannia gigas sp. nov 00 



Discussion 00 



Acknowledgements 00 



References 00 



SYNOPSIS. Two new species of the physonect siphonophore genus Bargmannia are described, and B. elongata Totton (1954) 

 and B. lata (Mapstone, 1998) are redescribed. The status of the genus and its retention within the family Pyrostephidae are 

 discussed. 



INTRODUCTION 



Totton (1954) established the genus Bargmannia, named after his 

 colleague Dr Helene Bargmann, to include the single species, B. 

 elongata; nectophores of which he had found in thirteen Discovery 

 samples, plus one from the Michael Sars Expedition (Leloup, 1955). 

 Because the structure of the nectophores differed so markedly from 

 those of all other known physonect siphonophores, Totton did not 

 give a detailed description of them; remarking only that the lateral 

 radial canals on the nectosac had straight courses. Totton (1965) 

 later noted that, although B. elongata was one of the most easily 

 recognised siphonophore species, nothing more had been published 

 on it since its original description. In fact, by the time of publication 

 of Totton's monograph, only Alvarino (1963, 1964) had mentioned 

 it; and then only in lists of siphonophore species collected in the 

 western Pacific. Totton included a brief description of a further 

 specimen collected at Discovery St 4246 (37°50'N, 13°22'W). re- 

 marking on the orange coloration of the stem. 



Since that time several authors have reportedly identified this 

 species from various collections. However, examination of both 

 Totton's material and that from more recent Discovery collections 

 (Mackie, Pugh & Purcell, 1987) appeared to indicate that Totton's 

 (1954, 1965) illustrations of Bargmannia elongata could be 

 referred to two species, and that his material also included a third 

 species. However, it was not until submersibles collected speci- 

 mens of this genus that this contention could be proved beyond 

 doubt. Study of this submersible material, together with that from 

 the Discovery collections, shows that there are at least four spe- 

 cies that may be referred to the genus Bargmannia. The second 

 species that Totton illustrated under the name B. elongata has 

 recently been described under the name B. lata. More detailed 

 descriptions of both these species, together with descriptions of 



two previously undescribed species, are given herein. 



Totton (1954) did not refer the genus Bargmannia to any of the 

 physonect families, although his description appears at the end of a 

 section dealing with various species of the family Agalmatidae. 

 Later, Totton (1965) placed the genus in the family Pyrostephidae, 

 which previously had been monotypic for the species Pyrostephos 

 vanhoeffeni Moser, 1925. However, his diagnosis of that family 

 applied only to the genus Pyrostephos, and included such features as 

 marked bends in the dorsal and lateral radial canals on the nectosac 

 of the nectophores. This character alone would exclude the genus 

 Bargmannia. Since then, Stepanjants (1967) placed the genus in the 

 catch-all family Agalmatidae, whereas Daniel (1974) retained it 

 within the family Pyrostephidae. Now that intact specimens have 

 been collected by submersibles it is possible to review the systematic 

 position of the genus Bargmannia. It is concluded that, for the 

 present at least, it should be retained within the family Pyrostephidae, 

 the diagnosis of which is adjusted accordingly. 



Family PYROSTEPHIDAE Moser, 1925 



DIAGNOSIS. Long-stemmed physonect siphonophores. Necto- 

 phores with large triangular thrust block; with lateral wedge-shaped 

 processes reduced or absent. With apico-, infra- and vertical (meso-) 

 lateral ridges; apico-laterals divide above ostial level. Adaxial wall 

 of nectosac lacking musculature; deeply hollowed. Long pallial 

 canal; short pedicular canal, giving rise, on nectosac, to only dorsal 

 and ventral radial canals; lateral radial canals arise separately from 

 dorsal. Dorsal and lateral radial canals either looped or straight. 

 Tentillum with straight (or twisted, but not tightly coiled) cnidoband; 

 lacking an involucrum; with terminal filament. Dactylozooids either 

 absent or modified to form peculiar palpacle-less oleocysts. Indi- 

 vidual specimens of single sex (dioecious), with gonophores budded 



) The Natural History Museum. 1999 



