900 



SCIENCE. 



[N. S. Vol. XI. No. 284. 



Missouri river, and continued in the Mis- 

 sourian series along the same stream north 

 from the city. Or, a direction takea south- 

 east of Kansas Citj' is equally advantageous, 

 besides passing through the Clinton district, 

 and reaching into the' old gorges in the 

 Mississippian series which are exposed on 

 the flanks of the Ozarks. Work along these 

 lines, both in the floral and faunal fields 

 were begun by the Missouri Geological Sur- 

 vey in connection with the detailed strati- 

 graphical cross sections, and much valuable 

 material obtained, but the efforts had to 

 be abandoned before the data were com- 

 plete. 



The exposures along the Des Moines and 

 Raccoon rivers, in central Iowa, afford an- 

 other excellent field for making up a stan- 

 dard paleobotanic section. Plant remains 

 occur in many localities and at numerous 

 horizons. Some exquisite things have been 

 observed. Van Meter affords beautiful ter- 

 nately divided fern fronds over a yard 

 across. Mud Creek, below the city of Des 

 Moines, furnishes, in profusion, plant-bear- 

 ing nodules similar to Mazon Creek, in Illi- 

 nois. Ford supplies structural specimens, 

 showing the wood cells and their workings 

 as perfectly as if they were taken from the 

 living plant. Knoxville has extensive plant 

 beds and the bluffs of the streams in Marion 

 county often have tons of shale-slabs with 

 plants in sight at one time. At one time 

 the Iowa Geological Survey began to get 

 material of this kind together for a report 

 on the coal floras of the State supplementary 

 to the reports on coal deposits. But since 

 the appearance of the first volume of the 

 latter nothing farther seems to have been 

 done regarding the plants. 



The main consideration, however, is the 

 fact that the Trans-Mississippian coal field 

 presents for the study of fossil plants a 

 wealth of material unrivalled in the whole 

 country. If standard paleobotanic sections 

 of the region are lacking, it is certainly not 



because the fossil botanist lacks the ma- 

 terial and opportunity to construct them. 

 Charles R. Keyes. 



ON THE ZOO-GEOGBAPHICAL DELATIONS OF 

 AFRICA.* 



The speaker prefaced his communication 

 by remarking that he had nothing abso- 

 lutely new to bring forward, but that, in- 

 asmuch as some views which seemed to 

 be contrary to evidence had been urged very 

 recently, a presentation of the conflicting 

 evidence was timely, if not necessary. The 

 views in question were broached in ' A 

 Geographical Historj^ of Mammals,' by Mr. 

 R. Lydekker, and the addretes of the retir- 

 ing president of the N"ew York Academy of 

 Sciences (Professor H. F. Osborn) published 

 last week in Science (April 13th). 



Beaumont's apologue of the shield has its 

 counterpart for the fauna of Africa. It has 

 two sides facing in opposite dii'ections, and 

 it cannot be understood without taking 

 both into consideration. The proposition 

 to combine Africa with Asia, Europe and 

 North America into a realm contrasted with 

 South America and Australia (or even to 

 combine again Africa and India against the 

 others) may apparently be justifled if we 

 look only to the present mammalian fauna, 

 but if we revert to the past and consider 

 other classes, we must be led to different 

 conclusions. 



The fishes are by far the most instructive 

 in their teaching. Very recent discoveries 

 recorded by Mr. Boulenger add force to 

 their testimony. Those animals represent 

 two very distinct assemblages. On the one 

 side, we have Cyprinids of genera occurring 

 also in India or very closely related to such 

 genera. On the other side, we see numer- 

 ous species belonging to families having no 

 representatives in India or elsewhere than 

 in tropical America. Such are the Cichlids, 



* A communication to the National Academy of 

 Sciences made April 18, 1900, by Dr. Tlieodore Gill. 



