72 GROWTH OF PLANTS 



nated American lotus seeds by treating them with acetone and then plac- 

 ing them in water. She assumed that fat-like substances were dissolved 

 out of the stomatal cavities, which extended deeper than the palisade 

 layer, allowing water to enter. Ether as the solvent made the coats per- 

 meable also, but killed the embryos. Shaw also maintains that the palisade 

 cells are impervious throughout their length. 



Hamly ^^ speaks of suberin caps over the palisade cells as causing the 

 water-resistance in sweet clover seeds. McKeever " found that treatment 

 of black locust seeds for 10 to 120 minutes Avith several wax solvents 

 (ether, xylene, acetone, and others) was effective. All removed consider- 

 able amounts of wax. The main effect seems to be hastening the germi- 

 nation by 10 or 15 days rather than increasing it after 27 days. Very early 

 Hohnel ^^ claimed that soakmg yellow lupine seeds in ether with immedi- 

 ate transfer to water softened the coats. Perennial lupine seeds did not 

 respond to this treatment. Verschaffelt,^26 working mainly with hard 

 honey locust seeds, found that placing them in ethyl alcohol and trans- 

 ferring them immediately to water led to swelling. He assumed that the 

 alcohol filled the rifts or interstices in the hard coats and furnished a 

 channel by which water could travel to the deeper layers of the coats. 

 Other simple alcohols were effective, but higher alcohols, ether, and other 

 fat solvents were not, because water is not sufficiently soluble in them to 

 reach the deeper layers; also some of them failed to fill the rifts in the 

 seeds. Many hard seeds of Caesalspinaceae and Mimosaceae acted like 

 honey locust seeds, but alcohol was less effective with Papilionaceae. The 

 first two groups have minute rifts all over the surface of the seeds, while 

 the latter has one rift at the hilum. According to Verschaffelt's interpre- 

 tation, the alcohol does not act as a wax or fat solvent but as a bridge for 

 conveying the water to the deeper layers of the coats. 



Some regions of the hard seed surface seem to be rendered water- 

 permeable more easily than others. Hutton and Porter ^^ showed that 

 dry, hard seeds of Amorpha and Lespedeza, when shaken in a bottle, 

 become water-permeable at the hilum. Hamly ^^ found that hard Meli- 

 lotus seeds were made water-permeable by moderate heating or mechanical 

 impact by producing a rift at the strophiole. This long and much worked 

 problem of what physical or chemical characteristics make hard seed coats 

 resistant to entrance of water evidently still needs thorough research atten- 

 tion. The chemical or physical method of water-proofing may vary with 

 different kinds of hard seeds. This is made probable by the contradictory 

 explanations offered above; also we must remember that little attention 

 has been given to the mechanism of hard-coatedness in several of the 

 families of plants. In some, not even the layer of the coat involved is 

 known. 



There is another reason for learning the mechanism or mechanisms of 

 hard-coatedness in seeds. Hard-coatedness is the world's best example of 

 highly effective water-proofing by thin layers. Man can well devote some 



