Discussion

B. paragaster can be distinguished from B. infundibulum (L. Agassiz) Mayer 1912 the type species, by its longer oral lobes, shorter comb rows, less deeply sunken sense organ and by the simple windings of the subventral meridional canals in the oral lobes; from B. vitrea (L, Agassiz) Mayer 1912 by possessing subventral comb rows only slightly longer than the subtentacular rows, whereas B. vitrea has the subventral rows about twice as long as the subtentacular; from B. elegans Mertens 1833 by the lack of body papillae, and from B. mikado Moser 1907. 1908 by the very much shorter subventral comb rows which are less than half the length of those displayed by B. mikado, which extend from the apical pole to nearly the lower border of the oral lobe. In B. paragaster the subventral rows do not extend beyond the level of the mouth. Further, the sense organ of B. mikado is far more deeply sunken than in B. paragaster. B. paragaster shows some similarity to B. ovalis Bigelow 1904 which has proportionally longer lobes than most other species of the genus. Bigelow (1912) considers, however, that B. oralis and B. hydactina Chun 1880 are varieties of B. vitrea so that the same character that separates B. vitrea from B. paragaster, namely the difference in length of the subventral comb rows, will also distinguish B. ovalis and B. hydactina from B. paragaster. B. chuni described by von Lendenfeld (1884) from South Australian waters can be distinguished from B. paragaster by the extraordinary thickness of its oral lobes, and by the fact that the paragastric canals lie at some distance from the stomach. In addition to the characters outlined above B. paragaster possesses several unique features that distinguish it from all the known species of the genus, viz.: double tentacle bases; branched paragastric canals that join the meridional subventral canals on the lower margin of the oral lobe and a short densely pigmented stomodeum and relatively long infundibulum. Unfortunately, the juvenile specimens do not throw any light on the manner in which the branched paragastric canals arise, as in both specimens no oral lobes are present and the normal non-branched condition of the canals is displayed. Double tentacle bases however seem to be characteristic of the species in the juvenile stage of development, and should assist in making possible early recognition of the species.