132 On the Examination of 



the year ; but since I am only using these facts as iUnstrations of 

 another subject, and am not -writing a paper on the colouring matters 

 of Algae, I wish to confine myself to what is requisite on the present 

 occasion. What I therefore now contend is that a simple compari^son 

 of these five spectra shows that they are due to the variable admix- 

 ture of four different colouring matters, which, for convenience, I 

 will call W, X, Y, and Z, from their respective absorption-bands, as 

 shown in the woodcut. In the first place there can be no doubt 

 that there is a simple substance X which gives rise to the single 

 band seen in spectrum No. 1. This same band occurs in No. 5, but 

 along with it is another, due to a blue-purple colour W, which is 

 obtained separate in solution by keeping the plant for some time in 

 a httle water. The spectrum of the Hght transmitted by this solu- 

 tion shows the single band "W and that of the splendid rose-coloured 

 fluorescence a single rose-coloured band, just on the red side of the 

 absorption-band. As will be seen on comparing the spectra, No. 4 

 differs from No. 5 in having a broad absorption-band Y. That both 

 the substances W and X are really present is shown by the spectrum 

 of the fluorescence of the aqueous solution obtained from that kind 

 of OsciJlatoria, which shows both the narrow rose-coloured and the 

 narrow gi'een-yellow bands, and thus, though I have not yet been 

 able to procm-e the substance Y separate from all others, I do not 

 hesitate to conclude that it does exist as a third independent colour- 

 ing matter. That there is a fourth, giving rise to the band Z, is 

 also clearly shown by comparing either spectra Nos. 1 and 2, or 3 

 and 4, and this conclusion is established by the fact that, on keeping 

 for a few days the mixed colouring matters dissolved by water from 

 the purple Floridea, and removing the deposit by filtration, a flesh- 

 coloured solution may be obtained, which gives a spectrum with 

 this single band Z. 



According to these principles the colouring matter of the various 

 plants may be expressed as follows : — 



1. Pahnella crucnta X 



2. BbA Floridea X -f- Y + Z 



3. Purple Floridea W' + X -f- Y + Z 



4. Oscillatoria grown in water AV + X + Y 



5. Microcoleus grown on a damp wall W + X 



The substance described by Kiitzing, Stokes, and Askenasy by 

 the name Phjcoerytlirin appears to have been a mixture similar to 

 2 or 3, along with perhaps another substance, which I have not yet 

 seen. Cohn's Phycocrjan must, I think, have been a mixture of 

 W, X, and Y with another, which was possibly an altered product, 

 giving an absorption-band not seen in the spectrum of the plants 

 themselves. I may here say that I have met with at least four 

 such products, and one, which gives a narrow band in the red, is 



