Cuscuta and its Host. 
661 
present in the haustorium in October, but were not formed in the sieve 
tubes of the main stem till November (Figs. 44 and 45). On the transverse 
plates the callus is sometimes nearly all collected on the parasite side of the 
plate (Fig. 45) ; in other cases, a continuous layer of callus lines all the 
walls of the elements of a sieve tube of the haustorium. Here, as in other 
cases, the callus is apparently deposited by the protoplasm. 1 
It is interesting to find that the haustorial sieve tubes do not terminate 
at the level opposite the phloem of the host, but are continued down the 
axis of the haustorium into the host pith (C. reflexa on Begonia , C. sp. on 
Solid'ago , &c.). 
Some attempt was made to determine the number of threads present 
in the longitudinal walls of a short young sieve tube of the haustorium 
(Fig. 43). The two groups, A and B, were made up of 11 and 16 threads 
respectively. The two groups together measured 11-2 \x across* the whole 
cell being 24-4//. across at this point and 64-4^ in length. From these 
numbers some estimation may be made of the large number of threads 
present in the wall of such a cell. 
III. On the Relations between Host and Parasite. 
A. Outlines of the Development of the Haustorium. 
The broad outlines of the development of the young haustorium in 
Cuscuta and its manner of penetrating the host have been described to 
some extent by Peirce. 2 A sucker-like organ first arises from the epidermis 
of the mother stem and adheres firmly to the host plant; its formation is 
then followed by the ingrowth of the true haustorium, which has originated 
endogenously, mainly from the cortical region just outside the pericycle, 
and seems very properly regarded as morphologically an adventitious 
root. 3 The cells of the sucker, or ‘ prehaustorium ’, dissolve their way into 
the host plant, partly by pressure, partly by the excretion of a ferment, 4 
and into the space thus made the haustorium grows, enlarging the opening 
and becoming surrounded by a mass of compacted dead cell-walls. 
The haustorium grows at its tip and finally becomes connected with 
the bundles of the host plant, its method of growth, either continuing in 
a straight line towards a single host bundle, or branching into two parts 
which diverge to join more than one host bundle, varying with the arrange¬ 
ment and size of the bundles of the host. 5 These variations in form do 
not concern us here. 
The following more detailed account is derived in great part from 
1 Hill, 19013, pp. 597-600; 1908, pp. 277-8; Sykes, 1908, pp. 316-17. 
2 Peirce, 1893, pp. 295 ff., and 1894, pp. 99 ft. 
3 Ibid., 1893, p. 305. 
4 Ibid., 1894, pp. 99-106 ; the invading cells were found to have the power of corroding starch 
grains. 5 See ibid., 1893. 
