210 
Transactions of the Royal Society of South Africa. 
of value 96, whereas in the uiisymmetrical series PhC^(CgH4X).3 the two 
i^roups CgH jX are united by the tautomerism C ,— CPh = C of value 103 + 
16 = 119. We will now calculate these dyes. 
(11 j Aurine-sodiiim in water. (CgH^O-^)--. 
According to the above theory we have tivo active rings CfiH.O/' of 
value 219 each, plus 
of value 96. This gives 438 + 96 = 534. This agrees exactly with observation. 
(12) Pararosaniline in irater. C^(CgH_,NHo);^. 
Here we have two active rings CgH^NHo of value 223 each, plus the same 
tautomerism ; hence the calculated value is 2 x 223 -f 96 = 542. Observation 
gives X 543. 
(13) Crystal -violet {N-hexamethylpararosaniline). (CoK^^Me.y)^- 
Here we have two active rings CgH^NMeo of value 251 each, plus the 
same tautomerism ; hence the calculated value is 502 -f- 96 = 598. Obser- 
vation gives ^ 596. 
It is known that in the case of Nos. 12 and 13 the addition of acid 
changes the colour, fuchsine becoming purple and crystal -violet green. 
The true explanation of this now lies to hand, viz. that the symmetry is 
destroyed, leaving two of the groups as C^H^NHo whilst the third becomes 
the ion CgH^NH.,^. The dyes become then derivatives of the unsymmetrical 
class mentioned just above PliC(CgH4X)o. Thus purple (acid) fuchsine is 
a substituted Doebner's violet, and green (acid) crystal-violet is a substi- 
tuted malachite-green. The ion CgN^NH^"*^ and the ion CgH^NHMej^ are 
non-tautomeric and appear both to have the value 27 ; if this number be 
substituted for the value 16 used for plain phenyl in calculating Doebner's 
violet and malachite -green, the calculated results are 549 (diaminobenz- 
hydrol) + 27 (or 576), and 605 (Micliler-hydrol) -f 27 (or 632), which are 
precisely the observed figures in acid parafuclisine and acid crystal -violet. 
(14) Paraoxy-derivative of malachite-green. 
This interesting substance is another crux of the colour chemist, being 
blue when neutral and purple-pink when alkaline. 
On my theory the alkaline phase should have nearly the same colour as 
the substance 
and the neutral phase should have nearly the same colour as 
