206 



NATURE 



[January 2, 1908 



Prof. A. C. Seward contributes to the Transactions of 

 ithe Geological Society of South Africa, vol. x., a descrip- 

 tion of a collection of Permo-Carboniferous plants from 

 Zululand and Natal. The material contained a large 

 number of specimens referable to Glossopteris, but few 

 .of the forms were distinct. Sporangia were discovered on 

 some of the leaves of Glossopteris indica, but this does not 

 preclude the possibility of the genus being a pteridosperm ; 

 in this respect the association of small winged seeds with 

 the leaves was noted. Other specimens referred to are 

 the genera Phyllotheca, Bolhrodendron, Vertebraria, and 

 Cordaites. The specimens do not furnish sufficient evi- 

 dence for determining the precise horizon in the Permo- 

 Carboniferous system of the coal-bearing strata of Zulu- 

 land and Natal. . 



The Memoirs of the College of Science and Engineering, 

 Kyoto Imperial University, Japan, of which the current 

 issue (vol. i.. No. 3) has been received, contains original 

 memoirs by members of the University. Of the thirteen 

 -papers, eleven are written in English and two in German. 

 The subjects dealt with comprise the ecjuilibrium between 

 reciprocal salt pairs, reaction between carbonic acid and 

 lead acetate in an aqueous solution, experiments on the 

 Tjtilisation of scrap metal, formation of amines from the 

 halogen imido esters, the refining of copper, electrolytic 

 dissociation of partially neutralised weak acids, short- 

 period magnetographs, the theory of the rotary converter, 

 Beckmann's rearrangement, determination of the solubility 

 ■of a given substance by means of Pulfrich's refractometer, 

 ■and dynamometer car experiments. The varied nature of 

 this list affords an indication of the large amount of atten- 

 tion that is now being devoted to research work in pure 

 science in Japan. Of the papers dealing with applied 

 science, that by Mr. D. Saito on the refining of copper 

 should be carefully studied by metallurgists. The author 

 Tias made a systematic study of the process of dry refining, 

 his investigations having been made upon the blister copper 

 from the Beshi mine in Japan. The blister copper, which 

 is comparatively pure, is refined in a reverberatory furnace 

 using coal as fuel, and the author finds that the greater 

 part of the impurities is oxidised in the earlier stages of 

 refining. Thus, after the end of the first rabbling, the 

 copper contains only 003 per cent, of iron and 0003 per 

 •cent, of sulphur, whilst after the second rabbling the iron 

 contents remain almost unchanged, and the copper is prac- 

 tically free from sulphur. If the copper could be re-melted 

 more quickly and the third rabbling period dispensed with, 

 there would be a great economy in fuel and labour. The 

 effect of the first poling is so great that the second poling 

 seems unnecessary, or at least could be shortened. 



Mr. J. W. Patterson, of the Technical College, West 

 Hartlepool, has sent us two very successful colour photo- 

 graphs of rock sections taken between crossed Nicols. 

 They were taken by the Lumi^re autochrome process, the 

 illuminant being an electric arc light, and reproduce very 

 satisfactorily the interference colours given by plagioclase 

 felspar, augite, and olivine. It is obvious that this places 

 in the hands of teachers of petrology and geology a very 

 useful aid for illustrating lectures. Autochrome photo- 

 graphs are most successful with slides which are fairly 

 transparent, and should be inspected in a strong light. 

 Mr. Patterson has also obtained photographs ef the inter- 

 ference figures yielded by uniaxial and biaxial crystals in 

 convergent polarised light. Some weeks ago we saw a 

 series of colour photographs of this kind exhibited in 

 Kelvingrove Museum, Glasgow. Three-colour photo- 



NO. 1 992, VOL. 77] 



graphic plates appeared two years ago as illustrations of 

 an annual report of the Geological Survey of the Trans- 

 vaal, and about the same time Prof. E. J. Garwood showed 

 some colour lantern slides of rock sections at the Geo- 

 logical Society, which were the finest of their kind we 

 have seen. They were taken by the Sanger Shephard 

 process, we believe. Undoubtedly methods of colour 

 photography will prove to be of great use in reproducing 

 microscopic slides, not only of rocks, but also of other 

 subjects. 



In Symons's Meteorological Magazine for December, 

 1907, Mr. W. Ellis, F.R.S., formerly superintendent of 

 the magnetical and meteorological department of Green- 

 wich Observatory, gives a useful summary of Greenwich 

 air-temperature observations published for the sixty-five 

 years 1841-1905. The lowest mean daily temperature, 

 37°.47, is reached on January 12 ; after February 12 the 

 rise towards spring begins, receiving, however, a slight 

 check in the last week of .'Vpril. The highest mean daily 

 temperature, 64°-oi, is reached on July 15; after 

 ."August 13 there is a continuous fall to the minimum 

 of winter. The mean annual temperature is 49°-s6 ; the 

 warmest year is 52°, in 1868, and the coldest 46°.28, in 

 1879. The mean monthly temperature is 38''-6, in January, 

 and 62°'7, in July. The absolute highest reading was 

 97°i, on July 15, 1881, and the absolute lowest 4°, on 

 January 9, 1841. The observations give no information 

 on secular change, for which purpose a much longer period 

 than sixty-five years is necessary ; there are several interest- 

 ing differences shown by dividing the series into groups, 

 but Mr. Ellis states that these are clearly due to accidental 

 causes. Nor is any influence traceable to sun-spot varia- 

 tion, which the author considers is practically insignificant 

 in all questions of weather change. 



At the meeting of the Royal Academy of Sciences of 

 Amsterdam of October 26, 1907, an interesting paper by 

 Dr. E. Van Everdingen was read on the relations between 

 mortality of infants and high temperatures. It had been 

 previously pointed out in a paper published by the Statis- 

 tical Bureau of .Amsterdam that a distinct maximum in 

 the mortality of children under one year of age existed in 

 the summer months, but an endeavour to find any con- 

 nection between this maximum in various places and the 

 monthly means of temperature only led to a negative 

 result, although it was still thought probable that the 

 mortality was due to fluctuations of temperature. Follow- 

 ing up this idea. Dr. Van Everdingen tabulated the 

 meteorological data for various places in several different 

 ways, one of which was to extract the days on which the 

 temperature exceeded 25° C. between the middle of one 

 month and that of another. In this case the agreement 

 between the deviations of mortality and the number of hot 

 days was so satisfactory that little doubt remains that the 

 high temperatures must be considered as the cause of the 

 increased mortality. The author expresses the hope that, 

 with the aid of other temperature limits and possibly with 

 other methods of grouping the observations, those com- 

 petent in medical matters will feel inclined to trace the 

 more direct relations of the phenomenon. 



In the Physical Review for November, 1907, Mr. VV. P. 

 White, of the geophysics laboratory of the Carnegie 

 Institution at Washington, makes a thorough examination 

 of the potentiometer methods of measuring temperature by 

 means of the resistance thermometer or the thermoelectric 

 junction, in order to determine the best arrangement to 

 use in melting-point measurements. He comes to the con- 



