Dr Scoresb'y on the Colours of the Bew-Drop. 53 



hand of the shadow of my head, and when examined, as above 

 recommended, from rlfjht to left^ the succession of colours wss 

 as follows — ^faint purple^ pink, red, orange, yellow, green, 

 bluish-white, resembling in lustre and fire the diamond. The 

 order here, omitting the two first tints, was that of the pri- 

 mary rainbow » 



Various attempts were made to determine whether there 

 were any particular angles at which the development of co- 

 lours was resoluble, but, as I have said, without being able to 

 reduce the phenomena to any given law. Measured from the 

 centre of the shadow of the head of the observer, towards the 

 right, for example, when the sun had considerable altitude 

 (such as 40° or 50°), 1 found a very large number of beautifully 

 prismatic drops, pendant on blades of grass, at different unequal 

 angular distances of from 5^° to 28° or 30°. Various angles 

 being measured by a pocket sextant, where the orange tint 

 appeared, gave, in numerous examinations of different dew- 

 drops, 51°, 12i°, 12^, ISr, several of 22^°, 28°, 34°, &c. In 

 all these cases the distance of the drop from the eye never ex- 

 ceeded 10 feet ; varying from 7 to 10 feet. The order of the 

 colours in these several cases (with but one exception that I 

 recollect) corresponded with that of the primary rainbow ; 

 but in many cases the only distinct colours were orange, yel- 

 low, and bluish-white. At an angle of 63°, on the same oc- 

 casion, a prismatic drop was examined, exhibiting the reverse 

 order, and thus indicating a double reflection. On examining 

 the drops closely, with a compound botanical microscope, I 

 found, as might have been anticipated, a great difference in 

 their relative positions and forms. Most of them w^ere pen- 

 dant on blades of grass, but the globules in many cases de- 

 viated considerably from the spherical form, — some being too 

 weighty for the attraction of cohesion, so that they assumed 

 an ovoidal form, and others being of so deficient a weight as 

 to appear, not as pendant ovoids but as semi-ovoids suspended 

 from the longer axis. For in all cases the part attached to 

 the blade of grass was necessarily flattened. Figures 3, 4, 5, 

 Plate II., shew some of the forms examined, whose colour and 

 angular distances have already been described. 



The order of the colours and the position of the ob- 

 server will be made more intelligible by reference to figure 6, 



