34 STUDIES IN ROOT-PARASITISM. 
very minor importance, as will be seen directly. The differ- 
ences from Olax haustoria, on the other hand, are so many and 
great that their mere enumeration would mean a recapitula- 
tion of much of this and the preceding paper in this series. In 
only three particulars has the haustorium of Cansjera been 
found to resemble that of Olax rather than that of Santalum: 
the haustorium is whitish and not coloured brown with age, 
no compound haustoria have been met with and secondary 
thickening in the wood is greater in the vascular loop than in 
the axis. In the characters now to be mentioned the hausto- 
ria of Cansjera differ as much from Olax as they do from 
Santalum. 
The haustorium of Cansjera is smaller and less regular 
than that of Santalum, the early irregularities being ap- 
parently due to the abundance of lenticels which are not 
found in Santalum, In extreme age the haustorium frequent- 
ly becomes fantastically lobed, the attachment to the host’s 
root being small if not minute. The general shape of the haus- 
torium thus becomes broad above and narrow below: the case 
in Santalum is the reverse, some of the oldest and largest haus- 
toria collected being smooth and conical or limpet-shaped. 
The development of the young haustorium is very similar 
to that in both the haustoria thus far considered, but when 
the nucleus is formed, its cells are sharply separated off from 
those above, the latter being elongated in a direction at right 
angles to the axis. When entrance is being effected, the 
cells of the upper part of the nucleus assume a striking appear- 
ance. They become much enlarged, develop dense, granular 
protoplasm and large nuclei and nucleoli—assume in fact the 
structure of glandular cells. The gland proper is small and 
slit-like and there is no duct. It thus would appear that, in 
Cansjera, the whole of the nucleus forms a gland, the so-called 
‘‘gland’’ perhaps acting as its lumen and duct. In Santalum, 
the cells of the upper part of the nucleus do not take on any 
peculiar form and, although sharply marked off from those 
above, remain small. The gland, on the other hand, is well 
