214 



NA TURE 



[July 2, 190: 



was the third South African species of the genus Precis in 

 which Mr. Marshall had produced incontrovertible evidence 

 of the specific identity of forms widely separated in colours, 

 patterns, shape, relation of upper- to under-side, &c., and 

 even instinct, including the selection of a particular type 

 of country. The president also showed a small series of ants, 

 part of a much larger collection made by the late W. J. 

 Burchell in Brazil between the years 1825 and 1830. Con- 

 sidering their great age, the specimens were wonderfully 

 well preserved, and were accompanied by remarkably exact 

 and detailed data, and, in many cases, interesting notes on 

 habits, instincts, &c. — Mr. O. E. Janson communicated a 

 paper on the genus Theodosia and other Eastern Goliath- 

 ides, with descriptions of some new species. — Colonel C. 

 Swinhoe communicated a paper on new genera and species 

 of the family Lymantriidic in the National Collection. — 

 Mr. G. W. Kirkaldy communicated a memoir on the Rhyn- 

 chota collected by Dr. Arthur Willey chiefly in Berara 

 and Lifu. — Prof. E. B. Poulton gave an account of ex- 

 periments in 1893, 1894, and 1896 on the colour relation 

 between certain lepidopterous larvjE and their surround- 

 ings, and especially the effect of lichen-covered bark upon 

 Odontopera bidcntata and Gastropacha qucrcifoJia. 



Mineralogical Society, June 9. — Dr. Hugo Muller, 

 president, in the chair. — Mr. H. F. Collins gave an account 

 of a remarkable mass of wollastonite with associated 

 minerals which occurs at Santa F6, State of Chiapas, 

 Mexico. This mass of nearly pure wollastonite covers an 

 area of 400 yards by 160 yards, and reaches to a depth of 

 more than 300 feet ; it is surrounded on all sides by granite, 

 felsite, and other igneous rocks, and is separated by a mile 

 from the nearest limestone. Near the outskirts of the mass 

 occur extremely large crystals of wollastonite, most of 

 which have been partially or entirely converted into quartz 

 or semi-opal. Here are also found masses of garnet and 

 of workable copper ores containing gold and silver. The 

 author exhibited and described specimens of wollastonite, 

 bornite in wollastonite, bornite in calcedony, gold-bearing 

 linnaiite, idocrase rock, and a remarkable intergrowth of 

 bornite and galena resembling graphic granite. — Prof. 

 H. A. Miers described the results he has obtained from 

 the observation of the growth of crystals by a new method. 

 The method consists in tracing the changes of angle upon 

 a crystal during its growth by measuring it at intervals 

 by means of a specially devised inverted goniometer, with- 

 out moving it from the solution in which it is growing. It 

 was found that an octahedron of alum yielded invariably 

 three images for each face, so that the crystal had really 

 the form of a very flat triakis-octahedron. Similar observ- 

 ations on other crystals lead to the conclusion that the faces 

 of a crystal are in general not faces with simple indices, 

 but vicinal planes slightly inclined to them, which change 

 their inclination during the growth of the crystal. By 

 determinations of the refractive index of the solution by 

 means of total reflection within the crystal, it was found 

 that in each case the liquid in contact with the growing 

 crystal is slightly supersaturated. 



Mathematical Society, June 11. — Prof. H. Lamb, presi- 

 dent, in the chair. — The president announced that, after the 

 conclusion of the current volume, some changes would be 

 made in the form of publication of the Proceedings, the 

 chief being an increase in the size of page and type. — The 

 following papers were communicated : — Major P. A. 

 MacMahon, The application of quaternions to the algebra 

 of invariants. — Mr. G. B. Mathewrs, Jacobi's construction 

 for quadric surfaces.— Mr. H. W. Richmond, Automorphic 

 functions in relation to the general theory of algebraic 

 curves. The object of the paper is to extend to curves in 

 space of three or more dimensions the methods which have 

 been developed by Poincard and Humbert for the parametric 

 representation of plane curves by means of automorphic 

 functions. Curves are classified by their genus (or de- 

 ficiency), their order and the number of dimensions of the 

 space in which they lie, and the properties of all the curves 

 in a class can be inferred from those of particular members 

 of the class. The genera i, 2, 3, 4, 5 are discussed in 

 detail. — Prof. L. E. Dickson, Addition to the paper on 

 four known simple groups of order 25920. — Prof. A. C. 

 Dixon made an informal communication On a method of 

 introducing the logarithmic function by means of geo- 

 metrical properties of conies. 



NO. 1757, VOL. 68] 



Edinburgh. 

 Royal Society, June i. — Dr. Munro in the chair. — 

 Mr. George Muirhead, commissioner for the Duke of 

 Richmond and Gordon, read a paper on the effect of 

 temperature on the taking of salmon by rod and fly on the 

 River Spev at Gordon Castle. From a careful examination 

 of the full statistics which had been kept for a number of 

 years, and a discussion of them in the light of various 

 possible meteorological and climatic causes, the conclusion 

 come to was that the number of salmon caught on a day 

 was determined, to a large extent, by the variation of 

 temperature during that day, the greater the variation of 

 temperature the smaller the catch. — Dr. W. Peddie read 

 a paper on the theory of colour vision. The theories which 

 give the best account of the facts of colour vision and 

 colour blindness are the Young-Helmholtz theory and 

 Hering's theory. Both are trichromatic theories, and, apart 

 from physiological or anatomical questions, both can, by 

 proper choice of fundamentals, be made to give a good 

 account of the main facts. The facts of one-eyed colour 

 blindness show that, on the Young-Helmholtz theory, colour 

 blindness rnust be regarded as due to fusion of at least two 

 fundamental sensations. But the curves of one sensation, 

 determined by observations on different eyes, differ con- 

 siderably among themselves. This indicates that a 

 broader basis for the theory may be desirable. This may 

 be sought for in a tetrachromatic theory. But any such 

 theory must explain the possibility of trichromatic repre- 

 sentation of all colours. The theory proposed assumes two 

 pairs of complementary stimulations, say, R,G, and G,,V. 

 In this respect it has a resemblance to Hering's theory. 

 But whereas, in Hering's theory, stimulation of one 

 member of a complementary pair means no stimulation of 

 the other member, in the proposed theory equal stimula- 

 tions of two such members gives white. It is shown that 

 four sets of equivalent trichromatic fundamentals must 

 exist. Assuming Helmholtz's fundamentals as such a set, 

 the four mathematically possible sets of tetrachromatic 

 equivalents, of which only one can exist physically, are 

 deduced ; and it is found that one of these does suit the 

 known facts of colour vision and colour blindness. 

 Choosing this set, the other three trichromatic equivalent 

 sets (Helmholtz's being the fourth) can be deduced. The per- 

 ceptibility curves (ordinates being differences of wave-length 

 just appreciable to the eye in the spectrum) found for one 

 of these sets is compared with that given by Helmholtz's 

 set. The comparison is found to be very satisfactory. The 

 nature of the tetrachromatic set shows that colour blindness 

 must be regarded as due to suppression of one comple- 

 mentary pair of sensations, while variations in normal eyes 

 are due to partial suppression. In this way the sensation 

 curves for different eyes may have greater fixity as regards 

 form, depending on wave-length, than in the trichromatic 

 set. This result is desirable if the sensation curves are to 

 be regarded as really corresponding to physioloeical stimu- 

 lation produced photochemically or photoelectrically. A 

 simple theory of such stimulation is given and shown to 

 lead to the required form of relation between the four 

 fundamentals. 



June 15. — The Hon. Lord M'Laren in the chair. — Dr. 

 Home and Dr. Peach read a paper on the Canonbie Coal- 

 field : its geological structure, &c. Though of limited ex- 

 tent, this coalfield has aroused considerable interest owing 

 to the important series of plants obtained from the beds 

 and to the questions bearing on the correlation of the 

 Carboniferous rocks of the Scottish border with those of 

 the north of England and centre of Scotland. About twenty 

 years ago it was assigned by the Geological Survey to the 

 Calciferous Sandstone series. At that time, however, great 

 ditTiculty was felt in correlating the subdivisions of the 

 Carboniferous rocks as there developed with those in the 

 midland valley of Scotland, owing to the marked variation 

 in some of the groups from the normal Scottish types. 

 The palajontological evidence then obtained Was not in 

 accord with these conclusions, for the plants seemed to show 

 that the coalfield really belonged to the true Coal-measures. 

 Last year the Canonbie area was reexamined by the Geo- 

 logical Survey and Mr. Kidston. Deep bores have been 

 sunk in recent years by His Grace the Duke of Buccleuch, 

 and these also have furnished important geological evidence. 

 By means of horizontal sections it was shown that the 



