SUMMARY AND CONCLUSION. 255 



and in Brazil, 1 some of them had already acquired the row of sharp ventral ridge- 

 scales which are so peculiar a feature of the surviving Ghvpea and allied genera. 

 These fishes have not hitherto been found in the English Chalk, but individuals of 

 at least one species are extremely abundant in the Upper Senonian of the Lebanon. 

 An extinct family, that of the Ctenothrissidse, confined to Upper Cretaceous 

 formations, is precocious in another respect. The two known genera exhibit the 

 pelvic fins so far forwards that their supports are in contact with the pectoral 

 arch — a specialisation which is otherwise met with only in the highest fishes of 

 the Acanthopterygian groups. 



With these specialised Clupeoids there are others of a more generalised grade, 

 beginning with Crossognathus from the Neocomian of Switzerland 2 and Hanover, 3 

 and represented in the English Chalk by Si/llasmus. There are also, as might be 

 expected, numerous Chirocentrids and Elopines, the former best known from the 

 North American Chalk (see p. 99), the latter from the English Chalk and the 

 Upper Cretaceous of the Lebanon and Brazil, 1 but both occurring also in 

 Australia. 5 They, again, are very little higher in type than some of the highest 

 Jurassic fishes ; but very few Lower Cretaceous forms have been discovered, and 

 until these are better known, possible origins can scarcely be discussed. With the 

 Elopines in the Upper Cretaceous, there are specialised offshoots more or less 

 closely related to the Albulidag and Osteoglossidae. These are represented only by 

 fragments in the English Chalk, but the Albulid Istieus is known by numerous 

 complete fishes from the Upper Senonian of Westphalia, while Anogmius is well 

 preserved in the Chalk of Kansas, U.S.A. (see p. 106). 



Istieus is essentially identical with the surviving deep-sea genus Bathythrissa ; and 

 it is interesting to add that the imperfectly known Tomognathus from the English 

 Chalk (see p. lo9) also bears much resemblance to some of the low-grade teleosteans 

 which are now confined to ocean depths. In fact, the more the Cretaceous 

 bony fishes are examined, the more evident are their relationships to members 

 of the existing abyssal fauna, rather than to fishes surviving in shallower depths. 



Of the Cretaceous "ganoids," all the families become extinct before the 



1 Diplomystus longicostatus, E. D. Cope, Proc. Auier. Phil. Soc, vol. xxiii (1886), p. 3 ; A. S. 

 Woodward, Aim. Mag. Nat. Hist. [6], vol. xv (1895), p. 2, pi. i, fig. 1. 



2 F. J Pictet, oj). cit. (1858), p. 18, pi. ii, and pi. iv, figs. 1—6. 



3 A. S. Woodward, Catal. Poss. Fishes Brit. Mus., pt, iv (1901), p. 350. 



4 A. S. Woodward, " On the Fossil Teleostean Genus Ehacolejns, Agass.," Proc. Zool. Soc, 1887, 

 pp. 535—541, pis. xlvi, xlvii ; also Catal. Fuss. Fishes Brit, Mus., pt. iv (1901), pp. 27— 32.— D. S. 

 Jordan and J. C. Branner, " The Cretaceous Fishes of Ceani, Brazil," Smithson. Miscell. Coll., vol. Hi 

 (1908), pp. 15-25, pis. iii— vii. 



5 Icldhyodectea mavuthonensis, B. Etheridge, jun., Rec. Australian Mus., vol. vi (1905), pp. 5 — 8, 

 pis. i, ii. — Portheus australis, A. S. Woodward, Ann. Mag. Nat. Hist, [6], vol. xiv(1894),p. 444, pi. x, 

 fig. 1. 



6 W. von der Marck, Palaeontographica, vol. xi (1863), pp.37 — 40, pis. iv, v; and foe. cit., vol. xxii 

 (1873), p. 59, pi. ii, fig. 2. 



