Chap. I. SUB-ORDERS OF CTENOPHORiE. 175 



but always tentacles, ihe CesHdcc and Ci/dippidce ; and 3°, those without loljes and 

 without tentacles, the Bcroidic. 



Here, as everywhere in our science, the total absence of a principle in assigning 

 a rank to the divisions adopted hy different authors is painfully felt. While 

 Eschscholtz considers the Beroids as an order, to which he first applied the name 

 of Ctenophor.e, Lesson considers the whole group simply as a family, with which 

 he unites most heterogeneous animals, belonging to several distinct classes ; and 

 Leuckart, whom Cregenbaur has followed in that respect, regards them as a distinct 

 class. The three families distinguished by Eschscholtz coincide very closely with 

 the three more comprehensive groups into which Gegenbaur arranges the five 

 families which he admits. Mertens, again, se^ijarates Cestum from Cydippe, as a 

 distinct famil}'', unfortunately retainmg the name Beroe for the family of the latter 

 genus, and applying that of Idyia to the family called Beroideaj by Eschscholtz 

 and that of Callianira to the family called Mnemiidae by Eschscholtz, who had 

 already used the name Callianiridre for the family to which he refers the genera 

 Cestum, Cydippe, and Callianira. It is in view of this confusion, probably, that 

 Gegenbaur has again changed the family name of Mnemiidse Esch. to Calymnida? 

 Gcgciib. ; Init in so doing he has only made the matter worse, since his fiimily 

 CalymnidfB again differs from the ti'ibe called Cal3'mmea3 by Lesson. This affords 

 another evidence of the absolute necessity of strictly adhering to the law of priority. 

 The family names first proposed by Eschscholtz cannot be discarded so long as 

 there remains a natural group of Ctenophora^ to which they can lie applied. We 

 shall see presently what is the value of the eight tribes distinguished by Lesson. 

 The first question that should now engage our attention is, whether the families 

 adopted by Eschscholtz and Gegenbaur bear family characters or not ; since I have 

 already shoAvn that the Ctenophorte, as a natural group, are neither a class, as 

 Leuckart and Gegenbaur admit, nor a flxmily, as Lesson would have it, but a 

 natural order of the class of Acalephs. 



Eschscholtz assigns the following characters to his three families of Ctenophorje : 

 CaUkmlridw, with small digestive cavity and tentacles ; Mncmiida;, with small digestive 

 cavity, liut Avithout tentacles ; Bcroidw, with large cavity of the ])ody acting as 

 digestive cavity. Gegenbaur characterizes his families as follows : CidUanirida', with 

 lateral, wing-like ajipendages supporting the locomotive flapj^ers ; Cahjmnida', with 

 two lobe-like appendages upon the sides of the mouth ; Cestidw, body riband-like, 

 expanding transversely ; CydipitidcB, body oval or rounded ; Bcroidcv, Ijody ovally 

 elongated. The characters assigned by Eschscholtz to his families are entirely 

 derived from their structure, without reference to form : Gegenbaur, on the con- 

 trary, distinguishes some of his families by anatomical characters, others Ijy their 

 form. Which of these two methods is correct ? for I mean, at present, only 



