December 29, 1910] 



NATURE 



281 



throughout British Columbia ; besides these, however, 

 observations for six years are available at Sand Heads 

 station, and for shorter periods at the others. On the 

 eastern coast longer periods are available, and the tables 

 for Quebec are based upon observations extending over 

 thirteen years. It is claimed that the tables for Quebec, 

 Father Point, Halifax, and St. John are now superior to 

 those of any other harbour on the Atlantic coast of North 

 America. 



We have just received Water Supply Papers Nos. 245, 



247, 250, 251, and 237, 239, published by the United States 



Geological Survey, in addition to the papers of the same 



series referred to elsewhere (p. 2S3). The first four papers 



deal with the surface waters of the Missouri and Lower 



Mississippi Basin, the Great Basin, and California, and 



ecord the gauge readings and discharge measurements 



nade in 1907-8. Covering as they do a large area where 



rainfall is slight, the results are interesting, though, of 



ourse, they extend over a short period only, and 3re in- 



•nded to be a preliminary investigation. River velocities 



.re determined by the Price current-meter, which is almost 



xclusively employed by the Survey, and in this way results 



■f much value are obtained rapidly and from a very wide 



irea. The other two papers treat of the quality of the 



-urface waters of Illinois and California, especially with 



•gard to their potability and their suitability for indus- 



:rial purposes. 



Is the last number of the Proceedings of the Royal 

 Society (vol. Ixxxiv., A, No. 572) is an important memoir 

 V Sir George Darwin on the tidal observations made 

 luring Sir Ernest Shackleton's Antarctic expedition of 

 1907. The observations are shown by Sir George Darwin 

 to demonstrate a tidal seiche in the Ross Sea, and from 

 its period Darwin concludes that the sea extends far 

 beneath the Great Ice Barrier into the Antarctic continent, 

 passing to the east of the Pole and for at least 10° of 

 latitude beyond it. He remarks that if this arm of the 

 sea extends across Antarctica to the Weddell Sea the 

 seiche would be much as the tidal observations indicate. 

 It was remarked in a note in Nature of May 12, on 

 the expedition by Lieut. Filchner, whose plan is based 

 on the assumption that Antarctica is divided into two 

 parts by a sound connecting the Ross Sea with the Weddell 

 Sea, that if the theory be correct some evidence in its 

 favour should have been forthcoming from the tidal 

 observations. Sir George Darwin's memoir shows that 

 the tides offer striking evidence in favour of the direct 

 connection between the Weddell and Ross Seas. 



.\mong several useful papers in the Journal of the 

 Scottish Meteorological Society for 1909 (recently pub- 

 :shed) there is one of especial interest by Dr. G. .A. Carse 

 nd Mr. D. MacOwan giving a brief resume of the more 

 nportant facts connected with atmospheric electricity, 

 l^escriptions are given of some of the earlier methods of 

 detecting the phenomena, and of Lord Kelvin's water- 

 dropping apparatus, which is most widely used for 

 measuring the atmospheric potential. Observations show 

 "hat, in general, this factor varies with the time (there 

 eing in most places a diurnal and annual variation), 

 tcd that, generally speaking, it increases in proportion to 

 he distance from an extended horizontal surface if the 

 listance between the points is not too great. It has, how- 

 ver, been found by balloon ascents that in fine weather 

 : diminishes with height above ground, thus indicating 

 hat electrification is largely confined to the lower levels 

 f the atmosphere. The annual variation has a maximum 

 bout mid-winter and a minimum in summer, but the 

 periods of the diurnal variation are much more complex. 



NO. 2148, VOL. 85] 



•Another factor of importance is the ionisation of the atmo- 

 sphere, and this is now being investigated more 

 thoroughly. A few of the more interesting of the various 

 theories accounting for the phenomena of atmospheric 

 electricity are briefly sketched, but, so far as known, none 

 has yet been promulgated which sufficiently explains all 

 the observed facts. 



To the Rendiconti R. Accad. Lincei of October 2 Dr. 

 Eredia communicates an interesting paper on the cold 

 period of June in Italy. This cold period has already 

 been shown to exist over a large part of Europe, and to 

 be due to the mean distribution of pressure at that period. 

 But as Italy possesses a valuable series of observations 

 available for the purpose, the author has taken advantage 

 of them to show that this cold period in Italy constitutes 

 a real climatoiogical factor. His tables show differences 

 of the ten-day means of temperature from each other 

 between the third decade of May and the first decade of 

 July, for 120 stations, for the period 1866-1906. He also 

 gives a map showing by various shadings the difference of 

 temperature between the first and second decades of June 

 for different regions. These clearly show that generally 

 there is a considerable fall of temperature in the second 

 decade of June, that it is much more marked in Upper 

 than in Lower Italy, and is considerably influenced by 

 geographical configuration. 



Ax interesting article by Dr. J. Mascart on actinometry 

 and on meteorology at Teneriffe is published in the Revue 

 ginerale des Sciences of November 15. The author points 

 out that in the determination of the solar constant a 

 difficulty arises at the outset ; according to the definition, 

 the receiving surface should be theoretically black for all 

 radiations, having the properties of an integral radiator. 

 Strictly speaking, this preliminary problem has not yet 

 been solved. He described the so-called " absolute " 

 instruments in use, which may be divided into two 

 groups : — (i) calorimetric actinometers, which contain a 

 liquid of known specific heat, of which that bj" Pouillet is 

 the oldest ; (2) those in which the electric energy necessary 

 to produce the same effect as the solar radiation is 

 measured ; to this class belong the actinometers of 

 Angstrom and F^ry. The Solar Committee has adopted 

 as a type Angstrom's compensation pyrheliometer. This 

 decision is excellent as regards uniformity of observations, 

 but might be harmful if it diminished the number of 

 measurements with other apparatus. Reference is made to 

 the observations made by the late Dr. W. Marcet and 

 others on the extreme dryness at times of the Peak of 

 Teneriffe, and on the electric phenomena there which 

 seem to be connected with the former. The author 

 considers the peak to be particularly favourable for 

 observations on atmospheric phenomena and their connec- 

 tion with actinometry ; also for observations of terrestrial 

 magnetism. He thinks that more attention should be 

 given to observations of zodiacal light, crepuscular rays, 

 and atmospheric polarisation ; these subjects are generally 

 omitted from meteorological text-books because they are 

 supposed to have no immediate connection with meteor- 

 ology, but with this view he does not agree. 



Is a communication made to the Illuminating Engineer- 

 ing Society on December 9th Prof. G. W. O. Howe 

 showed that the darkening of the glass bulbs of Osram 

 lamps sometimes noticed is due to the use of a slight 

 amount of copper in the leading-in wires. This copper 

 appears to be projected from the p)oint at which the fila- 

 ment is joined to the negative leading-in wire, and forms 

 on the inner surface of the bulb a distinct shadowgraph of 



