684 



SCIENCE. 



[N. S. Vol. XXIV. No. 622. 



problem and its practical importance should 

 justify its most active investigation. 



The author ^'ives in some detail, fully illus- 

 trating his thesis, the results of his early 

 studies in Mississippi along these lines, pre- 

 senting in Figs. 79, 80 and 81 the most stri- 

 king illustrations of how individuals of one 

 and the same species of post-oak, black-jack 

 oak and deciduous cypress persistently differ 

 in both stature and habit of growth when they 

 recur on the same soil types in different locali- 

 ties throughout the state; finally extending 

 the discussion to observations in the United 

 States at large and to Europe. I can not do 

 better, in closing the review of this valuable 

 work, than to quote again the author where he 

 is discussing the influence of lime in the soil 

 on the character of floras. 



What is a calcareous soil? The definition 

 adopted for this volume has been given in a 

 previous chapter; viz., that a soil must be con- 

 sidered calcareous so soon as it naturally svip- 

 ports a calciphile flora — the lime vegetation so 

 often referred to above and named in detail. 

 Upon this basis it has been seen that some 

 (sandy) soils containing only a little over one 

 tenth of one per cent, of lime show all the char- 

 acters and advantages of calcareous soils; while 

 in the ease of heavy clay soils, as has been shown, 

 the lime-percentages may rise to over one half 

 per cent, to produce native lime growth. 



At first thought it may appear to some that 

 the adoption of such a definition is a subter- 

 fuge to make observations harmonize with 

 theory, but it is not so. Every one will agree 

 that a moist soil, defining it from the stand- 

 point of plant nutrition, is one which will 

 yield moisture to a plant as rapidly as it is 

 needed. On this basis a sandy soil contain- 

 ing 4 per cent, of moisture is as moist as a 

 clay soil carrying 20 per cent., the physiolog- 

 ical difference being determined chiefly by the 

 relative amounts of internal soil surface in 

 the two cases. 



This volume should be introduced to a much 

 wider circle of students than those of the agri- 

 cultural colleges generally. It will be found 

 well suited to serve as the foundation of im- 



portant seminars in chemistry, in geology and 

 especially in plant physiology and ecology. 



E. H. King. 

 October 30, 1906. 



HATCH AND CORSTORPHINe's GEOLOGY OF SOUTH 

 AFRICA.^ 



The visit of the British Association to 

 South Africa was the occasion for the appear- 

 ance of two noteworthy books on the geology 

 of that region : Rogers's ' Geology of Cape 

 Colony,' and Hatch and Corstorphine's ' Geol- 

 ogy of South Africa.' The latter is the m.ore 

 general of the two, as it treats of much the 

 larger area; the former is somewhat more 

 detailed, as all of its space is devoted to the 

 formations that occur in the single colony 

 with which it is concerned. 



The small geological map, scale, 1 :5,000,000, 

 which serves as frontispiece to Hatch and 

 Corstorphine's book, provides a good introduc- 

 tion to the problems considered in the text. 

 The greater part of the area described is occu- 

 pied by the nearly horizontal beds of the Kar- 

 roo system, a vast body of continental deposits 

 which has shared the fate of other stratified 

 formations not containing marine fossils in 

 having been explained by earlier observers as 

 a lacustrine deposit, but which is now recog- 

 nized as of mixed origin. Its lowest ijiember 

 is the famous Dwyka glacial conglomerate, 

 or ' tillite,' as Penck has suggested it should 

 be called, unquestionably of glacial origin. 

 The overlying members of great thickness are 

 probably of mixed fluviatile and lacustrine 

 deposition, as they contain beds of coal and 

 fossils of reptiles, as well as numerous dikes 

 and sheets of dolerite. This great body of 

 continental formations occupies a geosynclinal 

 basin, some 600 miles east and west by 400 

 miles or more north and south. It is ob- 

 liquely truncated by the seacoast on the south- 

 east; there the ancient lands from which the 

 basin deposits were derived, appear to have 

 been lost in the Indian Ocean. On the south, 

 the Karroo system and the underlying forma- 

 tions are folded in long east-and-west anti- 



^Macmillan, 1905, 348 pages, 2 geol. maps, 89 

 figures and plates. $7.00. 



