species, its life history and habits, the nature of the damage it 
causes, the forest type and terrain, the size and accessibility of the 
area, a working knowledge of natural control factors affecting it, 
and the availability of trained personnel and funds. Aircraft are 
widely used in detection surveys and to a lesser extent in evaluation 
surveys. Conditions may be recorded by trained observers riding 
with the pilot (3, 339, 341) ; or by aerial photography (745, 2, 338, 
340, 746, 142). Unfortunately, evidence of infestation by many 
insect species is difficult or impossible to detect from the air. 
Where these insects are concerned, the only recourse is to use 
ground survey methods. To the extent that it is possible to use 
them, aerial surveys are far less time consuming and costly than 
comparable ground surveys. 
Ground surveys are not only required to detect infestations of 
many species but are essential for evaluation purposes. In most 
cases, they are highly complex and require the services of highly 
trained personnel. Ground surveys for evaluation purposes pre- 
sent many different sampling problems, such as when, where, and 
how to sample in order to obtain needed information within 
desired confidence limits. The choice of procedure is governed 
largely by the insect species and the intensity and size of the 
infestation. In addition to insect population density and size, data 
also may be collected to identify and record the abundance of 
natural control factors. In all cases information is sought and 
analyzed to determine two things: what likely will happen if no 
action is taken to suppress the infestation, and what might be 
expected to happen if suppressive measures are applied. 
Much progress has been made in recent years in reducing the 
time and cost of conducting surveys, particularly through the use 
of sequential sampling techniques. These techniques provide for 
flexibility in sample size in contrast to conventional techniques 
which usually specify a fixed number of sampling units to be 
examined. Units chosen at random from a sample are examined 
in sequence until the sample falls into one or more classes dis- 
tinguished by specified limits (743). Sampling plans have been 
published for several important forest insect pests (158, 157, 393, 
894, 542, 612, 692). 
Forest Insect Control 
The impact of naturally occurring adverse environmental fac- 
tors on forest insects is so great that the majority of species never 
occur in sufficient numbers to constitute an economic hazard. 
Even those insects that become sufficiently abundant at times to be 
considered ‘“‘pests” are also subject to control by these factors, 
though usually to a lesser degree and frequency. Depending on the 
insect and the value of the forest resources affected, a given de- 
gree of natural control may be economically effective. 
When natural control factors fail to hold populations to eco- 
nomically tolerable levels, it may be necessary to apply arti- 
ficial controls. Depending on the insect and its biology and eco- 
logical relationships, a wide variety of materials and methods are 
available for suppressing populations. These may be applied 
directly against the insect to interfere with its growth or repro- 
15 
