230 Scientific Intelligence. 



"2. Their restriction in altitude is only apparent. Their pres- 

 ent lines of altitude merely mark ancient or existing basin floors 

 or plains. 



" 3. They are derived from mineralized solutions brought to 

 the surface by capillarity, and are essentially replacements 

 (either mechanical or metasomatic) of soil or- of rock decomposed 

 in situ, or of both. 



"4. In the humid regions of India the tendency of change in 

 laterites is towards hydration and not towards dehydration. 



"The foregoing replacement hypothesis would appear to sup- 

 ply a fairly reasonable explanation for all the eccentricities of 

 laterite." 



Abundant references are given in this article and these seem to 

 have been examined with critical discrimination by the author. 



H. E. G. 



9. Contribution a V Etude des Roches alcalines de V Est- 

 Africain ; par H. Arsandaux. Ex. des Comp. rend. sci. de la 

 mission Duchesne-Fournet, 4°, pp. 96, 12 tab., 9 pi. Paris, 1906. 

 — That remarkable tectonic feature, the great Rift Valley of 

 Africa as it runs northward into Abyssinia, at Addis Abbeba 

 widens out into a huge depressed triangular area, called the Afar 

 region. On the west this is bounded b}^ the escarpment of the 

 Abyssinian highland, which continues north to Berbei-a on the Red 

 Sea ; in a similar way it is bounded on the southeast by the Somali 

 plateau. The third side of the triangle on the northeast is the 

 coast line of the Red Sea. French Somaliland comprises the 

 southern part of this area. This depressed region is dotted over 

 with volcanoes, some of which are still active, and is mostly 

 covered with extrusive igneous rocks. These, especially along 

 the southern border at the edge of the Somali plateau escarp- 

 ment, have been studied by the author on his journey from Jebuti 

 on the Red Sea to Addis Abbeba in Abyssinia, and to this are 

 added his later petrographical and chemical investigations of the 

 material collected, which form in fact the main part of the work. 



These rocks are of acid alkalic character and of three promi- 

 nent types, rhyolites with aegirite and riebeckite or comendites, 

 pantellarites with aegirite-augite and cossyrite and trachytes. 

 The author gives 13 analyses of these rocks but unfortunately 

 they are only partial ones, the less prominent oxides not having 

 been determined, and this, especially in the TiO„, makes a com- 

 parison with the rocks of Pantellaria less satisfactory than could 

 be wished. The author classifies the rocks according to the new 

 quantitative classification, which he uses with some modifications 

 of his own. His results are of value in adding further confirma- 

 tion to the existence of the remarkable petrographic province of 

 East Africa, in giving it much greater extension and in present- 

 ing additional evidence of the highly alkalic nature of its rocks, 

 in which soda predominates. They are of interest also in show- 

 ing that such types as those of Pantellaria, which have been con- 

 sidered local rarities, may, as the exploration and petrologic study 



