Chap. IT. CTEXOPIIOR.E SACCAT^E. 197 



distingiiii^liing tlic Mertcnsida) from the Cydippidns proper. As we have scon, the 

 spheromeres of tlic Cydippida) are ,«o uniform in their size, structure, and radiated 

 arrangement, that their confonnity to the type of the Ctenophorse, in their hilateral 

 disposition, is harely perceptihle : in the Mertensida?, on the contrary, the anterior 

 and posterior splieromeres are very different in their form and size, and even in 

 their structure, from the lateral ones, so that the bilateral symmetry of these Cte- 

 nophorfe is one of their most characteristic family distinctions. But this sjinmetry 

 is so peculiar, that, far from exhibiting a transition to the Cestida? and Calymnidaa 

 or Mnemiidce, as Gegenbaur believes, it presents the most striking contrast with 

 them. In the Cestidte, as well as in the Calymnidfe and Beroids proper, the com- 

 pression is lateral; that is to say, the lateral spheromeres are reduced in comparison 

 to the great development of the anterior and posterior pairs, or, in other words, 

 the coeliac diameter is the longest, and the diacoehac the shortest, of the two 

 transverse diameters : while in the Mertensidaj the compression is diametrically 

 opposite, the lateral spheromeres being greatly developed, Avhile the anterior and 

 posterior pairs are much reduced, so that here the coeliac diameter is much shorter 

 than the diacoeliac. Mortens, in his description of Beroe compressa, has already 

 alluded to this difference between the two types, and also noticed another structural 

 peculiarity coinciding Avith the antero-posterior compression in the form of the 

 circumscribed area, which is short and petaloid, while in the Cydippidse it is long, 

 with parallel sides. In conformity with this inequality of the spheromeres and 

 the prominence of the anterior and posterior pairs, the abactmal ,«ide of the larger 

 spheromeres projects more or less, thus giving a heart-shaped outline to the abactinal 

 pole, on which the circumscribed area is seen in a depression of the short coehac 

 diameter. A corresponding inequality is observed in the development of the vertical 

 ch}-miferous tubes and their rows of locomotive flappers. In the Cydippidae proper, 

 there is hardly a perceptible difference between them : in the Mertensidae, on the 

 contrary, the lateral pairs are much longer than the anterior and the posterior 

 pair. This again might easily be mistaken for a resemblance between the Mer- 

 tensidse and the Mnemiidse or CalymnidiP, were not the most developed chymiferous 

 tubes and rows of locomotive flappers m the latter those of the anterior and 

 posterior spheromeres, while in the Mertensidaj the most exten.sive chymiferous 

 tubes and rows of locomotive flappers are those of the lateral spheromere.s. Tlie 

 extraordinary development of the parts arranged in the direction of the diacoeliac 

 diameter constitutes, in fact, the most striking family character of the Mertensidae, 

 and determines its peculiar form. In accordance with this preponderance of the 

 sides, it is worthy of remark that the whole tentacular apparatus of the repre- 

 sentatives of this family is larger and more complicated than that of any other 

 group of Ctenophorte, in comparison to the absolute size of these animals. Mertens 



