ENDS AND CONNECTIONS OF THE BUNDLES. 083 



and more especially at the free ends of the transverse branchlets (cf Fig. 145, 

 P- 303)- 



Sect. 113. In typical roo/s, it is simply the meristematic group of the growing 

 point which forms the end of the vascular bundle. 



Those parts of Phanerogamic Parasites which are developed inside the host, 

 (intraniatrically), as well as their haustoria, behave differently. According to the 

 existing investigations, for the details of which the- special- works are to be compared^ 

 these organs are traversed by vascular strands, which are quite similar in structure to 

 the ends of bundles in the foliar expansions ; they form strands of vessels with short 

 elements, usually reticulated or pitted, accompanied by rows of moderately elongated 

 thin-walled cells, with acute ends ; or they are isolated vessels with very short, and 

 then usually very irregular elements, traversing the parenchyma, and often actually 

 interrupted by parenchyma, thus constituting isolated portions of vessels, or 

 tracheides, enclosed in the latter. These strands are, on the one hand, connected 

 with the xylem of the bundles of the extramatrical organs of the parasites ; on the 

 other hand, they extend at definite points to the limiting surface between the intra- 

 matrical tissue of the parasite and the tissue of the host, here entering into close 

 connection with the latter, and in fact becoming as a rule closely attached to the 

 equivalent tissue, i. e. to the vessels and woody elements of the host. The intimacy 

 of the connection may go so far that it becomes diflScult to determine the boundary 

 where the vessel of the parasite begins, and that of the host terminates. Sieve-tubes 

 have not been found in company with these vessels and strands of vessels. 



The following may be mentioned as examples, reference being made to the 

 numerous individual forms described in the special works. 



(i) The haustoria of the Cuscutas, Cassythae, Rhinanthaceae, and Santalacese. 

 In the first two groups they arise from the twining stem, and penetrate the stem and 

 foliar organs of the host, round which it is twisted. In the two families last- 

 mentioned they arise on the roots, penetrating the roots of the host. In most cases 

 they have the general form of conical warts, with the base adhering firmly to the 

 host, and from the middle of the attached surface an approximately cylindrical or 

 flat peg, the sucker, penetrates the tissue of the host. In the haustorium an axial, 

 broad, small-celled strand of parenchyma, the core, is to be distinguished from a 

 large-celled cortex ; the core is continued directly into the sucker. In Cuscula the 

 extramatrical portion is very little developed, the haustorium may be said to consist 

 only of the swollen sucker, which springs from the surface of the stem adhering to 

 the host. Vascular strands branch off from the xylem-groups of the bundles of the 

 main axis, so as to pass through the core usually until they reach the inner surface 

 of the sucker where it meets the vascular bundles or the wood of the host. In the 

 normally developed haustoria of Thesium, Santalum, and Osyrrs there are two thick 

 flat strands, which first diverge, with a curved course at the periphery of the flask- 

 shaped core, and converge again in the sucker; in Cuscuta, Cassytha, and the 

 Rhinanthaceae there is an axial strand, usually penetrating to the wedged-in end of 



' Graf zu Solms-Laubach, Bau und Entwickelung der Eraahrungsorgane parasitischer Phanero- 

 gamen; Pringsheim's Jahrb. Bd.VI. — Id. Das Haustorium der Loranthaceen, &c., Halle, i8;5. — 

 L. Koch, .Entw. d. Cuscuten, in Hanstein's Botan. Abhandl. Bd. II. Heft 3. 



