370 STUDIES IN SEEDS AND FRUITS 



first considering the " How " of seed-coloration before one 



can attempt to think of the " Why." One of the first things 



that struck me in this connection was the circumstance that 



Seed-colours many a pretty hue in seeds comes and- goes in the fruit before 



appear its seeds are exposed (by its dehiscence or its breaking down) 



fniiUs rioe '° *^^ ^^^' "^^^^ '^^ ^^^ ^^^'^ ^^ Abrus precatorius and Adenan- 

 thera pavonina, where the scarlet coloration of the normal rest- 

 ing seed is preceded by a pretty rose-pink colour in the soft 

 unripe seed. The seeds of Ravenala madagascariensis (the 

 Travellers' Palm), when wrapped in their bright blue arils in 

 the closed capsular fruit, are very different-looking objects 

 from the same seeds in the opening fruit, when the aril has 

 become a dirty brown. Then, again, on cutting open a young 

 fruit of Barringtonia speciosa, one is a little startled to find seeds,' 

 half an inch in size and not much more than bags of fluid, 

 coloured deep red, a hue which they lose altogether as the seed 

 matures in the centre of the fruit. One of the most striking 

 cases of seed-coloration is seen in the germination of the yellow 

 seed of Guilandina bonduc, which, as the hard shell swells with 

 the absorption of water, assumes a chocolate-brown hue. Such 

 a seed, when the swelling process has aflfected one-half of its 

 surface, is half yellow and half chocolate brown, and presents 

 thus a conspicuous contrast in coloration. 



From the various cases of seed-colouring just noticed, it is 

 obvious that it would be futile to look for an explanation, 

 before we learn more of the conditions attending and preceding 

 the coloration of seeds. One of the most important conditions 

 lies in the circumstance that seeds assume to a greater or less 

 Seeds colour degree their permanent colours in the closed fruit before they 

 w^^e ose ^j.g gxposed to the air either by its dehiscence or decay. This 

 is not absolutely essential for coloration, but it is as a rule 

 essential for normal colouring. I have handled a great many 

 immature seeds, and have found that generally such seeds 

 colour defectively when removed as soft seeds from the green 

 fruit. Those seeds that are white when immature and deep 

 brown or black when mature often sufFer least by being 



