BULLETIN OF THE UNITED STATES FISH COMMISSION. 193 



rior vertebrfB modified; the diapophyses much expanded. Inferior and 

 superior branchihyals wanting- or unossifled. Branchial processes in 

 tufts." To this the following- may be added as complementary and as 

 serving- to extend the diagnosis: Opercle a simple plate; mouth tooth- 

 less; opercular membrane iJersistentJy roofing over the gill-cliamhers of the 

 embryos. 



The term Lophobranch, it appears to me, is liable to lead to misap- 

 prehension, as the gills are in reality not tufted at all, but can be referred 

 to the ordinary pinnate type commonly found in a great many fishes. 

 Yarrell* observes in a foot-note : "The tufted filamentous gills of the 

 Lophobranchs are compared by Milne-Edwards to the filamentous 

 branchiie of a tadpole ; and Eathke, who has investigated their struct- 

 ure, informs us that each is framed of a short, delicate, ligamentous 

 stem, to which the respiratory processes are attached by repeated doub- 

 lings of the branchial membrane, the folds widening as they recede from 

 the base, so as to form an inverted cone or club-shaped tuft." Fila- 

 mentous tufts do not exist in the gills of Rippocamjms, so that the first 

 part of the foregoing quotation is erroneous ; the latter jiart of it, attrib- 

 uted to Eathke, is correct, except the last word, and in this the German 

 anatomist was possibly misunderstood. 



The true state of the case is as follows : there is a median stalk or rachis 

 to which the branchial leaflets are attached in a j)innate manner on each 

 side. The leaflets become larger as you go outwards, so that the pyra- 

 midal form of the compound branchial leaflets results ; this pyramid is 

 fixed by its apex to the outer side of the branchial arch. These inverted 

 Ijyramidal branchial structures are disposed in two series, usually four- 

 sided. There is therefore nothing at all in these structures which is not 

 represented homologically in the fish's gill of the ordinary type, since 

 the two series of vascular branchial appendages to each arch in Hippo- 

 campus are perfectly comparable with the bifurcated vascular branchial 

 appendages of such a form as Salmo. There is plain evidence that a pro- 

 cess of degeneration has taken place in the branchial apparatus of Sip- 

 pocampus ; the arches themselves have undergone reduction in length ; 

 the mesobranchial bony elements are reduced or aborted, and the num- 

 ber of vascular appendages is reduced very much below what is usual; 

 the greatest number of pinnate vascular branchial appendages ranged 

 in one row on the j)osterior margin of one arch of Hippocampus is about 

 ten, which is exceeded several times by the number found in Salmo, or 

 in many other common genera. The reduction in number of these ap- 

 pendages may have called for the extension of the area of the ultimate 

 branchial lamellae or pinnse, which is a marked feature in the gills of 

 the sea-horse. In other forms, as in Brevoortia, the ultimate vascular 

 I)inn8e on either side of the gill filaments, which are the active agents in 

 respiration, being richly supplied with capillary vessels, are very feebly 



* Britisli Fishes, II, 395. 

 Bull. U. S. F. C, 81 13 



