278 PEOF. W. A. HERDMAX ON 



As specimen after specimen of the livino- ascidian went through the same 

 ■series of colour-changes before our eyes this summer, it is probable that we 

 have seen all that can be determined from observation of the living animal, 

 and that any further investigation of the pigmentation must be undertaken 

 .by the chemist in the laboratory. 



Dr. Alfred Holt, of the Physical Chemistry department of the University 

 ■of Liverpool, has kindly undertaken to investigate the pigment or pigments 

 of these Hebridean colonies of Diazona from the chemical point of view. He 

 reports to me as follows on his investigation so far : — 



"Animal investigated was purple outside, but yellow-green inside. The 

 purple and yellow-green portions were investigated separately. 



" (1) Purple part. — Cut up into small fragments or ground with sand 

 and then extracted in a Soxhlet with absolute alcohol, the colour gradually 

 passed into solution. The solution thus obtained was bluish green in colour, 

 quite distinct from the colour of the inner part of the animal or from the 

 alcoholic solution obtained from the fresh animal on capture. This blue- 

 green solution on evaporation at a low temperature (below 100° C.) gave 

 a small residue which by transmitted light had no decided colour, bvit by 

 reflected lioht was purple, identical in colour with that of the animal. The 

 colour was soluble again in alcohol or zylene, solution taking a long time. 

 The zylene extract had a more pronounced blue tint than that in alcohol. 



"The addition of alkali to the blue-green solution changed the colour to 

 a greenish yellow. 



" (2) Green part. — The colouring matter of the inner green part was very 

 easily soluble in alcohol to a green solution, the colour of which was, 

 however, far more yellow than that of the purple part. On evaporation at 

 low temperature this yellow-green solution yielded a green residue, not 

 purple, and was again easily soluble in alcohol. The addition of alkali did- 

 not materially change its colour. 



"The oreen parts of the animal did not go purple in alcohol. The fresh 

 animal preserved in formaldehyde maintained its green colour, and this was 

 readily and completely soluble in alcohol. 



" Acid had no marked effect, nor acid nor alkaline hydrogen peroxide 

 in dilute solution. 



" From the absorption spectrum and general characters it would seem 

 that the pigment in the purple portion is very similar to that obtained by 

 Friediander from Murex hrandaris, and which was subsequently shown 

 by him to be an indigo derivative. If this be so, then the yellow-green 

 colour of the alcoholic solution from the inner (not purple) portion of the 

 animal can scarcely be due to this pigment. Microscopic examination of 

 this part shows numerous globules containing what looks like a yellow oil, 

 and it is probably the solution of this oil that gives the yellow-gri'cn colour 

 to the alcohol. The particular shade of tint may vary, dopentling on the 



