

44 



or younger, and both come in contact with the cortex of the host-root, while 

 in Thesium the outer pair is smaller and detaches from the host-root after 

 a new fold is formed inside. In the median line of each fold there runs 

 a striated band which consists of a number of cells stretched in the same 

 direction ; and these bands run all towards the proximal end, unite them- 

 selves together in their course to form a band on each side of the 

 central cylinder, and ultimately vanish near the base of haustorium. The 

 central cylinder which looks like an inverted flask in longitudinal section 

 and elliptical in cross-section, is surrounded by the parenchymatous cortex 

 which is divided into outer and inner parts by the band above mentioned. 

 The essential part of the cylinder consists of a pair of bundles of reticulated 

 vessels with a large mass of parenchymatous cells between them. So far 

 the haustoria of Bucldeya almost coincide in structure with those of Thesium. 



But the most striking and noteworthy structural change appears when 

 the haustorium begins its secondary growth due to the activity of 

 cambium between the cortex and the central cylinder. As it is 

 the case with the secondary growth of the stem in general, so also here the 

 cortex of the haustorium does not make any remarkable increase in its 

 breadth, while the central cylinder grows from year to year to a considerable 

 thickness. 



Owing to the growth of the central cylinder the haustorium which is 

 at first elliptical in cross-section becomes roundish, and afterwards by the 

 further activity of cambium, it resumes the elliptical shape, the longer axis ^ 

 however, taking the position of the shorter axis of the former. While thus 

 the transverse growth of the haustorium is going on, no secondary longi- 

 tudinal growth takes place, and then the haustorium assumes naturally a 

 somewhat discoidal shape. At this stage we can distinguish clearly a number of 

 m edullary rays which, consist of parenchymatous cells stretched transversally 

 and traverse the bundles to connect the central parenchyma with the 

 cortex. They run at first parallel, but as the cylinder grows they come 

 to run radially. As a whole, therefore, in the old haustorium the general 

 arrangement of its elements is comparable to that of the stem of dicotyle- 

 donous plants. 



As the result of the change mentioned above, the attaching fold of the 

 cortex and the sucking process 1} which can be easily distinguished in the 



1) The so called sucking process was distinguished in Thesium, Osyris and Santalum by 

 the previous authors, but it seems to me, judging from the haustorium of Buckleya, that this 

 part is nothing but the central cylinder, so that it is not necessary to distinguish it as a 

 special part. 



