3IO 



NA TURE 



[July -i^o, 1891 



the local weather predictions of the Blue Hill Observatory ; 

 M. W. Harrington, on weather prediction in the States and its 

 improve nent, together with several other similar papers. — The 

 zodiacal light as related to the aurora, by O. T. Sherman. The 

 author gives tables and curves constructed from a large number 

 of observations, showing (i) the relative elongation of the 

 zodiacal light, from observations taken in March, from 1801-86 ; 

 (2) corrections to the earth's calculated longitude, being that part 

 of the amount by which the observed position varied from the 

 calculated, which is probably due to zodiacal light ; (3) Fritz's 

 auroral numbers for Europe south of the polar circle ; and (4) 

 his relative numbers for Europe. The conclusions drawn from 

 the tables are that from 1806-27 there was no observation of the 

 zodiacal light, slight and irregular variation of the earth's motion, 

 and slight and irregular auroras. For the next fifty years each 

 period of elongation of the zodiacal light corresponded with a 

 maximum acceleration of the earth's motion, and a minimum in 

 the aurora. And further, that at the time when the zodiacal 

 light was beyond the earth's orbit, the auroras were few and 

 diminished in number. — Farwell's rainfall scheme. This article 

 {which is unsigned) states that Senator Farwell carried a Bill 

 through the last session of Congress, for testing the possibility of 

 the artificial production of rain by means of explosions. The 

 experiments, which are soon to be tried, are intrusted to the 

 Agricultural Department ; the officials, however, are said to have 

 little confidence in the success of the experiment. Mr. Fernow, 

 Chief of the Division of Forestry, gives a long report upon the 

 proposal, together with a summary of the literature of the subject. 

 American Tournal of Science, July. — The solar corona, an 

 instance of the Newtonian potential function in the case of 

 repulsion, by Prof. Frank H. Bigelow. This is a continuation 

 of the author's researches into the laws which regulate the 

 development of the various coronal forms. — Newtonite and 

 rectorite, two new minerals of the kaolinite group, by R. 

 N. Brackett and J. Francis Williams. Taking the composition 

 of kaolin as Al203,2Si02,2H20, the following series of hydrous 

 silicates of alumina may be derived by eliminating or introducing 

 a molecule of water : — 



Percentage Composition. 



A1203 



(1) AI2O3, 2Si02, H2O ... 42-52 



(2) AljOg, 2Si02, 2H2O ... 39-57 



(3) AI2O3, 2Si02, 3H2O ... 36-98 



(4) AI2O3, 2SiO.,, 4H2O ... 34-72 



From the facts and considerations stated in the present paper it 

 appears probable that three members are known out of the four 

 in the above series, viz. (i) rectorite, (2) kaolin and members 

 of the kaolinite group, and (4) newtonite.— On the intensity of 

 sound ; /ii. the energy used by organ-pipes, by Charles K. 

 Wead. From the results of experiments with different organ-stops 

 out, it appears that no exact conclusion can be drawn from the 

 loudness of the sound as to the relative quantity of wind 

 required to blow pipes of different construction ; thus, the soft 

 Dulciana stop of the organ upon which the experiments were 

 performed took more than half as much wind as the comparatively 

 loud Open Diapason, whilst the pipes of the Trumpet stop 

 required less energy than any others sounding the same note. 

 The results obtained in the case of different pipes of the same stop 

 indicate that the volume of air used per second, and therefore 

 the energy expended per second, varies as the ^-power of the 

 wave-length of the note, or inversely as the f-power of the 

 vibration-ratio. — New analyses of astrophyllite and tscheffkinite, 

 by L. E. Eakins. The analyses give R"4R'4Si(Si04)4 as the 

 general formula for astrophyllite. This agrees with that found 

 by Brogger from a discussion of analyses by Backstrom and 

 Konig. Tscheffkinite does not appear to be a mineral in any 

 strict construction of the word, but merely a mixture, — The 

 minerals in hollow spherulites of rhyolite from Glade Creek, 

 Wyoming, by J. P. Iddings and S. L. Penfield. The authors 

 find that in the rhyolite investigated fayalite occurs in association 

 with abundant quartz of a peculiar development, as the result of 

 the mineralizing action of vapours in the cooling acid lava. In 

 certain hollow spherulites the fayalite is replaced by hornblende 

 and biotite. — Bernardinite : is it a mineral or a fungus ?, by 

 Joseph Stanley Brown. From Mr. Brown's examination it 

 appears that the mineral resin from San Bernardino County, 

 California, described by Prof. Stillman in the American Journal 

 twelve years ago, is the fungus Polyporus officinalis, Fries. — 

 Development of Bilobites, by Dr. Charles E. Beecher. — 



Gmelinite from Nova Scotia, by Louis V. Pirsson. The optical 

 characters, cleavage, and chemical composition of this mineral 

 have been studied. The result of the crystallographic work points 

 to a distinct difference between it and chabazite, but with regard 

 to twinning and chemical constitution the two appear to be 

 identical. Indeed, gmelinite seems to bear much the same 

 relation to chabazite that enstatite does to hypersthene. — Analyses 

 of kamacite, tcenite, and plessite from the Welland meteoric 

 iron, by John M. Davidson. The conclusion is arrived at that 

 in the Welland siderolite only two distinct nickel-iron alloys 

 occur, viz. kamacite and tcBnite, and that the so-called plessite 

 is merely thin alternating lamellae of the two. 



American Journal of Mathematics, vol. xiii., No. 4. — In 

 this number J. Perrott's " Remarque au sujet du theoreme 

 d'Euclide sur I'infinile du nombre des nombres premiers" is 

 continued from No. 3, and concluded ; the author promising a 

 further article on " L'application du precede du geometre grec i 

 d'autres cas de la proposition de Lejeune Dirichlet." — The 

 following papers also appear : — Ether squirts, by Karl Pearson, 

 an attempt to specialize the form of ether motion which forms 

 an atom. The main portion of the paper is devoted to an 

 investigation of inter-atomic and inter-molecular forces. — On the 

 matrix which represents a vector, by C. H. Chapman. The 

 fundamental idea is that the linear and vector function of a vector 

 is simply the matrix of the third order. — Sur une forme nouvelle 

 de I'equation modulaireduhuitiemedegre, parF. Brioschi. — The 

 index to vol. xiii. is appended to this number, which concludes it. 



SOCIETIES AND ACADEMIES. 



Edinburgh. 



Royal Society, July 6. — The Hon. Lord McLaren, Vice- 

 President, in the chair. — Mr. John Aitken read a paper on the 

 solid and liquid particles in clouds (see p. 279, July23). — Prof.Tait 

 communicated a paper by Prof. Chrystal on a demonstration of 

 Lagrange's rule for the solution of the linear partial differential 

 equation, with some historical remarks on defective demonstra- 

 tions hitherto current. Prof. Chrystal's proof is purely analytical. 

 Prof. Tait remarked that, on quaternionic principles, the problem 

 may be regarded as follows. Let the equation be 



P/ + Q? = R. 

 where P, Q, and R, are given functions of x, y, and z, and/, q, 

 represent respectively the quantities dzjdx, dzjdy. By the in- 

 troduction of a new variable, «, this may be put into the form 



P^ + Q^ 

 ^dy 



dx 



R^« = o. 



dz 



NO. II 35, VOL. 44] 



But dujdx, dujdy, dujdz, are proportional to the direction 

 cosines of the normal to the surfaces = c, and therefore P, Q, R 

 are proportional to the direction cosines of a tangent line to 

 u = c. Hence we deduce, as the equations of a curve which 

 lies wholly on the surface, 



dx _ dy _ dz 

 P ~ Q ~ R ■ 



The integrals of these equations are known to have the form 

 V = a, w = fi, where a and fi are arbitrary constants. The 

 intersections of these surfaces fill space with a set of lines, and 

 the problem is to find a single general set of surfaces upon which 

 these lines will lie. Their equation is v =f{w), where f is an 

 arbitrary function. It is therefore the integral of the given 

 differential equation. — Prof. Tait read the fifth part of his 

 paper on the foundations of the kinetic theory of gases. He 

 has applied his expressson for the isothermals of a liquid and its 

 vapour to the case of ethyl oxide. The results are in remarkable 

 accordance with the direct observations of Drs. Ramsay and 

 Young. He has also applied the virial method to systems of 

 doublets, triplets, &c. The close correspondence of the results 

 calculated from his formula with Andrews's and Amagat's ob- 

 servations on carbonic acid was somewhat surprising when it 

 was considered that the theoretical results were deduced on the 

 assumption of smooth, hard, spherical molecules, while the 

 molecule of carbonic acid is very complex. In the present part 

 of his paper. Prof. Tait shows that, from the manner in which 

 the (approximate) virial equation is formed, no term depending 

 on internal actions in molecules themselves can appear in it 

 when the number of molecules is sufficiently large. He also 

 discusses the mechanism of equilibrium between liquid and 



