84 Prof. A. J. Brown. Selective Permeability of the [Jan. 23, 



first this appeared to be very improbable, as the property was exhibited by 

 the seeds in the presence of highly toxic acids and salts, which could hardly 

 fail to arrest its vital activity if brought into contact with protoplasm. 

 Experiments made with seeds which have been immersed in boiling water 

 during varying periods extending to one hour have afforded conclusive 

 evidence that the semi-permeability of the seed-coverings is not a function of 

 living protoplasm. 



Histological study of the seed-coverings indicates that their selective power 

 is confined to the testa, and probably to that portion which is derived from 

 the epidermis of the nucellus during the development of the seed. 



It appeared to be very desirable that the coverings should be removed and 

 their behaviour studied apart from that of the seed contents. Although this 

 has been attempted, the many experimental difficulties met with have not 

 been satisfactorily overcome, and hitherto no other means of investigating 

 their activity has been found than the somewhat unsatisfactory one of 

 experimenting with whole seeds. 



Behaviour of the Seed of H. vulgare as a Semi-permeable System possessing a 



strong affinity for Water. 



When sound seeds of H. vulgare are immersed in an aqueous solution con- 

 taining a substance which cannot pass through the outer semi-permeable 

 membrane, and water is absorbed, presumably the seed contents enter into 

 competition with the solute for the water. From this point of view, it 

 appeared to be important to ascertain to what extent water would be absorbed 

 from various solutions, and to compare the amounts with that absorbed when 

 the seeds were placed in water alone. 



The method adopted in all the experiments to be referred to was as follows : 

 — A known weight of selected air-dried seeds, usually about 5 grammes, 

 having been steeped in the solution under examination during some desired 

 period, the seeds were separated from the liquid by means of a small wire- 

 gauze strainer. After draining for a few minutes they were placed between 

 the folds of a soft, dry cloth and were rubbed gently to remove as much of 

 the adherent liquid as possible ; they were then weighed. It is obvious that 

 such a method cannot afford absolute values ; but experience shows that 

 concordant results may be arrived at without difficulty if all operations in 

 connection with draining and drying the seed are carried out in as constant 

 a manner as possible. The water absorbed is expressed as a percentage weight 

 calculated on the original dry weight of the seeds. 



The results of a series of observations with solutions of common salt con- 



