Nov. 1903.] BOTANICAL SOCIETY OF EDINBUKGII 4 21 



In D. calyclna the same arrangement of parts is seen as 

 in D. hulhi(ji:na, and on a larger scale, — the stock ending 

 in the swollen pro-bulb capping the bulb itself, or con- 

 tinuing farther downwards through one or more successive 

 bulbs. When the upper bulbs in a series are atropbied 

 and shrivelled, it is not clear from what portion ol tlie 

 axis the new stem sent up in the spring took its origin ; 

 this would depend on whether the rootstock died down 

 to the lowermost bulb, or retained its vitality for some 

 distance above that point. As the rootstock between the 

 bulbs of a series is sometimes seen enveloped in a loose 

 membranous covering, it would appear as if the previous 

 year's axis had perished, and that the existhig one had 

 sprung from a point below the upper bulb ; and if further 

 proof is required of the development of the bulbs taking 

 place from above downwards, the existence of this sheath, 

 which can only be the remains of a preceding rootstock, 

 will supply it. The presence of a new stem taking aii 

 independent course tlirough the soil to the surface is 

 seldom or never seen, the direction taken being always 

 exactly in the line of the old one, from whatever point it 

 may have arisen. A second stem is, however, not un- 

 frequently observed, in various species of Drosera, to spring 

 from a node of the rootstock at different levels between 

 the bulb and the surface of the ground, but only as a 

 branch, tliough usually within the same sheath as the 

 main axis. 



A recent examination of fresh specimens of various 

 species of Droscra confirms to a large degree the state- 

 ments above made, the distinction between the tissues of 

 the living plant and the remains of the growth of previous 

 years, as well as the arrangement of the scales on bulb and 

 rootstock, being much better seen. 



In D. stolonifcra, which is one of the forms having the 

 leaves arranged in rosettes, the large ]julb of a fresh 

 specimen (with tlowers partly expanded) was bisected, and 

 on removing one half of it a bud was displayed projecting 

 downwards into its substance from close to one side of tht 

 base of the rootstock. This bud was as yet no more than 

 about one and a lialf lines lonu', and diverged slightly to 

 one side, namely, inwards towards the axis of the plant. 



