October 1, 1886. 



SCIEWCE. 



307 



a point on the Oxus corresponding to the Avestern 

 limit of the district called Khoja Saleli, and that, 

 as the Afghans possessed this tract at the time of 

 the earlier agreement, they should be allowed to 

 retain it. This would, moreover, be in strict ac- 

 cordance with the principle laid do\vn on that 

 occasion; viz., that Afghanistan should be con- 

 sidered identical with the actual possessions of 

 the Ameer Shere Ali. 



To sum up the points presented by the Kham-i- 

 Ab question, Russia has in her favor the specific 

 mention of the name Khoja Saleh in the diplo- 

 matic documents. Beyond this fact, strong as it 

 undoubtedly is, Russia does not seem to possess a 

 weighty argument. Ou the other hand, there is 

 the Afghan right of possession, unquestioned by 

 anybody, going back for a long period, and con- 

 firmed in 1873. There is the recognition in 1873 

 of the Khanate of Akcha foi-ming part of the 

 dominions of Shere Ali, and consequently of Af- 

 ghanistan. Finally, there is the practical fact that 

 the Kham-i-Ab of the Afghans occupies almost 

 the identical geographical site of the ' Khodsha 

 Salor' of the Russians. Extraneous arguments 

 may be easily introduced into the case by irre- 

 sponsible writers ; but these are really all the con- 

 siderations that need affect the judgment of the 

 two governments. 



PACIFIC COAST WEATHER. 



Lieut. W. A. Glassfoed, in charge of the 

 Pacific coast division of the signal service at San 

 Francisco, has lately presented a paper to the Cali- 

 fornia academy of sciences on ' Weather types on 

 the Pacific coast.' These types diiier fi'om those of 

 the eastern United States in their relative lack of 

 progressive motion, and consequently in their 

 duration and in the less variability of the weather. 

 Distinct areas of low pressure are rare in southern 

 California, but increase in frequency northward, 

 until they are most numerous about Vancouver's 

 Island. The types recognized for the rainy season 

 (winter) are, 1°, North Pacific cyclonic ; low press- 

 ure over Oregon and Washington, high pressure 

 in the Great Basin, with southerly gales along the 

 coast, and general rains, heaviest in the north ; 

 2°, interior anticyclonic ; like the preceding, but 

 with less distinct cyclonic conditions ; the temper- 

 ature is high with south-easterly winds ; the 

 warm ' Santa Anna ' winds of Los Angeles occur 

 under this distribution of pressures ; 3°, North 

 Pacific anticyclonic ; high pressure in the north, 

 and low in the south, giving clear weather with 

 light, variable winds in the north, but with high 

 winds and southerly gales on the coast of CaUfor- 

 nia ; warm days and cool nights, often frosty ; 



the dreaded dry ' north wind ' of the Sacremento 

 and San Joaquin valley prevails at this time : 4°, 

 general cyclonic ; a rare type, with very low press- 

 vire on the coast, giving severe storms of high 

 southerly winds and heavy rain ; S'', South Pacific 

 anticyclonic ; moderately high pressure along the 

 south-western coast of California, and no distinct 

 centre of low pressure visible, but giving southerly 

 rain-bearing winds ; 6*^, sub-normal type ; irregu- 

 lar isobars and no decided gradients, with variable 

 winds and weather. During the dry season (sum- 

 mer), the weather is very constant, with high 

 pressure to the north-west over the cool ocean, 

 and low pressure over the hot land to the south- 

 east, northerly Avind^ and no rain. The change 

 from the wet season to the dry is indicated when 

 the air temperature on the coast rises permanently 

 over t le ocean temperatui-e. Lieutenant Glass- 

 ford has also compiled an extended table of the 

 rainfall on the Pacific coast from all soui'ces, in- 

 cluding some two hundred stations with records 

 varymg from one or tvpo years up to thirty-seven 

 (San Francisco and Sacramento). This was pub- 

 lished in the San Francisco daily Commercial 

 news for July 1. 1886. The maximum precipita- 

 tion is given for Neah Bay, Washington Territory, 

 where the annual average of nine years' record is 

 110 13 inches. Many other stations in the north 

 exceed fifty and sixty inches a year. In the south, 

 the minimum falls nearly to two inches, being 

 2.56 at Yuma, Arizona, from an eleven-year rec- 

 ord. The lowest of all is a three-years' average 

 for Bishop Creek, Cal., where the annual precipi- 

 tation is only 1.31 inches. The table gives the 

 months separately, as well as the yearly total, so 

 that the seasonal variation is well brought out. 

 In July and August only nine and ten stations re- 

 spectively have over an inch of rain, and these are 

 all in the north or in the interior ; while no rain 

 at all is given for eighty-two and ninety stations, 

 and a number more have only a trace or one or 

 two hundredths of an inch. 



DR. ROMANES ON PHYSIOLOGICAL 

 SELECTION. 



Dr. George J. Romanes, who, in more than a 

 literal sense, may be said to be the legatee of 

 Darwin, publishes in Nature (Aug. 5, 12, 19) 

 an abstract of a paper read before the Linnean 

 society, entitled '-Physiological selection : an ad- 

 ditional suggestion on the origin of species." 



The necessity of such an additional principle is 

 made evident by considering three objections to 

 natural selection as a theory of the origin of 

 species. 1°. The difference between species and 

 varieties in respect of mutual fertility. It is a fact 



