28 



WEALDEN AND PURBECK FOSSIL FISHES. 



less. Length of head with opercular apparatus exceeding three-quarters the 

 maximum depth of the trunk, and slightly less than one-quarter the total length 

 of the fish. Frontal profile steep and snout acute ; parietal bones about one third 

 as long as the frontals, which are four times as long as their maximum width, 

 narrow in front, and united by a nearly straight median suture ; two postorbital 

 plates, the lower large and much deeper than wide, and both wider than the 

 circumorbitals ; mandibular symphysis not much deepened or extended ; external 

 bones rugose and more or less tuberculated, except the maxilla and premaxilla, 

 wdiich are smooth. A single pair of supratemporal plates. Teeth on moderately 

 long pedicles, constricted below the crown, and all with a small pointed apex when 

 unworn. Opercuhun about three-fifths as wide as deep, its width contained three 

 times in the length of the head. Fin-fulcra very large, the principal dorsal fulcra 



Fio. l-i.—Lepidotus minor, Agassiz; restoration, from lialf to three-eiglitlis nat. size— Middle Purbeck 



Beds : Swanage, Dorset. 



more than half as long as the anterior dorsal fin-rays, and four to six directly 

 inserted in the ridge of the back ; pelvic fins arising midway between the pectorals 

 and the anal; dorsal fin much larger than the anal, but each with ten or eleven 

 rays. Scales smooth, usually more less serrated on the flank; scales of lateral 

 line notched ; dorsal ridge-scales acutely pointed and rather prominent. 



Descrijjtion of Sjjecimens. — This species varies considerably in the shape of the 

 trunk, and the fossils are often distorted by crushing. The degree of serration of 

 the flank-scales and the relative size of the fin-fulcra are also variable. Every 

 gradation, however, can be found between the deepened form with rounded l)ack, 

 shown in the type specimen and in our PI. VI, and the more slender type which 

 is figured by Branco {he. ciL, 1887), and represented here in Text-fig. 14. A 

 careful study of a large series of specimens also seems to show that the nature of 

 the serrations of the scales is not correlated either with the size of the fish or with 

 its shape. 



