502 



STUDIES IN SEEDS AND FRUITS 



Proportion of Parts in Moist Ripe Fruits of Prunus 

 COMMUNIS OF Different Size and Weight. 



These represent the average results for ten large and ten small fruits. 



accomplished. In the case of bared stones kept a year, I found 

 then when split open, and the seed removed, moisture was 

 deposited when the halves of the stone were placed with their 

 inner surfaces downward on the cold brass pan of the balance. 

 This did not, however, occur with the cut surfaces of the seed, 

 such seeds preserving a stable weight when exposed to the air for 

 days, subject only to ordinary hygroscopic variation. There is 

 no doubt, however, that the seed dries through the stone ; but 

 the above accidental indication in the pan of the balance led me 

 to think that ultimately the stone becomes impervious and that 

 free moisture is thus locked up in the seed-cavity. However, 

 this idea was disproved by a simple experiment. Of ten stones 

 that had been kept for about two years bared of their flesh, 

 belonging to the same set and possessing about the same weight, 

 five were kept in water for a week and the rest left exposed to 

 the air. At the end of that time it was found that whilst the 

 seeds in the dry stones were dry, a little shrunken, and averaged'! 

 I -06 grain in weight, those in the wet stones were moist and 

 swollen and averaged 1-46 grain in weight, the inner surface of 

 the stone being quite moist. 



(36) Pyrus Mains, pp. 324, 325. — Another ripe fruit weighing! 



716 grains and containing 5 seeds (4 grains) gave a pericarp-^ 

 proportion of 99-4 per cent, in the moist condition. ''> 



(37) Quercus Robur, p. 326. — It is not possible to give the exact data 



for the acorn, as the seed continues its growth after the pericarp 

 has ceased to grow, a matter discussed on p. 309. 



(38) Ravenala madagascariensis, p. 325. — The number of seeds refers 



only to those matured ; but the total seed-weight employed in 

 this table includes the aborted seeds, which are much more 

 numerous (see pp. 282, 359). 



(39) Ribes Grossularia, p. 325. — See Note 10, a, on p. 477. 



