A TAXONOMIC STUDY ON PYROSOMA — METCALF AND HOPKINS. 237 

 PYROSOMA ATLANTICUM (Peron, 1804). 



Plates 15, 34, and 35; plates 26-32; plate 36, figs. 50-53. 



There is a large group of closely related forms which we are in- 

 cluding in this species, classing as subspecies a number of forms 

 previously described as species, and also several new types in our 

 collections. To this group belong "Pyrosoma atlaniicum" (Peron, 

 1804), "P. giganteum" (Lesueur, 1815), "P. elegans" (Lesueur, 1815), 

 "Dipleurosomaellipticum" (Brooks, 1906), "P. triangulum" (Neumann, 

 1909, a), and our new forms liawaiiense, paradoxum, and echinatum. 



It is not possible to tell to which of our several subspecies belong 

 the forms described by Peron and Lesueur. Those colonies of this 

 group, collected in the Atlantic Ocean, which have been described in 

 sufficient detail to allow exact identification, are all of the sort which 

 we treat as the subspecies giganteum, and, from the large size often 

 attained, they seem to be of the type which has long been called 

 P. giganteum. The colonies which we take as the type and paratypes 

 of the species atlanticum (subspecies atlanticum) are from the Pacific 

 Ocean. They are distinct from the Atlantic forms generally classed 

 as giganteum. They are probabl} 7 the same as Huxley's southern 

 Pacific specimens which he called atlanticum and are surely the same 

 as Ritter's San Diego material, named by him giganteum. 



Somewhat intermediate between these atlanticum forms and the 

 true giganteum are a number of colonies which we do not describe 

 in detail, though we refer to them later under the name intermedium. 

 Here would be placed the forms discussed by Seeliger and Neumann 

 as P. atlanticum and its varieties Isevatum and tuberculosum. Peron's 

 P. atlanticum probably should be placed here. Pyrosoma minimum 

 we do not discuss. 



The problem of nomenclature is intricate and confused. There are 

 no descriptions of "atlanticum," not even Huxley's, which are clear 

 enough for certain identification. The earliest description of colonies 

 in this whole atlanticum group, which allow fairly confident identi- 

 fication, apply to large Atlantic specimens and are under the name 

 giganteum. Under these circumstances it might seem natural to 

 use the name giganteum for the species, holding the other forms as 

 subspecies, though this would involve redefinition of giganteum, 

 but, according to taxonomic conventions, we must retain the name 

 first applied to a member of this major species, though it is not pos- 

 sible to tell to which of the several subspecies the individuals first 

 described belonged. 



In 1804, Peron described the first colony of these animals dis- 

 covered, giving it the name Pyrosoma atlanticum. From his descrip- 

 tion, and the locality of its collection, we can say only that it is a 

 Pyrosoma, probably of our larger atlanticum group. The fuller 



