ON CHANGEABLE COLOURS. 137 



up by sliding underneath It a piece of paper, or of glass, as 

 the side of a funnel for instance, to allow the water to run 

 off. The colouvs of the pellible will continue visible after 

 it is removed ; and by lettin;^ the matter dry slowly we may 

 preserve them for an indefinite time with all their vividness. 

 But if we pa^s the linoer lightly over il, we shall collect 

 nothing but a green powder: the red, yellow, blue, and 

 purple, that appeared so brilliant, will be gone in an in- 

 stant. 



My second phenomenon respects the changeable colour of Changeable 

 several pai ts of the plumage of the peacock, and of some ^j^g^s. ' '' 



other birds, as the cock, pigeon, duck, and turkey. Here, 

 after a very deliberate compar;»rive examination, and the * 



most circumspect reflection, I have ultimately relinquished 

 the idea, that the^e colours are to be referred to rings. My 

 conviction was produced as follows. 



In the iirst place 1 considered, that these colours were not Not owing to 

 -the necessary result of a certain tenuity of parts; for on the ^®^^"^'y* 

 one hand several animals incontestably display in the slen- 

 der filaments of their hair, feathers, or dowii, various de- 

 grees of teiiuity, Irom tlie most imperceptible, vvitirout its 

 producing colour. Is not the white peacock itseif a staking 

 instanct of this? 



On the other hand have not many birds and insects co- 

 lours unchangeable in situation and iii tlieir reflections in 

 every point of view? 'i hose ot the wings of some butter- 

 flies are peifeciiy tixed, tiiough dependant on a down so tme 

 as to be scarcely visible, it is likewise proper to remark. Ail o^^ake ex- 

 that all these colours indicate opacity, as those of the jxa- ^^-!'t^'''« wings 

 cock s feathers ; the wii.^s oi flits excepted, in which tints 

 analogous to coloured rings are observed : but iheae mem- 

 branes have a sensible tiani&pareucy, like scales of niica or 

 blown glass. 



I alterward observed the change of colour of several fea- Feathers offhe 



thers of the birds mtntioised. in those of the peacock's pe^^^^^^'^ i^'l* 



tail, we see the lateral liiaments, on ciianging their position, 



pass suddenly from red to green. 1 he red is produced by Redbyper- 



a nearly perpendicular rellection of li<4hl,.the srreen by a l'^"^'cular, 

 ,,.,,. , , . , °. "^ screen by ob- 



very GblLc^ue retiection ; and there is no alternation of re- liquc reflection, 



flection 



