272 



THE JOURNAL OF BOTANY 



As illustx'ating the number of bulbils which may be produced, 

 I may mention that in one well-grown plant which I collected 

 there were a dozen lateral branches, often pinnately branched, and 

 some of these secondary branches bore as many as a dozen pairs of 

 bulbils. The large quantity of bulbils produced was brought home 

 to me by the following circumstance. In November some of the 

 slacks were found to contain one or two feet of water. Under the 

 influence of the wind, the detached bulbils had drifted towards the 

 leeward side of the slack in dense masses, like duckweed at the side 

 of a pond. This method of dispersal is a point of ecological in- 

 terest which would not be likely to occur to an observer who only 

 saw the slacks in a nearly dried-up condition at other times of the 

 year. Whilst as a rule favouring the dispersal of the plant, this 

 distribution by water is liable to prove prejudicial in circumstances 

 such as those mentioned, where the abnormal depth of water car- 

 ried the bulbils too far up on the sides of the dunes, and eventually 

 left them stranded there. During the succeeding period of dry 

 weather, these stranded plants were mostly buried by drifting sand 

 from the side of the dunes, and practically all trace of the drifted 

 masses of bulbils was obliterated. On making a careful search, I 

 found some of the young plants with their uppermost leaves just 

 showing above the sand, and on extracting them it was seen how 

 lengthened they had become in their effort to keep above the sand. 

 At this spot some of the parent plants were also exhumed, and 

 found to have bulbils still in the axils of the leaves and striking 

 root. In one such instance the bulbils and young plants de- 

 veloping from them were 15 mm. long. This continuance of the 

 bulbils in position in the leaf-axils until they begin to root is very 

 common. 



On examining a number of older plants, I found in some cases 

 that lateral branches originating in the axils of the radical leaves 

 were thickened and swollen just below one of the nodes at a point 

 some distance up the stem. Below this swollen portion the stem 

 and its lower leaves were yellow and evidently dying away. In some 

 instances an adventitious root had been put forth at one of the 

 nodes at right angles to the plane of the leaves. Probably such 

 branches ultimately separate from the parent plant, and develop 

 into new plants. Cuttings of these branches were, taken and 

 strewn on moist sand. These soon rooted, adventitious rootlets 

 developing from one or more of the lower nodes. These roots 

 sprang, occasionally singly, but mostly in pairs, one at each side, 

 either from the axils of the leaves or from the nodes of the stem 

 and at right angles to the leaves. 



Compound crystals of calcium oxalate occur in the leaves of all 

 parts of the plant. In the bulbil leaves at a little distance from 

 the nerve there is a broken irregular line of cells containing crystals. 

 These crystals average 0'030 mm. in diameter, and, as the bulbil 

 leaves get older, the crystals naturally become more numerous and 

 rather larger. 



The prevailing form of S. nodosa among the dunes is quite 

 eglandular, and I have not seen the glandular form (S. glandulosa 



