1069 
Haustorium of S trig a lutea . 
quite clearly in optical section up to the stage of Fig. 5. This is of interest 
as throwing light on the much-debated point as to whether the haustoria of 
root-parasites in general are to be regarded as much-modified lateral roots, 
or as special exogenous outgrowths. In the case of Striga the haustoria 
show no homologies with lateral roots, being markedly exogenous in their 
origin, which is from the cells of the subhypodermal or (more rarely) the 
hypodermal layer. It cannot of course be said with certainty that every 
haustorium has originated from one of these layers, but even in the older 
stages it is clear that it has taken its rise in one of the layers between 
epidermis and endodermis. The haustoria may arise on roots of any age — 
e. g. the young haustoria of Figs. 3 and 4 are growing from roots decidedly 
older than in the case of the more mature haustorium of Fig. 5. As to the 
cause of their originating at any particular point, it may perhaps be of 
significance that the early stages of development have only been seen 
on roots which were in close proximity to young roots of maize still covered 
with root-hairs, and that the length of the haustorium before it penetrates 
the host (e. g. such a haustorium as that of Fig. 5) is about that of a maize 
root-hair. These facts lead me to put forward the purely tentative sugges- 
tion that stimulus due to contact with a hair of the maize root may be the 
determining cause of their development. 
Superficially, the haustorium first appears as a slight elevation on 
the surface of the root, elongated parallel to its axis. In a side view 
(in optical section) this elevation is seen to be due to the elongation of 
several (3-5) cells of the subhypodermal or hypodermal layer (Fig. 2). 
These cells are situated in the same row, parallel to the axis of the root, and 
the developing haustorium is therefore elongated in that direction. They 
begin to divide by transverse and longitudinal walls, pushing up the over- 
lying layer or layers still further to form a protuberance which is cone-like 
in section (Fig. 4) but elliptical in surface view. Thus the young haus- 
torium of Fig. 3 is only 3-4 cells deep back to front (i. e. above the level 
of the mother-root), and that of the later stage shown in Fig. 4 is only 
4 cells deep. This later stage is formed from that of Fig. 3 (if this has not 
by now come in contact with a maize root) by its growing out still further 
into a blunt finger-like process (Fig. 4), which may be twice as long as the 
root from which it develops is broad. It never seems to develop further 
than this without contact with the root of the host. Only one case of self- 
attachment was seen ; in this the haustorium was still of the size and general 
shape of Fig. 4, and had penetrated to- the depth of about three cells into 
the cortex of the Striga root, with which its cells then gradually merged. 
None of the developments which are seen when the haustorium touches 
a foreign root had taken place. 
The cells of the developing haustorium have the appearance of ordinary 
cortical cells, and it never assumes the small-celled and meristematic 
