116 
Where possible this operation should be carried out in the 
autumn, and as the dressing is lable to crack it should be 
examined from time to time and renewed if necessary. The 
smaller wounds produced by pruning cannot be treated in this 
way owing to the necessity of pruning back to a bud which 
would be killed by such treatment. These small wounds fre. 
quently form the favourite place of entry for Nectria ditissima 
Fortunately it 1s now becoming recognized that the wholesale 
and indiscriminate pruning to which our fruit trees are gener- 
ally exposed is unnecessary. This fact will tend to keep down 
to a considerable extent the dreaded canker fungus. The one 
other method we can consider now of avoiding conditions which 
are likely to bring about infection is the one employed in com- 
bating the spread of some of the heteroecious rusts. For such 
fungi an alternative host plant is essential for their complete 
development, certain spore forms being produced on one host 
plant A which then infect the second host plant B. Their life 
cycle is only complete when the host A is again infected 
through the spores produced on B. By breaking the life-cycle 
one can obviously check the extended spread of such parasites. 
If A is to be cultivated, then it is only necessary to keep down 
the alternate host plant in the immediate vicinity. Numerous 
examples of the practical application of this principle might be 
quoted. Garden peas, for instance, are attacked by a rust 
(Uromyces pisi) which forms an aecidial stage on spurges. 
The spores produced on this host pass in turn to the cultivated 
garden peas, reproducing the well known disease. A second 
example is provided by Gymmnosporangium Sabinae. This rust 
is common on pears, the aecidial stage being formed in abund- 
ance on the foliage and fruits. The crop produced on infected 
trees is light and as a rule unsaleable. The teleutospore stage 
occurs on the Common Savin, where it forms characteristic jelly- 
like masses. If these hosts in the immediate neighbourhood of 
the pears are destroyed then, as the fungus cannot complete 
its life cycle, the epidemic ceases. To secure satisfactory re — 
sults by destroying the alternative host plant it 1s essential that 
the whole neighbourhood should be cleared. As this generally 
demands a certain amount of cooperation it cannot always be 
relied upon to give altogether satisfactory results. 
Setting aside as an obvious predisposing cause the using of 
soil for potting purposes which has already carried a diseased 
crop we may turn to a consideration of the best methods of 
disposing of tissues known to be diseased. The ideal method 
of course is to burn all such plant remains. In practice, how- 
ever, this is by no means as simple as it would appear to be at 
first sight. One cannot for instance readily burn all the 
diseased haulm of a crop of potatoes, neither can a farmer affor 
