514 Notices of Memoirs — Dr.- Spencer — Submarine Valleys, etc. 



some few other new forms from Gardenazza, and from the Lias 

 Coprolites of Ilsede, are fully described and figured in the 19 plates 

 accompanying this monograph. A comparison of the species from 

 the various localities mentioned above leads the author to consider 

 that they belong to one and the same Eadiolarian fauna. The 

 only new genus proposed, Cyclastriim, is included in the family 

 Porodiscida. A distinguishing feature of the Cittiglio Eadiolaria is 

 the large number of forms of the Order Cyrtoidea ; some of them, 

 moreover, are of unusually large size — one specimen of Stichocapse 

 Umberti, measuring 1-152 mm. by 0'16 mm. in length and breadth, 

 exceeds in size any fossil form of the group hitherto known. Though 

 the siliceous tests in these organisms are now for the most part 

 replaced by pyrites and raarcasite, their structural details have been 

 very perfectly preserved, and they can be determined with as much 

 precision as recent specimens. 



Thanks to this new contribution of Dr. Eiist, taken in connection 

 with his earlier work and that of Professor Parona, we are now 

 furnished with a fairly satisfactory standard of reference as to the 

 character of the Eadiolarian fauna of the summit of the Jurassic and 

 the base of the Cretaceous rocks in the Tyrol, Bavaria, and Northern 

 Italy. " G. J. H. 



II. — Eesemblances between the Declivities of High Plateaux 



AND THOSE OF SUBMARINE AnTILLEAN YaLLEYS. By J. W. 



Spencer. (Transactions of the Canadian Institute, vol. v, 1898, 



pp. 359-368.) 



[Communicated by Professor E. Hull, F.R.S.] 

 rj^HIS paper is a sequel to the " Eeconstruction of the Antillean 

 JL Continent," ' as in it the analysis of the slopes of the drowned 

 valleys had not been considered. Both in the land and in the 

 submarine valleys their gradients are of two kinds: (1) Those of 

 rivers which are flowing over continental plains, or upon the surface 

 of high tablelands, where the declivities of the streams are so gentle 

 as to be often reduced to even a foot per mile; (2) where the 

 valleys are descending from higher to lower plateaux, in which case 

 the descent is over a series of precipitous steps, separated by short 

 gradation planes, marking pauses in the elevation of the land. 

 Thus, if the mean descent of such a valley be taken, an average 

 gradient would be entirely misleading. While the mean slope may 

 reach from 100 to 200 feet per mile, it is found that in reality it is 

 composed of perhaps twenty abrupt steps with almost level flats 

 between. Or the steps may reach a height of five hundred feet or 

 more. Such features are seen descending from the Mexican plateaux 

 (of 8,000 feet in altitude) to the Gulf of Mexico. The valleys end 

 abruptly in amphitheatres indenting the floors of the tablelands and 

 dissecting them. 



In the drowned Antillean valleys long reaches have been dis- 

 covered with slopes of only a foot per mile like that of the 



1 Bull. Geol. Soc. Amer., vol. vii (1894), pp. 103-140. 



