80 DE. W. G. RIDEWOOD ON THE CRANIAL [May 3, 



present they may be fused to the bone, or may strip off with the 

 mucous membrane, and leave no scar upon the surface of the 

 bone. The matter is of some little importance because the mucous 

 membrane is customarily allowed to dry on the phai^yngeal 

 skeleton to prevent the constituent parts from separating, and 

 loose teeth thus appear as if they were rigidly attached. The 

 dentigerous plates on the fifth ceratobranchial are readily removable 

 in Elops and Megalojjs, which is rather remarkable, for the teeth 

 of the fifth ceratobranchial are usually firmly fixed to the bone 

 even in those cases in which there is considerable reduction in the 

 hyobranchial dentition as a whole. It is worthy of remark in 

 connection with the probable origin of the ectosteal constituent of 

 the glossohyal by the coalescence of tooth-bases (c/. vomer, &c., 

 p. 63), that in the Herring the glossohyal is a cartilage overlaid 

 by the ectosteal lamina ; and the teeth, although they may leave 

 scars on the bone when the mucous membrane is stripped off, are 

 not intimately attached to the bone. This evidently indicates a 

 process of degeneration in the lingual dentition, the first stages in 

 the transition to an edentulous state being marked by a I'eduction 

 of the basal parts of the teeth. 



The dentigerous plates lying on the pharyngeal surface of the 

 second, third, and fourth pharyngobranchials are not collected 

 together to foi-m an epipharyngeal apparatus as, for instance, in 

 the Cod, but they remain distinct. The cartilage of the fourth 

 phaiyngobranchial remains unossified, but in some forms, such as 

 Albula and Chirocentrus, the cartilage in drying shrinks upon the 

 underlying dentigerous plate in such a way as to give the 

 impression that itself is ossified. In Chatoessus the ectosteal bone, 

 here toothless, spreads over the mesial and dorsal sui'faces of the 

 cartilage, and gives the effect of a completely ossified fourth 

 pharyngobranchial. Swinnerton, it may be mentioned, has re- 

 corded the presence of a truly ossified fourth pharyngobranchial 

 in Cromeria (Zool. Jahrb., Abth. f. Anat. xviii. 1903, p. 65, 

 figs. H & K). 



Engraulis, Gonorhynchus, and Alepocephalus have a distinct 

 fifth epibranchial cai-tilage. The presence of this element in 

 Alepocephalus was pointed out by Gegenbaur, who has also figured 

 one in Clupea alosa (Morph. Jahrb. iv. Suppl. p. 24, and pi. 2. 

 fig. 13, Alosa vulgaris or Clupea vzdgaris). In Clupea Jiarengus 

 and C. finta the position of the fifth epibranchial is occupied by a 

 ligament ; in most genera there is no repi'esentative of this element 

 of the visceral skeleton. 



The epibranchial accessoiy organ of respiration produces, in 

 those Malacopterygians in which it is at all largely developed, 

 important modifications in the shape and size of the elements of 

 the fourth and fifth branchial arches. In Chatoessus ^ and to a 

 lesser extent in Chanos, the fifth ceratobrancliial and the fourth 

 epibranchial are considei-ably increased in width, but in Gono- 

 rhynchus these bones are only lengthened. For our knowledge of 

 the structure of the epibranchial organ of Clupeoid fishes we are 



