47 



existence of a high pressure in the air-passages, we must admit 

 that the opening of one of them by some mechanical Injury 

 would temporarily diminish that pressure. The turgor in the 

 cells bordering upon the canal would cause them to grow in the 

 direction of the least resistance, i. e., into the open space of the 

 canal, until the hairs thus formed would close the opening. That 

 the portion of the canal near the wound was most Hkely filled 

 with water may be concluded from the fact that in all sections, 

 both from fresh and alcohol material, the mucilas^e fillinir the 



spaces between the hairs, even at some distance above the wound, 

 abounds in bacteria, and even sometimes diatoms, of the same 

 kind as those that are found on the outer surface of the plant 

 Should these observations and conclusions prove correct, the 

 same arguments might be applied with equal force to account for 

 the origin of the curious formations known as tylosis, with which 

 the hairs in question have many analogies. 



EXPLANATION OF PLATES. 



PLA.TE LXXVIL— Fig, 1— Rootlet with enveloping sheath, sh. x 20. 

 Fig. 3 — Cross-section of stem ; c, cortical portion ; ic, intercellular air- 

 canals ; m, mestom ; h, hadrom canal; 1, leptom (The shaded part between 

 the leptom bundles — parenchyma cells — takes up more space than the tissue 

 it represents really does.) x 20. Fig- S—Longitudinal tangential section of 

 stem near node, through mestom (m, Fig. 3); h, hadrom canal; sd, spiral 

 ducts : st, sieve-tube, x 220. Fig. 4— Cross-section of floating leaf near upper 

 epidermis; e, epidermis cells; p, palisade cells; a, large air-cavity under 

 stoma X 600. Fig. 5 — Cross-section of intercellular air-canal in petiole, almost 

 filled with internal hairs, hr, and mucilage, mu ; one of the canals bordering 

 on it is lined with a thick layer of mucilage, mu. x 130. Pig. 6— Internal hair 

 beset with mucilage drops, x 430. Pie;. 7— Internal hair containing .larch. 

 X 220. 



Plate LXXVIII.— Fig. 8— Epidermis cell beginning to produce a hair ; 

 e, adjoining epidermis cells, x 600. Fig. 9— the same, after the formation of 

 a cross-partition, x 600. Fig. 10— Young hair; pc, pedicel cells; pr, proto- 

 plasma layer; mu, mucilage, x 600. Fig. 11— Hair with incipient mucilage 

 sac, ms ;' mu, mucilage ; pr, protoplasm (in circulation); pc, pedicel cells, x 500. 

 Fig, l3_Hair with mucilage sac x 500. Fig. 13— Hair with sac burst; muci- 

 lage has disappeared, and granular dead plasma remnants are left, x 500. 

 Fig. 14— Hair with two large mucilage drops, x 230. Fig. 15— Portion of hair 

 with mucilage much reduced; the dotted curves are to indicate the doublj- 

 concave shape of the drop (cf. text), x 500. Fig. 16 -Hair with remnants of 

 first sac and entire new one, x 230. Fig. 17— Same hair after bursting of 

 second sac (cf. text, p. 39). x 230. 



Cf. Melhnlv, 1. c. 



