Shoots in the assimilating state without prolongated internodes 



very seldom contain subterranean floral buds. I have examined a 



lot and only in some two or three cases I have succeeded in finding 



one (fig. 2). But in some other cases I have 



at the base of the culm found: 1) an old 



bud brownishly coloured from the humus in 



the soil, 2) then again after about 5 short 



nodes at the base of the first prolongated 



internode a fresh looking yellowish bud, both 



of them with fully developed seeds (fig. 3). 



The explanation of this might be, that the 



first floral bud was developed in autumn in 



the year before the prime panicle and the 



other at the same time as this. 



Not seldom two or three such buds are '^^I'^l ^^"d, a. Old sheats 

 „ , . T J ,1 1 removed. Natural size. 



found m succeeding nodes on the same culm, 



then the one or two uppermost usually do 

 not give fully developed seed. A very funny 

 exception shows fig. 4 in which every prolon- 

 gated internode in the culm produces a more 

 or less complete inflorescence. The solitary, 

 stalked spikelet in the second uppermost node 

 makes a kind of transition between the nor- 

 mally developed and the subterranean in- 

 florescence. 



Fig. ± 



Leaf shoot of 

 S. (lecumhens showing a 



The floral buds have a length of 5 — 13 



Fig. 3. S. decuinbens. 



Basal part of a culm 

 showing floral huds of 



different age; a, old 

 brownish bud, h and c, 



young yellowish buds. 

 Natural size. 



mm. generally about 7 mm. The bud con- 

 sists of the two-keeled prophyllum, which 

 has a single row of hairs along the upper part 

 of the keels, and a small spikelet. 



In the spikelet itself glumae are missing, 

 but paleae fully developed; they are pale in colour, the lower 

 without marked nerves, the upper two-keeled, but both much alike 

 vegetative subterranean budbracts in substance. 



The spikelet is one- or two-flowered ; in one-flowered spikelets 

 rudiments of a second flower are often found. 



The plane in which the spikelet is arranged seems to be the 

 ordinary one in Sieglingia, rectangular to the plane in which the 

 leaves in the main culm are placed. But the position of bracts 

 and seeds is often skew and twined, so much so, that it is difficult 



