Longevity of Seeds. 193 



Fifioher's results, and at any rate seeds of Sarjittaria liprhtly sand- 

 papered, and tlien kept at 28 dejr. C. to 30 dee. C, between well- 

 wetted filter paper, frerminate readily. Since the above was 

 written. Crocker (Bot. Gaz. 1907, p. 374) has shown that the 

 ditficulty of germinating the ripe seeds of various water plants is 

 due not to any inherent dormancv of the protoplasm, but to the 

 coats being imperfectly permeable, the seeds germinating readily 

 when the coats are filed or broken. Fischer's treatment, i.e., 

 the stiiiiul;)tion of H and HO ions, may perhaps enable the seed 

 to germinate under the stimulus of a lower partial pressure of 

 O and a lower temperature than otherwise, but is not an essentiail 

 condition for germination, which, given a free supply rf oxygen 

 and a sufficient temperature and supply of moisture, appears to 

 be rendily induced in the case of Sa(/itiaria at least. Correvon,^ 

 of Geneva, states that chemical stimuli, such as soaking in 0.25 

 per cent, acetic a,cid or 2 per cent, phosphoric acid is necessary 

 for the germination of the seeds of JunijJcrits Cedrus, but all 

 such actions may be the result of a change in the permeability 

 of the coverings of the seed. 



It must be remembered that in some cases ait least seeds per- 

 meable to water are covered before germination by continuous 

 semipermeable membranes, and such membranes may prevent the 

 entri' of the requisite amount of oxygen, while allowing water to 

 enter. Thus A. J. Brown^ has shown that the inner part of the 

 intact spermoderm of Hordeum, Avena, Triticum and Secale 

 forma a semi-permeable membrane which allows water and iodine 

 to enter, but not solutions of sulphuric acid up to 36 per cent, 

 strength, or of hydrochloric acid or metallic salts up to 5 per cent. 

 (1 per cent, nitric aoid penetrated slowly), and even after boiling 

 the seeds or treating them with iodine, the sulphuric acid may 

 still be unaible to penetrate. Hence the semi-permeable mem- 

 brane is in this case a non-vital one, and it appears to be that 

 ]>art of the spermoderm which is derived from the epidermis of 

 the nucellus. It may, in fact, be .something of the nature of an 

 internal cuticle. 



In the case of the paired burrs of Xanthium, one seed usually 

 terminates in the first year, the second in the next year, or some- 



1 Card. Chron., 1907. 



•2 Annals of Botany, vol. xxi., 1907, p. 70. 



