180 CLIMIJINU PARASITES. GKEEN-LEAVEÜ PARASITES. TOOTHWOUT. 



root-stock, which is generally onlj^ from i cm. to 2 cm. long, issue fleshy rootlets of 

 the thickness of a quill, but, in many species, as long and thick as a little finger. 

 These rootlets are abundantly supplied with starch, and, in course of time, elongate 

 till they measure 20 cm. They radiate in all directions in the black soil of the 

 meadow, wherein are buried the root-systems of grasses, sedges, and various other 

 plants, and fasten on to suitable hosts by means of one or two suckers yearly, and 

 repeating this process until at length their tips travel into earth devoid of roots, 

 where no more pi'ey is to be found, and there growth ceases. This explains also 

 why these long Pedicularis-roots never descend vertically in the earth, but remain 

 only in the upper strata of soil on a meadow, where a number of other roots are 

 interwoven together, and where it is most likely that the tapering growing-point 

 will meet with the root of some new host or other. 



The Alpine Bartsia (Bartsia alpina), one of the pei-ennial Rhinanthacea; 

 prevailing in the arctic regions as well as in mountainous parts of Europe on damp, 

 marshy, grass-covered spots, is distinguisheil by the sombre duskj- violet colouring 

 of its leaves, and has already been noticed amongst carnivorous plants. On the 

 secondary roots are suckers exactly like those of the Yellow-rattle (Rhinanthus), 

 and by means of these organs it clings to the fibrous roots of sedges and grasses, and 

 sucks their juices. The long, subterranean, runner-like stems, which are covered 

 with small, whitish scales, also bear, however, elongated absorption-cells (root-hairs), 

 which ax'e distinctly difierentiated, and take up nutriment from the vegetable mould 

 around. This Bartsia is, therefore, half-parasitic and half-saprophytic, and it is not 

 improbable that many other perennial Rhinanthaceae behave in the same way. 



The species of Pedicular is which constitute the most extensive group of 

 perennial green-leaved and parasitic Rhinanthacese are, it is true, destitute of 

 tubular absorption-cells (root-hairs) whether on the subtei'ranean stem-stnictures 

 or on the root-tip, with the exception of those which develop in the middle of the 

 suckei-s. But the construction of the epidermal cells on the roots, and the circum- 

 stance that these epidermal cells are always in intimate connection with dark 

 particles of humus, would favour the idea that these plants are capable of taking up 

 organic compounds from the mould of meadows in addition to the food acquired by 

 means of suckers from their hosts. This supposition is further supported by the 

 fact that I succeeded in rearing a species belonging to the Rhinanthacese, namely, 

 Odontites lutea, from a soil composed of a mixture of sand and humus, in which no 

 other plants wei-e rooted, so that the possibility of a withdrawal of nutritive matter 

 fi'om hosts was excluded. It is true that the plants thus reared remained 

 comparatively small and poor, and onlj^ developed few flowers and fruits. But at 

 anyrate they may be considered to prove that plants exist, which, though nomially 

 parasitic, are yet on occasion able to subsist in vegetable mould without the 

 assistance of hosts. 



The third series of parasitic flowering-plants is very restricted, contrasting in 

 this respect with the second series, composed of the numerous green-leaved 

 Santalacese and Rhinanthacese. The species belonging to it differ from those of the 



