6 THE AUD U'BSONY BUD rae 
being attempted by other communities in other states. The Michigan High- 
way Department has begun an extensive removal program of dead and dy- 
ing elms along its right-of-ways. This is indeed a step toward an overall 
sanitation program. Minnesota has indicated interest in a state-wide pro- 
gram of this type. 
To my knowledge no community has successfully controlled Dutch elm 
disease by using spray programs; nor has complete success resulted from 
the combination spray-sanitation programs. Conclusions cannot be drawn 
for the sanitation program because too little evidence is currently available. 
The lack of program preference has often resulted in the adoption of spray 
programs. The general policy is to use D.D.T. until something better is 
developed. If this attitude could be changed and the use of D.D.T. discon- 
tinued, research at the chemical companies would probably be increased on 
selective rather than broad-spectrum poisons. 
We have several reasons for skepticism regarding the validity of spray- 
ing. An interesting approach was recently announced by Kenneth E. F. 
Watt of the Statistical Research Service, Canadian Department of Agricul- 
ture, Ottawa. By use of a mathematical theory of insect control, long-range 
effects of various methods could be compared. The early results of his study 
indicate that the benefits from insecticide application may be “illusory.” He 
has devised computer programs that allow the machine to simulate popula- 
tion trends of an insect pest over 100 generations in the face of environ- 
mental hazards and artificial control measures. The computer takes into 
account weather and other factors that affect the insect population. 
Mr. Watt observed: “The general conclusion from these simulation studies 
is that insecticides do not, according to this model, depress pest population 
level as much as one would expect.” For some sets of circumstances, the 
sprayed population actually grows to exceed that which would naturally be 
supported in the given area if spraying had not been done. “‘In view of the 
importance of the indications from this primitive model, it would seem 
worthwhile to collect field data to check these findings.” Possibly reduced 
competition between pest and beneficial insects as an insecticide kills off 
large numbers may be a basic explanation for cases of insecticide failure. 
Such tendencies are indicated by several studies. The behavior, physiology, 
and genetics of insects also may show why spray programs are apt to be 
unsuccessful and might actually aid in the spread of Dutch elm fungus. 
INSECT BEHAVIOR. We have positive evidence that some insects may sense 
areas of insecticide application by taste or smell (e. g., Anopheles mosquito, 
house fly). In such instances the insects avoid the sprayed area. It is pos- 
sible that behavior of this type, since it enables its possessor to survive, 
could be passed on to future generations. Thus a particular insecticide could 
become useless on that species. If we associate this trait with elm bark 
beetles which breed in dead portions of elms, we can see that a dormant 
spray, before insects emerge from the bark, could cause the beetle popu- 
lation, or a part of it, to leave the treated area upon emergence and invade 
some neighboring region of unsprayed elms. Under normal conditions the 
beetles are capable of moving two or three miles from their brood sites. 
Driven by insecticides, they might travel greater distances. This phase of 
the problem requires consideration. 
GENETIC CONSIDERATIONS. This topic encompasses those discussed previ- 
ously. Extensive research, particularly with the fruit fly (Drosophila spp.), 
indicates that resistance to D.D.T. may be acquired by particular individu- 
als and resistant generations may follow. A second possibility is that some 
