August 31, 1883.) 



SCIENCE, 



porous, and generally shows no trace of ellVr- 

 vescence in hydroclilorie acid. In general ap- 

 |)earance it closely resenililes :in ini|juro siliceous 

 limestone from which the greater [jortion of 

 the carbonate of lime has been leached away. 

 . Having a thickness of about a hundred and 

 thirty metres, it becomes the chief topographi- 

 cal determinant of that region, and gives rise 

 to profound gorges and bold escarpments. 

 Throughout the greater portion of the mass, 

 it is uniformly fine-grained, but under the mi- 

 croscope has the structure of a tufa. 



Tile upper beds of the series, consisting of 

 thin fresh-water limestones, sandstones, shales, 

 and a large i)roportion of stratified tufas, with 

 conglomerates, have not been seen east of 

 Demirdji-kieui. The fossils collected were 

 found in this portion of the series; and it is 

 evident that the ejection of the andesites began 

 before the deposition of those beds was com- 

 jileted. 



Xinnerous oscillations of the land, as indi- 

 cated by the varying character of the strata, 

 must have occurred during the miocene and plio- 

 cene ; and, in all probability, these were con- 

 nected with the extrusion of the eruptive rocks 

 so abundant in that region. 



The massive rocks of the Troad belong in 

 l>art to those of pre-tertiary origin, but the 

 greater portion were extru<led since the begin- 

 ning of the tertiary jjeriod. The older group 

 includes biotite-hornblende-granite, tpiartz-por- 

 )ihyry, quartz-dioritc, augitc-porpiiyrite, inel- 

 aphyre, and serpentine, while the younger 

 group embraces liparites, andesites, augite- 

 andesitcs. basalts, and ne[)heline-basalt. 



The biotite-hornlilende-granite occurs in a 

 stock-like mass, forming the serrated ridge of 

 Chigri-dagh. It is distinctly younger than the 

 highly ci-ystalline stratified rocks which it i)en- 

 ctnites. and is especially interesting from the 

 fact, that, where it is altered, the fitanite is 

 changed to anatase. The alteration of titanite 

 and ilmenite to anatase is doubtless a common 

 an(i widely distributed occurrence; but, as the 

 crystals of anatase are so small, the}- have gen- 

 erally been overlooked. 



The quartz-porphyries are chiefly niicrogran- 

 ites. and are younger than the biotite-horn- 

 Men<le-granite through which they have been 

 extruded. The dikes in which lliey occur are 

 comparatively small, and do not exercise roncli 

 inflnence upon the topographical features of 

 the country. 



The quartz-diorites form a number of com- 

 paratively small slocke about the b<ise of 

 Mount Ida, and are evidently yonn}>:er than the 

 fjuartzose argillite of the middle zone Oif strati- 



fied rocks, which, in one case, has been meta- 

 iiiDrphoscd into a cordierite and anilalusitc 

 hornfels. It is to be especially noted that 

 these eruptive rocks do not, as fornu'riy siip- 

 liiised, enter into the struclure of .Mount Iila. 



The augite-porjjhyrites (diabase-porphy rites) 

 and melaphyres are. as far as yet known, lim- 

 ited to five outcrops, all lying in a line near 

 the southern coast of the Troad, and, with the 

 exception of tiiat between Ahmadja.and t^valar. 

 arc not imiiortant. At the locality just named 

 it is of especial interest from the fact that 

 melaphyre was the first rock extruded in tli.it 

 isolated (completely surrounded by tertiary 

 strata) volcanic centre, and was followed later 

 by mica-aiidesite, hornblende-andesite, augite- 

 andesite, basalt, and. late if not last, bva large 

 outpouring of liparite. 



The serpentine in the anterior part of the 

 Troad about C^ara-dagh has been derived 

 from olivine-enstatite rocks of a truly eruptive 

 nature. The almost entire absence of the 

 ciiaracteristic reticulated structure in some of 

 the serpentine from the Kemar valley leaves, 

 jieriiaps, some doubt as to the original rock 

 from which it has been derived. As previously 

 stated, the serpentine about the summit of 

 Mount Ida lias been derived from olivine-schist 

 which undoubtedl}- belongs to the stratified 

 rocks. 



Although the ancient eruptive rocks are ap- 

 ])arently not nearly so abundant as those of 

 more recent origin, yet they represent veiy 

 nearly the same range in chemical and miiiera- 

 logical composition. The granite and (juartz 

 ])oiplivries have their modern equivalents in 

 the liparites; the (|iiartz-diorites. in the mica 

 and liorublenile aiiilesit>s : the augite-iiorpliy- 

 rites, in the augite-andesites : the melaphj're, 

 in the basalt. However, no equivalents were 

 found for the ncplieline-basalts and the ancient 

 olivine-enstatite rocks. On the other hand, the 

 syenites, and their modern representatives 

 the trachytes, which were once sup|)oscd to he 

 abundant in the Troad, are now known to be at 

 most only very sparingly represented. 



The liparites occur in various types, with 

 many varieties, and are limited to the south- 

 ern part of the Troad. They appear also 

 south of Molivo on the island of Mitylene, and 

 at JSal Mosac south-west of Aivaly. . They are 

 generally in the stoii\- condition, but frequently 

 glassy upon the boundaries, and contain many 

 fragments of the andesites which they have 

 IK'netrated and overflowed. They always occur 

 in dikes, as at tiozloii-dagli and the great jda- 

 te.MU, which give rise to the peculiar drainage 

 of the Touzla Uiver. That some of the liparites 



