June, 1939 The Queensland Naturalist 
37 
Royal Society of Queensland by A. Norton in 1886.) 
The stump was not uprooted to verify the theory that 
its roots were grafted to those of an adjacent tree, but 
in view of the fact that root grafting is common in 
Eucalypts it seems a reasonable explanation. 
Habitual root parasitism, as distinct from more or 
less haphazard and dispensable root grafting, character- 
ises a considerable number of flowering plants belong- 
ing to different families. Special suctorial organs, the 
haustoria, are developed and rob the host plants of 
water, inorganic salts, and organic food materials. This 
habit is highly specialised and the plants must have a 
host. It is, however, a habit which has arisen in- 
dependently in different families, and though the 
haustoria have the same functions of parasitism, they 
differ in details of their structure. In the family 
Scrophulariaceae self-supporting plants such as the snap- 
dragon, penstemon and veronica are in the majority, 
but the evolution of two tribes, the Gerardiae and the 
Rhinantheae, has been along parasitic lines; Stiga, the 
sugar cane parasite, Melampyrum , Bartsio and Euphrasia 
are common examples. The family Santalaceae, with 
about 250 species, is parasitic, and in some species 
hailstorm are often found attacking almost any root with 
which they come into contact, even their own. Exocarvus 
cupressiformis , the native cherry, was the first Australian 
snecies to be investigated. Later investigations showed 
that all the Australian species examined had the same 
habit, but to a different degree. Some were very catholic 
in their taste, but others had a very narrow range of 
hosts, one species being apparently restricted to one par- 
ticular plant only. Extreme specialisation of this sort 
means that the existence of the species is somewhat pre- 
carious unless the host is common. Two important Aus- 
tralian santalaceus parasites are the sandalwood ( Fusanus 
sweat us) and the quondong (F. acuminatus). The san- 
dalwood is parasitic on mulga and other acacias for the 
most part, and the recognition of this fact makes possible 
its commercial cultivation. It is of no use to plant its 
seeds in cleared land ; they must have a nurse-tree to 
parasitise. The Indian sandalwood ( 8 ant alum album) is 
also parasitic, but it has a range of several hundred host 
species in its native rain forests. Before the spike disease 
of this important plant was understood, it was suggested 
that perhaps in its wholesale attack on its neighbours the 
sandalwood struck one which caused some derangement of 
