332 MR. H. MARSHALL WARD ON THE 
It will be noted that the largest production of new leaves occurred 
in the interval Sept. 23 to Oct. 15: this corresponds to a period 
of rapid vegetative growth which then took place during moist, 
cloudy, and warm weather. After this, the rapidity of growth 
ceased as the air became drier towards the end of the year. 
The greatest development of *leaf-disease," however, did not 
take place until the interval Nov. 9 to Dec. 25, i. e. about a 
month or six weeks later. This is in accordance with what is 
already known of the behaviour of the fungus, as, indeed, are all 
the details of this experiment. 
It is now possible to understand the explanation of some diffi- 
culties which have been raised in objection to the generalizations 
presented in this and preceding publications on this subject. 
One of the commonest popular difficulties freely expressed in 
connexion with the history of this and similar diseases of plants 
may be stated somewhat as follows :—1t being proved that the 
fungus in the leaves causes the disease-spots to appear and the 
leaves to fall, how can such huge effects arise from such appa- 
rently small causes ? The complete reply to this question would 
be a lengthy summary of our knowledge of the physiology of the 
leaf and its connexion with the life of the tree, It must be re- 
membered that the mycelium of Hemileia not only robs the leaf 
of valuable materials, on the manufacture of which a large expen- 
diture of energy was required, but it diverts the flow of nutri- 
tive substances. - Moreover, in occupying space in the tissues of 
the plant, it prevents these tissues from fulfilling functions of 
service to the coffee-tree. To replace the damage done, the leaves 
would require either to do more work in a given time, or to have 
a longer lease of life to work in; whereas they have less oppor- 
tunity of doing either. When the fungus is present in quantity 
there is less available substance sent down to the tree for the 
support of crop; and the life of the leaf is shortened. Hence the 
amount of fruit matured by the tree must be diminished unless 
one of two things happens—unless (1) more leaves are quickly 
formed and enabled to work for the benefit of the tree; or (2) 
more food-substances can be sent into the leaves and elaborated 
in the time allowed, so to speak, by the voracious pest. I have 
directly proved that the coffee-tree cannot produce crop if the 
leaves be stripped from its branches—a fact in accordance with 
what we otherwise know of similar cases. 
By cultivation and proper manuring, it is conceivable that 006 
