May 29, 1919] 



Mj^URE 



259 



settling of the particles. Application of Stokes's law 

 to this net settling gives a value for the sijie of the 

 particles very closely agreeing with that obtained by 

 the counting method [e.g. 22 x 10- * cm. anH 

 1-7x10-'' cm.), even though values to hand are taken 

 from old observations made when this net settling 

 was not appreciated and not closely observed. Ex- 

 periments are now in hand to determine this settling 

 very exactly.— W. E. Curtis : The value of the Ryd- 

 berg constant for spectral series. 



Royal Meteorological Society, May 21.— Sir Napier 

 Shaw, president, in the chair, — Capt. C. J. P. Cave 

 .Hid J. S. Dines : Further measurements on the rate 

 of ascent of pilot-balloons. The paper discusses ex- 

 periments made, in continuation of previous work, on 

 the rate of ascent of pilot-balloons measured in a 

 V losed building. The building used in the present 

 (.ase was the Royal Albert Hall, which is very suit- 

 able for the purpose, inasmuch as a clear height of 

 40 metres- is available from floor-level to the grid 

 at the centre of the domed roof. The formula in 

 general use in this country for^the rate of ascent 

 is, rising velocity V = g ^/L/ x-'W+L, where L = the 

 free lift and W the dead-weight of the balloon, and 

 q is a constant the value of which is to be determined 

 under different conditions. It had previously been 

 suggested that the value of q varied with different 

 degrees of loading of the balloon. Attention was 

 directed to this question, and quantitive results were 

 obtained. Measurements were also made with a 

 candle-lantern of the pattern used for night ascents 

 hung below the balloon. It was found that this pro- 

 duced no effect upon q. In timing the rate of ascent 

 in closed buildings a fine thread has generally been 

 attached to the neck, and has been drawn up from 

 the floor as the ascent proceeded. In the present case 

 experiments which were made with and without such 

 a thread showed that some correction is necessary 

 where a thread is used. The general results con- 

 firmed the value 5 = 84, which is used at the present 

 time, for balloons of the size generally adopted for 

 pilot-balloon work. This value gives velocities in 

 metres per minute when lift and dead-weight are 

 expressed in grams. — J. Edmund Clark and H. B. 

 Adames : Report on the observations for the pheno- 

 logical year, December, 1917, to November, 1918." 

 The excessive cold of December, 19 17, was followed 

 by three mild months, February in particular. Hence 

 by .April i blackthorn was in most parts blooming, 

 whereas after the very cold early months of 19 17 

 the mean date was thirty-five days later than in 19 18. 

 Rarely has the farm and garden promise at this date 

 been so satisfactory. Then came the mid-April bitter 

 weather, disastrous to the opening fruit-tree buds, 

 and a continuation of summer drought and coolness 

 continued the prejudicial conditions. A genial August 

 greatly favoured the earlier harvesting districts, but 

 the excessive wet in September caused damage and 

 loss elsewhere. The whole autumn was cool, but com- 

 parative dryness in October and November helped 

 finally in the harvesting of nearlv average field crops. 

 Potatoes gave a record for acreage and yield per acre, 

 but after storage there was serious loss from disease. 

 [ The migrant records support the interesting weather 

 f relationships shown by "the other tables. The .\pril 

 cold delayed the appearance of the sixteen earlier 

 birds two or three days more than the other ten. 

 The isophenal lines on the map indicate the districts 

 where the plants of Table III. blossomed simul- 

 taneously. Their course shows the marked influence 

 of elevation. On the Scime map are also shown the 

 isotherms for the first half of the year, and a com- 

 parison of these with the isophenes is a matter of 

 considerable interest. 



NO. 2587, VOL. 103] 



Paris, 

 Academy ol Sciences, May 5.--M. Leon Guignard in 

 the chair.— '.\. Lacroix and A. de Gramont : The 



presence of boron in some natur;U basic silico- 

 aluminates. A spectroscopic examination of saphirine, 

 kornerupine,.and grandidierite showed that boron had 

 been overlooked in the analyses. In the order named 

 there were present 0-75 per cent., 359 per cent,, and 

 2-8 1 per cent, of boric anhydride. The boron may be 

 considered as replacing aluminium isomorphically in 

 these minerals. The results of the spectroscopic 

 examination of other minerals for boron are given. — 

 H. Deslandres : Remarks on the constitution of the 

 atom and the properties of band spectra. A discus- 

 sion of the formulae expressing the general structure 

 of band spectra, with reference to the various hypo- 

 theses on the composition of the atom.— C. Depiret : 

 An attempt at the general chronological co-ordination 

 of the Quaternary epoch.- — Ed. Imbeaux : The 

 navigable waterways of Alsace and Lorraine. An 

 account of the actual position of water-carriage in 

 these provinces and the modifications which they will 

 want in the immediate future to meet the industrial 

 requirements, including the transport of coal from 

 the Saar basin, oil from Pechelbronn, iron from the 

 Lorraine mines, potash from Mulhouse, soda salts, 

 cements, and other industrial products. — M. De- 

 fourneaux : Some properties of electro-spherical poly- 

 nomials. — G. Julia : Uniform functions with an iso- 

 lated essential singular point.- — G. Guillaumin : Cer- 

 tain particular solutions of the problem of sandy flow 

 where the massif considered comprises two regions 

 governed by different laws. — Ed. Urbain and C. Seal : 

 The decomposition of dielectric liquids surrounding an 

 arc. It was necessary to use metallic electrodes in 

 these experiments, as the separated carbon then re- 

 mained in suspension in the liquid. If the liquid is 

 maintained at 1/;°, the decomposition products are 

 different from those obtained when the liquid is 

 allowed to boil. Some particulars of experiments with 

 tin tetrachloride, titanium tetrachloride, carbon tetra- 

 chloride, some hydrocarbons, and ketones are given. — 

 A. C. Vournasos": The normal nitrides of nickel and 

 cobalt. If nickel cyanide is heated with nickel oxide 

 to a temperature not exceeding 1000° C, the only 

 products of the reaction are carbon monoxide, nitrogen, 

 and metallic nickel. If, however, these two sub- 

 stances are rapidly heated to more than 2000° C, the 

 products are carbon monoxide and nickel nitride, 

 NiaNo. The corresponding cobalt nitride is formed 

 in a similar reaction.^ — \. Kling and R. Schmutz : The 

 estimation of traces of carbonyl chloride in air. The 

 air is passed through aqueous aniline, and the di- 

 phenylurea formed by the phosgene determined either 

 by weighing or by conversion into ammonia. For 

 quantities varying from 022 to 044 milligram of 

 COCl, per litre the error averaged 5 per cent, of the 

 amount present. — M. Picon : The action of the mono- 

 sodium derivative of acetylene upon some primary 

 alkvl iodides with branched chains. — E. Fleury : The 

 signification and role of lapiesation in the disaggrega- 

 tion of granitic rocks in Portugal. — G. Gullbert : The 

 prediction of barometric variations. — P. Thl^ry : New- 

 observations on the system of geological accidents 

 called the Faille des Cevennes. — L. L^ger and E. 

 Hesse : .\ new parasitic Coccidium of the trout. This 

 new species, for which the name Goussia truttae is 

 proposed, has been observed in wild trout from the 

 neighbourhood of Grenoble. A full description is 

 given. — S. Stefanescu : The co-ordination of the morpho- 

 logical characters and of the movements of rhe molars 

 of elephants and mastodons. — R. Fosse : The simul- 

 taneous oxidation of blood and £?lucose. L'rea is pro- 

 duced bv this oxidation. — G. Bertrand and Mme. M. 



