38 PROCEEDINGS OF THE 



sharks and skaies, aiul some of the fossils are essentially identical 

 with forms living at the jjresent day. The Cestraciontidjo, 

 Notidaniila^, Scylliida), fS(|naiini(lii', and Khinohatida), of which 

 tine skeletons are known from the J^ithographic >Stone ot (Ger- 

 many, were certainly already diiferentiated in Upper Jurassic 

 times. The iSj)iiiacidte were also neaily in existence, one tish 

 having been found which is only distinguished from the 8i)inacid8e 

 by retaining a small anal lin *. 



The typically modern sharks of the family Lamnidaj did not 

 appear until the Cretnceous period, and were not abundant until 

 the time when the Chalk tormation was deposited. The Car- 

 chariida3 arose even later, and their remains are not known from 

 rocks earlier than the Tertiary. The great skates also began to 

 be ditferentiated towards the end of tlie Cretaceous period, and by 

 the beginning of the Tertiary tliey were almost the same as in 

 existing seas, only with a different geographical distribution. 



Of progressive specialisations of certain skeletal structures 

 which can he followed among the fossils, perhaps the most 

 interesting is tlie evolution of the rostrum in the saw-fishes or 

 Pristida). We have already seen that Ehinobatidfe occur in the 

 Upper Jurassic. J{hl)iobatus-[ike fishes in the Upper Cretaceous 

 of the Lebanon, begin to have the rostrum elongated and 

 strengthened, and fringed with a row of sharp enamelled teeth 

 loosely fixed along each lateral edge t- Onchopristis X, f^'om the 

 Upper Cretaceous of Egypt, and Oncliosaurus §, from many 

 Cretaceous formations, have these rostral teeth larger and barbed 

 at the apex, with a tendency to reduction of the enamel. Pro- 

 pristis |(, from the Lower Eocene of Africa, has a stdl longer 

 stiffened rostrum, with the lateral teeth peg-shaped, loosely fixed 

 in very shallow sockets, and with scarcely any trace of enamel. 

 The modern Pristis, which ranges from the Middle Eocene 

 op wards, has the peg-shaped rostral teeth sunk in deep sockets of 

 cartilage. In tlie early stages of the development of the rostral 

 weapon there was thus much variety ; in the latest stage there is 

 only one form. 



This tolerably complete history of the Pristid rostrum, which 

 has only lately been discovered, is an illustration of the help 

 afforded by the fossilised extinct Ela^^mobranchs in undei'standing 

 the isolated survivors of the Subclass at the ])i'eseut day. Nearly 

 every addition to our knowledge of the fossils during recent 

 years, indeed, has helped to lill gaps in the series ; and researches 



* Protosjjindx annectens, A. S. Woodward, Proc. Zool. Soc. 1918 (1919), 

 p. 233, pi. 1. figs. 2, 3. 



t SclcrorJii/nchus atavun, A. S. Wooclward, Catal. Foss. Fislies Brit. Mus. 

 pt. i. (1889). p. 7(>, pi. 3. fig. J, and Proc. Zool. Soc. 1889, p. 449. See also 

 O. P. Hay, Bull. Aim-r. Mus. >!it. lli>>f. vol. xix. (1903), p. 398. 



X F. Siromer, Al)liaridl. k. bay. Akad. Wiss., uiatb.-pliys. CI. vol. xxviii. 

 Abh. 8 (1917), p. 12, pi. 1. 



§ E. Stromer, /oc. cit. (1917). p. 11. 



II E. Fraas, Neues Jahrb. fur Min., etc., 1907, vol. i. p. 1, pi. 1. 



