2 BULLETIN 250, U. S. DEPARTMENT OF AGRICULTURE. 



eaten to a greater or less extent. On 27 species the larvae fed very 

 slightly, and on 19 no feeding was noted. 



In 1905, when work to control the gipsy moth was resumed, after a 

 lapse of five years, it was found that the infestation had spread over 

 such a wide territory that extermination of the moth was impossible. 

 The following year (1906) an appropriation was made by Congress 

 and control work to prevent the spread of this insect was commenced 

 by the Bureau of Entomology. The infestation was so heavy in 

 eastern Massachusetts that the principal efforts, aside from parasite 

 introduction, was to clear the main highways in order to prevent the 

 distribution of caterpillars through the medium of passing vehicles. 

 As early as 1907 it was noticed by a number of observers that some 

 species of trees were more often defoliated than others. These ob- 

 servations also indicated that pine was one of the species which was 

 not readily attacked. In order to secure more information on this 

 subject a number of experiments were carried on by the writer under 

 the direction of Mr. A. H. Kirkland, who was superintendent of moth 

 work for the State of Massachusetts. 



The preliminary work was commenced in the spring of 1907, 

 newly hatched caterpillars of the gipsy moth being placed in jars 

 and furnished with pine foliage. In the feeding experiments it was 

 not possible to induce newly hatched caterpillars to feed and develop 

 on white pine, but when they were furnished with oak foliage no 

 serious difficulty was encountered. Several field experiments were 

 also carried on to determine whether a pine growth could be pro- 

 tected from the gipsy moth by placing bands of tanglefoot on the 

 trees in order to prevent caterpillars from climbing to the foliage. 

 These experiments were repeated the following season, and on ac- 

 count of the success of the field experiment, wherein several acres of 

 pine growth were protected by using tanglefoot bands, it was obvious 

 that more detailed information regarding food plants was necessary 

 to the proper conduct of the work than had been secured from the 

 experiments reported in 1896. Numerous field observations and 

 several experiments of greater scope were then conducted, but in 

 1911 it seemed desirable to investigate, in a thorough and systematic 

 way, the entire matter of preferred food plants. Accordingly plans 

 were formulated to carry on an elaborate series of laboratory experi- 

 ments, using first the more common trees occurring in the infested 

 ments at the Gipsy Moth Laboratory of the Bureau of Entomology at 

 Melrose Highlands, Mass., using first the more common trees occur- 

 ring in the infested region, with the idea of taking up the rarer species, 

 as well as the woodland shrubs and undergrowth, as soon as opportu- 

 nity permitted. The experiments were arranged so that two lots of 



