Dr Scoresby on the Colours of the Dew-Drop, 55 



S2)hericity of the dew-drops, and their more distinct separate- 

 ness, give this peculiar form a decided advantage. Similar 

 effects of colour also maybe seen during sunshine in any 

 globular vessel filled with water. In ver^ small globules of 

 glass, so filled, the phenomena more nearly approximate those 

 of the dew-drop, as the size of ordinary glass-vessels causes the 

 colour to be seen only at the extreme verge of the globe* 

 with a combination of tints, whilst in the small dew-drop the 

 different tints become generally resolved, -as to any discrimi- 

 nating power in the eye, into an uniform colour, varying only 

 by a change of the angle at which it is observed in respect to the 

 position of the sun. In all cases, however, in which these 

 phenomena are observed, the small telescope becomes a most 

 important acquisition, by removing the indistinctness and du- 

 biousness which necessarily belong to colour when the angle 

 subtended by the luminous coloured object is very minute. 



Whilst the departure from the true spherical form occasions, 

 in the colour of the dew-drop, such uncertainty as to the 

 angle at which the spectral phenomenon may appear ; the 

 regularity of the form of the rain-drop whilst descending 

 through the air, and the free and perfect operation of the 

 attraction of cohesion, yield obviously the fitting optical con- 

 ditions for the correspondency in the angular position, with 

 respect to the sun, of the rainbow. There should, however, 

 be, according to theory, a minute deviation from the true 

 spherical form in the descending rain-drops, arising from the 

 resistance of the air ; and that deviation ought to be different 

 in degree according to the magnitude of the rain-drops. For 

 as drops of unequal magnitude will descend with difi*erent 

 rates of speed, the resisting action of the air against the under 

 surface of the drops will occasion unequal measures of com- 

 pression ; whilst a shower of rain, consisting of drops of un- 

 equal size, should, it is presumed, afford spectral angles of 

 somewhat different magnitudes. Theory, I think, would fairly 

 lead to such a conclusion. And if so, may we not herein dis- 

 cern the cause, possibly, of that phenomenon in the rainbow 

 which, so far as I am aware, has not been explained, of super- 

 numerary arches ? On an occasion in which I once observed 

 a rainbow with three or four such arches of *itigular beautjr» 



