156 Royal Society : — Mr. H. C. Sorby on Analysis of Animal 



Fig. 3. 



Eed end. a b 



1. Neutral or somewhat acid. 



2. A little carbonate of soda. 



3. More carbonate of soda. 



Blue end. 





Acids produce no important change, and the effect of alkalies is 

 best seen by gradually adding carbonate of soda. This alters the 

 colour to a more and more blue purple, and the spectrum is changed 

 in the manner shown in fig. 3. The three bands seen in the neu- 

 tral solution may be referred to as b, c, and d ; and their centres 

 occur at equal intervals of about lj. When enough carbonate of 

 soda has been added to make it slightly purple, a fourth band, #, 

 makes its appearance, separated from b by the same equal interval of 

 1 2, whilst the other bands remain in the same position as at first, 

 only modified in intensity. The band a becomes darker and darker 

 as more carbonate is added, until, when the solution is a fine purple, 

 it is as dark as the others (see No. 2) ; and on adding more carbo- 

 nate it becomes still darker, and the bands c and d more faint, until 

 the solution is a purple blue ; and the spectrum has only the two 

 well-marked bands a and £>, shown by No. 3. 



The bright blue colouring-matter of the flowers of Lobelia speciosa 

 gives, when neutral y almost exactly the same spectrum as that 

 of Alkanet-root when alkaline (No. 3), having two well-marked 

 absorption-bands, whose centres are at 2| and 4| ; and on adding 

 carbonate of soda, the upper one is gradually removed, and the 

 centre of the lower is depressed to near 2\. More or less similar 

 results occur in the case of many other blue colouring-matters ; and 

 on adding a slight excess of acid the general absorption is raised, 

 and other bands may be developed higher up, at equal intervals ; but 

 when a large excess has been added, they are lost in a strong general 

 absorption. Too strong an alkali may also destroy narrow bands 

 in a similar manner, as is well seen in the case of Brazil-wood. 

 The neutral aqueous solution shows an absorption-band, made far 

 more distinct by the addition of bicarbonate of ammonia, which 

 makes it pink and very fluorescent. The spectrum is then 



4f-5f 7; 



but on adding excess of ammonia the solution ceases to be fluorescent, 

 the narrow absorption-band is lost, and the spectrum becomes 



3i.-4J— 8-..9|. 



