688 ON MOLLUSCS 



surface a calcareous layer as hard as ivory and of a different colour, 

 which covers the preceding layers 1 . 



Many bivalve and univalve shells have on their inner surface 

 lively and glancing colours, which must not be confounded with 

 the external colours. They are the play of colours from the refrac- 

 tion which is seen to arise on the transmission of white light 

 through a prism, and which causes the beautiful phenomenon of the 

 rain-bow. The mother of pearl therefore is not to be ascribed to 

 any special colouring matter, but to very fine streaks and folds of 

 the internal membrane of the shells 2 . It continues even after the 

 carbonate of lime has been removed by an acid, and CARPENTER 

 observed that the play of colour disappeared when he extended the 

 membrane with needles so that the folds were obliterated 3 . These 

 changes of colour are particularly beautiful in the genus Haliotis, a 

 species of which is used by the Japanese to ornament the lackered 

 furniture of their houses. 



Whenever the fluid, which forms this innermost layer of bi- 

 valves, is effused in the form of small drops, pearls arise, which 

 have frequently a very irregular shape. These may accordingly 

 exist in very different species of bivalves, also in turbinate uni- 

 valves; yet they arise usually in bivalves, and especially in Melea- 

 grina margaritifera and Unio m,argaritiferus. The former species 

 of conchifer is fished in the Persian gulph, the last is found in 

 fresh water in different parts of Europe, and also in our country. 

 According to HOME, pearls have undeveloped eggs for a nucleus, 

 which have remained accidentally on the outside of the mantle 

 within the shell. Such may often be the case, but there is no 



1 BRUGUIERE in the Journal d'ffist. Nat. I. 179*2, pp. 307 315, quoted by 

 SCHWEIGGER Handb. der Naturgesch. der sJcelettlosen ungeglied. Thiere, s. 68 1. A dif- 

 ferent opinion of BRUGUIERE, that the animal of Cyprcea leaves its shell, and forms a 

 new one, though adopted by LAMARCK and SCHWEIGGER, can no longer be defended ; 

 compare DESHAYES in the new edition of LAMARCK Hist. nat. des Animaux s. Verti- 

 Ires, x. p. 485. 



2 This was first announced by BREWSTER Phil. Trans. 1814. If a drop of sealing- 

 wax be allowed to fall on a mother of pearl shell, the surface of the wax, which has 

 been in contact with the shell, presents the same colours. It may be supposed that 

 this was an impression of the surface. But it appears, that in this experiment a thin 

 scale of the surface of the shell remains adhering to the wax. 



3 Annah of Nat. Hist. xn. p. 382. 



