Abstract
A criterion for the systematic study of the feeding habits of S. m. is suggested: namely, the use of number of crotches per sample as a basis of comparison. Observations made on more than 500 trees, including the examination of more than 100,000 crotches lead to the inference that the presence of wood which is either attracting or producing beetles in or near a given tree renders that tree more favorable for feeding attacks. Destruction of injured, dying and dead elm wood should not only aid materially in the control of the beetles by destroying their breeding places, but also should aid in reducing the spread of the disease by rendering healthy trees less subject to feeding attacks by possibly contaminated beetles.