April 8, 1920] 



NATURE 



181 



science had to be reported and new methods of working 

 arranged for, especially With regard to aviation. The 

 magazine contains a reproduction of a photograph of 

 the members of the conference. 



The Meteorological Office chart of the North Atlan- 

 tic Ocean for February contains some notes on the 

 origin and distribution of ice in these waters, and some 

 account of the ice patrol of the United States Govern, 

 ment which was resumed in 1919 after several years' 

 interruption during the war. Two coastguard cutters 

 have been detailed for the purpose of locating icebergs 

 and pack-ice in the vicinity of trans-Atlantic steamship 

 routes. During the months of April, May, and June 

 the two vessels alternate on patrol, each taking 

 fifteen days in the ice region, exclusive of the time 

 taken in going to and from Boston for coal and sup- 

 plies. Movements of ice are reported by wireless at 

 fixed hours daily. At 6 p.m. (75th meridian time) ice 

 information is sent broadcast with a 600-metre wave- 

 Itngth. The message is repeated three times. At 

 f).i5 p.m. the same information is sent out, using a 

 300-metre wave-length. At 4 a.m. a message defining 

 the native and southern limits of the ice is sent to the 

 New York Hydrographic Office. Ice information is 

 also sent at any hour to any ship With which the patrol 

 vessel can communicate. It will be recalled that this 

 is the work which was initiated by the Scotia subse- 

 quent to the loss of the Titanic. The chart also bears 

 an interesting map showing the drifts on the east coast 

 of Greenland, in Baffin Bay, and in Davis Strait of 

 various ships that have been imprisoned in the ice, and 

 of castaway crews during the last hundred vears. The 

 March chart gives an account of the relation of the 

 North Atlantic ice to currents and fogs. 



Cherrapunji, in the Khasi Hills in India, is often 

 cited as having the greatest known annual rainfall. 

 According to the Indian Meteorological Department, 

 the mean annual rainfall there is 426 in. The greatest 

 precipitation is said to have occurred in 1861, when 

 a rainfall of 905 in. was recorded, though doubt has 

 been expressed as to the accuracy of this record. It 

 appears, however, that the Cherrapunji rainfall is sur- 

 |)assed by records on the mountains in the Hawaian 

 islands. Thus Mount Waialeale is the peak (5080 ft.) 

 of the Island of Kauai, but is inaccessible except to 

 the most expert mountaineers. On this account it 

 was very difficult to maintain the station, and the 

 record has finally had to be discontinued. According 

 to the Monthly Weather Review (U.S. Dept. of Agric), 

 vol. xlvii.. No. 5, during the periods August 2, 191 1- 

 March 26, 1914 and May 31, 1915-^August 13, 1917, 

 a total of 1782 days, there was recorded on Mount 

 Waialeale a total precipitation of 2325 in., or an 

 average of 13047 in. per day. In a 365-day year this 

 would amount to an annual precipitation of about 

 476 in. No records were obtained during the years 

 1914 and 1918, but these years were considered the 

 wettest since the local Weather Bureau Office was 

 established in the Hawaian islands. Comparative 

 estimates from trustworthy records obtained at near- 

 by stations indicated that the rainfall at Waialeale 

 must have exceeded 600 in. From May 21, 19 15- 

 NO. 2632, VOL. 105] 



May 30, 19 16, the recorded rainfall of Mount Waia- 

 leale was 561 in. The Hawaian islands are known 

 for other very damp spots. Thus Puu Kukui, 5000 ft., 

 on the Island of Maui, had a seven-year average of 

 369 in. (maximum 562 in. in 1918). On the Island 

 of Hawaii, at a certain spot of ^ooo-ft. elevation, the 

 rainfall in 19 14 amounted to 504 in. At at least a 

 dozen other spots, all more than looo-ft. elevation, the 

 rainfall in each of the years 1914 and 1918 exceeded' 

 350 '"• 



Technologic Paper No. 123, by Mr. D. W. Kessler, 

 of the Bureau of Standards, Washington, is devoted 

 to the tests of the physical and chemical properties- 

 of fifty of the commercial marbles of the United 

 States. Marble has been selected as the first stone 

 to be tested, but the whole of the deposits of stone 

 in the country are to come under test in course of 

 time in order to provide the knowledge required by 

 architects in designing structures. The tests are of 

 tensile and compression strengths, specific gravity, 

 porosity, absorption of water, effect of freezing, 

 chemical composition, electrical resistivity, expansion- 

 with heat, and liability to warping. The trade name 

 and origin of each sample are given, and the tabulated 

 results of the tests fill twenty pages. The properties 

 of the samples differ widely, although the specific 

 gravities do not differ more than about 5 per cent, 

 from each other. On heating, each sample expands, 

 and on afterwards cooling fails to regain its original 

 dimensions. In consequence of this, marble sub- 

 jected to frequent heating and cooling is liable to 

 warp. 



Scientific Paper 352 of the Bureau of Standards, 

 Washington, gives the results of the measurements of 

 the expansion of forty samples of porcelain, about the 

 same number of samples of bakelite and similar mate- 

 rials, and about a dozen samples of marble and lime- 

 stone, made bv Messrs. W. H. Souder and P. Hidnert, 

 of the Bureau. The samples were in the form of rods 

 30 cm. long and i cm. square section, and were heated 

 in a horizontal electric furnace. The expansions were 

 measured by a pair of microscopes mounted on a bar 

 of invar. For the porcelain samples the coefficients 

 per degree Centigrade between 0° C. and 200° C. vary 

 from 2 to 20 millionths, according to the composition, 

 and between 200° C. and 400° C. from 3 to 11 mil- 

 lionths. Beryl porcelains have the smallest coeffi- 

 cients. For bakelite and similar materials no values 

 can be given, as there is so much contraction on again 

 bringing the material to its original temperature. The 

 marbles up to 100° C. have coefficients between 5 and 

 15x10-*, and at higher temperatures larger values. 

 On cooling to their original temperature they show a 

 permanent expansion. When cooled to -80° C. marble 

 expands to nearly the same extent as when heated to 

 80° C, so that it has its maximum density in the 

 neighbourhood of 0° C. 



The Science Reports of the University of Sendai, 

 Japan, for December, 1919, contain a paper by Mr. S. 

 Konno on the heat conductivities of metals below and 

 above their melting points. The metals were tested in 



