522 



SCIENCE. 



[Vol. I., No. 18. 



prove of this change, as he regards the cartographic 

 material on which the new estimate was made as of 

 less than mediocre value. — (Boll. soc. {/eogr. iial., 

 vii. 18S3, 241. Further discussion of the question 

 is given in AUi istit. veniito, ix. 1883, 179, 29.5.) 

 ■w. M. D. [1035 



{Ania.) 



Euphrates valley. — A corrected sketch-map of 

 M. v. Thielmann's route from Kerbela, near the Eu- 

 phrates, westward across the desert valley to Palmyra, 

 is prepared by R. Kiepert. It shows the great bar- 

 renness of the adjacent flat country, slightly indented 

 by dry stream-courses extending north-easterly to the 

 river. Some of these are two hundred feet below 

 the general surface, and sometimes contain pools 

 and springs. — (ZeitscAr. /. erdk. Berlin, xvii. 458.) 

 ■w. M. D. [1036 



Improvements in Persia. — Dr. J. E. Polak con- 

 cludes his account of an expedition to the Karagan 

 and Elvend regions in 1882 by noting the changes in 

 the country since his earlier visit in 1860. In addition 

 to the overland telegraph-line that connects India 

 with Europe, there are several shorter lines across 

 the country. The service is regular, and despatches 

 can be sent in English and French as well as Persian. 

 Many new roads have been constructed, and, although 

 not to be compared with the smooth highways of 

 Europe, they serve well for caravan traffic ; but roads 

 are still lacking in many districts. A responsible 

 postal-service is established, both for the interior and 

 for foreign correspondence; and a uniform currency 

 in gold and silver is introduced. With the improved 

 means of communication, letters of credit can now 

 take the place of a heavy supply of metallic money, 

 that travellers formerly found necessary. Railroads 

 are projected from the Caspian southward : they will 

 have the advantage of finding coal and wood near 

 their lines, but also the diificulties of heavy grades 

 between the coast and the interior tableland, and a 

 lack of good harbors at their termini. European 

 methods are introduced in many civil and military 

 arts, and a general tolerance of most sects and nation- 

 alities. Whether this improvement will continue or 

 not is doubtful, as the present Shah is over fifty 

 years old, and none of his sons give assurance of 

 carrying on his reforms. — (Mitth. geogr. ges. Wien, 

 xxvi. 1883, 106. ) w. M. D. [1037 



{Pacific Ocean.) 

 Polynesia. — An entertaining sketch of a three- 

 years' voyage to many of the island-groups in the 

 western Pacific is given by Dr. O. Finsch, who has 

 lately returned to Europe with large collections. 

 His studies were chiefly ethnological. Opportunity 

 for such investigation is rapidly disappearing; for 

 the local peculiarities of the natives on the various 

 island-groups are fast fading away under the influ- 

 ence of traders and missionaries. Among the natives 

 of the Marshall group, the making of large canoes 

 from the trunks of breadfruit-trees is already a lost 

 art. On one of the Caroline Islands, only about'three 

 hundred natives remain; and their earlier customs 

 have largely disappeared with their conversion to 

 Christianity. In the Melanesian Islands there has 

 been less change. The natives go naked, and retain 

 their cannibal fashions; and, by the absence of cer- 

 tain peculiarities not at all flattering to our civiliza- 

 tion, the lack of European influence is further proved. 

 Jlerr Finsch found the atolls monotonous. "They 

 are like American hotels : in knowing one, you know 

 them all." The higher islands have much more in- 

 terest. The irregularity of communication between 

 the different islands makes travel very difiBcult. One 



must wait for accidental opportunities. With a little 

 schooner of twenty tons, and a native crew of six or 

 seven men, much more could be done. — ( Verh. erdk. 

 Berl., 1882, 553.) w. M. D. [1038 



Philippine Islands. — Dr. S. Kneeland regards 

 this group, with many others south-east of Asia, as the 

 remains of a sunken continent, finding evidence for 

 this view in their broken outline, in the distribution 

 of races and their monuments, and in the numerous 

 volcanoes on the fracture along the border of the lost 

 land ; but this latter point is certainly open to ques- 

 tion, as is his opinion concerning the finished con- 

 dition of the earth, and the office of volcanoes as 

 safety-valves to earthquakes. Volcanic and seismic 

 phenomena are very marked on these islands. The 

 symmetrical cone of Mayou gave forth a continuous 

 stream of lava from its vei'y summit for the last five 

 months of 1881, and, in earlier years, has done great 

 damage to the villages on its flanks. The ruins of 

 the old town of Daraga, on the south-east, may still 

 be seen partly covered by the lava of 1814. Majay- 

 jay or Banajao, now dormant, formerly contained a 

 lake that was destroyed in the eruption of 1730. 

 'Large stones thrown from it are scattered far and 

 wide beyond its lava-flows.' From the lake of Bonbon, 

 seventy miles in circumference, rises the cinder-cone 

 of Taal, twelve hundred feet high, with a ragged crater 

 six miles around, within which is a sulphurous lake 

 giving forth suffocating fumes. The effect of earth- 

 quakes is seen in the change from heavy stone to light 

 wooden buildings of Spanish construction. The most 

 violent recent shocks at Manila were in 1863 and 

 1880. A meteorological observatory in charge of the 

 Jesuits publishes a daily weather bulletin ; January 

 and February have the coolest weather, with dry 

 north winds; April and May are hottest; and August 

 and September have the heaviest rains. Having an 

 extent from north to south over several degrees of 

 latitude, and a strongly broken surface, the islands 

 enjoy a remarkable variety of climate, and the pine 

 and maize flourish as well as the palm and orange. 

 The author's chief attention was given to ethno- 

 graphic questions, and some of his results have 

 already appeared in Science. — [Bull. Amer. geogr. 

 soc, 1883, no. 2.) w. M. D. [1039- 



BOTANY. 



Flowers of Turneraceae. — From studies by 

 Urban, it appears, that, of the eighty-three species, 

 fourteen are certainly homogone, and five probably 

 so, while forty-eight are dimorphic, and eight proba- 

 bly so. Six species are incompletely dimorphic; one 

 has six varieties homogone and six heterogone, and 

 one is unknown, with respect to the length of the 

 essential organs. Mathurina penduliflora, Piriqueta 

 capensis, Berneriana madagascariensis and odorata, 

 which depart most from the other Turneraceae, and 

 are remarkable for their geographical distribution, are 

 homogone. Aside from these, the homogone species 

 are represented in all genera, and in most of the 

 smaller groups of naturally related species, and they 

 are distributed as widely as the order. 



When a single individual of a species, found homo- 

 gone in many specimens from different localities, 

 shows an inclination to heterogony, this manifests 

 itself in the increased length of the style, while the 

 stamens retain their usual length. The northern- 

 most variety of Turnera ulmifolia is represented only 

 by the long-styled form. Certain species are char- 

 acterized as incompletely dimorphic. The long-styled 

 form is as it should be, while in the short-styled 

 flowers the branches of the stigma nearly or quite 

 reach the anthers. In these self-fertilization can 



