RHODALIIDAE. 301 



Angelopsis; hence his institution of the AngeUdae to inchide the two. Schneider 

 ('98) followed Fewkes in retaining Angela as the type genus of the family; and 

 Lens and Van Riemsdijk did the same; Haeckel ('88b) referred Angela doubtfully 

 to the Rhizophysaliae, placing it in his family Epibuliidae. Lesson's account 

 and figures of Angela clearly show that instead of being allied to Angelopsis 

 or to the Rhizophysaliae, in reahty it belongs to the Anthophysidae. Lesson's 

 brief description ('43, p. 496) is as follows: — "Corps entierement vesiculeux, ou 

 vessie aerienne evasee, subarrondie, elargie a sa base, a sommet comme mame- 

 lonne, et garni de valvules claustrales; plateau convert d'un grand nombre de 

 tubes digestifs, vermicules, allonges, cylindraces, dilatables en sac ventru a la 

 base, retreci au sommet, qui a une bouche arrondie et plissee sur les cotes. Du 

 plateau partent huit tentacules fins, tres-longs, formant tube, et garnis sur leur 

 cotes de petits sugoirs alternes, courts, termines par trois glandes " and this is a 

 very good account of the general appearance of .\nthophysa. Comparing 

 the description with the figures (Lesson, '43, pi. 9, figs. 1-le) drawn from life 

 by Rang, it is evident that the "tubes digestifs" are the palpons; the appear- 

 ance of a nipple ("mamelonne") at the apex of the "vessi aerienne" is the zone 

 of proliferation of the bracts. Although Rang did not discriminate between 

 siphons and palpons, his discovery of eight tentacles in a single row agrees with 

 arrangement of the cormidia in Anthophysa, while his determination of tricornu- 

 ate tenillae speaks highly for his powers of observation. In short, so close is 

 the correspondence between Angela and Anthophysa in all respects, that I have 

 no hesitation in placing it in that genus (p. 295). As a result, according to the 

 International rules of nomenclature (Art. 4), it is necessary to abandon the 

 family name Angelidae, at least in the sense in which it was used by Schneider 

 and by Lens and Van Riemsdijk. To replace it the name Rhodaliidae, pro- 

 posed in 1888 by Haeckel, must be used. Chun ('97b) has used Auronectae 

 and Auronectidae in a family sense but these are untenable as they are not 

 derived from a generic name. 



In reducing the group from the rank of an order to that of a family, I follow 

 all recent students of Siphonophores. Thus Claus ('89), Chun ('97b), Schneider 

 ('98), and Lens and Van Riemsdijk (:08) have all clearly shown that instead 

 of being worthy of ordinal rank on account of the presence of a supposedly unique 

 organ, the "aurophore," the animals in question are in reality closely allied to 

 the Physophorae in general. To illustrate the ease with which original descrip- 

 tions are accepted, and the difficulty of bringing criticisms of them before zoolo- 

 gists, Lens and Van Riemsdijk have pointed out that the order "Auronectae" 



