REJUVENESCENCE IN NATURE. 87 



immediately by a suppression of the leaf-formation, which 

 persists through the whole main superstructure of the 

 inflorescence, and only ceases in the spikelets. The leaves 

 often disappear entirely in this region, or they present 

 themselves as more or less evident thickenings, some- 

 times completely embracing the stem as rings,* and then 

 often curved up and down in wavy lines, sometimes only 

 partially embracing, and then mostly with decurrent 

 borders,f rarely ascending on the axis.J In Elymus 

 europcBus the lowest annular abortive leaf is often elon- 

 gated into a sharp tooth ; in Nardus all the subtending 

 leaves of the spikelets are tooth-like. Only few grasses 

 exhibit a considerable development of the lower leaves on 

 the axis of the inflorescence, such as occurs in Sesleria. 

 In Sesleria ccerulea they present themselves as tubular, 

 ochraceously-truncated, or irregularly excised, mem- 

 branous sheaths. A similar development of the first 

 leaf of the inflorescence occurs not unfrequently as an 

 accident and exceptionally, in other grasses. I have 

 observed a strange form of this phenomenon, and this 

 frequently, in Glyceria aquatica. The transitional leaf 

 occurring here, in the axil of which stands the first so- 

 called semi-verticil of the panicle, has an undeveloped 



* Thus, for instance, the lowest abortive leaf of Triticitm, Secale, Oryza, 

 &c., while the succeeding do not embrace ; in Glyceria fuitans and aquatica 

 several of the lower abortive leaves are annular. 



f Especially fine in Melica altissima and Phalaris arundinacea. Very 

 beautifully so in the lower abortive leaves of Lolium, Poa compressa, 

 Cynosurus, and Dactylis. The obliquely descending borders are even con- 

 fluent on the lowermost. The overlooking of these abortive leaves led to 

 the error of regarding the glume of the spikelet of Lolium as its subtending 

 leaf. This mistake might be pardoned in a superficial examination, but it is 

 incomprehensible how any one could see the true condition clearly, and even 

 draw it, and yet retain the false explanation, as Turpin has done. See his 

 plate of Lolium perenne in the ' Diet, des Sc. Nat.,' and the explanation 

 given of the figures. 



\ Thus, for example, the lowest and not always distinguishable abortive 

 leaf of Alopecurus agrestis. 



These ochraceous leaves are followed by one or the other unilaterally 

 developed bract; at the base of the uppermost lateral spikelet, on the 

 contrary, they vanish entirely, as in other grasses. In Oreochlna they are 

 not so strongly developed, and only the lowest are one-sided. See on this 

 point also Roper, ' Zur Flora Meklenburgs/ ii, 42. 



