282 



NATURE 



[April 29, 1920 



fortunate absence of much wind and by coming after 

 midnight. The track of serious damage rarely ex- 

 ceeded ^ mile in width and was i6^ miles long, the 

 hail beginning at 1.55 a.m. west of Holmwood Station 

 and ending near North Bromley Station, twenty-two 

 miles to the north-east, at about 2.30 a.m. (true time). 



Edinburgh. 

 Royal Society, March 15.— Prof. F. O. Bower, 

 president, in the chair.-^Capt. T. Bedford Franklin : 

 The effect of weather changes on soil temperatures. 

 In comparison with the variations of surface tem- 

 perature, the regular pulsations of temperature in 

 the soil follow well-known laws for amplitude 

 and retardation according to depth, but in these 

 regular pulsations there are fluctuations which 

 occur according to the weather and the state 

 of the soil. If the ratio of the ranges of tem- 

 perature at the 4-in. depth and the surface be 

 taken as the standard for measuring the heat transfer 

 in the soil, it is found that in a light loam soil this 

 range-ratio varies between 0-19 and 0-42 when active 

 percolation is not taking place in the soil, and between 

 0-42 and 0-85 when rain is actually falling or during 

 those long-period weather changes associated with the 

 passage of depressions north of these islands, whether 

 rain falls or not. Heat transfer in soil is thus carried 

 out by both conduction and percolation, and a sandy 

 soil that allows free percolation, with consequent high 

 values of the range-ratio, will heat up quicker in 

 spring than a clay soil which takes up and parts with 

 water only sluggishly. Apart from percolation, the 

 high values of the range-ratio in the south-westerly 

 cyclonic type of weather are particularly valuable in 

 causing rapid rises of soil temperature in spring. A 

 surface layer of frozen soil protects the lower depths 

 from rapid changes of temperature; an average sur- 

 face temperature of —10° C. would be necessary to 

 freeze ordinary soil to a depth of 4 in. in one night. 

 Snow is an even more efficient protection ; in Novem- 

 ber, 1919, the air temperature above 4 in. of snow fell 

 to — 15° C. without freezing the soil surface or caus- 

 ing any appreciable fluctuation in the temperature 

 4 in. beneath the surface of the soil. — D. Ferguson : 

 Geological observations in the South Shetlands, the 

 Palmer Archipelago, and the Danco coast, Graham- 

 land. — G. W." Tyrrell : A contribution to the petro- 

 graphy of the South Shetlands, the Palmer Archi- 

 pelago, and the Danco coast, Grahamland. — H. H. 

 Thomas : Petrographical notes on rocks from Decep- 

 tion Island and Roberts Island (South Shetlands), the 

 Danco coast, and adjacent islands,, Grahamland. 

 These papers contained a great many new facts re- 

 garding the geology and petrography of the rocks in 

 the islands named lying to the south of South America. 

 In a broad sense, the geological arrangement mi)?ht be 

 described as a mirror reflection of the arrangement 

 on the South American coast, and it was probable 

 that the two sets of strata were connected by an arc 

 passing east, and then bending round to the south and 

 to the west, but there was no evidence in support of 

 Suess's theory that this arc extended far to the east 

 so as to include the South Georgian group. — Miss 

 C. W. N. Sherrlff : A class of graduation formulae. — 

 Prof. L. Becker : The daily temperature curve. In 

 this paper the author developed a new mathematical 

 method of treating the variation of temperature, and 

 illustrated it by a discussion of a forty years' photo- 

 graphic record of temperature in Glasgow. 



Dublin. 

 Royal Dublin Society, March 23.— Dr. F. E. Hackett 

 in the chair. — Prof. James Wilson : The application of 

 the food-unit system to the fattening of sheep. A 

 NO. 2635, VOL. 105] 



summary of experiments in fattening sheep was pub- 

 lished by Mr. Herbert Ingle in the Transactions of the 

 Highland and Agricultural Society of Scotland for 

 1910 and 191 1. From this it is evident that the sheep 

 differs from the bullock by being better able to con- 

 sume roots, and, therefore, less dependent upon hay 

 and straw. The sheep is also considerably more econo- 

 mical as a producer of human food; for, while the 

 well-fed bullock of average size — say 9 cwt. — needs 

 from six to seven food units to produce a pound of 

 beef — a food unit is the quantity of any other food 

 which would have the same producing capacity as a 

 pound of barley — a well-fed sheep of average size — 

 say 120 lb. — produces a pound of mutton on five to 

 six food units. 



Royal Irish Academy, April 12.— Mr. W. G. Strick- 

 land, vice-president, in the chair. — E. Heron-Allen and 

 A. Earland : An experimental study of the Foramini- i 

 feral species Verneuilina polystropha, Reuss, and some 1 

 others, being a contribution to a discussion on " The 

 Origin, Evolution, and Transmission of Biological 

 Characters." In this paper the authors describe 

 normal and monstrous forms of Verneuilina poly- 

 stropha. The species exhibits characteristic di- 

 morphism in a long, tapering test which is megalo- 

 spheric, and a short test which is microspheric, but 

 in the dwarf variety, pusilla, the tapering test is 

 microspheric. Observations on the selection by Ver- 

 neuilina of fragments of heavy minerals, by mixing 

 crushed gems with the sand in the experirnental 

 tanks with which the authors worked, are described. 

 Variation in the shells of Massilina secans is also 

 described irt detail, one of the most remarkable 

 "monsters" being a perfectly twinned specimen which 

 had added a curved tube at the junction of the shells 

 to form a common aperture ; the whole of this 

 abnormal shell was chitinous. In another case a shell 

 was entirely chitinous except the terminal chamber, 

 which was perfectly and normally calcareous. Further 

 instances of shells combining the characters of two 

 distinct genera — such as have been recorded in former 

 papers by these authors — are given, and the opinion 

 is expressed that the accepted systems of classification 

 of the Foraminifera, founded as they are on the shape 

 and material of the test, must be regarded as largely 

 artificial and unscientific. 



Paris. 



- Academy of Sciences, March 29. — M. Henri Deslandres 

 in the chair. — G. Bigourdan : The observatory of J. S. 

 Bailly at the Louvre. — Prince Albert de Monaco : 

 Stray mines in the North Atlantic. A chart is given 

 showing the positions of sixty-eight mines located 

 between November 7, 1918, and February 9, 1920. 

 The predictions by the author in two earlier com- 

 munications have been fully confirmed. — A. Rateau : 

 Some considerations on flight at very high altitudes 

 and on the use of a turbo-compressor. An adverse 

 criticism of a recent communication on the same sub- 

 ject by M. Villey. — P. Vuillemin : The growth of fungi 

 discovered in the human nail by Louis Jannin. — 

 G. Julia : Families of functions of several variables. — 

 H. Mineur : Discontinuous solutions of a class of 

 functional equations.— B. de Fontviolant : The strength 

 of circular arches. — F. Kromm : A star with a large 

 proper motion. The star B.D. + 9-2636°, 9-1 magni- 

 tude, has an annual proper motion of nearly a second 

 of arc. — G. Sagnac : Newtonian light radiation and 

 the zones of silence in damped wireless telegraphy 

 signals. — M. Pauthenier : The ratio of the absolute 

 retardations in Kerr's phenomenon. — M. Lemarchands : 

 Study of the reactions of the metallurgy of zinc. 



