July 30, 1908] 



NA TURE 



The report of the meteorological observations made at 

 the Perth Observatory and other places in Western Aus- 

 tralia in 1906 contains much useful information. The 

 tables for Perth, which are given in great detail, include 

 monthly mean values of the principal elements from 1876; 

 monthly and yearly results are also given for some fifty 

 stations, and rainfall at a large number of places, with 

 averages for previous years. Mr. Cooke, the Government 

 astronomer, remarks that the readings from the various 

 localities may be considered trustworthy ; some very high 

 shade temperatures are recorded, e.g. ii6°-4 at Eyre 

 (January 22), and maxima exceeding 100° were recorded at 

 the majority of stations. In previous years readings of 117° 

 were observed at several places. The weather forecasts 

 are very accurate ; figures 1-5 are added to each forecast 

 to indicate its degree of probability. Eliminating those 

 stated at the time to be very doubtful, the success was 96 

 per cent. 



In the BullBtin dc la Socitte d'Eiicouragcment (vol. ex., 

 No. 6) there is an interesting paper on the combat 

 against fire-damp and carbonic oxide in collieries by Prof. 

 \. Gr^hant, in which illustrations are given of the fire- 

 damp detector designed by the author. In the same issue 

 Mr. S. Wologdine has a note on the heat-conductivity of 

 refractory materials. 



The recent engineering and manufacturing development 

 of Sweden has been rapid. With the opening of the rail- 

 way through the rich iron-ore fields north of the Arctic 

 circle and the extension of hydroelectric installations arc 

 now allied many important ship-canal and other projects, 

 which are fully described in a profusely illustrated article 

 by .Mr. J. G. Leigh in the Engineering Magazine (vol. 

 XXXV., Xo. 4). .Another interesting article in the same 

 number deals with the new processes for metal cutting 

 and autogenous welding. 



.At the installation of Dr. W. F. M. Goss as dean of 

 the college of engineering of tile University of Illinois, 

 several interesting addresses were delivered, which are 

 now published in a Bulletin (Xo. 21) of the University. 

 The subjects dealt with were : — significant events in the 

 development of the college of engineering, by Prof. Ira 

 O. Baker ; on the standing of the technical graduate in 

 !he engineering profession, by Mr. W. L. .Abbott ; on the 

 State College of Engineering, by Prof. W. E. M. Goss; 

 on the value of engineering research, by Mr. R. W. Hunt ; 

 and on the need of graduate courses in engineering, by 

 Mr. Willard A. Smith. 



.A p.'VPER on rail corrugation, read by Prof. C. A. Carus- 

 Wilson before the Tramway Congress on July 10, is re- 

 printed in Engi)ieeriiig. The investigations outlined appear 

 to show that the following conditions are necessary for 

 the formation of corrugations on grooved rails : — (i) in 

 the track (a) irregularities in gauge or level, (6) curvature, 

 or (c) a packed groove ; (2) a rail surface rough with sand 

 or gritty dust ; (3) wheels with check-cutting flanges ; and 

 (4) a critical speed. Corrugations cannot be formed unless 

 conditions (i), (2), (j), and (4) are all present at the same 

 time. These considerations suggest the lines on which it 

 mav be possible to avoid the formation of corrugations. 



The use of concrete and reinforced concrete pipe for 

 culverts in railway embankments has suggested to Prof. 

 A. X. Talbot an elaborate series of experiments, described 

 in Bulletin No. 22 of the University of Illinois. The results 

 throw light upon the resistance of pipe to embankment 

 pressures, and also upon the action of sewers under similar 



NO. 2022. VOL. 7^1 



conditions. Cast-iron pipes, concrete pipes, and reinforced 

 concrete pipes were tested. The specially prepared testing 

 apparatus used included a box of strong and stiff con- 

 struction, and the pipes were embedded in sand, and the 

 load applied through a saddle which rested on a sand 

 cushion. Auxiliary tests were made to connect the results 

 of the investigation with the strength of the materials. 



In a paper on space and mathematical reasoning {Mina, 

 xvii., 67, July), Mr. Leonard J. Russell discusses the 

 theories of Poincar^ and Russell, and puts forward a view 

 of space on the analogy of number, empty space being 

 found to have just such a significance as pure number has. 



In a note contributed to the Rendiconti del R. Istituto 

 Lombardo (2), xli., 11, on problems and methods of metallo- 

 graphy. Dr. Gaetano Maderna urges the importance of 

 metallographic researches, and directs attention to the fact 

 that this branch of study has been neglected in Italy, both 

 by the Government and private firms. 



PuoF. Fausto Morini contributes to the Rendiconto of 

 the Bologna Academy a short note on a monstrosity in 

 the fern .Ispedium lobatiwi, var. angulare, Metten, found 

 in the .Apennines of the Bologna region. One of the 

 lateral segments of the leaf is replaced by a very tiny 

 leaflet devoid of sori, but possessing all the character- 

 istics of an entire leaf, and the author considers that this 

 monstrosity is due to the development of an adventitious 

 gemma, the subsequent development of which has been 

 limited to the formation of the leaflet in question. 



In the .Atti of the Naples Academy of Physical and 

 Mathematical Sciences, xiii. (1908), Prof. G. de Lorenzo 

 discusses the origin of the volcanic craters of the Capo 

 Miseno and the island of Nisida. The papers are illus- 

 trated by photographs, and the observations point to some- 

 what similar conclusions in both cases, namely, that the 

 craters were caused by eruptions at first submarine in 

 character, but ultimately rising into the air, and that the 

 erosive action of air and water, and in particular of the 

 sea, is accountable for the present configuration of the 

 harbours in question. 



Two papers dealing with considerations relating to 

 statistics and probability have reached us. In the 

 Statistical Journal, Prof. F. Y. Edgeworth (June 30) dis- 

 cusses the probable errors of frequency constants. The 

 author alludes to the objections that have been raised to 

 the use of the term " probable error," and examines the 

 evidence for the existence of an inde.x of credibility of this 

 character from the consideration of a number of different 

 problems. A recently issued number (Studies in National 

 Deterioration, part iii.) of the Drapers' Company Research 

 Memoirs contains a paper by the late Ernest G. Pope on 

 marital infection in tuberculosis. While Prof. Karl Pear- 

 son in his editorial criticisms deviates in several points 

 from the opinions expressed by Mr. Pope, it may be safe 

 to infer from the statistics that the effect of infection on 

 the coeflicient of correlation is less than the effect which 

 may be attributable to " assortative mating." 



Mr. J. W. GiLTAV, of Delft, writes to say that the 

 optical illusion, mentioned by Dr. Terada in Nature of 

 July 16 (p. 255), and also in the following issue (p. 277), 

 reminds him of a similar effect noticed some weeks ago 

 when witnessing a bicycle race at the velodrome at 

 Scheveningen. " After having looked for some time at 

 the racers, I casually looked at the spectators : the whole 

 range of benches with the spectators appeared slowly to 

 slide around, in a direction contrary to that moved in by 

 the racers." 



