62 



WATURE 



[Nov. 20, 1879 



jhe obscurity attaching to the subject of Abyssinian ethno- 



ogy: 



IAghagha, prov. Agha 

 meder. 

 Hamra, prov. Lasta, ! 

 Tigre". 

 Falasha, mainly in Sem- 

 ji yen. 

 g Kwara, W 

 $) Aghaum 

 :3 •, Khamant, 



nd N. 



Semites. 



Tigr<?, N. and E. of 

 River Takazze\ 



Samhar. on coast near 

 Massowah. 



Shoho, S.W. of Mas- 

 sowah. 



Menia, N. ofHama- 



Habab, 



Bediuh, { N. SN.E. 

 TakwtS, / frontiers 

 Marea, Tigre. 



1 Barea, J 



i,l N. 



Table of Abyssinian Races 

 1 Mixed. 



(Hamites& Semites.) 

 ;Bogos, extreme N.E. 

 ' G-ngas,aboutGojam. 

 Guragwe, extreme S. 

 "Cunama, N.W. to- 

 wards Taka. 

 of Shea, S.E. comer. 

 Ainhara, between the 

 chiefly in I PJakazze and 



„ I Dembea. I |_Abai. 



<; I Fighen.S.W. from Lake King Theodore. 

 , Tsana. I 



Zalan. chiefly in N. Prince Alumayu. 



Amhara. 

 ! Witos, about Lake 

 I Tsana©. 



Of the languages three only are of any literary or 

 political interest : Glu'c, still surviving as the language of 

 the liturgy and Sacred writings, though scarcely under- 

 stood even by the clergy ; Tigrdi, its purest modern 

 representative, current throughout the kingdom of Tigre' 

 and generally north and east of the Takazzd; Amharna, 

 spoken with considerable dialectic variety in Amhara and 

 Shoa. All are written in a peculiar syllabic character 

 showing certain affinities to the Himyaritic rock inscrip- 

 tions of Marah and other parts of South Arabia. Am- 

 haric employs seven additional letters for sounds not 

 occurring in Ghez or Tigrai, making with the vocal 

 modifications a total of 249 distinct symbols. This was 

 the language of Prince Alumayu. A. H. Keane 



COLOUR-VISION AND COLOUR-BLINDNESS 



AS the notices of these subjects which have recently 

 appeared in Nature appear to me to do scant 

 justice to the received theory, will you permit me to call 

 attention to a portion of the evidence on which this theory 

 rests ? 



The Philosophical Transactions for 1S60 contain a 

 paper by Prof. Clerk Maxwell, in which actual measure- 

 ments are given of the quantitative relations between 

 various colours, some of the observations having been 

 taken by persons of normal vision, and others by a colour- 

 blind person. The instrument of observation consisted 

 of a species of spectroscope with three parallel slits, the 

 ■widths of these slits, and also the distances between them 

 being variable at pleasure. By this means three over- 

 lapping spectra are obtained, and any three spectral 

 colours can be mixed in any proportions. The observa- 

 tions showed that any four colours as presented to the eye 

 in a given spectrum are connected with each other by a 

 definite colour-equation, such as — 



3 A + 4B = 2C+6D, 

 which means that if the four colours A, B, C, D, as they 

 exist in the given spectrum, are increased in intensity 

 threefold, fourfold, twofold, and sixfold respectively, 

 and then mixed two and two, the mixture 3 A -f- 4B will 

 present exactly the same appearance as the mixture 

 2C+6D. This is only another way of saying that 

 colour as seen by normal vision contains three inde- 

 pendent variables, or requires three numbers for its 

 specification. Any three colours of the spectrum will 

 serve as the three specifying elements ; for example, if 

 we employ A, B, and C to specify D, the specification 

 will be — 



D*=\A +%B - $C. 

 Here we have one coefficient (that of C) with the negative 

 sign. The three primary colours are defined to be those 

 which will always have positive coefficients when they are 

 employed as the specifying elements. In plainer words, 

 all other colours can be exactly imitated by mixtures of the 

 primaries, whereas, in the above example, the colour D 

 cannot be imitated by a mixture of A, B, and C. 



The points of the spectrum at which the three primary 

 colours are found, will not necessarily be the points which 



most strongly excite the three elementary colour-sensa- 

 tions respectively. On the contrary, as a matter of fact, 

 the two extreme sensations (called by Maxwell the redand 

 the blue) are very feebly excited at the parts of the 

 spectrum where they are purest, namely, at the extreme 

 ends of the spectrum ; and the middle sensation, which is 

 largely adulterated with the other two even at the point 

 where it is purest (namely, at a point in the olive green, 

 which is, accordingly, one of the three primaries), has 

 not a maximum of intensity at this point, but increases 

 in intensity as the brightest part of the spectrum is 

 approached, and attains its maximum (for the solar 

 spectrum obtained with a flint glass prism), somewhere 

 between the fixed lines E and D. The determination of 

 the position of the middle primary in the spectrum, was 

 made with considerable precision in the paper referred 

 to ; but the faintness of the two extremities of the spec- 

 trum rendered wide slits necessary in examining these 

 regions, and thus introduced inaccuracy in determining 

 the positions of the two extreme primaries, which in later 

 publications Prof. Maxwell places at the very extremities 

 of the spectrum. 



The latter part of the paper of i860 consists of a post- 

 script containing observations made by a colour-blind 

 person. The colour-equations found by direct observa- 

 tion are given, and are shown to agree with the suppo- 

 sition that the observer' s vision was dichroic, the sensation 

 corresponding to the extreme red being absent. The 

 curves of intensity for each of the two elements in the 

 vision of the dichroic observer are given, side by side 

 with the three curves of intensity for the vision of a 

 trichroic observer, all these being directly calculated from 

 the observations, and the two dichroic curves appear to 

 be practically identical with two of the three trichroic 

 curves. 



Dr. Pole's objection to the received theory appears to 

 me to have no force except in so far as it is an objection 

 to a name. The colour which the colour-blind see in the 

 less refrangible half of the spectrum appears to be due 

 to the excitement of the middle one of the three elemen- 

 tary sensations of trichroic vision. Persons of normal 

 vision never get this sensation without large adulteration, 

 and hence ordinary language contains no appropriate 

 name for it. 



Prof. Hering's theory of colours, as expounded by Dr. 

 Pole (Nature, vol. xx. pp. 479, 480) seems inconsistent 

 with the fact (established by the observations of Prof. 

 Maxwell, Lord Rayleigh, and other competent observers) 

 that there is one definite colour-equation between any 

 four colours. For Prof. Hering's theory assumes four 

 elements of colour-sensation, R, G, B, Y, such that 

 R+ G = 0, B+ Y = o. 



It would follow that, with the help of the minus sign, all 

 colours could be specified in terms of R and B, and hence 

 by writing down the specifications of any three colours, and 

 employing the ordinary processes of elimination, a colour- 

 equation could be obtained between the three colours. 

 Prof. Hering's theory then leads to the result that there 

 is a definite colour-equation between any three colours ; 

 in other words, that when any three colours are given it 

 is possible to imitate one by a mixture of the other two. 

 This result is so utterly opposed to fact, that a theory 

 which leads to it cannot stand for a moment. 



J. D. Everett 



SOME OBSERVATIONS ON ELEUSS'S NEW 

 PROCESS OF DIVING AND REMAINING 

 UNDER WATER 



I HAVE recently had two opportunities of seeing a new 

 process of diving and of remaining for a long time 

 under water, called, after its inventor, Flcuss's process. 

 The peculiarity of it is that the diver takes down with him 

 such a good and wholesome supply of air-food, that he is 



