July i8, 1907] 



NA TURE 



279 



to snowfall for some fifty places. The monthly and yearly 

 summaries for the whole 507 stations include means for 

 past years, in some cases s^oing back to 1867. The maps 

 which accompany the vcliime are very, clear, and exhibit 

 the rainfall for the year ir)o() by lines showing equal 

 amounts (isohyets) ; the amounts on the "west coast 

 between iat. 5q'' and 62°, the usual track of the Atlantic 

 storms, range from 1200 mm. to 3000 mm. at some few 

 points, while inland the amounts vary from 400 mm. to 

 Soo mm. yearly. 



The nirector-nencral of Indian Observatories has issued 

 a memorandum, dated June 8, on the abnormal features of 

 th-" weather of tht past half-year, with a forecast of the 

 probable character of the south-west monsoon rains of 1907. 

 From the various conditions a.'fecting the question, all of 

 which are clearly stated, the outlook for the total rainfall of 

 India during the period June to September does not appear 

 on the whole to be unfavourable ; it seems, however, likelv 

 that the effect of the heavy and late snowfall will show 

 itself in north-west India in the delay of the establishment 

 of the monsoon, or in diminished rainfall. An interesting 

 fact in connection with the investigation is mentioned by 

 Dr. Walker, viz., that statistical analysis shows that when 

 estimating the amount of monsoon rainfall corresponding 

 to a gi\en pressure distribution and a given frequency of 

 sun-spots, the influence of solar activity upon Indian rain- 

 fall is almost exactly that which corresponds to the dis- 

 turbance in the pressure distribution caused by the solar 

 activity. 



In the Journal of the Meteorological Society of Japan 

 for April, Dr. T. Okada gives several interesting instances 

 of the occurrence of Fohn winds at Wonsan, one of the 

 seven observatories of the first order established by the 

 Japanese Government in Corea in 1904. The station lies 

 on the eastern coast, in Iat. 30° q' N., and is surrounded 

 by high mountain ranges, except on the east side. The 

 phenomenon is almost always associated with westerly 

 winds, and causes abnormally high temperature and 

 excessive dryness of the air as compared with that obtain- 

 ing at other stations. In another article Dr. Okada dis- 

 cusses the effect of snow upon the diurnal variation of 

 temperature in the lower parts of the atmosphere, based 

 on hourly observations at various depths at Hokkaido in 

 February last. The total daily heat exchange in the snow 

 on the ground was found to be approximately 19 gram- 

 calories per square centimetre ; the amount of the exchange 

 on clear days was nearly double that on cloudy days. 

 Other articles (in Japanese) deal with periodicity of earth- 

 quakes, and density of snow and evaporation from its 

 surface. 



Prof. H. Potonie has issued a fourth edition of his 

 pamphlet on the origin of coal (Berlin : Borntraeger 

 Brothers, 1907). In its present form it covers fortv-seven 

 pages, and contains twenty-eight admirably reproduced 

 illustrations. The author's views are well known, and we 

 are pleased to note that he intends shortly to publish an 

 exhaustive work on the subject. 



The British standard specification for material used in the 

 construction of railway rolling-stock (Report No. 24 of the 

 Engineering Standards Committee) has been re-published 

 in revised form. Several important alterations have been 

 made. In the specification for steel castings, the number 

 of tensile and bend tests required for waggon-wheel centres 

 has been reduced, as it was considered that the quantity 

 asked for under the existing specification was somewhat 

 in excess of that usually obtaining in general practice. 



NO. igbS. VOL. 76] 



For locomotive-wheel centres cast with heavy balance 

 weights it has been made permissible to reduce the height 

 of the fall in the drop test. In connection with the specifi- 

 cation for copper and brass tubes for locomotive boilers, an 

 alternative drifting test has been added to the clause deal- 

 ing with the bulging test. The principal alterations in the 

 specifications dealing with steel plates for locomotive 

 boilers, locomotive frames, and carriage and waggon 

 frames, are in connection with the new standard iv. 

 diameter test-piece for bars of more than i inch diameter. 

 This shorter test-piece has been introduced to reduce the 

 amount of material required for testing and the amount of 

 turning down when such is necessary. 



In the Journal of the Franklin Institute (vol. clxiii.. 

 No. 6) there is a lengthy paper by Dr. William Campbell 

 on the changes in structure ir> iron and steel, in which the 

 changes that tal<e place in the iron-carbon series are con- 

 sidered in the light of recent work. He considers that 

 the cementite-martensite series is the unstable one. 

 Absence of silicon and rapid cooling tend to cause white 

 cast iron to be formed. Grey cast irons are the martensite- 

 graphite series, which occur with much silicon and slow 

 cooling. The formation of graphite is due to the decom- 

 position of cemenlito by re-heating to temperatures of 

 about 1000° C. Most cast irons are a mixture of white 

 and grey, or cementite, martensite, and graphite. The 

 simultaneous occurrence of cementite and graphite in certain 

 specimens of siliconless irons cannot be explained satis- 

 factorily except by assuming that there are two systems, 

 (a) ferrite and graphite, and (b) ferrite and cementite. 

 This docs away with the necessity of assuming a reaction 

 between graphite and martensite to form cementite at 

 about 1000° C. In this paper the term martensite is used 

 in its old meaning, the solid solution of carbon in iron. 

 Now, however, the solid solution is known as austenite, 

 and martensite is regarded as a transition product. 



Notwithstanding the much improved statistics recently 

 issued by the Lunacy Commissioners, thoroughly satis- 

 factorv materials are still wanting for solving the question 

 whether the prevalence of insanity is or is not increasing. 

 The importance of the problem, especially in its bearing 

 on the persistently urged theory of progressive physical 

 deterioration, imparts special interest to a paper by Mr. 

 Noel A. Humphreys on the alleged increase of insanity, 

 published in the Journal of the Royal Statistical Society 

 (vol. Ixx., part ii.). This paper shows in a striking manner 

 the value of scientific statistics in checking crude figures. 

 The .luthor expresses a decided opinion that there is no 

 absolute proof of actual increase of occurring insanity in 

 England and Wales, and that the continued increase in 

 the number and proportion of the registered and certified 

 insane is due to changes in the degree and nature of mental 

 unsoundness for which asylum treatment is considered 

 necessary, and to the marked decline in the rate of dis- 

 charge (including deaths) from asylums. 



In an article entitled " The Measurement of Nerve 

 Force,'" contributed to the May number of the Contem- 

 porary Review, Dr. A. T. Schofield describes experiments 

 with an instrument called the " sthenometcr," which, it 

 is suggested, provides a means of measuring an unknown 

 " nerve force " emanating from the human organism. 

 The instrument consists essentially of a straw balanced 

 on a needle point and placed under a- glass case. When 

 a hand is brought close to the glass, at right angles to 

 the straw and with the tips of the fingers opposite the 

 end of the straw, a motion of the straw toward the hand 

 is obtained. Dr. Schofield concludes from evidence of this 



