554 



NA TURE 



[September 26, 1907 



have a peculiar significance in Japan, as the mean value 

 is tabulated of a group of two observations, one of which 

 is excepted ! 



The engineering experiment station of the University of 

 Illinois has published a Bulletin (No. 13), by Dr. N. 

 Clifford Ricker, professor of architecture, describing an 

 extension of the Dewey decimal system of classification 

 applied to architecture and building. The decimal classifi- 

 cation has been largely adopted in libraries in Europe and 

 America, and the proposed extension should prove useful 

 to architects and engineers for classifying collections of 

 lantern-slides and photographs, and for a card index to 

 technical periodicals. 



The Director-General of Indian Observatories has issued 

 a memorandum, dated August 8, with reference to the 

 probable monsoon rainfall, during August and September, 

 1907, based on data obtained since the publication of the 

 previous memorandum of June 8. .'\mong the chief factors 

 taken into consideration were the excess of pressure in 

 South America in July, while in the Indian Ocean the 

 deficiency still persisted. It has previously been pointed 

 out that these conditions are favourable to Indian rain- 

 fall, and Dr. Walker thinks it likely that the total amount 

 during August and September will reach or exceed the 

 average. 



September has so far proved exceptionally fine over the 

 entire country, and the whole period since the sth or 6th 

 of the month has been almost entirely rainless. At Green- 

 wich rain fell on each of the first five days, the aggregate 

 measurement being 0-44 inch, but no rain has fallen sub- 

 sequently, the dry weather continuing practically for three 

 weeks. At Yarmouth the rainfall to September 25 was 

 023 inch, whilst the average for the month is 2-41 inches, 

 and at both Clacton-on-Sea and at Dover the rainfall 

 amounts to 027 inch. The rain has been heavier and 

 more frequent in the north, and at Sumburgh Head there 

 have only been three days without rain, the total measure- 

 ment to September 24 being 2-64 inches, which is only 

 068 inch short of the average for the whole month. Much 

 mist or fog has prevailed during the past week in many 

 parts of the country, and radiation frost has occurred at 

 night. At Greenwich the exposed thermometer on the 

 grass fell to ■a,"--] on the morning of September 23, and 

 there have already been four frosts in the open, as shown 

 by the exposed thermometer, since the commencement of 

 the month. \ change in the type of weather is in pro- 

 gress, and the steadily falling barometer foreshadows the 

 setting in of unsettled conditions. 



In accordance with the decision of the International 

 Union for Cooperation in Solar Research that a re-deter- 

 mination of the wave-lengths of certain standard lines 

 should be carried out by independent observers by the 

 interference method of Drs. Fabry and Perot, Mr. A. H. 

 Pfund, of Johns Hopkins University, has, according to 

 a note in the Physical Review for .'\ugust, recently 

 measured the iron lines, and has obtained values which 

 differ from those of Fabry and Perot by less than one 

 part in a million. Mr. Pfund is now engaged in measur- 

 ing the wave-lengths of the titanium lines. 



The Zeitschrift filr Instrumentenkunde for August con- 

 tains a short account, by Dr. von Rohr, of the life and 

 work of the late Dr. S. Czapski, of Jena, so well known 

 for his masterly article on Abbe's theory of optical instru- 

 ments in Winkelmann's " Handbuch der Physik." He 

 was born in 1861, and after a university education became 

 Abbe's private assistant in 1885. He possessed a 

 NO. 1978, VOL. 76] 



wonderful power of grasping the essential points of any- 

 thing new brought to his notice, and Abbe found in him 

 a friend to whom he ultimately entrusted the publication 

 of his theories. 



Two papers from the pen of Dr. L. \. Bauer which 

 have appeared recently serve lo remind us of the 

 prominent position w^hich the United States is taking in 

 the extension of our knowledge of the magnetic state of 

 the earth. The first, in the Technology Quarterly for 

 June, summarises the recent results obtained from a de- 

 tailed survey of the United States and from the voyages 

 of the survey ship Galilee across the Pacific. The second 

 is the official report of the department of research in 

 terrestrial magnetism of the Carnegie Institution of 

 Washington, and deals with the voyages of the Galilee 

 from October, 1905, to October, 1906, in greater detail. 

 From the latter we gather that the charts of the Pacific 

 at present in use give variations of the compass less 

 than the true value by 1° or 2°, a very serious defect from 

 the navigator's point of view. 



Some curious observations made a few years ago by 

 Dr. A. Heydweiller as to the electrification of the human 

 body by the bending or stretching of the knee or elbow 

 joint receive their explanation in a paper by Drs. S. 

 Tereschin and A. Georgiewsky in the Physikalische Zeit- 

 schrift for September i. According to the latter, the 

 electrification produced is due entirely to friction of the 

 foot of the person experimented on on the insulating 

 stand on which he is placed, or, if he is clothed, to the 

 friction between body and underclothing or between under 

 and overclothing. For the electric charges produced in 

 these circumstances the human body is comparatively a 

 good conductor. 



In the Revue scienlifique for August 31, Dr. C. I'^ry 

 gives a short illustrated account of the new methods of 

 determining high temperatures in industrial operations. 

 For temperatures up to 700° C. he recommends a thermo- 

 electric couple of iron-constantan, from that to 1300° C. 

 one of platinum and its alloys, in each case in combin- 

 ation with a self-registering arrangement. Where the 

 thermo-couple would be injured if brought into direct 

 contact with the source of heat, he advocates the use of 

 his own pyrometer, in which the radiation from the source 

 is concentrated by a concave mirror on to the thermo- 

 junction. For sources of small dimensions at temperatures 

 above 900° C, optical pyrometers, e.g. Wanner 's, are the 

 most useful. 



The question of the improvement of the " small power 

 load," to which electric supply companies and borough 

 electricity committees are perforce paying more attention 

 at the present time than heretofore, is raised in an article 

 by Mr. H. S. Hatfield in the Electrician of September 13. 

 The difficulties attendant on the development of the small 

 power load, and the inability of the private lighting con- 

 sumer to avail himself of the offer of cheap power, have 

 been up to the present very great, owing to the fact that 

 the supply must be separately metered, and it is necessary 

 either to instal duplicate wiring or to use submeters. The 

 cost of a separate service generally bars the use of heat- 

 ing and power appliances by the small consumer. The 

 submeter system is free from the objection of first cost to 

 a great extent, and the meters may be removed and used 

 elsewhere, but so far this system has not been adopted 

 to any extent. The objections to the submeter system 

 have been that, although the capital expended on meters 

 would not be irrecoverable, still the cost of four or five 



