ZOOLOGY AND BOTANY, MICEOSCOPY, ETC. 715 



the pectorals are much smaller relatively, serving rather as balancers, the 

 principal work of locomotion being performed by the tail and caudal fin. 

 Among true Vertebrates, the only structures suggestive of the 

 cephalic appendages of the Ostracoderms are the external gills during 

 their early embryonic stages, including among these structures the 

 " balancers " of Amphibian larva?. The large size and anterior position 

 of the latter appendages make them especially suggestive of the oar-like 

 appendages of Tremataspis. 



Peculiar Modification in Permian Dipnoans.* — C. R. Eastman 

 describes a species of Sagenodus (S. pertemtis sp. n.), which occupies a 

 unique position amongst fossil Dipnoans in having a dentition adapted 

 for cutting instead of crushing, thus paralleling the conditions found in 

 certain Palaeozoic sharks -and in recent Gymnodonta. It may be 

 plausibly associated with the change from marine to brackish-water 

 conditions that took place during the Permian. A sharp cutting edge 

 is developed, and the author is struck by the fact that in the aberrant 

 series of Bdestus-\ike sharks, that flourished contemporaneously, a similar 

 departure occurs. In two of the most conservative and persistent groups 

 of fishes, namely the Ceratodonts and Cestracionts— both of which have 

 had a continuous existence ever since the Permian — the extreme of 

 variation was attained toward the close of the Palaeozoic. Another 

 interesting feature to be brought forward in connection with Sagenodus 

 pertenuis is its apparently wide distribution (e.g. in Russia as well as 

 Texas). Bearing in mind the world-wide scattering of the Edestus 

 series that took place during the late Palaeozoic, Eastman notes that the 

 stimulus which quickened variation and distribution was responded to 

 simultaneously by the two groups of fishes exceeding all others in longevity, 

 after which they relapsed into sluggishness. 



Lost Atlantis.f — R. F. Scharff has studied the land and freshwater 

 faunas of the Canary Islands, Madeira and the Azores, and comes to 

 conclusions very different from those of Wallace. There may have been 

 colonisation of these islands by animals transported by winds and 

 currents, but a large percentage of the fauna indicate connection with 

 present continents. 



He maintains that " Madeira and the Azores, up to Miocene times, 

 were connected with Portugal ; and that from Morocco to the Canary 

 Islands, and from them to South America, stretched a vast land which 

 extended southward certainly as far as St. Helena. This great continent 

 may have existed already in Secondary times, as Dr. Jhering suggested ; 

 and it probably began to subside in early Tertiary times." Scharff 

 believes that the northern portions of this Lost Atlantis persisted until 

 the Miocene, and that subsequently, -in early Pleistocene, there was 

 again a connection of the Atlantic islands with the Mediterranean 

 countries (Africa and Europe). These ideas agree with those of 

 A. E. Ortmann, at least as far as they admit the connection of West 

 Africa with South America. 



* Amer. Naturalist, xxxvii. (1903) pp. 493-5 (2 figs.). •' 

 t Proc. R. Irish Acad., xxiv. (1903) pp. 26S-302. 



