20 BULLETIN 1380, U. S. DEPARTMENT OF AGRICULTURE 
sterile layer over the bark, or this layer may be formed in small 
patches not connected with the more mature parts. This stage may- 
be followed by the appearance of small sterile pink or white pus- 
tules arranged in parallel rows in the cracks of the bark. This 
stage may appear over the surface covered with the thin white 
film or on areas not previously fruiting. A third stage appears in 
the form of irregularly rounded or elongated orange-red bodies. 
These are at first embedded in the cortex, but later rupture it and 
appear singly or in clusters on the surface. These bodies when 
they emerge resemble species of Xectria and consist of masses of 
spores. This spore-bearing stage was originally considered to be 
a distinct fungus and was described as Necator decretus Mass. It 
has been shown recently, however, that it is only a phase in the life 
history of the same fungus and is a rapid means of propagation. 
This stage is present on the material from the Amazon Valley. 
Xowell states that so far it has not been recognized in West Indian 
examples. - 
Inoculation experiments in the East have demonstrated that the 
fungus is capable of infecting its host directly through the unin- 
jured bark; hence it is not a wound fungus, though it appears that 
infection may take place through wounds. 
The type of injury caused is usually first shown by the yellowing 
of the leaves of the parts above the infection. The infection en- 
circles young stems or spreads from basal infections of branches to 
others at the point of union, and if complete girdling results all 
parts above the canker die. If the fungus is arrested by dry weather, 
the bark over the affected area cracks, causing the formation of 
open wounds. 
The vigor of the fungus is known to depend largely on moisture 
and exposure. Hence, the usual measures for reducing the rav- 
ages of the parasite in plantations have been to regulate spacing, to 
provide drainage, to avoid intercrops that serve as hosts, and to keep 
trees in a high state of vigor. Direct control would require that 
all infected branches be cut out and burned. The branches should 
be cut several inches beyond the visible zone of infection. This is 
necessary, for the reason that in practically all fungous infections 
of this nature the mycelium has advanced considerably beyond the 
immediate cankered area. All branch knots and larger wounds on 
the main stem should be thoroughly saturated with coal tar. Spray- 
ing uninfected trees regularly after rains may prevent infection by 
newly distributed spores. 
DIE-BACK 
The die-back fungus (Diplodia fh-eobromae (Pat.) Nowell) is 
very widely distributed in the Tropics and has been reported on a 
great many different kinds of plants from the West Indies, the East 
Indies, Oceania, South America, Africa, and southern Asia. The 
fungus was first reported from the American Tropics by Howard in 
1901. Pie found it causing a die-back on cacao in Grenada. 
During the present investigation it was found on the stems and 
fruits of cacao and mango under circumstances implying parasitic 
tendencies. 
