304 DB. J. W. GKEGOKY ON THE PALAEONTOLOGY [Aug. 1 895, 



fishes, mollusca, corals, and echinoids that there has heen no water- 

 way across the isthmus in Pleistocene times, for this must have 

 been closed prior to the development of the existing species. No 

 doubt the physical conditions of the two seas are somewhat different, 

 and hence variation may have gone on at an unusually rapid rate. 

 Nevertheless, the connexion between the two seas cannot have 

 lasted after the close of the Miocene. The two Agassiz, father 

 and son, have proposed to date it much farther back than this. 

 They referred the separation of the two oceans and the formation of 

 the Gulf Stream to the period at the close of the Cretaceous. 1 As 

 Prof. A. Agassiz stated it in 1883, ' soon after the end of the 

 Cretaceous period, the separation of the great Atlantic and Indo- 

 Pacific marine realms began.' 2 This is, however, probably con- 

 siderably too early, as that author admits in a letter published by 

 Jukes-Browne and Harrison. 3 As we have already seen, the strati- 

 graphical evidence advanced by Gabb and Belt renders a Miocene 

 submergence highly probable, although the operations for the 

 Panama Canal did not lead to the discovery of any Kainozoic 

 deposits on the summit of the pass chosen for it. According to 

 Dr. Dall — whose recently expressed opinion on this subject 4 carries 

 the greatest weight — the Kainozoic rocks of the higher levels are 

 all either Miocene or Eocene, while the summit of the ridge is 

 formed of ' Azoic ' rocks of uncertain age. 



The Connexion of North and South America. — The comparison of 

 the faunas, though conclusively proving the former connexion and 

 long-distant separation of the two oceans, does not, however, enable 

 us to fix the date of these events with much precision : for we do not 

 know the rate at which the faunas have varied. In order to restrict 

 the date within narrower limits we must turn to the migrations of 

 the land-fauna, and especially to the evidence of the fossil mammalia 

 of the Western States. This has recently been considered by Prof. 

 W. B. Scott, 5 who points out the occurrence of a glyptodont (Oaryo- 

 derma) of a South American type in the Loup Pork Beds. These 

 must accordingly be later than the connexion between the two 

 Americas. Prof. Scott therefore admits that, if the view of Dall 

 and Harris that this was not established till the end of the Miocene 



1 Louis Agassiz, Eeport upon Deep-Sea Dredgings in the Gulf Stream 

 during the third Cruise of the U.S. Steamer 'Bibb,' Bull. Mus. Oomp. Zool. 

 Harvard, vol. i. (1869) p. 377; Alex. Agassiz, 'Preliminary Eeport on the 

 Echini and Starfishes dredged in deep water between Cuba and the Florida 

 Beef, by L. F. de Pourtales,' Bull. Mus. Comp. Zool. Harvard, vol. i. (1869) 

 p. 305 ; and ' Three Cruises of the 'Blake,' ' ibid. vol. xiv. 1888, p. 116. 



2 Alexander Agassiz, Eeports ontheEesults of Dredging. . . . by the 'Blake,' 

 no. xxiv. pt. 1; 'Eeport on the Echini,' Mem. Mus. Comp. Zool. Harvard, 

 vol. x. no. 1 (1883), p. 83. 



3 Op. jam cit. pt. ii. Quart. Journ. Geol. Soc. vol. xlviii. (1892) p 223. 



4 W. H. Dall and G. D. Harris, ' Correlation Papers : Neocene,' Bull. "U.S. 

 Geol. Surv. no. 84 (1892), p. 151. 



6 W. B. Scott, ' The Mammalia of the Deep Eiver Beds,' Trans. Amer. Phil. 

 Soc. vol. xvii. (1894) p. 62. 



