2l2 
Jean Dufrenoy. 
THE ENDOTROPHIC MYCORHIZA OF ERICACE/E. 
By Jean Dufrenoy, 
Assistant, Station Biologique d’Arcachon. 
[With Four Figures in the Text]. 
HAT there is a necessary symbiosis between Ericaceae and 
endophytes, has been recently demonstrated by Dr. Chevely 
Rayner as far as Calluna vulgaris is concerned (1). Ovarial 
infection was proved by Rayner for Calluna, and reported by her 
for Vaccinium. 1 This makes it specially interesting to make a 
close study (3) of that very peculiar Ericaceous plant Arbutus 
Unedo, a shrub very common in Mediterranean countries and also 
in the S.W. of France, 2 but of which the mycorhiza does not seem 
to have been investigated so far, although in view of the edaphic 
peculiarities of Arbutus, it is certainly of considerable ecological 
interest. 
The roots of Arbutus are clothed with a dense mantle of hyphae. 
These are protected with a thick greyish membrane (Fig. I). Some 
of the branches they form have but a thin yellow membrane, and 
penetrate further into the tissues of the roots, whence they invade 
the whole plant. Haustoria are formed in the living cells of the 
host. These latter, as a rule, do not react, but the cytoplasm first 
becomes granular and then dwindles as the fungus develops. 
The cortex of the root often consists of empty cells, for here 
both the host cells and the invading fungus may ultimately die. 
Fungus-tubercles. 
Rootlets which have been inoculated early cease to growapically, 3 
and develop into small pear-shaped tubercles. Nearly all the 
epidermal cells develop into root-hairs. Algae and bacteria collect 
around them and form a mucus which stains readily with eosin. 
The fungus grows not only on their surface, but penetrates and 
invades the external layer of the cortex (Fig. I, 3). The base of the 
shoot develops into a very large tubercle from which many lateral 
1 Ovarial infection of Vaccinium was not admitted by Stahl, and no mention 
of it is made in the most recent American work on the subject (2). 
3 In the latter country this shrub is rather strangely limited to the old 
Pine forests of Arcachon and La Teste (60 kilometres south of Bordeaux) 
where the sand is covered by a peaty layer more than 20 eras. deep. Arbutus 
is never met with in the neighbouring Pine forests that have been grown in the 
sandy dunes of Gascony for the last century. This strict limitation has 
puzzled many naturalists and has been the object of several notes (4). 
s This, which was also observed in the cases of Vaccinium and of Arbutus, 
supports P. Vuillemin in the belief that even the tubercles of leguminous 
plants are abortive roots. 
/ 
